mirror of https://github.com/vladmandic/human
7653 lines
1.6 MiB
7653 lines
1.6 MiB
"use strict";/*
|
|
Human
|
|
homepage: <https://github.com/vladmandic/human>
|
|
author: <https://github.com/vladmandic>'
|
|
*/
|
|
|
|
var Human=(()=>{var zf=Object.defineProperty;var JR=Object.getOwnPropertyDescriptor;var QR=Object.getOwnPropertyNames;var e_=Object.prototype.hasOwnProperty;var t_=(e,t,n)=>t in e?zf(e,t,{enumerable:!0,configurable:!0,writable:!0,value:n}):e[t]=n;var ia=(e,t)=>{for(var n in t)zf(e,n,{get:t[n],enumerable:!0})},n_=(e,t,n,s)=>{if(t&&typeof t=="object"||typeof t=="function")for(let r of QR(t))!e_.call(e,r)&&r!==n&&zf(e,r,{get:()=>t[r],enumerable:!(s=JR(t,r))||s.enumerable});return e};var s_=e=>n_(zf({},"__esModule",{value:!0}),e);var ge=(e,t,n)=>(t_(e,typeof t!="symbol"?t+"":t,n),n),dv=(e,t,n)=>{if(!t.has(e))throw TypeError("Cannot "+n)};var Zd=(e,t,n)=>(dv(e,t,"read from private field"),n?n.call(e):t.get(e)),Yd=(e,t,n)=>{if(t.has(e))throw TypeError("Cannot add the same private member more than once");t instanceof WeakSet?t.add(e):t.set(e,n)},Jd=(e,t,n,s)=>(dv(e,t,"write to private field"),s?s.call(e,n):t.set(e,n),n);var Mxe={};ia(Mxe,{Human:()=>U4,default:()=>U4,defaults:()=>Ba,draw:()=>O4,env:()=>pe,match:()=>V4,models:()=>m1});function le(...e){let t=new Date,n=`${t.getHours().toString().padStart(2,"0")}:${t.getMinutes().toString().padStart(2,"0")}:${t.getSeconds().toString().padStart(2,"0")}.${t.getMilliseconds().toString().padStart(3,"0")}`;e&&console.log(n,"Human:",...e)}function pv(e,t){let n=e.endsWith("/")?"":"/",r=t.startsWith(".")||t.startsWith("/")||t.startsWith("http:")||t.startsWith("https:")||t.startsWith("file:")?`${t}`:`${e}${n}${t}`;if(!r.toLocaleLowerCase().includes(".json"))throw new Error(`modelpath error: expecting json file: ${r}`);return r}var ue=()=>typeof performance!="undefined"?performance.now():parseInt((Number(process.hrtime.bigint())/1e3/1e3).toString());function t3(e,t,n="config",s=[]){for(let r of Object.keys(t))if(typeof t[r]=="object")t3(e[r],t[r],r,s);else{let a=e&&typeof e[r]!="undefined";a||s.push({reason:"unknown property",where:`${n}.${r} = ${t[r]}`});let o=e&&typeof e[r]==typeof t[r];a&&!o&&s.push({reason:"property type mismatch",where:`${n}.${r} = ${t[r]}`,expected:typeof e[r]})}return t.debug&&n==="config"&&s.length>0&&le("invalid configuration",s),s}function Xt(...e){let t=n=>n&&typeof n=="object";return e.reduce((n,s)=>(Object.keys(s||{}).forEach(r=>{let a=n[r],o=s[r];Array.isArray(a)&&Array.isArray(o)?n[r]=a.concat(...o):t(a)&&t(o)?n[r]=Xt(a,o):n[r]=o}),n),{})}var Ba={backend:"",modelBasePath:"",cacheModels:!0,wasmPath:"",wasmPlatformFetch:!1,debug:!1,async:!0,warmup:"full",cacheSensitivity:.7,skipAllowed:!1,deallocate:!1,filter:{enabled:!0,equalization:!1,width:0,height:0,flip:!1,return:!0,brightness:0,contrast:0,sharpness:0,blur:0,saturation:0,hue:0,negative:!1,sepia:!1,vintage:!1,kodachrome:!1,technicolor:!1,polaroid:!1,pixelate:0},gesture:{enabled:!0},face:{enabled:!0,detector:{modelPath:"blazeface.json",rotation:!0,maxDetected:1,skipFrames:99,skipTime:2500,minConfidence:.2,iouThreshold:.1,mask:!1,return:!1},mesh:{enabled:!0,modelPath:"facemesh.json",keepInvalid:!1},attention:{enabled:!1,modelPath:"facemesh-attention.json"},iris:{enabled:!0,modelPath:"iris.json"},emotion:{enabled:!0,minConfidence:.1,skipFrames:99,skipTime:1500,modelPath:"emotion.json"},description:{enabled:!0,modelPath:"faceres.json",skipFrames:99,skipTime:3e3,minConfidence:.1},antispoof:{enabled:!1,skipFrames:99,skipTime:4e3,modelPath:"antispoof.json"},liveness:{enabled:!1,skipFrames:99,skipTime:4e3,modelPath:"liveness.json"}},body:{enabled:!0,modelPath:"movenet-lightning.json",maxDetected:-1,minConfidence:.3,skipFrames:1,skipTime:200},hand:{enabled:!0,rotation:!0,skipFrames:99,skipTime:1e3,minConfidence:.5,iouThreshold:.2,maxDetected:-1,landmarks:!0,detector:{modelPath:"handtrack.json"},skeleton:{modelPath:"handlandmark-full.json"}},object:{enabled:!1,modelPath:"mb3-centernet.json",minConfidence:.2,iouThreshold:.4,maxDetected:10,skipFrames:99,skipTime:2e3},segmentation:{enabled:!1,modelPath:"selfie.json",blur:8}};var je={};ia(je,{Abs:()=>il,Acos:()=>pc,Acosh:()=>hc,AdadeltaOptimizer:()=>j0,AdagradOptimizer:()=>q0,AdamOptimizer:()=>X0,AdamaxOptimizer:()=>K0,Add:()=>xa,AddN:()=>po,All:()=>fc,Any:()=>mc,ArgMax:()=>ho,ArgMin:()=>gc,Asin:()=>yc,Asinh:()=>Ac,Atan:()=>xc,Atan2:()=>vc,Atanh:()=>bc,AvgPool:()=>fo,AvgPool3D:()=>zp,AvgPool3DGrad:()=>Hm,AvgPoolGrad:()=>Gm,BackendWasm:()=>GC,BatchMatMul:()=>mo,BatchToSpaceND:()=>ll,Bincount:()=>jm,BroadcastArgs:()=>qm,BroadcastTo:()=>h6,Callback:()=>Ck,CallbackList:()=>R8,Cast:()=>go,Ceil:()=>yo,ClipByValue:()=>ba,Complex:()=>Lp,ComplexAbs:()=>Bp,Concat:()=>ul,Conv2D:()=>Ao,Conv2DBackpropFilter:()=>Xm,Conv2DBackpropInput:()=>xo,Conv3D:()=>Wp,Conv3DBackpropFilterV2:()=>Km,Conv3DBackpropInputV2:()=>Zm,Cos:()=>bo,Cosh:()=>vo,CropAndResize:()=>dl,Cumprod:()=>cl,Cumsum:()=>wo,CustomCallback:()=>D8,DataStorage:()=>Mp,DenseBincount:()=>Ym,DepthToSpace:()=>pl,DepthwiseConv2dNative:()=>ko,DepthwiseConv2dNativeBackpropFilter:()=>Jm,DepthwiseConv2dNativeBackpropInput:()=>Qm,Diag:()=>e0,Dilation2D:()=>Vp,Dilation2DBackpropFilter:()=>hm,Dilation2DBackpropInput:()=>pm,ENV:()=>ky,EarlyStopping:()=>Tk,Einsum:()=>Up,Elu:()=>So,EluGrad:()=>t0,Environment:()=>d6,Equal:()=>hl,Erf:()=>wc,Exp:()=>Co,ExpandDims:()=>fl,Expm1:()=>ml,FFT:()=>n0,Fill:()=>kc,FlipLeftRight:()=>gl,Floor:()=>To,FloorDiv:()=>No,FromPixels:()=>bp,FusedBatchNorm:()=>Eo,FusedConv2D:()=>Qa,FusedDepthwiseConv2D:()=>eo,GPGPUContext:()=>ju,GatherNd:()=>Al,GatherV2:()=>yl,GraphModel:()=>$h,Greater:()=>xl,GreaterEqual:()=>Ro,History:()=>_8,IFFT:()=>s0,Identity:()=>_o,Imag:()=>Gp,InputSpec:()=>rn,IsFinite:()=>Ic,IsInf:()=>Sc,IsNan:()=>Cc,KernelBackend:()=>cc,LRN:()=>Hp,LRNGrad:()=>a0,LayerVariable:()=>k8,LayersModel:()=>ha,LeakyRelu:()=>Do,Less:()=>bl,LessEqual:()=>vl,LinSpace:()=>r0,Log:()=>$o,Log1p:()=>Tc,LogSoftmax:()=>m6,LogicalAnd:()=>wl,LogicalNot:()=>kl,LogicalOr:()=>Nc,LogicalXor:()=>f6,LowerBound:()=>X_,MathBackendWebGL:()=>id,Max:()=>Po,MaxPool:()=>Oo,MaxPool3D:()=>jp,MaxPool3DGrad:()=>i0,MaxPoolGrad:()=>o0,MaxPoolWithArgmax:()=>l0,Maximum:()=>Fo,Mean:()=>Mo,Min:()=>zo,Minimum:()=>Lo,MirrorPad:()=>Bo,Mod:()=>Ec,MomentumOptimizer:()=>Z0,Multinomial:()=>u0,Multiply:()=>Wo,Neg:()=>Il,NonMaxSuppressionV3:()=>Cl,NonMaxSuppressionV4:()=>Rc,NonMaxSuppressionV5:()=>Tl,NotEqual:()=>Sl,OP_SCOPE_SUFFIX:()=>Ty,OneHot:()=>El,OnesLike:()=>Nl,Optimizer:()=>ka,OptimizerConstructors:()=>Wa,Pack:()=>Rl,PadV2:()=>Vo,Pool:()=>K_,Pow:()=>Uo,Prelu:()=>Go,Prod:()=>Ho,RMSPropOptimizer:()=>Y0,RNN:()=>Qr,Range:()=>_c,Rank:()=>A3,Real:()=>qp,RealDiv:()=>Io,Reciprocal:()=>Dc,Reduction:()=>Jn,Relu:()=>jo,Relu6:()=>Ko,Reshape:()=>_l,ResizeBilinear:()=>Xo,ResizeBilinearGrad:()=>d0,ResizeNearestNeighbor:()=>qo,ResizeNearestNeighborGrad:()=>c0,Reverse:()=>Dl,RotateWithOffset:()=>jl,Round:()=>$l,Rsqrt:()=>Zo,SGDOptimizer:()=>vh,ScatterNd:()=>Pl,SearchSorted:()=>p0,Select:()=>Fl,Selu:()=>$c,Sequential:()=>sc,Sigmoid:()=>Jo,Sign:()=>Pc,Sin:()=>Yo,Sinh:()=>Ml,Slice:()=>Ol,Softmax:()=>ti,Softplus:()=>Fc,SpaceToBatchND:()=>zl,SparseFillEmptyRows:()=>Xp,SparseReshape:()=>Oc,SparseSegmentMean:()=>Kp,SparseSegmentSum:()=>Zp,SparseToDense:()=>Yp,SplitV:()=>Ll,Sqrt:()=>Qo,Square:()=>Mc,SquaredDifference:()=>ni,Step:()=>ai,StridedSlice:()=>Bl,StringNGrams:()=>zc,StringSplit:()=>Jp,StringToHashBucketFast:()=>Qp,Sub:()=>si,Sum:()=>ei,SymbolicTensor:()=>Tr,Tan:()=>Wl,Tanh:()=>ri,Tensor:()=>nt,TensorBuffer:()=>fn,Tile:()=>va,TopK:()=>Vl,Transform:()=>Ul,Transpose:()=>Hr,Unique:()=>h0,Unpack:()=>Gl,UnsortedSegmentSum:()=>eh,UpperBound:()=>Z_,Variable:()=>kp,ZerosLike:()=>Hl,_FusedMatMul:()=>Ja,abs:()=>tn,acos:()=>jy,acosh:()=>qy,add:()=>ce,addN:()=>m0,all:()=>g0,any:()=>Cp,argMax:()=>Es,argMin:()=>Xy,asin:()=>Ky,asinh:()=>Zy,atan:()=>Yy,atan2:()=>Jy,atanh:()=>Qy,avgPool:()=>lh,avgPool3d:()=>tA,backend:()=>Ys,backend_util:()=>T,basicLSTMCell:()=>ow,batchNorm:()=>Wc,batchNorm2d:()=>nA,batchNorm3d:()=>sA,batchNorm4d:()=>rA,batchToSpaceND:()=>uh,bincount:()=>aA,booleanMaskAsync:()=>Ww,broadcastArgs:()=>iw,broadcastTo:()=>Gu,broadcast_util:()=>Xl,browser:()=>Js,buffer:()=>Le,callbacks:()=>ej,cast:()=>ye,ceil:()=>oA,clipByValue:()=>fs,clone:()=>On,complex:()=>fa,concat:()=>Ct,concat1d:()=>iA,concat2d:()=>Kl,concat3d:()=>lA,concat4d:()=>uA,constraints:()=>C8,conv1d:()=>y0,conv2d:()=>ma,conv2dTranspose:()=>A0,conv3d:()=>dA,conv3dTranspose:()=>pA,copyRegisteredKernels:()=>eD,cos:()=>ch,cosh:()=>x0,cosineWindow:()=>W0,cumprod:()=>Tp,cumsum:()=>b0,customGrad:()=>Kr,data:()=>Yk,denseBincount:()=>uw,deprecationWarn:()=>Fy,depthToSpace:()=>hA,depthwiseConv2d:()=>Vc,deregisterOp:()=>sj,device_util:()=>rh,diag:()=>cw,dilation2d:()=>fA,disableDeprecationWarnings:()=>T$,dispose:()=>ee,disposeVariables:()=>N$,div:()=>he,divNoNan:()=>mA,dot:()=>gA,dropout:()=>VA,einsum:()=>dw,elu:()=>Uc,enableDebugMode:()=>C$,enableProdMode:()=>Py,enclosingPowerOfTwo:()=>UA,engine:()=>sn,env:()=>q,equal:()=>Rs,erf:()=>yA,euclideanNorm:()=>bA,exp:()=>_s,expandDims:()=>Kt,expm1:()=>vA,eye:()=>v0,fft:()=>xh,fill:()=>Hc,findBackend:()=>My,findBackendFactory:()=>D$,floor:()=>jc,floorDiv:()=>Bc,forceHalfFloat:()=>c9,fused:()=>ec,gather:()=>qc,gatherND:()=>Hw,gather_util:()=>Ly,getBackend:()=>ss,getGradient:()=>g3,getKernel:()=>fm,getKernelsForBackend:()=>qr,getThreadsCount:()=>Eme,gpgpu_util:()=>VS,grad:()=>QF,grads:()=>eO,greater:()=>ys,greaterEqual:()=>li,ifft:()=>Qu,imag:()=>oh,image:()=>Se,inTopKAsync:()=>jw,initializers:()=>T8,input:()=>j8,io:()=>Ts,irfft:()=>M0,isFinite:()=>wA,isInf:()=>kA,isNaN:()=>IA,keep:()=>bn,kernel_impls:()=>pr,layers:()=>N8,leakyRelu:()=>dh,less:()=>w0,lessEqual:()=>ui,linalg:()=>jA,linspace:()=>gw,loadGraphModel:()=>oq,loadGraphModelSync:()=>iq,loadLayersModel:()=>cG,localResponseNormalization:()=>SA,log:()=>Ds,log1p:()=>ph,logSigmoid:()=>CA,logSoftmax:()=>I0,logSumExp:()=>S0,logicalAnd:()=>ur,logicalNot:()=>hh,logicalOr:()=>C0,logicalXor:()=>TA,losses:()=>r8,lowerBound:()=>Aw,matMul:()=>Qe,math:()=>L6,max:()=>mn,maxPool:()=>fh,maxPool3d:()=>NA,maxPoolWithArgmax:()=>xw,maximum:()=>Jr,mean:()=>Bt,memory:()=>ym,meshgrid:()=>bw,metrics:()=>kk,min:()=>ga,minimum:()=>Xc,mirrorPad:()=>EA,mod:()=>Yl,model:()=>lG,models:()=>Ik,moments:()=>mh,movingAverage:()=>Vw,mul:()=>L,multiRNNCell:()=>vw,multinomial:()=>ww,neg:()=>Dt,nextFrame:()=>qA,norm:()=>Gc,notEqual:()=>Qi,oneHot:()=>Zu,ones:()=>Ns,onesLike:()=>$s,op:()=>V,outerProduct:()=>kw,pad:()=>Qs,pad1d:()=>Iw,pad2d:()=>Sw,pad3d:()=>Cw,pad4d:()=>Tw,pool:()=>RA,pow:()=>ya,prelu:()=>yh,print:()=>_y,prod:()=>_A,profile:()=>E$,rand:()=>Nw,randomGamma:()=>Ew,randomNormal:()=>N0,randomStandardNormal:()=>Rw,randomUniform:()=>Kc,range:()=>Ju,ready:()=>Lc,real:()=>Yu,reciprocal:()=>PA,registerBackend:()=>ql,registerCallbackConstructor:()=>dG,registerGradient:()=>g6,registerKernel:()=>dr,registerOp:()=>nj,regularizers:()=>Sk,relu:()=>Pr,relu6:()=>E0,removeBackend:()=>_$,reshape:()=>U,reverse:()=>Zs,reverse1d:()=>_w,reverse2d:()=>Dw,reverse3d:()=>$w,reverse4d:()=>Pw,rfft:()=>bh,round:()=>R0,rsqrt:()=>_0,scalar:()=>Ce,scatterND:()=>Uw,scatter_util:()=>By,searchSorted:()=>T0,selu:()=>D0,separableConv2d:()=>$0,sequential:()=>uG,serialization:()=>de,setBackend:()=>Oy,setPlatform:()=>$$,setThreadsCount:()=>Nme,setWasmPath:()=>Tme,setWasmPaths:()=>nb,setWebGLContext:()=>x2,setdiff1dAsync:()=>Fw,sigmoid:()=>Cn,sign:()=>FA,signal:()=>s8,sin:()=>P0,sinh:()=>F0,slice:()=>Oe,slice1d:()=>Ah,slice2d:()=>O0,slice3d:()=>ci,slice4d:()=>no,slice_util:()=>Ut,softmax:()=>Jl,softplus:()=>Zl,spaceToBatchND:()=>gh,sparse:()=>a8,sparseToDense:()=>Gw,spectral:()=>n8,split:()=>Zt,sqrt:()=>Nn,square:()=>vt,squaredDifference:()=>z0,squeeze:()=>st,stack:()=>on,step:()=>Ql,stridedSlice:()=>OA,string:()=>o8,sub:()=>fe,sum:()=>ke,sumOutType:()=>sh,tan:()=>MA,tanh:()=>Yi,tensor:()=>ct,tensor1d:()=>Ft,tensor2d:()=>ir,tensor3d:()=>zy,tensor4d:()=>Ow,tensor5d:()=>Mw,tensor6d:()=>zw,tensor_util:()=>Nr,test_util:()=>ew,tidy:()=>Y,tile:()=>qs,time:()=>R$,topk:()=>zA,train:()=>Fi,transpose:()=>et,truncatedNormal:()=>L0,unique:()=>LA,unregisterGradient:()=>Q_,unregisterKernel:()=>J_,unsortedSegmentSum:()=>B0,unstack:()=>En,upcastType:()=>Mn,upperBound:()=>Lw,util:()=>v,valueAndGrad:()=>tO,valueAndGrads:()=>nO,variable:()=>BA,variableGrads:()=>yw,version:()=>Uh,version_converter:()=>uq,version_core:()=>Hy,version_layers:()=>m5,version_wasm:()=>Rme,version_webgl:()=>Ine,webgl:()=>Sne,webgl_util:()=>pS,webgpu:()=>yT,where:()=>zn,whereAsync:()=>WA,zeros:()=>Wt,zerosLike:()=>it});var r_=Object.create,Ay=Object.defineProperty,a_=Object.getOwnPropertyDescriptor,Q7=Object.getOwnPropertyNames,o_=Object.getPrototypeOf,i_=Object.prototype.hasOwnProperty,ln=(e,t)=>function(){return t||(0,e[Q7(e)[0]])((t={exports:{}}).exports,t),t.exports},Ve=(e,t)=>{for(var n in t)Ay(e,n,{get:t[n],enumerable:!0})},l_=(e,t,n,s)=>{if(t&&typeof t=="object"||typeof t=="function")for(let r of Q7(t))!i_.call(e,r)&&r!==n&&Ay(e,r,{get:()=>t[r],enumerable:!(s=a_(t,r))||s.enumerable});return e},uo=(e,t,n)=>(n=e!=null?r_(o_(e)):{},l_(t||!e||!e.__esModule?Ay(n,"default",{value:e,enumerable:!0}):n,e)),u_=ln({"node_modules/.pnpm/long@4.0.0/node_modules/long/src/long.js"(e,t){t.exports=s;var n=null;try{n=new WebAssembly.Instance(new WebAssembly.Module(new Uint8Array([0,97,115,109,1,0,0,0,1,13,2,96,0,1,127,96,4,127,127,127,127,1,127,3,7,6,0,1,1,1,1,1,6,6,1,127,1,65,0,11,7,50,6,3,109,117,108,0,1,5,100,105,118,95,115,0,2,5,100,105,118,95,117,0,3,5,114,101,109,95,115,0,4,5,114,101,109,95,117,0,5,8,103,101,116,95,104,105,103,104,0,0,10,191,1,6,4,0,35,0,11,36,1,1,126,32,0,173,32,1,173,66,32,134,132,32,2,173,32,3,173,66,32,134,132,126,34,4,66,32,135,167,36,0,32,4,167,11,36,1,1,126,32,0,173,32,1,173,66,32,134,132,32,2,173,32,3,173,66,32,134,132,127,34,4,66,32,135,167,36,0,32,4,167,11,36,1,1,126,32,0,173,32,1,173,66,32,134,132,32,2,173,32,3,173,66,32,134,132,128,34,4,66,32,135,167,36,0,32,4,167,11,36,1,1,126,32,0,173,32,1,173,66,32,134,132,32,2,173,32,3,173,66,32,134,132,129,34,4,66,32,135,167,36,0,32,4,167,11,36,1,1,126,32,0,173,32,1,173,66,32,134,132,32,2,173,32,3,173,66,32,134,132,130,34,4,66,32,135,167,36,0,32,4,167,11])),{}).exports}catch(P){}function s(P,C,F){this.low=P|0,this.high=C|0,this.unsigned=!!F}s.prototype.__isLong__,Object.defineProperty(s.prototype,"__isLong__",{value:!0});function r(P){return(P&&P.__isLong__)===!0}s.isLong=r;var a={},o={};function i(P,C){var F,G,K;return C?(P>>>=0,(K=0<=P&&P<256)&&(G=o[P],G)?G:(F=u(P,(P|0)<0?-1:0,!0),K&&(o[P]=F),F)):(P|=0,(K=-128<=P&&P<128)&&(G=a[P],G)?G:(F=u(P,P<0?-1:0,!1),K&&(a[P]=F),F))}s.fromInt=i;function l(P,C){if(isNaN(P))return C?b:A;if(C){if(P<0)return b;if(P>=g)return R}else{if(P<=-y)return $;if(P+1>=y)return E}return P<0?l(-P,C).neg():u(P%m|0,P/m|0,C)}s.fromNumber=l;function u(P,C,F){return new s(P,C,F)}s.fromBits=u;var c=Math.pow;function p(P,C,F){if(P.length===0)throw Error("empty string");if(P==="NaN"||P==="Infinity"||P==="+Infinity"||P==="-Infinity")return A;if(typeof C=="number"?(F=C,C=!1):C=!!C,F=F||10,F<2||36<F)throw RangeError("radix");var G;if((G=P.indexOf("-"))>0)throw Error("interior hyphen");if(G===0)return p(P.substring(1),C,F).neg();for(var K=l(c(F,8)),z=A,Z=0;Z<P.length;Z+=8){var J=Math.min(8,P.length-Z),te=parseInt(P.substring(Z,Z+J),F);if(J<8){var B=l(c(F,J));z=z.mul(B).add(l(te))}else z=z.mul(K),z=z.add(l(te))}return z.unsigned=C,z}s.fromString=p;function d(P,C){return typeof P=="number"?l(P,C):typeof P=="string"?p(P,C):u(P.low,P.high,typeof C=="boolean"?C:P.unsigned)}s.fromValue=d;var h=1<<16,f=1<<24,m=h*h,g=m*m,y=g/2,x=i(f),A=i(0);s.ZERO=A;var b=i(0,!0);s.UZERO=b;var w=i(1);s.ONE=w;var k=i(1,!0);s.UONE=k;var S=i(-1);s.NEG_ONE=S;var E=u(-1,2147483647,!1);s.MAX_VALUE=E;var R=u(-1,-1,!0);s.MAX_UNSIGNED_VALUE=R;var $=u(0,-2147483648,!1);s.MIN_VALUE=$;var _=s.prototype;_.toInt=function(){return this.unsigned?this.low>>>0:this.low},_.toNumber=function(){return this.unsigned?(this.high>>>0)*m+(this.low>>>0):this.high*m+(this.low>>>0)},_.toString=function(C){if(C=C||10,C<2||36<C)throw RangeError("radix");if(this.isZero())return"0";if(this.isNegative())if(this.eq($)){var F=l(C),G=this.div(F),K=G.mul(F).sub(this);return G.toString(C)+K.toInt().toString(C)}else return"-"+this.neg().toString(C);for(var z=l(c(C,6),this.unsigned),Z=this,J="";;){var te=Z.div(z),B=Z.sub(te.mul(z)).toInt()>>>0,oe=B.toString(C);if(Z=te,Z.isZero())return oe+J;for(;oe.length<6;)oe="0"+oe;J=""+oe+J}},_.getHighBits=function(){return this.high},_.getHighBitsUnsigned=function(){return this.high>>>0},_.getLowBits=function(){return this.low},_.getLowBitsUnsigned=function(){return this.low>>>0},_.getNumBitsAbs=function(){if(this.isNegative())return this.eq($)?64:this.neg().getNumBitsAbs();for(var C=this.high!=0?this.high:this.low,F=31;F>0&&(C&1<<F)==0;F--);return this.high!=0?F+33:F+1},_.isZero=function(){return this.high===0&&this.low===0},_.eqz=_.isZero,_.isNegative=function(){return!this.unsigned&&this.high<0},_.isPositive=function(){return this.unsigned||this.high>=0},_.isOdd=function(){return(this.low&1)===1},_.isEven=function(){return(this.low&1)===0},_.equals=function(C){return r(C)||(C=d(C)),this.unsigned!==C.unsigned&&this.high>>>31===1&&C.high>>>31===1?!1:this.high===C.high&&this.low===C.low},_.eq=_.equals,_.notEquals=function(C){return!this.eq(C)},_.neq=_.notEquals,_.ne=_.notEquals,_.lessThan=function(C){return this.comp(C)<0},_.lt=_.lessThan,_.lessThanOrEqual=function(C){return this.comp(C)<=0},_.lte=_.lessThanOrEqual,_.le=_.lessThanOrEqual,_.greaterThan=function(C){return this.comp(C)>0},_.gt=_.greaterThan,_.greaterThanOrEqual=function(C){return this.comp(C)>=0},_.gte=_.greaterThanOrEqual,_.ge=_.greaterThanOrEqual,_.compare=function(C){if(r(C)||(C=d(C)),this.eq(C))return 0;var F=this.isNegative(),G=C.isNegative();return F&&!G?-1:!F&&G?1:this.unsigned?C.high>>>0>this.high>>>0||C.high===this.high&&C.low>>>0>this.low>>>0?-1:1:this.sub(C).isNegative()?-1:1},_.comp=_.compare,_.negate=function(){return!this.unsigned&&this.eq($)?$:this.not().add(w)},_.neg=_.negate,_.add=function(C){r(C)||(C=d(C));var F=this.high>>>16,G=this.high&65535,K=this.low>>>16,z=this.low&65535,Z=C.high>>>16,J=C.high&65535,te=C.low>>>16,B=C.low&65535,oe=0,Q=0,ae=0,ie=0;return ie+=z+B,ae+=ie>>>16,ie&=65535,ae+=K+te,Q+=ae>>>16,ae&=65535,Q+=G+J,oe+=Q>>>16,Q&=65535,oe+=F+Z,oe&=65535,u(ae<<16|ie,oe<<16|Q,this.unsigned)},_.subtract=function(C){return r(C)||(C=d(C)),this.add(C.neg())},_.sub=_.subtract,_.multiply=function(C){if(this.isZero())return A;if(r(C)||(C=d(C)),n){var F=n.mul(this.low,this.high,C.low,C.high);return u(F,n.get_high(),this.unsigned)}if(C.isZero())return A;if(this.eq($))return C.isOdd()?$:A;if(C.eq($))return this.isOdd()?$:A;if(this.isNegative())return C.isNegative()?this.neg().mul(C.neg()):this.neg().mul(C).neg();if(C.isNegative())return this.mul(C.neg()).neg();if(this.lt(x)&&C.lt(x))return l(this.toNumber()*C.toNumber(),this.unsigned);var G=this.high>>>16,K=this.high&65535,z=this.low>>>16,Z=this.low&65535,J=C.high>>>16,te=C.high&65535,B=C.low>>>16,oe=C.low&65535,Q=0,ae=0,ie=0,me=0;return me+=Z*oe,ie+=me>>>16,me&=65535,ie+=z*oe,ae+=ie>>>16,ie&=65535,ie+=Z*B,ae+=ie>>>16,ie&=65535,ae+=K*oe,Q+=ae>>>16,ae&=65535,ae+=z*B,Q+=ae>>>16,ae&=65535,ae+=Z*te,Q+=ae>>>16,ae&=65535,Q+=G*oe+K*B+z*te+Z*J,Q&=65535,u(ie<<16|me,Q<<16|ae,this.unsigned)},_.mul=_.multiply,_.divide=function(C){if(r(C)||(C=d(C)),C.isZero())throw Error("division by zero");if(n){if(!this.unsigned&&this.high===-2147483648&&C.low===-1&&C.high===-1)return this;var F=(this.unsigned?n.div_u:n.div_s)(this.low,this.high,C.low,C.high);return u(F,n.get_high(),this.unsigned)}if(this.isZero())return this.unsigned?b:A;var G,K,z;if(this.unsigned){if(C.unsigned||(C=C.toUnsigned()),C.gt(this))return b;if(C.gt(this.shru(1)))return k;z=b}else{if(this.eq($)){if(C.eq(w)||C.eq(S))return $;if(C.eq($))return w;var Z=this.shr(1);return G=Z.div(C).shl(1),G.eq(A)?C.isNegative()?w:S:(K=this.sub(C.mul(G)),z=G.add(K.div(C)),z)}else if(C.eq($))return this.unsigned?b:A;if(this.isNegative())return C.isNegative()?this.neg().div(C.neg()):this.neg().div(C).neg();if(C.isNegative())return this.div(C.neg()).neg();z=A}for(K=this;K.gte(C);){G=Math.max(1,Math.floor(K.toNumber()/C.toNumber()));for(var J=Math.ceil(Math.log(G)/Math.LN2),te=J<=48?1:c(2,J-48),B=l(G),oe=B.mul(C);oe.isNegative()||oe.gt(K);)G-=te,B=l(G,this.unsigned),oe=B.mul(C);B.isZero()&&(B=w),z=z.add(B),K=K.sub(oe)}return z},_.div=_.divide,_.modulo=function(C){if(r(C)||(C=d(C)),n){var F=(this.unsigned?n.rem_u:n.rem_s)(this.low,this.high,C.low,C.high);return u(F,n.get_high(),this.unsigned)}return this.sub(this.div(C).mul(C))},_.mod=_.modulo,_.rem=_.modulo,_.not=function(){return u(~this.low,~this.high,this.unsigned)},_.and=function(C){return r(C)||(C=d(C)),u(this.low&C.low,this.high&C.high,this.unsigned)},_.or=function(C){return r(C)||(C=d(C)),u(this.low|C.low,this.high|C.high,this.unsigned)},_.xor=function(C){return r(C)||(C=d(C)),u(this.low^C.low,this.high^C.high,this.unsigned)},_.shiftLeft=function(C){return r(C)&&(C=C.toInt()),(C&=63)===0?this:C<32?u(this.low<<C,this.high<<C|this.low>>>32-C,this.unsigned):u(0,this.low<<C-32,this.unsigned)},_.shl=_.shiftLeft,_.shiftRight=function(C){return r(C)&&(C=C.toInt()),(C&=63)===0?this:C<32?u(this.low>>>C|this.high<<32-C,this.high>>C,this.unsigned):u(this.high>>C-32,this.high>=0?0:-1,this.unsigned)},_.shr=_.shiftRight,_.shiftRightUnsigned=function(C){if(r(C)&&(C=C.toInt()),C&=63,C===0)return this;var F=this.high;if(C<32){var G=this.low;return u(G>>>C|F<<32-C,F>>>C,this.unsigned)}else return C===32?u(F,0,this.unsigned):u(F>>>C-32,0,this.unsigned)},_.shru=_.shiftRightUnsigned,_.shr_u=_.shiftRightUnsigned,_.toSigned=function(){return this.unsigned?u(this.low,this.high,!1):this},_.toUnsigned=function(){return this.unsigned?this:u(this.low,this.high,!0)},_.toBytes=function(C){return C?this.toBytesLE():this.toBytesBE()},_.toBytesLE=function(){var C=this.high,F=this.low;return[F&255,F>>>8&255,F>>>16&255,F>>>24,C&255,C>>>8&255,C>>>16&255,C>>>24]},_.toBytesBE=function(){var C=this.high,F=this.low;return[C>>>24,C>>>16&255,C>>>8&255,C&255,F>>>24,F>>>16&255,F>>>8&255,F&255]},s.fromBytes=function(C,F,G){return G?s.fromBytesLE(C,F):s.fromBytesBE(C,F)},s.fromBytesLE=function(C,F){return new s(C[0]|C[1]<<8|C[2]<<16|C[3]<<24,C[4]|C[5]<<8|C[6]<<16|C[7]<<24,F)},s.fromBytesBE=function(C,F){return new s(C[4]<<24|C[5]<<16|C[6]<<8|C[7],C[0]<<24|C[1]<<16|C[2]<<8|C[3],F)}}}),c_=ln({"(disabled):node_modules/.pnpm/node-fetch@2.6.7/node_modules/node-fetch/browser.js"(){}}),d_=ln({"(disabled):util"(){}}),p_=ln({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/alea.js"(e,t){(function(n,s,r){function a(u){var c=this,p=l();c.next=function(){var d=2091639*c.s0+c.c*23283064365386963e-26;return c.s0=c.s1,c.s1=c.s2,c.s2=d-(c.c=d|0)},c.c=1,c.s0=p(" "),c.s1=p(" "),c.s2=p(" "),c.s0-=p(u),c.s0<0&&(c.s0+=1),c.s1-=p(u),c.s1<0&&(c.s1+=1),c.s2-=p(u),c.s2<0&&(c.s2+=1),p=null}function o(u,c){return c.c=u.c,c.s0=u.s0,c.s1=u.s1,c.s2=u.s2,c}function i(u,c){var p=new a(u),d=c&&c.state,h=p.next;return h.int32=function(){return p.next()*4294967296|0},h.double=function(){return h()+(h()*2097152|0)*11102230246251565e-32},h.quick=h,d&&(typeof d=="object"&&o(d,p),h.state=function(){return o(p,{})}),h}function l(){var u=4022871197,c=function(p){p=String(p);for(var d=0;d<p.length;d++){u+=p.charCodeAt(d);var h=.02519603282416938*u;u=h>>>0,h-=u,h*=u,u=h>>>0,h-=u,u+=h*4294967296}return(u>>>0)*23283064365386963e-26};return c}s&&s.exports?s.exports=i:r&&r.amd?r(function(){return i}):this.alea=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),h_=ln({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/xor128.js"(e,t){(function(n,s,r){function a(l){var u=this,c="";u.x=0,u.y=0,u.z=0,u.w=0,u.next=function(){var d=u.x^u.x<<11;return u.x=u.y,u.y=u.z,u.z=u.w,u.w^=u.w>>>19^d^d>>>8},l===(l|0)?u.x=l:c+=l;for(var p=0;p<c.length+64;p++)u.x^=c.charCodeAt(p)|0,u.next()}function o(l,u){return u.x=l.x,u.y=l.y,u.z=l.z,u.w=l.w,u}function i(l,u){var c=new a(l),p=u&&u.state,d=function(){return(c.next()>>>0)/4294967296};return d.double=function(){do var h=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(h+f)/(1<<21);while(m===0);return m},d.int32=c.next,d.quick=d,p&&(typeof p=="object"&&o(p,c),d.state=function(){return o(c,{})}),d}s&&s.exports?s.exports=i:r&&r.amd?r(function(){return i}):this.xor128=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),f_=ln({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/xorwow.js"(e,t){(function(n,s,r){function a(l){var u=this,c="";u.next=function(){var d=u.x^u.x>>>2;return u.x=u.y,u.y=u.z,u.z=u.w,u.w=u.v,(u.d=u.d+362437|0)+(u.v=u.v^u.v<<4^(d^d<<1))|0},u.x=0,u.y=0,u.z=0,u.w=0,u.v=0,l===(l|0)?u.x=l:c+=l;for(var p=0;p<c.length+64;p++)u.x^=c.charCodeAt(p)|0,p==c.length&&(u.d=u.x<<10^u.x>>>4),u.next()}function o(l,u){return u.x=l.x,u.y=l.y,u.z=l.z,u.w=l.w,u.v=l.v,u.d=l.d,u}function i(l,u){var c=new a(l),p=u&&u.state,d=function(){return(c.next()>>>0)/4294967296};return d.double=function(){do var h=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(h+f)/(1<<21);while(m===0);return m},d.int32=c.next,d.quick=d,p&&(typeof p=="object"&&o(p,c),d.state=function(){return o(c,{})}),d}s&&s.exports?s.exports=i:r&&r.amd?r(function(){return i}):this.xorwow=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),m_=ln({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/xorshift7.js"(e,t){(function(n,s,r){function a(l){var u=this;u.next=function(){var p=u.x,d=u.i,h,f,m;return h=p[d],h^=h>>>7,f=h^h<<24,h=p[d+1&7],f^=h^h>>>10,h=p[d+3&7],f^=h^h>>>3,h=p[d+4&7],f^=h^h<<7,h=p[d+7&7],h=h^h<<13,f^=h^h<<9,p[d]=f,u.i=d+1&7,f};function c(p,d){var h,f,m=[];if(d===(d|0))f=m[0]=d;else for(d=""+d,h=0;h<d.length;++h)m[h&7]=m[h&7]<<15^d.charCodeAt(h)+m[h+1&7]<<13;for(;m.length<8;)m.push(0);for(h=0;h<8&&m[h]===0;++h);for(h==8?f=m[7]=-1:f=m[h],p.x=m,p.i=0,h=256;h>0;--h)p.next()}c(u,l)}function o(l,u){return u.x=l.x.slice(),u.i=l.i,u}function i(l,u){l==null&&(l=+new Date);var c=new a(l),p=u&&u.state,d=function(){return(c.next()>>>0)/4294967296};return d.double=function(){do var h=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(h+f)/(1<<21);while(m===0);return m},d.int32=c.next,d.quick=d,p&&(p.x&&o(p,c),d.state=function(){return o(c,{})}),d}s&&s.exports?s.exports=i:r&&r.amd?r(function(){return i}):this.xorshift7=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),g_=ln({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/xor4096.js"(e,t){(function(n,s,r){function a(l){var u=this;u.next=function(){var p=u.w,d=u.X,h=u.i,f,m;return u.w=p=p+1640531527|0,m=d[h+34&127],f=d[h=h+1&127],m^=m<<13,f^=f<<17,m^=m>>>15,f^=f>>>12,m=d[h]=m^f,u.i=h,m+(p^p>>>16)|0};function c(p,d){var h,f,m,g,y,x=[],A=128;for(d===(d|0)?(f=d,d=null):(d=d+"\0",f=0,A=Math.max(A,d.length)),m=0,g=-32;g<A;++g)d&&(f^=d.charCodeAt((g+32)%d.length)),g===0&&(y=f),f^=f<<10,f^=f>>>15,f^=f<<4,f^=f>>>13,g>=0&&(y=y+1640531527|0,h=x[g&127]^=f+y,m=h==0?m+1:0);for(m>=128&&(x[(d&&d.length||0)&127]=-1),m=127,g=4*128;g>0;--g)f=x[m+34&127],h=x[m=m+1&127],f^=f<<13,h^=h<<17,f^=f>>>15,h^=h>>>12,x[m]=f^h;p.w=y,p.X=x,p.i=m}c(u,l)}function o(l,u){return u.i=l.i,u.w=l.w,u.X=l.X.slice(),u}function i(l,u){l==null&&(l=+new Date);var c=new a(l),p=u&&u.state,d=function(){return(c.next()>>>0)/4294967296};return d.double=function(){do var h=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(h+f)/(1<<21);while(m===0);return m},d.int32=c.next,d.quick=d,p&&(p.X&&o(p,c),d.state=function(){return o(c,{})}),d}s&&s.exports?s.exports=i:r&&r.amd?r(function(){return i}):this.xor4096=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),y_=ln({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/tychei.js"(e,t){(function(n,s,r){function a(l){var u=this,c="";u.next=function(){var d=u.b,h=u.c,f=u.d,m=u.a;return d=d<<25^d>>>7^h,h=h-f|0,f=f<<24^f>>>8^m,m=m-d|0,u.b=d=d<<20^d>>>12^h,u.c=h=h-f|0,u.d=f<<16^h>>>16^m,u.a=m-d|0},u.a=0,u.b=0,u.c=-1640531527,u.d=1367130551,l===Math.floor(l)?(u.a=l/4294967296|0,u.b=l|0):c+=l;for(var p=0;p<c.length+20;p++)u.b^=c.charCodeAt(p)|0,u.next()}function o(l,u){return u.a=l.a,u.b=l.b,u.c=l.c,u.d=l.d,u}function i(l,u){var c=new a(l),p=u&&u.state,d=function(){return(c.next()>>>0)/4294967296};return d.double=function(){do var h=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(h+f)/(1<<21);while(m===0);return m},d.int32=c.next,d.quick=d,p&&(typeof p=="object"&&o(p,c),d.state=function(){return o(c,{})}),d}s&&s.exports?s.exports=i:r&&r.amd?r(function(){return i}):this.tychei=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),A_=ln({"(disabled):crypto"(){}}),x_=ln({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/seedrandom.js"(e,t){(function(n,s,r){var a=256,o=6,i=52,l="random",u=r.pow(a,o),c=r.pow(2,i),p=c*2,d=a-1,h;function f(w,k,S){var E=[];k=k==!0?{entropy:!0}:k||{};var R=x(y(k.entropy?[w,b(s)]:w==null?A():w,3),E),$=new m(E),_=function(){for(var P=$.g(o),C=u,F=0;P<c;)P=(P+F)*a,C*=a,F=$.g(1);for(;P>=p;)P/=2,C/=2,F>>>=1;return(P+F)/C};return _.int32=function(){return $.g(4)|0},_.quick=function(){return $.g(4)/4294967296},_.double=_,x(b($.S),s),(k.pass||S||function(P,C,F,G){return G&&(G.S&&g(G,$),P.state=function(){return g($,{})}),F?(r[l]=P,C):P})(_,R,"global"in k?k.global:this==r,k.state)}function m(w){var k,S=w.length,E=this,R=0,$=E.i=E.j=0,_=E.S=[];for(S||(w=[S++]);R<a;)_[R]=R++;for(R=0;R<a;R++)_[R]=_[$=d&$+w[R%S]+(k=_[R])],_[$]=k;(E.g=function(P){for(var C,F=0,G=E.i,K=E.j,z=E.S;P--;)C=z[G=d&G+1],F=F*a+z[d&(z[G]=z[K=d&K+C])+(z[K]=C)];return E.i=G,E.j=K,F})(a)}function g(w,k){return k.i=w.i,k.j=w.j,k.S=w.S.slice(),k}function y(w,k){var S=[],E=typeof w,R;if(k&&E=="object")for(R in w)try{S.push(y(w[R],k-1))}catch($){}return S.length?S:E=="string"?w:w+"\0"}function x(w,k){for(var S=w+"",E,R=0;R<S.length;)k[d&R]=d&(E^=k[d&R]*19)+S.charCodeAt(R++);return b(k)}function A(){try{var w;return h&&(w=h.randomBytes)?w=w(a):(w=new Uint8Array(a),(n.crypto||n.msCrypto).getRandomValues(w)),b(w)}catch(E){var k=n.navigator,S=k&&k.plugins;return[+new Date,n,S,n.screen,b(s)]}}function b(w){return String.fromCharCode.apply(0,w)}if(x(r.random(),s),typeof t=="object"&&t.exports){t.exports=f;try{h=A_()}catch(w){}}else typeof define=="function"&&define.amd?define(function(){return f}):r["seed"+l]=f})(typeof self!="undefined"?self:e,[],Math)}}),Wm=ln({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/index.js"(e,t){var n=p_(),s=h_(),r=f_(),a=m_(),o=g_(),i=y_(),l=x_();l.alea=n,l.xor128=s,l.xorwow=r,l.xorshift7=a,l.xor4096=o,l.tychei=i,t.exports=l}}),e6=ln({"(disabled):node_modules/.pnpm/string_decoder@1.3.0/node_modules/string_decoder/lib/string_decoder.js"(){}}),xy=ln({"(disabled):fs"(){}}),um=ln({"(disabled):path"(){}}),b_=ln({"(disabled):worker_threads"(){}}),v_=ln({"(disabled):perf_hooks"(){}}),w_=ln({"(disabled):os"(){}}),k_=ln({"node_modules/.pnpm/@tensorflow+tfjs-backend-wasm@3.19.0_hek32lflchivueqv5i4vgonghu/node_modules/@tensorflow/tfjs-backend-wasm/wasm-out/tfjs-backend-wasm-threaded-simd.js"(e,t){var n=(()=>{var s=typeof document!="undefined"&&document.currentScript?document.currentScript.src:void 0;return typeof __filename!="undefined"&&(s=s||__filename),function(r){r=r||{};function a(){return $e.buffer!=qn&&Ar($e.buffer),sf}function o(){return $e.buffer!=qn&&Ar($e.buffer),rf}function i(){return $e.buffer!=qn&&Ar($e.buffer),zd}function l(){return $e.buffer!=qn&&Ar($e.buffer),af}function u(){return $e.buffer!=qn&&Ar($e.buffer),of}function c(){return $e.buffer!=qn&&Ar($e.buffer),lf}function p(){return $e.buffer!=qn&&Ar($e.buffer),uf}var d=typeof r!="undefined"?r:{},h,f;d.ready=new Promise(function(N,O){h=N,f=O});var m;typeof process!="undefined"&&process.listeners&&(m={uncaughtException:process.listeners("uncaughtException"),unhandledRejection:process.listeners("unhandledRejection")});var g=Object.assign({},d),y=[],x="./this.program",A=(N,O)=>{throw O},b=typeof window=="object",w=typeof importScripts=="function",k=typeof process=="object"&&typeof process.versions=="object"&&typeof process.versions.node=="string",S=d.ENVIRONMENT_IS_PTHREAD||!1,E="";function R(N){return d.locateFile?d.locateFile(N,E):E+N}var $,_,P,C;function F(N){if(N instanceof Xd)return;B("exiting due to exception: "+N)}var G,K,z;if(k){w?E=um().dirname(E)+"/":E=__dirname+"/",z=()=>{K||(G=xy(),K=um())},$=function(H,ne){return z(),H=K.normalize(H),G.readFileSync(H,ne?void 0:"utf8")},P=O=>{var H=$(O,!0);return H.buffer||(H=new Uint8Array(H)),H},_=(O,H,ne)=>{z(),O=K.normalize(O),G.readFile(O,function(Ae,ve){Ae?ne(Ae):H(ve.buffer)})},process.argv.length>1&&(x=process.argv[1].replace(/\\/g,"/")),y=process.argv.slice(2),process.on("uncaughtException",function(O){if(!(O instanceof Xd))throw O}),process.on("unhandledRejection",function(O){throw O}),A=(O,H)=>{if(Ti())throw process.exitCode=O,H;F(H),process.exit(O)},d.inspect=function(){return"[Emscripten Module object]"};let N;try{N=b_()}catch(O){throw console.error('The "worker_threads" module is not supported in this node.js build - perhaps a newer version is needed?'),O}global.Worker=N.Worker}else(b||w)&&(w?E=self.location.href:typeof document!="undefined"&&document.currentScript&&(E=document.currentScript.src),typeof s!="undefined"&&s&&(E=s),E.indexOf("blob:")!==0?E=E.substr(0,E.replace(/[?#].*/,"").lastIndexOf("/")+1):E="",k||($=N=>{var O=new XMLHttpRequest;return O.open("GET",N,!1),O.send(null),O.responseText},w&&(P=N=>{var O=new XMLHttpRequest;return O.open("GET",N,!1),O.responseType="arraybuffer",O.send(null),new Uint8Array(O.response)}),_=(N,O,H)=>{var ne=new XMLHttpRequest;ne.open("GET",N,!0),ne.responseType="arraybuffer",ne.onload=()=>{if(ne.status==200||ne.status==0&&ne.response){O(ne.response);return}H()},ne.onerror=H,ne.send(null)}),C=N=>document.title=N);k&&typeof performance=="undefined"&&(global.performance=v_().performance);var Z=console.log.bind(console),J=console.warn.bind(console);k&&(z(),Z=N=>G.writeSync(1,N+`
|
|
`),J=N=>G.writeSync(2,N+`
|
|
`));var te=d.print||Z,B=d.printErr||J;Object.assign(d,g),g=null,d.arguments&&(y=d.arguments),d.thisProgram&&(x=d.thisProgram),d.quit&&(A=d.quit);var oe=4;function Q(N){Q.shown||(Q.shown={}),Q.shown[N]||(Q.shown[N]=1,B(N))}function ae(N,O){if(typeof WebAssembly.Function=="function"){for(var H={i:"i32",j:"i64",f:"f32",d:"f64"},ne={parameters:[],results:O[0]=="v"?[]:[H[O[0]]]},Ae=1;Ae<O.length;++Ae)ne.parameters.push(H[O[Ae]]);return new WebAssembly.Function(ne,N)}var ve=[1,0,1,96],Ne=O.slice(0,1),Me=O.slice(1),Lt={i:127,j:126,f:125,d:124};ve.push(Me.length);for(var Ae=0;Ae<Me.length;++Ae)ve.push(Lt[Me[Ae]]);Ne=="v"?ve.push(0):ve=ve.concat([1,Lt[Ne]]),ve[1]=ve.length-2;var wr=new Uint8Array([0,97,115,109,1,0,0,0].concat(ve,[2,7,1,1,101,1,102,0,0,7,5,1,1,102,0,0])),kr=new WebAssembly.Module(wr),Mf=new WebAssembly.Instance(kr,{e:{f:N}}),Kd=Mf.exports.f;return Kd}var ie=[],me;function we(){if(ie.length)return ie.pop();try{Gs.grow(1)}catch(N){throw N instanceof RangeError?"Unable to grow wasm table. Set ALLOW_TABLE_GROWTH.":N}return Gs.length-1}function Re(N,O){for(var H=N;H<N+O;H++){var ne=Su(H);ne&&me.set(ne,H)}}var _e=0,Be=N=>{_e=N},He=Atomics.load,ot=Atomics.store,pt=Atomics.compareExchange,ht;d.wasmBinary&&(ht=d.wasmBinary);var At=d.noExitRuntime||!0;typeof WebAssembly!="object"&&wu("no native wasm support detected");var $e,Tt,It=!1,Gn;function Qt(N,O){N||wu(O)}function bs(N){var O=d["_"+N];return O}function pn(N,O,H,ne,Ae){var ve={string:function(ks){var Du=0;if(ks!=null&&ks!==0){var cv=(ks.length<<2)+1;Du=_u(cv),aa(ks,Du,cv)}return Du},array:function(ks){var Du=_u(ks.length);return oa(ks,Du),Du}};function Ne(ks){return O==="string"?jn(ks):O==="boolean"?Boolean(ks):ks}var Me=bs(N),Lt=[],wr=0;if(ne)for(var kr=0;kr<ne.length;kr++){var Mf=ve[H[kr]];Mf?(wr===0&&(wr=Qg()),Lt[kr]=Mf(ne[kr])):Lt[kr]=ne[kr]}var Kd=Me.apply(null,Lt);function YR(ks){return wr!==0&&$f(wr),Ne(ks)}return Kd=YR(Kd),Kd}function Hn(N,O,H,ne){H=H||[];var Ae=H.every(function(Ne){return Ne==="number"}),ve=O!=="string";return ve&&Ae&&!ne?bs(N):function(){return pn(N,O,H,arguments,ne)}}var vs=1;function ws(N){var O=new TextDecoder(N);this.decode=H=>(H.buffer instanceof SharedArrayBuffer&&(H=new Uint8Array(H)),O.decode.call(O,H))}var Pn=typeof TextDecoder!="undefined"?new ws("utf8"):void 0;function Us(N,O,H){for(var ne=O+H,Ae=O;N[Ae]&&!(Ae>=ne);)++Ae;if(Ae-O>16&&N.subarray&&Pn)return Pn.decode(N.subarray(O,Ae));for(var ve="";O<Ae;){var Ne=N[O++];if(!(Ne&128)){ve+=String.fromCharCode(Ne);continue}var Me=N[O++]&63;if((Ne&224)==192){ve+=String.fromCharCode((Ne&31)<<6|Me);continue}var Lt=N[O++]&63;if((Ne&240)==224?Ne=(Ne&15)<<12|Me<<6|Lt:Ne=(Ne&7)<<18|Me<<12|Lt<<6|N[O++]&63,Ne<65536)ve+=String.fromCharCode(Ne);else{var wr=Ne-65536;ve+=String.fromCharCode(55296|wr>>10,56320|wr&1023)}}return ve}function jn(N,O){return N?Us(o(),N,O):""}function ra(N,O,H,ne){if(!(ne>0))return 0;for(var Ae=H,ve=H+ne-1,Ne=0;Ne<N.length;++Ne){var Me=N.charCodeAt(Ne);if(Me>=55296&&Me<=57343){var Lt=N.charCodeAt(++Ne);Me=65536+((Me&1023)<<10)|Lt&1023}if(Me<=127){if(H>=ve)break;O[H++]=Me}else if(Me<=2047){if(H+1>=ve)break;O[H++]=192|Me>>6,O[H++]=128|Me&63}else if(Me<=65535){if(H+2>=ve)break;O[H++]=224|Me>>12,O[H++]=128|Me>>6&63,O[H++]=128|Me&63}else{if(H+3>=ve)break;O[H++]=240|Me>>18,O[H++]=128|Me>>12&63,O[H++]=128|Me>>6&63,O[H++]=128|Me&63}}return O[H]=0,H-Ae}function aa(N,O,H){return ra(N,o(),O,H)}function xu(N){for(var O=0,H=0;H<N.length;++H){var ne=N.charCodeAt(H);ne>=55296&&ne<=57343&&(ne=65536+((ne&1023)<<10)|N.charCodeAt(++H)&1023),ne<=127?++O:ne<=2047?O+=2:ne<=65535?O+=3:O+=4}return O}var Oa=typeof TextDecoder!="undefined"?new ws("utf-16le"):void 0;function oa(N,O){a().set(N,O)}function Md(N,O,H){for(var ne=0;ne<N.length;++ne)a()[O++>>0]=N.charCodeAt(ne);H||(a()[O>>0]=0)}function bu(N,O){return N%O>0&&(N+=O-N%O),N}var qn,sf,rf,zd,af,of,G4,lf,uf;S&&(qn=d.buffer);function Ar(N){qn=N,d.HEAP8=sf=new Int8Array(N),d.HEAP16=zd=new Int16Array(N),d.HEAP32=of=new Int32Array(N),d.HEAPU8=rf=new Uint8Array(N),d.HEAPU16=af=new Uint16Array(N),d.HEAPU32=G4=new Uint32Array(N),d.HEAPF32=lf=new Float32Array(N),d.HEAPF64=uf=new Float64Array(N)}var cf=d.INITIAL_MEMORY||16777216;if(S)$e=d.wasmMemory,qn=d.buffer;else if(d.wasmMemory)$e=d.wasmMemory;else if($e=new WebAssembly.Memory({initial:cf/65536,maximum:32768,shared:!0}),!($e.buffer instanceof SharedArrayBuffer))throw B("requested a shared WebAssembly.Memory but the returned buffer is not a SharedArrayBuffer, indicating that while the browser has SharedArrayBuffer it does not have WebAssembly threads support - you may need to set a flag"),k&&console.log("(on node you may need: --experimental-wasm-threads --experimental-wasm-bulk-memory and also use a recent version)"),Error("bad memory");$e&&(qn=$e.buffer),cf=qn.byteLength,Ar(qn);var Gs,vu=[],Ma=[],b1=[],df=[],Ci=!1,v1=!1,pf=0;function Ti(){return At||pf>0}function Xn(){if(d.preRun)for(typeof d.preRun=="function"&&(d.preRun=[d.preRun]);d.preRun.length;)H4(d.preRun.shift());gf(vu)}function Ld(){Ci=!0,!S&&gf(Ma)}function w1(){S||(ze.terminateAllThreads(),v1=!0)}function k1(){if(!S){if(d.postRun)for(typeof d.postRun=="function"&&(d.postRun=[d.postRun]);d.postRun.length;)Bd(d.postRun.shift());gf(df)}}function H4(N){vu.unshift(N)}function j4(N){Ma.unshift(N)}function Bd(N){df.unshift(N)}var za=0,hf=null,xr=null;function Wd(N){za++,d.monitorRunDependencies&&d.monitorRunDependencies(za)}function q4(N){if(za--,d.monitorRunDependencies&&d.monitorRunDependencies(za),za==0&&(hf!==null&&(clearInterval(hf),hf=null),xr)){var O=xr;xr=null,O()}}d.preloadedImages={},d.preloadedAudios={};function wu(N){S?postMessage({cmd:"onAbort",arg:N}):d.onAbort&&d.onAbort(N),N="Aborted("+N+")",B(N),It=!0,Gn=1,N+=". Build with -s ASSERTIONS=1 for more info.";var O=new WebAssembly.RuntimeError(N);throw f(O),O}var I1="data:application/octet-stream;base64,";function Vd(N){return N.startsWith(I1)}function ff(N){return N.startsWith("file://")}var Kn;Kn="tfjs-backend-wasm-threaded-simd.wasm",Vd(Kn)||(Kn=R(Kn));function mf(N){try{if(N==Kn&&ht)return new Uint8Array(ht);if(P)return P(N);throw"both async and sync fetching of the wasm failed"}catch(O){wu(O)}}function ku(){if(!ht&&(b||w)){if(typeof fetch=="function"&&!ff(Kn))return fetch(Kn,{credentials:"same-origin"}).then(function(N){if(!N.ok)throw"failed to load wasm binary file at '"+Kn+"'";return N.arrayBuffer()}).catch(function(){return mf(Kn)});if(_)return new Promise(function(N,O){_(Kn,function(H){N(new Uint8Array(H))},O)})}return Promise.resolve().then(function(){return mf(Kn)})}function S1(){var N={env:Nf,wasi_snapshot_preview1:Nf};function O(Ne,Me){var Lt=Ne.exports;if(d.asm=Lt,D1(d.asm.emscripten_tls_init),Gs=d.asm.__indirect_function_table,j4(d.asm.__wasm_call_ctors),Tt=Me,!S){var wr=ze.unusedWorkers.length;ze.unusedWorkers.forEach(function(kr){ze.loadWasmModuleToWorker(kr,function(){--wr||q4("wasm-instantiate")})})}}S||Wd("wasm-instantiate");function H(Ne){O(Ne.instance,Ne.module)}function ne(Ne){return ku().then(function(Me){return WebAssembly.instantiate(Me,N)}).then(function(Me){return Me}).then(Ne,function(Me){B("failed to asynchronously prepare wasm: "+Me),wu(Me)})}function Ae(){return!ht&&typeof WebAssembly.instantiateStreaming=="function"&&!Vd(Kn)&&!ff(Kn)&&typeof fetch=="function"?fetch(Kn,{credentials:"same-origin"}).then(function(Ne){var Me=WebAssembly.instantiateStreaming(Ne,N);return Me.then(H,function(Lt){return B("wasm streaming compile failed: "+Lt),B("falling back to ArrayBuffer instantiation"),ne(H)})}):ne(H)}if(d.instantiateWasm)try{var ve=d.instantiateWasm(N,O);return ve}catch(Ne){return B("Module.instantiateWasm callback failed with error: "+Ne),!1}return Ae().catch(f),{}}var X4,K4,C1={};function gf(N){for(;N.length>0;){var O=N.shift();if(typeof O=="function"){O(d);continue}var H=O.func;typeof H=="number"?O.arg===void 0?Su(H)():Su(H)(O.arg):H(O.arg===void 0?null:O.arg)}}function Iu(N){var O=Qg(),H=N();return $f(O),H}function aR(N){return N}function Z4(N){var O=/\b_Z[\w\d_]+/g;return N.replace(O,function(H){var ne=H;return H===ne?H:ne+" ["+H+"]"})}function T1(N){u()[N>>2]=0;var O=ze.pthreads[N];delete ze.pthreads[N],O.worker.terminate(),Jg(N),ze.runningWorkers.splice(ze.runningWorkers.indexOf(O.worker),1),O.worker.pthread=void 0}function N1(N){var O=ze.pthreads[N];O.worker.postMessage({cmd:"cancel"})}function yf(N){var O=ze.pthreads[N];if(O){u()[N>>2]=0;var H=O.worker;ze.returnWorkerToPool(H)}}function Af(N){XR(N)}function E1(N){if(N instanceof Xd||N=="unwind")return Gn;A(1,N)}var ze={unusedWorkers:[],runningWorkers:[],tlsInitFunctions:[],init:function(){S?ze.initWorker():ze.initMainThread()},initMainThread:function(){for(var N=8,O=0;O<N;++O)ze.allocateUnusedWorker()},initWorker:function(){At=!1},pthreads:{},setExitStatus:function(N){Gn=N},terminateAllThreads:function(){for(var N in ze.pthreads){var O=ze.pthreads[N];O&&O.worker&&ze.returnWorkerToPool(O.worker)}for(var H=0;H<ze.unusedWorkers.length;++H){var ne=ze.unusedWorkers[H];ne.terminate()}ze.unusedWorkers=[]},returnWorkerToPool:function(N){ze.runWithoutMainThreadQueuedCalls(function(){delete ze.pthreads[N.pthread.threadInfoStruct],ze.unusedWorkers.push(N),ze.runningWorkers.splice(ze.runningWorkers.indexOf(N),1),Jg(N.pthread.threadInfoStruct),N.pthread=void 0})},runWithoutMainThreadQueuedCalls:function(N){u()[uv>>2]=0;try{N()}finally{u()[uv>>2]=1}},receiveObjectTransfer:function(N){},threadInit:function(){for(var N in ze.tlsInitFunctions)ze.tlsInitFunctions[N]()},loadWasmModuleToWorker:function(N,O){N.onmessage=H=>{var ne=H.data,Ae=ne.cmd;if(N.pthread&&(ze.currentProxiedOperationCallerThread=N.pthread.threadInfoStruct),ne.targetThread&&ne.targetThread!=Df()){var ve=ze.pthreads[ne.targetThread];ve?ve.worker.postMessage(ne,ne.transferList):B('Internal error! Worker sent a message "'+Ae+'" to target pthread '+ne.targetThread+", but that thread no longer exists!"),ze.currentProxiedOperationCallerThread=void 0;return}Ae==="processQueuedMainThreadWork"?rv():Ae==="spawnThread"?bf(ne):Ae==="cleanupThread"?yf(ne.thread):Ae==="killThread"?T1(ne.thread):Ae==="cancelThread"?N1(ne.thread):Ae==="loaded"?(N.loaded=!0,O&&O(N),N.runPthread&&(N.runPthread(),delete N.runPthread)):Ae==="print"?te("Thread "+ne.threadId+": "+ne.text):Ae==="printErr"?B("Thread "+ne.threadId+": "+ne.text):Ae==="alert"?alert("Thread "+ne.threadId+": "+ne.text):ne.target==="setimmediate"?N.postMessage(ne):Ae==="onAbort"?d.onAbort&&d.onAbort(ne.arg):B("worker sent an unknown command "+Ae),ze.currentProxiedOperationCallerThread=void 0},N.onerror=H=>{var ne="worker sent an error!";throw B(ne+" "+H.filename+":"+H.lineno+": "+H.message),H},k&&(N.on("message",function(H){N.onmessage({data:H})}),N.on("error",function(H){N.onerror(H)}),N.on("detachedExit",function(){})),N.postMessage({cmd:"load",urlOrBlob:d.mainScriptUrlOrBlob||s,wasmMemory:$e,wasmModule:Tt})},allocateUnusedWorker:function(){var N=R("tfjs-backend-wasm-threaded-simd.worker.js");ze.unusedWorkers.push(new Worker(N))},getNewWorker:function(){return ze.unusedWorkers.length==0&&(ze.allocateUnusedWorker(),ze.loadWasmModuleToWorker(ze.unusedWorkers[0])),ze.unusedWorkers.pop()}};function R1(){var N=Df(),O=u()[N+44>>2],H=u()[N+48>>2],ne=O-H;lv(O,ne),$f(O)}d.establishStackSpace=R1;function xf(N){if(S)return Ri(1,0,N);try{Af(N)}catch(O){E1(O)}}var Ni=[];function Su(N){var O=Ni[N];return O||(N>=Ni.length&&(Ni.length=N+1),Ni[N]=O=Gs.get(N)),O}function _1(N,O){return Su(N)(O)}d.invokeEntryPoint=_1;function Y4(){var N=new Error;if(!N.stack){try{throw new Error}catch(O){N=O}if(!N.stack)return"(no stack trace available)"}return N.stack.toString()}function D1(N,O,H){ze.tlsInitFunctions.push(N)}function J4(N,O){Gs.set(N,O),Ni[N]=O}var Ei;k?Ei=()=>{var N=process.hrtime();return N[0]*1e3+N[1]/1e6}:S?Ei=()=>performance.now()-d.__performance_now_clock_drift:Ei=()=>performance.now();var $1=!0;function P1(N){return u()[sv()>>2]=N,N}function F1(N,O){var H;if(N===0)H=Date.now();else if((N===1||N===4)&&$1)H=Ei();else return P1(28),-1;return u()[O>>2]=H/1e3|0,u()[O+4>>2]=H%1e3*1e3*1e3|0,0}function O1(N,O){return F1(N,O)}function M1(N){av(N,!w,1,!b),ze.threadInit()}function z1(N){S?postMessage({cmd:"cleanupThread",thread:N}):yf(N)}function bf(N){var O=ze.getNewWorker();if(!O)return 6;ze.runningWorkers.push(O);var H=ze.pthreads[N.pthread_ptr]={worker:O,threadInfoStruct:N.pthread_ptr};O.pthread=H;var ne={cmd:"run",start_routine:N.startRoutine,arg:N.arg,threadInfoStruct:N.pthread_ptr};return O.runPthread=()=>{ne.time=performance.now(),O.postMessage(ne,N.transferList)},O.loaded&&(O.runPthread(),delete O.runPthread),0}function L1(N,O,H,ne){if(typeof SharedArrayBuffer=="undefined")return B("Current environment does not support SharedArrayBuffer, pthreads are not available!"),6;var Ae=[],ve=0;if(S&&(Ae.length===0||ve))return ov(687865856,N,O,H,ne);if(ve)return ve;var Ne={startRoutine:H,pthread_ptr:N,arg:ne,transferList:Ae};return S?(Ne.cmd="spawnThread",postMessage(Ne,Ae),0):bf(Ne)}function B1(){return 2097152}function W1(N,O){if(N==O)postMessage({cmd:"processQueuedMainThreadWork"});else if(S)postMessage({targetThread:N,cmd:"processThreadQueue"});else{var H=ze.pthreads[N],ne=H&&H.worker;if(!ne)return;ne.postMessage({cmd:"processThreadQueue"})}return 1}function V1(){wu("")}function U1(){k||w||Q("Blocking on the main thread is very dangerous, see https://emscripten.org/docs/porting/pthreads.html#blocking-on-the-main-browser-thread")}function vf(){return 2147483648}function G1(N,O,H){o().copyWithin(N,O,O+H)}function H1(){return k?w_().cpus().length:navigator.hardwareConcurrency}function Ri(N,O){var H=arguments.length-2,ne=arguments;return Iu(function(){for(var Ae=H,ve=_u(Ae*8),Ne=ve>>3,Me=0;Me<H;Me++){var Lt=ne[2+Me];p()[Ne+Me]=Lt}return iv(N,Ae,ve,O)})}var Ud=[];function j1(N,O,H){Ud.length=O;for(var ne=H>>3,Ae=0;Ae<O;Ae++)Ud[Ae]=p()[ne+Ae];var ve=N<0,Ne=ve?C1[-N-1]:dg[N];return Ne.apply(null,Ud)}function q1(N){try{return $e.grow(N-qn.byteLength+65535>>>16),Ar($e.buffer),1}catch(O){}}function X1(N){var O=o().length;if(N=N>>>0,N<=O)return!1;var H=vf();if(N>H)return!1;for(var ne=1;ne<=4;ne*=2){var Ae=O*(1+.2/ne);Ae=Math.min(Ae,N+100663296);var ve=Math.min(H,bu(Math.max(N,Ae),65536)),Ne=q1(ve);if(Ne)return!0}return!1}var Je={inEventHandler:0,removeAllEventListeners:function(){for(var N=Je.eventHandlers.length-1;N>=0;--N)Je._removeHandler(N);Je.eventHandlers=[],Je.deferredCalls=[]},registerRemoveEventListeners:function(){Je.removeEventListenersRegistered||(b1.push(Je.removeAllEventListeners),Je.removeEventListenersRegistered=!0)},deferredCalls:[],deferCall:function(N,O,H){function ne(Ne,Me){if(Ne.length!=Me.length)return!1;for(var Lt in Ne)if(Ne[Lt]!=Me[Lt])return!1;return!0}for(var Ae in Je.deferredCalls){var ve=Je.deferredCalls[Ae];if(ve.targetFunction==N&&ne(ve.argsList,H))return}Je.deferredCalls.push({targetFunction:N,precedence:O,argsList:H}),Je.deferredCalls.sort(function(Ne,Me){return Ne.precedence<Me.precedence})},removeDeferredCalls:function(N){for(var O=0;O<Je.deferredCalls.length;++O)Je.deferredCalls[O].targetFunction==N&&(Je.deferredCalls.splice(O,1),--O)},canPerformEventHandlerRequests:function(){return Je.inEventHandler&&Je.currentEventHandler.allowsDeferredCalls},runDeferredCalls:function(){if(!!Je.canPerformEventHandlerRequests())for(var N=0;N<Je.deferredCalls.length;++N){var O=Je.deferredCalls[N];Je.deferredCalls.splice(N,1),--N,O.targetFunction.apply(null,O.argsList)}},eventHandlers:[],removeAllHandlersOnTarget:function(N,O){for(var H=0;H<Je.eventHandlers.length;++H)Je.eventHandlers[H].target==N&&(!O||O==Je.eventHandlers[H].eventTypeString)&&Je._removeHandler(H--)},_removeHandler:function(N){var O=Je.eventHandlers[N];O.target.removeEventListener(O.eventTypeString,O.eventListenerFunc,O.useCapture),Je.eventHandlers.splice(N,1)},registerOrRemoveHandler:function(N){var O=function(Ae){++Je.inEventHandler,Je.currentEventHandler=N,Je.runDeferredCalls(),N.handlerFunc(Ae),Je.runDeferredCalls(),--Je.inEventHandler};if(N.callbackfunc)N.eventListenerFunc=O,N.target.addEventListener(N.eventTypeString,O,N.useCapture),Je.eventHandlers.push(N),Je.registerRemoveEventListeners();else for(var H=0;H<Je.eventHandlers.length;++H)Je.eventHandlers[H].target==N.target&&Je.eventHandlers[H].eventTypeString==N.eventTypeString&&Je._removeHandler(H--)},queueEventHandlerOnThread_iiii:function(N,O,H,ne,Ae){Iu(function(){var ve=_u(12);u()[ve>>2]=H,u()[ve+4>>2]=ne,u()[ve+8>>2]=Ae,Yg(N,637534208,O,ne,ve)})},getTargetThreadForEventCallback:function(N){switch(N){case 1:return 0;case 2:return ze.currentProxiedOperationCallerThread;default:return N}},getNodeNameForTarget:function(N){return N?N==window?"#window":N==screen?"#screen":N&&N.nodeName?N.nodeName:"":""},fullscreenEnabled:function(){return document.fullscreenEnabled||document.webkitFullscreenEnabled}};function K1(N){var O=xu(N)+1,H=Zg(O);return aa(N,H,O),H}function Z1(N,O,H,ne){Iu(function(){var Ae=_u(12),ve=0;O&&(ve=K1(O)),u()[Ae>>2]=ve,u()[Ae+4>>2]=H,u()[Ae+8>>2]=ne,Yg(N,657457152,0,ve,Ae)})}function Y1(N,O,H,ne){O=O?jn(O):"",Z1(N,O,H,ne)}function J1(N){return N>2?jn(N):N}var Q1=[0,typeof document!="undefined"?document:0,typeof window!="undefined"?window:0];function eg(N){N=J1(N);var O=Q1[N]||(typeof document!="undefined"?document.querySelector(N):void 0);return O}function Gd(N){return eg(N)}function wf(N,O,H){var ne=Gd(N);if(!ne)return-4;if(ne.canvasSharedPtr&&(u()[ne.canvasSharedPtr>>2]=O,u()[ne.canvasSharedPtr+4>>2]=H),ne.offscreenCanvas||!ne.controlTransferredOffscreen){ne.offscreenCanvas&&(ne=ne.offscreenCanvas);var Ae=!1;if(ne.GLctxObject&&ne.GLctxObject.GLctx){var ve=ne.GLctxObject.GLctx.getParameter(2978);Ae=ve[0]===0&&ve[1]===0&&ve[2]===ne.width&&ve[3]===ne.height}ne.width=O,ne.height=H,Ae&&ne.GLctxObject.GLctx.viewport(0,0,O,H)}else if(ne.canvasSharedPtr){var Ne=u()[ne.canvasSharedPtr+8>>2];return Y1(Ne,N,O,H),1}else return-4;return 0}function kf(N,O,H){return S?Ri(2,1,N,O,H):wf(N,O,H)}function tg(N,O,H){var ne=Gd(N);return ne?wf(N,O,H):kf(N,O,H)}function ng(){throw"unwind"}function sg(N){var O=N.getExtension("ANGLE_instanced_arrays");if(O)return N.vertexAttribDivisor=function(H,ne){O.vertexAttribDivisorANGLE(H,ne)},N.drawArraysInstanced=function(H,ne,Ae,ve){O.drawArraysInstancedANGLE(H,ne,Ae,ve)},N.drawElementsInstanced=function(H,ne,Ae,ve,Ne){O.drawElementsInstancedANGLE(H,ne,Ae,ve,Ne)},1}function rg(N){var O=N.getExtension("OES_vertex_array_object");if(O)return N.createVertexArray=function(){return O.createVertexArrayOES()},N.deleteVertexArray=function(H){O.deleteVertexArrayOES(H)},N.bindVertexArray=function(H){O.bindVertexArrayOES(H)},N.isVertexArray=function(H){return O.isVertexArrayOES(H)},1}function ag(N){var O=N.getExtension("WEBGL_draw_buffers");if(O)return N.drawBuffers=function(H,ne){O.drawBuffersWEBGL(H,ne)},1}function og(N){return!!(N.multiDrawWebgl=N.getExtension("WEBGL_multi_draw"))}var zt={counter:1,buffers:[],programs:[],framebuffers:[],renderbuffers:[],textures:[],shaders:[],vaos:[],contexts:{},offscreenCanvases:{},queries:[],stringCache:{},unpackAlignment:4,recordError:function(O){zt.lastError||(zt.lastError=O)},getNewId:function(N){for(var O=zt.counter++,H=N.length;H<O;H++)N[H]=null;return O},getSource:function(N,O,H,ne){for(var Ae="",ve=0;ve<O;++ve){var Ne=ne?u()[ne+ve*4>>2]:-1;Ae+=jn(u()[H+ve*4>>2],Ne<0?void 0:Ne)}return Ae},createContext:function(N,O){N.getContextSafariWebGL2Fixed||(N.getContextSafariWebGL2Fixed=N.getContext,N.getContext=function(Ae,ve){var Ne=N.getContextSafariWebGL2Fixed(Ae,ve);return Ae=="webgl"==Ne instanceof WebGLRenderingContext?Ne:null});var H=N.getContext("webgl",O);if(!H)return 0;var ne=zt.registerContext(H,O);return ne},registerContext:function(N,O){var H=Zg(8);u()[H+4>>2]=Df();var ne={handle:H,attributes:O,version:O.majorVersion,GLctx:N};return N.canvas&&(N.canvas.GLctxObject=ne),zt.contexts[H]=ne,(typeof O.enableExtensionsByDefault=="undefined"||O.enableExtensionsByDefault)&&zt.initExtensions(ne),H},makeContextCurrent:function(N){return zt.currentContext=zt.contexts[N],d.ctx=Tf=zt.currentContext&&zt.currentContext.GLctx,!(N&&!Tf)},getContext:function(N){return zt.contexts[N]},deleteContext:function(N){zt.currentContext===zt.contexts[N]&&(zt.currentContext=null),typeof Je=="object"&&Je.removeAllHandlersOnTarget(zt.contexts[N].GLctx.canvas),zt.contexts[N]&&zt.contexts[N].GLctx.canvas&&(zt.contexts[N].GLctx.canvas.GLctxObject=void 0),nv(zt.contexts[N].handle),zt.contexts[N]=null},initExtensions:function(N){if(N||(N=zt.currentContext),!N.initExtensionsDone){N.initExtensionsDone=!0;var O=N.GLctx;sg(O),rg(O),ag(O),O.disjointTimerQueryExt=O.getExtension("EXT_disjoint_timer_query"),og(O);var H=O.getSupportedExtensions()||[];H.forEach(function(ne){!ne.includes("lose_context")&&!ne.includes("debug")&&O.getExtension(ne)})}}},ig=["default","low-power","high-performance"];function lg(N,O){var H=O>>2,ne=u()[H+6],Ae={alpha:!!u()[H+0],depth:!!u()[H+1],stencil:!!u()[H+2],antialias:!!u()[H+3],premultipliedAlpha:!!u()[H+4],preserveDrawingBuffer:!!u()[H+5],powerPreference:ig[ne],failIfMajorPerformanceCaveat:!!u()[H+7],majorVersion:u()[H+8],minorVersion:u()[H+9],enableExtensionsByDefault:u()[H+10],explicitSwapControl:u()[H+11],proxyContextToMainThread:u()[H+12],renderViaOffscreenBackBuffer:u()[H+13]},ve=Gd(N);if(!ve||Ae.explicitSwapControl)return 0;var Ne=zt.createContext(ve,Ae);return Ne}function ug(N,O){return lg(N,O)}var Cu={mappings:{},buffers:[null,[],[]],printChar:function(N,O){var H=Cu.buffers[N];O===0||O===10?((N===1?te:B)(Us(H,0)),H.length=0):H.push(O)},varargs:void 0,get:function(){Cu.varargs+=4;var N=u()[Cu.varargs-4>>2];return N},getStr:function(N){var O=jn(N);return O},get64:function(N,O){return N}};function If(N){return S?Ri(3,1,N):0}function Sf(N,O,H,ne,Ae){if(S)return Ri(4,1,N,O,H,ne,Ae)}function Cf(N,O,H,ne){if(S)return Ri(5,1,N,O,H,ne);for(var Ae=0,ve=0;ve<H;ve++){var Ne=u()[O>>2],Me=u()[O+4>>2];O+=8;for(var Lt=0;Lt<Me;Lt++)Cu.printChar(N,o()[Ne+Lt]);Ae+=Me}return u()[ne>>2]=Ae,0}function cg(N){Be(N)}ze.init();var Tf,dg=[null,xf,kf,If,Sf,Cf],Q4=!1,Nf={__clock_gettime:O1,__emscripten_init_main_thread_js:M1,__emscripten_thread_cleanup:z1,__pthread_create_js:L1,_emscripten_default_pthread_stack_size:B1,_emscripten_notify_thread_queue:W1,abort:V1,emscripten_check_blocking_allowed:U1,emscripten_get_heap_max:vf,emscripten_get_now:Ei,emscripten_memcpy_big:G1,emscripten_num_logical_cores:H1,emscripten_receive_on_main_thread_js:j1,emscripten_resize_heap:X1,emscripten_set_canvas_element_size:tg,emscripten_unwind_to_js_event_loop:ng,emscripten_webgl_create_context:ug,exit:Af,fd_close:If,fd_seek:Sf,fd_write:Cf,memory:$e||d.wasmMemory,setTempRet0:cg},ev=S1(),pg=d.___wasm_call_ctors=function(){return(pg=d.___wasm_call_ctors=d.asm.__wasm_call_ctors).apply(null,arguments)},hg=d._init=function(){return(hg=d._init=d.asm.init).apply(null,arguments)},fg=d._init_with_threads_count=function(){return(fg=d._init_with_threads_count=d.asm.init_with_threads_count).apply(null,arguments)},mg=d._get_threads_count=function(){return(mg=d._get_threads_count=d.asm.get_threads_count).apply(null,arguments)},gg=d._register_tensor=function(){return(gg=d._register_tensor=d.asm.register_tensor).apply(null,arguments)},yg=d._dispose_data=function(){return(yg=d._dispose_data=d.asm.dispose_data).apply(null,arguments)},Ag=d._dispose=function(){return(Ag=d._dispose=d.asm.dispose).apply(null,arguments)},xg=d._Abs=function(){return(xg=d._Abs=d.asm.Abs).apply(null,arguments)},bg=d._Add=function(){return(bg=d._Add=d.asm.Add).apply(null,arguments)},vg=d._AddN=function(){return(vg=d._AddN=d.asm.AddN).apply(null,arguments)},wg=d._All=function(){return(wg=d._All=d.asm.All).apply(null,arguments)},kg=d._Any=function(){return(kg=d._Any=d.asm.Any).apply(null,arguments)},Ig=d._ArgMax=function(){return(Ig=d._ArgMax=d.asm.ArgMax).apply(null,arguments)},Sg=d._AvgPool=function(){return(Sg=d._AvgPool=d.asm.AvgPool).apply(null,arguments)},Cg=d._BatchMatMul=function(){return(Cg=d._BatchMatMul=d.asm.BatchMatMul).apply(null,arguments)},Tg=d._Ceil=function(){return(Tg=d._Ceil=d.asm.Ceil).apply(null,arguments)},Ng=d._ClipByValue=function(){return(Ng=d._ClipByValue=d.asm.ClipByValue).apply(null,arguments)},Eg=d._Conv2D=function(){return(Eg=d._Conv2D=d.asm.Conv2D).apply(null,arguments)},Rg=d._Conv2DBackpropInput=function(){return(Rg=d._Conv2DBackpropInput=d.asm.Conv2DBackpropInput).apply(null,arguments)},_g=d._Cos=function(){return(_g=d._Cos=d.asm.Cos).apply(null,arguments)},Dg=d._Cosh=function(){return(Dg=d._Cosh=d.asm.Cosh).apply(null,arguments)},$g=d._CropAndResize=function(){return($g=d._CropAndResize=d.asm.CropAndResize).apply(null,arguments)},Pg=d._Cumprod=function(){return(Pg=d._Cumprod=d.asm.Cumprod).apply(null,arguments)},Fg=d._Cumsum=function(){return(Fg=d._Cumsum=d.asm.Cumsum).apply(null,arguments)},Og=d._DepthToSpace=function(){return(Og=d._DepthToSpace=d.asm.DepthToSpace).apply(null,arguments)},Mg=d._DepthwiseConv2dNative=function(){return(Mg=d._DepthwiseConv2dNative=d.asm.DepthwiseConv2dNative).apply(null,arguments)},zg=d._Elu=function(){return(zg=d._Elu=d.asm.Elu).apply(null,arguments)},Lg=d._Equal=function(){return(Lg=d._Equal=d.asm.Equal).apply(null,arguments)},Bg=d._Exp=function(){return(Bg=d._Exp=d.asm.Exp).apply(null,arguments)},Wg=d._FlipLeftRight=function(){return(Wg=d._FlipLeftRight=d.asm.FlipLeftRight).apply(null,arguments)},Vg=d._Floor=function(){return(Vg=d._Floor=d.asm.Floor).apply(null,arguments)},Ug=d._FloorDiv=function(){return(Ug=d._FloorDiv=d.asm.FloorDiv).apply(null,arguments)},Gg=d._FusedBatchNorm=function(){return(Gg=d._FusedBatchNorm=d.asm.FusedBatchNorm).apply(null,arguments)},Hg=d._FusedConv2D=function(){return(Hg=d._FusedConv2D=d.asm.FusedConv2D).apply(null,arguments)},Ef=d._FusedDepthwiseConv2D=function(){return(Ef=d._FusedDepthwiseConv2D=d.asm.FusedDepthwiseConv2D).apply(null,arguments)},Rf=d._Gather=function(){return(Rf=d._Gather=d.asm.Gather).apply(null,arguments)},Hd=d._GatherNd=function(){return(Hd=d._GatherNd=d.asm.GatherNd).apply(null,arguments)},jg=d._Greater=function(){return(jg=d._Greater=d.asm.Greater).apply(null,arguments)},qg=d._GreaterEqual=function(){return(qg=d._GreaterEqual=d.asm.GreaterEqual).apply(null,arguments)},Tu=d._LeakyRelu=function(){return(Tu=d._LeakyRelu=d.asm.LeakyRelu).apply(null,arguments)},jd=d._Less=function(){return(jd=d._Less=d.asm.Less).apply(null,arguments)},qd=d._LessEqual=function(){return(qd=d._LessEqual=d.asm.LessEqual).apply(null,arguments)},tv=d._Log=function(){return(tv=d._Log=d.asm.Log).apply(null,arguments)},Nu=d._LogicalAnd=function(){return(Nu=d._LogicalAnd=d.asm.LogicalAnd).apply(null,arguments)},Eu=d._LogicalNot=function(){return(Eu=d._LogicalNot=d.asm.LogicalNot).apply(null,arguments)},Xg=d._LogicalOr=function(){return(Xg=d._LogicalOr=d.asm.LogicalOr).apply(null,arguments)},X=d._LogicalXor=function(){return(X=d._LogicalXor=d.asm.LogicalXor).apply(null,arguments)},se=d._Max=function(){return(se=d._Max=d.asm.Max).apply(null,arguments)},xe=d._MaxPool=function(){return(xe=d._MaxPool=d.asm.MaxPool).apply(null,arguments)},De=d._Maximum=function(){return(De=d._Maximum=d.asm.Maximum).apply(null,arguments)},ft=d._Mean=function(){return(ft=d._Mean=d.asm.Mean).apply(null,arguments)},yt=d._Min=function(){return(yt=d._Min=d.asm.Min).apply(null,arguments)},tt=d._Minimum=function(){return(tt=d._Minimum=d.asm.Minimum).apply(null,arguments)},Ze=d._MirrorPad=function(){return(Ze=d._MirrorPad=d.asm.MirrorPad).apply(null,arguments)},en=d._Multiply=function(){return(en=d._Multiply=d.asm.Multiply).apply(null,arguments)},br=d._Neg=function(){return(br=d._Neg=d.asm.Neg).apply(null,arguments)},vr=d._NonMaxSuppressionV3=function(){return(vr=d._NonMaxSuppressionV3=d.asm.NonMaxSuppressionV3).apply(null,arguments)},Ru=d._NonMaxSuppressionV4=function(){return(Ru=d._NonMaxSuppressionV4=d.asm.NonMaxSuppressionV4).apply(null,arguments)},_i=d._NonMaxSuppressionV5=function(){return(_i=d._NonMaxSuppressionV5=d.asm.NonMaxSuppressionV5).apply(null,arguments)},Kg=d._NotEqual=function(){return(Kg=d._NotEqual=d.asm.NotEqual).apply(null,arguments)},Zn=d._OneHot=function(){return(Zn=d._OneHot=d.asm.OneHot).apply(null,arguments)},La=d._PadV2=function(){return(La=d._PadV2=d.asm.PadV2).apply(null,arguments)},_f=d._Pow=function(){return(_f=d._Pow=d.asm.Pow).apply(null,arguments)},oR=d._Prelu=function(){return(oR=d._Prelu=d.asm.Prelu).apply(null,arguments)},iR=d._Prod=function(){return(iR=d._Prod=d.asm.Prod).apply(null,arguments)},lR=d._RealDiv=function(){return(lR=d._RealDiv=d.asm.RealDiv).apply(null,arguments)},uR=d._Relu=function(){return(uR=d._Relu=d.asm.Relu).apply(null,arguments)},cR=d._Relu6=function(){return(cR=d._Relu6=d.asm.Relu6).apply(null,arguments)},dR=d._ResizeBilinear=function(){return(dR=d._ResizeBilinear=d.asm.ResizeBilinear).apply(null,arguments)},pR=d._ResizeNearestNeighbor=function(){return(pR=d._ResizeNearestNeighbor=d.asm.ResizeNearestNeighbor).apply(null,arguments)},hR=d._Reverse=function(){return(hR=d._Reverse=d.asm.Reverse).apply(null,arguments)},fR=d._RotateWithOffset=function(){return(fR=d._RotateWithOffset=d.asm.RotateWithOffset).apply(null,arguments)},mR=d._Round=function(){return(mR=d._Round=d.asm.Round).apply(null,arguments)},gR=d._Rsqrt=function(){return(gR=d._Rsqrt=d.asm.Rsqrt).apply(null,arguments)},yR=d._ScatterNd=function(){return(yR=d._ScatterNd=d.asm.ScatterNd).apply(null,arguments)},AR=d._SelectV2=function(){return(AR=d._SelectV2=d.asm.SelectV2).apply(null,arguments)},xR=d._Sigmoid=function(){return(xR=d._Sigmoid=d.asm.Sigmoid).apply(null,arguments)},bR=d._Sin=function(){return(bR=d._Sin=d.asm.Sin).apply(null,arguments)},vR=d._Softmax=function(){return(vR=d._Softmax=d.asm.Softmax).apply(null,arguments)},wR=d._SparseFillEmptyRows=function(){return(wR=d._SparseFillEmptyRows=d.asm.SparseFillEmptyRows).apply(null,arguments)},kR=d._SparseReshape=function(){return(kR=d._SparseReshape=d.asm.SparseReshape).apply(null,arguments)},IR=d._SparseSegmentReduction=function(){return(IR=d._SparseSegmentReduction=d.asm.SparseSegmentReduction).apply(null,arguments)},SR=d._Sqrt=function(){return(SR=d._Sqrt=d.asm.Sqrt).apply(null,arguments)},CR=d._Square=function(){return(CR=d._Square=d.asm.Square).apply(null,arguments)},TR=d._SquaredDifference=function(){return(TR=d._SquaredDifference=d.asm.SquaredDifference).apply(null,arguments)},NR=d._Step=function(){return(NR=d._Step=d.asm.Step).apply(null,arguments)},ER=d._StridedSlice=function(){return(ER=d._StridedSlice=d.asm.StridedSlice).apply(null,arguments)},RR=d._Sub=function(){return(RR=d._Sub=d.asm.Sub).apply(null,arguments)},_R=d._Sum=function(){return(_R=d._Sum=d.asm.Sum).apply(null,arguments)},DR=d._Tan=function(){return(DR=d._Tan=d.asm.Tan).apply(null,arguments)},$R=d._Tanh=function(){return($R=d._Tanh=d.asm.Tanh).apply(null,arguments)},PR=d._Tile=function(){return(PR=d._Tile=d.asm.Tile).apply(null,arguments)},FR=d._TopK=function(){return(FR=d._TopK=d.asm.TopK).apply(null,arguments)},OR=d._Transform=function(){return(OR=d._Transform=d.asm.Transform).apply(null,arguments)},MR=d._Transpose=function(){return(MR=d._Transpose=d.asm.Transpose).apply(null,arguments)},zR=d.__FusedMatMul=function(){return(zR=d.__FusedMatMul=d.asm._FusedMatMul).apply(null,arguments)},Zg=d._malloc=function(){return(Zg=d._malloc=d.asm.malloc).apply(null,arguments)},nv=d._free=function(){return(nv=d._free=d.asm.free).apply(null,arguments)},LR=d._emscripten_tls_init=function(){return(LR=d._emscripten_tls_init=d.asm.emscripten_tls_init).apply(null,arguments)},sv=d.___errno_location=function(){return(sv=d.___errno_location=d.asm.__errno_location).apply(null,arguments)},Df=d._pthread_self=function(){return(Df=d._pthread_self=d.asm.pthread_self).apply(null,arguments)},rv=d._emscripten_main_thread_process_queued_calls=function(){return(rv=d._emscripten_main_thread_process_queued_calls=d.asm.emscripten_main_thread_process_queued_calls).apply(null,arguments)},BR=d.__emscripten_thread_crashed=function(){return(BR=d.__emscripten_thread_crashed=d.asm._emscripten_thread_crashed).apply(null,arguments)},av=d.__emscripten_thread_init=function(){return(av=d.__emscripten_thread_init=d.asm._emscripten_thread_init).apply(null,arguments)},WR=d._emscripten_current_thread_process_queued_calls=function(){return(WR=d._emscripten_current_thread_process_queued_calls=d.asm.emscripten_current_thread_process_queued_calls).apply(null,arguments)},VR=d._emscripten_main_browser_thread_id=function(){return(VR=d._emscripten_main_browser_thread_id=d.asm.emscripten_main_browser_thread_id).apply(null,arguments)},UR=d._emscripten_sync_run_in_main_thread_2=function(){return(UR=d._emscripten_sync_run_in_main_thread_2=d.asm.emscripten_sync_run_in_main_thread_2).apply(null,arguments)},ov=d._emscripten_sync_run_in_main_thread_4=function(){return(ov=d._emscripten_sync_run_in_main_thread_4=d.asm.emscripten_sync_run_in_main_thread_4).apply(null,arguments)},iv=d._emscripten_run_in_main_runtime_thread_js=function(){return(iv=d._emscripten_run_in_main_runtime_thread_js=d.asm.emscripten_run_in_main_runtime_thread_js).apply(null,arguments)},Yg=d._emscripten_dispatch_to_thread_=function(){return(Yg=d._emscripten_dispatch_to_thread_=d.asm.emscripten_dispatch_to_thread_).apply(null,arguments)},Jg=d.__emscripten_thread_free_data=function(){return(Jg=d.__emscripten_thread_free_data=d.asm._emscripten_thread_free_data).apply(null,arguments)},GR=d.__emscripten_thread_exit=function(){return(GR=d.__emscripten_thread_exit=d.asm._emscripten_thread_exit).apply(null,arguments)},HR=d._memalign=function(){return(HR=d._memalign=d.asm.memalign).apply(null,arguments)},lv=d._emscripten_stack_set_limits=function(){return(lv=d._emscripten_stack_set_limits=d.asm.emscripten_stack_set_limits).apply(null,arguments)},Qg=d.stackSave=function(){return(Qg=d.stackSave=d.asm.stackSave).apply(null,arguments)},$f=d.stackRestore=function(){return($f=d.stackRestore=d.asm.stackRestore).apply(null,arguments)},_u=d.stackAlloc=function(){return(_u=d.stackAlloc=d.asm.stackAlloc).apply(null,arguments)},jR=d.dynCall_iijjiiii=function(){return(jR=d.dynCall_iijjiiii=d.asm.dynCall_iijjiiii).apply(null,arguments)},qR=d.dynCall_jiji=function(){return(qR=d.dynCall_jiji=d.asm.dynCall_jiji).apply(null,arguments)},uv=d.__emscripten_allow_main_runtime_queued_calls=21672;d.cwrap=Hn,d.keepRuntimeAlive=Ti,d.PThread=ze,d.PThread=ze,d.wasmMemory=$e,d.ExitStatus=Xd;var Pf;function Xd(N){this.name="ExitStatus",this.message="Program terminated with exit("+N+")",this.status=N}xr=function N(){Pf||e3(),Pf||(xr=N)};function e3(N){if(N=N||y,za>0)return;if(S){h(d),Ld(),postMessage({cmd:"loaded"});return}if(Xn(),za>0)return;function O(){Pf||(Pf=!0,d.calledRun=!0,!It&&(Ld(),h(d),d.onRuntimeInitialized&&d.onRuntimeInitialized(),k1()))}d.setStatus?(d.setStatus("Running..."),setTimeout(function(){setTimeout(function(){d.setStatus("")},1),O()},1)):O()}d.run=e3;function XR(N,O){if(Gn=N,!O&&S)throw xf(N),"unwind";Ti()||w1(),KR(N)}function KR(N){Gn=N,Ti()||(ze.terminateAllThreads(),d.onExit&&d.onExit(N),It=!0),A(N,new Xd(N))}if(d.preInit)for(typeof d.preInit=="function"&&(d.preInit=[d.preInit]);d.preInit.length>0;)d.preInit.pop()();e3();var Ff;m&&(Ff={uncaughtException:process.listeners("uncaughtException").filter(function(N){return!m.uncaughtException.indexOf(N)>-1}),unhandledRejection:process.listeners("unhandledRejection").filter(function(N){return!m.unhandledRejection.indexOf(N)>-1})});var Of;if(typeof WasmBackendModule!="undefined")Of=WasmBackendModule;else if(typeof r!="undefined")Of=r;else throw new Error("Could not find wasm module in post.js");if(Ff){var ZR=Of._dispose;Of._dispose=function(){ZR(),Ff.uncaughtException.forEach(function(N){process.removeListener("uncaughtException",N)}),Ff.unhandledRejection.forEach(function(N){process.removeListener("unhandledRejection",N)})}}return r.ready}})();typeof e=="object"&&typeof t=="object"?t.exports=n:typeof define=="function"&&define.amd?define([],function(){return n}):typeof e=="object"&&(e.WasmBackendModuleThreadedSimd=n)}}),I_=ln({"node_modules/.pnpm/@tensorflow+tfjs-backend-wasm@3.19.0_hek32lflchivueqv5i4vgonghu/node_modules/@tensorflow/tfjs-backend-wasm/wasm-out/tfjs-backend-wasm-threaded-simd.worker.js"(e,t){t.exports.wasmWorkerContents=`"use strict";var Module={};var ENVIRONMENT_IS_NODE=typeof process==="object"&&typeof process.versions==="object"&&typeof process.versions.node==="string";if(ENVIRONMENT_IS_NODE){var nodeWorkerThreads=require("worker_threads");var parentPort=nodeWorkerThreads.parentPort;parentPort.on("message",function(data){onmessage({data:data})});var fs=require("fs");Object.assign(global,{self:global,require:require,Module:Module,location:{href:__filename},Worker:nodeWorkerThreads.Worker,importScripts:function(f){(0,eval)(fs.readFileSync(f,"utf8"))},postMessage:function(msg){parentPort.postMessage(msg)},performance:global.performance||{now:function(){return Date.now()}}})}function threadPrintErr(){var text=Array.prototype.slice.call(arguments).join(" ");if(ENVIRONMENT_IS_NODE){fs.writeSync(2,text+"
|
|
");return}console.error(text)}function threadAlert(){var text=Array.prototype.slice.call(arguments).join(" ");postMessage({cmd:"alert",text:text,threadId:Module["_pthread_self"]()})}var err=threadPrintErr;self.alert=threadAlert;Module["instantiateWasm"]=((info,receiveInstance)=>{var instance=new WebAssembly.Instance(Module["wasmModule"],info);receiveInstance(instance);Module["wasmModule"]=null;return instance.exports});self.onmessage=(e=>{try{if(e.data.cmd==="load"){Module["wasmModule"]=e.data.wasmModule;Module["wasmMemory"]=e.data.wasmMemory;Module["buffer"]=Module["wasmMemory"].buffer;Module["ENVIRONMENT_IS_PTHREAD"]=true;if(typeof e.data.urlOrBlob==="string"){importScripts(e.data.urlOrBlob)}else{var objectUrl=URL.createObjectURL(e.data.urlOrBlob);importScripts(objectUrl);URL.revokeObjectURL(objectUrl)}WasmBackendModuleThreadedSimd(Module).then(function(instance){Module=instance})}else if(e.data.cmd==="run"){Module["__performance_now_clock_drift"]=performance.now()-e.data.time;Module["__emscripten_thread_init"](e.data.threadInfoStruct,0,0,1);Module["establishStackSpace"]();Module["PThread"].receiveObjectTransfer(e.data);Module["PThread"].threadInit();try{var result=Module["invokeEntryPoint"](e.data.start_routine,e.data.arg);if(Module["keepRuntimeAlive"]()){Module["PThread"].setExitStatus(result)}else{Module["__emscripten_thread_exit"](result)}}catch(ex){if(ex!="unwind"){if(ex instanceof Module["ExitStatus"]){if(Module["keepRuntimeAlive"]()){}else{Module["__emscripten_thread_exit"](ex.status)}}else{throw ex}}}}else if(e.data.cmd==="cancel"){if(Module["_pthread_self"]()){Module["__emscripten_thread_exit"](-1)}}else if(e.data.target==="setimmediate"){}else if(e.data.cmd==="processThreadQueue"){if(Module["_pthread_self"]()){Module["_emscripten_current_thread_process_queued_calls"]()}}else if(e.data.cmd==="processProxyingQueue"){if(Module["_pthread_self"]()){Module["_emscripten_proxy_execute_queue"](e.data.queue)}}else{err("worker.js received unknown command "+e.data.cmd);err(e.data)}}catch(ex){err("worker.js onmessage() captured an uncaught exception: "+ex);if(ex&&ex.stack)err(ex.stack);if(Module["__emscripten_thread_crashed"]){Module["__emscripten_thread_crashed"]()}throw ex}});`}}),S_=ln({"node_modules/.pnpm/@tensorflow+tfjs-backend-wasm@3.19.0_hek32lflchivueqv5i4vgonghu/node_modules/@tensorflow/tfjs-backend-wasm/wasm-out/tfjs-backend-wasm.js"(e,t){var n=(()=>{var s=typeof document!="undefined"&&document.currentScript?document.currentScript.src:void 0;return typeof __filename!="undefined"&&(s=s||__filename),function(r){r=r||{};var a=typeof r!="undefined"?r:{},o,i;a.ready=new Promise(function(X,se){o=X,i=se});var l;typeof process!="undefined"&&process.listeners&&(l={uncaughtException:process.listeners("uncaughtException"),unhandledRejection:process.listeners("unhandledRejection")});var u=Object.assign({},a),c=[],p="./this.program",d=(X,se)=>{throw se},h=typeof window=="object",f=typeof importScripts=="function",m=typeof process=="object"&&typeof process.versions=="object"&&typeof process.versions.node=="string",g="";function y(X){return a.locateFile?a.locateFile(X,g):g+X}var x,A,b,w;function k(X){if(X instanceof jd)return;_("exiting due to exception: "+X)}var S,E,R;m?(f?g=um().dirname(g)+"/":g=__dirname+"/",R=()=>{E||(S=xy(),E=um())},x=function(se,xe){return R(),se=E.normalize(se),S.readFileSync(se,xe?void 0:"utf8")},b=X=>{var se=x(X,!0);return se.buffer||(se=new Uint8Array(se)),se},A=(X,se,xe)=>{R(),X=E.normalize(X),S.readFile(X,function(De,ft){De?xe(De):se(ft.buffer)})},process.argv.length>1&&(p=process.argv[1].replace(/\\/g,"/")),c=process.argv.slice(2),process.on("uncaughtException",function(X){if(!(X instanceof jd))throw X}),process.on("unhandledRejection",function(X){throw X}),d=(X,se)=>{if(zd())throw process.exitCode=X,se;k(se),process.exit(X)},a.inspect=function(){return"[Emscripten Module object]"}):(h||f)&&(f?g=self.location.href:typeof document!="undefined"&&document.currentScript&&(g=document.currentScript.src),s&&(g=s),g.indexOf("blob:")!==0?g=g.substr(0,g.replace(/[?#].*/,"").lastIndexOf("/")+1):g="",x=X=>{var se=new XMLHttpRequest;return se.open("GET",X,!1),se.send(null),se.responseText},f&&(b=X=>{var se=new XMLHttpRequest;return se.open("GET",X,!1),se.responseType="arraybuffer",se.send(null),new Uint8Array(se.response)}),A=(X,se,xe)=>{var De=new XMLHttpRequest;De.open("GET",X,!0),De.responseType="arraybuffer",De.onload=()=>{if(De.status==200||De.status==0&&De.response){se(De.response);return}xe()},De.onerror=xe,De.send(null)},w=X=>document.title=X);var $=a.print||console.log.bind(console),_=a.printErr||console.warn.bind(console);Object.assign(a,u),u=null,a.arguments&&(c=a.arguments),a.thisProgram&&(p=a.thisProgram),a.quit&&(d=a.quit);var P=4;function C(X){C.shown||(C.shown={}),C.shown[X]||(C.shown[X]=1,_(X))}function F(X,se){if(typeof WebAssembly.Function=="function"){for(var xe={i:"i32",j:"i64",f:"f32",d:"f64"},De={parameters:[],results:se[0]=="v"?[]:[xe[se[0]]]},ft=1;ft<se.length;++ft)De.parameters.push(xe[se[ft]]);return new WebAssembly.Function(De,X)}var yt=[1,0,1,96],tt=se.slice(0,1),Ze=se.slice(1),en={i:127,j:126,f:125,d:124};yt.push(Ze.length);for(var ft=0;ft<Ze.length;++ft)yt.push(en[Ze[ft]]);tt=="v"?yt.push(0):yt=yt.concat([1,en[tt]]),yt[1]=yt.length-2;var br=new Uint8Array([0,97,115,109,1,0,0,0].concat(yt,[2,7,1,1,101,1,102,0,0,7,5,1,1,102,0,0])),vr=new WebAssembly.Module(br),Ru=new WebAssembly.Instance(vr,{e:{f:X}}),_i=Ru.exports.f;return _i}var G=[],K;function z(){if(G.length)return G.pop();try{Oa.grow(1)}catch(X){throw X instanceof RangeError?"Unable to grow wasm table. Set ALLOW_TABLE_GROWTH.":X}return Oa.length-1}function Z(X,se){for(var xe=X;xe<X+se;xe++){var De=Wd(xe);De&&K.set(De,xe)}}var J=0,te=X=>{J=X},B;a.wasmBinary&&(B=a.wasmBinary);var oe=a.noExitRuntime||!0;typeof WebAssembly!="object"&&Ci("no native wasm support detected");var Q,ae=!1,ie;function me(X,se){X||Ci(se)}function we(X){var se=a["_"+X];return se}function Re(X,se,xe,De,ft){var yt={string:function(Zn){var La=0;if(Zn!=null&&Zn!==0){var _f=(Zn.length<<2)+1;La=Hd(_f),At(Zn,La,_f)}return La},array:function(Zn){var La=Hd(Zn.length);return It(Zn,La),La}};function tt(Zn){return se==="string"?pt(Zn):se==="boolean"?Boolean(Zn):Zn}var Ze=we(X),en=[],br=0;if(De)for(var vr=0;vr<De.length;vr++){var Ru=yt[xe[vr]];Ru?(br===0&&(br=Ef()),en[vr]=Ru(De[vr])):en[vr]=De[vr]}var _i=Ze.apply(null,en);function Kg(Zn){return br!==0&&Rf(br),tt(Zn)}return _i=Kg(_i),_i}function _e(X,se,xe,De){xe=xe||[];var ft=xe.every(function(tt){return tt==="number"}),yt=se!=="string";return yt&&ft&&!De?we(X):function(){return Re(X,se,xe,arguments,De)}}var Be=1,He=typeof TextDecoder!="undefined"?new TextDecoder("utf8"):void 0;function ot(X,se,xe){for(var De=se+xe,ft=se;X[ft]&&!(ft>=De);)++ft;if(ft-se>16&&X.subarray&&He)return He.decode(X.subarray(se,ft));for(var yt="";se<ft;){var tt=X[se++];if(!(tt&128)){yt+=String.fromCharCode(tt);continue}var Ze=X[se++]&63;if((tt&224)==192){yt+=String.fromCharCode((tt&31)<<6|Ze);continue}var en=X[se++]&63;if((tt&240)==224?tt=(tt&15)<<12|Ze<<6|en:tt=(tt&7)<<18|Ze<<12|en<<6|X[se++]&63,tt<65536)yt+=String.fromCharCode(tt);else{var br=tt-65536;yt+=String.fromCharCode(55296|br>>10,56320|br&1023)}}return yt}function pt(X,se){return X?ot(Hn,X,se):""}function ht(X,se,xe,De){if(!(De>0))return 0;for(var ft=xe,yt=xe+De-1,tt=0;tt<X.length;++tt){var Ze=X.charCodeAt(tt);if(Ze>=55296&&Ze<=57343){var en=X.charCodeAt(++tt);Ze=65536+((Ze&1023)<<10)|en&1023}if(Ze<=127){if(xe>=yt)break;se[xe++]=Ze}else if(Ze<=2047){if(xe+1>=yt)break;se[xe++]=192|Ze>>6,se[xe++]=128|Ze&63}else if(Ze<=65535){if(xe+2>=yt)break;se[xe++]=224|Ze>>12,se[xe++]=128|Ze>>6&63,se[xe++]=128|Ze&63}else{if(xe+3>=yt)break;se[xe++]=240|Ze>>18,se[xe++]=128|Ze>>12&63,se[xe++]=128|Ze>>6&63,se[xe++]=128|Ze&63}}return se[xe]=0,xe-ft}function At(X,se,xe){return ht(X,Hn,se,xe)}function $e(X){for(var se=0,xe=0;xe<X.length;++xe){var De=X.charCodeAt(xe);De>=55296&&De<=57343&&(De=65536+((De&1023)<<10)|X.charCodeAt(++xe)&1023),De<=127?++se:De<=2047?se+=2:De<=65535?se+=3:se+=4}return se}var Tt=typeof TextDecoder!="undefined"?new TextDecoder("utf-16le"):void 0;function It(X,se){pn.set(X,se)}function Gn(X,se,xe){for(var De=0;De<X.length;++De)pn[se++>>0]=X.charCodeAt(De);xe||(pn[se>>0]=0)}function Qt(X,se){return X%se>0&&(X+=se-X%se),X}var bs,pn,Hn,vs,ws,Pn,Us,jn,ra;function aa(X){bs=X,a.HEAP8=pn=new Int8Array(X),a.HEAP16=vs=new Int16Array(X),a.HEAP32=Pn=new Int32Array(X),a.HEAPU8=Hn=new Uint8Array(X),a.HEAPU16=ws=new Uint16Array(X),a.HEAPU32=Us=new Uint32Array(X),a.HEAPF32=jn=new Float32Array(X),a.HEAPF64=ra=new Float64Array(X)}var xu=a.INITIAL_MEMORY||16777216,Oa,oa=[],Md=[],bu=[],qn=!1,sf=!1,rf=0;function zd(){return oe||rf>0}function af(){if(a.preRun)for(typeof a.preRun=="function"&&(a.preRun=[a.preRun]);a.preRun.length;)uf(a.preRun.shift());Bd(oa)}function of(){qn=!0,Bd(Md)}function G4(){sf=!0}function lf(){if(a.postRun)for(typeof a.postRun=="function"&&(a.postRun=[a.postRun]);a.postRun.length;)cf(a.postRun.shift());Bd(bu)}function uf(X){oa.unshift(X)}function Ar(X){Md.unshift(X)}function cf(X){bu.unshift(X)}var Gs=0,vu=null,Ma=null;function b1(X){Gs++,a.monitorRunDependencies&&a.monitorRunDependencies(Gs)}function df(X){if(Gs--,a.monitorRunDependencies&&a.monitorRunDependencies(Gs),Gs==0&&(vu!==null&&(clearInterval(vu),vu=null),Ma)){var se=Ma;Ma=null,se()}}a.preloadedImages={},a.preloadedAudios={};function Ci(X){a.onAbort&&a.onAbort(X),X="Aborted("+X+")",_(X),ae=!0,ie=1,X+=". Build with -s ASSERTIONS=1 for more info.";var se=new WebAssembly.RuntimeError(X);throw i(se),se}var v1="data:application/octet-stream;base64,";function pf(X){return X.startsWith(v1)}function Ti(X){return X.startsWith("file://")}var Xn;Xn="tfjs-backend-wasm.wasm",pf(Xn)||(Xn=y(Xn));function Ld(X){try{if(X==Xn&&B)return new Uint8Array(B);if(b)return b(X);throw"both async and sync fetching of the wasm failed"}catch(se){Ci(se)}}function w1(){if(!B&&(h||f)){if(typeof fetch=="function"&&!Ti(Xn))return fetch(Xn,{credentials:"same-origin"}).then(function(X){if(!X.ok)throw"failed to load wasm binary file at '"+Xn+"'";return X.arrayBuffer()}).catch(function(){return Ld(Xn)});if(A)return new Promise(function(X,se){A(Xn,function(xe){X(new Uint8Array(xe))},se)})}return Promise.resolve().then(function(){return Ld(Xn)})}function k1(){var X={env:Iu,wasi_snapshot_preview1:Iu};function se(tt,Ze){var en=tt.exports;a.asm=en,Q=a.asm.memory,aa(Q.buffer),Oa=a.asm.__indirect_function_table,Ar(a.asm.__wasm_call_ctors),df("wasm-instantiate")}b1("wasm-instantiate");function xe(tt){se(tt.instance)}function De(tt){return w1().then(function(Ze){return WebAssembly.instantiate(Ze,X)}).then(function(Ze){return Ze}).then(tt,function(Ze){_("failed to asynchronously prepare wasm: "+Ze),Ci(Ze)})}function ft(){return!B&&typeof WebAssembly.instantiateStreaming=="function"&&!pf(Xn)&&!Ti(Xn)&&typeof fetch=="function"?fetch(Xn,{credentials:"same-origin"}).then(function(tt){var Ze=WebAssembly.instantiateStreaming(tt,X);return Ze.then(xe,function(en){return _("wasm streaming compile failed: "+en),_("falling back to ArrayBuffer instantiation"),De(xe)})}):De(xe)}if(a.instantiateWasm)try{var yt=a.instantiateWasm(X,se);return yt}catch(tt){return _("Module.instantiateWasm callback failed with error: "+tt),!1}return ft().catch(i),{}}var H4,j4;function Bd(X){for(;X.length>0;){var se=X.shift();if(typeof se=="function"){se(a);continue}var xe=se.func;typeof xe=="number"?se.arg===void 0?Wd(xe)():Wd(xe)(se.arg):xe(se.arg===void 0?null:se.arg)}}function za(X){return X}function hf(X){var se=/\b_Z[\w\d_]+/g;return X.replace(se,function(xe){var De=xe;return xe===De?xe:De+" ["+xe+"]"})}var xr=[];function Wd(X){var se=xr[X];return se||(X>=xr.length&&(xr.length=X+1),xr[X]=se=Oa.get(X)),se}function q4(){var X=new Error;if(!X.stack){try{throw new Error}catch(se){X=se}if(!X.stack)return"(no stack trace available)"}return X.stack.toString()}function wu(X,se){Oa.set(X,se),xr[X]=se}function I1(){Ci("")}function Vd(){return 2147483648}function ff(X,se,xe){Hn.copyWithin(X,se,se+xe)}function Kn(X){try{return Q.grow(X-bs.byteLength+65535>>>16),aa(Q.buffer),1}catch(se){}}function mf(X){var se=Hn.length;X=X>>>0;var xe=Vd();if(X>xe)return!1;for(var De=1;De<=4;De*=2){var ft=se*(1+.2/De);ft=Math.min(ft,X+100663296);var yt=Math.min(xe,Qt(Math.max(X,ft),65536)),tt=Kn(yt);if(tt)return!0}return!1}var ku={mappings:{},buffers:[null,[],[]],printChar:function(X,se){var xe=ku.buffers[X];se===0||se===10?((X===1?$:_)(ot(xe,0)),xe.length=0):xe.push(se)},varargs:void 0,get:function(){ku.varargs+=4;var X=Pn[ku.varargs-4>>2];return X},getStr:function(X){var se=pt(X);return se},get64:function(X,se){return X}};function S1(X){return 0}function X4(X,se,xe,De,ft){}function K4(X,se,xe,De){for(var ft=0,yt=0;yt<xe;yt++){var tt=Pn[se>>2],Ze=Pn[se+4>>2];se+=8;for(var en=0;en<Ze;en++)ku.printChar(X,Hn[tt+en]);ft+=Ze}return Pn[De>>2]=ft,0}function C1(X){te(X)}var gf=!1,Iu={abort:I1,emscripten_get_heap_max:Vd,emscripten_memcpy_big:ff,emscripten_resize_heap:mf,fd_close:S1,fd_seek:X4,fd_write:K4,setTempRet0:C1},aR=k1(),Z4=a.___wasm_call_ctors=function(){return(Z4=a.___wasm_call_ctors=a.asm.__wasm_call_ctors).apply(null,arguments)},T1=a._init=function(){return(T1=a._init=a.asm.init).apply(null,arguments)},N1=a._init_with_threads_count=function(){return(N1=a._init_with_threads_count=a.asm.init_with_threads_count).apply(null,arguments)},yf=a._get_threads_count=function(){return(yf=a._get_threads_count=a.asm.get_threads_count).apply(null,arguments)},Af=a._register_tensor=function(){return(Af=a._register_tensor=a.asm.register_tensor).apply(null,arguments)},E1=a._dispose_data=function(){return(E1=a._dispose_data=a.asm.dispose_data).apply(null,arguments)},ze=a._dispose=function(){return(ze=a._dispose=a.asm.dispose).apply(null,arguments)},R1=a._Abs=function(){return(R1=a._Abs=a.asm.Abs).apply(null,arguments)},xf=a._Add=function(){return(xf=a._Add=a.asm.Add).apply(null,arguments)},Ni=a._AddN=function(){return(Ni=a._AddN=a.asm.AddN).apply(null,arguments)},Su=a._All=function(){return(Su=a._All=a.asm.All).apply(null,arguments)},_1=a._Any=function(){return(_1=a._Any=a.asm.Any).apply(null,arguments)},Y4=a._ArgMax=function(){return(Y4=a._ArgMax=a.asm.ArgMax).apply(null,arguments)},D1=a._AvgPool=function(){return(D1=a._AvgPool=a.asm.AvgPool).apply(null,arguments)},J4=a._BatchMatMul=function(){return(J4=a._BatchMatMul=a.asm.BatchMatMul).apply(null,arguments)},Ei=a._Ceil=function(){return(Ei=a._Ceil=a.asm.Ceil).apply(null,arguments)},$1=a._ClipByValue=function(){return($1=a._ClipByValue=a.asm.ClipByValue).apply(null,arguments)},P1=a._Conv2D=function(){return(P1=a._Conv2D=a.asm.Conv2D).apply(null,arguments)},F1=a._Conv2DBackpropInput=function(){return(F1=a._Conv2DBackpropInput=a.asm.Conv2DBackpropInput).apply(null,arguments)},O1=a._Cos=function(){return(O1=a._Cos=a.asm.Cos).apply(null,arguments)},M1=a._Cosh=function(){return(M1=a._Cosh=a.asm.Cosh).apply(null,arguments)},z1=a._CropAndResize=function(){return(z1=a._CropAndResize=a.asm.CropAndResize).apply(null,arguments)},bf=a._Cumprod=function(){return(bf=a._Cumprod=a.asm.Cumprod).apply(null,arguments)},L1=a._Cumsum=function(){return(L1=a._Cumsum=a.asm.Cumsum).apply(null,arguments)},B1=a._DepthToSpace=function(){return(B1=a._DepthToSpace=a.asm.DepthToSpace).apply(null,arguments)},W1=a._DepthwiseConv2dNative=function(){return(W1=a._DepthwiseConv2dNative=a.asm.DepthwiseConv2dNative).apply(null,arguments)},V1=a._Elu=function(){return(V1=a._Elu=a.asm.Elu).apply(null,arguments)},U1=a._Equal=function(){return(U1=a._Equal=a.asm.Equal).apply(null,arguments)},vf=a._Exp=function(){return(vf=a._Exp=a.asm.Exp).apply(null,arguments)},G1=a._FlipLeftRight=function(){return(G1=a._FlipLeftRight=a.asm.FlipLeftRight).apply(null,arguments)},H1=a._Floor=function(){return(H1=a._Floor=a.asm.Floor).apply(null,arguments)},Ri=a._FloorDiv=function(){return(Ri=a._FloorDiv=a.asm.FloorDiv).apply(null,arguments)},Ud=a._FusedBatchNorm=function(){return(Ud=a._FusedBatchNorm=a.asm.FusedBatchNorm).apply(null,arguments)},j1=a._FusedConv2D=function(){return(j1=a._FusedConv2D=a.asm.FusedConv2D).apply(null,arguments)},q1=a._FusedDepthwiseConv2D=function(){return(q1=a._FusedDepthwiseConv2D=a.asm.FusedDepthwiseConv2D).apply(null,arguments)},X1=a._Gather=function(){return(X1=a._Gather=a.asm.Gather).apply(null,arguments)},Je=a._GatherNd=function(){return(Je=a._GatherNd=a.asm.GatherNd).apply(null,arguments)},K1=a._Greater=function(){return(K1=a._Greater=a.asm.Greater).apply(null,arguments)},Z1=a._GreaterEqual=function(){return(Z1=a._GreaterEqual=a.asm.GreaterEqual).apply(null,arguments)},Y1=a._LeakyRelu=function(){return(Y1=a._LeakyRelu=a.asm.LeakyRelu).apply(null,arguments)},J1=a._Less=function(){return(J1=a._Less=a.asm.Less).apply(null,arguments)},Q1=a._LessEqual=function(){return(Q1=a._LessEqual=a.asm.LessEqual).apply(null,arguments)},eg=a._Log=function(){return(eg=a._Log=a.asm.Log).apply(null,arguments)},Gd=a._LogicalAnd=function(){return(Gd=a._LogicalAnd=a.asm.LogicalAnd).apply(null,arguments)},wf=a._LogicalNot=function(){return(wf=a._LogicalNot=a.asm.LogicalNot).apply(null,arguments)},kf=a._LogicalOr=function(){return(kf=a._LogicalOr=a.asm.LogicalOr).apply(null,arguments)},tg=a._LogicalXor=function(){return(tg=a._LogicalXor=a.asm.LogicalXor).apply(null,arguments)},ng=a._Max=function(){return(ng=a._Max=a.asm.Max).apply(null,arguments)},sg=a._MaxPool=function(){return(sg=a._MaxPool=a.asm.MaxPool).apply(null,arguments)},rg=a._Maximum=function(){return(rg=a._Maximum=a.asm.Maximum).apply(null,arguments)},ag=a._Mean=function(){return(ag=a._Mean=a.asm.Mean).apply(null,arguments)},og=a._Min=function(){return(og=a._Min=a.asm.Min).apply(null,arguments)},zt=a._Minimum=function(){return(zt=a._Minimum=a.asm.Minimum).apply(null,arguments)},ig=a._MirrorPad=function(){return(ig=a._MirrorPad=a.asm.MirrorPad).apply(null,arguments)},lg=a._Multiply=function(){return(lg=a._Multiply=a.asm.Multiply).apply(null,arguments)},ug=a._Neg=function(){return(ug=a._Neg=a.asm.Neg).apply(null,arguments)},Cu=a._NonMaxSuppressionV3=function(){return(Cu=a._NonMaxSuppressionV3=a.asm.NonMaxSuppressionV3).apply(null,arguments)},If=a._NonMaxSuppressionV4=function(){return(If=a._NonMaxSuppressionV4=a.asm.NonMaxSuppressionV4).apply(null,arguments)},Sf=a._NonMaxSuppressionV5=function(){return(Sf=a._NonMaxSuppressionV5=a.asm.NonMaxSuppressionV5).apply(null,arguments)},Cf=a._NotEqual=function(){return(Cf=a._NotEqual=a.asm.NotEqual).apply(null,arguments)},cg=a._OneHot=function(){return(cg=a._OneHot=a.asm.OneHot).apply(null,arguments)},Tf=a._PadV2=function(){return(Tf=a._PadV2=a.asm.PadV2).apply(null,arguments)},dg=a._Pow=function(){return(dg=a._Pow=a.asm.Pow).apply(null,arguments)},Q4=a._Prelu=function(){return(Q4=a._Prelu=a.asm.Prelu).apply(null,arguments)},Nf=a._Prod=function(){return(Nf=a._Prod=a.asm.Prod).apply(null,arguments)},ev=a._RealDiv=function(){return(ev=a._RealDiv=a.asm.RealDiv).apply(null,arguments)},pg=a._Relu=function(){return(pg=a._Relu=a.asm.Relu).apply(null,arguments)},hg=a._Relu6=function(){return(hg=a._Relu6=a.asm.Relu6).apply(null,arguments)},fg=a._ResizeBilinear=function(){return(fg=a._ResizeBilinear=a.asm.ResizeBilinear).apply(null,arguments)},mg=a._ResizeNearestNeighbor=function(){return(mg=a._ResizeNearestNeighbor=a.asm.ResizeNearestNeighbor).apply(null,arguments)},gg=a._Reverse=function(){return(gg=a._Reverse=a.asm.Reverse).apply(null,arguments)},yg=a._RotateWithOffset=function(){return(yg=a._RotateWithOffset=a.asm.RotateWithOffset).apply(null,arguments)},Ag=a._Round=function(){return(Ag=a._Round=a.asm.Round).apply(null,arguments)},xg=a._Rsqrt=function(){return(xg=a._Rsqrt=a.asm.Rsqrt).apply(null,arguments)},bg=a._ScatterNd=function(){return(bg=a._ScatterNd=a.asm.ScatterNd).apply(null,arguments)},vg=a._SelectV2=function(){return(vg=a._SelectV2=a.asm.SelectV2).apply(null,arguments)},wg=a._Sigmoid=function(){return(wg=a._Sigmoid=a.asm.Sigmoid).apply(null,arguments)},kg=a._Sin=function(){return(kg=a._Sin=a.asm.Sin).apply(null,arguments)},Ig=a._Softmax=function(){return(Ig=a._Softmax=a.asm.Softmax).apply(null,arguments)},Sg=a._SparseFillEmptyRows=function(){return(Sg=a._SparseFillEmptyRows=a.asm.SparseFillEmptyRows).apply(null,arguments)},Cg=a._SparseReshape=function(){return(Cg=a._SparseReshape=a.asm.SparseReshape).apply(null,arguments)},Tg=a._SparseSegmentReduction=function(){return(Tg=a._SparseSegmentReduction=a.asm.SparseSegmentReduction).apply(null,arguments)},Ng=a._Sqrt=function(){return(Ng=a._Sqrt=a.asm.Sqrt).apply(null,arguments)},Eg=a._Square=function(){return(Eg=a._Square=a.asm.Square).apply(null,arguments)},Rg=a._SquaredDifference=function(){return(Rg=a._SquaredDifference=a.asm.SquaredDifference).apply(null,arguments)},_g=a._Step=function(){return(_g=a._Step=a.asm.Step).apply(null,arguments)},Dg=a._StridedSlice=function(){return(Dg=a._StridedSlice=a.asm.StridedSlice).apply(null,arguments)},$g=a._Sub=function(){return($g=a._Sub=a.asm.Sub).apply(null,arguments)},Pg=a._Sum=function(){return(Pg=a._Sum=a.asm.Sum).apply(null,arguments)},Fg=a._Tan=function(){return(Fg=a._Tan=a.asm.Tan).apply(null,arguments)},Og=a._Tanh=function(){return(Og=a._Tanh=a.asm.Tanh).apply(null,arguments)},Mg=a._Tile=function(){return(Mg=a._Tile=a.asm.Tile).apply(null,arguments)},zg=a._TopK=function(){return(zg=a._TopK=a.asm.TopK).apply(null,arguments)},Lg=a._Transform=function(){return(Lg=a._Transform=a.asm.Transform).apply(null,arguments)},Bg=a._Transpose=function(){return(Bg=a._Transpose=a.asm.Transpose).apply(null,arguments)},Wg=a.__FusedMatMul=function(){return(Wg=a.__FusedMatMul=a.asm._FusedMatMul).apply(null,arguments)},Vg=a._malloc=function(){return(Vg=a._malloc=a.asm.malloc).apply(null,arguments)},Ug=a._free=function(){return(Ug=a._free=a.asm.free).apply(null,arguments)},Gg=a.___errno_location=function(){return(Gg=a.___errno_location=a.asm.__errno_location).apply(null,arguments)},Hg=a._emscripten_main_thread_process_queued_calls=function(){return(Hg=a._emscripten_main_thread_process_queued_calls=a.asm.emscripten_main_thread_process_queued_calls).apply(null,arguments)},Ef=a.stackSave=function(){return(Ef=a.stackSave=a.asm.stackSave).apply(null,arguments)},Rf=a.stackRestore=function(){return(Rf=a.stackRestore=a.asm.stackRestore).apply(null,arguments)},Hd=a.stackAlloc=function(){return(Hd=a.stackAlloc=a.asm.stackAlloc).apply(null,arguments)},jg=a.dynCall_iijjiiii=function(){return(jg=a.dynCall_iijjiiii=a.asm.dynCall_iijjiiii).apply(null,arguments)},qg=a.dynCall_jiji=function(){return(qg=a.dynCall_jiji=a.asm.dynCall_jiji).apply(null,arguments)};a.cwrap=_e;var Tu;function jd(X){this.name="ExitStatus",this.message="Program terminated with exit("+X+")",this.status=X}Ma=function X(){Tu||qd(),Tu||(Ma=X)};function qd(X){if(X=X||c,Gs>0||(af(),Gs>0))return;function se(){Tu||(Tu=!0,a.calledRun=!0,!ae&&(of(),o(a),a.onRuntimeInitialized&&a.onRuntimeInitialized(),lf()))}a.setStatus?(a.setStatus("Running..."),setTimeout(function(){setTimeout(function(){a.setStatus("")},1),se()},1)):se()}a.run=qd;function tv(X){ie=X,zd()||(a.onExit&&a.onExit(X),ae=!0),d(X,new jd(X))}if(a.preInit)for(typeof a.preInit=="function"&&(a.preInit=[a.preInit]);a.preInit.length>0;)a.preInit.pop()();qd();var Nu;l&&(Nu={uncaughtException:process.listeners("uncaughtException").filter(function(X){return!l.uncaughtException.indexOf(X)>-1}),unhandledRejection:process.listeners("unhandledRejection").filter(function(X){return!l.unhandledRejection.indexOf(X)>-1})});var Eu;if(typeof r!="undefined")Eu=r;else if(typeof WasmBackendModuleThreadedSimd!="undefined")Eu=WasmBackendModuleThreadedSimd;else throw new Error("Could not find wasm module in post.js");if(Nu){var Xg=Eu._dispose;Eu._dispose=function(){Xg(),Nu.uncaughtException.forEach(function(X){process.removeListener("uncaughtException",X)}),Nu.unhandledRejection.forEach(function(X){process.removeListener("unhandledRejection",X)})}}return r.ready}})();typeof e=="object"&&typeof t=="object"?t.exports=n:typeof define=="function"&&define.amd?define([],function(){return n}):typeof e=="object"&&(e.WasmBackendModule=n)}}),C_=1e-7,T_=1e-4,Mp=class{constructor(e,t){this.backend=e,this.dataMover=t,this.data=new WeakMap,this.dataIdsCount=0}get(e){return this.data.has(e)||this.dataMover.moveData(this.backend,e),this.data.get(e)}set(e,t){this.dataIdsCount++,this.data.set(e,t)}has(e){return this.data.has(e)}delete(e){return this.dataIdsCount--,this.data.delete(e)}numDataIds(){return this.dataIdsCount}},cc=class{refCount(e){return Hs("refCount")}incRef(e){return Hs("incRef")}timerAvailable(){return!0}time(e){return Hs("time")}read(e){return Hs("read")}readSync(e){return Hs("readSync")}readToGPU(e,t){return Hs("readToGPU")}numDataIds(){return Hs("numDataIds")}disposeData(e,t){return Hs("disposeData")}write(e,t,n){return Hs("write")}move(e,t,n,s,r){return Hs("move")}memory(){return Hs("memory")}floatPrecision(){return Hs("floatPrecision")}epsilon(){return this.floatPrecision()===32?C_:T_}dispose(){return Hs("dispose")}};function Hs(e){throw new Error(`'${e}' not yet implemented or not found in the registry. This kernel may not be supported by the tfjs backend you have chosen`)}function t6(e){let t=e.length,n=0;for(;t>0;)n=Math.random()*t|0,t--,cm(e,t,n)}function N_(e,t){if(e.length!==t.length)throw new Error(`Array sizes must match to be shuffled together First array length was ${e.length}Second array length was ${t.length}`);let n=e.length,s=0;for(;n>0;)s=Math.random()*n|0,n--,cm(e,n,s),cm(t,n,s)}function xp(e,t,n){return Math.max(e,Math.min(t,n))}function E_(e){return e%2===0?e:e+1}function cm(e,t,n){let s=e[t];e[t]=e[n],e[n]=s}function R_(e){let t=0;for(let n=0;n<e.length;n++)t+=e[n];return t}function __(e,t){let n=Math.random();return t*n+(1-n)*e}function D_(e,t){let n=0;for(let s=0;s<e.length;s++){let r=Number(e[s])-Number(t[s]);n+=r*r}return n}function M(e,t){if(!e)throw new Error(typeof t=="string"?t:t())}function ns(e,t,n=""){M(co(e,t),()=>n+` Shapes ${e} and ${t} must match`)}function ol(e){M(e!=null,()=>"The input to the tensor constructor must be a non-null value.")}function Xi(e,t=[],n=!1){if(t==null&&(t=[]),Array.isArray(e)||Fn(e)&&!n)for(let s=0;s<e.length;++s)Xi(e[s],t,n);else t.push(e);return t}function Et(e){if(e.length===0)return 1;let t=e[0];for(let n=1;n<e.length;n++)t*=e[n];return t}function $_(e){return e.length===0}function co(e,t){if(e===t)return!0;if(e==null||t==null||e.length!==t.length)return!1;for(let n=0;n<e.length;n++)if(e[n]!==t[n])return!1;return!0}function qu(e){return e%1===0}function P_(e){if(Math.tanh!=null)return Math.tanh(e);if(e===1/0)return 1;if(e===-1/0)return-1;{let t=Math.exp(2*e);return(t-1)/(t+1)}}function F_(e){let t=Math.ceil(Math.sqrt(e));return[t,Math.ceil(e/t)]}function O_(e){let t=new Uint32Array(e);for(let n=0;n<e;++n)t[n]=n;return t6(t),t}function fp(e,t){return t<=e.length?e:e+" ".repeat(t-e.length)}function M_(e,t=s=>0,n){return new Promise((s,r)=>{let a=0,o=()=>{if(e()){s();return}a++;let i=t(a);if(n!=null&&a>=n){r();return}setTimeout(o,i)};o()})}function z_(e,t){let n=1,s=-1;for(let a=0;a<e.length;++a)if(e[a]>=0)n*=e[a];else if(e[a]===-1){if(s!==-1)throw Error(`Shapes can only have 1 implicit size. Found -1 at dim ${s} and dim ${a}`);s=a}else if(e[a]<0)throw Error(`Shapes can not be < 0. Found ${e[a]} at dim ${a}`);if(s===-1){if(t>0&&t!==n)throw Error(`Size(${t}) must match the product of shape ${e}`);return e}if(n===0)throw Error(`Cannot infer the missing size in [${e}] when there are 0 elements`);if(t%n!==0)throw Error(`The implicit shape can't be a fractional number. Got ${t} / ${n}`);let r=e.slice();return r[s]=t/n,r}function cr(e,t){let n=t.length;return e=e==null?t.map((s,r)=>r):[].concat(e),M(e.every(s=>s>=-n&&s<n),()=>`All values in axis param must be in range [-${n}, ${n}) but got axis ${e}`),M(e.every(s=>qu(s)),()=>`All values in axis param must be integers but got axis ${e}`),e.map(s=>s<0?n+s:s)}function n6(e,t){let n=[],s=[],r=t!=null&&Array.isArray(t)&&t.length===0,a=t==null||r?null:cr(t,e).sort(),o=0;for(let i=0;i<e.length;++i){if(a!=null){if(a[o]===i&&e[i]!==1)throw new Error(`Can't squeeze axis ${i} since its dim '${e[i]}' is not 1`);(a[o]==null||a[o]>i)&&e[i]===1&&(n.push(e[i]),s.push(i)),a[o]<=i&&o++}e[i]!==1&&(n.push(e[i]),s.push(i))}return{newShape:n,keptDims:s}}function s6(e,t){let n=null;if(e==null||e==="float32")n=new Float32Array(t);else if(e==="int32")n=new Int32Array(t);else if(e==="bool")n=new Uint8Array(t);else throw new Error(`Unknown data type ${e}`);return n}function r6(e,t){let n=null;if(e==null||e==="float32")n=new Float32Array(t);else if(e==="int32")n=new Int32Array(t);else if(e==="bool")n=new Uint8Array(t);else if(e==="string")n=new Array(t);else throw new Error(`Unknown data type ${e}`);return n}function a6(e,t){for(let n=0;n<e.length;n++){let s=e[n];if(isNaN(s)||!isFinite(s))throw Error(`A tensor of type ${t} being uploaded contains ${s}.`)}}function o6(e){return e==="bool"||e==="complex64"||e==="float32"||e==="int32"||e==="string"}function L_(e,t){return!(t==="complex64"||t==="float32"&&e!=="complex64"||t==="int32"&&e!=="float32"&&e!=="complex64"||t==="bool"&&e==="bool")}function Fn(e){return e instanceof Float32Array||e instanceof Int32Array||e instanceof Uint8Array||e instanceof Uint8ClampedArray}function m3(e){if(e==="float32"||e==="int32")return 4;if(e==="complex64")return 8;if(e==="bool")return 1;throw new Error(`Unknown dtype ${e}`)}function i6(e){if(e==null)return 0;let t=0;return e.forEach(n=>t+=n.length),t}function Ga(e){return typeof e=="string"||e instanceof String}function l6(e){return typeof e=="boolean"}function u6(e){return typeof e=="number"}function Vm(e){return Array.isArray(e)?Vm(e[0]):e instanceof Float32Array?"float32":e instanceof Int32Array||e instanceof Uint8Array||e instanceof Uint8ClampedArray?"int32":u6(e)?"float32":Ga(e)?"string":l6(e)?"bool":"float32"}function Ya(e){return!!(e&&e.constructor&&e.call&&e.apply)}function dm(e,t){for(let n=t;n<e;++n)if(e%n===0)return n;return e}function dc(e){let t=e.length;if(t<2)return[];let n=new Array(t-1);n[t-2]=e[t-1];for(let s=t-3;s>=0;--s)n[s]=n[s+1]*e[s+1];return n}function c6(e,t,n,s=!1){let r=new Array;if(t.length===1){let a=t[0]*(s?2:1);for(let o=0;o<a;o++)r[o]=n[e+o]}else{let a=t[0],o=t.slice(1),i=o.reduce((l,u)=>l*u)*(s?2:1);for(let l=0;l<a;l++)r[l]=c6(e+l*i,o,n,s)}return r}function Wu(e,t,n=!1){if(e.length===0)return t[0];let s=e.reduce((r,a)=>r*a)*(n?2:1);if(s===0)return[];if(s!==t.length)throw new Error(`[${e}] does not match the input size ${t.length}${n?" for a complex tensor":""}.`);return c6(0,e,t,n)}function by(e,t){let n=Um(e,t);for(let s=0;s<n.length;s++)n[s]=1;return n}function Um(e,t){if(t==null||t==="float32"||t==="complex64")return new Float32Array(e);if(t==="int32")return new Int32Array(e);if(t==="bool")return new Uint8Array(e);throw new Error(`Unknown data type ${t}`)}function B_(e,t){let n=e.reduce((s,r)=>s*r,1);if(t==null||t==="float32")return Wu(e,new Float32Array(n));if(t==="int32")return Wu(e,new Int32Array(n));if(t==="bool")return Wu(e,new Uint8Array(n));throw new Error(`Unknown data type ${t}`)}function vy(e){e.forEach(t=>{M(Number.isInteger(t)&&t>=0,()=>`Tensor must have a shape comprised of positive integers but got shape [${e}].`)})}function W_(e,t,n){if(t===0)return 0;if(t===1)return e[0];let s=e[e.length-1];for(let r=0;r<e.length-1;++r)s+=n[r]*e[r];return s}function V_(e,t,n){if(t===0)return[];if(t===1)return[e];let s=new Array(t);for(let r=0;r<s.length-1;++r)s[r]=Math.floor(e/n[r]),e-=s[r]*n[r];return s[s.length-1]=e,s}function wy(e){return e&&e.then&&typeof e.then=="function"}var hv="tfjsflags",d6=class{constructor(e){this.global=e,this.flags={},this.flagRegistry={},this.urlFlags={},this.getQueryParams=U_,this.populateURLFlags()}setPlatform(e,t){this.platform!=null&&(q().getBool("IS_TEST")||q().getBool("PROD")||console.warn(`Platform ${this.platformName} has already been set. Overwriting the platform with ${e}.`)),this.platformName=e,this.platform=t}registerFlag(e,t,n){if(this.flagRegistry[e]={evaluationFn:t,setHook:n},this.urlFlags[e]!=null){let s=this.urlFlags[e];q().getBool("IS_TEST")||q().getBool("PROD")||console.warn(`Setting feature override from URL ${e}: ${s}.`),this.set(e,s)}}async getAsync(e){return e in this.flags?this.flags[e]:(this.flags[e]=await this.evaluateFlag(e),this.flags[e])}get(e){if(e in this.flags)return this.flags[e];let t=this.evaluateFlag(e);if(wy(t))throw new Error(`Flag ${e} cannot be synchronously evaluated. Please use getAsync() instead.`);return this.flags[e]=t,this.flags[e]}getNumber(e){return this.get(e)}getBool(e){return this.get(e)}getFlags(){return this.flags}get features(){return this.flags}set(e,t){if(this.flagRegistry[e]==null)throw new Error(`Cannot set flag ${e} as it has not been registered.`);this.flags[e]=t,this.flagRegistry[e].setHook!=null&&this.flagRegistry[e].setHook(t)}evaluateFlag(e){if(this.flagRegistry[e]==null)throw new Error(`Cannot evaluate flag '${e}': no evaluation function found.`);return this.flagRegistry[e].evaluationFn()}setFlags(e){this.flags=Object.assign({},e)}reset(){this.flags={},this.urlFlags={},this.populateURLFlags()}populateURLFlags(){if(typeof this.global=="undefined"||typeof this.global.location=="undefined"||typeof this.global.location.search=="undefined")return;let e=this.getQueryParams(this.global.location.search);hv in e&&e[hv].split(",").forEach(n=>{let[s,r]=n.split(":");this.urlFlags[s]=H_(s,r)})}};function U_(e){let t={};return e.replace(/[?&]([^=?&]+)(?:=([^&]*))?/g,(n,...s)=>(G_(t,s[0],s[1]),s.join("="))),t}function G_(e,t,n){e[decodeURIComponent(t)]=decodeURIComponent(n||"")}function H_(e,t){if(t=t.toLowerCase(),t==="true"||t==="false")return t==="true";if(`${+t}`===t)return+t;throw new Error(`Could not parse value flag value ${t} for flag ${e}.`)}function q(){return ky}var ky=null;function j_(e){ky=e}var n3;function p6(){if(n3==null){let e;if(typeof window!="undefined")e=window;else if(typeof global!="undefined")e=global;else if(typeof process!="undefined")e=process;else if(typeof self!="undefined")e=self;else throw new Error("Could not find a global object");n3=e}return n3}function q_(){let e=p6();return e._tfGlobals==null&&(e._tfGlobals=new Map),e._tfGlobals}function Iy(e,t){let n=q_();if(n.has(e))return n.get(e);{let s=t();return n.set(e,s),n.get(e)}}var il="Abs",pc="Acos",hc="Acosh",xa="Add",po="AddN",fc="All",mc="Any",ho="ArgMax",gc="ArgMin",yc="Asin",Ac="Asinh",xc="Atan",bc="Atanh",vc="Atan2",fo="AvgPool",Gm="AvgPoolGrad",zp="AvgPool3D",Hm="AvgPool3DGrad",mo="BatchMatMul",ll="BatchToSpaceND",jm="Bincount",h6="BroadcastTo",qm="BroadcastArgs",go="Cast",yo="Ceil",ba="ClipByValue",Lp="Complex",Bp="ComplexAbs",ul="Concat",Ao="Conv2D",Xm="Conv2DBackpropFilter",xo="Conv2DBackpropInput",Wp="Conv3D",Km="Conv3DBackpropFilterV2",Zm="Conv3DBackpropInputV2",bo="Cos",vo="Cosh",cl="Cumprod",wo="Cumsum",dl="CropAndResize",Ym="DenseBincount",pl="DepthToSpace",ko="DepthwiseConv2dNative",Jm="DepthwiseConv2dNativeBackpropFilter",Qm="DepthwiseConv2dNativeBackpropInput",e0="Diag",Vp="Dilation2D",pm="Dilation2DBackpropInput",hm="Dilation2DBackpropFilter",Io="RealDiv",Up="Einsum",So="Elu",t0="EluGrad",wc="Erf",hl="Equal",Co="Exp",fl="ExpandDims",ml="Expm1",n0="FFT",kc="Fill",gl="FlipLeftRight",To="Floor",No="FloorDiv",Eo="FusedBatchNorm",yl="GatherV2",Al="GatherNd",xl="Greater",Ro="GreaterEqual",_o="Identity",s0="IFFT",Gp="Imag",Ic="IsFinite",Sc="IsInf",Cc="IsNan",Do="LeakyRelu",bl="Less",vl="LessEqual",r0="LinSpace",$o="Log",Tc="Log1p",wl="LogicalAnd",kl="LogicalNot",Nc="LogicalOr",f6="LogicalXor",m6="LogSoftmax",X_="LowerBound",Hp="LRN",a0="LRNGrad",Po="Max",Fo="Maximum",Oo="MaxPool",o0="MaxPoolGrad",jp="MaxPool3D",i0="MaxPool3DGrad",l0="MaxPoolWithArgmax",Mo="Mean",zo="Min",Lo="Minimum",Bo="MirrorPad",Ec="Mod",u0="Multinomial",Wo="Multiply",Il="Neg",Sl="NotEqual",Cl="NonMaxSuppressionV3",Rc="NonMaxSuppressionV4",Tl="NonMaxSuppressionV5",Nl="OnesLike",El="OneHot",Rl="Pack",Vo="PadV2",K_="Pool",Uo="Pow",Go="Prelu",Ho="Prod",_c="Range",qp="Real",Dc="Reciprocal",jo="Relu",_l="Reshape",qo="ResizeNearestNeighbor",c0="ResizeNearestNeighborGrad",Xo="ResizeBilinear",d0="ResizeBilinearGrad",Ko="Relu6",Dl="Reverse",$l="Round",Zo="Rsqrt",Pl="ScatterNd",p0="SearchSorted",Fl="Select",$c="Selu",Ol="Slice",Yo="Sin",Ml="Sinh",Pc="Sign",Jo="Sigmoid",Fc="Softplus",Qo="Sqrt",ei="Sum",zl="SpaceToBatchND",Ll="SplitV",ti="Softmax",Xp="SparseFillEmptyRows",Oc="SparseReshape",Kp="SparseSegmentMean",Zp="SparseSegmentSum",Yp="SparseToDense",ni="SquaredDifference",Mc="Square",Bl="StridedSlice",zc="StringNGrams",Jp="StringSplit",Qp="StringToHashBucketFast",si="Sub",Wl="Tan",ri="Tanh",va="Tile",Vl="TopK",Ul="Transform",Hr="Transpose",h0="Unique",Gl="Unpack",eh="UnsortedSegmentSum",Z_="UpperBound",Hl="ZerosLike",ai="Step",bp="FromPixels",jl="RotateWithOffset",Ja="_FusedMatMul",Qa="FusedConv2D",eo="FusedDepthwiseConv2D";function Ua(...e){q().getBool("IS_TEST")||q().getBool("PROD")||console.warn(...e)}function Y_(...e){q().getBool("IS_TEST")||q().getBool("PROD")||console.log(...e)}var Xu=Iy("kernelRegistry",()=>new Map),vp=Iy("gradRegistry",()=>new Map);function fm(e,t){let n=Sy(e,t);return Xu.get(n)}function g3(e){return vp.get(e)}function qr(e){let t=Xu.entries(),n=[];for(;;){let{done:s,value:r}=t.next();if(s)break;let[a,o]=r,[i]=a.split("_");i===e&&n.push(o)}return n}function dr(e){let{kernelName:t,backendName:n}=e,s=Sy(t,n);Xu.has(s)&&Ua(`The kernel '${t}' for backend '${n}' is already registered`),Xu.set(s,e)}function g6(e){let{kernelName:t}=e;vp.has(t)&&q().getBool("DEBUG")&&Ua(`Overriding the gradient for '${t}'`),vp.set(t,e)}function J_(e,t){let n=Sy(e,t);if(!Xu.has(n))throw new Error(`The kernel '${e}' for backend '${t}' is not registered`);Xu.delete(n)}function Q_(e){if(!vp.has(e))throw new Error(`The gradient '${e}' for backend is not registered`);vp.delete(e)}function eD(e,t){qr(e).forEach(s=>{let r=Object.assign({},s,{backendName:t});dr(r)})}function Sy(e,t){return`${t}_${e}`}var v={};Ve(v,{arraysEqual:()=>co,assert:()=>M,assertNonNegativeIntegerDimensions:()=>vy,assertNonNull:()=>ol,assertShapesMatch:()=>ns,bytesFromStringArray:()=>i6,bytesPerElement:()=>m3,checkConversionForErrors:()=>a6,clamp:()=>xp,computeStrides:()=>dc,createScalarValue:()=>oD,createShuffledIndices:()=>O_,decodeString:()=>mm,distSquared:()=>D_,encodeString:()=>nh,fetch:()=>lD,fingerPrint64:()=>aD,flatten:()=>Xi,getArrayFromDType:()=>r6,getTypedArrayFromDType:()=>s6,hasEncodingLoss:()=>L_,hexToLong:()=>th,indexToLoc:()=>V_,inferDtype:()=>Vm,inferFromImplicitShape:()=>z_,isBoolean:()=>l6,isFunction:()=>Ya,isInt:()=>qu,isNumber:()=>u6,isPromise:()=>wy,isScalarShape:()=>$_,isString:()=>Ga,isTypedArray:()=>Fn,isValidDtype:()=>o6,locToIndex:()=>W_,makeOnesTypedArray:()=>by,makeZerosNestedTypedArray:()=>B_,makeZerosTypedArray:()=>Um,nearestDivisor:()=>dm,nearestLargerEven:()=>E_,now:()=>wp,parseAxisParam:()=>cr,randUniform:()=>__,repeatedTry:()=>M_,rightPad:()=>fp,shuffle:()=>t6,shuffleCombo:()=>N_,sizeFromShape:()=>Et,sizeToSquarishShape:()=>F_,squeezeShape:()=>n6,sum:()=>R_,swap:()=>cm,tanh:()=>P_,toNestedArray:()=>Wu,toTypedArray:()=>f0});var fv=uo(u_()),Mi=fv.default||fv;function th(e){return Mi.fromString(e,!0,16)}var y6=th("c3a5c85c97cb3127"),Pi=th("b492b66fbe98f273"),Yn=th("9ae16a3b2f90404f");function y3(e){return e.xor(e.shru(47))}function A6(e,t,n){let s=e.slice(t,t+n);return Mi.fromBytes(Array.from(s),!0,!0)}function Nt(e,t){return A6(e,t,8)}function mv(e,t){return A6(e,t,4)}function xn(e,t){return t===0?e:e.shru(t).or(e.shl(64-t))}function qa(e,t,n=th("9ddfea08eb382d69")){let s=e.xor(t).mul(n);s=s.xor(s.shru(47));let r=t.xor(s).mul(n);return r=r.xor(r.shru(47)),r=r.mul(n),r}function tD(e,t,n,s,r,a){r=r.add(e),a=xn(a.add(r).add(s),21);let o=r;return r=r.add(t),r=r.add(n),a=a.add(xn(r,44)),[r.add(s),a.add(o)]}function Lf(e,t,n,s){return tD(Nt(e,t),Nt(e,t+8),Nt(e,t+16),Nt(e,t+24),n,s)}function nD(e,t=e.length){if(t>=8){let n=Yn.add(t*2),s=Nt(e,0).add(Yn),r=Nt(e,t-8),a=xn(r,37).mul(n).add(s),o=xn(s,25).add(r).mul(n);return qa(a,o,n)}if(t>=4){let n=Yn.add(t*2),s=mv(e,0);return qa(s.shl(3).add(t),mv(e,t-4),n)}if(t>0){let n=e[0],s=e[t>>1],r=e[t-1],a=n+(s<<8),o=t+(r<<2);return y3(Yn.mul(a).xor(y6.mul(o))).mul(Yn)}return Yn}function sD(e,t=e.length){let n=Yn.add(t*2),s=Nt(e,0).mul(Pi),r=Nt(e,8),a=Nt(e,t-8).mul(n),o=Nt(e,t-16).mul(Yn);return qa(xn(s.add(r),43).add(xn(a,30)).add(o),s.add(xn(r.add(Yn),18)).add(a),n)}function rD(e,t=e.length){let n=Yn.add(t*2),s=Nt(e,0).mul(Yn),r=Nt(e,8),a=Nt(e,t-8).mul(n),o=Nt(e,t-16).mul(Yn),i=xn(s.add(r),43).add(xn(a,30)).add(o),l=qa(i,s.add(xn(r.add(Yn),18)).add(a),n),u=Nt(e,16).mul(n),c=Nt(e,24),p=i.add(Nt(e,t-32)).mul(n),d=l.add(Nt(e,t-24)).mul(n);return qa(xn(u.add(c),43).add(xn(p,30)).add(d),u.add(xn(c.add(s),18)).add(p),n)}function aD(e,t=e.length){let n=Mi.fromNumber(81,!0);if(t<=32)return t<=16?nD(e,t):sD(e,t);if(t<=64)return rD(e,t);let s=n,r=n.mul(Pi).add(113),a=y3(r.mul(Yn).add(113)).mul(Yn),o=[Mi.UZERO,Mi.UZERO],i=[Mi.UZERO,Mi.UZERO];s=s.mul(Yn).add(Nt(e,0));let l=0,u=(t-1>>6)*64,c=u+(t-1&63)-63;do s=xn(s.add(r).add(o[0]).add(Nt(e,l+8)),37).mul(Pi),r=xn(r.add(o[1]).add(Nt(e,l+48)),42).mul(Pi),s=s.xor(i[1]),r=r.add(o[0]).add(Nt(e,l+40)),a=xn(a.add(i[0]),33).mul(Pi),o=Lf(e,l,o[1].mul(Pi),s.add(i[0])),i=Lf(e,l+32,a.add(i[1]),r.add(Nt(e,l+16))),[a,s]=[s,a],l+=64;while(l!==u);let p=Pi.add(a.and(255).shl(1));return l=c,i[0]=i[0].add(t-1&63),o[0]=o[0].add(i[0]),i[0]=i[0].add(o[0]),s=xn(s.add(r).add(o[0]).add(Nt(e,l+8)),37).mul(p),r=xn(r.add(o[1]).add(Nt(e,l+48)),42).mul(p),s=s.xor(i[1].mul(9)),r=r.add(o[0].mul(9).add(Nt(e,l+40))),a=xn(a.add(i[0]),33).mul(p),o=Lf(e,l,o[1].mul(p),s.add(i[0])),i=Lf(e,l+32,a.add(i[1]),r.add(Nt(e,l+16))),[a,s]=[s,a],qa(qa(o[0],i[0],p).add(y3(r).mul(y6)).add(a),qa(o[1],i[1],p).add(s),p)}function oD(e,t){return t==="string"?nh(e):f0([e],t)}function iD(e,t){return e instanceof Float32Array&&t==="float32"||e instanceof Int32Array&&t==="int32"||e instanceof Uint8Array&&t==="bool"}function f0(e,t){if(t==="string")throw new Error("Cannot convert a string[] to a TypedArray");if(Array.isArray(e)&&(e=Xi(e)),q().getBool("DEBUG")&&a6(e,t),iD(e,t))return e;if(t==null||t==="float32"||t==="complex64")return new Float32Array(e);if(t==="int32")return new Int32Array(e);if(t==="bool"){let n=new Uint8Array(e.length);for(let s=0;s<n.length;++s)Math.round(e[s])!==0&&(n[s]=1);return n}else throw new Error(`Unknown data type ${t}`)}function wp(){return q().platform.now()}function lD(e,t){return q().platform.fetch(e,t)}function nh(e,t="utf-8"){return t=t||"utf-8",q().platform.encode(e,t)}function mm(e,t="utf-8"){return t=t||"utf-8",q().platform.decode(e,t)}var uD=class{constructor(e,t){this.backendTimer=e,this.logger=t,t==null&&(this.logger=new dD)}profileKernel(e,t,n){let s,r=()=>{s=n()},a,o=wp();if(this.backendTimer.timerAvailable())a=this.backendTimer.time(r);else{r();for(let l of s)l.dataSync();a=Promise.resolve({kernelMs:wp()-o})}if(q().getBool("CHECK_COMPUTATION_FOR_ERRORS"))for(let l=0;l<s.length;l++){let u=s[l];u.data().then(c=>{cD(c,u.dtype,e)})}return{kernelName:e,outputs:s,inputs:t,timeMs:a.then(l=>l.kernelMs),extraInfo:a.then(l=>l.getExtraProfileInfo!=null?l.getExtraProfileInfo():"")}}logKernelProfile(e){let{kernelName:t,outputs:n,timeMs:s,inputs:r,extraInfo:a}=e;n.forEach(o=>{Promise.all([o.data(),s,a]).then(i=>{this.logger.logKernelProfile(t,o,i[0],i[1],r,i[2])})})}};function cD(e,t,n){if(t!=="float32")return!1;for(let s=0;s<e.length;s++){let r=e[s];if(isNaN(r)||!isFinite(r))return console.warn(`Found ${r} in the result of '${n}'`),!0}return!1}var dD=class{logKernelProfile(e,t,n,s,r,a){let o=typeof s=="number"?fp(`${s}ms`,9):s.error,i=fp(e,25),l=t.rank,u=t.size,c=fp(t.shape.toString(),14),p="";for(let d in r){let h=r[d];if(h!=null){let f=h.shape||t.shape,m=f.length;p+=`${d}: ${m}D ${m>0?f:""} `}}console.log(`%c${i} %c${o} %c${l}D ${c} %c${u} %c${p} %c${a}`,"font-weight:bold","color:red","color:blue","color: orange","color: green","color: steelblue")}};function pD(e,t,n){let s={},r={};for(let l=0;l<t.length;l++)s[t[l].id]=!0;for(let l=0;l<e.length;l++){let u=e[l],c=u.inputs;for(let p in c){let d=c[p],h=!1;for(let f=0;f<t.length;f++)if(s[d.id]){u.outputs.forEach(m=>s[m.id]=!0),h=!0,r[u.id]=!0;break}if(h)break}}let a={};a[n.id]=!0;let o={};for(let l=e.length-1;l>=0;l--){let u=e[l],c=u.inputs;for(let p=0;p<u.outputs.length;p++)if(a[u.outputs[p].id]){for(let d in c)a[c[d].id]=!0,o[u.id]=!0;break}}let i=[];for(let l=0;l<e.length;l++){let u=e[l];if(r[u.id]&&o[u.id]){let c={};for(let d in u.inputs){let h=u.inputs[d];s[h.id]&&(c[d]=h)}let p=Object.assign({},u);p.inputs=c,p.outputs=u.outputs,i.push(p)}}return i}function hD(e,t,n,s){for(let r=t.length-1;r>=0;r--){let a=t[r],o=[];if(a.outputs.forEach(l=>{let u=e[l.id];u!=null?o.push(u):o.push(null)}),a.gradient==null)throw new Error(`Cannot compute gradient: gradient function not found for ${a.kernelName}.`);let i=a.gradient(o);for(let l in a.inputs){if(!(l in i))throw new Error(`Cannot backprop through input ${l}. Available gradients found: ${Object.keys(i)}.`);let u=n(()=>i[l]());if(u.dtype!=="float32")throw new Error(`Error in gradient for op ${a.kernelName}. The gradient of input ${l} must have 'float32' dtype, but has '${u.dtype}'`);let c=a.inputs[l];if(!co(u.shape,c.shape))throw new Error(`Error in gradient for op ${a.kernelName}. The gradient of input '${l}' has shape '${u.shape}', which does not match the shape of the input '${c.shape}'`);if(e[c.id]==null)e[c.id]=u;else{let p=e[c.id];e[c.id]=s(p,u),p.dispose()}}}}var gv=20,Qd=3,s3=7;function fD(e,t,n,s){let r=dc(t),a=mD(e,t,n,r),o=t.length,i=Qf(e,t,n,r,a),l=["Tensor"];return s&&(l.push(` dtype: ${n}`),l.push(` rank: ${o}`),l.push(` shape: [${t}]`),l.push(" values:")),l.push(i.map(u=>" "+u).join(`
|
|
`)),l.join(`
|
|
`)}function mD(e,t,n,s){let r=Et(t),a=s[s.length-1],o=new Array(a).fill(0),i=t.length,l=n==="complex64"?rp(e):e;if(i>1)for(let u=0;u<r/a;u++){let c=u*a;for(let p=0;p<a;p++)o[p]=Math.max(o[p],sp(l[c+p],0,n).length)}return o}function sp(e,t,n){let s;return Array.isArray(e)?s=`${parseFloat(e[0].toFixed(s3))} + ${parseFloat(e[1].toFixed(s3))}j`:Ga(e)?s=`'${e}'`:n==="bool"?s=x6(e):s=parseFloat(e.toFixed(s3)).toString(),fp(s,t)}function x6(e){return e===0?"false":"true"}function Qf(e,t,n,s,r,a=!0){let o=n==="complex64"?2:1,i=t[0],l=t.length;if(l===0){if(n==="complex64"){let m=rp(e);return[sp(m[0],0,n)]}return n==="bool"?[x6(e[0])]:[e[0].toString()]}if(l===1){if(i>gv){let g=Qd*o,y=Array.from(e.slice(0,g)),x=Array.from(e.slice((i-Qd)*o,i*o));return n==="complex64"&&(y=rp(y),x=rp(x)),["["+y.map((A,b)=>sp(A,r[b],n)).join(", ")+", ..., "+x.map((A,b)=>sp(A,r[i-Qd+b],n)).join(", ")+"]"]}let m=n==="complex64"?rp(e):Array.from(e);return["["+m.map((g,y)=>sp(g,r[y],n)).join(", ")+"]"]}let u=t.slice(1),c=s.slice(1),p=s[0]*o,d=[];if(i>gv){for(let m=0;m<Qd;m++){let g=m*p,y=g+p;d.push(...Qf(e.slice(g,y),u,n,c,r,!1))}d.push("...");for(let m=i-Qd;m<i;m++){let g=m*p,y=g+p;d.push(...Qf(e.slice(g,y),u,n,c,r,m===i-1))}}else for(let m=0;m<i;m++){let g=m*p,y=g+p;d.push(...Qf(e.slice(g,y),u,n,c,r,m===i-1))}let h=l===2?",":"";d[0]="["+d[0]+h;for(let m=1;m<d.length-1;m++)d[m]=" "+d[m]+h;let f=`,
|
|
`;for(let m=2;m<l;m++)f+=`
|
|
`;return d[d.length-1]=" "+d[d.length-1]+"]"+(a?"":f),d}function rp(e){let t=[];for(let n=0;n<e.length;n+=2)t.push([e[n],e[n+1]]);return t}var fn=class{constructor(e,t,n){if(this.dtype=t,this.shape=e.slice(),this.size=Et(e),n!=null){let s=n.length;M(s===this.size,()=>`Length of values '${s}' does not match the size inferred by the shape '${this.size}'.`)}if(t==="complex64")throw new Error("complex64 dtype TensorBuffers are not supported. Please create a TensorBuffer for the real and imaginary parts separately and call tf.complex(real, imag).");this.values=n||r6(t,this.size),this.strides=dc(e)}set(e,...t){t.length===0&&(t=[0]),M(t.length===this.rank,()=>`The number of provided coordinates (${t.length}) must match the rank (${this.rank})`);let n=this.locToIndex(t);this.values[n]=e}get(...e){e.length===0&&(e=[0]);let t=0;for(let s of e){if(s<0||s>=this.shape[t]){let r=`Requested out of range element at ${e}. Buffer shape=${this.shape}`;throw new Error(r)}t++}let n=e[e.length-1];for(let s=0;s<e.length-1;++s)n+=this.strides[s]*e[s];return this.values[n]}locToIndex(e){if(this.rank===0)return 0;if(this.rank===1)return e[0];let t=e[e.length-1];for(let n=0;n<e.length-1;++n)t+=this.strides[n]*e[n];return t}indexToLoc(e){if(this.rank===0)return[];if(this.rank===1)return[e];let t=new Array(this.shape.length);for(let n=0;n<t.length-1;++n)t[n]=Math.floor(e/this.strides[n]),e-=t[n]*this.strides[n];return t[t.length-1]=e,t}get rank(){return this.shape.length}toTensor(){return Ir().makeTensor(this.values,this.shape,this.dtype)}},Ir=null,zu=null,gD=null;function yD(e){Ir=e}function AD(e){zu=e}function xD(e){gD=e}var nt=class{constructor(e,t,n,s){this.kept=!1,this.isDisposedInternal=!1,this.shape=e.slice(),this.dtype=t||"float32",this.size=Et(e),this.strides=dc(e),this.dataId=n,this.id=s,this.rankType=this.rank<5?this.rank.toString():"higher"}get rank(){return this.shape.length}async buffer(){let e=await this.data();return zu.buffer(this.shape,this.dtype,e)}bufferSync(){return zu.buffer(this.shape,this.dtype,this.dataSync())}async array(){let e=await this.data();return Wu(this.shape,e,this.dtype==="complex64")}arraySync(){return Wu(this.shape,this.dataSync(),this.dtype==="complex64")}async data(){this.throwIfDisposed();let e=Ir().read(this.dataId);if(this.dtype==="string"){let t=await e;try{return t.map(n=>mm(n))}catch(n){throw new Error("Failed to decode the string bytes into utf-8. To get the original bytes, call tensor.bytes().")}}return e}dataToGPU(e){return this.throwIfDisposed(),Ir().readToGPU(this.dataId,e)}dataSync(){this.throwIfDisposed();let e=Ir().readSync(this.dataId);if(this.dtype==="string")try{return e.map(t=>mm(t))}catch(t){throw new Error("Failed to decode the string bytes into utf-8. To get the original bytes, call tensor.bytes().")}return e}async bytes(){this.throwIfDisposed();let e=await Ir().read(this.dataId);return this.dtype==="string"?e:new Uint8Array(e.buffer)}dispose(){this.isDisposed||(Ir().disposeTensor(this),this.isDisposedInternal=!0)}get isDisposed(){return this.isDisposedInternal}throwIfDisposed(){if(this.isDisposed)throw new Error("Tensor is disposed.")}print(e=!1){return zu.print(this,e)}clone(){return this.throwIfDisposed(),zu.clone(this)}toString(e=!1){let t=this.dataSync();return fD(t,this.shape,this.dtype,e)}cast(e){return this.throwIfDisposed(),zu.cast(this,e)}variable(e=!0,t,n){return this.throwIfDisposed(),Ir().makeVariable(this,e,t,n)}};Object.defineProperty(nt,Symbol.hasInstance,{value:e=>!!e&&e.data!=null&&e.dataSync!=null&&e.throwIfDisposed!=null});function re(){return Iy("Tensor",()=>nt)}re();var kp=class extends nt{constructor(e,t,n,s){super(e.shape,e.dtype,e.dataId,s),this.trainable=t,this.name=n}assign(e){if(e.dtype!==this.dtype)throw new Error(`dtype of the new value (${e.dtype}) and previous value (${this.dtype}) must match`);if(!co(e.shape,this.shape))throw new Error(`shape of the new value (${e.shape}) and previous value (${this.shape}) must match`);Ir().disposeTensor(this),this.dataId=e.dataId,Ir().incRef(this,null)}dispose(){Ir().disposeVariable(this),this.isDisposedInternal=!0}};Object.defineProperty(kp,Symbol.hasInstance,{value:e=>e instanceof nt&&e.assign!=null&&e.assign instanceof Function});var Nr={};Ve(Nr,{assertTypesMatch:()=>b6,getTensorsInContainer:()=>Cy,isTensorInList:()=>vD,makeTypesMatch:()=>Gt});var A3;(function(e){e.R0="R0",e.R1="R1",e.R2="R2",e.R3="R3",e.R4="R4",e.R5="R5",e.R6="R6"})(A3||(A3={}));var x3;(function(e){e.float32="float32",e.int32="int32",e.bool="int32",e.complex64="complex64"})(x3||(x3={}));var b3;(function(e){e.float32="float32",e.int32="int32",e.bool="bool",e.complex64="complex64"})(b3||(b3={}));var v3;(function(e){e.float32="float32",e.int32="float32",e.bool="float32",e.complex64="complex64"})(v3||(v3={}));var w3;(function(e){e.float32="complex64",e.int32="complex64",e.bool="complex64",e.complex64="complex64"})(w3||(w3={}));var bD={float32:v3,int32:x3,bool:b3,complex64:w3};function Mn(e,t){if(e==="string"||t==="string"){if(e==="string"&&t==="string")return"string";throw new Error(`Can not upcast ${e} with ${t}`)}return bD[e][t]}function sh(e){return Mn(e,"int32")}function Gt(e,t){if(e.dtype===t.dtype)return[e,t];let n=Mn(e.dtype,t.dtype);return[e.cast(n),t.cast(n)]}function b6(e,t){M(e.dtype===t.dtype,()=>`The dtypes of the first(${e.dtype}) and second(${t.dtype}) input must match`)}function vD(e,t){return t.some(n=>n.id===e.id)}function Cy(e){let t=[];return v6(e,t,new Set),t}function v6(e,t,n){if(e==null)return;if(e instanceof nt){t.push(e);return}if(!wD(e))return;let s=e;for(let r in s){let a=s[r];n.has(a)||(n.add(a),v6(a,t,n))}}function wD(e){return Array.isArray(e)||typeof e=="object"}function r3(e){return e.kernelName!=null}var yv=class{constructor(){this.registeredVariables={},this.nextTapeNodeId=0,this.numBytes=0,this.numTensors=0,this.numStringTensors=0,this.numDataBuffers=0,this.gradientDepth=0,this.kernelDepth=0,this.scopeStack=[],this.numDataMovesStack=[],this.nextScopeId=0,this.tensorInfo=new WeakMap,this.profiling=!1,this.activeProfile={newBytes:0,newTensors:0,peakBytes:0,kernels:[],result:null,get kernelNames(){return Array.from(new Set(this.kernels.map(e=>e.name)))}}}dispose(){for(let e in this.registeredVariables)this.registeredVariables[e].dispose()}},Ip=class{constructor(e){this.ENV=e,this.registry={},this.registryFactory={},this.pendingBackendInitId=0,this.state=new yv}async ready(){if(this.pendingBackendInit!=null)return this.pendingBackendInit.then(()=>{});if(this.backendInstance!=null)return;let e=this.getSortedBackends();for(let t=0;t<e.length;t++){let n=e[t];if(await this.initializeBackend(n).success){await this.setBackend(n);return}}throw new Error("Could not initialize any backends, all backend initializations failed.")}get backend(){if(this.pendingBackendInit!=null)throw new Error(`Backend '${this.backendName}' has not yet been initialized. Make sure to await tf.ready() or await tf.setBackend() before calling other methods`);if(this.backendInstance==null){let{name:e,asyncInit:t}=this.initializeBackendsAndReturnBest();if(t)throw new Error(`The highest priority backend '${e}' has not yet been initialized. Make sure to await tf.ready() or await tf.setBackend() before calling other methods`);this.setBackend(e)}return this.backendInstance}backendNames(){return Object.keys(this.registryFactory)}findBackend(e){if(!(e in this.registry))if(e in this.registryFactory){let{asyncInit:t}=this.initializeBackend(e);if(t)return null}else return null;return this.registry[e]}findBackendFactory(e){return e in this.registryFactory?this.registryFactory[e].factory:null}registerBackend(e,t,n=1){return e in this.registryFactory?(Ua(`${e} backend was already registered. Reusing existing backend factory.`),!1):(this.registryFactory[e]={factory:t,priority:n},!0)}async setBackend(e){if(this.registryFactory[e]==null)throw new Error(`Backend name '${e}' not found in registry`);if(this.backendName=e,this.registry[e]==null){this.backendInstance=null;let{success:t,asyncInit:n}=this.initializeBackend(e);if(!(n?await t:t))return!1}return this.backendInstance=this.registry[e],this.setupRegisteredKernels(),this.profiler=new uD(this.backendInstance),!0}setupRegisteredKernels(){qr(this.backendName).forEach(t=>{t.setupFunc!=null&&t.setupFunc(this.backendInstance)})}disposeRegisteredKernels(e){qr(e).forEach(n=>{n.disposeFunc!=null&&n.disposeFunc(this.registry[e])})}initializeBackend(e){let t=this.registryFactory[e];if(t==null)throw new Error(`Cannot initialize backend ${e}, no registration found.`);try{let n=t.factory();if(n&&!(n instanceof cc)&&typeof n.then=="function"){let s=++this.pendingBackendInitId,r=n.then(a=>s<this.pendingBackendInitId?!1:(this.registry[e]=a,this.pendingBackendInit=null,!0)).catch(a=>(s<this.pendingBackendInitId||(this.pendingBackendInit=null,Ua(`Initialization of backend ${e} failed`),Ua(a.stack||a.message)),!1));return this.pendingBackendInit=r,{success:r,asyncInit:!0}}else return this.registry[e]=n,{success:!0,asyncInit:!1}}catch(n){return Ua(`Initialization of backend ${e} failed`),Ua(n.stack||n.message),{success:!1,asyncInit:!1}}}removeBackend(e){if(!(e in this.registryFactory))throw new Error(`${e} backend not found in registry`);this.backendName===e&&this.pendingBackendInit!=null&&this.pendingBackendInitId++,e in this.registry&&(this.disposeRegisteredKernels(e),this.registry[e].dispose(),delete this.registry[e]),delete this.registryFactory[e],this.backendName===e&&(this.pendingBackendInit=null,this.backendName=null,this.backendInstance=null)}getSortedBackends(){if(Object.keys(this.registryFactory).length===0)throw new Error("No backend found in registry.");return Object.keys(this.registryFactory).sort((e,t)=>this.registryFactory[t].priority-this.registryFactory[e].priority)}initializeBackendsAndReturnBest(){let e=this.getSortedBackends();for(let t=0;t<e.length;t++){let n=e[t],{success:s,asyncInit:r}=this.initializeBackend(n);if(r||s)return{name:n,asyncInit:r}}throw new Error("Could not initialize any backends, all backend initializations failed.")}moveData(e,t){let n=this.state.tensorInfo.get(t),s=n.backend,r=this.readSync(t),a=s.refCount(t);s.disposeData(t,!0),n.backend=e,e.move(t,r,n.shape,n.dtype,a),this.shouldCheckForMemLeaks()&&this.state.numDataMovesStack[this.state.numDataMovesStack.length-1]++}tidy(e,t){let n=null;if(t==null){if(typeof e!="function")throw new Error("Please provide a function to tidy()");t=e}else{if(typeof e!="string"&&!(e instanceof String))throw new Error("When calling with two arguments, the first argument to tidy() must be a string");if(typeof t!="function")throw new Error("When calling with two arguments, the 2nd argument to tidy() must be a function");n=e}let s;return this.scopedRun(()=>this.startScope(n),()=>this.endScope(s),()=>(s=t(),s instanceof Promise&&console.error("Cannot return a Promise inside of tidy."),s))}scopedRun(e,t,n){e();try{let s=n();return t(),s}catch(s){throw t(),s}}nextTensorId(){return Ip.nextTensorId++}nextVariableId(){return Ip.nextVariableId++}clone(e){let t=W.runKernel(_o,{x:e}),n={x:e},s=a=>({x:()=>{let o="float32",i={x:a},l={dtype:o};return W.runKernel(go,i,l)}}),r=[];return this.addTapeNode(this.state.activeScope.name,n,[t],s,r,{}),t}runKernel(e,t,n){if(this.backendName==null&&this.backend,!(fm(e,this.backendName)!=null))throw new Error(`Kernel '${e}' not registered for backend '${this.backendName}'`);return this.runKernelFunc({kernelName:e,inputs:t,attrs:n})}shouldCheckForMemLeaks(){return this.ENV.getBool("IS_TEST")}checkKernelForMemLeak(e,t,n){let s=this.backend.numDataIds(),r=0;n.forEach(i=>{r+=i.dtype==="complex64"?3:1});let a=this.state.numDataMovesStack[this.state.numDataMovesStack.length-1],o=s-t-r-a;if(o>0)throw new Error(`Backend '${this.backendName}' has an internal memory leak (${o} data ids) after running '${e}'`)}runKernelFunc(e){let t,n=[],s=this.isTapeOn(),r=this.state.numBytes,a=this.state.numTensors;this.shouldCheckForMemLeaks()&&this.state.numDataMovesStack.push(0);let o;this.backendName==null&&this.backend;let i,l=r3(e)?e.kernelName:this.state.activeScope!=null?this.state.activeScope.name:"";if(r3(e)){let{kernelName:h,inputs:f,attrs:m}=e;this.backendName==null&&this.backend;let g=fm(h,this.backendName);M(g!=null,()=>`Cannot find registered kernel '${h}' for backend '${this.backendName}'`),o=()=>{let y=this.backend.numDataIds();i=g.kernelFunc({inputs:f,attrs:m,backend:this.backend});let x=Array.isArray(i)?i:[i];this.shouldCheckForMemLeaks()&&this.checkKernelForMemLeak(h,y,x);let A=x.map(b=>b.rank!=null?b:this.makeTensorFromTensorInfo(b));if(s){let b=this.getTensorsForGradient(h,f,A);n=this.saveTensorsForBackwardMode(b)}return A}}else{let{forwardFunc:h}=e,f=m=>{!s||(n=m.map(g=>this.keep(this.clone(g))))};o=()=>{let m=this.backend.numDataIds();i=this.tidy(()=>h(this.backend,f));let g=Array.isArray(i)?i:[i];return this.shouldCheckForMemLeaks()&&this.checkKernelForMemLeak(l,m,g),g}}let{inputs:u,attrs:c}=e,p=r3(e)?null:e.backwardsFunc,d;return this.scopedRun(()=>this.state.kernelDepth++,()=>this.state.kernelDepth--,()=>{!this.ENV.getBool("DEBUG")&&!this.state.profiling?t=o():(d=this.profiler.profileKernel(l,u,()=>o()),this.ENV.getBool("DEBUG")&&this.profiler.logKernelProfile(d),t=d.outputs)}),s&&this.addTapeNode(l,u,t,p,n,c),this.state.profiling&&this.state.activeProfile.kernels.push({name:l,bytesAdded:this.state.numBytes-r,totalBytesSnapshot:this.state.numBytes,tensorsAdded:this.state.numTensors-a,totalTensorsSnapshot:this.state.numTensors,inputShapes:Object.keys(u).map(h=>u[h]!=null?u[h].shape:null),outputShapes:t.map(h=>h.shape),kernelTimeMs:d.timeMs,extraInfo:d.extraInfo}),Array.isArray(i)?t:t[0]}saveTensorsForBackwardMode(e){return e.map(n=>this.keep(this.clone(n)))}getTensorsForGradient(e,t,n){let s=g3(e);if(s!=null){let r=s.inputsToSave||[],a=s.outputsToSave||[],o;s.saveAllInputs?(M(Array.isArray(t),()=>"saveAllInputs is true, expected inputs to be an array."),o=Object.keys(t).map(l=>t[l])):o=r.map(l=>t[l]);let i=n.filter((l,u)=>a[u]);return o.concat(i)}return[]}makeTensor(e,t,n,s){if(e==null)throw new Error("Values passed to engine.makeTensor() are null");n=n||"float32",s=s||this.backend;let r=e;n==="string"&&Ga(e[0])&&(r=e.map(i=>nh(i)));let a=s.write(r,t,n),o=new nt(t,n,a,this.nextTensorId());if(this.trackTensor(o,s),n==="string"){let i=this.state.tensorInfo.get(a),l=i6(r);this.state.numBytes+=l-i.bytes,i.bytes=l}return o}makeTensorFromDataId(e,t,n,s){n=n||"float32";let r={dataId:e,shape:t,dtype:n};return this.makeTensorFromTensorInfo(r,s)}makeTensorFromTensorInfo(e,t){let{dataId:n,shape:s,dtype:r}=e,a=new nt(s,r,n,this.nextTensorId());return this.trackTensor(a,t),a}makeVariable(e,t=!0,n,s){n=n||this.nextVariableId().toString(),s!=null&&s!==e.dtype&&(e=e.cast(s));let r=new kp(e,t,n,this.nextTensorId());if(this.state.registeredVariables[r.name]!=null)throw new Error(`Variable with name ${r.name} was already registered`);return this.state.registeredVariables[r.name]=r,this.incRef(r,this.backend),r}trackTensor(e,t){this.state.numTensors++,e.dtype==="string"&&this.state.numStringTensors++;let n=0;e.dtype!=="complex64"&&e.dtype!=="string"&&(n=e.size*m3(e.dtype)),this.state.numBytes+=n,this.state.tensorInfo.has(e.dataId)||(this.state.numDataBuffers++,this.state.tensorInfo.set(e.dataId,{backend:t||this.backend,dtype:e.dtype,shape:e.shape,bytes:n})),e instanceof kp||this.track(e)}incRef(e,t){this.trackTensor(e,t),this.backend.incRef(e.dataId)}removeDataId(e,t){this.state.tensorInfo.has(e)&&this.state.tensorInfo.get(e).backend===t&&(this.state.tensorInfo.delete(e),this.state.numDataBuffers--)}disposeTensor(e){if(!this.state.tensorInfo.has(e.dataId))return;let t=this.state.tensorInfo.get(e.dataId);if(this.state.numTensors--,e.dtype==="string"&&(this.state.numStringTensors--,this.state.numBytes-=t.bytes),e.dtype!=="complex64"&&e.dtype!=="string"){let n=e.size*m3(e.dtype);this.state.numBytes-=n}t.backend.disposeData(e.dataId)&&this.removeDataId(e.dataId,t.backend)}disposeVariables(){for(let e in this.state.registeredVariables){let t=this.state.registeredVariables[e];this.disposeVariable(t)}}disposeVariable(e){this.disposeTensor(e),this.state.registeredVariables[e.name]!=null&&delete this.state.registeredVariables[e.name]}memory(){let e=this.backend.memory();return e.numTensors=this.state.numTensors,e.numDataBuffers=this.state.numDataBuffers,e.numBytes=this.state.numBytes,this.state.numStringTensors>0&&(e.unreliable=!0,e.reasons==null&&(e.reasons=[]),e.reasons.push("Memory usage by string tensors is approximate (2 bytes per character)")),e}async profile(e){this.state.profiling=!0;let t=this.state.numBytes,n=this.state.numTensors;this.state.activeProfile.kernels=[],this.state.activeProfile.result=await e(),this.state.profiling=!1,this.state.activeProfile.peakBytes=Math.max(...this.state.activeProfile.kernels.map(s=>s.totalBytesSnapshot)),this.state.activeProfile.newBytes=this.state.numBytes-t,this.state.activeProfile.newTensors=this.state.numTensors-n;for(let s of this.state.activeProfile.kernels)s.kernelTimeMs=await s.kernelTimeMs,s.extraInfo=await s.extraInfo;return this.state.activeProfile}isTapeOn(){return this.state.gradientDepth>0&&this.state.kernelDepth===0}addTapeNode(e,t,n,s,r,a){let o={id:this.state.nextTapeNodeId++,kernelName:e,inputs:t,outputs:n,saved:r},i=g3(e);i!=null&&(s=i.gradFunc),s!=null&&(o.gradient=l=>(l=l.map((u,c)=>{if(u==null){let p=n[c],d=Um(p.size,p.dtype);return this.makeTensor(d,p.shape,p.dtype)}return u}),s(l.length>1?l:l[0],r,a))),this.state.activeTape.push(o)}keep(e){return e.kept=!0,e}startTape(){this.state.gradientDepth===0&&(this.state.activeTape=[]),this.state.gradientDepth++}endTape(){this.state.gradientDepth--}startScope(e){let t={track:[],name:"unnamed scope",id:this.state.nextScopeId++};e&&(t.name=e),this.state.scopeStack.push(t),this.state.activeScope=t}endScope(e){let t=Cy(e),n=new Set(t.map(r=>r.id));for(let r=0;r<this.state.activeScope.track.length;r++){let a=this.state.activeScope.track[r];!a.kept&&!n.has(a.id)&&a.dispose()}let s=this.state.scopeStack.pop();this.state.activeScope=this.state.scopeStack.length===0?null:this.state.scopeStack[this.state.scopeStack.length-1],t.forEach(r=>{!r.kept&&r.scopeId===s.id&&this.track(r)})}gradients(e,t,n,s=!1){if(M(t.length>0,()=>"gradients() received an empty list of xs."),n!=null&&n.dtype!=="float32")throw new Error(`dy must have 'float32' dtype, but has '${n.dtype}'`);let r=this.scopedRun(()=>this.startTape(),()=>this.endTape(),()=>this.tidy("forward",e));M(r instanceof nt,()=>"The result y returned by f() must be a tensor.");let a=pD(this.state.activeTape,t,r);if(!s&&a.length===0&&t.length>0)throw new Error("Cannot compute gradient of y=f(x) with respect to x. Make sure that the f you passed encloses all operations that lead from x to y.");return this.tidy("backward",()=>{let o={};o[r.id]=n==null?kD(r.shape):n,hD(o,a,l=>this.tidy(l),ID);let i=t.map(l=>o[l.id]);return this.state.gradientDepth===0&&(this.state.activeTape.forEach(l=>{for(let u of l.saved)u.dispose()}),this.state.activeTape=null),{value:r,grads:i}})}customGrad(e){return M(Ya(e),()=>"The f passed in customGrad(f) must be a function."),(...t)=>{M(t.every(o=>o instanceof nt),()=>"The args passed in customGrad(f)(x1, x2,...) must all be tensors");let n,s={};t.forEach((o,i)=>{s[i]=o});let r=(o,i)=>(n=e(...t,i),M(n.value instanceof nt,()=>"The function f passed in customGrad(f) must return an object where `obj.value` is a tensor"),M(Ya(n.gradFunc),()=>"The function f passed in customGrad(f) must return an object where `obj.gradFunc` is a function."),n.value),a=(o,i)=>{let l=n.gradFunc(o,i),u=Array.isArray(l)?l:[l];M(u.length===t.length,()=>"The function f passed in customGrad(f) must return an object where `obj.gradFunc` is a function that returns the same number of tensors as inputs passed to f(...)."),M(u.every(p=>p instanceof nt),()=>"The function f passed in customGrad(f) must return an object where `obj.gradFunc` is a function that returns a list of only tensors.");let c={};return u.forEach((p,d)=>{c[d]=()=>p}),c};return this.runKernelFunc({forwardFunc:r,backwardsFunc:a,inputs:s})}}readSync(e){return this.state.tensorInfo.get(e).backend.readSync(e)}read(e){return this.state.tensorInfo.get(e).backend.read(e)}readToGPU(e,t){return this.state.tensorInfo.get(e).backend.readToGPU(e,t)}async time(e){let t=wp(),n=await this.backend.time(e);return n.wallMs=wp()-t,n}track(e){return this.state.activeScope!=null&&(e.scopeId=this.state.activeScope.id,this.state.activeScope.track.push(e)),e}get registeredVariables(){return this.state.registeredVariables}reset(){this.pendingBackendInitId++,this.state.dispose(),this.ENV.reset(),this.state=new yv;for(let e in this.registry)this.disposeRegisteredKernels(e),this.registry[e].dispose(),delete this.registry[e];this.backendName=null,this.backendInstance=null,this.pendingBackendInit=null}};Ip.nextTensorId=0;Ip.nextVariableId=0;function kD(e){let t=by(Et(e),"float32");return W.makeTensor(t,e,"float32")}function w6(){let e=p6();if(e._tfengine==null){let t=new d6(e);e._tfengine=new Ip(t)}return j_(e._tfengine.ENV),yD(()=>e._tfengine),e._tfengine}var W=w6();function ID(e,t){let n={a:e,b:t};return W.runKernel(xa,n)}var rh={};Ve(rh,{isBrowser:()=>k6,isMobile:()=>TD,mockIsMobile:()=>CD});function SD(){return typeof navigator!="undefined"&&navigator!=null}var k3;function CD(e){k3=e}function TD(e){if(k3!==void 0)return k3;if(e||SD()){if(e||(e=navigator),e.product==="ReactNative")return!0;let t=e.userAgent||e.vendor||(typeof window!="undefined"?window.opera:"");if(!t){let n=e;return n.userAgentData&&n.userAgentData.mobile}return/(android|bb\d+|meego).+mobile|avantgo|bada\/|blackberry|blazer|compal|elaine|fennec|hiptop|iemobile|ip(hone|od)|iris|kindle|lge |maemo|midp|mmp|mobile.+firefox|netfront|opera m(ob|in)i|palm( os)?|phone|p(ixi|re)\/|plucker|pocket|psp|series(4|6)0|symbian|treo|up\.(browser|link)|vodafone|wap|windows ce|xda|xiino/i.test(t)||/1207|6310|6590|3gso|4thp|50[1-6]i|770s|802s|a wa|abac|ac(er|oo|s\-)|ai(ko|rn)|al(av|ca|co)|amoi|an(ex|ny|yw)|aptu|ar(ch|go)|as(te|us)|attw|au(di|\-m|r |s )|avan|be(ck|ll|nq)|bi(lb|rd)|bl(ac|az)|br(e|v)w|bumb|bw\-(n|u)|c55\/|capi|ccwa|cdm\-|cell|chtm|cldc|cmd\-|co(mp|nd)|craw|da(it|ll|ng)|dbte|dc\-s|devi|dica|dmob|do(c|p)o|ds(12|\-d)|el(49|ai)|em(l2|ul)|er(ic|k0)|esl8|ez([4-7]0|os|wa|ze)|fetc|fly(\-|_)|g1 u|g560|gene|gf\-5|g\-mo|go(\.w|od)|gr(ad|un)|haie|hcit|hd\-(m|p|t)|hei\-|hi(pt|ta)|hp( i|ip)|hs\-c|ht(c(\-| |_|a|g|p|s|t)|tp)|hu(aw|tc)|i\-(20|go|ma)|i230|iac( |\-|\/)|ibro|idea|ig01|ikom|im1k|inno|ipaq|iris|ja(t|v)a|jbro|jemu|jigs|kddi|keji|kgt( |\/)|klon|kpt |kwc\-|kyo(c|k)|le(no|xi)|lg( g|\/(k|l|u)|50|54|\-[a-w])|libw|lynx|m1\-w|m3ga|m50\/|ma(te|ui|xo)|mc(01|21|ca)|m\-cr|me(rc|ri)|mi(o8|oa|ts)|mmef|mo(01|02|bi|de|do|t(\-| |o|v)|zz)|mt(50|p1|v )|mwbp|mywa|n10[0-2]|n20[2-3]|n30(0|2)|n50(0|2|5)|n7(0(0|1)|10)|ne((c|m)\-|on|tf|wf|wg|wt)|nok(6|i)|nzph|o2im|op(ti|wv)|oran|owg1|p800|pan(a|d|t)|pdxg|pg(13|\-([1-8]|c))|phil|pire|pl(ay|uc)|pn\-2|po(ck|rt|se)|prox|psio|pt\-g|qa\-a|qc(07|12|21|32|60|\-[2-7]|i\-)|qtek|r380|r600|raks|rim9|ro(ve|zo)|s55\/|sa(ge|ma|mm|ms|ny|va)|sc(01|h\-|oo|p\-)|sdk\/|se(c(\-|0|1)|47|mc|nd|ri)|sgh\-|shar|sie(\-|m)|sk\-0|sl(45|id)|sm(al|ar|b3|it|t5)|so(ft|ny)|sp(01|h\-|v\-|v )|sy(01|mb)|t2(18|50)|t6(00|10|18)|ta(gt|lk)|tcl\-|tdg\-|tel(i|m)|tim\-|t\-mo|to(pl|sh)|ts(70|m\-|m3|m5)|tx\-9|up(\.b|g1|si)|utst|v400|v750|veri|vi(rg|te)|vk(40|5[0-3]|\-v)|vm40|voda|vulc|vx(52|53|60|61|70|80|81|83|85|98)|w3c(\-| )|webc|whit|wi(g |nc|nw)|wmlb|wonu|x700|yas\-|your|zeto|zte\-/i.test(t.substr(0,4))}return!1}function k6(){return typeof window!="undefined"&&window.document!=null||typeof WorkerGlobalScope!="undefined"}var lr=q();lr.registerFlag("DEBUG",()=>!1,e=>{e&&console.warn("Debugging mode is ON. The output of every math call will be downloaded to CPU and checked for NaNs. This significantly impacts performance.")});lr.registerFlag("IS_BROWSER",()=>k6());lr.registerFlag("IS_NODE",()=>typeof process!="undefined"&&typeof process.versions!="undefined"&&typeof process.versions.node!="undefined");lr.registerFlag("IS_CHROME",()=>typeof navigator!="undefined"&&navigator!=null&&navigator.userAgent!=null&&/Chrome/.test(navigator.userAgent)&&/Google Inc/.test(navigator.vendor));lr.registerFlag("PROD",()=>!1);lr.registerFlag("TENSORLIKE_CHECK_SHAPE_CONSISTENCY",()=>lr.getBool("DEBUG"));lr.registerFlag("DEPRECATION_WARNINGS_ENABLED",()=>!0);lr.registerFlag("IS_TEST",()=>!1);lr.registerFlag("CHECK_COMPUTATION_FOR_ERRORS",()=>!0);lr.registerFlag("WRAP_TO_IMAGEBITMAP",()=>!1);lr.registerFlag("ENGINE_COMPILE_ONLY",()=>!1);function Xr(e,t){let n=e;if(Fn(e))return t==="string"?[]:[e.length];if(!Array.isArray(e))return[];let s=[];for(;Array.isArray(n)||Fn(n)&&t!=="string";)s.push(n.length),n=n[0];return Array.isArray(e)&&q().getBool("TENSORLIKE_CHECK_SHAPE_CONSISTENCY")&&I6(e,s,[]),s}function I6(e,t,n){if(n=n||[],!Array.isArray(e)&&!Fn(e)){M(t.length===0,()=>`Element arr[${n.join("][")}] is a primitive, but should be an array/TypedArray of ${t[0]} elements`);return}M(t.length>0,()=>`Element arr[${n.join("][")}] should be a primitive, but is an array of ${e.length} elements`),M(e.length===t[0],()=>`Element arr[${n.join("][")}] should have ${t[0]} elements, but has ${e.length} elements`);let s=t.slice(1);for(let r=0;r<e.length;++r)I6(e[r],s,n.concat(r))}function Av(e,t,n,s){if(e!=="string_or_numeric"){if(e==null)throw new Error("Expected dtype cannot be null.");if(e!=="numeric"&&e!==t||e==="numeric"&&t==="string")throw new Error(`Argument '${n}' passed to '${s}' must be ${e} tensor, but got ${t} tensor`)}}function D(e,t,n,s="numeric"){if(e instanceof nt)return Av(s,e.dtype,t,n),e;let r=Vm(e);if(r!=="string"&&["bool","int32","float32"].indexOf(s)>=0&&(r=s),Av(s,r,t,n),e==null||!Fn(e)&&!Array.isArray(e)&&typeof e!="number"&&typeof e!="boolean"&&typeof e!="string"){let l=e==null?"null":e.constructor.name;throw new Error(`Argument '${t}' passed to '${n}' must be a Tensor or TensorLike, but got '${l}'`)}let a=Xr(e,r);!Fn(e)&&!Array.isArray(e)&&(e=[e]);let i=r!=="string"?f0(e,r):Xi(e,[],!0);return W.makeTensor(i,a,r)}function Sp(e,t,n,s="numeric"){if(!Array.isArray(e))throw new Error(`Argument ${t} passed to ${n} must be a \`Tensor[]\` or \`TensorLike[]\``);return e.map((a,o)=>D(a,`${t}[${o}]`,n,s))}var Ty="__op";function V(e){let t=Object.keys(e);if(t.length!==1)throw new Error(`Please provide an object with a single key (operation name) mapping to a function. Got an object with ${t.length} keys.`);let n=t[0],s=e[n];n.endsWith("_")&&(n=n.substring(0,n.length-1)),n=n+Ty;let r=(...a)=>{W.startScope(n);try{let o=s(...a);return wy(o)&&console.error("Cannot return a Promise inside of tidy."),W.endScope(o),o}catch(o){throw W.endScope(null),o}};return Object.defineProperty(r,"name",{value:n,configurable:!0}),r}function ND(e,t){let n=D(e,"real","complex"),s=D(t,"imag","complex");ns(n.shape,s.shape,`real and imag shapes, ${n.shape} and ${s.shape}, must match in call to tf.complex().`);let r={real:n,imag:s};return W.runKernel(Lp,r)}var fa=V({complex_:ND});function oi(e,t,n,s){if(s==null&&(s=Vm(e)),s==="complex64")throw new Error("Cannot construct a complex64 tensor directly. Please use tf.complex(real, imag).");if(!Fn(e)&&!Array.isArray(e)&&typeof e!="number"&&typeof e!="boolean"&&typeof e!="string")throw new Error("values passed to tensor(values) must be a number/boolean/string or an array of numbers/booleans/strings, or a TypedArray");if(t!=null){vy(t);let r=Et(t),a=Et(n);M(r===a,()=>`Based on the provided shape, [${t}], the tensor should have ${r} values but has ${a}`);for(let o=0;o<n.length;++o){let i=n[o],l=o===n.length-1?i!==Et(t.slice(o)):!0;M(n[o]===t[o]||!l,()=>`Error creating a new Tensor. Inferred shape (${n}) does not match the provided shape (${t}). `)}}return!Fn(e)&&!Array.isArray(e)&&(e=[e]),t=t||n,e=s!=="string"?f0(e,s):Xi(e,[],!0),W.makeTensor(e,t,s)}function ct(e,t,n){let s=Xr(e,n);return oi(e,t,s,n)}var I3={float32:4,float16:2,int32:4,uint16:2,uint8:1,bool:1,complex64:8},gm=4;async function ED(e,t){let n=[],s=[],r=Array.isArray(e)?e.map(o=>o.name):Object.keys(e);for(let o=0;o<r.length;++o){let i=r[o],l=Array.isArray(e)?e[o].tensor:e[i];if(l.dtype!=="float32"&&l.dtype!=="int32"&&l.dtype!=="bool"&&l.dtype!=="string"&&l.dtype!=="complex64")throw new Error(`Unsupported dtype in weight '${i}': ${l.dtype}`);let u={name:i,shape:l.shape,dtype:l.dtype};if(l.dtype==="string"){let c=new Promise(async p=>{let d=await l.bytes(),h=d.reduce((g,y)=>g+y.length,0)+gm*d.length,f=new Uint8Array(h),m=0;for(let g=0;g<d.length;g++){let y=d[g],x=new Uint8Array(new Uint32Array([y.length]).buffer);f.set(x,m),m+=gm,f.set(y,m),m+=y.length}p(f)});s.push(c)}else s.push(l.data());t!=null&&(u.group=t),n.push(u)}let a=await Promise.all(s);return{data:RD(a),specs:n}}function S6(e,t){let n={},s,r=0;for(let a of t){let o=a.name,i=a.dtype,l=a.shape,u=Et(l),c;if("quantization"in a){let p=a.quantization;if(p.dtype==="uint8"||p.dtype==="uint16"){if(!("min"in p&&"scale"in p))throw new Error(`Weight ${a.name} with quantization ${p.dtype} doesn't have corresponding metadata min and scale.`)}else if(p.dtype==="float16"){if(i!=="float32")throw new Error(`Weight ${a.name} is quantized with ${p.dtype} which only supports weights of type float32 not ${i}.`)}else throw new Error(`Weight ${a.name} has unknown quantization dtype ${p.dtype}. Supported quantization dtypes are: 'uint8', 'uint16', and 'float16'.`);let d=I3[p.dtype],h=e.slice(r,r+u*d),f=p.dtype==="uint8"?new Uint8Array(h):new Uint16Array(h);if(i==="float32")if(p.dtype==="uint8"||p.dtype==="uint16"){c=new Float32Array(f.length);for(let m=0;m<f.length;m++){let g=f[m];c[m]=g*p.scale+p.min}}else if(p.dtype==="float16")s===void 0&&(s=OD()),c=s(f);else throw new Error(`Unsupported quantization type ${p.dtype} for weight type float32.`);else if(i==="int32"){if(p.dtype!=="uint8"&&p.dtype!=="uint16")throw new Error(`Unsupported quantization type ${p.dtype} for weight type int32.`);c=new Int32Array(f.length);for(let m=0;m<f.length;m++){let g=f[m];c[m]=Math.round(g*p.scale+p.min)}}else throw new Error(`Unsupported dtype in weight '${o}': ${i}`);r+=u*d}else if(i==="string"){let p=Et(a.shape);c=[];for(let d=0;d<p;d++){let h=new Uint32Array(e.slice(r,r+gm))[0];r+=gm;let f=new Uint8Array(e.slice(r,r+h));c.push(f),r+=h}}else{let p=I3[i],d=e.slice(r,r+u*p);if(i==="float32")c=new Float32Array(d);else if(i==="int32")c=new Int32Array(d);else if(i==="bool")c=new Uint8Array(d);else if(i==="complex64"){c=new Float32Array(d);let h=new Float32Array(c.length/2),f=new Float32Array(c.length/2);for(let y=0;y<h.length;y++)h[y]=c[y*2],f[y]=c[y*2+1];let m=ct(h,l,"float32"),g=ct(f,l,"float32");n[o]=fa(m,g),m.dispose(),g.dispose()}else throw new Error(`Unsupported dtype in weight '${o}': ${i}`);r+=u*p}i!=="complex64"&&(n[o]=ct(c,l,i))}return n}function RD(e){if(e===null)throw new Error(`Invalid input value: ${JSON.stringify(e)}`);let t=0,n=[];e.forEach(a=>{if(t+=a.byteLength,n.push(a.byteLength===a.buffer.byteLength?a:new a.constructor(a)),!(a instanceof Float32Array||a instanceof Int32Array||a instanceof Uint8Array))throw new Error(`Unsupported TypedArray subtype: ${a.constructor.name}`)});let s=new Uint8Array(t),r=0;return n.forEach(a=>{s.set(new Uint8Array(a.buffer),r),r+=a.byteLength}),s.buffer}var Ny=typeof Buffer!="undefined"&&(typeof Blob=="undefined"||typeof atob=="undefined"||typeof btoa=="undefined");function xv(e){return Ny?Buffer.byteLength(e):new Blob([e]).size}function _D(e){if(Ny)return Buffer.from(e).toString("base64");let t=new Uint8Array(e),n="";for(let s=0,r=t.length;s<r;s++)n+=String.fromCharCode(t[s]);return btoa(n)}function DD(e){if(Ny){let s=Buffer.from(e,"base64");return s.buffer.slice(s.byteOffset,s.byteOffset+s.byteLength)}let t=atob(e),n=new Uint8Array(t.length);for(let s=0;s<t.length;++s)n.set([t.charCodeAt(s)],s);return n.buffer}function Ey(e){if(e.length===1)return e[0];let t=0;e.forEach(r=>{t+=r.byteLength});let n=new Uint8Array(t),s=0;return e.forEach(r=>{n.set(new Uint8Array(r),s),s+=r.byteLength}),n.buffer}function bv(e){let t="/";for(e=e.trim();e.endsWith(t);)e=e.slice(0,e.length-1);let n=e.split(t);return n[n.length-1]}function C6(e,t){let n={modelTopology:e.modelTopology,format:e.format,generatedBy:e.generatedBy,convertedBy:e.convertedBy,weightsManifest:t};return e.signature!=null&&(n.signature=e.signature),e.userDefinedMetadata!=null&&(n.userDefinedMetadata=e.userDefinedMetadata),e.modelInitializer!=null&&(n.modelInitializer=e.modelInitializer),e.trainingConfig!=null&&(n.trainingConfig=e.trainingConfig),n}async function Ry(e,t){let n={modelTopology:e.modelTopology,format:e.format,generatedBy:e.generatedBy,convertedBy:e.convertedBy};if(e.trainingConfig!=null&&(n.trainingConfig=e.trainingConfig),e.weightsManifest!=null){let[s,r]=await t(e.weightsManifest);n.weightSpecs=s,n.weightData=r}return e.signature!=null&&(n.signature=e.signature),e.userDefinedMetadata!=null&&(n.userDefinedMetadata=e.userDefinedMetadata),e.modelInitializer!=null&&(n.modelInitializer=e.modelInitializer),n}function ah(e){if(e.modelTopology instanceof ArrayBuffer)throw new Error("Expected JSON model topology, received ArrayBuffer.");return{dateSaved:new Date,modelTopologyType:"JSON",modelTopologyBytes:e.modelTopology==null?0:xv(JSON.stringify(e.modelTopology)),weightSpecsBytes:e.weightSpecs==null?0:xv(JSON.stringify(e.weightSpecs)),weightDataBytes:e.weightData==null?0:e.weightData.byteLength}}function $D(){let e=n=>{let s=n<<13,r=0;for(;(s&8388608)===0;)r-=8388608,s<<=1;return s&=-8388609,r+=947912704,s|r},t=new Uint32Array(2048);t[0]=0;for(let n=1;n<1024;n++)t[n]=e(n);for(let n=1024;n<2048;n++)t[n]=939524096+(n-1024<<13);return t}function PD(){let e=new Uint32Array(64);e[0]=0,e[31]=1199570944,e[32]=2147483648,e[63]=3347054592;for(let t=1;t<31;t++)e[t]=t<<23;for(let t=33;t<63;t++)e[t]=2147483648+(t-32<<23);return e}function FD(){let e=new Uint32Array(64);for(let t=0;t<64;t++)e[t]=1024;return e[0]=e[32]=0,e}function OD(){let e=$D(),t=PD(),n=FD();return s=>{let r=new ArrayBuffer(4*s.length),a=new Uint32Array(r);for(let o=0;o<s.length;o++){let i=s[o],l=e[n[i>>10]+(i&1023)]+t[i>>10];a[o]=l}return new Float32Array(r)}}var qt=class{constructor(){this.saveRouters=[],this.loadRouters=[]}static getInstance(){return qt.instance==null&&(qt.instance=new qt),qt.instance}static registerSaveRouter(e){qt.getInstance().saveRouters.push(e)}static registerLoadRouter(e){qt.getInstance().loadRouters.push(e)}static getSaveHandlers(e){return qt.getHandlers(e,"save")}static getLoadHandlers(e,t){return qt.getHandlers(e,"load",t)}static getHandlers(e,t,n){let s=[];return(t==="load"?qt.getInstance().loadRouters:qt.getInstance().saveRouters).forEach(a=>{let o=a(e,n);o!==null&&s.push(o)}),s}},MD=e=>qt.registerSaveRouter(e),zD=e=>qt.registerLoadRouter(e),LD=e=>qt.getSaveHandlers(e),BD=(e,t)=>qt.getLoadHandlers(e,t),S3="tensorflowjs",C3=1,Wi="models_store",Ha="model_info_store";function T6(){if(!q().getBool("IS_BROWSER"))throw new Error("Failed to obtain IndexedDB factory because the current environmentis not a web browser.");let e=typeof window=="undefined"?self:window,t=e.indexedDB||e.mozIndexedDB||e.webkitIndexedDB||e.msIndexedDB||e.shimIndexedDB;if(t==null)throw new Error("The current browser does not appear to support IndexedDB.");return t}function T3(e){let t=e.result;t.createObjectStore(Wi,{keyPath:"modelPath"}),t.createObjectStore(Ha,{keyPath:"modelPath"})}var Ki=class{constructor(e){if(this.indexedDB=T6(),e==null||!e)throw new Error("For IndexedDB, modelPath must not be null, undefined or empty.");this.modelPath=e}async save(e){if(e.modelTopology instanceof ArrayBuffer)throw new Error("BrowserLocalStorage.save() does not support saving model topology in binary formats yet.");return this.databaseAction(this.modelPath,e)}async load(){return this.databaseAction(this.modelPath)}databaseAction(e,t){return new Promise((n,s)=>{let r=this.indexedDB.open(S3,C3);r.onupgradeneeded=()=>T3(r),r.onsuccess=()=>{let a=r.result;if(t==null){let o=a.transaction(Wi,"readonly"),l=o.objectStore(Wi).get(this.modelPath);l.onsuccess=()=>{if(l.result==null)return a.close(),s(new Error(`Cannot find model with path '${this.modelPath}' in IndexedDB.`));n(l.result.modelArtifacts)},l.onerror=u=>(a.close(),s(l.error)),o.oncomplete=()=>a.close()}else{let o=ah(t),i=a.transaction(Ha,"readwrite"),l=i.objectStore(Ha),u=l.put({modelPath:this.modelPath,modelArtifactsInfo:o}),c;u.onsuccess=()=>{c=a.transaction(Wi,"readwrite");let d=c.objectStore(Wi).put({modelPath:this.modelPath,modelArtifacts:t,modelArtifactsInfo:o});d.onsuccess=()=>n({modelArtifactsInfo:o}),d.onerror=h=>{l=i.objectStore(Ha);let f=l.delete(this.modelPath);f.onsuccess=()=>(a.close(),s(d.error)),f.onerror=m=>(a.close(),s(d.error))}},u.onerror=p=>(a.close(),s(u.error)),i.oncomplete=()=>{c==null?a.close():c.oncomplete=()=>a.close()}}},r.onerror=a=>s(r.error)})}};Ki.URL_SCHEME="indexeddb://";var N6=e=>q().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(Ki.URL_SCHEME)?WD(e.slice(Ki.URL_SCHEME.length)):null;qt.registerSaveRouter(N6);qt.registerLoadRouter(N6);function WD(e){return new Ki(e)}function VD(e){return e.startsWith(Ki.URL_SCHEME)?e.slice(Ki.URL_SCHEME.length):e}var UD=class{constructor(){this.indexedDB=T6()}async listModels(){return new Promise((e,t)=>{let n=this.indexedDB.open(S3,C3);n.onupgradeneeded=()=>T3(n),n.onsuccess=()=>{let s=n.result,r=s.transaction(Ha,"readonly"),o=r.objectStore(Ha).getAll();o.onsuccess=()=>{let i={};for(let l of o.result)i[l.modelPath]=l.modelArtifactsInfo;e(i)},o.onerror=i=>(s.close(),t(o.error)),r.oncomplete=()=>s.close()},n.onerror=s=>t(n.error)})}async removeModel(e){return e=VD(e),new Promise((t,n)=>{let s=this.indexedDB.open(S3,C3);s.onupgradeneeded=()=>T3(s),s.onsuccess=()=>{let r=s.result,a=r.transaction(Ha,"readwrite"),o=a.objectStore(Ha),i=o.get(e),l;i.onsuccess=()=>{if(i.result==null)return r.close(),n(new Error(`Cannot find model with path '${e}' in IndexedDB.`));{let u=o.delete(e),c=()=>{l=r.transaction(Wi,"readwrite");let d=l.objectStore(Wi).delete(e);d.onsuccess=()=>t(i.result.modelArtifactsInfo),d.onerror=h=>n(i.error)};u.onsuccess=c,u.onerror=p=>(c(),r.close(),n(i.error))}},i.onerror=u=>(r.close(),n(i.error)),a.oncomplete=()=>{l==null?r.close():l.oncomplete=()=>r.close()}},s.onerror=r=>n(s.error)})}},da="/",Lu="tensorflowjs_models",E6="info",GD="model_topology",HD="weight_specs",jD="weight_data",qD="model_metadata";function R6(e){return{info:[Lu,e,E6].join(da),topology:[Lu,e,GD].join(da),weightSpecs:[Lu,e,HD].join(da),weightData:[Lu,e,jD].join(da),modelMetadata:[Lu,e,qD].join(da)}}function _6(e){for(let t of Object.values(e))window.localStorage.removeItem(t)}function XD(e){let t=e.split(da);if(t.length<3)throw new Error(`Invalid key format: ${e}`);return t.slice(1,t.length-1).join(da)}function KD(e){return e.startsWith(Zi.URL_SCHEME)?e.slice(Zi.URL_SCHEME.length):e}var Zi=class{constructor(e){if(!q().getBool("IS_BROWSER")||typeof window=="undefined"||typeof window.localStorage=="undefined")throw new Error("The current environment does not support local storage.");if(this.LS=window.localStorage,e==null||!e)throw new Error("For local storage, modelPath must not be null, undefined or empty.");this.modelPath=e,this.keys=R6(this.modelPath)}async save(e){if(e.modelTopology instanceof ArrayBuffer)throw new Error("BrowserLocalStorage.save() does not support saving model topology in binary formats yet.");{let t=JSON.stringify(e.modelTopology),n=JSON.stringify(e.weightSpecs),s=ah(e);try{this.LS.setItem(this.keys.info,JSON.stringify(s)),this.LS.setItem(this.keys.topology,t),this.LS.setItem(this.keys.weightSpecs,n),this.LS.setItem(this.keys.weightData,_D(e.weightData));let r={format:e.format,generatedBy:e.generatedBy,convertedBy:e.convertedBy,signature:e.signature!=null?e.signature:void 0,userDefinedMetadata:e.userDefinedMetadata!=null?e.userDefinedMetadata:void 0,modelInitializer:e.modelInitializer!=null?e.modelInitializer:void 0,trainingConfig:e.trainingConfig!=null?e.trainingConfig:void 0};return this.LS.setItem(this.keys.modelMetadata,JSON.stringify(r)),{modelArtifactsInfo:s}}catch(r){throw _6(this.keys),new Error(`Failed to save model '${this.modelPath}' to local storage: size quota being exceeded is a possible cause of this failure: modelTopologyBytes=${s.modelTopologyBytes}, weightSpecsBytes=${s.weightSpecsBytes}, weightDataBytes=${s.weightDataBytes}.`)}}}async load(){let e=JSON.parse(this.LS.getItem(this.keys.info));if(e==null)throw new Error(`In local storage, there is no model with name '${this.modelPath}'`);if(e.modelTopologyType!=="JSON")throw new Error("BrowserLocalStorage does not support loading non-JSON model topology yet.");let t={},n=JSON.parse(this.LS.getItem(this.keys.topology));if(n==null)throw new Error(`In local storage, the topology of model '${this.modelPath}' is missing.`);t.modelTopology=n;let s=JSON.parse(this.LS.getItem(this.keys.weightSpecs));if(s==null)throw new Error(`In local storage, the weight specs of model '${this.modelPath}' are missing.`);t.weightSpecs=s;let r=this.LS.getItem(this.keys.modelMetadata);if(r!=null){let o=JSON.parse(r);t.format=o.format,t.generatedBy=o.generatedBy,t.convertedBy=o.convertedBy,o.signature!=null&&(t.signature=o.signature),o.userDefinedMetadata!=null&&(t.userDefinedMetadata=o.userDefinedMetadata),o.modelInitializer!=null&&(t.modelInitializer=o.modelInitializer),o.trainingConfig!=null&&(t.trainingConfig=o.trainingConfig)}let a=this.LS.getItem(this.keys.weightData);if(a==null)throw new Error(`In local storage, the binary weight values of model '${this.modelPath}' are missing.`);return t.weightData=DD(a),t}};Zi.URL_SCHEME="localstorage://";var D6=e=>q().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(Zi.URL_SCHEME)?ZD(e.slice(Zi.URL_SCHEME.length)):null;qt.registerSaveRouter(D6);qt.registerLoadRouter(D6);function ZD(e){return new Zi(e)}var YD=class{constructor(){M(q().getBool("IS_BROWSER"),()=>"Current environment is not a web browser"),M(typeof window=="undefined"||typeof window.localStorage!="undefined",()=>"Current browser does not appear to support localStorage"),this.LS=window.localStorage}async listModels(){let e={},t=Lu+da,n=da+E6;for(let s=0;s<this.LS.length;++s){let r=this.LS.key(s);if(r.startsWith(t)&&r.endsWith(n)){let a=XD(r);e[a]=JSON.parse(this.LS.getItem(r))}}return e}async removeModel(e){e=KD(e);let t=R6(e);if(this.LS.getItem(t.info)==null)throw new Error(`Cannot find model at path '${e}'`);let n=JSON.parse(this.LS.getItem(t.info));return _6(t),n}},Vu="://",ds=class{constructor(){this.managers={}}static getInstance(){return ds.instance==null&&(ds.instance=new ds),ds.instance}static registerManager(e,t){M(e!=null,()=>"scheme must not be undefined or null."),e.endsWith(Vu)&&(e=e.slice(0,e.indexOf(Vu))),M(e.length>0,()=>"scheme must not be an empty string.");let n=ds.getInstance();M(n.managers[e]==null,()=>`A model store manager is already registered for scheme '${e}'.`),n.managers[e]=t}static getManager(e){let t=ds.getInstance().managers[e];if(t==null)throw new Error(`Cannot find model manager for scheme '${e}'`);return t}static getSchemes(){return Object.keys(ds.getInstance().managers)}};function em(e){if(e.indexOf(Vu)===-1)throw new Error(`The url string provided does not contain a scheme. Supported schemes are: ${ds.getSchemes().join(",")}`);return{scheme:e.split(Vu)[0],path:e.split(Vu)[1]}}async function $6(e,t,n=!1){M(e!==t,()=>`Old path and new path are the same: '${e}'`);let s=qt.getLoadHandlers(e);M(s.length>0,()=>`Copying failed because no load handler is found for source URL ${e}.`),M(s.length<2,()=>`Copying failed because more than one (${s.length}) load handlers for source URL ${e}.`);let r=s[0],a=qt.getSaveHandlers(t);M(a.length>0,()=>`Copying failed because no save handler is found for destination URL ${t}.`),M(a.length<2,()=>`Copying failed because more than one (${s.length}) save handlers for destination URL ${t}.`);let o=a[0],i=em(e).scheme,l=em(e).path,u=i===em(e).scheme,c=await r.load();n&&u&&await ds.getManager(i).removeModel(l);let p=await o.save(c);return n&&!u&&await ds.getManager(i).removeModel(l),p.modelArtifactsInfo}async function JD(){let e=ds.getSchemes(),t={};for(let n of e){let s=await ds.getManager(n).listModels();for(let r in s){let a=n+Vu+r;t[a]=s[r]}}return t}async function QD(e){let t=em(e);return ds.getManager(t.scheme).removeModel(t.path)}async function e$(e,t){return $6(e,t,!1)}async function t$(e,t){return $6(e,t,!0)}var n$=class{fetch(e,t){return fetch(e,t)}now(){return performance.now()}encode(e,t){if(t!=="utf-8"&&t!=="utf8")throw new Error(`Browser's encoder only supports utf-8, but got ${t}`);return this.textEncoder==null&&(this.textEncoder=new TextEncoder),this.textEncoder.encode(e)}decode(e,t){return new TextDecoder(t).decode(e)}};if(q().get("IS_BROWSER")){q().setPlatform("browser",new n$);try{ds.registerManager(Zi.URL_SCHEME,new YD)}catch(e){}try{ds.registerManager(Ki.URL_SCHEME,new UD)}catch(e){}}var s$={importFetch:()=>c_()},a3,r$=class{constructor(){this.util=d_(),this.textEncoder=new this.util.TextEncoder}fetch(e,t){return q().global.fetch!=null?q().global.fetch(e,t):(a3==null&&(a3=s$.importFetch()),a3(e,t))}now(){let e=process.hrtime();return e[0]*1e3+e[1]/1e6}encode(e,t){if(t!=="utf-8"&&t!=="utf8")throw new Error(`Node built-in encoder only supports utf-8, but got ${t}`);return this.textEncoder.encode(e)}decode(e,t){return e.length===0?"":new this.util.TextDecoder(t).decode(e)}};q().get("IS_NODE")&&!q().get("IS_BROWSER")&&q().setPlatform("node",new r$);function Le(e,t="float32",n){return t=t||"float32",vy(e),new fn(e,t,n)}function a$(e,t){let n=D(e,"x","cast");if(!o6(t))throw new Error(`Failed to cast to unknown dtype ${t}`);if(t==="string"&&n.dtype!=="string"||t!=="string"&&n.dtype==="string")throw new Error("Only strings can be casted to strings");let s={x:n},r={dtype:t};return W.runKernel(go,s,r)}var ye=V({cast_:a$});function o$(e){let n={x:D(e,"x","clone","string_or_numeric")};return W.runKernel(_o,n)}var On=V({clone_:o$});function _y(e,t=!1){console.log(e.toString(t))}w6();var i$={buffer:Le,cast:ye,clone:On,print:_y};AD(i$);var Ts={};Ve(Ts,{browserFiles:()=>f$,browserHTTPRequest:()=>x$,concatenateArrayBuffers:()=>Ey,copyModel:()=>e$,decodeWeights:()=>S6,encodeWeights:()=>ED,fromMemory:()=>v$,fromMemorySync:()=>z6,getLoadHandlers:()=>BD,getModelArtifactsForJSON:()=>Ry,getModelArtifactsInfoForJSON:()=>ah,getSaveHandlers:()=>LD,http:()=>$y,isHTTPScheme:()=>N3,listModels:()=>JD,loadWeights:()=>m$,moveModel:()=>t$,registerLoadRouter:()=>zD,registerSaveRouter:()=>MD,removeModel:()=>QD,weightsLoaderFactory:()=>F6,withSaveHandler:()=>w$,withSaveHandlerSync:()=>k$});var l$="model",u$=".json",c$=".weights.bin";function vv(e){return new Promise(t=>setTimeout(t)).then(e)}var Ku=class{constructor(e){if(!q().getBool("IS_BROWSER"))throw new Error("browserDownloads() cannot proceed because the current environment is not a browser.");e.startsWith(Ku.URL_SCHEME)&&(e=e.slice(Ku.URL_SCHEME.length)),(e==null||e.length===0)&&(e=l$),this.modelJsonFileName=e+u$,this.weightDataFileName=e+c$}async save(e){if(typeof document=="undefined")throw new Error("Browser downloads are not supported in this environment since `document` is not present");let t=window.URL.createObjectURL(new Blob([e.weightData],{type:"application/octet-stream"}));if(e.modelTopology instanceof ArrayBuffer)throw new Error("BrowserDownloads.save() does not support saving model topology in binary formats yet.");{let n=[{paths:["./"+this.weightDataFileName],weights:e.weightSpecs}],s=C6(e,n),r=window.URL.createObjectURL(new Blob([JSON.stringify(s)],{type:"application/json"})),a=this.modelJsonAnchor==null?document.createElement("a"):this.modelJsonAnchor;if(a.download=this.modelJsonFileName,a.href=r,await vv(()=>a.dispatchEvent(new MouseEvent("click"))),e.weightData!=null){let o=this.weightDataAnchor==null?document.createElement("a"):this.weightDataAnchor;o.download=this.weightDataFileName,o.href=t,await vv(()=>o.dispatchEvent(new MouseEvent("click")))}return{modelArtifactsInfo:ah(e)}}}};Ku.URL_SCHEME="downloads://";var d$=class{constructor(e){if(e==null||e.length<1)throw new Error(`When calling browserFiles, at least 1 file is required, but received ${e}`);this.jsonFile=e[0],this.weightsFiles=e.slice(1)}async load(){return new Promise((e,t)=>{let n=new FileReader;n.onload=s=>{let r=JSON.parse(s.target.result),a=r.modelTopology;if(a==null){t(new Error(`modelTopology field is missing from file ${this.jsonFile.name}`));return}if(r.weightsManifest==null){t(new Error(`weightManifest field is missing from file ${this.jsonFile.name}`));return}if(this.weightsFiles.length===0){e({modelTopology:a});return}let i=Ry(r,l=>this.loadWeights(l));e(i)},n.onerror=s=>t(`Failed to read model topology and weights manifest JSON from file '${this.jsonFile.name}'. BrowserFiles supports loading Keras-style tf.Model artifacts only.`),n.readAsText(this.jsonFile)})}loadWeights(e){let t=[],n=[];for(let a of e)t.push(...a.weights),n.push(...a.paths);let s=this.checkManifestAndWeightFiles(e),r=n.map(a=>this.loadWeightsFile(a,s[a]));return Promise.all(r).then(a=>[t,Ey(a)])}loadWeightsFile(e,t){return new Promise((n,s)=>{let r=new FileReader;r.onload=a=>{let o=a.target.result;n(o)},r.onerror=a=>s(`Failed to weights data from file of path '${e}'.`),r.readAsArrayBuffer(t)})}checkManifestAndWeightFiles(e){let t=[],n=this.weightsFiles.map(r=>bv(r.name)),s={};for(let r of e)r.paths.forEach(a=>{let o=bv(a);if(t.indexOf(o)!==-1)throw new Error(`Duplicate file basename found in weights manifest: '${o}'`);if(t.push(o),n.indexOf(o)===-1)throw new Error(`Weight file with basename '${o}' is not provided.`);s[a]=this.weightsFiles[n.indexOf(o)]});if(t.length!==this.weightsFiles.length)throw new Error(`Mismatch in the number of files in weights manifest (${t.length}) and the number of weight files provided (${this.weightsFiles.length}).`);return s}},p$=e=>q().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(Ku.URL_SCHEME)?h$(e.slice(Ku.URL_SCHEME.length)):null;qt.registerSaveRouter(p$);function h$(e="model"){return new Ku(e)}function f$(e){return new d$(e)}function wv(e,t,n,s){o(e),n=n==null?0:n,s=s==null?1:s,i(n,s);let r=0,a=l=>(l.then(u=>{let c=n+ ++r/e.length*(s-n);return t(c),u}),l);function o(l){M(l!=null&&Array.isArray(l)&&l.length>0,()=>"promises must be a none empty array")}function i(l,u){M(l>=0&&l<=1,()=>`Progress fraction must be in range [0, 1], but got startFraction ${l}`),M(u>=0&&u<=1,()=>`Progress fraction must be in range [0, 1], but got endFraction ${u}`),M(u>=l,()=>`startFraction must be no more than endFraction, but got startFraction ${l} and endFraction ${u}`)}return Promise.all(e.map(a))}async function P6(e,t){t==null&&(t={});let n=t.fetchFunc==null?q().platform.fetch:t.fetchFunc,s=e.map(p=>n(p,t.requestInit,{isBinary:!0})),r=0,a=.5,i=(t.onProgress==null?await Promise.all(s):await wv(s,t.onProgress,r,a)).map(p=>p.arrayBuffer()),l=.5,u=1;return t.onProgress==null?await Promise.all(i):await wv(i,t.onProgress,l,u)}async function m$(e,t="",n,s){return F6(o=>P6(o,{requestInit:s}))(e,t,n)}function F6(e){return async(t,n="",s)=>{let r=t.map(()=>!1),a={},o=s!=null?s.map(()=>!1):[],i=[];if(t.forEach((h,f)=>{let m=0;h.weights.forEach(g=>{let y="quantization"in g?g.quantization.dtype:g.dtype,x=I3[y]*Et(g.shape),A=()=>{r[f]=!0,a[f]==null&&(a[f]=[]),a[f].push({manifestEntry:g,groupOffset:m,sizeBytes:x})};s!=null?s.forEach((b,w)=>{b===g.name&&(A(),o[w]=!0)}):A(),i.push(g.name),m+=x})}),!o.every(h=>h)){let h=s.filter((f,m)=>!o[m]);throw new Error(`Could not find weights in manifest with names: ${h.join(", ")}.
|
|
Manifest JSON has weights with names: ${i.join(", ")}.`)}let l=r.reduce((h,f,m)=>(f&&h.push(m),h),[]),u=[];l.forEach(h=>{t[h].paths.forEach(f=>{let m=n+(n.endsWith("/")?"":"/")+f;u.push(m)})});let c=await e(u),p={},d=0;return l.forEach(h=>{let f=t[h].paths.length,m=0;for(let b=0;b<f;b++)m+=c[d+b].byteLength;let g=new ArrayBuffer(m),y=new Uint8Array(g),x=0;for(let b=0;b<f;b++){let w=new Uint8Array(c[d+b]);y.set(w,x),x+=w.byteLength}a[h].forEach(b=>{let w=g.slice(b.groupOffset,b.groupOffset+b.sizeBytes),k=S6(w,[b.manifestEntry]);for(let S in k)p[S]=k[S]}),d+=f}),p}}var g$="application/octet-stream",y$="application/json",Dy=class{constructor(e,t){if(this.DEFAULT_METHOD="POST",t==null&&(t={}),this.weightPathPrefix=t.weightPathPrefix,this.onProgress=t.onProgress,this.weightUrlConverter=t.weightUrlConverter,t.fetchFunc!=null?(M(typeof t.fetchFunc=="function",()=>"Must pass a function that matches the signature of `fetch` (see https://developer.mozilla.org/en-US/docs/Web/API/Fetch_API)"),this.fetch=t.fetchFunc):this.fetch=q().platform.fetch,M(e!=null&&e.length>0,()=>"URL path for http must not be null, undefined or empty."),Array.isArray(e)&&M(e.length===2,()=>`URL paths for http must have a length of 2, (actual length is ${e.length}).`),this.path=e,t.requestInit!=null&&t.requestInit.body!=null)throw new Error("requestInit is expected to have no pre-existing body, but has one.");this.requestInit=t.requestInit||{}}async save(e){if(e.modelTopology instanceof ArrayBuffer)throw new Error("BrowserHTTPRequest.save() does not support saving model topology in binary formats yet.");let t=Object.assign({method:this.DEFAULT_METHOD},this.requestInit);t.body=new FormData;let n=[{paths:["./model.weights.bin"],weights:e.weightSpecs}],s=C6(e,n);t.body.append("model.json",new Blob([JSON.stringify(s)],{type:y$}),"model.json"),e.weightData!=null&&t.body.append("model.weights.bin",new Blob([e.weightData],{type:g$}),"model.weights.bin");let r=await this.fetch(this.path,t);if(r.ok)return{modelArtifactsInfo:ah(e),responses:[r]};throw new Error(`BrowserHTTPRequest.save() failed due to HTTP response status ${r.status}.`)}async load(){let e=await this.fetch(this.path,this.requestInit);if(!e.ok)throw new Error(`Request to ${this.path} failed with status code ${e.status}. Please verify this URL points to the model JSON of the model to load.`);let t;try{t=await e.json()}catch(r){let a=`Failed to parse model JSON of response from ${this.path}.`;throw this.path.endsWith(".pb")?a+=" Your path contains a .pb file extension. Support for .pb models have been removed in TensorFlow.js 1.0 in favor of .json models. You can re-convert your Python TensorFlow model using the TensorFlow.js 1.0 conversion scripts or you can convert your.pb models with the 'pb2json'NPM script in the tensorflow/tfjs-converter repository.":a+=" Please make sure the server is serving valid JSON for this request.",new Error(a)}let n=t.modelTopology,s=t.weightsManifest;if(n==null&&s==null)throw new Error(`The JSON from HTTP path ${this.path} contains neither model topology or manifest for weights.`);return Ry(t,r=>this.loadWeights(r))}async loadWeights(e){let t=Array.isArray(this.path)?this.path[1]:this.path,[n,s]=A$(t),r=this.weightPathPrefix||n,a=[];for(let u of e)a.push(...u.weights);let o=[],i=[];for(let u of e)for(let c of u.paths)this.weightUrlConverter!=null?i.push(this.weightUrlConverter(c)):o.push(r+c+s);this.weightUrlConverter&&o.push(...await Promise.all(i));let l=await P6(o,{requestInit:this.requestInit,fetchFunc:this.fetch,onProgress:this.onProgress});return[a,Ey(l)]}};Dy.URL_SCHEME_REGEX=/^https?:\/\//;function A$(e){let t=e.lastIndexOf("/"),n=e.lastIndexOf("?"),s=e.substring(0,t),r=n>t?e.substring(n):"";return[s+"/",r]}function N3(e){return e.match(Dy.URL_SCHEME_REGEX)!=null}var O6=(e,t)=>{if(typeof fetch=="undefined"&&(t==null||t.fetchFunc==null))return null;{let n=!0;if(Array.isArray(e)?n=e.every(s=>N3(s)):n=N3(e),n)return $y(e,t)}return null};qt.registerSaveRouter(O6);qt.registerLoadRouter(O6);function $y(e,t){return new Dy(e,t)}function x$(e,t){return $y(e,t)}var o3=class{constructor(e){this.modelArtifacts=e}load(){return this.modelArtifacts}},M6=class{constructor(e){this.saveHandler=e}save(e){return this.saveHandler(e)}},b$=class{constructor(e){e.load&&(this.load=()=>Promise.resolve(e.load())),e.save&&(this.save=t=>Promise.resolve(e.save(t)))}};function v$(e,t,n,s){let r=arguments;return new b$(z6(...r))}function z6(e,t,n,s){return arguments.length===1?e.modelTopology!=null||e.weightSpecs!=null?new o3(e):(console.warn("Please call tf.io.fromMemory() with only one argument. The argument should be of type ModelArtifacts. The multi-argument signature of tf.io.fromMemory() has been deprecated and will be removed in a future release."),new o3({modelTopology:e})):(console.warn("Please call tf.io.fromMemory() with only one argument. The argument should be of type ModelArtifacts. The multi-argument signature of tf.io.fromMemory() has been deprecated and will be removed in a future release."),new o3({modelTopology:e,weightSpecs:t,weightData:n,trainingConfig:s}))}function w$(e){return new M6(e)}function k$(e){return new M6(e)}var L6={};Ve(L6,{confusionMatrix:()=>L$});function I$(e,t,n=!1,s=!1){let r=D(e,"a","matMul"),a=D(t,"b","matMul");[r,a]=Gt(r,a);let o={a:r,b:a},i={transposeA:n,transposeB:s};return W.runKernel(mo,o,i)}var Qe=V({matMul_:I$});function S$(e,t,n=1,s=0){if(t<2)throw new Error(`Error in oneHot: depth must be >=2, but it is ${t}`);let a={indices:D(e,"indices","oneHot","int32")},o={depth:t,onValue:n,offValue:s};return W.runKernel(El,a,o)}var Zu=V({oneHot_:S$});function Py(){q().set("PROD",!0)}function C$(){q().set("DEBUG",!0)}function T$(){q().set("DEPRECATION_WARNINGS_ENABLED",!1),console.warn("TensorFlow.js deprecation warnings have been disabled.")}function Fy(e){q().getBool("DEPRECATION_WARNINGS_ENABLED")&&console.warn(e+" You can disable deprecation warnings with tf.disableDeprecationWarnings().")}xD(Fy);function N$(){W.disposeVariables()}function sn(){return W}function ym(){return W.memory()}function E$(e){return W.profile(e)}function Y(e,t){return W.tidy(e,t)}function ee(e){Cy(e).forEach(n=>n.dispose())}function bn(e){return W.keep(e)}function R$(e){return W.time(e)}function Oy(e){return W.setBackend(e)}function Lc(){return W.ready()}function ss(){return W.backendName}function _$(e){W.removeBackend(e)}function My(e){return W.findBackend(e)}function D$(e){return W.findBackendFactory(e)}function ql(e,t,n=1){return W.registerBackend(e,t,n)}function Ys(){return W.backend}function $$(e,t){q().setPlatform(e,t)}function P$(e){let n={input:D(e,"input","imag")};return W.runKernel(Gp,n)}var oh=V({imag_:P$});function F$(e){let n={x:D(e,"x","neg")};return W.runKernel(Il,n)}var Dt=V({neg_:F$});function O$(e){let n={input:D(e,"input","real")};return W.runKernel(qp,n)}var Yu=V({real_:O$});function M$(e,t,n){let s=D(e,"x","transpose");if(t==null&&(t=s.shape.map((o,i)=>i).reverse()),M(s.rank===t.length,()=>`Error in transpose: rank of input ${s.rank} must match length of perm ${t}.`),t.forEach(o=>{M(o>=0&&o<s.rank,()=>`All entries in 'perm' must be between 0 and ${s.rank-1} but got ${t}`)}),s.rank<=1)return s.clone();let r={x:s},a={perm:t};return s.dtype==="complex64"?Y(()=>{let o=Yu(s),i=oh(s);return o=W.runKernel(Hr,{x:o},a),i=W.runKernel(Hr,{x:i},a),n&&(i=Dt(i)),fa(o,i)}):W.runKernel(Hr,r,a)}var et=V({transpose_:M$});function z$(e,t,n){let s=D(e,"labels","confusionMatrix"),r=D(t,"predictions","confusionMatrix");M(n==null||n>0&&Number.isInteger(n),()=>`If provided, numClasses must be a positive integer, but got ${n}`),M(s.rank===1,()=>`Expected the rank of labels to be 1, but got ${s.rank}`),M(r.rank===1,()=>`Expected the rank of predictions to be 1, but got ${r.rank}`),M(s.shape[0]===r.shape[0],()=>`Mismatch in the number of examples: ${s.shape[0]} vs. ${r.shape[0]}. Labels and predictions should have the same number of elements.`),M(n>0&&Number.isInteger(n),()=>`numClasses is required to be a positive integer, but got ${n}`);let a=Zu(ye(s,"int32"),n),o=Zu(ye(r,"int32"),n),i=et(a),l=Qe(i,o);return ye(l,"int32")}var L$=V({confusionMatrix_:z$}),Xl={};Ve(Xl,{assertAndGetBroadcastShape:()=>kt,getBroadcastDims:()=>B6,getReductionAxes:()=>an});function B6(e,t){let n=e.length,s=[];for(let r=0;r<n;r++){let a=n-1-r,o=e[a]||1;(t[t.length-1-r]||1)>1&&o===1&&s.unshift(a)}return s}function an(e,t){let n=[];for(let s=0;s<t.length;s++){let r=e[e.length-s-1],a=t.length-s-1,o=t[a];(r==null||r===1&&o>1)&&n.unshift(a)}return n}function kt(e,t){let n=[],s=Math.max(e.length,t.length);for(let r=0;r<s;r++){let a=e[e.length-r-1];a==null&&(a=1);let o=t[t.length-r-1];if(o==null&&(o=1),a===1)n.unshift(o);else if(o===1)n.unshift(a);else if(a!==o){let i=`Operands could not be broadcast together with shapes ${e} and ${t}.`;throw Error(i)}else n.unshift(a)}return n}var Js={};Ve(Js,{fromPixels:()=>j$,fromPixelsAsync:()=>G$,toPixels:()=>H$});function zy(e,t,n){if(ol(e),t!=null&&t.length!==3)throw new Error("tensor3d() requires shape to have three numbers");let s=Xr(e,n);if(s.length!==3&&s.length!==1)throw new Error("tensor3d() requires values to be number[][][] or flat/TypedArray");if(s.length===1&&t==null)throw new Error("tensor3d() requires shape to be provided when `values` are a flat array");return oi(e,t,s,n)}var Di;function W6(e,t=3){if(t>4)throw new Error("Cannot construct Tensor with more than 4 channels from pixels.");if(e==null)throw new Error("pixels passed to tf.browser.fromPixels() can not be null");let n=!1,s=!1,r=!1,a=!1,o=!1,i=!1;if(e.data instanceof Uint8Array)n=!0;else if(typeof ImageData!="undefined"&&e instanceof ImageData)s=!0;else if(typeof HTMLVideoElement!="undefined"&&e instanceof HTMLVideoElement)r=!0;else if(typeof HTMLImageElement!="undefined"&&e instanceof HTMLImageElement)a=!0;else if(e.getContext!=null)o=!0;else if(typeof ImageBitmap!="undefined"&&e instanceof ImageBitmap)i=!0;else throw new Error(`pixels passed to tf.browser.fromPixels() must be either an HTMLVideoElement, HTMLImageElement, HTMLCanvasElement, ImageData in browser, or OffscreenCanvas, ImageData in webworker or {data: Uint32Array, width: number, height: number}, but was ${e.constructor.name}`);if(r&&r&&e.readyState<2)throw new Error("The video element has not loaded data yet. Please wait for `loadeddata` event on the <video> element.");if(fm(bp,W.backendName)!=null){let f={pixels:e},m={numChannels:t};return W.runKernel(bp,f,m)}let[u,c]=r?[e.videoWidth,e.videoHeight]:[e.width,e.height],p;if(o)p=e.getContext("2d").getImageData(0,0,u,c).data;else if(s||n)p=e.data;else if(a||r||i){if(Di==null)if(typeof document=="undefined")if(typeof OffscreenCanvas!="undefined"&&typeof OffscreenCanvasRenderingContext2D!="undefined")Di=new OffscreenCanvas(1,1).getContext("2d");else throw new Error("Cannot parse input in current context. Reason: OffscreenCanvas Context2D rendering is not supported.");else Di=document.createElement("canvas").getContext("2d",{willReadFrequently:!0});Di.canvas.width=u,Di.canvas.height=c,Di.drawImage(e,0,0,u,c),p=Di.getImageData(0,0,u,c).data}let d;if(t===4)d=new Int32Array(p);else{let f=u*c;d=new Int32Array(f*t);for(let m=0;m<f;m++)for(let g=0;g<t;++g)d[m*t+g]=p[m*4+g]}return zy(d,[c,u,t],"int32")}function B$(e){return e!=null&&e.data instanceof Uint8Array}function W$(){return typeof window!="undefined"&&typeof ImageBitmap!="undefined"&&window.hasOwnProperty("createImageBitmap")}function V$(e){return e!=null&&e.width!==0&&e.height!==0}function U$(e){return W$()&&!(e instanceof ImageBitmap)&&V$(e)&&!B$(e)}async function G$(e,t=3){let n=null;if(q().getBool("WRAP_TO_IMAGEBITMAP")&&U$(e)){let s;try{s=await createImageBitmap(e,{premultiplyAlpha:"none"})}catch(r){s=null}s!=null&&s.width===e.width&&s.height===e.height?n=s:n=e}else n=e;return W6(n,t)}async function H$(e,t){let n=D(e,"img","toPixels");if(!(e instanceof nt)){let u=n;n=ye(u,"int32"),u.dispose()}if(n.rank!==2&&n.rank!==3)throw new Error(`toPixels only supports rank 2 or 3 tensors, got rank ${n.rank}.`);let[s,r]=n.shape.slice(0,2),a=n.rank===2?1:n.shape[2];if(a>4||a===2)throw new Error(`toPixels only supports depth of size 1, 3 or 4 but got ${a}`);if(n.dtype!=="float32"&&n.dtype!=="int32")throw new Error(`Unsupported type for toPixels: ${n.dtype}. Please use float32 or int32 tensors.`);let o=await n.data(),i=n.dtype==="float32"?255:1,l=new Uint8ClampedArray(r*s*4);for(let u=0;u<s*r;++u){let c=[0,0,0,255];for(let d=0;d<a;d++){let h=o[u*a+d];if(n.dtype==="float32"){if(h<0||h>1)throw new Error(`Tensor values for a float32 Tensor must be in the range [0 - 1] but encountered ${h}.`)}else if(n.dtype==="int32"&&(h<0||h>255))throw new Error(`Tensor values for a int32 Tensor must be in the range [0 - 255] but encountered ${h}.`);a===1?(c[0]=h*i,c[1]=h*i,c[2]=h*i):c[d]=h*i}let p=u*4;l[p+0]=Math.round(c[0]),l[p+1]=Math.round(c[1]),l[p+2]=Math.round(c[2]),l[p+3]=Math.round(c[3])}if(t!=null){t.width=r,t.height=s;let u=t.getContext("2d"),c=new ImageData(l,r,s);u.putImageData(c,0,0)}return n!==e&&n.dispose(),l}var j$=V({fromPixels_:W6}),Ly={};Ve(Ly,{prepareAndValidate:()=>V6});function V6(e,t){let n=e.shape.length,s=t.shape.length;if(n<1)throw new Error(`tf.gatherND() expects the input to be rank 1 or higher, but the rank was ${n}.`);if(s<1)throw new Error(`tf.gatherND() expects the indices to be rank 1 or higher, but the rank was ${s}.`);if(t.dtype!=="int32")throw new Error(`tf.gatherND() expects the indices to be int32 type, but the dtype was ${t.dtype}.`);if(t.shape[s-1]>n)throw new Error(`index innermost dimension length must be <= tensor rank; saw: ${t.shape[s-1]} vs. ${n}`);if(Et(e.shape)===0)throw new Error(`Requested more than 0 entries, but input is empty. Input shape: ${e.shape}.`);let r=t.shape,a=r[r.length-1],o=1;for(let p=0;p<r.length-1;++p)o*=r[p];let i=e.shape,l=r.slice();l.pop();let u=1;for(let p=a;p<n;++p)u*=i[p],l.push(i[p]);let c=[...dc(e.shape).map(p=>p/u),1].slice(0,a);return[l,o,u,c]}var By={};Ve(By,{calculateShapes:()=>U6,validateInput:()=>Vy,validateUpdateShape:()=>Wy});function Wy(e,t,n){let s=t.rank>1?t.shape[t.rank-1]:1,r=t.rank>1?t.rank-1:1,a=`Must have updates.shape = indices.shape[:batchDim] + shape[sliceDim:], got updates.shape: ${n.shape}, indices.shape: ${t.shape}, shape: ${e}, sliceDim: ${s}, and batchDim: ${r}.`;if(n.rank<r)throw new Error(a+` update.rank < ${r}. `);if(e.length<s+(n.rank-r))throw new Error(a+` Output shape length < ${s+(n.rank-r)}`);if(n.rank!==r+e.length-s)throw new Error(a+` update.rank != ${r+e.length-s}`);for(let o=0;o<r;++o)if(n.shape[o]!==t.shape[o])throw new Error(a+` updates.shape[${o}] (${n.shape[o]}) != indices.shape[${o}] (${t.shape[o]}).`);for(let o=0;o<n.rank-r;++o)if(n.shape[o+r]!==e[o+s])throw new Error(a+` updates.shape[${o+r}] (${n.shape[o+r]}) != shape[${o+r}] (${e[o+r]})`)}function Vy(e,t,n){if(t.rank<1)throw new Error(`tf.scatterND() expects the indices to be rank 1 or higher, but the rank was ${t.rank}.`);if(e.rank<1)throw new Error(`tf.scatterND() expects the updates to be rank 1 or higher, but the rank was ${e.rank}.`);if(t.dtype!=="int32")throw new Error(`The dtype of 'indices' should be int32, but got dtype: ${t.dtype}`);if(n.length<1)throw new Error(`Output rank must be greater or equal to 1, but got shape: ${n}`);if(n.length===0){if(t.size===0)throw new Error(`Indices specified for empty output. indices shape: ${t.shape}`);if(e.size===0)throw new Error(`Updates specified for empty output. updates shape: ${e.shape}`)}Wy(n,t,e)}function U6(e,t,n){let s=t.shape.length,r=s>1?t.shape[s-1]:1,a=n.length,o=1;for(let p=r;p<a;++p)o*=n[p];let i=r<1?1:r,l=Et(t.shape)/i,u=[...dc(n.slice(0,r)),1],c=Et(n);return{sliceRank:r,numUpdates:l,sliceSize:o,strides:u,outputSize:c}}var Ut={};Ve(Ut,{assertParamsValid:()=>X$,computeFlatOffset:()=>Q$,computeOutShape:()=>Z$,getNormalizedAxes:()=>Y$,isSliceContinous:()=>J$,maskToAxes:()=>K$,parseSliceParams:()=>J6,sliceInfo:()=>eP,startForAxis:()=>Z6,startIndicesWithElidedDims:()=>q6,stopForAxis:()=>Y6,stopIndicesWithElidedDims:()=>X6,stridesForAxis:()=>K6,stridesWithElidedDims:()=>G6});var E3=-2,q$=-1;function X$(e,t,n){let s=e.shape.length;M(s===t.length,()=>`Error in slice${s}D: Length of begin ${t} must match the rank of the array (${s}).`),M(s===n.length,()=>`Error in slice${s}D: Length of size ${n} must match the rank of the array (${s}).`);for(let r=0;r<s;++r)M(t[r]+n[r]<=e.shape[r],()=>`Error in slice${s}D: begin[${r}] + size[${r}] (${t[r]+n[r]}) would overflow input.shape[${r}] (${e.shape[r]})`)}function K$(e){let t=[],n=0;for(;e>0;)e&1&&t.push(n),e/=2,n++;return t}function Z$(e,t,n){let s=[];for(let r=0;r<e.length;r++)s[r]=Math.ceil((t[r]-e[r])/n[r]);return s}function G6(e,t,n,s){let r=[...e];for(let a=r.length;a<s.length;a++)r.push(1);for(let a=0;a<n;a++)a===0?r[t]=1:(r.splice(t,0,1),r.pop());return r}function H6(e,t,n){return n<=e?n:n-(t-1)}function j6(e,t){let n=[];for(let s=0;s<e;s++)n.push(t+s);return n}function Y$(e,t,n,s,r,a,o,i,l){let u=e.length,c=new Array(u),p=new Array(u),d=new Array(u);if(t.length&&n>0){let h=t[0],f=n+1;c=q6(o,h,f,s,e),p=X6(i,h,f,r,e),d=G6(a,h,f,e)}else for(let h=0;h<u;h++)c[h]=Z6(o,s,a,e,h,l),p[h]=Y6(i,r,a,e,h,l),d[h]=K6(a,h,l);return{begin:c,end:p,strides:d}}function q6(e,t,n,s,r){let a=[...r],o=j6(n,t);for(let i=0;i<a.length;i++)if(o.indexOf(i)>-1)a[i]=0;else{let l=H6(t,n,i),u=s[l];e&1<<l&&(u=0),a[i]=u}return a}function X6(e,t,n,s,r){let a=[...r],o=j6(n,t);for(let i=0;i<a.length;i++)if(o.indexOf(i)>-1)a[i]=Number.MAX_SAFE_INTEGER;else{let l=H6(t,n,i),u=s[l];e&1<<l&&(u=Number.MAX_SAFE_INTEGER),a[i]=u}for(let i=0;i<a.length;i++){let l=r[i];a[i]<0&&(a[i]+=l),a[i]=xp(0,a[i],r[i])}return a}function K6(e,t,n){let s=e[t];return(n&1<<t||s==null)&&(s=1),s}function Z6(e,t,n,s,r,a){let o=t[r],i=n[r]||1;(e&1<<r||a&1<<r||o==null)&&(i>0?o=Number.MIN_SAFE_INTEGER:o=Number.MAX_SAFE_INTEGER);let l=s[r];return o<0&&(o+=l),o=xp(0,o,l-1),o}function Y6(e,t,n,s,r,a){let o=t[r],i=n[r]||1;(e&1<<r||a&1<<r||o==null)&&(i>0?o=Number.MAX_SAFE_INTEGER:o=Number.MIN_SAFE_INTEGER);let l=s[r];return o<0&&(o+=l),i>0?o=xp(0,o,l):o=xp(-1,o,l-1),o}function J$(e,t,n){let s=n.length;for(let r=0;r<n.length;r++)if(n[r]>1){s=r;break}for(let r=s+1;r<n.length;r++)if(t[r]>0||n[r]!==e[r])return!1;return!0}function Q$(e,t){let n=e.length>0?e[e.length-1]:1;for(let s=0;s<e.length-1;s++)n+=e[s]*t[s];return n}function J6(e,t,n){let s,r=e.shape.length;typeof t=="number"?s=[t,...new Array(r-1).fill(0)]:t.length<r?s=t.concat(new Array(r-t.length).fill(0)):s=t.slice(),s.forEach(o=>{M(o!==-1,()=>"slice() does not support negative begin indexing.")});let a;return n==null?a=new Array(r).fill(-1):typeof n=="number"?a=[n,...new Array(r-1).fill(-1)]:n.length<r?a=n.concat(new Array(r-n.length).fill(-1)):a=n,a=a.map((o,i)=>o>=0?o:(M(o===-1,()=>`Negative size values should be exactly -1 but got ${o} for the slice() size at index ${i}.`),e.shape[i]-s[i])),[s,a]}function eP(e,t,n,s,r,a,o,i,l){let u;if(s==null?(u=new Array(t.length),u.fill(1)):u=s,o!=null&&(o&o-1)!==0)throw new Error("Multiple ellipses in slice is not allowed.");let c=!1,p={dims:u.length,numAddAxisAfterEllipsis:0,begin:t.slice(),end:n.slice(),strides:u.slice(),beginMask:r,endMask:a,ellipsisMask:o,newAxisMask:i,shrinkAxisMask:l};for(let A=0;A<p.dims;A++)c&&(1<<A&i)!==0&&p.numAddAxisAfterEllipsis++,1<<A&o&&(c=!0);c||(p.ellipsisMask|=1<<p.dims,p.dims++);let d={dims:e.length,beginMask:0,endMask:0,beginValid:!1,endValid:!1};tP(p,d);let h=!0,f=!0,m=!0,g=[],y=[];for(let A=0;A<e.length;++A){if(d.strides[A]===0)throw Error(`strides[${A}] must be non-zero`);let b=!!(d.shrinkAxisMask&1<<A),w=e[A];if(w===-1){g.push(b?1:-1);continue}let k=[d.beginMask&1<<A,d.endMask&1<<A],S=[d.strides[A]>0?0:-1,d.strides[A]>0?w:w-1];if(b&&d.strides[A]<=0)throw Error("only stride 1 allowed on non-range indexing.");m=m&&d.strides[A]===1;let E=!!(d.beginMask&1<<A&&d.endMask&1<<A);if(d.beginValid&&d.endValid){if(b){let P=d.begin[A]<0?w+d.begin[A]:d.begin[A];if(d.begin[A]=P,d.end[A]=d.begin[A]+1,P<0||P>=w)throw Error(`slice index ${d.begin[A]} of dimension ${A} out of bounds.`)}else d.begin[A]=kv(d.begin[A],0,d.strides[A],w,k,S),d.end[A]=kv(d.end[A],1,d.strides[A],w,k,S);let _=d.strides[A]===1&&d.begin[A]===0&&d.end[A]===w;h=h&&_,f=f&&(A===0&&d.strides[A]===1||_)}else h=h&&d.strides[A]===1&&E,f=f&&(A===0&&d.strides[A]===1||E);let R,$=!1;if(d.beginValid&&d.endValid?(R=d.end[A]-d.begin[A],$=!0):b?(R=1,$=!0):E&&w>=0&&(d.strides[A]<0?R=-w:R=w,$=!0),$){let _;R===0||R<0!=d.strides[A]<0?_=0:_=Math.trunc(R/d.strides[A])+(R%d.strides[A]!==0?1:0),g.push(_)}else g.push(-1)}for(let A=0;A<d.finalShapeGatherIndices.length;++A){let b=d.finalShapeGatherIndices[A];b>=0?y.push(g[b]):b===E3&&y.push(1)}return{finalShapeSparse:y.filter((A,b)=>d.finalShapeGatherIndices[b]!==E3),finalShape:y,isIdentity:h,sliceDim0:f,isSimpleSlice:m,begin:d.begin,end:d.end,strides:d.strides}}function tP(e,t){t.beginMask=0,t.endMask=0,t.shrinkAxisMask=0;let n=0;t.beginValid=e.begin!=null,t.endValid=e.end!=null,t.begin=new Array(t.dims),t.end=new Array(t.dims),t.strides=new Array(t.dims),t.finalShapeGatherIndices=[],t.finalShapeGatherIndicesSparse=[],t.inputShapeGatherIndicesSparse=new Array(t.dims);for(let s=0;s<e.dims;s++)if(1<<s&e.ellipsisMask){let r=Math.min(t.dims-(e.dims-s)+1+e.numAddAxisAfterEllipsis,t.dims);for(;n<r;n++)t.begin[n]=0,t.end[n]=0,t.strides[n]=1,t.beginMask|=1<<n,t.endMask|=1<<n,t.finalShapeGatherIndices.push(n),t.finalShapeGatherIndicesSparse.push(-1),t.inputShapeGatherIndicesSparse[n]=s}else if(1<<s&e.newAxisMask)t.finalShapeGatherIndices.push(E3),t.finalShapeGatherIndicesSparse.push(-1);else{if(n===t.begin.length)throw Error(`Index out of range using input dim ${n}; input has only ${t.dims} dims, ${t.begin.length}.`);e.begin!=null&&(t.begin[n]=e.begin[s]),e.end!=null&&(t.end[n]=e.end[s]),t.strides[n]=e.strides[s],e.beginMask&1<<s&&(t.beginMask|=1<<n),e.endMask&1<<s&&(t.endMask|=1<<n),e.shrinkAxisMask&1<<s?(t.finalShapeGatherIndices.push(q$),t.finalShapeGatherIndicesSparse.push(-1),t.shrinkAxisMask|=1<<n):(t.finalShapeGatherIndices.push(n),t.finalShapeGatherIndicesSparse.push(s)),t.inputShapeGatherIndicesSparse[n]=s,n++}}function kv(e,t,n,s,r,a){if(r[t])return n>0?a[t]:a[t+1&1];{let o=e<0?s+e:e;return o<a[0]?a[0]:o>a[1]?a[1]:o}}var de={};Ve(de,{Serializable:()=>Q6,SerializationMap:()=>zi,registerClass:()=>ii});var Q6=class{getClassName(){return this.constructor.className}static fromConfig(e,t){return new e(t)}},zi=class{constructor(){this.classNameMap={}}static getMap(){return zi.instance==null&&(zi.instance=new zi),zi.instance}static register(e){zi.getMap().classNameMap[e.className]=[e,e.fromConfig]}};function ii(e){M(e.className!=null,()=>"Class being registered does not have the static className property defined."),M(typeof e.className=="string",()=>"className is required to be a string, but got type "+typeof e.className),M(e.className.length>0,()=>"Class being registered has an empty-string as its className, which is disallowed."),zi.register(e)}var ew={};Ve(ew,{TEST_EPSILON_FLOAT16:()=>tw,encodeStrings:()=>nw,expectArrayBuffersEqual:()=>lP,expectArraysClose:()=>sP,expectArraysEqual:()=>aP,expectNumbersClose:()=>oP,expectPromiseToFail:()=>rP,expectValuesInRange:()=>iP,testEpsilon:()=>Uy});var nP=.001,tw=.1;function sP(e,t,n){return n==null&&(n=Uy()),R3(e,t,(s,r)=>Gy(s,r,n))}function Uy(){return W.backend.floatPrecision()===32?nP:tw}function R3(e,t,n){let s=!0;if((Fn(e)||Fn(t))&&(s=!1),Fn(e)&&Fn(t)&&(s=!0),s){let o=e.constructor.name,i=t.constructor.name;if(o!==i)throw new Error(`Arrays are of different type. Actual: ${o}. Expected: ${i}`)}if(Array.isArray(e)&&Array.isArray(t)){let o=Xr(e),i=Xr(t);if(!co(o,i))throw new Error(`Arrays have different shapes. Actual: [${o}]. Expected: [${i}]`)}let r=Fn(e)?e:Xi(e),a=Fn(t)?t:Xi(t);if(r.length!==a.length)throw new Error(`Arrays have different lengths actual: ${r.length} vs expected: ${a.length}.
|
|
Actual: ${r}.
|
|
Expected: ${a}.`);for(let o=0;o<a.length;++o){let i=r[o],l=a[o];if(!n(i,l))throw new Error(`Arrays differ: actual[${o}] = ${i}, expected[${o}] = ${l}.
|
|
Actual: ${r}.
|
|
Expected: ${a}.`)}}function rP(e,t){e().then(()=>t.fail(),()=>t())}function aP(e,t){let n=typeof t=="string"||typeof t=="number"||typeof t=="boolean"?[t]:t;return Ga(e)||Ga(e[0])||Ga(t)||Ga(t[0])?R3(e,n,(s,r)=>s==r):R3(e,t,(s,r)=>Gy(s,r,0))}function oP(e,t,n){if(n==null&&(n=Uy()),!Gy(e,t,n))throw new Error(`Numbers differ: actual === ${e}, expected === ${t}`)}function Gy(e,t,n){return!isFinite(e)&&!isFinite(t)?!0:!(isNaN(e)||isNaN(t)||Math.abs(e-t)>n)}function iP(e,t,n){for(let s=0;s<e.length;s++)if(e[s]<t||e[s]>n)throw new Error(`Value out of range:${e[s]} low: ${t}, high: ${n}`)}function lP(e,t){let n=new Float32Array(e),s=new Float32Array(t);if(n.length!==s.length)throw new Error(`Expected ArrayBuffer to be of length ${s.length}, but it was ${n.length}`);for(let r=0;r<s.length;r++)if(n[r]!==s[r])throw new Error(`Expected ArrayBuffer value at ${r} to be ${s[r]} but got ${n[r]} instead`)}function nw(e){for(let t=0;t<e.length;t++){let n=e[t];Array.isArray(n)?nw(n):e[t]=nh(n)}return e}var Hy="3.19.0";function uP(e,t){let n=D(e,"a","add"),s=D(t,"b","add");[n,s]=Gt(n,s);let r={a:n,b:s};return W.runKernel(xa,r)}var ce=V({add_:uP});function cP(e,t){let n=D(e,"a","floorDiv"),s=D(t,"b","floorDiv");[n,s]=Gt(n,s);let r={a:n,b:s};return W.runKernel(No,r)}var Bc=V({floorDiv_:cP});function dP(e,t){let n=D(e,"a","div"),s=D(t,"b","div");if([n,s]=Gt(n,s),n.dtype==="int32"&&s.dtype==="int32")return Bc(n,s);let r={a:n,b:s},a={};return W.runKernel(Io,r,a)}var he=V({div_:dP});function pP(e,t){let n=D(e,"a","mul"),s=D(t,"b","mul");[n,s]=Gt(n,s);let r={a:n,b:s};return W.runKernel(Wo,r)}var L=V({mul_:pP});function hP(e){let t=D(e,"x","abs");if(t.dtype==="complex64"){let n={x:t};return W.runKernel(Bp,n)}else{let n={x:t};return W.runKernel(il,n)}}var tn=V({abs_:hP});function fP(e){let n={x:D(e,"x","acos")};return W.runKernel(pc,n)}var jy=V({acos_:fP});function mP(e){let n={x:D(e,"x","acosh")};return W.runKernel(hc,n)}var qy=V({acosh_:mP});function gP(e){M(Array.isArray(e),()=>"The argument passed to tf.addN() must be a list of tensors"),M(e.length>=1,()=>`Must pass at least one tensor to tf.addN(), but got ${e.length}`);let t=e.map((r,a)=>D(r,`tensors${a}`,"addN")),n=t[0];t.forEach(r=>{if(r.dtype!==n.dtype)throw new Error("All tensors passed to tf.addN() must have the same dtype")}),t.forEach(r=>{if(!co(r.shape,n.shape))throw new Error("All tensors passed to tf.addN() must have the same shape")});let s=t;return W.runKernel(po,s)}var m0=V({addN_:gP});function yP(e,t=null,n=!1){let r={x:D(e,"x","all","bool")},a={axis:t,keepDims:n};return W.runKernel(fc,r,a)}var g0=V({all_:yP});function AP(e,t=null,n=!1){let r={x:D(e,"x","any","bool")},a={axis:t,keepDims:n};return W.runKernel(mc,r,a)}var Cp=V({any_:AP});function xP(e,t=0){let s={x:D(e,"x","argMax")},r={axis:t};return W.runKernel(ho,s,r)}var Es=V({argMax_:xP});function bP(e,t=0){let s={x:D(e,"x","argMin")},r={axis:t};return W.runKernel(gc,s,r)}var Xy=V({argMin_:bP});function vP(e){let n={x:D(e,"x","asin")};return W.runKernel(yc,n)}var Ky=V({asin_:vP});function wP(e){let n={x:D(e,"x","asinh")};return W.runKernel(Ac,n)}var Zy=V({asinh_:wP});function kP(e){let n={x:D(e,"x","atan")};return W.runKernel(xc,n)}var Yy=V({atan_:kP});function IP(e,t){let n=D(e,"a","atan2"),s=D(t,"b","atan2");[n,s]=Gt(n,s);let r={a:n,b:s};return W.runKernel(vc,r)}var Jy=V({atan2_:IP});function SP(e){let n={x:D(e,"x","atanh")};return W.runKernel(bc,n)}var Qy=V({atanh_:SP});function CP(e,t,n,s,r="NHWC",a){let o=e[3],i=[...t,o],l=aw(r);return ih(e,i,n,a,s,null,null,l)}function sw(e,t,n,s,r,a,o="channelsLast"){let[i,l]=Am(t),u;if(o==="channelsLast")u=[i,l,e[3],e[3]];else if(o==="channelsFirst")u=[i,l,e[1],e[1]];else throw new Error(`Unknown dataFormat ${o}`);return ih(e,u,n,s,r,a,!1,o)}function TP(e,t,n,s,r,a,o="NDHWC"){let[i,l,u]=_3(t),c,p;if(o==="NDHWC")p="channelsLast",c=[i,l,u,e[4],e[4]];else if(o==="NCDHW")p="channelsFirst",c=[i,l,u,e[1],e[1]];else throw new Error(`Unknown dataFormat ${o}`);return rw(e,c,n,s,r,!1,p,a)}function ih(e,t,n,s,r,a,o=!1,i="channelsLast"){let[l,u,c,p]=[-1,-1,-1,-1];if(i==="channelsLast")[l,u,c,p]=e;else if(i==="channelsFirst")[l,p,u,c]=e;else throw new Error(`Unknown dataFormat ${i}`);let[d,h,,f]=t,[m,g]=Am(n),[y,x]=Am(s),A=Uu(d,y),b=Uu(h,x),{padInfo:w,outHeight:k,outWidth:S}=RP(r,u,c,m,g,A,b,a,i),E=o?f*p:f,R;return i==="channelsFirst"?R=[l,E,k,S]:i==="channelsLast"&&(R=[l,k,S,E]),{batchSize:l,dataFormat:i,inHeight:u,inWidth:c,inChannels:p,outHeight:k,outWidth:S,outChannels:E,padInfo:w,strideHeight:m,strideWidth:g,filterHeight:d,filterWidth:h,effectiveFilterHeight:A,effectiveFilterWidth:b,dilationHeight:y,dilationWidth:x,inShape:e,outShape:R,filterShape:t}}function rw(e,t,n,s,r,a=!1,o="channelsLast",i){let[l,u,c,p,d]=[-1,-1,-1,-1,-1];if(o==="channelsLast")[l,u,c,p,d]=e;else if(o==="channelsFirst")[l,d,u,c,p]=e;else throw new Error(`Unknown dataFormat ${o}`);let[h,f,m,,g]=t,[y,x,A]=_3(n),[b,w,k]=_3(s),S=Uu(h,b),E=Uu(f,w),R=Uu(m,k),{padInfo:$,outDepth:_,outHeight:P,outWidth:C}=_P(r,u,c,p,y,x,A,S,E,R,i),F=a?g*d:g,G;return o==="channelsFirst"?G=[l,F,_,P,C]:o==="channelsLast"&&(G=[l,_,P,C,F]),{batchSize:l,dataFormat:o,inDepth:u,inHeight:c,inWidth:p,inChannels:d,outDepth:_,outHeight:P,outWidth:C,outChannels:F,padInfo:$,strideDepth:y,strideHeight:x,strideWidth:A,filterDepth:h,filterHeight:f,filterWidth:m,effectiveFilterDepth:S,effectiveFilterHeight:E,effectiveFilterWidth:R,dilationDepth:b,dilationHeight:w,dilationWidth:k,inShape:e,outShape:G,filterShape:t}}function NP(e,t,n,s,r){s==null&&(s=eA(e,t,n));let a=e[0],o=e[1],i=Gi((a-t+2*s)/n+1,r),l=Gi((o-t+2*s)/n+1,r);return[i,l]}function EP(e,t,n,s,r,a){r==null&&(r=eA(e,t,s));let o=e[0],i=e[1],l=e[2],u=Gi((o-t+2*r)/s+1,a),c=Gi((i-t+2*r)/s+1,a),p=Gi((l-t+2*r)/s+1,a);return[u,c,p,n]}function eA(e,t,n,s=1){let r=Uu(t,s);return Math.floor((e[0]*(n-1)-n+r)/2)}function Am(e){return typeof e=="number"?[e,e,e]:e.length===2?[e[0],e[1],1]:e}function _3(e){return typeof e=="number"?[e,e,e]:e}function Uu(e,t){return t<=1?e:e+(e-1)*(t-1)}function RP(e,t,n,s,r,a,o,i,l){let u,c,p;if(typeof e=="number"){u={top:e,bottom:e,left:e,right:e,type:e===0?"VALID":"NUMBER"};let h=NP([t,n],a,s,e,i);c=h[0],p=h[1]}else if(e==="same"){c=Math.ceil(t/s),p=Math.ceil(n/r);let d=Math.max(0,(c-1)*s+a-t),h=Math.max(0,(p-1)*r+o-n),f=Math.floor(d/2),m=d-f,g=Math.floor(h/2),y=h-g;u={top:f,bottom:m,left:g,right:y,type:"SAME"}}else if(e==="valid")u={top:0,bottom:0,left:0,right:0,type:"VALID"},c=Math.ceil((t-a+1)/s),p=Math.ceil((n-o+1)/r);else if(typeof e=="object"){let d=l==="channelsLast"?e[1][0]:e[2][0],h=l==="channelsLast"?e[1][1]:e[2][1],f=l==="channelsLast"?e[2][0]:e[3][0],m=l==="channelsLast"?e[2][1]:e[3][1];u={top:d,bottom:h,left:f,right:m,type:d===0&&h===0&&f===0&&m===0?"VALID":"EXPLICIT"},c=Gi((t-a+d+h)/s+1,i),p=Gi((n-o+f+m)/r+1,i)}else throw Error(`Unknown padding parameter: ${e}`);return{padInfo:u,outHeight:c,outWidth:p}}function _P(e,t,n,s,r,a,o,i,l,u,c){let p,d,h,f;if(typeof e=="number"){p={top:e,bottom:e,left:e,right:e,front:e,back:e,type:e===0?"VALID":"NUMBER"};let g=EP([t,n,s,1],i,1,r,e,c);d=g[0],h=g[1],f=g[2]}else if(e==="same"){d=Math.ceil(t/r),h=Math.ceil(n/a),f=Math.ceil(s/o);let m=(d-1)*r+i-t,g=(h-1)*a+l-n,y=(f-1)*o+u-s,x=Math.floor(m/2),A=m-x,b=Math.floor(g/2),w=g-b,k=Math.floor(y/2),S=y-k;p={top:b,bottom:w,left:k,right:S,front:x,back:A,type:"SAME"}}else if(e==="valid")p={top:0,bottom:0,left:0,right:0,front:0,back:0,type:"VALID"},d=Math.ceil((t-i+1)/r),h=Math.ceil((n-l+1)/a),f=Math.ceil((s-u+1)/o);else throw Error(`Unknown padding parameter: ${e}`);return{padInfo:p,outDepth:d,outHeight:h,outWidth:f}}function Gi(e,t){if(!t)return Math.trunc(e);switch(t){case"round":return Math.round(e);case"ceil":return Math.ceil(e);case"floor":return Math.floor(e);default:throw new Error(`Unknown roundingMode ${t}`)}}function to(e){let[t,n,s]=Am(e);return t===1&&n===1&&s===1}function Yr(e,t){return to(e)||to(t)}function aw(e){if(e==="NHWC")return"channelsLast";if(e==="NCHW")return"channelsFirst";throw new Error(`Unknown dataFormat ${e}`)}function rs(e,t,n){if(n!=null){if(typeof t=="string")throw Error(`Error in ${e}: pad must be an integer when using dimRoundingMode ${n} but got pad ${t}.`);if(typeof t=="number")M(qu(t),()=>`Error in ${e}: pad must be an integer when using dimRoundingMode ${n} but got pad ${t}.`);else if(typeof t=="object")t.forEach(s=>{s.forEach(r=>{M(qu(r),()=>`Error in ${e}: pad must be an integer when using dimRoundingMode ${n} but got pad ${r}.`)})});else throw Error(`Error in ${e}: Unknown padding parameter: ${t}`)}}function DP(e,t){let s={x:D(e,"x","reshape","string_or_numeric")},r={shape:t};return W.runKernel(_l,s,r)}var U=V({reshape_:DP});function $P(e,t,n,s,r){let a=D(e,"x","avgPool","float32"),o=1;M(Yr(n,o),()=>`Error in avgPool: Either strides or dilations must be 1. Got strides ${n} and dilations '${o}'`);let i=a,l=!1;a.rank===3&&(l=!0,i=U(a,[1,a.shape[0],a.shape[1],a.shape[2]])),M(i.rank===4,()=>`Error in avgPool: x must be rank 4 but got rank ${i.rank}.`),rs("avgPool",s,r);let u={x:i},c={filterSize:t,strides:n,pad:s,dimRoundingMode:r},p=W.runKernel(fo,u,c);return p=ye(p,a.dtype),l?U(p,[p.shape[1],p.shape[2],p.shape[3]]):p}var lh=V({avgPool_:$P});function PP(e,t,n,s,r,a="NDHWC"){let o=D(e,"x","avgPool3d","float32"),i=o,l=!1;o.rank===4&&(l=!0,i=U(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),M(i.rank===5,()=>`Error in avgPool3d: x must be rank 5 but got rank ${i.rank}.`),M(a==="NDHWC",()=>`Error in avgPool3d: Only NDHWC is currently supported, but got dataFormat of ${a}`),rs("avgPool3d",s,r);let u={x:i},c={filterSize:t,strides:n,pad:s,dimRoundingMode:r,dataFormat:a},p=W.runKernel(zp,u,c);return p=ye(p,i.dtype),l?U(p,[p.shape[1],p.shape[2],p.shape[3],p.shape[4]]):p}var tA=V({avgPool3d_:PP});function FP(e,t=0){M(e.length>=1,()=>"Pass at least one tensor to concat");let n=Sp(e,"tensors","concat","string_or_numeric");if(n[0].dtype==="complex64"&&n.forEach(a=>{if(a.dtype!=="complex64")throw new Error(`Cannot concatenate complex64 tensors with a tensor
|
|
with dtype ${a.dtype}. `)}),n.length===1)return On(n[0]);let s=n,r={axis:t};return W.runKernel(ul,s,r)}var Ct=V({concat_:FP});function OP(e){let n={x:D(e,"x","sigmoid","float32")};return W.runKernel(Jo,n)}var Cn=V({sigmoid_:OP});function MP(e,t,n){let s=D(e,"x","slice","string_or_numeric");if(s.rank===0)throw new Error("Slicing scalar is not possible");let r={x:s},a={begin:t,size:n};return W.runKernel(Ol,r,a)}var Oe=V({slice_:MP});function zP(e){let n={x:D(e,"x","tanh","float32")};return W.runKernel(ri,n)}var Yi=V({tanh_:zP});function LP(e,t,n,s,r,a){let o=D(e,"forgetBias","basicLSTMCell"),i=D(t,"lstmKernel","basicLSTMCell"),l=D(n,"lstmBias","basicLSTMCell"),u=D(s,"data","basicLSTMCell"),c=D(r,"c","basicLSTMCell"),p=D(a,"h","basicLSTMCell"),d=Ct([u,p],1),h=Qe(d,i),f=ce(h,l),m=f.shape[0],g=f.shape[1]/4,y=[m,g],x=Oe(f,[0,0],y),A=Oe(f,[0,g],y),b=Oe(f,[0,g*2],y),w=Oe(f,[0,g*3],y),k=ce(L(Cn(x),Yi(A)),L(c,Cn(ce(o,b)))),S=L(Yi(k),Cn(w));return[k,S]}var ow=V({basicLSTMCell_:LP});function BP(e,t,n){let s=D(e,"x","batchToSpaceND"),r=t.reduce((i,l)=>i*l);M(s.rank>=1+t.length,()=>`input rank is ${s.rank} but should be > than blockShape.length ${t.length}`),M(n.length===t.length,()=>`crops.length is ${n.length} but should be equal to blockShape.length ${t.length}`),M(s.shape[0]%r===0,()=>`input tensor batch is ${s.shape[0]} but is not divisible by the product of the elements of blockShape ${t.join(" * ")} === ${r}`);let a={x:s},o={blockShape:t,crops:n};return W.runKernel(ll,a,o)}var uh=V({batchToSpaceND_:BP});function WP(e){let t;return e.rank===0||e.rank===1?t=U(e,[1,1,1,e.size]):e.rank===2?t=U(e,[1,1,e.shape[0],e.shape[1]]):e.rank===3?t=U(e,[1,e.shape[0],e.shape[1],e.shape[2]]):t=e,t}function VP(e,t,n,s,r,a){a==null&&(a=.001);let o=D(e,"x","batchNorm"),i=D(t,"mean","batchNorm"),l=D(n,"variance","batchNorm"),u;r!=null&&(u=D(r,"scale","batchNorm"));let c;s!=null&&(c=D(s,"offset","batchNorm")),M(i.rank===l.rank,()=>"Batch normalization gradient requires mean and variance to have equal ranks."),M(c==null||i.rank===c.rank,()=>"Batch normalization gradient requires mean and offset to have equal ranks."),M(u==null||i.rank===u.rank,()=>"Batch normalization gradient requires mean and scale to have equal ranks.");let d={x:WP(o),scale:u,offset:c,mean:i,variance:l},h={varianceEpsilon:a},f=W.runKernel(Eo,d,h);return U(f,o.shape)}var Wc=V({batchNorm_:VP});function UP(e,t,n,s,r,a){let o=D(e,"x","batchNorm"),i=D(t,"mean","batchNorm"),l=D(n,"variance","batchNorm"),u;r!=null&&(u=D(r,"scale","batchNorm"));let c;return s!=null&&(c=D(s,"offset","batchNorm")),M(o.rank===2,()=>`Error in batchNorm2D: x must be rank 2 but got rank ${o.rank}.`),M(i.rank===2||i.rank===1,()=>`Error in batchNorm2D: mean must be rank 2 or rank 1 but got rank ${i.rank}.`),M(l.rank===2||l.rank===1,()=>`Error in batchNorm2D: variance must be rank 2 or rank 1 but got rank ${l.rank}.`),u!=null&&M(u.rank===2||u.rank===1,()=>`Error in batchNorm2D: scale must be rank 2 or rank 1 but got rank ${u.rank}.`),c!=null&&M(c.rank===2||c.rank===1,()=>`Error in batchNorm2D: offset must be rank 2 or rank 1 but got rank ${c.rank}.`),Wc(o,i,l,c,u,a)}var nA=V({batchNorm2d_:UP});function GP(e,t,n,s,r,a){let o=D(e,"x","batchNorm"),i=D(t,"mean","batchNorm"),l=D(n,"variance","batchNorm"),u;r!=null&&(u=D(r,"scale","batchNorm"));let c;return s!=null&&(c=D(s,"offset","batchNorm")),M(o.rank===3,()=>`Error in batchNorm3D: x must be rank 3 but got rank ${o.rank}.`),M(i.rank===3||i.rank===1,()=>`Error in batchNorm3D: mean must be rank 3 or rank 1 but got rank ${i.rank}.`),M(l.rank===3||l.rank===1,()=>`Error in batchNorm3D: variance must be rank 3 or rank 1 but got rank ${l.rank}.`),u!=null&&M(u.rank===3||u.rank===1,()=>`Error in batchNorm3D: scale must be rank 3 or rank 1 but got rank ${u.rank}.`),c!=null&&M(c.rank===3||c.rank===1,()=>`Error in batchNorm3D: offset must be rank 3 or rank 1 but got rank ${c.rank}.`),Wc(o,i,l,c,u,a)}var sA=V({batchNorm3d_:GP});function HP(e,t,n,s,r,a){let o=D(e,"x","batchNorm"),i=D(t,"mean","batchNorm"),l=D(n,"variance","batchNorm"),u;r!=null&&(u=D(r,"scale","batchNorm"));let c;return s!=null&&(c=D(s,"offset","batchNorm")),M(o.rank===4,()=>`Error in batchNorm4D: x must be rank 4 but got rank ${o.rank}.`),M(i.rank===4||i.rank===1,()=>`Error in batchNorm4D: mean must be rank 4 or rank 1 but got rank ${i.rank}.`),M(l.rank===4||l.rank===1,()=>`Error in batchNorm4D: variance must be rank 4 or rank 1 but got rank ${l.rank}.`),u!=null&&M(u.rank===4||u.rank===1,()=>`Error in batchNorm4D: scale must be rank 4 or rank 1 but got rank ${u.rank}.`),c!=null&&M(c.rank===4||c.rank===1,()=>`Error in batchNorm4D: offset must be rank 4 or rank 1 but got rank ${c.rank}.`),Wc(o,i,l,c,u,a)}var rA=V({batchNorm4d_:HP});function jP(e,t,n){let s=D(e,"x","bincount"),r=D(t,"weights","bincount");M(s.dtype==="int32",()=>`Error in bincount: input dtype must be int32, but got ${s.dtype}`),M(n>=0,()=>`size must be non-negative, but got ${n}.`),M(r.size===s.size||r.size===0,()=>`Error in bincount: weights must have the same size as input or0-length, but got input shape: ${s.shape}, weights shape: ${r.shape}.`);let a={x:s,weights:r},o={size:n};return W.runKernel(jm,a,o)}var aA=V({bincount_:jP});function qP(e,t){let n=D(e,"s0","broadcastArgs","int32"),s=D(t,"s1","broadcastArgs","int32");if(n.rank!==1)throw new Error(`broadcastArgs(): first input must be a vector (rank=1). Has rank ${n.rank}`);if(s.rank!==1)throw new Error(`broadcastArgs(): second input must be a vector (rank=1). Has rank ${s.rank}`);let r={s0:n,s1:s};return W.runKernel(qm,r)}var iw=V({broadcastArgs_:qP});function XP(e,t){let n=D(e,"broadcastTo","x"),s=n.shape;if(t.some(u=>!(u>0)||u%1!==0))throw new Error(`broadcastTo(): Invalid broadcast shape [${t}].`);if(t.length<n.rank)throw new Error(`broadcastTo(): shape.length=${t.length} < input.rank=${n.rank}.`);if(t.length>n.rank){let u=n.shape.slice();for(;u.length<t.length;)u.unshift(1);n=U(n,u)}let r=n.shape,a=Array.from(t);for(let u=t.length-1;u>=0;u--)if(r[u]===t[u])a[u]=1;else if(n.shape[u]!==1)throw new Error(`broadcastTo(): [${s}] cannot be broadcast to [${t}].`);if(a.map((u,c)=>u>1?c:-1).filter(u=>u>=0).length===0)return On(n);let i={x:n},l={reps:a};return W.runKernel(va,i,l)}var Gu=V({broadcastTo_:XP});function KP(e){let n={x:D(e,"x","ceil","float32")};return W.runKernel(yo,n)}var oA=V({ceil_:KP});function ZP(e,t,n){let s=D(e,"x","clipByValue");M(t<=n,()=>`Error in clip: min (${t}) must be less than or equal to max (${n}).`);let r={x:s},a={clipValueMin:t,clipValueMax:n};return W.runKernel(ba,r,a)}var fs=V({clipByValue_:ZP});function YP(e){return Ct(e,0)}var iA=V({concat1d_:YP});function JP(e,t){return Ct(e,t)}var Kl=V({concat2d_:JP});function QP(e,t){return Ct(e,t)}var lA=V({concat3d_:QP});function eF(e,t){return Ct(e,t)}var uA=V({concat4d_:eF});function tF(e,t,n,s,r="NHWC",a=[1,1],o){let i=D(e,"x","conv2d","float32"),l=D(t,"filter","conv2d","float32"),u=i,c=!1;i.rank===3&&(c=!0,u=U(i,[1,i.shape[0],i.shape[1],i.shape[2]])),M(u.rank===4,()=>`Error in conv2d: input must be rank 4, but got rank ${u.rank}.`),M(l.rank===4,()=>`Error in conv2d: filter must be rank 4, but got rank ${l.rank}.`),rs("conv2d",s,o);let p=r==="NHWC"?u.shape[3]:u.shape[1];M(p===l.shape[2],()=>`Error in conv2d: depth of input (${p}) must match input depth for filter ${l.shape[2]}.`),M(Yr(n,a),()=>`Error in conv2D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`);let d={x:u,filter:l},h={strides:n,pad:s,dataFormat:r,dilations:a,dimRoundingMode:o},f=W.runKernel(Ao,d,h);return c?U(f,[f.shape[1],f.shape[2],f.shape[3]]):f}var ma=V({conv2d_:tF});function nF(e,t,n,s,r="NWC",a=1,o){let i=D(e,"x","conv1d"),l=D(t,"filter","conv1d"),u=i,c=!1;i.rank===2&&(c=!0,u=U(i,[1,i.shape[0],i.shape[1]])),M(u.rank===3,()=>`Error in conv1d: input must be rank 3, but got rank ${u.rank}.`),M(l.rank===3,()=>`Error in conv1d: filter must be rank 3, but got rank ${l.rank}.`),rs("conv1d",s,o),M(u.shape[2]===l.shape[1],()=>`Error in conv1d: depth of input (${u.shape[2]}) must match input depth for filter ${l.shape[1]}.`),M(Yr(n,a),()=>`Error in conv1D: Either stride or dilation must be 1. Got stride ${n} and dilation '${a}'`),M(r==="NWC",()=>`Error in conv1d: got dataFormat of ${r} but only NWC is currently supported.`);let p=U(l,[1,l.shape[0],l.shape[1],l.shape[2]]),d=U(u,[u.shape[0],1,u.shape[1],u.shape[2]]),g=ma(d,p,[1,n],s,"NHWC",[1,a],o);return c?U(g,[g.shape[2],g.shape[3]]):U(g,[g.shape[0],g.shape[2],g.shape[3]])}var y0=V({conv1d_:nF});function sF(e,t,n,s,r,a="NHWC",o){M(e.length===t.rank,()=>`Length of inShape (${e.length}) and rank of dy (${t.rank}) must match`);let i=e,l=t,u=!1;t.rank===3&&(u=!0,l=U(t,[1,t.shape[0],t.shape[1],t.shape[2]]),i=[1,e[0],e[1],e[2]]),M(i.length===4,()=>`Error in conv2dDerInput: inShape must be length 4, but got length ${i.length}.`),M(l.rank===4,()=>`Error in conv2dDerInput: dy must be rank 4, but got rank ${l.rank}`),M(n.rank===4,()=>`Error in conv2dDerInput: filter must be rank 4, but got rank ${n.rank}`);let c=a==="NHWC"?i[3]:i[1],p=a==="NHWC"?l.shape[3]:l.shape[1];M(c===n.shape[2],()=>`Error in conv2dDerInput: depth of input (${c}) must match input depth for filter ${n.shape[2]}.`),M(p===n.shape[3],()=>`Error in conv2dDerInput: depth of output (${p}) must match output depth for filter ${n.shape[3]}.`),rs("conv2dDerInput",r,o);let d={dy:l,filter:n},h={strides:s,pad:r,dataFormat:a,dimRoundingMode:o,inputShape:i},f=W.runKernel(xo,d,h);return u?U(f,[f.shape[1],f.shape[2],f.shape[3]]):f}var cA=V({conv2DBackpropInput_:sF});function rF(e,t,n,s,r,a){let o=D(e,"x","conv2dTranspose"),i=D(t,"filter","conv2dTranspose");return cA(n,o,i,s,r,"NHWC",a)}var A0=V({conv2dTranspose_:rF});function aF(e,t,n,s,r="NDHWC",a=[1,1,1]){let o=D(e,"x","conv3d"),i=D(t,"filter","conv3d"),l=o,u=!1;o.rank===4&&(u=!0,l=U(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),M(l.rank===5,()=>`Error in conv3d: input must be rank 5, but got rank ${l.rank}.`),M(i.rank===5,()=>`Error in conv3d: filter must be rank 5, but got rank ${i.rank}.`),M(l.shape[4]===i.shape[3],()=>`Error in conv3d: depth of input (${l.shape[4]}) must match input depth for filter ${i.shape[3]}.`),M(Yr(n,a),()=>`Error in conv3D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`),M(r==="NDHWC",()=>`Error in conv3d: got dataFormat of ${r} but only NDHWC is currently supported.`);let c={x:l,filter:i},p={strides:n,pad:s,dataFormat:r,dilations:a},d=W.runKernel(Wp,c,p);return u?U(d,[d.shape[1],d.shape[2],d.shape[3],d.shape[4]]):d}var dA=V({conv3d_:aF});function oF(e,t,n,s,r){M(e.length===t.rank,()=>`Length of inShape (${e.length}) and rank of dy (${t.rank}) must match`);let a=e,o=t,i=!1;t.rank===4&&(i=!0,o=U(t,[1,t.shape[0],t.shape[1],t.shape[2],t.shape[3]]),a=[1,e[0],e[1],e[2],e[3]]);let l=a[4],u=o.shape[4];M(a.length===5,()=>`Error in conv3dDerInput: inShape must be length 5, but got length ${a.length}.`),M(o.rank===5,()=>`Error in conv3dDerInput: dy must be rank 5, but got rank ${o.rank}`),M(n.rank===5,()=>`Error in conv3dDerInput: filter must be rank 5, but got rank ${n.rank}`),M(l===n.shape[3],()=>`Error in conv3dDerInput: depth of input (${l}) must match input depth for filter ${n.shape[3]}.`),M(u===n.shape[4],()=>`Error in conv3dDerInput: depth of output (${u}) must match output depth for filter ${n.shape[4]}.`);let c={dy:o,filter:n},p={pad:r,strides:s,inputShape:a},d=W.runKernel(Zm,c,p);return i?U(d,[d.shape[1],d.shape[2],d.shape[3],d.shape[4]]):d}var lw=V({conv3DBackpropInput_:oF});function iF(e,t,n,s,r){let a=D(e,"x","conv3dTranspose"),o=D(t,"filter","conv3dTranspose");return lw(n,a,o,s,r)}var pA=V({conv3dTranspose_:iF});function lF(e){let n={x:D(e,"x","cos","float32")};return W.runKernel(bo,n)}var ch=V({cos_:lF});function uF(e){let n={x:D(e,"x","cosh","float32")};return W.runKernel(vo,n)}var x0=V({cosh_:uF});function cF(e,t=0,n=!1,s=!1){let a={x:D(e,"x","cumprod")},o={axis:t,exclusive:n,reverse:s};return W.runKernel(cl,a,o)}var Tp=V({cumprod_:cF});function dF(e,t=0,n=!1,s=!1){let a={x:D(e,"x","cumsum")},o={axis:t,exclusive:n,reverse:s};return W.runKernel(wo,a,o)}var b0=V({cumsum_:dF});function pF(e,t,n,s=!1){let r=D(e,"x","denseBincount"),a=D(t,"weights","denseBincount");M(r.dtype==="int32",()=>`Error in denseBincount: input dtype must be int32, but got ${r.dtype}`),M(r.rank<=2,()=>`Error in denseBincount: input must be at most rank 2, but got rank ${r.rank}.`),M(n>=0,()=>`size must be non-negative, but got ${n}.`),M(a.size===r.size||a.size===0,()=>`Error in denseBincount: weights must have the same shape as x or 0-length, but got x shape: ${r.shape}, weights shape: ${a.shape}.`);let o={x:r,weights:a},i={size:n,binaryOutput:s};return W.runKernel(Ym,o,i)}var uw=V({denseBincount_:pF});function hF(e,t,n="NHWC"){let s=D(e,"x","depthToSpace","float32"),r=n==="NHWC"?s.shape[1]:s.shape[2],a=n==="NHWC"?s.shape[2]:s.shape[3],o=n==="NHWC"?s.shape[3]:s.shape[1];M(t>1,()=>`blockSize should be > 1 for depthToSpace, but was: ${t}`),M(r*t>=0,()=>`Negative dimension size caused by overflow when multiplying
|
|
${r} and ${t} for depthToSpace with input shape
|
|
${s.shape}`),M(a*t>=0,()=>`Negative dimension size caused by overflow when multiplying
|
|
${a} and ${t} for depthToSpace with input shape
|
|
${s.shape}`),M(o%(t*t)===0,()=>`Dimension size must be evenly divisible by ${t*t} but is ${o} for depthToSpace with input shape ${s.shape}`);let i={x:s},l={blockSize:t,dataFormat:n};return W.runKernel(pl,i,l)}var hA=V({depthToSpace_:hF});function fF(e,t,n,s,r="NHWC",a=[1,1],o){let i=D(e,"x","depthwiseConv2d","float32"),l=D(t,"filter","depthwiseConv2d","float32"),u=i,c=!1;i.rank===3&&(c=!0,u=U(i,[1,i.shape[0],i.shape[1],i.shape[2]])),M(u.rank===4,()=>`Error in depthwiseConv2d: input must be rank 4, but got rank ${u.rank}.`),M(l.rank===4,()=>`Error in depthwiseConv2d: filter must be rank 4, but got rank ${l.rank}.`);let p=r==="NHWC"?u.shape[3]:u.shape[1];M(p===l.shape[2],()=>`Error in depthwiseConv2d: number of input channels (${p}) must match the inChannels dimension in filter ${l.shape[2]}.`),rs("depthwiseConv2d",s,o);let d={x:u,filter:l},h={strides:n,pad:s,dataFormat:r,dilations:a,dimRoundingMode:o},f=W.runKernel(ko,d,h);return c?U(f,[f.shape[1],f.shape[2],f.shape[3]]):f}var Vc=V({depthwiseConv2d_:fF});function mF(e){let n={x:D(e,"x","diag")};return W.runKernel(e0,n)}var cw=V({diag_:mF});function gF(e,t,n,s,r=[1,1],a="NHWC"){let o=D(e,"x","dilation2d"),i=D(t,"filter","dilation2d");M(o.rank===3||o.rank===4,()=>`Error in dilation2d: input must be rank 3 or 4, but got rank ${o.rank}.`),M(i.rank===3,()=>`Error in dilation2d: filter must be rank 3, but got rank ${i.rank}.`),M(a==="NHWC",()=>`Error in dilation2d: Only NHWC is currently supported, but got dataFormat of ${a}`);let l=o,u=!1;o.rank===3&&(l=U(o,[1,o.shape[0],o.shape[1],o.shape[2]]),u=!0);let c={x:l,filter:i},p={strides:n,pad:s,dilations:r},d=W.runKernel(Vp,c,p);return u?U(d,[d.shape[1],d.shape[2],d.shape[3]]):d}var fA=V({dilation2d_:gF});function yF(e,t){let n=D(e,"a","equal","string_or_numeric"),s=D(t,"b","equal","string_or_numeric");[n,s]=Gt(n,s),kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(hl,r)}var Rs=V({equal_:yF});function AF(e,t,n){let s=D(t,"a","where"),r=D(n,"b","where"),a=D(e,"condition","where","bool"),o=kt(kt(a.shape,s.shape),r.shape),i=Gu(a,o),l=Gu(s,o),u=Gu(r,o),c={condition:i,t:l,e:u};return W.runKernel(Fl,c)}var zn=V({where_:AF});function xF(e){let n={x:D(e,"x","zerosLike")};return W.runKernel(Hl,n)}var it=V({zerosLike_:xF});function bF(e,t){let n=D(e,"a","div"),s=D(t,"b","div");[n,s]=Gt(n,s);let r=he(n,s),a=it(r),o=Rs(s,a);return zn(o,a,r)}var mA=V({divNoNan_:bF});function vF(e,t){let n=D(e,"t1","dot"),s=D(t,"t2","dot");M((n.rank===1||n.rank===2)&&(s.rank===1||s.rank===2),()=>`Error in dot: inputs must all be rank 1 or 2, but got ranks ${n.rank} and ${s.rank}.`);let r=n.rank===1?n.size:n.shape[1],a=s.rank===1?s.size:s.shape[0];if(M(r===a,()=>`Error in dot: inner dimensions of inputs must match, but got ${r} and ${a}.`),n.rank===1&&s.rank===1){let o=U(n,[1,-1]),i=U(s,[-1,1]),l=Qe(o,i);return U(l,[])}else if(n.rank===1&&s.rank===2){let o=U(n,[1,-1]),i=U(s,[s.shape[0],s.shape[1]]),l=Qe(o,i);return U(l,[l.size])}else if(n.rank===2&&s.rank===1){let o=U(s,[-1,1]),i=Qe(n,o);return U(i,[i.size])}else{let o=U(s,[s.shape[0],s.shape[1]]);return Qe(n,o)}}var gA=V({dot_:vF});function wF(e,...t){let n=t.map((r,a)=>D(r,`tensors${a}`,"einsum")),s={equation:e};return W.runKernel(Up,n,s)}var dw=V({einsum_:wF});function kF(e){let n={x:D(e,"x","elu","float32")};return W.runKernel(So,n)}var Uc=V({elu_:kF});function IF(e){let t=D(e,"x","erf");M(t.dtype==="int32"||t.dtype==="float32",()=>"Input dtype must be `int32` or `float32`."),t.dtype==="int32"&&(t=ye(t,"float32"));let n={x:t};return W.runKernel(wc,n)}var yA=V({erf_:IF});function AA(e,t){for(let n=0;n<e.length;++n)if(e[e.length-n-1]!==t-1-n)return!1;return!0}function pw(e,t,n){let s=e.length+t.length,r=[],a=0,o=0;for(let i=0;i<s;i++)n.indexOf(i)===-1?r.push(e[a++]):r.push(t[o++]);return r}function hw(e,t){let n=[],s=e.length;for(let a=0;a<s;a++)t.indexOf(a)===-1&&n.push(e[a]);let r=t.map(a=>e[a]);return[n,r]}function Ji(e,t){let n=t.map(s=>1);return pw(e,n,t)}function SF(e,t,n){M(AA(t,n),()=>`${e} supports only inner-most axes for now. Got axes ${t} and rank-${n} input.`)}function fw(e,t){if(AA(e,t))return null;let n=[];for(let s=0;s<t;++s)e.indexOf(s)===-1&&n.push(s);return e.forEach(s=>n.push(s)),n}function xA(e){return e.map((t,n)=>[n,t]).sort((t,n)=>t[1]-n[1]).map(t=>t[0])}function CF(e,t){let n=[];for(let s=t-e;s<t;++s)n.push(s);return n}function TF(e,t=null,n=!1){let r={x:D(e,"x","max")},a={reductionIndices:t,keepDims:n};return W.runKernel(Po,r,a)}var mn=V({max_:TF});function NF(e,t=null,n=!1){let r={x:D(e,"x","min")},a={axis:t,keepDims:n};return W.runKernel(zo,r,a)}var ga=V({min_:NF});function EF(e,t){let n=D(e,"base","pow"),s=D(t,"exp","pow");[n,s]=Gt(n,s);let r={a:n,b:s};return W.runKernel(Uo,r)}var ya=V({pow_:EF});function Ce(e,t){if((Fn(e)&&t!=="string"||Array.isArray(e))&&t!=="complex64")throw new Error("Error creating a new Scalar: value must be a primitive (number|boolean|string)");if(t==="string"&&Fn(e)&&!(e instanceof Uint8Array))throw new Error("When making a scalar from encoded string, the value must be `Uint8Array`.");return oi(e,[],[],t)}function RF(e){let n={x:D(e,"x","sqrt","float32")};return W.runKernel(Qo,n)}var Nn=V({sqrt_:RF});function _F(e){let t=D(e,"x","square"),n={};return W.runKernel("Square",{x:t},n)}var vt=V({square_:_F});function DF(e,t=null,n=!1){let s=D(e,"x","sum");s.dtype==="bool"&&(s=ye(s,"int32"));let r={x:s},a={axis:t,keepDims:n};return W.runKernel(ei,r,a)}var ke=V({sum_:DF});function $F(e,t="euclidean",n=null,s=!1){e=D(e,"x","norm");let r=mw(e,t,n),a=r.shape;if(s){let o=cr(n,e.shape);a=Ji(r.shape,o)}return U(r,a)}function mw(e,t,n=null){if(e.rank===0)return tn(e);if(e.rank!==1&&n===null)return mw(U(e,[-1]),t,n);if(e.rank===1||typeof n=="number"||Array.isArray(n)&&n.length===1){if(t===1)return ke(tn(e),n);if(t===1/0)return mn(tn(e),n);if(t===-1/0)return ga(tn(e),n);if(t==="euclidean"||t===2)return Nn(ke(ya(tn(e),Ce(2,"int32")),n));throw new Error(`Error in norm: invalid ord value: ${t}`)}if(Array.isArray(n)&&n.length===2){if(t===1)return mn(ke(tn(e),n[0]),n[1]-1);if(t===1/0)return mn(ke(tn(e),n[1]),n[0]);if(t===-1/0)return ga(ke(tn(e),n[1]),n[0]);if(t==="fro"||t==="euclidean")return Nn(ke(vt(e),n));throw new Error(`Error in norm: invalid ord value: ${t}`)}throw new Error(`Error in norm: invalid axis: ${n}`)}var Gc=V({norm_:$F});function PF(e,t=null,n=!1){return Gc(e,"euclidean",t,n)}var bA=V({euclideanNorm_:PF});function FF(e){let n={x:D(e,"x","exp")};return W.runKernel(Co,n)}var _s=V({exp_:FF});function OF(e,t=0){let n=D(e,"x","expandDims","string_or_numeric");M(t<=n.rank,()=>"Axis must be <= rank of the tensor");let s={input:n},r={dim:t};return W.runKernel(fl,s,r)}var Kt=V({expandDims_:OF});function MF(e){let n={x:D(e,"x","expm1")};return W.runKernel(ml,n)}var vA=V({expm1_:MF});function zF(e,t){let n=D(e,"x","tile","string_or_numeric");M(n.rank===t.length,()=>`Error in transpose: rank of input ${n.rank} must match length of reps ${t}.`);let s={x:n},r={reps:t};return W.runKernel(va,s,r)}var qs=V({tile_:zF});function LF(e,t,n,s="float32"){t==null&&(t=e);let r=Le([e,t],s),a=e<=t?e:t;for(let i=0;i<a;++i)r.set(1,i,i);let o=U(r.toTensor(),[e,t]);if(n==null)return o;if(n.length===1)return qs(Kt(o,0),[n[0],1,1]);if(n.length===2)return qs(Kt(Kt(o,0),0),[n[0],n[1],1,1]);if(n.length===3)return qs(Kt(Kt(Kt(o,0),0),0),[n[0],n[1],n[2],1,1]);throw new Error(`eye() currently supports only 1D and 2D batchShapes, but received ${n.length}D.`)}var v0=V({eye_:LF});function Hc(e,t,n){let s={shape:e,value:t,dtype:n};return W.runKernel(kc,{},s)}function BF(e){let n={x:D(e,"x","floor","float32")};return W.runKernel(To,n)}var jc=V({floor_:BF});function WF(e,t,n=0,s=0){let r=D(e,"x","gather"),a=D(t,"indices","gather","int32"),o={x:r,indices:a},i={axis:n,batchDims:s};return W.runKernel(yl,o,i)}var qc=V({gather_:WF});function VF(e,t){let n=D(e,"a","greater","string_or_numeric"),s=D(t,"b","greater","string_or_numeric");[n,s]=Gt(n,s),kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(xl,r)}var ys=V({greater_:VF});function UF(e,t){let n=D(e,"a","greaterEqual","string_or_numeric"),s=D(t,"b","greaterEqual","string_or_numeric");[n,s]=Gt(n,s),kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(Ro,r)}var li=V({greaterEqual_:UF});function GF(e){let n={x:D(e,"x","isFinite")};return W.runKernel(Ic,n)}var wA=V({isFinite_:GF});function HF(e){let n={x:D(e,"x","isInf")};return W.runKernel(Sc,n)}var kA=V({isInf_:HF});function jF(e){let n={x:D(e,"x","isNaN")};return W.runKernel(Cc,n)}var IA=V({isNaN_:jF});function qF(e,t=.2){let s={x:D(e,"x","leakyRelu")},r={alpha:t};return W.runKernel(Do,s,r)}var dh=V({leakyRelu_:qF});function XF(e,t){let n=D(e,"a","less","string_or_numeric"),s=D(t,"b","less","string_or_numeric");[n,s]=Gt(n,s),kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(bl,r)}var w0=V({less_:XF});function KF(e,t){let n=D(e,"a","lessEqual","string_or_numeric"),s=D(t,"b","lessEqual","string_or_numeric");[n,s]=Gt(n,s),kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(vl,r)}var ui=V({lessEqual_:KF});function gw(e,t,n){if(n<=0)throw new Error("The number of values should be positive.");let s={start:e,stop:t,num:n};return W.runKernel(r0,{},s)}function ZF(e,t=5,n=1,s=1,r=.5){let a=D(e,"x","localResponseNormalization");M(a.rank===4||a.rank===3,()=>`Error in localResponseNormalization: x must be rank 3 or 4 but got
|
|
rank ${a.rank}.`),M(qu(t),()=>`Error in localResponseNormalization: depthRadius must be an integer but got depthRadius ${t}.`);let o=a,i=!1;a.rank===3&&(i=!0,o=U(a,[1,a.shape[0],a.shape[1],a.shape[2]]));let l={x:o},u={depthRadius:t,bias:n,alpha:s,beta:r},c=W.runKernel(Hp,l,u);return i?U(c,[c.shape[1],c.shape[2],c.shape[3]]):c}var SA=V({localResponseNormalization_:ZF});function YF(e){let n={x:D(e,"x","log","float32")};return W.runKernel($o,n)}var Ds=V({log_:YF});function JF(e){let n={x:D(e,"x","log1p")};return W.runKernel(Tc,n)}var ph=V({log1p_:JF});function QF(e){return M(Ya(e),()=>"The f passed in grad(f) must be a function"),(t,n)=>{let s=D(t,"x","tf.grad","string_or_numeric"),r=n!=null?D(n,"dy","tf.grad"):null;return W.tidy(()=>{let{value:a,grads:o}=W.gradients(()=>e(s),[s],r);return r!=null&&ns(a.shape,r.shape,"The shape of dy passed in grad(f)(x, dy) must match the shape returned by f(x)"),k0(o),o[0]})}}function eO(e){return M(Ya(e),()=>"The f passed in grads(f) must be a function"),(t,n)=>{M(Array.isArray(t),()=>"The args passed in grads(f)(args) must be an array of `Tensor`s or `TensorLike`s");let s=Sp(t,"args","tf.grads","string_or_numeric"),r=n!=null?D(n,"dy","tf.grads"):null;return W.tidy(()=>{let{value:a,grads:o}=W.gradients(()=>e(...s),s,r);return r!=null&&ns(a.shape,r.shape,"The shape of dy passed in grads(f)([x1,...], dy) must match the shape returned by f([x1,...])"),k0(o),o})}}function tO(e){return M(Ya(e),()=>"The f passed in valueAndGrad(f) must be a function"),(t,n)=>{M(t instanceof nt,()=>"The x passed in valueAndGrad(f)(x) must be a tensor"),M(n==null||n instanceof nt,()=>"The dy passed in valueAndGrad(f)(x, dy) must be a tensor");let{grads:s,value:r}=W.gradients(()=>e(t),[t],n);return k0(s),{grad:s[0],value:r}}}function nO(e){return M(Ya(e),()=>"The f passed in valueAndGrads(f) must be a function"),(t,n)=>{M(Array.isArray(t)&&t.every(r=>r instanceof nt),()=>"The args passed in valueAndGrads(f)(args) must be array of tensors"),M(n==null||n instanceof nt,()=>"The dy passed in valueAndGrads(f)(args, dy) must be a tensor");let s=W.gradients(()=>e(...t),t,n);return n!=null&&ns(s.value.shape,n.shape,"The shape of dy passed in valueAndGrads(f)([x1,...], dy) must match the shape returned by f([x1,...])"),k0(s.grads),s}}function yw(e,t){M(Ya(e),()=>"The f passed in variableGrads(f) must be a function"),M(t==null||Array.isArray(t)&&t.every(u=>u instanceof kp),()=>"The varList passed in variableGrads(f, varList) must be an array of variables");let n=t!=null;if(!n){t=[];for(let u in W.registeredVariables)t.push(W.registeredVariables[u])}let s=n?t.filter(u=>!u.trainable):null,r=t.length;t=t.filter(u=>u.trainable),M(t.length>0,()=>`variableGrads() expects at least one of the input variables to be trainable, but none of the ${r} variables is trainable.`);let a=!0,{value:o,grads:i}=W.gradients(e,t,null,a);M(i.some(u=>u!=null),()=>"Cannot find a connection between any variable and the result of the loss function y=f(x). Please make sure the operations that use variables are inside the function f passed to minimize()."),M(o.rank===0,()=>`The f passed in variableGrads(f) must return a scalar, but it returned a rank-${o.rank} tensor`);let l={};return t.forEach((u,c)=>{i[c]!=null&&(l[u.name]=i[c])}),s!=null&&s.forEach(u=>l[u.name]=null),{value:o,grads:l}}function Kr(e){return W.customGrad(e)}function k0(e){if(e.filter(n=>n==null).length>0)throw new Error(`Cannot compute gradient of y=f(x) with respect to x. Make sure that
|
|
the f you passed encloses all operations that lead from x to y.`)}function sO(e){let n={x:D(e,"x","softplus")};return W.runKernel(Fc,n)}var Zl=V({softplus_:sO});function rO(e){let t=D(e,"x","logSigmoid");return Kr(s=>({value:Dt(Zl(Dt(s))),gradFunc:o=>L(o,Cn(Dt(s)))}))(t)}var CA=V({logSigmoid_:rO});function aO(e,t){let n=D(e,"a","sub"),s=D(t,"b","sub");[n,s]=Gt(n,s);let r={a:n,b:s};return W.runKernel(si,r)}var fe=V({sub_:aO});function oO(e,t=-1){let n=D(e,"logits","logSoftmax");if(t===-1&&(t=n.rank-1),t!==n.rank-1)throw Error(`Log Softmax along a non-last dimension is not yet supported. Logits was rank ${n.rank} and axis was ${t}`);return Kr((r,a)=>{let i=mn(r,t,!0),l=fe(r,i),u=fe(ye(l,"float32"),Ds(ke(_s(l),t,!0)));return a([u]),{value:u,gradFunc:(p,d)=>{let[h]=d,f=!0,m=_s(h);return fe(p,L(ke(p,t,f),m))}}})(n)}var I0=V({logSoftmax_:oO});function iO(e,t=null,n=!1){let s=D(e,"x","logSumExp"),r=cr(t,s.shape),a=mn(s,r,!0),o=fe(s,a),i=_s(o),l=ke(i,r),u=Ds(l),c=ce(U(a,u.shape),u);if(n){let p=Ji(c.shape,r);return U(c,p)}return c}var S0=V({logSumExp_:iO});function lO(e,t){let n=D(e,"a","logicalAnd","bool"),s=D(t,"b","logicalAnd","bool");kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(wl,r)}var ur=V({logicalAnd_:lO});function uO(e){let n={x:D(e,"x","logicalNot","bool")};return W.runKernel(kl,n)}var hh=V({logicalNot_:uO});function cO(e,t){let n=D(e,"a","logicalOr","bool"),s=D(t,"b","logicalOr","bool");kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(Nc,r)}var C0=V({logicalOr_:cO});function dO(e,t){let n=D(e,"a","logicalXor","bool"),s=D(t,"b","logicalXor","bool");return kt(n.shape,s.shape),ur(C0(e,t),hh(ur(e,t)))}var TA=V({logicalXor_:dO}),Bf=2147483648;function pO(e,t,n="left"){let s=D(e,"sortedSequence","searchSorted"),r=D(t,"values","searchSorted"),a=s.shape[s.shape.length-1],o=r.shape[r.shape.length-1],i=U(s,[-1,a]),l=U(r,[-1,o]);if(i.rank<2)throw new Error("Sorted input argument must be at least 2-dimensional");if(i.shape[0]!==l.shape[0])throw new Error("Leading dimension of 'sortedSequence' and 'values' must match.");if(Et(l.shape)>=Bf)throw new Error(`values tensor size must less than ${Bf}`);if(i.shape[1]>=Bf)throw new Error(`trailing dim_size must less than ${Bf} for int32 output type, was ${i.shape[1]}`);let u={sortedSequence:i,values:l},c={side:n};return W.runKernel(p0,u,c)}var T0=V({searchSorted_:pO});function Aw(e,t){return T0(e,t,"left")}function hO(e,t,n,s,r){let a=D(e,"x","maxPool"),o=1,i=a,l=!1;a.rank===3&&(l=!0,i=U(a,[1,a.shape[0],a.shape[1],a.shape[2]])),M(i.rank===4,()=>`Error in maxPool: input must be rank 4 but got rank ${i.rank}.`),M(Yr(n,o),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${n} and dilations '${o}'`),rs("maxPool",s,r);let u={x:i},c={filterSize:t,strides:n,pad:s,dimRoundingMode:r},p=W.runKernel(Oo,u,c);return l?U(p,[p.shape[1],p.shape[2],p.shape[3]]):p}var fh=V({maxPool_:hO});function fO(e,t=[1,1,1],n,s,r,a="NDHWC"){let o=D(e,"x","maxPool3d"),i=o,l=!1;o.rank===4&&(l=!0,i=U(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),M(i.rank===5,()=>`Error in maxPool3d: x must be rank 5 but got rank ${i.rank}.`),M(a==="NDHWC",()=>`Error in maxPool3d: Only NDHWC is currently supported, but got dataFormat of ${a}`),rs("maxPool3d",s,r);let u={x:i},c={filterSize:t,strides:n,pad:s,dimRoundingMode:r,dataFormat:a},p=W.runKernel(jp,u,c);return l?U(p,[p.shape[1],p.shape[2],p.shape[3],p.shape[4]]):p}var NA=V({maxPool3d_:fO});function mO(e,t,n,s,r=!1){let o={x:D(e,"x","maxPoolWithArgmax")},i={filterSize:t,strides:n,pad:s,includeBatchInIndex:r},l=W.runKernel(l0,o,i);return{result:l[0],indexes:l[1]}}var xw=V({maxPoolWithArgmax_:mO});function gO(e,t){let n=D(e,"a","maximum"),s=D(t,"b","maximum");[n,s]=Gt(n,s),n.dtype==="bool"&&(n=ye(n,"int32"),s=ye(s,"int32")),kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(Fo,r)}var Jr=V({maximum_:gO});function yO(e,t=null,n=!1){let r={x:D(e,"x","mean")},a={axis:t,keepDims:n};return W.runKernel(Mo,r,a)}var Bt=V({mean_:yO});function Wt(e,t="float32"){if(t==="complex64"){let s=Wt(e,"float32"),r=Wt(e,"float32");return fa(s,r)}let n=Um(Et(e),t);return W.makeTensor(n,e,t)}function Ns(e,t="float32"){if(t==="complex64"){let s=Ns(e,"float32"),r=Wt(e,"float32");return fa(s,r)}let n=by(Et(e),t);return W.makeTensor(n,e,t)}function bw(e,t,{indexing:n="xy"}={}){if(n!=="xy"&&n!=="ij")throw new TypeError(`${n} is not a valid third argument to meshgrid`);if(e===void 0)return[];let s=D(e,"x","meshgrid",e instanceof nt?e.dtype:"float32");if(t===void 0)return[s];let r=D(t,"y","meshgrid",t instanceof nt?t.dtype:"float32"),a=Et(s.shape),o=Et(r.shape);return n==="xy"?(s=U(s,[1,-1]),r=U(r,[-1,1]),[Qe(Ns([o,1],s.dtype),s),Qe(r,Ns([1,a],r.dtype))]):(s=U(s,[-1,1]),r=U(r,[1,-1]),[Qe(s,Ns([1,o],s.dtype)),Qe(Ns([a,1],r.dtype),r)])}function AO(e,t){let n=D(e,"a","minimum"),s=D(t,"b","minimum");[n,s]=Gt(n,s),n.dtype==="bool"&&(n=ye(n,"int32"),s=ye(s,"int32")),kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(Lo,r)}var Xc=V({minimum_:AO});function xO(e,t,n){M(n==="reflect"||n==="symmetric",()=>`Invalid mode. Mode must be either reflect or symmetric. Got ${n}.`);let s=D(e,"x","mirrorPad");if(s.rank===0)throw new Error("mirrorPad(scalar) is not defined. Pass non-scalar to mirrorPad");M(t.length===s.rank,()=>`Padding doesn't match input. Must be ${s.rank}. Got ${t.length}.`);let r=n==="reflect"?1:0;for(let i=0;i<s.rank;i++)M(t[i].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),M(t[i][0]>=0&&t[i][0]<=s.shape[i]-r&&t[i][1]>=0&&t[i][1]<=s.shape[i]-r,()=>`Padding in dimension ${i} cannot be greater than or equal to ${s.shape[i]-r} or less than 0 for input of shape ${s.shape}`);let a={paddings:t,mode:n},o={x:s};return W.runKernel(Bo,o,a)}var EA=V({mirrorPad_:xO});function bO(e,t){let n=D(e,"a","mod"),s=D(t,"b","mod");[n,s]=Gt(n,s);let r={a:n,b:s};return W.runKernel(Ec,r)}var Yl=V({mod_:bO});function vO(e,t=null,n=!1){e=D(e,"x","moments");let s=cr(t,e.shape),r=Bt(e,s,n),a=r.shape;n||(a=Ji(r.shape,s));let o=vt(fe(ye(e,"float32"),U(r,a))),i=Bt(o,s,n);return{mean:r,variance:i}}var mh=V({moments_:vO});function wO(e,t,n,s){let r=D(t,"data","multiRNNCell"),a=Sp(n,"c","multiRNNCell"),o=Sp(s,"h","multiRNNCell"),i=r,l=[];for(let p=0;p<e.length;p++){let d=e[p](i,a[p],o[p]);l.push(d[0]),l.push(d[1]),i=d[1]}let u=[],c=[];for(let p=0;p<l.length;p+=2)u.push(l[p]),c.push(l[p+1]);return[u,c]}var vw=V({multiRNNCell_:wO});function kO(e,t,n,s=!1){let r=D(e,"logits","multinomial"),a=r.size,o=r.rank;if(a<2)throw new Error(`Error in multinomial: you need at least 2 outcomes, but got ${a}.`);if(o>2)throw new Error(`Rank of probabilities must be 1 or 2, but is ${o}`);n=n||Math.random();let l={logits:o===1?U(r,[1,-1]):r},u={numSamples:t,seed:n,normalized:s},c=W.runKernel(u0,l,u);return o===1?U(c,[c.size]):c}var ww=V({multinomial_:kO});function IO(e,t){let n=D(e,"a","notEqual","string_or_numeric"),s=D(t,"b","notEqual","string_or_numeric");[n,s]=Gt(n,s),kt(n.shape,s.shape);let r={a:n,b:s};return W.runKernel(Sl,r)}var Qi=V({notEqual_:IO});function SO(e){let n={x:D(e,"x","onesLike")};return W.runKernel(Nl,n)}var $s=V({onesLike_:SO});function CO(e,t){let n=D(e,"v1","outerProduct"),s=D(t,"v2","outerProduct");M(n.rank===1&&s.rank===1,()=>`Error in outerProduct: inputs must be rank 1, but got ranks ${n.rank} and ${s.rank}.`);let r=U(n,[-1,1]),a=U(s,[1,-1]);return Qe(r,a)}var kw=V({outerProduct_:CO});function TO(e,t,n=0){let s=D(e,"x","pad");if(s.rank===0)throw new Error("pad(scalar) is not defined. Pass non-scalar to pad");let r={paddings:t,constantValue:n},a={x:s};return W.runKernel(Vo,a,r)}var Qs=V({pad_:TO});function NO(e,t,n=0){return M(t.length===2,()=>"Invalid number of paddings. Must be length of 2."),Qs(e,[t],n)}var Iw=V({pad1d_:NO});function EO(e,t,n=0){return M(t.length===2&&t[0].length===2&&t[1].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),Qs(e,t,n)}var Sw=V({pad2d_:EO});function RO(e,t,n=0){return M(t.length===3&&t[0].length===2&&t[1].length===2&&t[2].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),Qs(e,t,n)}var Cw=V({pad3d_:RO});function _O(e,t,n=0){return M(t.length===4&&t[0].length===2&&t[1].length===2&&t[2].length===2&&t[3].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),Qs(e,t,n)}var Tw=V({pad4d_:_O});function DO(e,t,n){let s=D(e,"x","spaceToBatchND");M(s.rank>=1+t.length,()=>`input rank ${s.rank} should be > than [blockShape] ${t.length}`),M(n.length===t.length,()=>`paddings.shape[0] ${n.length} must be equal to [blockShape] ${t.length}`),M(s.shape.reduce((o,i,l)=>l>0&&l<=t.length?o&&(i+n[l-1][0]+n[l-1][1])%t[l-1]===0:o,!0),()=>`input spatial dimensions ${s.shape.slice(1)} with paddings ${n.toString()} must be divisible by blockShapes ${t.toString()}`);let r={x:s},a={blockShape:t,paddings:n};return W.runKernel(zl,r,a)}var gh=V({spaceToBatchND_:DO});function $O(e,t,n,s,r,a,o){r==null&&(r=[1,1]),a==null&&(a=1),s===0&&(s="valid");let i=D(e,"x","maxPool"),l=i,u=!1;i.rank===3&&(u=!0,l=U(i,[1,i.shape[0],i.shape[1],i.shape[2]])),M(Yr(a,r),()=>`Error in pool: Either strides or dilations must be 1. Got strides ${a} and dilations '${r}'`);let c=sw(l.shape,t,a,r,s),p=[c.dilationHeight,c.dilationWidth],d;s==="same"?d=FO([c.filterHeight,c.filterWidth],p):d=[[0,0],[0,0]];let h=p[0]===1&&p[1]===1,[f,m]=PO([c.inHeight,c.inWidth],p,d),g=h?s:"valid",y=h?l:gh(l,p,f),A=(n==="avg"?()=>lh(y,t,a,g,o):()=>fh(y,t,a,g,o))(),b=h?A:uh(A,p,m);return u?U(b,[b.shape[1],b.shape[2],b.shape[3]]):b}function PO(e,t,n){let s=n.map(c=>c[0]),r=n.map(c=>c[1]),a=e.concat(s,r),o=t.map((c,p)=>(c-a[p]%c)%c),i=r.map((c,p)=>c+o[p]),l=t.map((c,p)=>[s[p],i[p]]),u=t.map((c,p)=>[0,o[p]]);return[l,u]}function FO(e,t){let s=e.map((o,i)=>o+(o-1)*(t[i]-1)).map(o=>o-1),r=s.map(o=>Math.floor(o/2)),a=s.map((o,i)=>o-r[i]);return s.map((o,i)=>[r[i],a[i]])}var RA=V({pool_:$O});function OO(e,t){let n=D(e,"x","prelu"),s=D(t,"alpha","prelu"),r={x:n,alpha:s};return W.runKernel(Go,r)}var yh=V({prelu_:OO});function MO(e,t=null,n=!1){let s=D(e,"x","prod");s.dtype==="bool"&&(s=ye(s,"int32"));let r={x:s},a={axis:t,keepDims:n};return W.runKernel(Ho,r,a)}var _A=V({prod_:MO});function zO(e,t,n){let s=Et(e),r=null;if(n==null||n==="float32")r=new Float32Array(s);else if(n==="int32")r=new Int32Array(s);else if(n==="bool")r=new Uint8Array(s);else throw new Error(`Unknown data type ${n}`);for(let a=0;a<s;a++)r[a]=t();return W.makeTensor(r,e,n)}var Nw=V({rand_:zO}),DA=uo(Wm()),$A=class{constructor(e,t,n,s,r){this.mean=e,this.stdDev=t,this.dtype=n,this.nextVal=NaN,this.truncated=s,this.truncated&&(this.upper=this.mean+this.stdDev*2,this.lower=this.mean-this.stdDev*2);let a=r||Math.random();this.random=DA.alea(a.toString())}nextValue(){if(!isNaN(this.nextVal)){let s=this.nextVal;return this.nextVal=NaN,s}let e,t,n=!1;for(;!n;){let s,r,a;do s=2*this.random()-1,r=2*this.random()-1,a=s*s+r*r;while(a>=1||a===0);let o=Math.sqrt(-2*Math.log(a)/a);e=this.mean+this.stdDev*s*o,t=this.mean+this.stdDev*r*o,(!this.truncated||this.isValidTruncated(e))&&(n=!0)}return(!this.truncated||this.isValidTruncated(t))&&(this.nextVal=this.convertValue(t)),this.convertValue(e)}convertValue(e){return this.dtype==null||this.dtype==="float32"?e:Math.round(e)}isValidTruncated(e){return e<=this.upper&&e>=this.lower}},LO=class{constructor(e,t,n,s){this.alpha=e,this.beta=1/t,this.dtype=n;let r=s||Math.random();this.randu=DA.alea(r.toString()),this.randn=new $A(0,1,n,!1,this.randu()),e<1?this.d=e+2/3:this.d=e-1/3,this.c=1/Math.sqrt(9*this.d)}nextValue(){let e,t,n,s,r,a;for(;;){do s=this.randn.nextValue(),a=1+this.c*s;while(a<=0);if(a*=a*a,e=s*s,t=1-.331*e*e,n=.5*e+this.d*(1-a+Math.log(a)),r=this.randu(),r<t||Math.log(r)<n)break}return a=1/this.beta*this.d*a,this.alpha<1&&(a*=Math.pow(this.randu(),1/this.alpha)),this.convertValue(a)}convertValue(e){return this.dtype==="float32"?e:Math.round(e)}},BO=class{constructor(e=0,t=1,n,s){if(this.canReturnFloat=()=>this.dtype==null||this.dtype==="float32",this.min=e,this.range=t-e,this.dtype=n,s==null&&(s=Math.random()),typeof s=="number"&&(s=s.toString()),!this.canReturnFloat()&&this.range<=1)throw new Error(`The difference between ${e} - ${t} <= 1 and dtype is not float`);this.random=DA.alea(s)}convertValue(e){return this.canReturnFloat()?e:Math.round(e)}nextValue(){return this.convertValue(this.min+this.range*this.random())}};function WO(e,t,n=1,s="float32",r){if(n==null&&(n=1),s==null&&(s="float32"),s!=="float32"&&s!=="int32")throw new Error(`Unsupported data type ${s}`);let a=new LO(t,n,s,r),o=Le(e,s);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var Ew=V({randomGamma_:WO});function VO(e,t=0,n=1,s,r){if(s!=null&&s==="bool")throw new Error(`Unsupported data type ${s}`);let a=new $A(t,n,s,!1,r),o=Le(e,s);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var N0=V({randomNormal_:VO});function UO(e,t,n){if(t!=null&&t==="bool")throw new Error(`Unsupported data type ${t}`);return N0(e,0,1,t,n)}var Rw=V({randomStandardNormal_:UO});function GO(e,t=0,n=1,s="float32",r){let a=Le(e,s),o=new BO(t,n,null,r);for(let i=0;i<a.values.length;i++)a.values[i]=o.nextValue();return a.toTensor()}var Kc=V({randomUniform_:GO});function Ju(e,t,n=1,s="float32"){if(n===0)throw new Error("Cannot have a step of zero");let r={start:e,stop:t,step:n,dtype:s};return W.runKernel(_c,{},r)}function HO(e){let n={x:D(e,"x","reciprocal")};return W.runKernel(Dc,n)}var PA=V({reciprocal_:HO});function jO(e){let n={x:D(e,"x","relu")};return W.runKernel(jo,n)}var Pr=V({relu_:jO});function qO(e){let n={x:D(e,"x","relu6")};return W.runKernel(Ko,n)}var E0=V({relu6_:qO});function XO(e,t){let s={x:D(e,"x","reverse")},r={dims:t};return W.runKernel(Dl,s,r)}var Zs=V({reverse_:XO});function KO(e){let t=D(e,"x","reverse");return M(t.rank===1,()=>`Error in reverse1D: x must be rank 1 but got rank ${t.rank}.`),Zs(t,0)}var _w=V({reverse1d_:KO});function ZO(e,t){let n=D(e,"x","reverse");return M(n.rank===2,()=>`Error in reverse2D: x must be rank 2 but got rank ${n.rank}.`),Zs(n,t)}var Dw=V({reverse2d_:ZO});function YO(e,t){let n=D(e,"x","reverse");return M(n.rank===3,()=>`Error in reverse3D: x must be rank 3 but got rank ${n.rank}.`),Zs(n,t)}var $w=V({reverse3d_:YO});function JO(e,t){let n=D(e,"x","reverse");return M(n.rank===4,()=>`Error in reverse4D: x must be rank 4 but got rank ${n.rank}.`),Zs(n,t)}var Pw=V({reverse4d_:JO});function QO(e){let n={x:D(e,"x","round")};return W.runKernel($l,n)}var R0=V({round_:QO});function eM(e){let n={x:D(e,"x","rsqrt","float32")};return W.runKernel(Zo,n)}var _0=V({rsqrt_:eM});function tM(e){let n={x:D(e,"x","selu")};return W.runKernel($c,n)}var D0=V({selu_:tM});function nM(e,t,n,s,r,a=[1,1],o="NHWC"){let i=D(e,"x","separableConv2d"),l=D(t,"depthwiseFilter","separableConv2d"),u=D(n,"pointwiseFilter","separableConv2d"),c=i,p=!1;if(i.rank===3&&(p=!0,c=U(i,[1,i.shape[0],i.shape[1],i.shape[2]])),o==="NCHW")throw new Error("separableConv2d currently does not support dataFormat NCHW; only NHWC is supported");M(c.rank===4,()=>`Error in separableConv2d: input must be rank 4, but got rank ${c.rank}.`),M(l.rank===4,()=>`Error in separableConv2d: depthwise filter must be rank 4, but got rank ${l.rank}.`),M(u.rank===4,()=>`Error in separableConv2d: pointwise filter must be rank 4, but got rank ${l.rank}.`),M(u.shape[0]===1,()=>`Error in separableConv2d: the first dimension of pointwise filter must be 1, but got ${u.shape[0]}.`),M(u.shape[1]===1,()=>`Error in separableConv2d: the second dimension of pointwise filter must be 1, but got ${u.shape[1]}.`);let d=l.shape[2],h=l.shape[3];M(u.shape[2]===d*h,()=>`Error in separableConv2d: the third dimension of pointwise filter must be ${d*h}, but got ${u.shape[2]}.`);let f=Vc(c,l,s,r,o,a),g=ma(f,u,1,"valid",o);return p?U(g,[g.shape[1],g.shape[2],g.shape[3]]):g}var $0=V({separableConv2d_:nM});async function sM(e,t){let n=D(e,"x","setdiff1d"),s=D(t,"y","setdiff1d");M(n.dtype===s.dtype,()=>`x and y should have the same dtype, but got x (${n.dtype}) and y (${s.dtype}).`),M(n.rank===1,()=>`x should be 1D tensor, but got x (${n.shape}).`),M(s.rank===1,()=>`y should be 1D tensor, but got y (${s.shape}).`);let r=await n.data(),a=await s.data(),o=new Set(a),i=0;for(let c=0;c<r.length;c++)o.has(r[c])||i++;let l=new fn([i],n.dtype),u=new fn([i],"int32");for(let c=0,p=0;c<r.length;c++)o.has(r[c])||(l.values[p]=r[c],u.values[p]=c,p++);return[l.toTensor(),u.toTensor()]}var Fw=sM;function rM(e){let n={x:D(e,"x","sign")};return W.runKernel(Pc,n)}var FA=V({sign_:rM});function aM(e){let n={x:D(e,"x","sin","float32")};return W.runKernel(Yo,n)}var P0=V({sin_:aM});function oM(e){let n={x:D(e,"x","sinh")};return W.runKernel(Ml,n)}var F0=V({sinh_:oM});function iM(e,t,n){let s=D(e,"x","slice1d");return M(s.rank===1,()=>`slice1d expects a rank-1 tensor, but got a rank-${s.rank} tensor`),Oe(s,[t],[n])}var Ah=V({slice1d_:iM});function lM(e,t,n){let s=D(e,"x","slice2d");return M(s.rank===2,()=>`slice2d expects a rank-2 tensor, but got a rank-${s.rank} tensor`),Oe(s,t,n)}var O0=V({slice2d_:lM});function uM(e,t,n){let s=D(e,"x","slice3d");return M(s.rank===3,()=>`slice3d expects a rank-3 tensor, but got a rank-${s.rank} tensor`),Oe(s,t,n)}var ci=V({slice3d_:uM});function cM(e,t,n){let s=D(e,"x","slice4d");return M(s.rank===4,()=>`slice4d expects a rank-4 tensor, but got a rank-${s.rank} tensor`),Oe(s,t,n)}var no=V({slice4d_:cM});function dM(e,t=-1){let n=D(e,"logits","softmax","float32");if(t===-1&&(t=n.rank-1),t!==n.rank-1)throw Error(`Softmax along a non-last dimension is not yet supported. Logits was rank ${n.rank} and dim was ${t}`);let s={logits:n},r={dim:t};return W.runKernel(ti,s,r)}var Jl=V({softmax_:dM});function pM(e){M(e.dtype==="complex64",()=>`The dtype for tf.spectral.fft() must be complex64 but got ${e.dtype}.`);let t={input:e};return W.runKernel(n0,t)}var xh=V({fft_:pM});function hM(e){M(e.dtype==="complex64",()=>`The dtype for tf.spectral.ifft() must be complex64 but got ${e.dtype}.`);let t={input:e};return W.runKernel(s0,t)}var Qu=V({ifft_:hM});function fM(e){let t=e.shape[e.shape.length-1],n=e.size/t,s;if(t<=2){let r=U(e,[n,t]);s=Qu(r)}else{let r=[n,2*(t-1)],a=U(Yu(e),[n,t]),o=U(oh(e),[n,t]),i=Zs(Oe(a,[0,1],[n,t-2]),1),l=L(Zs(Oe(o,[0,1],[n,t-2]),1),Ce(-1)),u=Ct([a,i],1),c=Ct([o,l],1),p=U(fa(u,c),[r[0],r[1]]);s=Qu(p)}if(s=Yu(s),e.rank===3&&e.shape[0]!==0){let r=s,a=e.shape[0];s=U(s,[a,s.shape[0]/a,s.shape[1]]),r.dispose()}return s}var M0=V({irfft_:fM});function mM(e,t,n=0){let r={x:D(e,"x","split")},a={numOrSizeSplits:t,axis:n};return W.runKernel(Ll,r,a)}var Zt=V({split_:mM});function gM(e,t){M(e.dtype==="float32",()=>`The dtype for rfft() must be real value but got ${e.dtype}`);let n=e.shape[e.shape.length-1],s=e.size/n,r;if(t!=null&&t<n){let f=e.shape.map(g=>0),m=e.shape.map(g=>g);m[e.shape.length-1]=t,r=Oe(e,f,m),n=t}else if(t!=null&&t>n){let f=e.shape.map(m=>m);f[e.shape.length-1]=t-n,r=Ct([e,Wt(f)],e.shape.length-1),n=t}else r=e;let a=it(r),o=U(fa(r,a),[s,n]),i=xh(o),l=Math.floor(n/2)+1,u=Yu(i),c=oh(i),p=Zt(u,[l,n-l],u.shape.length-1),d=Zt(c,[l,n-l],c.shape.length-1),h=r.shape.slice();return h[r.shape.length-1]=l,U(fa(p[0],d[0]),h)}var bh=V({rfft_:gM});function yM(e,t){let n=D(e,"a","squaredDifference"),s=D(t,"b","squaredDifference");[n,s]=Gt(n,s),kt(n.shape,s.shape);let r={a:n,b:s},a={};return W.runKernel(ni,r,a)}var z0=V({squaredDifference_:yM});function AM(e,t){let n=D(e,"x","squeeze","string_or_numeric");return U(n,n6(n.shape,t).newShape)}var st=V({squeeze_:AM});function xM(e,t=0){let n=Sp(e,"tensors","stack","string_or_numeric");M(n.length>=1,()=>"Pass at least one tensor to tf.stack"),n.length>0&&M(t<=n[0].rank,()=>"Axis must be <= rank of the tensor");let s=n,r={axis:t};return W.runKernel(Rl,s,r)}var on=V({stack_:xM});function bM(e,t=0){let s={x:D(e,"x","step")},r={alpha:t};return W.runKernel(ai,s,r)}var Ql=V({step_:bM});function vM(e,t,n,s,r=0,a=0,o=0,i=0,l=0){let c={x:D(e,"x","stridedSlice","string_or_numeric")},p={begin:t,end:n,strides:s,beginMask:r,endMask:a,ellipsisMask:o,newAxisMask:i,shrinkAxisMask:l};return W.runKernel(Bl,c,p)}var OA=V({stridedSlice_:vM});function wM(e){let n={x:D(e,"x","tan","float32")};return W.runKernel(Wl,n)}var MA=V({tan_:wM});function Ft(e,t){ol(e);let n=Xr(e,t);if(n.length!==1)throw new Error("tensor1d() requires values to be a flat/TypedArray");return oi(e,null,n,t)}function ir(e,t,n){if(ol(e),t!=null&&t.length!==2)throw new Error("tensor2d() requires shape to have two numbers");let s=Xr(e,n);if(s.length!==2&&s.length!==1)throw new Error("tensor2d() requires values to be number[][] or flat/TypedArray");if(s.length===1&&t==null)throw new Error("tensor2d() requires shape to be provided when `values` are a flat/TypedArray");return oi(e,t,s,n)}function Ow(e,t,n){if(ol(e),t!=null&&t.length!==4)throw new Error("tensor4d() requires shape to have four numbers");let s=Xr(e,n);if(s.length!==4&&s.length!==1)throw new Error("tensor4d() requires values to be number[][][][] or flat/TypedArray");if(s.length===1&&t==null)throw new Error("tensor4d() requires shape to be provided when `values` are a flat array");return oi(e,t,s,n)}function Mw(e,t,n){if(ol(e),t!=null&&t.length!==5)throw new Error("tensor5d() requires shape to have five numbers");let s=Xr(e,n);if(s.length!==5&&s.length!==1)throw new Error("tensor5d() requires values to be number[][][][][] or flat/TypedArray");if(s.length===1&&t==null)throw new Error("tensor5d() requires shape to be provided when `values` are a flat array");return oi(e,t,s,n)}function zw(e,t,n){if(ol(e),t!=null&&t.length!==6)throw new Error("tensor6d() requires shape to have six numbers");let s=Xr(e,n);if(s.length!==6&&s.length!==1)throw new Error("tensor6d() requires values to be number[][][][][][] or flat/TypedArray");if(s.length===1&&t==null)throw new Error("tensor6d() requires shape to be provided when `values` are a flat array");return t=t||s,oi(e,t,s,n)}function kM(e,t=1,n=!0){let s=D(e,"x","topk");if(s.rank===0)throw new Error("topk() expects the input to be of rank 1 or higher");let r=s.shape[s.shape.length-1];if(t<0)throw new Error(`'k' passed to topk() must be >= 0 but got ${t}`);if(t>r)throw new Error(`'k' passed to topk() must be <= the last dimension (${r}) but got ${t}`);let a={x:s},o={k:t,sorted:n},[i,l]=W.runKernel(Vl,a,o);return{values:i,indices:l}}var zA=V({topk_:kM});function IM(e,t=0,n=1,s,r){if(s!=null&&s==="bool")throw new Error("Unsupported data type $ { dtype }");let a=new $A(t,n,s,!0,r),o=Le(e,s);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var L0=V({truncatedNormal_:IM});function SM(e,t=0){let n=D(e,"x","unique","string_or_numeric");M(n.rank>0,()=>"The input tensor must be at least 1D");let s={x:n},r={axis:t},[a,o]=W.runKernel(h0,s,r);return{values:a,indices:o}}var LA=V({unique_:SM});function CM(e,t,n){let s=D(e,"x","unsortedSegmentSum"),r=D(t,"segmentIds","unsortedSegmentSum","int32");M(qu(n),()=>"numSegments must be of dtype int");let a={x:s,segmentIds:r},o={numSegments:n};return W.runKernel(eh,a,o)}var B0=V({unsortedSegmentSum_:CM});function TM(e,t=0){let n=D(e,"x","unstack","string_or_numeric");M(t>=-n.shape.length&&t<n.shape.length,()=>`Axis = ${t} is not in [-${n.shape.length}, ${n.shape.length})`);let s={value:n},r={axis:t};return W.runKernel(Gl,s,r)}var En=V({unstack_:TM});function Lw(e,t){return T0(e,t,"right")}function BA(e,t=!0,n,s){return W.makeVariable(e,t,n,s)}function Bw(e,t){let n=[];for(let a=0;a<t.length;a++)t[a]&&n.push(a);let s=Le(e,"int32"),r=Le([n.length,e.length],"int32");for(let a=0;a<n.length;a++){let o=s.indexToLoc(n[a]),i=a*e.length;r.values.set(o,i)}return r.toTensor()}async function NM(e){let t=D(e,"condition","whereAsync","bool"),n=await t.data(),s=Bw(t.shape,n);return e!==t&&t.dispose(),s}var WA=NM;async function EM(e,t,n){let s=D(e,"tensor","boolMask"),r=D(t,"mask","boolMask","bool"),a=n==null?0:n,o=r.rank,i=s.shape;M(o>0,()=>"mask cannot be scalar"),ns(i.slice(a,a+o),r.shape,"mask's shape must match the first K dimensions of tensor's shape,");let l=1;for(let m=a;m<a+o;m++)l*=i[m];let u=i.slice(0,a).concat([l],i.slice(a+o)),c=U(s,u),p=U(r,[-1]),d=await WA(p),h=st(d,[1]),f=qc(c,h,a);return e!==s&&s.dispose(),t!==r&&r.dispose(),h.dispose(),c.dispose(),p.dispose(),d.dispose(),f}var Ww=EM;function RM(e,t,n,s,r=!0){let a=D(e,"v","movingAverage"),o=D(t,"x","movingAverage"),i=D(n,"decay","movingAverage");b6(a,o),M(co(a.shape,o.shape),()=>"Shape mismatch in v and x");let l=Ce(1),u=fe(l,i),c=L(fe(o,a),u);if(r){M(s!=null,()=>"When using zeroDebias: true, step is required.");let p=D(s,"step","movingAverage");c=he(c,fe(l,ya(i,p)))}return ce(a,c)}var Vw=V({movingAverage_:RM});function _M(e,t,n){let s=D(e,"indices","scatterND","int32"),r=D(t,"updates","scatterND");Vy(r,s,n);let a={indices:s,updates:r},o={shape:n};return W.runKernel(Pl,a,o)}var Uw=V({scatterND_:_M});function DM(e,t,n,s){if(e.dtype!=="int32")throw new Error(`tf.sparseToDense() expects the indices to be int32 type, but the dtype was ${e.dtype}.`);if(e.rank>2)throw new Error(`sparseIndices should be a scalar, vector, or matrix, but got shape ${e.shape}.`);let r=e.rank>0?e.shape[0]:1,a=e.rank>1?e.shape[1]:1;if(n.length!==a)throw new Error(`outputShape has incorrect number of elements:, ${n.length}, should be: ${a}.`);let o=t.size;if(!(t.rank===0||t.rank===1&&o===r))throw new Error(`sparseValues has incorrect shape ${t.shape}, should be [] or [${r}]`);if(t.dtype!==s.dtype)throw new Error("sparseValues.dtype must match defaultValues.dtype")}function $M(e,t,n,s=0){let r=D(e,"sparseIndices","sparseToDense","int32"),a=D(t,"sparseValues","sparseToDense","string_or_numeric"),o=D(s,"defaultValue","sparseToDense",a.dtype);DM(r,a,n,o);let i={sparseIndices:r,sparseValues:a,defaultValue:o},l={outputShape:n};return W.runKernel(Yp,i,l)}var Gw=V({sparseToDense_:$M});function PM(e,t){let n=D(t,"indices","gatherND","int32"),r={params:D(e,"x","gatherND","string_or_numeric"),indices:n};return W.runKernel(Al,r)}var Hw=V({gatherND_:PM});function FM(e,t){if(t==null)return e.shape.slice();if(co(e.shape,t))return t;if(e.shape.length===t.length){let n=[];for(let s=0;s<e.shape.length;s++)t[s]==null&&e.shape[s]!=null?n.push(e.shape[s]):n.push(t[s]);return n}return t}function OM(e,t,n,s){let r=D(e,"x","dropout");if(M(r.dtype==="float32",()=>`x has to be a floating point tensor since it's going to be scaled, but got a ${r.dtype} tensor instead.`),M(t>=0&&t<1,()=>`rate must be a float in the range [0, 1), but got ${t}.`),t===0)return e instanceof nt?r.clone():r;let a=FM(r,n),o=1-t,i=he(jc(ce(Kc(a,0,1,"float32",s),o)),o);return L(r,i)}var VA=V({dropout_:OM});function UA(e){return Math.floor(Math.pow(2,Math.ceil(Math.log(e)/Math.log(2))))}function W0(e,t,n){let s=1-e%2,r=new Float32Array(e);for(let a=0;a<e;++a){let o=2*Math.PI*a/(e+s-1);r[a]=t-n*Math.cos(o)}return Ft(r,"float32")}async function MM(e,t,n=1){let s=D(e,"predictions","inTopK"),r=D(t,"targets","inTopK");M(s.rank>1,()=>`inTopK() expects the predictions to be of rank 2 or higher, but got ${s.rank}`),M(s.rank-1===r.rank,()=>`predictions rank should be 1 larger than targets rank, but got predictions rank ${s.rank} and targets rank ${r.rank}`),ns(s.shape.slice(0,s.shape.length-1),r.shape,"predictions's shape should be align with the targets' shape, except the last dimension.");let a=s.shape[s.shape.length-1];M(n>0&&n<=a,()=>`'k' passed to inTopK() must be > 0 && <= the predictions last dimension (${a}), but got ${n}`);let o=await s.data(),i=await r.data(),[l,u]=[o.length/a,a],c=s6("bool",l);for(let p=0;p<l;p++){let d=p*u,h=o.subarray(d,d+u),f=[];for(let m=0;m<h.length;m++)f.push({value:h[m],index:m});f.sort((m,g)=>g.value-m.value),c[p]=0;for(let m=0;m<n;m++)if(f[m].index===i[p]){c[p]=1;break}}return e!==s&&s.dispose(),t!==r&&r.dispose(),ct(c,r.shape,"bool")}var jw=MM,ec={};Ve(ec,{conv2d:()=>BM,depthwiseConv2d:()=>GM,matMul:()=>jM});function zM(e,t,n,s,r,a="NHWC",o){let i=e;e.rank===3&&(i=U(e,[1,e.shape[0],e.shape[1],e.shape[2]]));let l=t;l.rank===3&&(l=U(t,[1,t.shape[0],t.shape[1],t.shape[2]])),M(i.rank===4,()=>`Error in conv2dDerFilter: input must be rank 4, but got shape ${i.shape}.`),M(l.rank===4,()=>`Error in conv2dDerFilter: dy must be rank 4, but got shape ${l.shape}.`),M(n.length===4,()=>`Error in conv2dDerFilter: filterShape must be length 4, but got ${n}.`);let u=a==="NHWC"?i.shape[3]:i.shape[1],c=a==="NHWC"?l.shape[3]:l.shape[1];M(u===n[2],()=>`Error in conv2dDerFilter: depth of input ${u}) must match input depth in filter (${n[2]}.`),M(c===n[3],()=>`Error in conv2dDerFilter: depth of dy (${c}) must match output depth for filter (${n[3]}).`),rs("conv2dDerFilter",r,o);let p={x:i,dy:l},d={strides:s,pad:r,dataFormat:a,dimRoundingMode:o,filterShape:n};return W.runKernel(Xm,p,d)}var GA=V({conv2DBackpropFilter_:zM});function V0(e,t,n){if(n==null||n==="linear")return e;if(n==="relu")return L(e,Ql(t));throw new Error(`Cannot compute gradient for fused activation ${n}.`)}function U0(e,t){let n=t,s=an(e.shape,t.shape);return s.length>0&&(n=ke(n,s)),U(n,e.shape)}function G0(e,t,n,s){if(t==="linear")return e;if(t==="relu")return Pr(e);if(t==="elu")return Uc(e);if(t==="relu6")return E0(e);if(t==="prelu")return yh(e,n);if(t==="leakyrelu")return dh(e,s);if(t==="sigmoid")return Cn(e);throw new Error(`Unknown fused activation ${t}.`)}var H0=(e,t)=>!(e>0)||t==="linear";function LM({x:e,filter:t,strides:n,pad:s,dataFormat:r="NHWC",dilations:a=[1,1],dimRoundingMode:o,bias:i,activation:l="linear",preluActivationWeights:u,leakyreluAlpha:c}){if(l=l||"linear",H0(W.state.gradientDepth,l)===!1){M(r==="NHWC",()=>`Error in fused conv2d: got dataFormat of ${r} but only NHWC is currently supported for the case of gradient depth is 0 and the activation is not linear.`);let k=ma(e,t,n,s,r,a,o);return i!=null&&(k=ce(k,i)),G0(k,l,u,c)}let p=D(e,"x","conv2d","float32"),d=D(t,"filter","conv2d","float32"),h=p,f=!1;p.rank===3&&(f=!0,h=U(p,[1,p.shape[0],p.shape[1],p.shape[2]])),M(h.rank===4,()=>`Error in fused conv2d: input must be rank 4, but got rank ${h.rank}.`),M(d.rank===4,()=>`Error in fused conv2d: filter must be rank 4, but got rank ${d.rank}.`),rs("fused conv2d",s,o);let m=r==="NHWC"?h.shape[3]:h.shape[1];M(d.shape[2]===m,()=>`Error in conv2d: depth of input (${m}) must match input depth for filter ${d.shape[2]}.`),M(Yr(n,a),()=>`Error in conv2D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`);let g=ih(h.shape,d.shape,n,a,s,o),y;i!=null&&(y=D(i,"bias","fused conv2d"),[y]=Gt(y,p),r==="NHWC"?kt(g.outShape,y.shape):(M(y.shape.length<=1,()=>`Error in fused conv2d: only supports scalar or 1-D Tensor bias for NCHW format but got the bias of rank-${y.shape.length}.`),M(y.shape.length===0||y.shape[0]===g.outChannels||y.shape[0]===1,()=>`Error in fused conv2d: bias shape (${y.shape}) is not compatible with the number of output channels (${g.outChannels})`)));let x;if(u!=null){let k=u.shape;if(M(k.length<=1||k.length===3,()=>`Error in fused conv2d: only supports scalar, 1-D Tensor or 3-D Tensor PReLU activation weights but got a tensor of rank-${k.length}.`),k.length===1)M(k[0]===1||k[0]===g.outChannels,()=>`Error in fused conv2d: PReLU activation weights (${k}) is not compatible with the number of output channels (${g.outChannels}).`);else if(k.length===3)try{kt(k,g.outShape)}catch(S){let E=`Error in fused conv2d: PReLU activation weights (${k}) is not compatible with the output shape of the conv2d (${g.outShape}).`;throw Error(E)}x=D(u,"prelu weights","fused conv2d")}let A=(k,S)=>{M(r==="NHWC",()=>`Error in gradient of fused conv2D: got dataFormat of ${r} but only NHWC is currently supported.`);let[E,R,$,_]=S,P=V0(k,$,l);M(to(a),()=>`Error in gradient of fused conv2D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${a}'`);let C=cA(R.shape,P,E,n,s),F=GA(R,P,E.shape,n,s),G=[C,F];if(_!=null){let K=U0(_,P);G.push(K)}return G},b={x:h,filter:d,bias:y,preluActivationWeights:x},w={strides:n,pad:s,dataFormat:r,dilations:a,dimRoundingMode:o,activation:l,leakyreluAlpha:c};return i==null?Kr((S,E,R)=>{let $=W.runKernel(Qa,b,w);return R([E,S,$]),f&&($=U($,[$.shape[1],$.shape[2],$.shape[3]])),{value:$,gradFunc:A}})(h,d):Kr((S,E,R,$)=>{let _=W.runKernel(Qa,b,w);return $([E,S,_,R]),f&&(_=U(_,[_.shape[1],_.shape[2],_.shape[3]])),{value:_,gradFunc:A}})(h,d,y)}var BM=V({fusedConv2d_:LM});function WM(e,t,n,s,r,a=[1,1],o){let i=e;e.rank===3&&(i=U(e,[1,e.shape[0],e.shape[1],e.shape[2]]));let l=t;l.rank===3&&(l=U(t,[1,t.shape[0],t.shape[1],t.shape[2]]));let u={x:i,dy:l},c={strides:s,pad:r,dimRoundingMode:o,dilations:a,filterShape:n};return W.runKernel(Jm,u,c)}var qw=V({depthwiseConv2dNativeBackpropFilter_:WM});function VM(e,t,n,s,r,a=[1,1],o){let i=t,l=!1;t.rank===3&&(l=!0,i=U(t,[1,t.shape[0],t.shape[1],t.shape[2]]));let u={dy:i,filter:n},c={strides:s,pad:r,dimRoundingMode:o,dilations:a,inputShape:e},p=W.runKernel(Qm,u,c);return l?U(p,[p.shape[1],p.shape[2],p.shape[3]]):p}var Xw=V({depthwiseConv2dNativeBackpropInput_:VM});function UM({x:e,filter:t,strides:n,pad:s,dataFormat:r="NHWC",dilations:a=[1,1],dimRoundingMode:o,bias:i,activation:l="linear",preluActivationWeights:u,leakyreluAlpha:c}){if(H0(W.state.gradientDepth,l)===!1){let w=Vc(e,t,n,s,r,a,o);return i!=null&&(w=ce(w,i)),G0(w,l,u,c)}let p=D(e,"x","depthwiseConv2d","float32"),d=D(t,"filter","depthwiseConv2d","float32"),h=p,f=!1;p.rank===3&&(f=!0,h=U(p,[1,p.shape[0],p.shape[1],p.shape[2]])),M(h.rank===4,()=>`Error in fused depthwiseConv2d: input must be rank 4, but got rank ${h.rank}.`),M(d.rank===4,()=>`Error in fused depthwiseConv2d: filter must be rank 4, but got rank ${d.rank}.`),M(h.shape[3]===d.shape[2],()=>`Error in fused depthwiseConv2d: number of input channels (${h.shape[3]}) must match the inChannels dimension in filter ${d.shape[2]}.`),a==null&&(a=[1,1]),M(Yr(n,a),()=>`Error in fused depthwiseConv2d: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`),rs("fused depthwiseConv2d",s,o);let m=ih(h.shape,d.shape,n,a,s,o,!0),g;i!=null&&(g=D(i,"bias","fused conv2d"),[g]=Gt(g,p),kt(m.outShape,g.shape));let y;u!=null&&(y=D(u,"prelu weights","fused depthwiseConv2d"));let x=(w,k)=>{M(to(a),()=>`Error in gradient of fused depthwiseConv2d: dilation rates greater than 1 are not yet supported. Got dilations '${a}'`);let[S,E,R,$]=k,_=V0(w,R,l),P=Xw(E.shape,_,S,n,s,a,o),C=qw(E,_,S.shape,n,s,a,o);if($!=null){let F=U0(g,_);return[P,C,F]}return[P,C]},A={x:h,filter:d,bias:g,preluActivationWeights:y},b={strides:n,pad:s,dataFormat:r,dilations:a,dimRoundingMode:o,activation:l,leakyreluAlpha:c};return i==null?Kr((k,S,E)=>{let R=W.runKernel(eo,A,b);return E([S,k,R]),f&&(R=U(R,[R.shape[1],R.shape[2],R.shape[3]])),{value:R,gradFunc:x}})(h,d):Kr((k,S,E,R)=>{let $=W.runKernel(eo,A,b);return R([S,k,$,E]),f&&($=U($,[$.shape[1],$.shape[2],$.shape[3]])),{value:$,gradFunc:x}})(h,d,g)}var GM=V({fusedDepthwiseConv2d_:UM});function HM({a:e,b:t,transposeA:n=!1,transposeB:s=!1,bias:r,activation:a="linear",preluActivationWeights:o,leakyreluAlpha:i=.2}){if(H0(W.state.gradientDepth,a)===!1){let _=Qe(e,t,n,s);return r!=null&&(_=ce(_,r)),G0(_,a,o,i)}let l=D(e,"a","fused matMul"),u=D(t,"b","fused matMul");[l,u]=Gt(l,u);let c=n?l.shape[l.rank-2]:l.shape[l.rank-1],p=s?u.shape[u.rank-1]:u.shape[u.rank-2],d=n?l.shape[l.rank-1]:l.shape[l.rank-2],h=s?u.shape[u.rank-2]:u.shape[u.rank-1],f=l.shape.slice(0,-2),m=u.shape.slice(0,-2),g=Et(f),y=Et(m);M(c===p,()=>`Error in fused matMul: inner shapes (${c}) and (${p}) of Tensors with shapes ${l.shape} and ${u.shape} and transposeA=${n} and transposeB=${s} must match.`);let A=kt(l.shape.slice(0,-2),u.shape.slice(0,-2)).concat([d,h]),b=n?U(l,[g,c,d]):U(l,[g,d,c]),w=s?U(u,[y,h,p]):U(u,[y,p,h]),k;r!=null&&(k=D(r,"bias","fused matMul"),[k]=Gt(k,l),kt(A,k.shape));let S;o!=null&&(S=D(o,"prelu weights","fused matMul"));let E=(_,P)=>{let[C,F,G,K]=P,z=V0(U(_,G.shape),G,a),Z,J;if(!n&&!s?(Z=Qe(z,F,!1,!0),J=Qe(C,z,!0,!1)):!n&&s?(Z=Qe(z,F,!1,!1),J=Qe(z,C,!0,!1)):n&&!s?(Z=Qe(F,z,!1,!0),J=Qe(C,z,!1,!1)):(Z=Qe(F,z,!0,!0),J=Qe(z,C,!0,!0)),r!=null){let te=U0(K,z);return[Z,J,te]}else return[Z,J]},R={a:b,b:w,bias:k,preluActivationWeights:S},$={transposeA:n,transposeB:s,activation:a,leakyreluAlpha:i};return r==null?Kr((P,C,F)=>{let G=W.runKernel(Ja,R,$);return F([P,C,G]),{value:U(G,A),gradFunc:E}})(b,w):Kr((P,C,F,G)=>{let K=W.runKernel(Ja,R,$);return G([P,C,K,F]),{value:U(K,A),gradFunc:E}})(b,w,k)}var jM=V({fusedMatMul_:HM});function qM(e){return W0(e,.54,.46)}var XM=V({hammingWindow_:qM});function KM(e){return W0(e,.5,.5)}var Kw=V({hannWindow_:KM});function ZM(e,t,n,s=!1,r=0){let a=0,o=[];for(;a+t<=e.size;)o.push(Oe(e,a,t)),a+=n;if(s)for(;a<e.size;){let i=a+t-e.size,l=Ct([Oe(e,a,t-i),Hc([i],r)]);o.push(l),a+=n}return o.length===0?ir([],[0,t]):U(Ct(o),[o.length,t])}var Zw=V({frame_:ZM});function YM(e,t,n,s,r=Kw){s==null&&(s=UA(t));let a=Zw(e,t,n),o=L(a,r(t));return bh(o,s)}var JM=V({stft_:YM});function QM(e,t,n,s,r="bilinear",a=0){let o=D(e,"image","cropAndResize"),i=D(t,"boxes","cropAndResize","float32"),l=D(n,"boxInd","cropAndResize","int32"),u=i.shape[0];M(o.rank===4,()=>`Error in cropAndResize: image must be rank 4,but got rank ${o.rank}.`),M(i.rank===2&&i.shape[1]===4,()=>`Error in cropAndResize: boxes must be have size [${u},4] but had shape ${i.shape}.`),M(l.rank===1&&l.shape[0]===u,()=>`Error in cropAndResize: boxInd must be have size [${u}] but had shape ${i.shape}.`),M(s.length===2,()=>`Error in cropAndResize: cropSize must be of length 2, but got length ${s.length}.`),M(s[0]>=1&&s[1]>=1,()=>`cropSize must be atleast [1,1], but was ${s}`),M(r==="bilinear"||r==="nearest",()=>`method must be bilinear or nearest, but was ${r}`);let c={image:o,boxes:i,boxInd:l},p={method:r,extrapolationValue:a,cropSize:s};return W.runKernel(dl,c,p)}var ez=V({cropAndResize_:QM});function tz(e){let t=D(e,"image","flipLeftRight","float32");M(t.rank===4,()=>`Error in flipLeftRight: image must be rank 4,but got rank ${t.rank}.`);let n={image:t};return W.runKernel(gl,n,{})}var nz=V({flipLeftRight_:tz});function sz(e){let t=D(e,"image","grayscaleToRGB"),n=t.rank-1,s=t.shape[n];M(t.rank>=2,()=>`Error in grayscaleToRGB: images must be at least rank 2, but got rank ${t.rank}.`),M(s===1,()=>`Error in grayscaleToRGB: last dimension of a grayscale image should be size 1, but got size ${s}.`);let r=new Array(t.rank);return r.fill(1,0,n),r[n]=3,qs(t,r)}var rz=V({grayscaleToRGB_:sz});function az(e,t,n=0,s=.5){let r=D(e,"image","rotateWithOffset","float32");M(r.rank===4,()=>`Error in rotateWithOffset: image must be rank 4,but got rank ${r.rank}.`);let a={image:r},o={radians:t,fillValue:n,center:s};return W.runKernel(jl,a,o)}var oz=V({rotateWithOffset_:az});function Zc(e,t,n,s,r,a){s==null&&(s=.5),r==null&&(r=Number.NEGATIVE_INFINITY),a==null&&(a=0);let o=e.shape[0];return n=Math.min(n,o),M(0<=s&&s<=1,()=>`iouThreshold must be in [0, 1], but was '${s}'`),M(e.rank===2,()=>`boxes must be a 2D tensor, but was of rank '${e.rank}'`),M(e.shape[1]===4,()=>`boxes must have 4 columns, but 2nd dimension was ${e.shape[1]}`),M(t.rank===1,()=>"scores must be a 1D tensor"),M(t.shape[0]===o,()=>`scores has incompatible shape with boxes. Expected ${o}, but was ${t.shape[0]}`),M(0<=a&&a<=1,()=>`softNmsSigma must be in [0, 1], but was '${a}'`),{maxOutputSize:n,iouThreshold:s,scoreThreshold:r,softNmsSigma:a}}function iz(e,t,n,s=.5,r=Number.NEGATIVE_INFINITY){let a=D(e,"boxes","nonMaxSuppression","float32"),o=D(t,"scores","nonMaxSuppression","float32"),i=Zc(a,o,n,s,r);n=i.maxOutputSize,s=i.iouThreshold,r=i.scoreThreshold;let l={maxOutputSize:n,iouThreshold:s,scoreThreshold:r};return W.runKernel(Cl,{boxes:a,scores:o},l)}var lz=V({nonMaxSuppression_:iz});function uz(e,t,n){let s=cz(e,t,n),r=s<0?-(s+1):s;e.splice(r,0,t)}function cz(e,t,n){return pz(e,t,n||dz)}function dz(e,t){return e>t?1:e<t?-1:0}function pz(e,t,n){let s=0,r=e.length,a=0,o=!1;for(;s<r;){a=s+(r-s>>>1);let i=n(t,e[a]);i>0?s=a+1:(r=a,o=!i)}return o?s:-s-1}function Yw(e,t,n,s,r){return HA(e,t,n,s,r,0)}function Jw(e,t,n,s,r,a){return HA(e,t,n,s,r,0,!1,a,!0)}function Qw(e,t,n,s,r,a){return HA(e,t,n,s,r,a,!0)}function HA(e,t,n,s,r,a,o=!1,i=!1,l=!1){let u=[];for(let g=0;g<t.length;g++)t[g]>r&&u.push({score:t[g],boxIndex:g,suppressBeginIndex:0});u.sort(Iv);let c=a>0?-.5/a:0,p=[],d=[];for(;p.length<n&&u.length>0;){let g=u.pop(),{score:y,boxIndex:x,suppressBeginIndex:A}=g;if(y<r)break;let b=!1;for(let w=p.length-1;w>=A;--w){let k=hz(e,x,p[w]);if(k>=s){b=!0;break}if(g.score=g.score*fz(s,c,k),g.score<=r)break}g.suppressBeginIndex=p.length,b||(g.score===y?(p.push(x),d.push(g.score)):g.score>r&&uz(u,g,Iv))}let h=p.length,f=n-h;i&&f>0&&(p.push(...new Array(f).fill(0)),d.push(...new Array(f).fill(0)));let m={selectedIndices:p};return o&&(m.selectedScores=d),l&&(m.validOutputs=h),m}function hz(e,t,n){let s=e.subarray(t*4,t*4+4),r=e.subarray(n*4,n*4+4),a=Math.min(s[0],s[2]),o=Math.min(s[1],s[3]),i=Math.max(s[0],s[2]),l=Math.max(s[1],s[3]),u=Math.min(r[0],r[2]),c=Math.min(r[1],r[3]),p=Math.max(r[0],r[2]),d=Math.max(r[1],r[3]),h=(i-a)*(l-o),f=(p-u)*(d-c);if(h<=0||f<=0)return 0;let m=Math.max(a,u),g=Math.max(o,c),y=Math.min(i,p),x=Math.min(l,d),A=Math.max(y-m,0)*Math.max(x-g,0);return A/(h+f-A)}function fz(e,t,n){let s=Math.exp(t*n*n);return n<=e?s:0}function Iv(e,t){return e.score-t.score||e.score===t.score&&t.boxIndex-e.boxIndex}async function mz(e,t,n,s=.5,r=Number.NEGATIVE_INFINITY){let a=D(e,"boxes","nonMaxSuppressionAsync"),o=D(t,"scores","nonMaxSuppressionAsync"),i=Zc(a,o,n,s,r);n=i.maxOutputSize,s=i.iouThreshold,r=i.scoreThreshold;let l=await Promise.all([a.data(),o.data()]),u=l[0],c=l[1],{selectedIndices:p}=Yw(u,c,n,s,r);return a!==e&&a.dispose(),o!==t&&o.dispose(),Ft(p,"int32")}var gz=mz;function yz(e,t,n,s=.5,r=Number.NEGATIVE_INFINITY,a=0){let o=D(e,"boxes","nonMaxSuppression"),i=D(t,"scores","nonMaxSuppression"),l=Zc(o,i,n,s,r,a);n=l.maxOutputSize,s=l.iouThreshold,r=l.scoreThreshold,a=l.softNmsSigma;let u={boxes:o,scores:i},c={maxOutputSize:n,iouThreshold:s,scoreThreshold:r,softNmsSigma:a},p=W.runKernel(Tl,u,c);return{selectedIndices:p[0],selectedScores:p[1]}}var Az=V({nonMaxSuppressionWithScore_:yz});async function xz(e,t,n,s=.5,r=Number.NEGATIVE_INFINITY,a=0){let o=D(e,"boxes","nonMaxSuppressionAsync"),i=D(t,"scores","nonMaxSuppressionAsync"),l=Zc(o,i,n,s,r,a);n=l.maxOutputSize,s=l.iouThreshold,r=l.scoreThreshold,a=l.softNmsSigma;let u=await Promise.all([o.data(),i.data()]),c=u[0],p=u[1],{selectedIndices:d,selectedScores:h}=Qw(c,p,n,s,r,a);return o!==e&&o.dispose(),i!==t&&i.dispose(),{selectedIndices:Ft(d,"int32"),selectedScores:Ft(h)}}var bz=xz;function vz(e,t,n,s=.5,r=Number.NEGATIVE_INFINITY,a=!1){let o=D(e,"boxes","nonMaxSuppression"),i=D(t,"scores","nonMaxSuppression"),l=Zc(o,i,n,s,r,null),u=l.maxOutputSize,c=l.iouThreshold,p=l.scoreThreshold,d={boxes:o,scores:i},h={maxOutputSize:u,iouThreshold:c,scoreThreshold:p,padToMaxOutputSize:a},f=W.runKernel(Rc,d,h);return{selectedIndices:f[0],validOutputs:f[1]}}var wz=V({nonMaxSuppressionPadded_:vz});async function kz(e,t,n,s=.5,r=Number.NEGATIVE_INFINITY,a=!1){let o=D(e,"boxes","nonMaxSuppressionAsync"),i=D(t,"scores","nonMaxSuppressionAsync"),l=Zc(o,i,n,s,r,null),u=l.maxOutputSize,c=l.iouThreshold,p=l.scoreThreshold,[d,h]=await Promise.all([o.data(),i.data()]),{selectedIndices:f,validOutputs:m}=Jw(d,h,u,c,p,a);return o!==e&&o.dispose(),i!==t&&i.dispose(),{selectedIndices:Ft(f,"int32"),validOutputs:Ce(m,"int32")}}var Iz=kz;function Sz(e,t,n=!1,s=!1){let r=D(e,"images","resizeBilinear");M(r.rank===3||r.rank===4,()=>`Error in resizeBilinear: x must be rank 3 or 4, but got rank ${r.rank}.`),M(t.length===2,()=>`Error in resizeBilinear: new shape must 2D, but got shape ${t}.`),M(s===!1||n===!1,()=>"Error in resizeBilinear: If halfPixelCenters is true, alignCorners must be false.");let a=r,o=!1;r.rank===3&&(o=!0,a=U(r,[1,r.shape[0],r.shape[1],r.shape[2]]));let[]=t,i={images:a},l={alignCorners:n,halfPixelCenters:s,size:t},u=W.runKernel(Xo,i,l);return o?U(u,[u.shape[1],u.shape[2],u.shape[3]]):u}var e8=V({resizeBilinear_:Sz});function Cz(e,t,n=!1,s=!1){let r=D(e,"images","resizeNearestNeighbor");M(r.rank===3||r.rank===4,()=>`Error in resizeNearestNeighbor: x must be rank 3 or 4, but got rank ${r.rank}.`),M(t.length===2,()=>`Error in resizeNearestNeighbor: new shape must 2D, but got shape ${t}.`),M(r.dtype==="float32"||r.dtype==="int32",()=>"`images` must have `int32` or `float32` as dtype"),M(s===!1||n===!1,()=>"Error in resizeNearestNeighbor: If halfPixelCenters is true, alignCorners must be false.");let a=r,o=!1;r.rank===3&&(o=!0,a=U(r,[1,r.shape[0],r.shape[1],r.shape[2]]));let[]=t,i={images:a},l={alignCorners:n,halfPixelCenters:s,size:t},u=W.runKernel(qo,i,l);return o?U(u,[u.shape[1],u.shape[2],u.shape[3]]):u}var t8=V({resizeNearestNeighbor_:Cz});function Tz(e,t="binary",n=!1,s=.5){let r=D(e,"image","threshold"),a=.2989,o=.587,i=.114,l=r.shape[0]*r.shape[1],u=L(Ft([s]),255),c,p,d,h;if(M(r.rank===3,()=>`Error in threshold: image must be rank 3,but got rank ${r.rank}.`),M(r.shape[2]===3||r.shape[2]===1,()=>`Error in threshold: image color channel must be equal to 3 or 1but got ${r.shape[2]}.`),M(r.dtype==="int32"||r.dtype==="float32",()=>`Error in dtype: image dtype must be int32 or float32,but got dtype ${r.dtype}.`),M(t==="otsu"||t==="binary",()=>`Method must be binary or otsu, but was ${t}`),r.shape[2]===3){[c,p,d]=Zt(r,[1,1,1],-1);let g=L(c,a),y=L(p,o),x=L(d,i);h=ce(ce(g,y),x)}else h=e;if(t==="otsu"){let g=aA(ye(R0(h),"int32"),ct([]),256);u=Nz(g,l)}let f=n?ui(h,u):ys(h,u);return ye(L(f,255),"int32")}function Nz(e,t){let n=Ft([-1]),s=Ft([0]),r=Ft([0]),a,o,i,l,u,c;for(let p=0;p<e.size-1;p++){a=Oe(e,0,p+1),o=Oe(e,p+1),u=he(ke(a),t),c=he(ke(o),t);let d=ke(L(a,Ju(0,a.size)));i=he(d,ke(a));let h=Hc(o.shape,a.size),f=ce(Ju(0,o.size),h),m=L(o,f);l=he(ke(m),ke(o));let g=fe(i,l),y=fe(i,l),x=L(u,c);r=L(L(x,g),y);let A=ys(r,s);s=zn(A,r,s),n=zn(A,Ft([p]),n)}return n}var Ez=V({threshold_:Tz});function Rz(e,t,n="nearest",s="constant",r=0,a){let o=D(e,"image","transform","float32"),i=D(t,"transforms","transform","float32");M(o.rank===4,()=>`Error in transform: image must be rank 4,but got rank ${o.rank}.`),M(i.rank===2&&(i.shape[0]===o.shape[0]||i.shape[0]===1)&&i.shape[1]===8,()=>"Error in transform: Input transform should be batch x 8 or 1 x 8"),M(a==null||a.length===2,()=>`Error in transform: outputShape must be [height, width] or null, but got ${a}.`);let l={image:o,transforms:i},u={interpolation:n,fillMode:s,fillValue:r,outputShape:a};return W.runKernel(Ul,l,u)}var _z=V({transform_:Rz});function Dz(e,t,n){M(t%1===0,()=>`bandPart(): numLower must be an integer, got ${t}.`),M(n%1===0,()=>`bandPart(): numUpper must be an integer, got ${n}.`);let s=D(e,"a","bandPart");M(s.rank>=2,()=>`bandPart(): Rank must be at least 2, got ${s.rank}.`);let r=s.shape,[a,o]=s.shape.slice(-2);if(!(t<=a))throw new Error(`bandPart(): numLower (${t}) must not be greater than the number of rows (${a}).`);if(!(n<=o))throw new Error(`bandPart(): numUpper (${n}) must not be greater than the number of columns (${o}).`);t<0&&(t=a),n<0&&(n=o);let i=U(Ju(0,a,1,"int32"),[-1,1]),l=Ju(0,o,1,"int32"),u=fe(i,l),c=ur(ui(u,Ce(+t,"int32")),li(u,Ce(-n,"int32"))),p=Wt([a,o],s.dtype);return U(on(En(U(s,[-1,a,o])).map(d=>zn(c,d,p))),r)}var $z=V({bandPart_:Dz});function Pz(e){let t;if(Array.isArray(e)){t=!1,M(e!=null&&e.length>0,()=>"Gram-Schmidt process: input must not be null, undefined, or empty");let r=e[0].shape[0];for(let a=1;a<e.length;++a)M(e[a].shape[0]===r,()=>`Gram-Schmidt: Non-unique lengths found in the input vectors: (${e[a].shape[0]} vs. ${r})`)}else t=!0,e=Zt(e,e.shape[0],0).map(r=>st(r,[0]));M(e.length<=e[0].shape[0],()=>`Gram-Schmidt: Number of vectors (${e.length}) exceeds number of dimensions (${e[0].shape[0]}).`);let n=[],s=e;for(let r=0;r<e.length;++r)n.push(W.tidy(()=>{let a=s[r];if(r>0)for(let o=0;o<r;++o){let i=L(ke(L(n[o],a)),n[o]);a=fe(a,i)}return he(a,Gc(a,"euclidean"))}));return t?on(n,0):n}var Fz=V({gramSchmidt_:Pz});function Oz(e,t=!1){if(M(e.rank>=2,()=>`qr() requires input tensor to have a rank >= 2, but got rank ${e.rank}`),e.rank===2)return Sv(e,t);{let n=e.shape.slice(0,e.shape.length-2).reduce((l,u)=>l*u),s=En(U(e,[n,e.shape[e.shape.length-2],e.shape[e.shape.length-1]]),0),r=[],a=[];s.forEach(l=>{let[u,c]=Sv(l,t);r.push(u),a.push(c)});let o=U(on(r,0),e.shape),i=U(on(a,0),e.shape);return[o,i]}}function Sv(e,t=!1){return W.tidy(()=>{M(e.shape.length===2,()=>`qr2d() requires a 2D Tensor, but got a ${e.shape.length}D Tensor.`);let n=e.shape[0],s=e.shape[1],r=v0(n),a=On(e),o=ir([[1]],[1,1]),i=On(o),l=n>=s?s:n;for(let u=0;u<l;++u){let c=a,p=i,d=r;[i,a,r]=W.tidy(()=>{let h=Oe(a,[u,u],[n-u,1]),f=Gc(h),m=Oe(a,[u,u],[1,1]),g=zn(ys(m,0),ir([[-1]]),ir([[1]])),y=fe(m,L(g,f)),x=he(h,y);x.shape[0]===1?i=On(o):i=Ct([o,Oe(x,[1,0],[x.shape[0]-1,x.shape[1]])],0);let A=Dt(he(Qe(g,y),f)),b=Oe(a,[u,0],[n-u,s]),w=L(A,i),k=et(i);if(u===0)a=fe(b,Qe(w,Qe(k,b)));else{let R=fe(b,Qe(w,Qe(k,b)));a=Ct([Oe(a,[0,0],[u,s]),R],0)}let S=et(w),E=Oe(r,[0,u],[n,r.shape[1]-u]);if(u===0)r=fe(E,Qe(Qe(E,i),S));else{let R=fe(E,Qe(Qe(E,i),S));r=Ct([Oe(r,[0,0],[n,u]),R],1)}return[i,a,r]}),ee([c,p,d])}return!t&&n>s&&(r=Oe(r,[0,0],[n,s]),a=Oe(a,[0,0],[s,s])),[r,a]})}var Mz=V({qr_:Oz}),Jn;(function(e){e[e.NONE=0]="NONE",e[e.MEAN=1]="MEAN",e[e.SUM=2]="SUM",e[e.SUM_BY_NONZERO_WEIGHTS=3]="SUM_BY_NONZERO_WEIGHTS"})(Jn||(Jn={}));function zz(e,t,n=Jn.SUM_BY_NONZERO_WEIGHTS){let s=D(e,"losses","computeWeightedLoss"),r=null;t!=null&&(r=D(t,"weights","computeWeightedLoss"));let a=r==null?s:L(s,r);if(n===Jn.NONE)return a;if(n===Jn.SUM)return ke(a);if(n===Jn.MEAN){if(r==null)return Bt(a);{let o=s.size/r.size,i=he(ke(a),ke(r));return o>1?he(i,Ce(o)):i}}if(n===Jn.SUM_BY_NONZERO_WEIGHTS){if(r==null)return he(ke(a),Ce(s.size));{let o=L(r,Ns(s.shape)),i=ye(ke(Qi(o,Ce(0))),"float32");return he(ke(a),i)}}throw Error(`Unknown reduction: ${n}`)}var wa=V({computeWeightedLoss_:zz});function Lz(e,t,n,s=Jn.SUM_BY_NONZERO_WEIGHTS){let r=D(e,"labels","absoluteDifference"),a=D(t,"predictions","absoluteDifference"),o=null;n!=null&&(o=D(n,"weights","absoluteDifference")),ns(r.shape,a.shape,"Error in absoluteDifference: ");let i=tn(fe(r,a));return wa(i,o,s)}var Bz=V({absoluteDifference_:Lz});function Wz(e,t,n,s,r=Jn.SUM_BY_NONZERO_WEIGHTS){let a=D(e,"labels","cosineDistance"),o=D(t,"predictions","cosineDistance"),i=null;s!=null&&(i=D(s,"weights","cosineDistance")),ns(a.shape,o.shape,"Error in cosineDistance: ");let l=Ce(1),u=fe(l,ke(L(a,o),n,!0));return wa(u,i,r)}var Vz=V({cosineDistance_:Wz});function Uz(e,t,n,s=Jn.SUM_BY_NONZERO_WEIGHTS){let r=D(e,"labels","hingeLoss"),a=D(t,"predictions","hingeLoss"),o=null;n!=null&&(o=D(n,"weights","hingeLoss")),ns(r.shape,a.shape,"Error in hingeLoss: ");let i=Ce(1);r=fe(L(Ce(2),r),i);let l=Pr(fe(i,L(r,a)));return wa(l,o,s)}var Gz=V({hingeLoss_:Uz});function Hz(e,t,n,s=1,r=Jn.SUM_BY_NONZERO_WEIGHTS){let a=D(e,"labels","huberLoss"),o=D(t,"predictions","huberLoss"),i=null;n!=null&&(i=D(n,"weights","huberLoss")),ns(a.shape,o.shape,"Error in huberLoss: ");let l=Ce(s),u=tn(fe(o,a)),c=Xc(u,l),p=fe(u,c),d=ce(L(Ce(.5),vt(c)),L(l,p));return wa(d,i,r)}var jz=V({huberLoss_:Hz});function qz(e,t,n,s=1e-7,r=Jn.SUM_BY_NONZERO_WEIGHTS){let a=D(e,"labels","logLoss"),o=D(t,"predictions","logLoss"),i=null;n!=null&&(i=D(n,"weights","logLoss")),ns(a.shape,o.shape,"Error in logLoss: ");let l=Ce(1),u=Ce(s),c=Dt(L(a,Ds(ce(o,u)))),p=L(fe(l,a),Ds(ce(fe(l,o),u))),d=fe(c,p);return wa(d,i,r)}var Xz=V({logLoss_:qz});function Kz(e,t,n,s=Jn.SUM_BY_NONZERO_WEIGHTS){let r=D(e,"labels","meanSquaredError"),a=D(t,"predictions","meanSquaredError"),o=null;n!=null&&(o=D(n,"weights","meanSquaredError")),ns(r.shape,a.shape,"Error in meanSquaredError: ");let i=z0(r,a);return wa(i,o,s)}var Zz=V({meanSquaredError_:Kz});function Yz(e,t){let n=D(e,"labels","sigmoidCrossEntropyWithLogits"),s=D(t,"logits","sigmoidCrossEntropyWithLogits");ns(n.shape,s.shape,"Error in sigmoidCrossEntropyWithLogits: ");let r=Pr(s),a=L(s,n),o=ph(_s(Dt(tn(s))));return ce(fe(r,a),o)}function Jz(e,t,n,s=0,r=Jn.SUM_BY_NONZERO_WEIGHTS){let a=D(e,"multiClassLabels","sigmoidCrossEntropy"),o=D(t,"logits","sigmoidCrossEntropy"),i=null;if(n!=null&&(i=D(n,"weights","sigmoidCrossEntropy")),ns(a.shape,o.shape,"Error in sigmoidCrossEntropy: "),s>0){let u=Ce(s),c=Ce(1),p=Ce(.5);a=ce(L(a,fe(c,u)),L(p,u))}let l=Yz(a,o);return wa(l,i,r)}var Qz=V({sigmoidCrossEntropy_:Jz});function eL(e,t,n=-1){if(n===-1&&(n=t.rank-1),n!==t.rank-1)throw Error(`Softmax cross entropy along a non-last dimension is not yet supported. Labels / logits was rank ${t.rank} and dim was ${n}`);return Kr((r,a,o)=>{let l=S0(a,[n],!0),u=fe(ye(a,"float32"),l);o([r,u]);let c=Dt(L(u,r));return{value:ke(c,[n]),gradFunc:(h,f)=>{let[m,g]=f,y=Ji(h.shape,[n]);return[L(U(h,y),fe(ye(m,"float32"),_s(g))),L(U(h,y),fe(_s(g),ye(m,"float32")))]}}})(e,t)}function tL(e,t,n,s=0,r=Jn.SUM_BY_NONZERO_WEIGHTS){let a=D(e,"onehotLabels","softmaxCrossEntropy"),o=D(t,"logits","softmaxCrossEntropy"),i=null;if(n!=null&&(i=D(n,"weights","softmaxCrossEntropy")),ns(a.shape,o.shape,"Error in softmaxCrossEntropy: "),s>0){let u=Ce(s),c=Ce(1),p=Ce(a.shape[1]);a=ce(L(a,fe(c,u)),he(u,p))}let l=eL(a,o);return wa(l,i,r)}var nL=V({softmaxCrossEntropy_:tL});function sL(e,t,n,s){let r=D(e,"indices","sparseFillEmptyRows","int32"),a=D(t,"values","sparseFillEmptyRows"),o=D(n,"denseShape","sparseFillEmptyRows","int32"),i=D(s,"defaultValue","sparseFillEmptyRows",a.dtype);if(r.rank!==2)throw new Error(`Indices should be Tensor2D but received shape
|
|
${r.shape}`);if(a.rank!==1)throw new Error(`Values should be Tensor1D but received shape ${a.shape}`);if(o.rank!==1)throw new Error(`Dense shape should be Tensor1D but received shape ${o.shape}`);if(i.rank!==0)throw new Error(`Default value should be a scalar but received shape ${i.shape}`);let l={indices:r,values:a,denseShape:o,defaultValue:i},u=W.runKernel(Xp,l);return{outputIndices:u[0],outputValues:u[1],emptyRowIndicator:u[2],reverseIndexMap:u[3]}}var rL=V({sparseFillEmptyRows_:sL});function aL(e,t,n){let s=D(e,"inputIndices","sparseReshape","int32"),r=D(t,"inputShape","sparseReshape","int32"),a=D(n,"newShape","sparseReshape","int32");if(s.rank!==2)throw new Error(`Input indices should be Tensor2D but received shape
|
|
${s.shape}`);if(r.rank!==1)throw new Error(`Input shape should be Tensor1D but received shape ${r.shape}`);if(a.rank!==1)throw new Error(`New shape should be Tensor1D but received shape ${a.shape}`);let o={inputIndices:s,inputShape:r,newShape:a},i=W.runKernel(Oc,o);return{outputIndices:i[0],outputShape:i[1]}}var oL=V({sparseReshape_:aL});function iL(e,t,n){let s=D(e,"data","sparseSegmentMean"),r=D(t,"indices","sparseSegmentMean","int32"),a=D(n,"segmentIds","sparseSegmentMean","int32");if(s.rank<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(r.rank!==1)throw new Error(`Indices should be Tensor1D but received shape
|
|
${r.shape}`);if(a.rank!==1)throw new Error(`Segment ids should be Tensor1D but received shape
|
|
${a.shape}`);let o={data:s,indices:r,segmentIds:a};return W.runKernel(Kp,o)}var lL=V({sparseSegmentMean_:iL});function uL(e,t,n){let s=D(e,"data","sparseSegmentSum"),r=D(t,"indices","sparseSegmentSum","int32"),a=D(n,"segmentIds","sparseSegmentSum","int32");if(s.rank<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(r.rank!==1)throw new Error(`Indices should be Tensor1D but received shape
|
|
${r.shape}`);if(a.rank!==1)throw new Error(`Segment ids should be Tensor1D but received shape
|
|
${a.shape}`);let o={data:s,indices:r,segmentIds:a};return W.runKernel(Zp,o)}var cL=V({sparseSegmentSum_:uL});function dL(e,t,n,s,r,a,o,i){let l=D(e,"data","stringNGrams","string");if(l.dtype!=="string")throw new Error("Data must be of datatype string");if(l.shape.length!==1)throw new Error(`Data must be a vector, saw: ${l.shape}`);let u=D(t,"dataSplits","stringNGrams");if(u.dtype!=="int32")throw new Error("Data splits must be of datatype int32");let c={separator:n,nGramWidths:s,leftPad:r,rightPad:a,padWidth:o,preserveShortSequences:i},p={data:l,dataSplits:u},d=W.runKernel(zc,p,c);return{nGrams:d[0],nGramsSplits:d[1]}}var pL=V({stringNGrams_:dL});function hL(e,t,n=!0){let s=D(e,"input","stringSplit","string"),r=D(t,"delimiter","stringSplit","string");if(s.rank!==1)throw new Error(`Input should be Tensor1D but received shape ${s.shape}`);if(r.rank!==0)throw new Error(`Delimiter should be a scalar but received shape ${r.shape}`);let a={skipEmpty:n},o={input:s,delimiter:r},i=W.runKernel(Jp,o,a);return{indices:i[0],values:i[1],shape:i[2]}}var fL=V({stringSplit_:hL});function mL(e,t){let n=D(e,"input","stringToHashBucketFast","string"),s={numBuckets:t};if(t<=0)throw new Error("Number of buckets must be at least 1");let r={input:n};return W.runKernel(Qp,r,s)}var gL=V({stringToHashBucketFast_:mL}),n8={fft:xh,ifft:Qu,rfft:bh,irfft:M0},s8={hammingWindow:XM,hannWindow:Kw,frame:Zw,stft:JM},Se={flipLeftRight:nz,grayscaleToRGB:rz,resizeNearestNeighbor:t8,resizeBilinear:e8,rotateWithOffset:oz,cropAndResize:ez,nonMaxSuppression:lz,nonMaxSuppressionAsync:gz,nonMaxSuppressionWithScore:Az,nonMaxSuppressionWithScoreAsync:bz,nonMaxSuppressionPadded:wz,nonMaxSuppressionPaddedAsync:Iz,threshold:Ez,transform:_z},jA={bandPart:$z,gramSchmidt:Fz,qr:Mz},r8={absoluteDifference:Bz,computeWeightedLoss:wa,cosineDistance:Vz,hingeLoss:Gz,huberLoss:jz,logLoss:Xz,meanSquaredError:Zz,sigmoidCrossEntropy:Qz,softmaxCrossEntropy:nL},a8={sparseFillEmptyRows:rL,sparseReshape:oL,sparseSegmentMean:lL,sparseSegmentSum:cL},o8={stringNGrams:pL,stringSplit:fL,stringToHashBucketFast:gL},ka=class extends Q6{minimize(e,t=!1,n){let{value:s,grads:r}=this.computeGradients(e,n);if(n!=null){let a=n.map(o=>({name:o.name,tensor:r[o.name]}));this.applyGradients(a)}else this.applyGradients(r);return ee(r),t?s:(s.dispose(),null)}get iterations(){return this.iterations_==null&&(this.iterations_=0),this.iterations_}incrementIterations(){this.iterations_=this.iterations+1}computeGradients(e,t){return yw(e,t)}dispose(){this.iterations_!=null&&ee(this.iterations_)}async saveIterations(){return this.iterations_==null&&(this.iterations_=0),{name:"iter",tensor:Ce(this.iterations_,"int32")}}async getWeights(){throw new Error("getWeights() is not implemented for this optimizer yet.")}async setWeights(e){throw new Error(`setWeights() is not implemented for this optimizer class ${this.getClassName()}`)}async extractIterations(e){return this.iterations_=(await e[0].tensor.data())[0],e.slice(1)}};Object.defineProperty(ka,Symbol.hasInstance,{value:e=>e.minimize!=null&&e.computeGradients!=null&&e.applyGradients!=null});var j0=class extends ka{constructor(e,t,n=null){super(),this.learningRate=e,this.rho=t,this.epsilon=n,this.accumulatedGrads=[],this.accumulatedUpdates=[],n==null&&(this.epsilon=W.backend.epsilon())}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,s)=>{let r=W.registeredVariables[n],a=!1;this.accumulatedGrads[s]==null&&(this.accumulatedGrads[s]={originalName:`${n}/accum_grad`,variable:Y(()=>it(r).variable(a))}),this.accumulatedUpdates[s]==null&&(this.accumulatedUpdates[s]={originalName:`${n}/accum_var`,variable:Y(()=>it(r).variable(a))});let o=Array.isArray(e)?e[s].tensor:e[n];if(o==null)return;let i=this.accumulatedGrads[s].variable,l=this.accumulatedUpdates[s].variable;Y(()=>{let u=ce(L(i,this.rho),L(vt(o),1-this.rho)),c=L(he(Nn(ce(l,this.epsilon)),Nn(ce(i,this.epsilon))),o),p=ce(L(l,this.rho),L(vt(c),1-this.rho));i.assign(u),l.assign(p);let d=ce(L(c,-this.learningRate),r);r.assign(d)})}),this.incrementIterations()}dispose(){this.accumulatedUpdates!=null&&(ee(this.accumulatedGrads.map(e=>e.variable)),ee(this.accumulatedUpdates.map(e=>e.variable)))}async getWeights(){let e=[...this.accumulatedGrads,...this.accumulatedUpdates];return[await this.saveIterations()].concat(e.map(t=>({name:t.originalName,tensor:t.variable})))}async setWeights(e){e=await this.extractIterations(e);let t=e.length/2,n=!1;this.accumulatedGrads=e.slice(0,t).map(s=>({originalName:s.name,variable:s.tensor.variable(n)})),this.accumulatedUpdates=e.slice(t,t*2).map(s=>({originalName:s.name,variable:s.tensor.variable(n)}))}getConfig(){return{learningRate:this.learningRate,rho:this.rho,epsilon:this.epsilon}}static fromConfig(e,t){return new e(t.learningRate,t.rho,t.epsilon)}};j0.className="Adadelta";ii(j0);var q0=class extends ka{constructor(e,t=.1){super(),this.learningRate=e,this.initialAccumulatorValue=t,this.accumulatedGrads=[]}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,s)=>{let r=W.registeredVariables[n];this.accumulatedGrads[s]==null&&(this.accumulatedGrads[s]={originalName:`${n}/accumulator`,variable:Y(()=>Hc(r.shape,this.initialAccumulatorValue).variable(!1))});let a=Array.isArray(e)?e[s].tensor:e[n];if(a==null)return;let o=this.accumulatedGrads[s].variable;Y(()=>{let i=ce(o,vt(a));o.assign(i);let l=ce(L(he(a,Nn(ce(i,W.backend.epsilon()))),-this.learningRate),r);r.assign(l)})}),this.incrementIterations()}dispose(){this.accumulatedGrads!=null&&ee(this.accumulatedGrads.map(e=>e.variable))}async getWeights(){return[await this.saveIterations()].concat(this.accumulatedGrads.map(e=>({name:e.originalName,tensor:e.variable})))}async setWeights(e){e=await this.extractIterations(e);let t=!1;this.accumulatedGrads=e.map(n=>({originalName:n.name,variable:n.tensor.variable(t)}))}getConfig(){return{learningRate:this.learningRate,initialAccumulatorValue:this.initialAccumulatorValue}}static fromConfig(e,t){return new e(t.learningRate,t.initialAccumulatorValue)}};q0.className="Adagrad";ii(q0);var X0=class extends ka{constructor(e,t,n,s=null){super(),this.learningRate=e,this.beta1=t,this.beta2=n,this.epsilon=s,this.accumulatedFirstMoment=[],this.accumulatedSecondMoment=[],Y(()=>{this.accBeta1=Ce(t).variable(),this.accBeta2=Ce(n).variable()}),s==null&&(this.epsilon=W.backend.epsilon())}applyGradients(e){let t=Array.isArray(e)?e.map(n=>n.name):Object.keys(e);Y(()=>{let n=fe(1,this.accBeta1),s=fe(1,this.accBeta2);t.forEach((r,a)=>{let o=W.registeredVariables[r],i=!1;this.accumulatedFirstMoment[a]==null&&(this.accumulatedFirstMoment[a]={originalName:`${r}/m`,variable:Y(()=>it(o).variable(i))}),this.accumulatedSecondMoment[a]==null&&(this.accumulatedSecondMoment[a]={originalName:`${r}/v`,variable:Y(()=>it(o).variable(i))});let l=Array.isArray(e)?e[a].tensor:e[r];if(l==null)return;let u=this.accumulatedFirstMoment[a].variable,c=this.accumulatedSecondMoment[a].variable,p=ce(L(u,this.beta1),L(l,1-this.beta1)),d=ce(L(c,this.beta2),L(vt(l),1-this.beta2)),h=he(p,n),f=he(d,s);u.assign(p),c.assign(d);let m=ce(L(he(h,ce(Nn(f),this.epsilon)),-this.learningRate),o);o.assign(m)}),this.accBeta1.assign(L(this.accBeta1,this.beta1)),this.accBeta2.assign(L(this.accBeta2,this.beta2))}),this.incrementIterations()}dispose(){this.accBeta1.dispose(),this.accBeta2.dispose(),this.accumulatedFirstMoment!=null&&ee(this.accumulatedFirstMoment.map(e=>e.variable)),this.accumulatedSecondMoment!=null&&ee(this.accumulatedSecondMoment.map(e=>e.variable))}async getWeights(){let e=[...this.accumulatedFirstMoment,...this.accumulatedSecondMoment];return[await this.saveIterations()].concat(e.map(t=>({name:t.originalName,tensor:t.variable})))}async setWeights(e){e=await this.extractIterations(e),Y(()=>{this.accBeta1.assign(ya(this.beta1,this.iterations_+1)),this.accBeta2.assign(ya(this.beta2,this.iterations_+1))});let t=e.length/2,n=!1;this.accumulatedFirstMoment=e.slice(0,t).map(s=>({originalName:s.name,variable:s.tensor.variable(n)})),this.accumulatedSecondMoment=e.slice(t,t*2).map(s=>({originalName:s.name,variable:s.tensor.variable(n)}))}getConfig(){return{learningRate:this.learningRate,beta1:this.beta1,beta2:this.beta2,epsilon:this.epsilon}}static fromConfig(e,t){return new e(t.learningRate,t.beta1,t.beta2,t.epsilon)}};X0.className="Adam";ii(X0);var K0=class extends ka{constructor(e,t,n,s=null,r=0){super(),this.learningRate=e,this.beta1=t,this.beta2=n,this.epsilon=s,this.decay=r,this.accumulatedFirstMoment=[],this.accumulatedWeightedInfNorm=[],Y(()=>{this.iteration=Ce(0).variable(),this.accBeta1=Ce(t).variable()}),s==null&&(this.epsilon=W.backend.epsilon())}applyGradients(e){let t=Array.isArray(e)?e.map(n=>n.name):Object.keys(e);Y(()=>{let n=fe(1,this.accBeta1),s=he(-this.learningRate,ce(L(this.iteration,this.decay),1));t.forEach((r,a)=>{let o=W.registeredVariables[r],i=!1;this.accumulatedFirstMoment[a]==null&&(this.accumulatedFirstMoment[a]={originalName:`${r}/m`,variable:it(o).variable(i)}),this.accumulatedWeightedInfNorm[a]==null&&(this.accumulatedWeightedInfNorm[a]={originalName:`${r}/v`,variable:it(o).variable(i)});let l=Array.isArray(e)?e[a].tensor:e[r];if(l==null)return;let u=this.accumulatedFirstMoment[a].variable,c=this.accumulatedWeightedInfNorm[a].variable,p=ce(L(u,this.beta1),L(l,1-this.beta1)),d=L(c,this.beta2),h=tn(l),f=Jr(d,h);u.assign(p),c.assign(f);let m=ce(L(he(s,n),he(p,ce(f,this.epsilon))),o);o.assign(m)}),this.iteration.assign(ce(this.iteration,1)),this.accBeta1.assign(L(this.accBeta1,this.beta1))}),this.incrementIterations()}dispose(){this.accBeta1.dispose(),this.iteration.dispose(),this.accumulatedFirstMoment!=null&&ee(this.accumulatedFirstMoment.map(e=>e.variable)),this.accumulatedWeightedInfNorm!=null&&ee(this.accumulatedWeightedInfNorm.map(e=>e.variable))}async getWeights(){throw new Error("getWeights() is not implemented for Adamax yet.")}async setWeights(e){throw new Error("setWeights() is not implemented for Adamax yet.")}getConfig(){return{learningRate:this.learningRate,beta1:this.beta1,beta2:this.beta2,epsilon:this.epsilon,decay:this.decay}}static fromConfig(e,t){return new e(t.learningRate,t.beta1,t.beta2,t.epsilon,t.decay)}};K0.className="Adamax";ii(K0);var vh=class extends ka{constructor(e){super(),this.learningRate=e,this.setLearningRate(e)}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,s)=>{let r=Array.isArray(e)?e[s].tensor:e[n];if(r==null)return;let a=W.registeredVariables[n];Y(()=>{let o=ce(L(this.c,r),a);a.assign(o)})}),this.incrementIterations()}setLearningRate(e){this.learningRate=e,this.c!=null&&this.c.dispose(),this.c=bn(Ce(-e))}dispose(){this.c.dispose()}async getWeights(){return[await this.saveIterations()]}async setWeights(e){if(e=await this.extractIterations(e),e.length!==0)throw new Error("SGD optimizer does not have settable weights.")}getConfig(){return{learningRate:this.learningRate}}static fromConfig(e,t){return new e(t.learningRate)}};vh.className="SGD";ii(vh);var Z0=class extends vh{constructor(e,t,n=!1){super(e),this.learningRate=e,this.momentum=t,this.useNesterov=n,this.accumulations=[],this.m=Ce(this.momentum)}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,s)=>{let r=W.registeredVariables[n];this.accumulations[s]==null&&(this.accumulations[s]={originalName:`${n}/momentum`,variable:Y(()=>it(r).variable(!1))});let a=this.accumulations[s].variable,o=Array.isArray(e)?e[s].tensor:e[n];o!=null&&Y(()=>{let i,l=ce(L(this.m,a),o);this.useNesterov?i=ce(L(this.c,ce(o,L(l,this.m))),r):i=ce(L(this.c,l),r),a.assign(l),r.assign(i)})}),this.incrementIterations()}dispose(){this.m.dispose(),this.accumulations!=null&&ee(this.accumulations.map(e=>e.variable))}setMomentum(e){this.momentum=e}async getWeights(){return[await this.saveIterations()].concat(this.accumulations.map(e=>({name:e.originalName,tensor:e.variable})))}async setWeights(e){e=await this.extractIterations(e);let t=!1;this.accumulations=e.map(n=>({originalName:n.name,variable:n.tensor.variable(t)}))}getConfig(){return{learningRate:this.learningRate,momentum:this.momentum,useNesterov:this.useNesterov}}static fromConfig(e,t){return new e(t.learningRate,t.momentum,t.useNesterov)}};Z0.className="Momentum";ii(Z0);var Y0=class extends ka{constructor(e,t=.9,n=0,s=null,r=!1){if(super(),this.learningRate=e,this.decay=t,this.momentum=n,this.epsilon=s,this.accumulatedMeanSquares=[],this.accumulatedMoments=[],this.accumulatedMeanGrads=[],this.centered=r,s==null&&(this.epsilon=W.backend.epsilon()),e==null)throw new Error("learningRate for RMSPropOptimizer must be defined.")}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,s)=>{let r=W.registeredVariables[n],a=!1;this.accumulatedMeanSquares[s]==null&&(this.accumulatedMeanSquares[s]={originalName:`${n}/rms`,variable:Y(()=>it(r).variable(a))}),this.accumulatedMoments[s]==null&&(this.accumulatedMoments[s]={originalName:`${n}/momentum`,variable:Y(()=>it(r).variable(a))}),this.accumulatedMeanGrads[s]==null&&this.centered&&(this.accumulatedMeanGrads[s]={originalName:`${n}/mg`,variable:Y(()=>it(r).variable(a))});let o=Array.isArray(e)?e[s].tensor:e[n];if(o==null)return;let i=this.accumulatedMeanSquares[s].variable,l=this.accumulatedMoments[s].variable;Y(()=>{let u=ce(L(i,this.decay),L(vt(o),1-this.decay));if(this.centered){let c=this.accumulatedMeanGrads[s].variable,p=ce(L(c,this.decay),L(o,1-this.decay)),d=he(L(o,this.learningRate),Nn(fe(u,ce(vt(p),this.epsilon)))),h=ce(L(l,this.momentum),d);i.assign(u),c.assign(p),l.assign(h);let f=fe(r,h);r.assign(f)}else{let c=ce(L(i,this.decay),L(vt(o),1-this.decay)),p=ce(L(l,this.momentum),he(L(o,this.learningRate),Nn(ce(c,this.epsilon))));i.assign(c),l.assign(p);let d=fe(r,p);r.assign(d)}})}),this.incrementIterations()}dispose(){this.accumulatedMeanSquares!=null&&ee(this.accumulatedMeanSquares.map(e=>e.variable)),this.accumulatedMeanGrads!=null&&this.centered&&ee(this.accumulatedMeanGrads.map(e=>e.variable)),this.accumulatedMoments!=null&&ee(this.accumulatedMoments.map(e=>e.variable))}async getWeights(){let e=[...this.accumulatedMeanSquares,...this.accumulatedMoments];return this.centered&&e.push(...this.accumulatedMeanGrads),[await this.saveIterations()].concat(e.map(t=>({name:t.originalName,tensor:t.variable})))}async setWeights(e){e=await this.extractIterations(e);let t=this.centered?e.length/3:e.length/2,n=!1;this.accumulatedMeanSquares=e.slice(0,t).map(s=>({originalName:s.name,variable:s.tensor.variable(n)})),this.accumulatedMoments=e.slice(t,t*2).map(s=>({originalName:s.name,variable:s.tensor.variable(n)})),this.centered&&(this.accumulatedMeanGrads=e.slice(t*2,t*3).map(s=>({originalName:s.name,variable:s.tensor.variable(n)})))}getConfig(){return{learningRate:this.learningRate,decay:this.decay,momentum:this.momentum,epsilon:this.epsilon,centered:this.centered}}static fromConfig(e,t){return new e(t.learningRate,t.decay,t.momentum,t.epsilon,t.centered)}};Y0.className="RMSProp";ii(Y0);var Wa=class{static sgd(e){return new vh(e)}static momentum(e,t,n=!1){return new Z0(e,t,n)}static rmsprop(e,t=.9,n=0,s=null,r=!1){return new Y0(e,t,n,s,r)}static adam(e=.001,t=.9,n=.999,s=null){return new X0(e,t,n,s)}static adadelta(e=.001,t=.95,n=null){return new j0(e,t,n)}static adamax(e=.002,t=.9,n=.999,s=null,r=0){return new K0(e,t,n,s,r)}static adagrad(e,t=.1){return new q0(e,t)}},Fi={sgd:Wa.sgd,momentum:Wa.momentum,adadelta:Wa.adadelta,adagrad:Wa.adagrad,rmsprop:Wa.rmsprop,adamax:Wa.adamax,adam:Wa.adam},yL=(()=>typeof requestAnimationFrame!="undefined"?requestAnimationFrame:typeof setImmediate!="undefined"?setImmediate:e=>e())();function qA(){return new Promise(e=>yL(()=>e()))}var T={};Ve(T,{ERF_A1:()=>NL,ERF_A2:()=>EL,ERF_A3:()=>RL,ERF_A4:()=>_L,ERF_A5:()=>DL,ERF_P:()=>TL,PARALLELIZE_THRESHOLD:()=>XA,SELU_SCALE:()=>l8,SELU_SCALEALPHA:()=>i8,applyActivation:()=>G0,assertAndGetBroadcastShape:()=>kt,assertAxesAreInnerMostDims:()=>SF,assertParamsConsistent:()=>AL,assignToTypedArray:()=>zL,axesAreInnerMostDims:()=>AA,calculateShapes:()=>U6,checkEinsumDimSizes:()=>GL,checkPadOnDimRoundingMode:()=>rs,combineLocations:()=>pw,complexWithEvenIndex:()=>FL,complexWithOddIndex:()=>OL,computeConv2DInfo:()=>ih,computeConv3DInfo:()=>rw,computeDefaultPad:()=>eA,computeDilation2DInfo:()=>CP,computeOptimalWindowSize:()=>bL,computeOutAndReduceShapes:()=>hw,computeOutShape:()=>xL,computePool2DInfo:()=>sw,computePool3DInfo:()=>TP,convertConv2DDataFormat:()=>aw,decodeEinsumEquation:()=>VL,eitherStridesOrDilationsAreOne:()=>Yr,expandShapeToKeepDim:()=>Ji,exponent:()=>BL,exponents:()=>LL,fromStringArrayToUint8:()=>dB,fromUint8ToStringArray:()=>cB,getAxesPermutation:()=>fw,getBroadcastDims:()=>B6,getComplexWithIndex:()=>ML,getEinsumComputePath:()=>HL,getEinsumPermutation:()=>UL,getFusedBiasGradient:()=>U0,getFusedDyActivation:()=>V0,getImageCenter:()=>vL,getInnerMostAxes:()=>CF,getPermuted:()=>kL,getReductionAxes:()=>an,getReshaped:()=>wL,getReshapedPermuted:()=>IL,getSliceBeginCoords:()=>SL,getSliceSize:()=>CL,getSparseFillEmptyRowsIndicesDenseShapeMismatch:()=>KL,getSparseFillEmptyRowsNegativeIndexErrorMessage:()=>ZL,getSparseFillEmptyRowsOutOfRangeIndexErrorMessage:()=>YL,getSparseReshapeEmptyTensorZeroOutputDimErrorMessage:()=>eB,getSparseReshapeInputOutputMismatchErrorMessage:()=>nB,getSparseReshapeInputOutputMultipleErrorMessage:()=>tB,getSparseReshapeMultipleNegativeOneOutputDimErrorMessage:()=>JL,getSparseReshapeNegativeOutputDimErrorMessage:()=>QL,getSparseSegmentReductionIndicesOutOfRangeErrorMessage:()=>oB,getSparseSegmentReductionNegativeSegmentIdsErrorMessage:()=>sB,getSparseSegmentReductionNonIncreasingSegmentIdsErrorMessage:()=>rB,getSparseSegmentReductionSegmentIdOutOfRangeErrorMessage:()=>aB,getUndoAxesPermutation:()=>xA,isIdentityPermutation:()=>jL,log:()=>Y_,mergeRealAndImagArrays:()=>$L,prepareAndValidate:()=>V6,prepareSplitSize:()=>XL,segment_util:()=>u8,shouldFuse:()=>H0,slice_util:()=>Ut,splitRealAndImagArrays:()=>PL,tupleValuesAreOne:()=>to,upcastType:()=>Mn,validateInput:()=>Vy,validateUpdateShape:()=>Wy,warn:()=>Ua});function AL(e,t){let n=e[0].length;e.forEach((r,a)=>{M(r.length===n,()=>`Error in concat${n}D: rank of tensors[${a}] must be the same as the rank of the rest (${n})`)}),M(t>=0&&t<n,()=>`Error in concat${n}D: axis must be between 0 and ${n-1}.`);let s=e[0];e.forEach((r,a)=>{for(let o=0;o<n;o++)M(o===t||r[o]===s[o],()=>`Error in concat${n}D: Shape of tensors[${a}] (${r}) does not match the shape of the rest (${s}) along the non-concatenated axis ${a}.`)})}function xL(e,t){let n=e[0].slice();for(let s=1;s<e.length;s++)n[t]+=e[s][t];return n}var XA=30;function bL(e){return e<=XA?e:dm(e,Math.floor(Math.sqrt(e)))}function vL(e,t,n){let s=n*(typeof e=="number"?e:e[0]),r=t*(typeof e=="number"?e:e[1]);return[s,r]}function wL(e,t,n,s=!0){let r=[];if(s)r=r.concat(t.slice(0)),r.push(e[0]/n),r=r.concat(e.slice(1));else{r=r.concat(e[0]);let a=t.length;for(let o=0;o<a;++o)r=r.concat([e[o+1]/t[o],t[o]]);r=r.concat(e.slice(a+1))}return r}function kL(e,t,n=!0){let s=[];if(n){s.push(t);for(let r=t+1;r<e;++r)r<=2*t?(s.push(r),s.push(r-(t+1))):s.push(r)}else{let r=[],a=[];for(let o=1;o<e;++o)o>=t*2+1||o%2===1?a.push(o):r.push(o);s.push(...r),s.push(0),s.push(...a)}return s}function IL(e,t,n,s=!0){let r=[];s?r.push(e[0]/n):r.push(e[0]*n);for(let a=1;a<e.length;++a)a<=t.length?s?r.push(t[a-1]*e[a]):r.push(e[a]/t[a-1]):r.push(e[a]);return r}function SL(e,t){let n=[0];for(let s=0;s<t;++s)n.push(e[s][0]);return n}function CL(e,t,n){let s=e.slice(0,1);for(let r=0;r<n;++r)s.push(e[r+1]-t[r][0]-t[r][1]);return s}var i8=1.7580993408473768,l8=1.0507009873554805,TL=.3275911,NL=.254829592,EL=-.284496736,RL=1.421413741,_L=-1.453152027,DL=1.061405429;function $L(e,t){if(e.length!==t.length)throw new Error(`Cannot merge real and imag arrays of different lengths. real:${e.length}, imag: ${t.length}.`);let n=new Float32Array(e.length*2);for(let s=0;s<n.length;s+=2)n[s]=e[s/2],n[s+1]=t[s/2];return n}function PL(e){let t=new Float32Array(e.length/2),n=new Float32Array(e.length/2);for(let s=0;s<e.length;s+=2)t[s/2]=e[s],n[s/2]=e[s+1];return{real:t,imag:n}}function FL(e){let t=Math.ceil(e.length/4),n=new Float32Array(t),s=new Float32Array(t);for(let r=0;r<e.length;r+=4)n[Math.floor(r/4)]=e[r],s[Math.floor(r/4)]=e[r+1];return{real:n,imag:s}}function OL(e){let t=Math.floor(e.length/4),n=new Float32Array(t),s=new Float32Array(t);for(let r=2;r<e.length;r+=4)n[Math.floor(r/4)]=e[r],s[Math.floor(r/4)]=e[r+1];return{real:n,imag:s}}function ML(e,t){let n=e[t*2],s=e[t*2+1];return{real:n,imag:s}}function zL(e,t,n,s){e[s*2]=t,e[s*2+1]=n}function LL(e,t){let n=new Float32Array(e/2),s=new Float32Array(e/2);for(let r=0;r<Math.ceil(e/2);r++){let a=(t?2:-2)*Math.PI*(r/e);n[r]=Math.cos(a),s[r]=Math.sin(a)}return{real:n,imag:s}}function BL(e,t,n){let s=(n?2:-2)*Math.PI*(e/t),r=Math.cos(s),a=Math.sin(s);return{real:r,imag:a}}var i3="->",WL=/->/g,Cv=",",Tv="...";function VL(e,t){e=e.replace(/\s/g,"");let n=(e.length-e.replace(WL,"").length)/i3.length;if(n<1)throw new Error("Equations without an arrow are not supported.");if(n>1)throw new Error(`Equation must contain exactly one arrow ("${i3}").`);let[s,r]=e.split(i3);M(s.indexOf(Tv)===-1,()=>`The ellipsis notation ("${Tv}") is not supported yet.`);let a=s.split(Cv),o=a.length;if(t!==o)throw new Error(`Expected ${o} input tensors, received ${t}`);if(o>2)throw new Error("Support for more than 2 input tensors is not implemented yet.");let i=[];for(let d=0;d<r.length;++d){let h=r[d];if(!a.some(f=>f.indexOf(h)!==-1))throw new Error(`Output subscripts contain the label ${h} not present in the input subscripts.`);i.indexOf(h)===-1&&i.push(h)}for(let d=0;d<s.length;++d){let h=s[d];i.indexOf(h)===-1&&h!==Cv&&i.push(h)}let l=new Array(a.length);for(let d=0;d<o;++d){if(new Set(a[d].split("")).size!==a[d].length)throw new Error(`Found duplicate axes in input component ${a[d]}. Support for duplicate axes in input is not implemented yet.`);l[d]=[];for(let h=0;h<a[d].length;++h)l[d].push(i.indexOf(a[d][h]))}let u=i.length,c=r.length,p=[];for(let d=c;d<u;++d)p.push(d);return{allDims:i,summedDims:p,idDims:l}}function UL(e,t){let n=new Array(e);n.fill(-1);for(let r=0;r<t.length;++r)n[t[r]]=r;let s=[];for(let r=0;r<e;++r)n[r]===-1&&s.push(r);return n=n.filter(r=>r!==-1),{permutationIndices:n,expandDims:s}}function GL(e,t,n){let s=new Array(e);for(let r=0;r<n.length;++r){let a=n[r].shape;for(let o=0;o<t[r].length;++o)s[t[r][o]]===void 0?s[t[r][o]]=a[o]:M(s[t[r][o]]===a[o],()=>`Expected dimension ${s[t[r][o]]} at axis ${o} of input shaped ${JSON.stringify(a)}, but got dimension ${a[o]}`)}}function HL(e,t){let n=e,s=[],r=0;e.length===0&&n.push(-1),r=e.length+1;for(let o=0;o<r;++o)s.push([]);let a=[];for(let o=0;o<n.length;++o){let i=n[o],l=qL(t,i);for(let u of l)a.indexOf(u)===-1&&(s[o].push(u),a.push(u))}return{path:n,steps:s}}function jL(e){return e.every((t,n)=>t===n)}function qL(e,t){let n=[];for(let s=0;s<e.length;++s)(e[s].length===0||e[s].indexOf(t)!==-1||t===-1)&&n.push(s);return n}function XL(e,t,n=0){let s=[];if(typeof t=="number")M(e.shape[n]%t===0,()=>"Number of splits must evenly divide the axis."),s=new Array(t).fill(e.shape[n]/t);else{let r=t.reduce((o,i)=>(i===-1&&(o+=1),o),0);M(r<=1,()=>"There should be only one negative value in split array.");let a=t.indexOf(-1);if(a!==-1){let o=t.reduce((i,l)=>l>0?i+l:i);t[a]=e.shape[n]-o}M(e.shape[n]===t.reduce((o,i)=>o+i),()=>"The sum of sizes must match the size of the axis dimension."),s=t}return s}function KL(e){return`Received SparseTensor with denseShape[0] = 0 but
|
|
indices.shape[0] = ${e}`}function ZL(e,t){return`indices(${e}, 0) is invalid: ${t} < 0`}function YL(e,t,n){return`indices(${e}, 0) is invalid: ${t} >= ${n}`}function JL(e,t){return`only one output dimension may be -1, not both ${e} and ${t}`}function QL(e,t){return`size ${e} must be non-negative, not ${t}`}function eB(){return"reshape cannot infer the missing input size for an empty tensor unless all specified input sizes are non-zero"}function tB(e,t){let n=Et(e),s=Et(t);return`Input to reshape is a SparseTensor with ${n}
|
|
dense values, but the requested shape requires a multiple of ${s}. inputShape=${e} outputShape= ${t}`}function nB(e,t){let n=Et(e),s=Et(t);return`Input to reshape is a tensor with ${n} dense values, but the requested shape has ${s}. inputShape=${e} outputShape=${t}`}function sB(){return"segment ids must be >= 0"}function rB(){return"segment ids are not increasing"}function aB(e,t){return`Segment id ${e} out of range [0, ${t}), possibly because segmentIds input is not sorted.`}function oB(e,t,n){return`Bad: indices[${e}] == ${t} out of range [0, ${n})`}var u8={};Ve(u8,{collectGatherOpShapeInfo:()=>uB,computeOutShape:()=>lB,segOpComputeOptimalWindowSize:()=>iB});function iB(e,t){let n=!1,s;for(e<=XA?(s=e,n=!0):s=dm(e,Math.floor(Math.sqrt(e)));!n;)s>t||s===e?n=!0:s=dm(e,s+1);return s}function lB(e,t,n){let s=[],r=e.length;for(let a=0;a<r;a++)a!==t?s.push(e[a]):s.push(n);return s}function uB(e,t,n,s){let r=t.shape.length,a=e.shape.length;if(s!==0&&(s<-r||s>r))throw new Error(`Expect batchDims in the range of [-${r}, ${r}], but got ${s}`);if(s<0&&(s+=r),s>a)throw new Error(`batchDims (${s}) must be less than rank(x) (
|
|
${a}).`);if(n<s)throw new Error(`batchDims (${s}) must be less than or equal to axis (${n}).`);for(let p=0;p<s;++p)if(e.shape[p]!==t.shape[p])throw new Error(`x.shape[${p}]: ${e.shape[p]} should be equal to indices.shape[${p}]: ${t.shape[p]}.`);let o=e.shape[n],i=[],l=1,u=1,c=1;for(let p=0;p<s;++p)i.push(e.shape[p]),l*=e.shape[p];for(let p=s;p<n;p++)i.push(e.shape[p]),u*=e.shape[p];for(let p=s;p<r;p++)i.push(t.shape[p]);for(let p=n+1;p<a;p++)i.push(e.shape[p]),c*=e.shape[p];return{batchSize:l,sliceSize:c,outerSize:u,dimSize:o,outputShape:i}}function cB(e){try{return e.map(t=>mm(t))}catch(t){throw new Error(`Failed to decode encoded string bytes into utf-8, error: ${t}`)}}function dB(e){return e.map(t=>nh(t))}var pr={};Ve(pr,{nonMaxSuppressionV3Impl:()=>Yw,nonMaxSuppressionV4Impl:()=>Jw,nonMaxSuppressionV5Impl:()=>Qw,whereImpl:()=>Bw});var c8={kernelName:il,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(e,Ql(ye(n,"float32"),-1))}}},pB={kernelName:pc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>{let s=vt(ye(n,"float32")),r=Nn(fe(Ce(1),s));return Dt(he(e,r))}}}},hB={kernelName:hc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>{let s=Nn(fe(vt(ye(n,"float32")),1));return he(e,s)}}}},fB={kernelName:xa,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t,r=kt(n.shape,s.shape);return{a:()=>{let i=e,l=an(n.shape,r);return l.length>0&&(i=ke(i,l)),U(i,n.shape)},b:()=>{let i=e,l=an(s.shape,r);return l.length>0&&(i=ke(i,l)),U(i,s.shape)}}}},mB={kernelName:po,saveAllInputs:!0,gradFunc:(e,t)=>{let n={};return t.forEach((s,r)=>{n[r]=()=>e.clone()}),n}},gB={kernelName:ho,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>it(n)}}},yB={kernelName:gc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>it(n)}}},AB={kernelName:yc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>he(e,Nn(fe(Ce(1),vt(ye(n,"float32")))))}}},xB={kernelName:Ac,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>{let s=Nn(ce(Ce(1),vt(ye(n,"float32"))));return he(e,s)}}}},bB={kernelName:vc,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t,r=kt(n.shape,s.shape);return{a:()=>{let i=ce(vt(n),vt(s)),l=L(e,he(s,i)),u=an(n.shape,r);return u.length>0&&(l=ke(l,u)),U(l,n.shape)},b:()=>{let i=ce(vt(n),vt(s)),l=Dt(L(e,he(n,i))),u=an(s.shape,r);return u.length>0&&(l=ke(l,u)),U(l,s.shape)}}}},vB={kernelName:xc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>he(e,ce(vt(ye(n,"float32")),1))}}},wB={kernelName:bc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>he(e,fe(Ce(1),vt(ye(n,"float32"))))}}};function kB(e,t,n,s,r,a){let o=D(e,"dy","avgPool3dGrad"),i=D(t,"input","avgPool3dGrad"),l=o,u=i,c=!1;i.rank===4&&(c=!0,l=U(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]]),u=U(i,[1,i.shape[0],i.shape[1],i.shape[2],i.shape[3]])),M(l.rank===5,()=>`Error in avgPool3dGrad: dy must be rank 5 but got rank ${l.rank}.`),M(u.rank===5,()=>`Error in avgPool3dGrad: input must be rank 5 but got rank ${u.rank}.`),rs("avgPool3dGrad",r,a);let p={dy:l,input:u},d={filterSize:n,strides:s,pad:r,dimRoundingMode:a},h=W.runKernel(Hm,p,d);return c?U(h,[h.shape[1],h.shape[2],h.shape[3],h.shape[4]]):h}var IB=V({avgPool3dGrad_:kB}),SB={kernelName:zp,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{filterSize:r,strides:a,pad:o,dimRoundingMode:i}=n;return{x:()=>IB(e,s,r,a,o,i)}}};function CB(e,t,n,s,r){let a=D(e,"dy","avgPoolGrad"),o=D(t,"input","avgPoolGrad");M(o.rank===a.rank,()=>`Rank of input (${o.rank}) does not match rank of dy (${a.rank})`);let i=o,l=a,u=!1;o.rank===3&&(u=!0,i=U(o,[1,o.shape[0],o.shape[1],o.shape[2]]),l=U(a,[1,a.shape[0],a.shape[1],a.shape[2]])),M(l.rank===4,()=>`Error in avgPoolGrad: dy must be rank 4 but got rank ${l.rank}.`),M(i.rank===4,()=>`Error in avgPoolGrad: input must be rank 4 but got rank ${i.rank}.`);let c={dy:l,input:i},p={filterSize:n,strides:s,pad:r},d=W.runKernel(Gm,c,p);return u?U(d,[d.shape[1],d.shape[2],d.shape[3]]):d}var TB=V({avgPoolGrad_:CB}),NB={kernelName:fo,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{filterSize:r,strides:a,pad:o}=n;return{x:()=>TB(e,s,r,a,o)}}},EB={kernelName:mo,inputsToSave:["a","b"],gradFunc:(e,t,n)=>{let[s,r]=t,{transposeA:a,transposeB:o}=n;return!a&&!o?{a:()=>Qe(e,r,!1,!0),b:()=>Qe(s,e,!0,!1)}:!a&&o?{a:()=>Qe(e,r,!1,!1),b:()=>Qe(e,s,!0,!1)}:a&&!o?{a:()=>Qe(r,e,!1,!0),b:()=>Qe(s,e,!1,!1)}:{a:()=>Qe(r,e,!0,!0),b:()=>Qe(e,s,!0,!0)}}},RB={kernelName:ll,gradFunc:(e,t,n)=>{let{blockShape:s,crops:r}=n;return{x:()=>gh(e,s,r)}}},_B={kernelName:h6,gradFunc:(e,t,n)=>{let s=n,r=s.inputShape,a=s.shape,o=Array.from(a);for(let l=r.length-1;l>=0;l--)if(r[l]===a[l])o[l]=1;else if(r[l]!==1)throw new Error(`broadcastTo(): [${r}] cannot be broadcast to [${a}].`);let i=[];for(let l=0;l<o.length;l++)o[l]>1&&i.push(l);return{x:()=>ke(e,i,!0)}}},DB={kernelName:go,gradFunc:e=>({x:()=>e.clone()})},$B={kernelName:yo,gradFunc:e=>({x:()=>it(e)})},PB={kernelName:ba,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{clipValueMin:r,clipValueMax:a}=n;return{x:()=>zn(ur(li(s,r),ui(s,a)),e,it(e))}}},FB={kernelName:Bp,inputsToSave:["x"],gradFunc:c8.gradFunc},OB={kernelName:ul,saveAllInputs:!0,gradFunc:(e,t,n)=>{let s=t.map(l=>l.shape),{axis:r}=n,a=cr(r,t[0].shape)[0],o=s.map(l=>l[a]);return Zt(e,o,a).map(l=>()=>l)}},MB={kernelName:Ao,inputsToSave:["x","filter"],gradFunc:(e,t,n)=>{let[s,r]=t,{dilations:a,strides:o,pad:i,dataFormat:l}=n;return M(to(a),()=>`Error in gradient of conv2D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${a}'`),{x:()=>cA(s.shape,e,r,o,i,l),filter:()=>GA(s,e,r.shape,o,i,l)}}},zB={kernelName:xo,inputsToSave:["dy","filter"],gradFunc:(e,t,n)=>{let[s,r]=t,{strides:a,pad:o,dataFormat:i,dimRoundingMode:l}=n;return{dy:()=>ma(e,r,a,o,i,1,l),filter:()=>GA(e,s,r.shape,a,o,i,l)}}};function LB(e,t,n,s,r){let a=e;e.rank===4&&(a=U(e,[1,e.shape[0],e.shape[1],e.shape[2],e.shape[3]]));let o=t;o.rank===4&&(o=U(t,[1,t.shape[0],t.shape[1],t.shape[2],t.shape[3]])),M(a.rank===5,()=>`Error in conv3dDerFilter: input must be rank 5, but got shape ${a.shape}.`),M(o.rank===5,()=>`Error in conv3dDerFilter: dy must be rank 5, but got shape ${o.shape}.`),M(n.length===5,()=>`Error in conv3dDerFilter: filterShape must be length 5, but got ${n}.`),M(a.shape[4]===n[3],()=>`Error in conv3dDerFilter: depth of input ${a.shape[4]}) must match input depth in filter (${n[3]}.`),M(o.shape[4]===n[4],()=>`Error in conv3dDerFilter: depth of dy (${o.shape[4]}) must match output depth for filter (${n[4]}).`);let i={x:a,dy:o},l={strides:s,pad:r,filterShape:n};return W.runKernel(Km,i,l)}var BB=V({conv3DBackpropFilter_:LB}),WB={kernelName:Wp,inputsToSave:["x","filter"],gradFunc:(e,t,n)=>{let{dilations:s,strides:r,pad:a}=n;M(to(s),()=>`Error in gradient of conv3D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${s}'`);let[o,i]=t;return{x:()=>lw(o.shape,e,i,r,a),filter:()=>BB(o,e,i.shape,r,a)}}},VB={kernelName:bo,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(Dt(P0(ye(n,"float32"))),e)}}},UB={kernelName:vo,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(F0(ye(n,"float32")),e)}}},GB={kernelName:wo,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{axis:r,exclusive:a,reverse:o}=n;return{x:()=>{let i=fw([r],s.rank),l=b0(e,r,a,!o);return i!=null&&(l=et(l,i)),l}}}},HB={kernelName:ko,inputsToSave:["x","filter"],gradFunc:(e,t,n)=>{let{dilations:s,strides:r,pad:a,dimRoundingMode:o}=n,i=s==null?[1,1]:s;M(to(i),()=>`Error in gradient of depthwiseConv2dNative: dilation rates greater than 1 are not yet supported. Got dilations '${i}'`);let[l,u]=t;return M(l.rank===4,()=>`Error in gradient of depthwiseConv2dNative: input must be rank 4, but got rank ${l.rank}.`),M(u.rank===4,()=>`Error in gradient of depthwiseConv2dNative: filter must be rank 4, but got rank ${u.rank}.`),M(l.shape[3]===u.shape[2],()=>`Error in gradient of depthwiseConv2d: number of input channels (${l.shape[3]}) must match the inChannels dimension in filter ${u.shape[2]}.`),M(Yr(r,i),()=>`Error in gradient of depthwiseConv2d: Either strides or dilations must be 1. Got strides ${r} and dilations '${i}'.`),rs("depthwiseConv2d",a,o),{x:()=>Xw(l.shape,e,u,r,a,i,o),filter:()=>qw(l,e,u.shape,r,a,i,o)}}},jB={kernelName:Vp,inputsToSave:["x","filter"],gradFunc:(e,t,n)=>{let[s,r]=t,a={x:s,filter:r,dy:e},o={x:s,filter:r,dy:e};return{x:()=>W.runKernel(pm,a,n),filter:()=>W.runKernel(hm,o,n)}}},qB={kernelName:So,outputsToSave:[!0],gradFunc:(e,t)=>{let[n]=t,s={dy:e,y:n};return{x:()=>W.runKernel(t0,s)}}},XB={kernelName:wc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t,s=L(_s(Dt(vt(n))),2/Math.sqrt(Math.PI));return{x:()=>L(e,s)}}},KB={kernelName:Co,outputsToSave:[!0],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(e,n)}}},ZB={kernelName:fl,inputsToSave:["input"],gradFunc:(e,t)=>{let[n]=t;return{input:()=>U(e,n.shape)}}},YB={kernelName:ml,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(e,_s(n))}}},JB={kernelName:To,gradFunc:e=>({x:()=>it(e)})},QB={kernelName:No,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t,r=kt(n.shape,s.shape);return{a:()=>{let i=he(e,ye(s,"float32")),l=an(n.shape,r);return l.length>0?U(ke(i,l),n.shape):i},b:()=>{let i=L(e,ye(n,"float32")),l=an(s.shape,r);l.length>0&&(i=U(ke(i,l),s.shape));let u=vt(s);return Dt(he(i,ye(u,"float32")))}}}},eW={kernelName:Eo,inputsToSave:["x","mean","variance","scale"],gradFunc:(e,t,n)=>{let{varianceEpsilon:s}=n,[r,a,o,i]=t,l=i==null?Ce(1):i,u=an(a.shape,r.shape),c=[];if(a.rank===1){for(let b=0;b<r.shape.length-1;++b)c.push(r.shape[b]);c.push(1)}let p=fe(r,a),d=L(e,l),h=_0(ce(o,Ce(s))),f=L(L(L(h,h),h),Ce(-.5));return{x:()=>a.rank===1?U(L(L(e,qs(U(h,[1,1,1,a.shape[0]]),c)),l),r.shape):U(L(L(e,h),l),r.shape),mean:()=>{let b=L(L(h,Ce(-1)),d);return a.rank===1&&(b=ke(b,u)),U(b,a.shape)},variance:()=>{let b=L(L(f,p),d);return a.rank===1&&(b=ke(b,u)),U(b,a.shape)},scale:()=>{let b=L(p,h),w=L(e,b);return a.rank===1&&(w=ke(w,u)),U(w,a.shape)},offset:()=>{let b=e;return a.rank===1&&(b=ke(b,u)),U(b,a.shape)}}}},tW={kernelName:yl,inputsToSave:["x","indices"],gradFunc:(e,t,n)=>{let[s,r]=t,{axis:a}=n,o=cr(a,s.shape)[0];return{x:()=>{let l=s.shape,u=r.size,c=l.slice(0,o),p=c.length,d=l.slice(a,l.length).slice(1),h=d.length,f=Nv(0,p),m=Nv(p+1,p+1+h),g=Ev([c,[u],d]),y=U(e,g),x=U(r,[u]),A=Ev([[p],f,m]),b=et(y,A),w=B0(b,x,s.shape[o]),k=xA(A);return w=et(w,k),w},indices:()=>r}}};function Nv(e,t){let n=[];for(let s=e;s<t;++s)n.push(s);return n}function Ev(e){let t=[];for(let n=0;n<e.length;++n)for(let s=0;s<e[n].length;++s)t.push(e[n][s]);return t}var nW={kernelName:Ro,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t;return{a:()=>it(n),b:()=>it(s)}}},sW={kernelName:_o,gradFunc:e=>({x:()=>ye(e,"float32")})},rW={kernelName:Ic,gradFunc:e=>({x:()=>it(e)})},aW={kernelName:Sc,gradFunc:e=>({x:()=>it(e)})},oW={kernelName:Cc,gradFunc:e=>({x:()=>it(e)})},iW={kernelName:Do,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{alpha:r}=n,a=ys(s,0);return{x:()=>zn(a,e,L(e,r))}}},lW={kernelName:Tc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>he(e,ce(n,1))}}},uW={kernelName:$o,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>he(e,ye(n,"float32"))}}},cW={kernelName:m6,inputsToSave:[],outputsToSave:[!0],gradFunc:(e,t,n)=>{let[s]=t,{axis:r}=n;return{logits:()=>{let o=_s(s);return fe(e,L(ke(e,r,!0),o))}}}};function dW(e,t,n,s=5,r=1,a=1,o=.5){let i={x:e,y:t,dy:n},l={depthRadius:s,bias:r,alpha:a,beta:o};return W.runKernel(a0,i,l)}var pW=V({localResponseNormalizationBackprop_:dW}),hW={kernelName:Hp,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let[s,r]=t,{depthRadius:a,bias:o,alpha:i,beta:l}=n;return{x:()=>pW(s,r,e,a,o,i,l)}}};function d8(e,t,n,s){return t.rank<n.rank&&(t=U(t,Ji(t.shape,s))),e.rank<n.rank&&(e=U(e,Ji(e.shape,s))),{x:()=>L(e,ye(Rs(n,t),e.dtype))}}var Rv={kernelName:Po,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let s=n,{reductionIndices:r}=s,a=t[0],o=t[1],i=cr(r,a.shape),l=d8(e,o,a,i);return{x:()=>l.x()}}},fW={kernelName:Fo,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t;return{a:()=>L(e,ye(li(n,s),"float32")),b:()=>L(e,ye(w0(n,s),"float32"))}}};function mW(e,t,n,s,r,a,o){let i=D(e,"dy","maxPool3dGrad"),l=D(t,"input","maxPool3dGrad"),u=D(n,"output","maxPool3dGrad"),c=i,p=l,d=u,h=!1;l.rank===4&&(h=!0,c=U(i,[1,i.shape[0],i.shape[1],i.shape[2],i.shape[3]]),p=U(l,[1,l.shape[0],l.shape[1],l.shape[2],l.shape[3]]),d=U(u,[1,u.shape[0],u.shape[1],u.shape[2],u.shape[3]])),M(c.rank===5,()=>`Error in maxPool3dGrad: dy must be rank 5 but got rank ${c.rank}.`),M(p.rank===5,()=>`Error in maxPool3dGrad: input must be rank 5 but got rank ${p.rank}.`),M(d.rank===5,()=>`Error in maxPool3dGrad: output must be rank 5 but got rank ${d.rank}.`),rs("maxPool3dGrad",a,o);let f={dy:c,input:p,output:d},m={filterSize:s,strides:r,pad:a,dimRoundingMode:o},g=W.runKernel(i0,f,m);return h?U(g,[g.shape[1],g.shape[2],g.shape[3],g.shape[4]]):g}var gW=V({maxPool3dGrad_:mW}),yW={kernelName:jp,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let[s,r]=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=n;return{x:()=>gW(e,s,r,a,o,i,l)}}};function AW(e,t,n,s,r,a,o){let i=D(e,"dy","maxPoolGrad"),l=D(t,"input","maxPoolGrad"),u=D(n,"output","maxPoolGrad");M(l.rank===i.rank,()=>`Rank of input (${l.rank}) does not match rank of dy (${i.rank})`),M(i.rank===4,()=>`Error in maxPoolGrad: dy must be rank 4 but got rank ${i.rank}.`),M(l.rank===4,()=>`Error in maxPoolGrad: input must be rank 4 but got rank ${l.rank}.`),rs("maxPoolGrad",a,o);let c={dy:i,input:l,output:u},p={filterSize:s,strides:r,pad:a,dimRoundingMode:o};return W.runKernel(o0,c,p)}var xW=V({maxPoolGrad_:AW}),bW={kernelName:Oo,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let[s,r]=t,{filterSize:a,strides:o,pad:i}=n;return{x:()=>xW(e,s,r,a,o,i)}}},vW={kernelName:Mo,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{axis:r}=n,a=cr(r,s.shape),i=hw(s.shape,a)[1],l=Et(i);return{x:()=>{let c=s.shape.slice();a.forEach(h=>{c[h]=1});let p=U(e,c);return he(L(p,Ns(s.shape,"float32")),l)}}}},wW={kernelName:zo,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let s=n,{axis:r}=s,[a,o]=t,i=cr(r,a.shape),l=d8(e,o,a,i);return{x:()=>l.x()}}},kW={kernelName:Lo,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t;return{a:()=>L(e,ye(ui(n,s),"float32")),b:()=>L(e,ye(ys(n,s),"float32"))}}},IW={kernelName:Bo,inputsToSave:["x"],gradFunc:(e,t,n)=>{let s=t[0],{paddings:r}=n,a=r.map(o=>o[0]);return{x:()=>Oe(e,a,s.shape)}}},SW={kernelName:Ec,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t,r=kt(n.shape,s.shape);return{a:()=>{let i=an(n.shape,r);return i.length>0?U(ke(e,i),n.shape):e},b:()=>{let i=L(e,Dt(jc(he(n,s)))),l=an(s.shape,r);return l.length>0?U(ke(i,l),s.shape):i}}}},CW={kernelName:Wo,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t,r=kt(n.shape,s.shape);return{a:()=>{let i=L(e,ye(s,"float32")),l=an(n.shape,r);return l.length>0?U(ke(i,l),n.shape):i},b:()=>{let i=L(e,ye(n,"float32")),l=an(s.shape,r);return l.length>0?U(ke(i,l),s.shape):i}}}},TW={kernelName:Il,gradFunc:e=>({x:()=>Dt(e)})},NW={kernelName:El,inputsToSave:["indices"],gradFunc:(e,t)=>{let n=t[0];return{indices:()=>Wt(n.shape,"float32")}}},EW={kernelName:Nl,gradFunc:e=>({x:()=>it(e)})},RW={kernelName:Rl,saveAllInputs:!0,gradFunc:(e,t,n)=>{let{axis:s}=n;return En(e,s).map(a=>()=>a)}},_v={kernelName:Vo,inputsToSave:["x"],gradFunc:(e,t,n)=>{let s=t[0],{paddings:r}=n,a=r.map(o=>o[0]);return{x:()=>Oe(e,a,s.shape)}}},_W={kernelName:Uo,inputsToSave:["a","b"],outputsToSave:[!0],gradFunc:(e,t)=>{let[n,s,r]=t,a=n,o=s,i=kt(a.shape,o.shape);return{a:()=>{let c=ye(o,"float32"),p=L(e,L(c,ya(a,fe(c,Ce(1))))),d=an(a.shape,i);return d.length>0&&(p=ke(p,d)),U(p,a.shape)},b:()=>{let c=ys(a,0),p=zn(c,Ds(a),it(a)),d=L(e,L(r,p)),h=an(o.shape,i);return h.length>0&&(d=ke(d,h)),U(d,o.shape)}}}},DW={kernelName:Go,inputsToSave:["x","alpha"],gradFunc:(e,t)=>{let[n,s]=t,r=ys(n,0);return{x:()=>zn(r,e,L(e,s)),alpha:()=>{let a=zn(r,it(e),L(e,n)),o=an(s.shape,e.shape);return o.length>0&&(a=ke(a,o)),U(a,s.shape)}}}};function $W(e,t,n){let s=e.shape.slice();s[n]=1;let r=U(t,s),a=Tp(e,n,!0,!1),o=Tp(e,n,!0,!0),i=L(a,o);return L(r,i)}function PW(e,t,n){let s=e.shape.length,r=s-n.length,a=T.getAxesPermutation(n,s),o=e;a!=null&&(o=et(e,a));let i=o.shape.slice(),u=i.splice(s-n.length,n.length).reduce((d,h)=>d*h,1);i.push(u);let c=o.reshape(i),p=$W(c,t,r);if(p=p.reshape(o.shape),a!=null){let d=T.getUndoAxesPermutation(a);p=et(p,d)}return p}var FW={kernelName:Ho,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{axis:r}=n,a=[];return r==null?a=s.shape.map((o,i)=>i):typeof r=="number"?a=[r]:a=r,{x:()=>PW(s,e,a)}}},OW={kernelName:Io,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t,r=kt(n.shape,s.shape);return{a:()=>{let i=he(e,ye(s,"float32")),l=an(n.shape,r);return l.length>0?U(ke(i,l),n.shape):i},b:()=>{let i=L(e,ye(n,"float32")),l=an(s.shape,r);l.length>0&&(i=U(ke(i,l),s.shape));let u=vt(s);return Dt(he(i,ye(u,"float32")))}}}},MW={kernelName:Dc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>he(e,Dt(vt(n)))}}},zW={kernelName:Ko,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t,s=L(ui(n,6),Ql(n));return{x:()=>L(e,ye(s,"float32"))}}},LW={kernelName:jo,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(e,ye(Ql(n),"float32"))}}},BW={kernelName:_l,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>U(e,n.shape)}}},WW={kernelName:Xo,inputsToSave:["images"],gradFunc:(e,t,n)=>{let[s]=t,r={dy:e,images:s};return{images:()=>W.runKernel(d0,r,n)}}},VW={kernelName:qo,inputsToSave:["images"],gradFunc:(e,t,n)=>{let[s]=t,r={dy:e,images:s};return{images:()=>W.runKernel(c0,r,n)}}},UW={kernelName:Dl,gradFunc:(e,t,n)=>{let{dims:s}=n,r=cr(s,e.shape);return{x:()=>Zs(e,r)}}},GW={kernelName:$l,gradFunc:e=>({x:()=>it(e)})},HW={kernelName:Zo,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Dt(he(e,L(ya(n,1.5),2)))}}},jW={kernelName:Fl,inputsToSave:["condition"],gradFunc:(e,t)=>{let[n]=t;return{condition:()=>ye(it(n),"float32"),t:()=>L(e,ye(n,e.dtype)),e:()=>L(e,ye(hh(n),e.dtype))}}},qW={kernelName:$c,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>{let s=ys(n,Ce(0)),r=Ce(i8),a=Ce(l8),o=L(e,a),i=L(L(e,r),_s(ye(n,"float32")));return zn(s,o,i)}}}},XW={kernelName:Jo,outputsToSave:[!0],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(e,L(n,fe(Ce(1),n)))}}},KW={kernelName:Pc,gradFunc:e=>({x:()=>it(e)})},ZW={kernelName:Yo,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(ch(ye(n,"float32")),e)}}},YW={kernelName:Ml,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(x0(ye(n,"float32")),e)}}},JW={kernelName:Ol,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{begin:r,size:a}=n,o=s.shape,[i,l]=J6(s,r,a),u=[];for(let c=0;c<e.rank;c++)u.push([i[c],o[c]-i[c]-l[c]]);return{x:()=>Qs(e,u)}}},QW={kernelName:ti,outputsToSave:[!0],gradFunc:(e,t,n)=>{let[s]=t,{dim:r}=n,a=!0,o=L(e,s);return{logits:()=>fe(o,L(ke(o,[r],a),s))}}},eV={kernelName:Fc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(e,Cn(n))}}},Dv={kernelName:zl,gradFunc:(e,t,n)=>{let{blockShape:s,paddings:r}=n;return{x:()=>uh(e,s,r)}}},$v={kernelName:Ll,gradFunc:(e,t,n)=>{let{axis:s}=n;return{x:()=>Ct(e,s)}}},tV={kernelName:Qo,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>he(e,L(Nn(ye(n,"float32")),2))}}},nV={kernelName:Mc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(e,L(ye(n,"float32"),2))}}},sV={kernelName:ni,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t,r=Ce(2);return{a:()=>L(e,L(r,fe(n,s))),b:()=>L(e,L(r,fe(s,n)))}}},rV={kernelName:ai,gradFunc:e=>({x:()=>it(e)})},aV={kernelName:si,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,s]=t,r=kt(n.shape,s.shape);return{a:()=>{let i=e,l=an(n.shape,r);return l.length>0&&(i=ke(i,l)),U(i,n.shape)},b:()=>{let i=e,l=an(s.shape,r);return l.length>0&&(i=ke(i,l)),U(Dt(i),s.shape)}}}},oV={kernelName:ei,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,r=s.shape.slice(),{axis:a}=n;cr(a,s.shape).forEach(u=>{r[u]=1});let i=U(e,r),l=L(i,Ns(s.shape,"float32"));return{x:()=>l}}},iV={kernelName:Wl,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>he(e,vt(ch(n)))}}},lV={kernelName:ri,outputsToSave:[!0],gradFunc:(e,t)=>{let[n]=t;return{x:()=>L(fe(Ce(1),vt(n)),e)}}},uV={kernelName:va,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[s]=t,{reps:r}=n;return{x:()=>{let o=it(s);if(s.rank===1)for(let i=0;i<r[0];++i)o=ce(o,Oe(e,[i*s.shape[0]],[s.shape[0]]));else if(s.rank===2)for(let i=0;i<r[0];++i)for(let l=0;l<r[1];++l)o=ce(o,Oe(e,[i*s.shape[0],l*s.shape[1]],[s.shape[0],s.shape[1]]));else if(s.rank===3)for(let i=0;i<r[0];++i)for(let l=0;l<r[1];++l)for(let u=0;u<r[2];++u)o=ce(o,Oe(e,[i*s.shape[0],l*s.shape[1],u*s.shape[2]],[s.shape[0],s.shape[1],s.shape[2]]));else if(s.rank===4)for(let i=0;i<r[0];++i)for(let l=0;l<r[1];++l)for(let u=0;u<r[2];++u)for(let c=0;c<r[3];++c)o=ce(o,Oe(e,[i*s.shape[0],l*s.shape[1],u*s.shape[2],c*s.shape[3]],[s.shape[0],s.shape[1],s.shape[2],s.shape[3]]));else throw new Error(`Gradient for tile operation is not implemented for rank-${s.rank} tensors yet.`);return o}}}},cV={kernelName:Hr,gradFunc:(e,t,n)=>{let s=n,{perm:r}=s,a=xA(r);return{x:()=>et(e,a)}}},dV={kernelName:Gl,gradFunc:(e,t,n)=>{let s=n,{axis:r}=s;return{value:()=>on(e,r)}}},pV={kernelName:eh,inputsToSave:["segmentIds"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>hV(e,n)}}};function hV(e,t){let n=Jr(t,it(t)),s=qc(e,n),r=li(t,Ce(0,"int32")),a=s.rank-r.rank;for(let i=0;i<a;++i)r=Kt(r,i+1);r=ur(r,Ns(s.shape,"bool"));let o=it(s);return zn(r,s,o)}var fV={kernelName:Hl,gradFunc:e=>({x:()=>it(e)})},mV=[c8,pB,hB,fB,mB,gB,yB,AB,xB,bB,vB,wB,SB,NB,EB,RB,_B,DB,$B,PB,FB,OB,zB,MB,WB,VB,UB,GB,HB,jB,OW,qB,XB,KB,ZB,YB,QB,JB,eW,tW,nW,sW,rW,aW,oW,iW,lW,uW,cW,hW,Rv,Rv,fW,yW,bW,vW,wW,kW,IW,SW,CW,TW,NW,EW,RW,_v,_v,_W,DW,FW,MW,zW,LW,BW,WW,VW,UW,GW,HW,jW,qW,XW,KW,ZW,YW,JW,QW,eV,Dv,Dv,$v,$v,tV,sV,nV,rV,aV,oV,iV,lV,uV,cV,dV,pV,fV];for(let e of mV)g6(e);re().prototype.abs=function(){return this.throwIfDisposed(),tn(this)};re().prototype.acos=function(){return this.throwIfDisposed(),jy(this)};re().prototype.acosh=function(){return this.throwIfDisposed(),qy(this)};re().prototype.add=function(e){return this.throwIfDisposed(),ce(this,e)};re().prototype.all=function(e,t){return this.throwIfDisposed(),g0(this,e,t)};re().prototype.any=function(e,t){return this.throwIfDisposed(),Cp(this,e,t)};re().prototype.argMax=function(e){return this.throwIfDisposed(),Es(this,e)};re().prototype.argMin=function(e){return this.throwIfDisposed(),Xy(this,e)};re().prototype.asScalar=function(){return this.throwIfDisposed(),M(this.size===1,()=>"The array must have only 1 element."),U(this,[])};re().prototype.asType=function(e){return this.throwIfDisposed(),ye(this,e)};re().prototype.as1D=function(){return this.throwIfDisposed(),U(this,[this.size])};re().prototype.as2D=function(e,t){return this.throwIfDisposed(),U(this,[e,t])};re().prototype.as3D=function(e,t,n){return this.throwIfDisposed(),U(this,[e,t,n])};re().prototype.as4D=function(e,t,n,s){return this.throwIfDisposed(),U(this,[e,t,n,s])};re().prototype.as5D=function(e,t,n,s,r){return this.throwIfDisposed(),U(this,[e,t,n,s,r])};re().prototype.asin=function(){return this.throwIfDisposed(),Ky(this)};re().prototype.asinh=function(){return this.throwIfDisposed(),Zy(this)};re().prototype.atan=function(){return this.throwIfDisposed(),Yy(this)};re().prototype.atan2=function(e){return this.throwIfDisposed(),Jy(this,e)};re().prototype.atanh=function(){return this.throwIfDisposed(),Qy(this)};re().prototype.avgPool=function(e,t,n,s){return this.throwIfDisposed(),lh(this,e,t,n,s)};re().prototype.batchToSpaceND=function(e,t){return this.throwIfDisposed(),uh(this,e,t)};re().prototype.batchNorm=function(e,t,n,s,r){return this.throwIfDisposed(),Wc(this,e,t,n,s,r)};re().prototype.broadcastTo=function(e){return this.throwIfDisposed(),Gu(this,e)};re().prototype.cast=function(e){return this.throwIfDisposed(),ye(this,e)};re().prototype.ceil=function(){return this.throwIfDisposed(),oA(this)};re().prototype.clipByValue=function(e,t){return this.throwIfDisposed(),fs(this,e,t)};re().prototype.concat=function(e,t){return this.throwIfDisposed(),e instanceof nt&&(e=[e]),Ct([this,...e],t)};re().prototype.conv1d=function(e,t,n,s,r,a){return this.throwIfDisposed(),y0(this,e,t,n,s,r,a)};re().prototype.conv2dTranspose=function(e,t,n,s,r){return this.throwIfDisposed(),A0(this,e,t,n,s,r)};re().prototype.conv2d=function(e,t,n,s,r,a){return this.throwIfDisposed(),ma(this,e,t,n,s,r,a)};re().prototype.cos=function(){return this.throwIfDisposed(),ch(this)};re().prototype.cosh=function(){return this.throwIfDisposed(),x0(this)};re().prototype.cumprod=function(e,t,n){return this.throwIfDisposed(),Tp(this,e,t,n)};re().prototype.cumsum=function(e,t,n){return this.throwIfDisposed(),b0(this,e,t,n)};re().prototype.depthToSpace=function(e,t){return this.throwIfDisposed(),hA(this,e,t)};re().prototype.depthwiseConv2d=function(e,t,n,s,r,a){return this.throwIfDisposed(),Vc(this,e,t,n,s,r,a)};re().prototype.dilation2d=function(e,t,n,s,r){return this.throwIfDisposed(),fA(this,e,t,n,s,r)};re().prototype.divNoNan=function(e){return this.throwIfDisposed(),mA(this,e)};re().prototype.div=function(e){return this.throwIfDisposed(),he(this,e)};re().prototype.dot=function(e){return this.throwIfDisposed(),gA(this,e)};re().prototype.elu=function(){return this.throwIfDisposed(),Uc(this)};re().prototype.equal=function(e){return this.throwIfDisposed(),Rs(this,e)};re().prototype.erf=function(){return this.throwIfDisposed(),yA(this)};re().prototype.euclideanNorm=function(e,t){return this.throwIfDisposed(),bA(this,e,t)};re().prototype.exp=function(){return this.throwIfDisposed(),_s(this)};re().prototype.expandDims=function(e){return this.throwIfDisposed(),Kt(this,e)};re().prototype.expm1=function(){return this.throwIfDisposed(),vA(this)};re().prototype.fft=function(){return this.throwIfDisposed(),xh(this)};re().prototype.flatten=function(){return this.throwIfDisposed(),U(this,[this.size])};re().prototype.floor=function(){return this.throwIfDisposed(),jc(this)};re().prototype.floorDiv=function(e){return this.throwIfDisposed(),Bc(this,e)};re().prototype.gather=function(e,t){return this.throwIfDisposed(),qc(this,e,t)};re().prototype.greaterEqual=function(e){return this.throwIfDisposed(),li(this,e)};re().prototype.greater=function(e){return this.throwIfDisposed(),ys(this,e)};re().prototype.ifft=function(){return this.throwIfDisposed(),Qu(this)};re().prototype.irfft=function(){return this.throwIfDisposed(),M0(this)};re().prototype.isFinite=function(){return this.throwIfDisposed(),wA(this)};re().prototype.isInf=function(){return this.throwIfDisposed(),kA(this)};re().prototype.isNaN=function(){return this.throwIfDisposed(),IA(this)};re().prototype.leakyRelu=function(e){return this.throwIfDisposed(),dh(this,e)};re().prototype.lessEqual=function(e){return this.throwIfDisposed(),ui(this,e)};re().prototype.less=function(e){return this.throwIfDisposed(),w0(this,e)};re().prototype.localResponseNormalization=function(e,t,n,s){return this.throwIfDisposed(),SA(this,e,t,n,s)};re().prototype.logSigmoid=function(){return this.throwIfDisposed(),CA(this)};re().prototype.logSoftmax=function(e){return this.throwIfDisposed(),I0(this,e)};re().prototype.logSumExp=function(e,t){return this.throwIfDisposed(),S0(this,e,t)};re().prototype.log=function(){return this.throwIfDisposed(),Ds(this)};re().prototype.log1p=function(){return this.throwIfDisposed(),ph(this)};re().prototype.logicalAnd=function(e){return this.throwIfDisposed(),ur(this,e)};re().prototype.logicalNot=function(){return this.throwIfDisposed(),hh(this)};re().prototype.logicalOr=function(e){return this.throwIfDisposed(),C0(this,e)};re().prototype.logicalXor=function(e){return this.throwIfDisposed(),TA(this,e)};re().prototype.matMul=function(e,t,n){return this.throwIfDisposed(),Qe(this,e,t,n)};re().prototype.maxPool=function(e,t,n,s){return this.throwIfDisposed(),fh(this,e,t,n,s)};re().prototype.max=function(e,t){return this.throwIfDisposed(),mn(this,e,t)};re().prototype.maximum=function(e){return this.throwIfDisposed(),Jr(this,e)};re().prototype.mean=function(e,t){return this.throwIfDisposed(),Bt(this,e,t)};re().prototype.min=function(e,t){return this.throwIfDisposed(),ga(this,e,t)};re().prototype.minimum=function(e){return this.throwIfDisposed(),Xc(this,e)};re().prototype.mirrorPad=function(e,t){return this.throwIfDisposed(),EA(this,e,t)};re().prototype.mod=function(e){return this.throwIfDisposed(),Yl(this,e)};re().prototype.mul=function(e){return this.throwIfDisposed(),L(this,e)};re().prototype.neg=function(){return this.throwIfDisposed(),Dt(this)};re().prototype.norm=function(e,t,n){return this.throwIfDisposed(),Gc(this,e,t,n)};re().prototype.notEqual=function(e){return this.throwIfDisposed(),Qi(this,e)};re().prototype.oneHot=function(e,t=1,n=0){return this.throwIfDisposed(),Zu(this,e,t,n)};re().prototype.onesLike=function(){return this.throwIfDisposed(),$s(this)};re().prototype.pad=function(e,t){return this.throwIfDisposed(),Qs(this,e,t)};re().prototype.pool=function(e,t,n,s,r,a){return this.throwIfDisposed(),RA(this,e,t,n,s,r,a)};re().prototype.pow=function(e){return this.throwIfDisposed(),ya(this,e)};re().prototype.prelu=function(e){return this.throwIfDisposed(),yh(this,e)};re().prototype.prod=function(e,t){return this.throwIfDisposed(),_A(this,e,t)};re().prototype.reciprocal=function(){return this.throwIfDisposed(),PA(this)};re().prototype.relu=function(){return this.throwIfDisposed(),Pr(this)};re().prototype.relu6=function(){return this.throwIfDisposed(),E0(this)};re().prototype.reshapeAs=function(e){return this.throwIfDisposed(),U(this,e.shape)};re().prototype.reshape=function(e){return this.throwIfDisposed(),U(this,e)};re().prototype.resizeBilinear=function(e,t,n){return this.throwIfDisposed(),e8(this,e,t,n)};re().prototype.resizeNearestNeighbor=function(e,t,n){return this.throwIfDisposed(),t8(this,e,t,n)};re().prototype.reverse=function(e){return this.throwIfDisposed(),Zs(this,e)};re().prototype.rfft=function(){return this.throwIfDisposed(),bh(this)};re().prototype.round=function(){return this.throwIfDisposed(),R0(this)};re().prototype.rsqrt=function(){return this.throwIfDisposed(),_0(this)};re().prototype.selu=function(){return this.throwIfDisposed(),D0(this)};re().prototype.separableConv2d=function(e,t,n,s,r,a){return this.throwIfDisposed(),$0(this,e,t,n,s,r,a)};re().prototype.sigmoid=function(){return this.throwIfDisposed(),Cn(this)};re().prototype.sign=function(){return this.throwIfDisposed(),FA(this)};re().prototype.sin=function(){return this.throwIfDisposed(),P0(this)};re().prototype.sinh=function(){return this.throwIfDisposed(),F0(this)};re().prototype.slice=function(e,t){return this.throwIfDisposed(),Oe(this,e,t)};re().prototype.softmax=function(e){return this.throwIfDisposed(),Jl(this,e)};re().prototype.softplus=function(){return this.throwIfDisposed(),Zl(this)};re().prototype.spaceToBatchND=function(e,t){return this.throwIfDisposed(),gh(this,e,t)};re().prototype.split=function(e,t){return this.throwIfDisposed(),Zt(this,e,t)};re().prototype.sqrt=function(){return this.throwIfDisposed(),Nn(this)};re().prototype.square=function(){return this.throwIfDisposed(),vt(this)};re().prototype.squaredDifference=function(e){return this.throwIfDisposed(),z0(this,e)};re().prototype.squeeze=function(e){return this.throwIfDisposed(),st(this,e)};re().prototype.stack=function(e,t){this.throwIfDisposed();let n=e instanceof nt?[this,e]:[this,...e];return on(n,t)};re().prototype.step=function(e){return this.throwIfDisposed(),Ql(this,e)};re().prototype.stridedSlice=function(e,t,n,s,r,a,o,i){return this.throwIfDisposed(),OA(this,e,t,n,s,r,a,o,i)};re().prototype.sub=function(e){return this.throwIfDisposed(),fe(this,e)};re().prototype.sum=function(e,t){return this.throwIfDisposed(),ke(this,e,t)};re().prototype.tan=function(){return this.throwIfDisposed(),MA(this)};re().prototype.tanh=function(){return this.throwIfDisposed(),Yi(this)};re().prototype.tile=function(e){return this.throwIfDisposed(),qs(this,e)};re().prototype.toBool=function(){return this.throwIfDisposed(),ye(this,"bool")};re().prototype.toFloat=function(){return this.throwIfDisposed(),ye(this,"float32")};re().prototype.toInt=function(){return this.throwIfDisposed(),ye(this,"int32")};re().prototype.topk=function(e,t){return this.throwIfDisposed(),zA(this,e,t)};re().prototype.transpose=function(e){return this.throwIfDisposed(),et(this,e)};re().prototype.unique=function(e){return this.throwIfDisposed(),LA(this,e)};re().prototype.unsortedSegmentSum=function(e,t){return this.throwIfDisposed(),B0(this,e,t)};re().prototype.unstack=function(e){return this.throwIfDisposed(),En(this,e)};re().prototype.where=function(e,t){return this.throwIfDisposed(),zn(e,this,t)};re().prototype.zerosLike=function(){return this.throwIfDisposed(),it(this)};var la=class extends Error{constructor(e){super(e),Object.setPrototypeOf(this,la.prototype)}},Cr=class extends Error{constructor(e){super(e),Object.setPrototypeOf(this,Cr.prototype)}},j=class extends Error{constructor(e){super(e),Object.setPrototypeOf(this,j.prototype)}},qe=class extends Error{constructor(e){super(e),Object.setPrototypeOf(this,qe.prototype)}},p8=class extends Error{constructor(e){super(e),Object.setPrototypeOf(this,p8.prototype)}},h8=class{constructor(e){this.maxEntries=e||100,this.cache=new Map}get(e){let t;return this.cache.has(e)&&(t=this.cache.get(e),this.cache.delete(e),this.cache.set(e,t)),t}put(e,t){if(this.cache.has(e))this.cache.delete(e);else if(this.cache.size>=this.maxEntries){let n=this.cache.keys().next().value;this.cache.delete(n)}this.cache.set(e,t)}getMaxEntries(){return this.maxEntries}setMaxEntries(e){if(e<0)throw new Error(`The maxEntries of LRU caches must be at least 0, but got ${e}.`);if(this.maxEntries>e)for(let t=0;t<this.maxEntries-e;t++){let n=this.cache.keys().next().value;this.cache.delete(n)}this.maxEntries=e}};function el(e,t){if(Array.isArray(e)){let n=[];for(let s=0;s<t;s++)n=n.concat(e);return n}else{let n=new Array(t);return n.fill(e),n}}function Vr(e,t){if(!e)throw new p8(t)}function Pv(e,t){let n=0;for(let s of e)s===t&&n++;return n}function ps(e){return e.length===1?e[0]:e}function _t(e){return Array.isArray(e)?e:[e]}function ua(e){let n=e.replace(/(.)([A-Z][a-z0-9]+)/g,"$1_$2").replace(/([a-z])([A-Z])/g,"$1_$2").toLowerCase();return n[0]!=="_"?n:"private"+n}function Li(e){return e.length<=1||e.indexOf("_")===-1?e:e.replace(/[_]+(\w|$)/g,(t,n)=>n.toUpperCase())}var sr={};function KA(e){if(e==null)return null;let t={};return t.className=e.getClassName(),t.config=e.getConfig(),t}function D3(e){if(!(e==null||typeof e!="object"))if(Array.isArray(e))e.forEach(t=>D3(t));else{let t=Object.keys(e);for(let n of t){let s=e[n];s!=null&&typeof s=="object"&&(!Array.isArray(s)&&s.type==="ndarray"&&typeof s.value=="number"?e[n]=s.value:D3(s))}}}function wh(e,t={},n={},s="object",r=!1){if(typeof e=="string"){let a=e,o;if(a in n)o=n[a];else if(a in sr)o=sr[a];else if(o=t[a],o==null)throw new j(`Unknown ${s}: ${e}. This may be due to one of the following reasons:
|
|
1. The ${s} is defined in Python, in which case it needs to be ported to TensorFlow.js or your JavaScript code.
|
|
2. The custom ${s} is defined in JavaScript, but is not registered properly with tf.serialization.registerClass().`);return o}else{let a=e;if(a.className==null||a.config==null)throw new j(`${s}: Improper config format: ${JSON.stringify(a)}.
|
|
'className' and 'config' must set.`);let o=a.className,i,l;if(o in n?[i,l]=n[o]:o in sr?[i,l]=sr.className:o in t&&([i,l]=t[o]),i==null)throw new j(`Unknown ${s}: ${o}. This may be due to one of the following reasons:
|
|
1. The ${s} is defined in Python, in which case it needs to be ported to TensorFlow.js or your JavaScript code.
|
|
2. The custom ${s} is defined in JavaScript, but is not registered properly with tf.serialization.registerClass().`);if(l!=null){let u={};for(let h of Object.keys(sr))u[h]=sr[h];for(let h of Object.keys(n))u[h]=n[h];let c=a.config;c.customObjects=u;let p=Object.assign({},sr);for(let h of Object.keys(n))sr[h]=n[h];D3(a.config);let d=l(i,a.config,n,r);return sr=Object.assign({},p),d}else{let u=Object.assign({},sr);for(let p of Object.keys(n))sr[p]=n[p];let c=new i(a.config);return sr=Object.assign({},u),c}}}function gV(e,t){return e<t?-1:e>t?1:0}function Wf(e,t){return-1*gV(e,t)}function Xa(e){if(e==null)return e;let t=[];for(let n of e)t.indexOf(n)===-1&&t.push(n);return t}function yV(e){if(e==null)throw new j(`Invalid value in obj: ${JSON.stringify(e)}`);for(let t in e)if(e.hasOwnProperty(t))return!1;return!0}function eu(e,t,n){if(n!=null&&e.indexOf(n)<0)throw new j(`${n} is not a valid ${t}. Valid values are ${e} or null/undefined.`)}function ZA(e,t,n=0,s=1/0){return Vr(n>=0),Vr(s>=n),Array.isArray(e)&&e.length>=n&&e.length<=s&&e.every(r=>typeof r===t)}function vn(e,t){Array.isArray(e)?(v.assert(e.length>0,()=>`${t} is unexpectedly an empty array.`),e.forEach((n,s)=>vn(n,`element ${s+1} of ${t}`))):v.assert(Number.isInteger(e)&&e>0,()=>`Expected ${t} to be a positive integer, but got ${f8(e)}.`)}function f8(e){return e===null?"null":Array.isArray(e)?"["+e.map(t=>f8(t)).join(",")+"]":typeof e=="string"?`"${e}"`:`${e}`}function AV(e,t,n){let s=n!=null?n():v.now(),r;return(...o)=>{let i=n!=null?n():v.now();return i-s<t||(s=i,r=e(...o)),r}}function m8(e){return e==="relu"?"relu":e==="linear"?"linear":e==="elu"?"elu":null}var xV=0;function g8(){return xV++}var Vf={};function J0(e=""){return e in Vf||(Vf[e]=0),Vf[e]+=1,e+Vf[e].toString()}var bV=["channelsFirst","channelsLast"],vV=["nearest","bilinear"],wV=["valid","same","causal"],kV=["max","avg"],IV=["sum","mul","concat","ave"],$u=new Map;function Yt(e){eu(bV,"DataFormat",e)}function SV(e){eu(vV,"InterpolationFormat",e)}function er(e){eu(wV,"PaddingMode",e)}function y8(e){eu(kV,"PoolMode",e)}var mp=[],Fv="/";function Hi(e,t){mp.push(e);try{let n=t();return mp.pop(),n}catch(n){throw mp.pop(),n}}function CV(){return mp.length===0?"":mp.join(Fv)+Fv}function A8(e){if(!b8(e))throw new Error("Not a valid tensor name: '"+e+"'");return CV()+e}function x8(e){if(!b8(e))throw new Error("Not a valid tensor name: '"+e+"'");$u.has(e)||$u.set(e,0);let t=$u.get(e);if($u.set(e,$u.get(e)+1),t>0){let n=`${e}_${t}`;return $u.set(n,1),n}else return e}var TV=new RegExp(/^[A-Za-z0-9][-A-Za-z0-9\._\/]*$/);function b8(e){return!!e.match(TV)}function NV(e){return e===parseInt(e.toString(),10)}function Ka(e,t,n){t==null&&(t=0),n==null&&(n=e.length);let s=1;for(let r=t;r<n;++r)s*=e[r];return s}function tc(e){if(e.length===0)return Number.NaN;let t=Number.POSITIVE_INFINITY;for(let n=0;n<e.length;n++){let s=e[n];s<t&&(t=s)}return t}function so(e){if(e.length===0)return Number.NaN;let t=Number.NEGATIVE_INFINITY;for(let n=0;n<e.length;n++){let s=e[n];s>t&&(t=s)}return t}function _r(e,t){if(t<e)throw new j(`end (${t}) < begin (${e}) is forbidden.`);let n=[];for(let s=e;s<t;++s)n.push(s);return n}var l3;function hn(){return l3==null&&(l3=Ys().epsilon()),l3}function Dr(){return"channelsLast"}function Q0(e,t){return ye(e,t)}function kh(e,t=-1){let n=e.shape.slice();return t<0&&(t=n.length+t+1),n.splice(t,0,1),U(e,n)}function EV(e,t){return Y(()=>{if(e.shape.length!==2)throw new j(`repeat() expects a rank-2 tensor, but received a rank-${e.shape.length} tensor.`);let n=kh(e,1);return $3(n,[1,t,1])})}function RV(e){let t=[Ka(e.shape)];return U(e,t)}function _V(e){if(e.rank<=1)throw new j(`batchFlatten requires a minimum rank of 2. Got rank: ${e.rank}.`);let t=[e.shape[0],Ka(e.shape,1)];return U(e,t)}function ji(e,t,n){return Y(()=>{switch(e.rank){case 1:return Ah(e,t,n);case 2:return O0(e,[t,0],[n,e.shape[1]]);case 3:return ci(e,[t,0,0],[n,e.shape[1],e.shape[2]]);case 4:return no(e,[t,0,0,0],[n,e.shape[1],e.shape[2],e.shape[3]]);case 5:return Oe(e,[t,0,0,0,0],[n,e.shape[1],e.shape[2],e.shape[3],e.shape[4]]);case 6:return Oe(e,[t,0,0,0,0,0],[n,e.shape[1],e.shape[2],e.shape[3],e.shape[4],e.shape[5]]);default:throw new j(`sliceAlongFirstAxis() received an unsupported tensor rank: ${e.rank}`)}})}function u3(e,t,n){return Y(()=>{switch(e.rank){case 1:return Ah(e,t,n);case 2:return O0(e,[0,t],[e.shape[0],n]);case 3:return ci(e,[0,0,t],[e.shape[0],e.shape[1],n]);case 4:return no(e,[0,0,0,t],[e.shape[0],e.shape[1],e.shape[2],n]);default:throw new j(`sliceAlongLastAxis() received an unsupported tensor rank: ${e.rank}`)}})}function Uf(e,t,n,s){return Y(()=>{switch(e.rank){case 1:return Ah(e,t,n);case 2:switch(s){case 1:return ji(e,t,n);case 2:return u3(e,t,n);default:throw new j(`The axis is not within the rank of the tensor ${s}`)}case 3:switch(s){case 1:return ji(e,t,n);case 2:return ci(e,[0,t,0],[e.shape[0],n,e.shape[2]]);case 3:return u3(e,t,n);default:throw new j(`The axis is not within the rank of the tensor ${s}`)}case 4:switch(s){case 1:return ji(e,t,n);case 2:return no(e,[0,t,0,0],[e.shape[0],n,e.shape[2],e.shape[3]]);case 3:return no(e,[0,0,t,0],[e.shape[0],e.shape[1],n,e.shape[3]]);case 4:return u3(e,t,n);default:throw new j(`The axis is not within the rank of the tensor ${s}`)}default:throw new j(`sliceAlongLastAxis() received an unsupported tensor rank: ${e.rank}`)}})}function YA(e,t=-1){let n;return t<0&&(n=e[0].rank,n!==0?t=n:t=0),t===e[0].rank&&(t=-1),Ct(e,t)}function Ov(e,t){switch(e.rank){case 1:return iA([e,t]);case 2:return Kl([e,t],0);case 3:return lA([e,t],0);case 4:return uA([e,t],0);default:throw new j(`concatAlongFirstAxis() received an unsupported tensor rank: ${e.rank}`)}}function $3(e,t){if(Array.isArray(t)||(t=[t]),e.rank!==t.length)throw new j(`The length of input n (${t.length}) does not match the number of dimensions in input x (${e.rank})`);return qs(e,t)}function e2(e,t=0,n=1,s,r){return N0(e,t,n,s,r)}function jr(e,t,n,s){if(e.rank<2||t.rank<2)throw new qe(`dot requires both inputs to be rank >= 2 but got x shape = ${e.shape} and y shape = ${t.shape}`);if(t.rank>=3){let r=e.shape.slice(-1)[0],a=t.shape.slice(-2)[0];if(r!==a)throw new qe(`If rank y >= 3, then the second last dim of y must equal the last dim of x but got x shape = ${e.shape} and y shape = ${t.shape}`)}if(e.rank===2&&t.rank===2)return ec.matMul({a:e,b:t,transposeA:!1,transposeB:!1,bias:s?P3(e.rank,s,Dr()):null,activation:n});{let r=e.shape.slice(),a=r.pop();e=U(e,[-1,a]);let o=t.shape.slice(),i=o.pop(),l=o.pop(),u=[...o,i],c=Array.from({length:t.rank},(f,m)=>m===0?t.rank-2:m<=t.rank-2?m-1:m);t=U(et(t,c),[l,-1]);let p=[...r,...u],d=!1,h=!1;return U(ec.matMul({a:e,b:t,transposeA:d,transposeB:h,bias:s?P3(e.rank,s,Dr()):null,activation:n}),p)}}function v8(e,t,n){return Y(()=>(Array.isArray(t)?t=Ft(t,"int32"):t=ye(t,"int32"),qc(e,t,n)))}function Ih(e){return L(e,e)}function P3(e,t,n){let s=t.shape;if(t.rank!==1&&t.rank!==e)throw new j(`Unexpected bias dimensions: ${t.rank}; expected it to be 1 or ${e}`);if(e===5){if(n==="channelsFirst")return s.length===1?U(t,[1,s[0],1,1,1]):U(t,[1,s[3],s[0],s[1],s[2]]);if(n==="channelsLast")return s.length===1?U(t,[1,1,1,1,s[0]]):U(t,[1].concat(s))}else if(e===4){if(n==="channelsFirst")return s.length===1?U(t,[1,s[0],1,1]):U(t,[1,s[2],s[0],s[1]]);if(n==="channelsLast")return s.length===1?U(t,[1,1,1,s[0]]):U(t,[1].concat(s))}else if(e===3){if(n==="channelsFirst")return s.length===1?U(t,[1,s[0],1]):U(t,[1,s[1],s[0]]);if(n==="channelsLast")return s.length===1?U(t,[1,1,s[0]]):U(t,[1].concat(s))}else if(e<3)return t;throw new j(`Unsupported input rank by biasAdd: ${t.rank}`)}function Fr(e,t,n){return Y(()=>(n==null&&(n=Dr()),Yt(n),ce(e,P3(e.rank,t,n))))}function DV(e,t=1){if(t!==1)throw new qe(`Support for alpha values other than 1 (${t}) is not implemented yet.`);return Uc(e)}function $V(e){return Y(()=>he(e,ce(tn(e),1)))}function w8(e,t,n,s){return Y(()=>VA(e,t,n,s))}function PV(e){return Y(()=>{let t=ce(.5,L(.2,e));return fs(t,0,1)})}function Sh(e,t,n=!1){return n?e():t()}var FV=["fanIn","fanOut","fanAvg"],OV=["normal","uniform","truncatedNormal"];function MV(e){eu(FV,"FanMode",e)}function zV(e){eu(OV,"Distribution",e)}var hr=class extends de.Serializable{fromConfigUsesCustomObjects(){return!1}getConfig(){return{}}},JA=class extends hr{apply(e,t){return Wt(e,t)}};JA.className="Zeros";de.registerClass(JA);var t2=class extends hr{apply(e,t){return Ns(e,t)}};t2.className="Ones";de.registerClass(t2);var QA=class extends hr{constructor(e){if(super(),typeof e!="object")throw new j(`Expected argument of type ConstantConfig but got ${e}`);if(e.value===void 0)throw new j(`config must have value set but got ${e}`);this.value=e.value}apply(e,t){return Y(()=>L(Ce(this.value),Ns(e,t)))}getConfig(){return{value:this.value}}};QA.className="Constant";de.registerClass(QA);var e5=class extends hr{constructor(e){super(),this.DEFAULT_MINVAL=-.05,this.DEFAULT_MAXVAL=.05,this.minval=e.minval||this.DEFAULT_MINVAL,this.maxval=e.maxval||this.DEFAULT_MAXVAL,this.seed=e.seed}apply(e,t){return Kc(e,this.minval,this.maxval,t)}getConfig(){return{minval:this.minval,maxval:this.maxval,seed:this.seed}}};e5.className="RandomUniform";de.registerClass(e5);var t5=class extends hr{constructor(e){super(),this.DEFAULT_MEAN=0,this.DEFAULT_STDDEV=.05,this.mean=e.mean||this.DEFAULT_MEAN,this.stddev=e.stddev||this.DEFAULT_STDDEV,this.seed=e.seed}apply(e,t){if(t=t||"float32",t!=="float32"&&t!=="int32")throw new qe(`randomNormal does not support dType ${t}.`);return e2(e,this.mean,this.stddev,t,this.seed)}getConfig(){return{mean:this.mean,stddev:this.stddev,seed:this.seed}}};t5.className="RandomNormal";de.registerClass(t5);var n5=class extends hr{constructor(e){super(),this.DEFAULT_MEAN=0,this.DEFAULT_STDDEV=.05,this.mean=e.mean||this.DEFAULT_MEAN,this.stddev=e.stddev||this.DEFAULT_STDDEV,this.seed=e.seed}apply(e,t){if(t=t||"float32",t!=="float32"&&t!=="int32")throw new qe(`truncatedNormal does not support dType ${t}.`);return L0(e,this.mean,this.stddev,t,this.seed)}getConfig(){return{mean:this.mean,stddev:this.stddev,seed:this.seed}}};n5.className="TruncatedNormal";de.registerClass(n5);var s5=class extends hr{constructor(e){super(),this.gain=e.gain!=null?e.gain:1}apply(e,t){return Y(()=>{if(e.length!==2||e[0]!==e[1])throw new j("Identity matrix initializer can only be used for 2D square matrices.");return L(this.gain,v0(e[0]))})}getConfig(){return{gain:this.gain}}};s5.className="Identity";de.registerClass(s5);function LV(e,t="channelsLast"){let n,s;if(Yt(t),e.length===2)n=e[0],s=e[1];else if([3,4,5].indexOf(e.length)!==-1){if(t==="channelsFirst"){let r=Ka(e,2);n=e[1]*r,s=e[0]*r}else if(t==="channelsLast"){let r=Ka(e,0,e.length-2);n=e[e.length-2]*r,s=e[e.length-1]*r}}else{let r=Ka(e);n=Math.sqrt(r),s=Math.sqrt(r)}return[n,s]}var ms=class extends hr{constructor(e){if(super(),e.scale<0)throw new j(`scale must be a positive float. Got: ${e.scale}`);this.scale=e.scale==null?1:e.scale,this.mode=e.mode==null?"fanIn":e.mode,MV(this.mode),this.distribution=e.distribution==null?"normal":e.distribution,zV(this.distribution),this.seed=e.seed}apply(e,t){let n=LV(e),s=n[0],r=n[1],a=this.scale;if(this.mode==="fanIn"?a/=Math.max(1,s):this.mode==="fanOut"?a/=Math.max(1,r):a/=Math.max(1,(s+r)/2),this.distribution==="normal"){let o=Math.sqrt(a);if(t=t||"float32",t!=="float32"&&t!=="int32")throw new qe(`${this.getClassName()} does not support dType ${t}.`);return L0(e,0,o,t,this.seed)}else{let o=Math.sqrt(3*a);return Kc(e,-o,o,t)}}getConfig(){return{scale:this.scale,mode:this.mode,distribution:this.distribution,seed:this.seed}}};ms.className="VarianceScaling";de.registerClass(ms);var n2=class extends ms{constructor(e){super({scale:1,mode:"fanAvg",distribution:"uniform",seed:e==null?null:e.seed})}getClassName(){return ms.className}};n2.className="GlorotUniform";de.registerClass(n2);var s2=class extends ms{constructor(e){super({scale:1,mode:"fanAvg",distribution:"normal",seed:e==null?null:e.seed})}getClassName(){return ms.className}};s2.className="GlorotNormal";de.registerClass(s2);var r2=class extends ms{constructor(e){super({scale:2,mode:"fanIn",distribution:"normal",seed:e==null?null:e.seed})}getClassName(){return ms.className}};r2.className="HeNormal";de.registerClass(r2);var a2=class extends ms{constructor(e){super({scale:2,mode:"fanIn",distribution:"uniform",seed:e==null?null:e.seed})}getClassName(){return ms.className}};a2.className="HeUniform";de.registerClass(a2);var o2=class extends ms{constructor(e){super({scale:1,mode:"fanIn",distribution:"normal",seed:e==null?null:e.seed})}getClassName(){return ms.className}};o2.className="LeCunNormal";de.registerClass(o2);var i2=class extends ms{constructor(e){super({scale:1,mode:"fanIn",distribution:"uniform",seed:e==null?null:e.seed})}getClassName(){return ms.className}};i2.className="LeCunNormal";de.registerClass(i2);var r5=class extends hr{constructor(e){if(super(),this.DEFAULT_GAIN=1,this.gain=e.gain==null?this.DEFAULT_GAIN:e.gain,this.seed=e.seed,this.seed!=null)throw new qe("Random seed is not implemented for Orthogonal Initializer yet.")}apply(e,t){return Y(()=>{if(e.length<2)throw new qe("Shape must be at least 2D.");e[0]*e[1]>2e3&&console.warn(`Orthogonal initializer is being called on a matrix with more than 2000 (${e[0]*e[1]}) elements: Slowness may result.`);let n=e[0]>e[1]?[e[1],e[0]]:e,s=e2(n,0,1,"float32"),r=jA.gramSchmidt(s);return e[0]>e[1]&&(r=et(r)),L(this.gain,r)})}getConfig(){return{gain:this.gain,seed:this.seed}}};r5.className="Orthogonal";de.registerClass(r5);var Mv={constant:"Constant",glorotNormal:"GlorotNormal",glorotUniform:"GlorotUniform",heNormal:"HeNormal",heUniform:"HeUniform",identity:"Identity",leCunNormal:"LeCunNormal",leCunUniform:"LeCunUniform",ones:"Ones",orthogonal:"Orthogonal",randomNormal:"RandomNormal",randomUniform:"RandomUniform",truncatedNormal:"TruncatedNormal",varianceScaling:"VarianceScaling",zeros:"Zeros"};function zv(e,t={}){return wh(e,de.SerializationMap.getMap().classNameMap,t,"initializer")}function Vt(e){return KA(e)}function Ot(e){if(typeof e=="string"){let t=e in Mv?Mv[e]:e;if(t==="GlorotNormal")return new s2;if(t==="GlorotUniform")return new n2;if(t==="HeNormal")return new r2;if(t==="HeUniform")return new a2;if(t==="LeCunNormal")return new o2;if(t==="LeCunUniform")return new i2;{let n={};return n.className=t,n.config={},zv(n)}}else return e instanceof hr?e:zv(e)}function F3(e){return Array.isArray(e)&&Array.isArray(e[0])}function xm(e){return e.length===0?[]:Array.isArray(e[0])?e:[e]}function Xe(e){let t;if(Array.isArray(e)){if(e.length!==1)throw new j(`Expected Tensor length to be 1; got ${e.length}`);t=e[0]}else t=e;return t}function xt(e){if(Array.isArray(e)&&Array.isArray(e[0])){if(e.length===1)return e=e,e[0];throw new j(`Expected exactly 1 Shape; got ${e.length}`)}else return e}function bm(e){let t=0;for(let n of e)n.shape.length===0?t+=1:t+=n.shape.reduce((s,r)=>s*r);return t}var Lv="Variable",k8=class{constructor(e,t="float32",n=Lv,s=!0,r=null){this.dtype=t==null?"float32":t,this.shape=e.shape,this.id=g8(),n=n==null?Lv:n,this.originalName=A8(n),this.name=x8(this.originalName),this.trainable_=s,this.constraint=r,this.val=BA(e,this.trainable_,this.name,this.dtype)}read(){return this.assertNotDisposed(),this.val}write(e){return this.assertNotDisposed(),BV(this.val,e),this.val.id!==e.id&&(this.val.assign(e),this.constraint!=null&&this.val.assign(this.constraint.apply(this.val))),this}dispose(){this.assertNotDisposed(),this.val.dispose()}assertNotDisposed(){if(this.val.isDisposed)throw new Error(`LayersVariable ${this.name} is already disposed.`)}get trainable(){return this.trainable_}set trainable(e){this.trainable_=e,this.val.trainable=e}};function BV(e,t){if(e.shape.toString()!==t.shape.toString())throw new Error("Shape mismatch: "+JSON.stringify(e.shape)+" vs. "+JSON.stringify(t.shape))}function O3(e){return e.map(t=>t.read())}function a5(e){e.forEach(t=>{t[0].write(t[1])})}var rn=class{constructor(e){this.dtype=e.dtype,this.shape=e.shape,e.shape!=null?this.ndim=e.shape.length:this.ndim=e.ndim,this.maxNDim=e.maxNDim,this.minNDim=e.minNDim,this.axes=e.axes||{}}},Tr=class{constructor(e,t,n,s,r,a,o){this.dtype=e,this.shape=t,this.sourceLayer=n,this.inputs=s,this.callArgs=r,this.outputTensorIndex=o,this.id=g8(),a!=null&&(this.originalName=A8(a),this.name=x8(this.originalName)),this.rank=t.length}},WV=0,l2=class{constructor(e,t){this.callArgs=t,this.id=WV++,this.outboundLayer=e.outboundLayer,this.inboundLayers=e.inboundLayers,this.nodeIndices=e.nodeIndices,this.tensorIndices=e.tensorIndices,this.inputTensors=e.inputTensors,this.outputTensors=e.outputTensors,this.inputMasks=e.inputMasks,this.outputMasks=e.outputMasks,this.inputShapes=e.inputShapes,this.outputShapes=e.outputShapes;for(let n of e.inboundLayers)n!=null&&n.outboundNodes.push(this);e.outboundLayer.inboundNodes.push(this)}getConfig(){let e=[];for(let t of this.inboundLayers)t!=null?e.push(t.name):e.push(null);return{outboundLayer:this.outboundLayer?this.outboundLayer.name:null,inboundLayers:e,nodeIndices:this.nodeIndices,tensorIndices:this.tensorIndices}}},VV=0,ut=class extends de.Serializable{constructor(e={}){super(),this._callHook=null,this._addedWeightNames=[],this._stateful=!1,this.id=VV++,this.activityRegularizer=null,this.inputSpec=null,this.supportsMasking=!1,this._trainableWeights=[],this._nonTrainableWeights=[],this._losses=[],this._updates=[],this._built=!1,this.inboundNodes=[],this.outboundNodes=[];let t=e.name;if(!t){let n=this.getClassName();t=ua(n)+"_"+J0(n)}if(this.name=t,this.trainable_=e.trainable==null?!0:e.trainable,e.inputShape!=null||e.batchInputShape!=null){let n;if(e.batchInputShape!=null)n=e.batchInputShape;else if(e.inputShape!=null){let r=null;e.batchSize!=null&&(r=e.batchSize),n=[r].concat(e.inputShape)}this.batchInputShape=n;let s=e.dtype;s==null&&(s=e.inputDType),s==null&&(s="float32"),this.dtype=s}e.weights!=null?this.initialWeights=e.weights:this.initialWeights=null,this._refCount=null,this.fastWeightInitDuringBuild=!1}static nodeKey(e,t){return e.name+"_ib-"+t.toString()}getNodeAtIndex(e,t){if(this.inboundNodes.length===0)throw new Cr(`The layer has never been called and thus has no defined ${t}.`);if(this.inboundNodes.length<=e)throw new j(`Asked to get ${t} at node ${e}, but the layer has only ${this.inboundNodes.length} inbound nodes.`);return this.inboundNodes[e]}getInputAt(e){return ps(this.getNodeAtIndex(e,"input").inputTensors)}getOutputAt(e){return ps(this.getNodeAtIndex(e,"output").outputTensors)}get input(){if(this.inboundNodes.length>1)throw new la(`Layer ${this.name} has multiple inbound nodes, hence the notion of "layer input" is ill-defined. Use \`getInputAt(nodeIndex)\` instead.`);if(this.inboundNodes.length===0)throw new la(`Layer ${this.name} is not connected, no input to return.`);return ps(this.getNodeAtIndex(0,"input").inputTensors)}get output(){if(this.inboundNodes.length===0)throw new la(`Layer ${this.name} has no inbound nodes.`);if(this.inboundNodes.length>1)throw new la(`Layer ${this.name} has multiple inbound nodes, hence the notion of "layer output" is ill-defined. Use \`getOutputAt(nodeIndex)\` instead.`);return ps(this.getNodeAtIndex(0,"output").outputTensors)}get losses(){return this._losses}calculateLosses(){return this.losses.map(e=>e())}get updates(){return this._updates}get built(){return this._built}set built(e){this._built=e}get trainable(){return this.trainable_}set trainable(e){this._trainableWeights.forEach(t=>t.trainable=e),this.trainable_=e}get trainableWeights(){return this.trainable_?this._trainableWeights.filter(e=>e.trainable):[]}set trainableWeights(e){this._trainableWeights=e}get nonTrainableWeights(){return this.trainable?this._trainableWeights.filter(e=>!e.trainable).concat(this._nonTrainableWeights):this._trainableWeights.concat(this._nonTrainableWeights)}set nonTrainableWeights(e){this._nonTrainableWeights=e}get weights(){return this.trainableWeights.concat(this.nonTrainableWeights)}get stateful(){return this._stateful}resetStates(){if(!this.stateful)throw new Error("Cannot call the resetStates() method of a non-stateful Layer object.")}assertInputCompatibility(e){if(e=_t(e),this.inputSpec==null||this.inputSpec.length===0)return;let t=_t(this.inputSpec);if(e.length!==t.length)throw new j(`Layer ${this.name} expects ${t.length} inputs, but it received ${e.length} input tensors. Input received: ${e}`);for(let n=0;n<e.length;n++){let s=e[n],r=t[n];if(r==null)continue;let a=s.rank;if(r.ndim!=null&&a!==r.ndim)throw new j(`Input ${n} is incompatible with layer ${this.name}: expected ndim=${r.ndim}, found ndim=${a}`);if(r.maxNDim!=null&&a>r.maxNDim)throw new j(`Input ${n} is incompatible with layer ${this.name}: expected max_ndim=${r.maxNDim}, found ndim=${a}`);if(r.minNDim!=null&&a<r.minNDim)throw new j(`Input ${n} is incompatible with layer ${this.name}: expected min_ndim=${r.minNDim}, found ndim=${a}.`);if(r.dtype!=null&&s.dtype!==r.dtype)throw new j(`Input ${n} is incompatible with layer ${this.name} : expected dtype=${r.dtype}, found dtype=${s.dtype}.`);if(r.axes){let o=s.shape;for(let i in r.axes){let l=Number(i),u=r.axes[i],c=l>=0?o[l]:o[o.length+l];if(u!=null&&[u,null].indexOf(c)===-1)throw new j(`Input ${n} is incompatible with layer ${this.name}: expected axis ${l} of input shape to have value ${u} but got shape ${o}.`)}}if(r.shape!=null)for(let o=0;o<r.shape.length;++o){let i=r.shape[o],l=s.shape[o];if(i!=null&&l!=null&&i!==l)throw new j(`Input ${n} is incompatible with layer ${this.name}: expected shape=${r.shape}, found shape=${s.shape}.`)}}}call(e,t){return e}invokeCallHook(e,t){this._callHook!=null&&this._callHook(e,t)}setCallHook(e){this._callHook=e}clearCallHook(){this._callHook=null}apply(e,t){t=t||{},this.assertNotDisposed();let n=_t(e),s=!0;for(let a of n)if(!(a instanceof Tr)){s=!1;break}let r=!0;for(let a of n)if(a instanceof Tr){r=!1;break}if(s===r)throw new j("Arguments to apply() must be all SymbolicTensors or all Tensors");return Hi(this.name,()=>{if(!this.built){this.assertInputCompatibility(e);let a=[];for(let o of _t(e))a.push(o.shape);this.build(ps(a)),this.built=!0,this.initialWeights&&this.setWeights(this.initialWeights),this._refCount===null&&r&&(this._refCount=1)}if(this.assertInputCompatibility(e),r){let a=this.call(e,t),o=_t(a),i=[];for(let l of o)n.indexOf(l)!==-1&&(l=l.clone()),i.push(l);if(a=ps(i),this.activityRegularizer!=null)throw new qe("Layer invocation in the presence of activity regularizer(s) is not supported yet.");return a}else{let a=UV(e),o=this.computeOutputShape(a),i,l=GV(e);if(this.warnOnIncompatibleInputShape(Array.isArray(e)?a[0]:a),o!=null&&o.length>0&&Array.isArray(o[0])?i=o.map((u,c)=>new Tr(l,u,this,_t(e),t,this.name,c)):i=new Tr(l,o,this,_t(e),t,this.name),this.addInboundNode(e,i,null,null,a,o,t),this._refCount++,this.activityRegularizer!=null)throw new qe("Layer invocation in the presence of activity regularizer(s) is not supported yet.");return i}})}warnOnIncompatibleInputShape(e){if(this.batchInputShape!=null)if(e.length!==this.batchInputShape.length)console.warn(`The rank of the input tensor provided (shape: ${JSON.stringify(e)}) does not match that of the batchInputShape (${JSON.stringify(this.batchInputShape)}) of the layer ${this.name}`);else{let t=!1;this.batchInputShape.forEach((n,s)=>{n!=null&&e[s]!=null&&e[s]!==n&&(t=!0)}),t&&console.warn(`The shape of the input tensor (${JSON.stringify(e)}) does not match the expectation of layer ${this.name}: ${JSON.stringify(this.batchInputShape)}`)}}get outputShape(){if(this.inboundNodes==null||this.inboundNodes.length===0)throw new la(`The layer ${this.name} has never been called and thus has no defined output shape.`);let e=[];for(let t of this.inboundNodes){let n=JSON.stringify(t.outputShapes);e.indexOf(n)===-1&&e.push(n)}if(e.length===1){let t=this.inboundNodes[0].outputShapes;return Array.isArray(t)&&Array.isArray(t[0])&&t.length===1?t[0]:t}else throw new la(`The layer ${this.name} has multiple inbound nodes with different output shapes. Hence the notion of "output shape" is ill-defined for the layer.`)}countParams(){if(!this.built)throw new Cr(`You tried to call countParams() on ${this.name}, but the layer is not built yet. Build it first by calling build(batchInputShape).`);return bm(this.weights)}build(e){this.built=!0}getWeights(e=!1){return O3(e?this.trainableWeights:this.weights)}setWeights(e){Y(()=>{let t=this.weights;if(t.length!==e.length)throw new j(`You called setWeights(weights) on layer "${this.name}" with a weight list of length ${e.length}, but the layer was expecting ${t.length} weights. Provided weights: ${e}...`);if(t.length===0)return;let n=[],s=O3(t);for(let r=0;r<s.length;++r){let a=s[r],o=t[r],i=e[r];if(!v.arraysEqual(a.shape,i.shape))throw new j(`Layer weight shape ${a.shape} not compatible with provided weight shape ${i.shape}`);n.push([o,i])}a5(n)})}addWeight(e,t,n,s,r,a,o,i){if(this._addedWeightNames.indexOf(e)!==-1)throw new j(`Duplicate weight name ${e} for layer ${this.name}`);this._addedWeightNames.push(e),n==null&&(n="float32"),this.fastWeightInitDuringBuild&&(s=i!=null?i():Ot("zeros"));let l=s.apply(t,n),u=new k8(l,n,e,a,o);return l.dispose(),r!=null&&this.addLoss(()=>r.apply(u.read())),a==null&&(a=!0),a?this._trainableWeights.push(u):this._nonTrainableWeights.push(u),u}setFastWeightInitDuringBuild(e){this.fastWeightInitDuringBuild=e}addLoss(e){e==null||Array.isArray(e)&&e.length===0||(e=_t(e),this._losses!==void 0&&this._losses!==null&&this.losses.push(...e))}computeOutputShape(e){return e}computeMask(e,t){if(!this.supportsMasking){if(t!=null)if(Array.isArray(t))t.forEach(n=>{if(n!=null)throw new TypeError(`Layer ${this.name} does not support masking, but was passed an inputMask.`)});else throw new TypeError(`Layer ${this.name} does not support masking, but was passed an inputMask.`);return null}return t}addInboundNode(e,t,n,s,r,a,o=null){let i=_t(e);t=_t(t),n=_t(n),s=_t(s),r=xm(r),a=xm(a);let l=[],u=[],c=[];for(let p of i)l.push(p.sourceLayer),u.push(p.nodeIndex),c.push(p.tensorIndex);new l2({outboundLayer:this,inboundLayers:l,nodeIndices:u,tensorIndices:c,inputTensors:i,outputTensors:t,inputMasks:n,outputMasks:s,inputShapes:r,outputShapes:a},o);for(let p=0;p<t.length;p++)t[p].sourceLayer=this,t[p].nodeIndex=this.inboundNodes.length-1,t[p].tensorIndex=p}getConfig(){let e={name:this.name,trainable:this.trainable};return this.batchInputShape!=null&&(e.batchInputShape=this.batchInputShape),this.dtype!=null&&(e.dtype=this.dtype),e}disposeWeights(){return this.weights.forEach(e=>e.dispose()),this.weights.length}assertNotDisposed(){if(this._refCount===0)throw new Error(`Layer '${this.name}' is already disposed.`)}dispose(){if(!this.built)throw new Error(`Cannot dispose Layer ${this.name} because it has not been built yet.`);if(this._refCount===null)throw new Error(`Cannot dispose Layer ${this.name} because it has not been used yet.`);this.assertNotDisposed();let e=0;return--this._refCount===0&&(e=this.disposeWeights()),{refCountAfterDispose:this._refCount,numDisposedVariables:e}}};function UV(e){e=_t(e);let t=[];for(let n of e)t.push(n.shape);return ps(t)}function GV(e){return"float32"}function I8(e,t,n){if((t==null||n!=null&&n>0)&&(t=e.sourceLayer,n=e.nodeIndex),t.inboundNodes.length===0)return[e];{let s=t.inboundNodes[n];if(s.inboundLayers.length===0)return s.inputTensors;{let r=[];for(let a=0;a<s.inboundLayers.length;a++){let o=s.inputTensors[a],i=s.inboundLayers[a],l=s.nodeIndices[a],u=I8(o,i,l);for(let c of u)r.indexOf(c)===-1&&r.push(c)}return r}}}var Yc=class extends ut{constructor(e){if(super({dtype:e.dtype,name:e.name!=null?e.name:J0("input").toString()}),e.batchSize==null&&(e.batchSize=null),e.sparse==null&&(e.sparse=!1),this.trainable=!1,this.built=!0,this.sparse=e.sparse,e.inputShape!=null&&e.batchInputShape!=null)throw new j("Only provide the inputShape OR batchInputShape argument to inputLayer, not both at the same time.");let t=e.batchInputShape;if(t==null){if(e.inputShape==null)throw new j("An InputLayer should be passed either a `batchInputShape` or an `inputShape`.");t=[e.batchSize].concat(e.inputShape)}else if(e.batchSize!=null)throw new j("Cannot specify batchSize if batchInputShape is specified when creating an InputLayer.");let n=e.dtype||"float32";this.batchInputShape=t,this.dtype=n,this.inputSpec=[{shape:t}];let s=new Tr(this.dtype,this.batchInputShape,this,[],{},this.name);s.nodeIndex=0,s.tensorIndex=0,new l2({outboundLayer:this,inboundLayers:[],nodeIndices:[],tensorIndices:[],inputTensors:[s],outputTensors:[s],inputMasks:[null],outputMasks:[null],inputShapes:[t],outputShapes:[t]})}apply(e,t){throw new j(`Cannot pass any input to an InputLayer's apply() method. InputLayer name: ${this.name}`)}dispose(){return{refCountAfterDispose:this._refCount,numDisposedVariables:0}}getConfig(){return{batchInputShape:this.batchInputShape,dtype:this.dtype,sparse:this.sparse,name:this.name}}};Yc.className="InputLayer";de.registerClass(Yc);function S8(e){if(e.batchShape==null&&e.shape==null)throw new Error("Please provide to Input either a `shape` or a `batchShape` argument. Note that `shape` does not include the batch dimension.");if(e.batchShape!=null&&e.shape!=null)throw new j("Please provide either a `shape` or `batchShape` argument to Input, but not both.");let t=e.batchShape;e.shape!=null&&t==null&&(t=[null].concat(e.shape));let n=e.dtype;return n==null&&(n="float32"),new Yc({batchInputShape:t,name:e.name,dtype:n,sparse:e.sparse}).inboundNodes[0].outputTensors[0]}function HV(e,t){if(e.dtype==null||e.dtype===t.dtype)return t;try{return ye(t,e.dtype)}catch(n){throw new j(`The dtype of the feed (${t.dtype}) can not be cast to the dtype of the key '${e.name}' (${e.dtype}).`)}}var Vi=class{constructor(e){if(this.id2Value={},this.id2Mask={},this.name2Id={},e instanceof Vi)for(let t in e.id2Value)this.id2Value[t]=e.id2Value[t],t in e.id2Mask&&(this.id2Mask[t]=e.id2Mask[t]);else{if(e==null)return;for(let t of e)this.add(t.key,t.value)}}add(e,t,n){if(this.id2Value[e.id]==null)this.id2Value[e.id]=HV(e,t),this.name2Id[e.name]=e.id,n!=null&&(this.id2Mask[e.id]=n);else throw new j(`Duplicate key: name=${e.name}, id=${e.id}`);return this}addFeed(e){this.add(e.key,e.value)}hasKey(e){return this.id2Value[e.id]!=null}names(){return Object.keys(this.name2Id)}getValue(e){if(e instanceof Tr){if(this.id2Value[e.id]==null)throw new j(`Nonexistent key: ${e.name}`);return this.id2Value[e.id]}else{let t=this.name2Id[e];if(t==null)throw new j(`Feed dict has no SymbolicTensor name: ${e}`);return this.id2Value[t]}}getMask(e){if(e instanceof Tr){if(this.id2Value[e.id]==null)throw new j(`Nonexistent key: ${e.name}`);return this.id2Mask[e.id]}else{let t=this.name2Id[e];if(t==null)throw new j(`Feed dict has no SymbolicTensor name: ${e}`);return this.id2Mask[t]}}disposeMasks(){this.id2Mask!=null&&ee(this.id2Mask)}},vm=new h8,wm=new h8;function jV(e){vm!=null&&vm.setMaxEntries(e),wm!=null&&wm.setMaxEntries(e)}function ap(e,t,n,s){let r=n==null?!1:n.training,a=Array.isArray(e),o=a?e:[e],i=o.map(f=>f.name),l=[],u=t.names();for(let f of i)u.indexOf(f)!==-1?l.push(t.getValue(f)):l.push(null);s!=null&&(s.maxNumTensors=-1/0,s.minNumTensors=1/0);let c=i.join(",")+"|"+t.names().sort().join(","),p=vm.get(c),d;if(p==null){let f=qV(o,t);p=f.sorted,d=f.recipientCounts,vm.put(c,p),wm.put(c,d)}d={},r||Object.assign(d,wm.get(c));let h=new Vi(t);for(let f=0;f<p.length;++f){if(s!=null){let R=ym().numTensors;R>s.maxNumTensors&&(s.maxNumTensors=R),R<s.minNumTensors&&(s.minNumTensors=R)}let m=p[f],g=m.sourceLayer;if(g instanceof Yc)continue;let y=[],x=[],A=[],b=!1;for(let R of m.inputs){let $=h.getValue(R),_=h.getMask(R);y.push($),x.push(_),_!=null&&(b=!0),r||(d[R.name]--,d[R.name]===0&&!t.hasKey(R)&&i.indexOf(R.name)===-1&&!$.isDisposed&&R.sourceLayer.stateful!==!0&&A.push($))}b&&(n=n||{},n.mask=x[0]);let w=_t(g.apply(y,n)),k=null;g.supportsMasking&&(k=g.computeMask(y,x));let S=KV(m),E=Array.isArray(S)?S:[S];for(let R=0;R<E.length;++R){h.hasKey(E[R])||h.add(E[R],w[R],Array.isArray(k)?k[0]:k);let $=i.indexOf(E[R].name);$!==-1&&(l[$]=w[R])}r||ee(A)}return h.disposeMasks(),a?l:l[0]}function qV(e,t){v.assert(e!=null&&e.length>0,()=>"Expected at least one fetch, got none");let n=[],s={};if(e.length===1){let r=Bv(e[0],t);n=r.sorted,s=r.recipientMap}else{let r=new Set;for(let a of e){let{sorted:o,recipientMap:i}=Bv(a,t);for(let l of o)r.has(l.name)||(n.push(l),r.add(l.name));for(let l in i)s[l]==null&&(s[l]=new Set),i[l].forEach(u=>s[l].add(u))}}return{sorted:n,recipientCounts:XV(s)}}function XV(e){let t={};for(let n in e)t[n]=e[n].size;return t}function Bv(e,t){let n=new Set,s=[],r={};for(let i of t.names())n.add(i);let a=[],o=[];for(a.push(e);a.length>0;){let i=a[a.length-1];if(n.has(i.name)){a.pop();continue}let l=o[o.length-1]===a.length-1;if(i.inputs.length===0||l)a.pop(),s.push(i),n.add(i.name),l&&o.pop();else{o.push(a.length-1);for(let u of i.inputs)r[u.name]==null&&(r[u.name]=new Set),r[u.name].add(i.name),!n.has(u.name)&&a.push(u)}}return{sorted:s,recipientMap:r}}function KV(e){let t;if(e.sourceLayer.inboundNodes.length===1)t=e.sourceLayer.output;else{let n=null;for(let s=0;s<e.sourceLayer.inboundNodes.length;++s)for(let r of e.sourceLayer.inboundNodes[s].outputTensors)if(r.id===e.id){n=s;break}t=e.sourceLayer.getOutputAt(n)}return t}var ZV=q();ZV.registerFlag("TOPOLOGICAL_SORT_CACHE_MAX_ENTRIES",()=>100,jV);var C8={};Ve(C8,{maxNorm:()=>YV,minMaxNorm:()=>eU,nonNeg:()=>QV,unitNorm:()=>JV});function o5(e,t){return Y(()=>Nn(ke(L(e,e),t,!0)))}var Ch=class extends de.Serializable{getConfig(){return{}}},i5=class extends Ch{constructor(e){super(),this.defaultMaxValue=2,this.defaultAxis=0,this.maxValue=e.maxValue!=null?e.maxValue:this.defaultMaxValue,this.axis=e.axis!=null?e.axis:this.defaultAxis}apply(e){return Y(()=>{let t=o5(e,this.axis),n=fs(t,0,this.maxValue);return L(e,he(n,ce(hn(),t)))})}getConfig(){return{maxValue:this.maxValue,axis:this.axis}}};i5.className="MaxNorm";de.registerClass(i5);var l5=class extends Ch{constructor(e){super(),this.defaultAxis=0,this.axis=e.axis!=null?e.axis:this.defaultAxis}apply(e){return Y(()=>he(e,ce(hn(),o5(e,this.axis))))}getConfig(){return{axis:this.axis}}};l5.className="UnitNorm";de.registerClass(l5);var u5=class extends Ch{apply(e){return Pr(e)}};u5.className="NonNeg";de.registerClass(u5);var c5=class extends Ch{constructor(e){super(),this.defaultMinValue=0,this.defaultMaxValue=1,this.defaultRate=1,this.defaultAxis=0,this.minValue=e.minValue!=null?e.minValue:this.defaultMinValue,this.maxValue=e.maxValue!=null?e.maxValue:this.defaultMaxValue,this.rate=e.rate!=null?e.rate:this.defaultRate,this.axis=e.axis!=null?e.axis:this.defaultAxis}apply(e){return Y(()=>{let t=o5(e,this.axis),n=ce(L(this.rate,fs(t,this.minValue,this.maxValue)),L(1-this.rate,t));return L(e,he(n,ce(hn(),t)))})}getConfig(){return{minValue:this.minValue,maxValue:this.maxValue,rate:this.rate,axis:this.axis}}};c5.className="MinMaxNorm";de.registerClass(c5);var Wv={maxNorm:"MaxNorm",minMaxNorm:"MinMaxNorm",nonNeg:"NonNeg",unitNorm:"UnitNorm"};function gn(e){return KA(e)}function Vv(e,t={}){return wh(e,de.SerializationMap.getMap().classNameMap,t,"constraint")}function yn(e){if(e==null)return null;if(typeof e=="string"){let n={className:e in Wv?Wv[e]:e,config:{}};return Vv(n)}else return e instanceof Ch?e:Vv(e)}function YV(e){return new i5(e)}function JV(e){return new l5(e)}function QV(){return new u5}function eU(e){return new c5(e)}var T8={};Ve(T8,{constant:()=>sU,glorotNormal:()=>cU,glorotUniform:()=>uU,heNormal:()=>dU,heUniform:()=>pU,identity:()=>iU,leCunNormal:()=>hU,leCunUniform:()=>fU,ones:()=>nU,orthogonal:()=>mU,randomNormal:()=>aU,randomUniform:()=>rU,truncatedNormal:()=>oU,varianceScaling:()=>lU,zeros:()=>tU});function tU(){return new JA}function nU(){return new t2}function sU(e){return new QA(e)}function rU(e){return new e5(e)}function aU(e){return new t5(e)}function oU(e){return new n5(e)}function iU(e){return new s5(e)}function lU(e){return new ms(e)}function uU(e){return new n2(e)}function cU(e){return new s2(e)}function dU(e){return new r2(e)}function pU(e){return new a2(e)}function hU(e){return new o2(e)}function fU(e){return new i2(e)}function mU(e){return new r5(e)}var N8={};Ve(N8,{Layer:()=>ut,RNN:()=>Qr,RNNCell:()=>_h,activation:()=>WG,add:()=>ZG,alphaDropout:()=>$H,average:()=>YG,averagePooling1d:()=>yx,averagePooling2d:()=>Ax,averagePooling3d:()=>xx,avgPool1d:()=>oH,avgPool2d:()=>lH,avgPool3d:()=>cH,avgPooling1d:()=>iH,avgPooling2d:()=>uH,avgPooling3d:()=>dH,batchNormalization:()=>sH,bidirectional:()=>SH,concatenate:()=>JG,conv1d:()=>DG,conv2d:()=>$G,conv2dTranspose:()=>PG,conv3d:()=>FG,conv3dTranspose:()=>OG,convLstm2d:()=>vH,convLstm2dCell:()=>wH,cropping2D:()=>zG,dense:()=>VG,depthwiseConv2d:()=>BG,dot:()=>nH,dropout:()=>UG,elu:()=>CG,embedding:()=>KG,flatten:()=>HG,gaussianDropout:()=>DH,gaussianNoise:()=>_H,globalAveragePooling1d:()=>pH,globalAveragePooling2d:()=>hH,globalMaxPool1d:()=>TH,globalMaxPool2d:()=>NH,globalMaxPooling1d:()=>xk,globalMaxPooling2d:()=>bk,gru:()=>mH,gruCell:()=>gH,input:()=>j8,inputLayer:()=>SG,layerNormalization:()=>rH,leakyReLU:()=>NG,lstm:()=>yH,lstmCell:()=>AH,masking:()=>PH,maxPool1d:()=>EH,maxPool2d:()=>RH,maxPooling1d:()=>vk,maxPooling2d:()=>wk,maxPooling3d:()=>fH,maximum:()=>QG,minimum:()=>eH,multiply:()=>tH,permute:()=>XG,prelu:()=>EG,reLU:()=>TG,repeatVector:()=>jG,reshape:()=>qG,rnn:()=>kH,separableConv2d:()=>MG,simpleRNN:()=>xH,simpleRNNCell:()=>bH,softmax:()=>RG,spatialDropout1d:()=>GG,stackedRNNCells:()=>IH,thresholdedReLU:()=>_G,timeDistributed:()=>CH,upSampling2d:()=>LG,zeroPadding2d:()=>aH});async function Va(e){if(e==null)return;let t=[],n=[],s=[];for(let r in e){let a=e[r];if(typeof a!="number"){let o=a;t.push(o.data()),n.push(r),s.push(o)}}if(t.length>0){let r=await Promise.all(t);for(let a=0;a<r.length;++a)e[n[a]]=r[a][0];ee(s)}}function E8(e){if(e!=null)for(let t in e){let n=e[t];typeof n!="number"&&n.dispose()}}var Uv;(function(e){e[e.SILENT=0]="SILENT",e[e.VERBOSE=1]="VERBOSE"})(Uv||(Uv={}));var gU=125,nc=class{constructor(){this.validationData=null}setParams(e){this.params=e}async onEpochBegin(e,t){}async onEpochEnd(e,t){}async onBatchBegin(e,t){}async onBatchEnd(e,t){}async onTrainBegin(e){}async onTrainEnd(e){}setModel(e){}},R8=class{constructor(e,t=10){e==null&&(e=[]),this.callbacks=e,this.queueLength=t}append(e){this.callbacks.push(e)}setParams(e){for(let t of this.callbacks)t.setParams(e)}setModel(e){for(let t of this.callbacks)t.setModel(e)}async onEpochBegin(e,t){t==null&&(t={});for(let n of this.callbacks)await n.onEpochBegin(e,t)}async onEpochEnd(e,t){t==null&&(t={});for(let n of this.callbacks)await n.onEpochEnd(e,t)}async onBatchBegin(e,t){t==null&&(t={});for(let n of this.callbacks)await n.onBatchBegin(e,t)}async onBatchEnd(e,t){t==null&&(t={});for(let n of this.callbacks)await n.onBatchEnd(e,t)}async onTrainBegin(e){e==null&&(e={});for(let t of this.callbacks)await t.onTrainBegin(e)}async onTrainEnd(e){e==null&&(e={});for(let t of this.callbacks)await t.onTrainEnd(e)}},yU=class extends nc{constructor(){super()}async onEpochBegin(e){this.seen=0,this.totals={}}async onBatchEnd(e,t){t==null&&(t={});let n=t.size==null?0:t.size;this.seen+=n;for(let s in t){let r=t[s];if(typeof r=="number")this.totals.hasOwnProperty(s)||(this.totals[s]=0),this.totals[s]=this.totals[s]+r*n;else{let a;s in this.totals?a=this.totals[s]:this.totals[s]=0;let o=Y(()=>ce(this.totals[s],L(r,n)));this.totals[s]=o,a!=null&&a.dispose()}}}async onEpochEnd(e,t){if(t!=null)for(let n of this.params.metrics)this.totals[n]!=null&&(typeof this.totals[n]=="number"?t[n]=this.totals[n]/this.seen:Y(()=>{let s=L(he(1,this.seen),this.totals[n]);t[n]=s,this.totals[n].dispose(),bn(t[n])}))}},_8=class extends nc{async onTrainBegin(e){this.epoch=[],this.history={}}async onEpochEnd(e,t){t==null&&(t={}),this.epoch.push(e);for(let n in t)this.history[n]==null&&(this.history[n]=[]),this.history[n].push(t[n])}async syncData(){let e=[],t=[],n=[];for(let r in this.history){let a=this.history[r];for(let o=0;o<a.length;++o)if(typeof a[o]!="number"){let i=a[o];e.push(i.data()),t.push(r),n.push(o)}}let s=await Promise.all(e);for(let r=0;r<s.length;++r)this.history[t[r]][n[r]].dispose(),this.history[t[r]][n[r]]=s[r][0]}},D8=class extends nc{constructor(e,t){if(super(),this.currentEpoch=0,this.nowFunc=e.nowFunc,this.nextFrameFunc=e.nextFrameFunc||qA,this.yieldEvery=t||"auto",this.yieldEvery==="auto"&&(this.yieldEvery=gU),this.yieldEvery==="never"&&e.onYield!=null)throw new Error("yieldEvery is `never` but you provided an `onYield` callback. Either change `yieldEvery` or remove the callback");v.isNumber(this.yieldEvery)&&(this.maybeWait=AV(this.maybeWait.bind(this),this.yieldEvery,this.nowFunc)),this.trainBegin=e.onTrainBegin,this.trainEnd=e.onTrainEnd,this.epochBegin=e.onEpochBegin,this.epochEnd=e.onEpochEnd,this.batchBegin=e.onBatchBegin,this.batchEnd=e.onBatchEnd,this.yield=e.onYield}async maybeWait(e,t,n){let s=[];this.yield!=null&&(await Va(n),s.push(this.yield(e,t,n))),s.push(this.nextFrameFunc()),await Promise.all(s)}async onEpochBegin(e,t){this.currentEpoch=e,this.epochBegin!=null&&(await Va(t),await this.epochBegin(e,t))}async onEpochEnd(e,t){let n=[];this.epochEnd!=null&&(await Va(t),n.push(this.epochEnd(e,t))),this.yieldEvery==="epoch"&&n.push(this.nextFrameFunc()),await Promise.all(n)}async onBatchBegin(e,t){this.batchBegin!=null&&(await Va(t),await this.batchBegin(e,t))}async onBatchEnd(e,t){let n=[];this.batchEnd!=null&&(await Va(t),n.push(this.batchEnd(e,t))),this.yieldEvery==="batch"?n.push(this.nextFrameFunc()):v.isNumber(this.yieldEvery)&&n.push(this.maybeWait(this.currentEpoch,e,t)),await Promise.all(n)}async onTrainBegin(e){this.trainBegin!=null&&(await Va(e),await this.trainBegin(e))}async onTrainEnd(e){this.trainEnd!=null&&(await Va(e),await this.trainEnd(e))}};function $8(e,t){return e==null&&(e={}),e instanceof nc?[e]:Array.isArray(e)&&e[0]instanceof nc?e:_t(e).map(s=>new D8(s,t))}var ar=class{constructor(){}static registerCallbackConstructor(e,t){v.assert(e>=0&&Number.isInteger(e),()=>`Verbosity level is expected to be an integer >= 0, but got ${e}`),ar.checkForDuplicate(t),ar.constructors[e]==null&&(ar.constructors[e]=[]),ar.constructors[e].push(t)}static checkForDuplicate(e){for(let t in ar.constructors)ar.constructors[+t].forEach(s=>{if(s===e)throw new j("Duplicate callback constructor.")})}static clear(){ar.constructors={}}static createCallbacks(e){let t=[];for(let n in ar.constructors){let s=+n;e>=s&&t.push(...ar.constructors[s])}return t.map(n=>new n)}};ar.constructors={};function P8(e,t,n,s,r,a,o,i,l){let u=new _8,c=[new yU,...ar.createCallbacks(t)];e!=null&&c.push(...e),c.push(u);let p=new R8(c);return p.setParams({epochs:n,initialEpoch:s,samples:r,steps:a,batchSize:o,verbose:t,doValidation:i,metrics:l}),{callbackList:p,history:u}}function Er(e,t={},n=!1){return wh(e,de.SerializationMap.getMap().classNameMap,t,"layer",n)}function km(e,t){return Y(()=>{e.dtype!=="float32"&&(e=ye(e,"float32"));let n=ke(Ih(e),t,!0),s=Hc(n.shape,hn()),r=Nn(Jr(n,s));return he(e,r)})}function tu(e,t){return Y(()=>Bt(Ih(fe(t,e)),-1))}function u2(e,t){return Y(()=>Bt(tn(fe(t,e)),-1))}function Jc(e,t){return Y(()=>{let n=fe(e,t),s=fs(tn(e),hn(),Number.MAX_VALUE),r=tn(he(n,s));return L(100,Bt(r,-1))})}function AU(e,t){return Y(()=>{let n=fs(t,hn(),Number.MAX_VALUE),s=Ds(ce(1,n)),r=fs(e,hn(),Number.MAX_VALUE),a=Ds(ce(1,r));return Bt(Ih(fe(s,a)),-1)})}function xU(e,t){return Y(()=>{let n=Jr(0,fe(1,L(e,t)));return Bt(Ih(n),-1)})}function bU(e,t){return Y(()=>{let n=Jr(0,fe(1,L(e,t)));return Bt(n,-1)})}function vU(e,t){return Y(()=>{let n=ke(L(e,t),-1),s=mn(L(fe(1,e),t),-1);return Jr(0,ce(1,fe(s,n)))})}function wU(e,t){return Y(()=>{let n=Math.log(2),s=fe(t,e),r=fe(ce(s,Zl(L(-2,s))),n);return Bt(r,-1)})}function Np(e,t,n=!1){return Y(()=>{if(n)t=Jl(t);else{let s=ke(t,t.shape.length-1,!0);t=he(t,s)}return t=fs(t,hn(),1-hn()),Dt(ke(L(ye(e,"float32"),Ds(t)),t.shape.length-1))})}function Im(e,t,n=!1){return Y(()=>{let s=ye(jc(RV(e)),"int32");t=fs(t,hn(),1-hn());let r=t.shape,a=U(Zu(s,r[r.length-1]),r);return Np(a,t,n)})}function kU(e,t){if(!v.arraysEqual(e.shape,t.shape))throw new j(`logits and labels must have the same shape, but got shapes ${JSON.stringify(e.shape)} and ${JSON.stringify(t.shape)}`);return Y(()=>{let n=Pr(t),s=Dt(tn(t));return ce(fe(n,L(t,e)),ph(_s(s)))})}function c2(e,t){return Y(()=>{let n;return n=fs(t,hn(),1-hn()),n=Ds(he(n,fe(1,n))),Bt(kU(e,n),-1)})}function IU(e,t){return Y(()=>{let n=fs(e,hn(),1),s=fs(t,hn(),1);return ke(L(e,Ds(he(n,s))),-1)})}function SU(e,t){return Y(()=>{let n=Ds(ce(hn(),t));return Bt(fe(t,L(e,n)),-1)})}function d5(e,t){return Y(()=>{let n=km(e,-1),s=km(t,-1),r=L(n,s);return Dt(ke(r,-1))})}var Sm={meanSquaredError:tu,meanAbsoluteError:u2,meanAbsolutePercentageError:Jc,meanSquaredLogarithmicError:AU,squaredHinge:xU,hinge:bU,categoricalHinge:vU,logcosh:wU,categoricalCrossentropy:Np,sparseCategoricalCrossentropy:Im,binaryCrossentropy:c2,kullbackLeiblerDivergence:IU,poisson:SU,cosineProximity:d5};function c3(e){if(typeof e=="string"){if(e in Sm)return Sm[e];let t=`Unknown loss ${e}`;throw e.toLowerCase().includes("softmaxcrossentropy")&&(t=`Unknown loss ${e}. Use "categoricalCrossentropy" as the string name for tf.losses.softmaxCrossEntropy`),new j(t)}else return e}function p5(e,t){return Y(()=>{let n=L(.5,$s(t)),s=Q0(ys(t,n),e.dtype);return Bt(Rs(e,s),-1)})}function h5(e,t){return Y(()=>Q0(Rs(Es(e,-1),Es(t,-1)),"float32"))}function F8(e,t){return Y(()=>ye(ke(ur(Rs(e,1),Rs(t,1))),"float32"))}function CU(e,t){return Y(()=>ye(ke(ur(Rs(e,1),Rs(t,0))),"float32"))}function TU(e,t){return Y(()=>ye(ke(ur(Rs(e,0),Rs(t,1))),"float32"))}function O8(e,t){return Y(()=>{let n=F8(e,t),s=TU(e,t),r=ce(n,s);return ye(zn(ys(r,0),he(n,r),0),"float32")})}function NU(e,t){return Y(()=>{let n=F8(e,t),s=CU(e,t),r=ce(n,s);return ye(zn(ys(r,0),he(n,r),0),"float32")})}function M8(e,t){return c2(e,t)}function z8(e,t){return e.rank===t.rank&&(e=st(e,[e.rank-1])),t=Es(t,-1),t.dtype!==e.dtype&&(t=ye(t,e.dtype)),ye(Rs(e,t),"float32")}var EU=tu,RU=tu,_U=u2,DU=u2,$U=Jc,PU=Jc,f5=Np,FU=d5,L8=Im,Cm={binaryAccuracy:p5,categoricalAccuracy:h5,precision:O8,categoricalCrossentropy:f5,sparseCategoricalCrossentropy:L8,mse:EU,MSE:RU,mae:_U,MAE:DU,mape:$U,MAPE:PU,cosine:FU};function OU(e){if(typeof e=="string"&&e in Cm)return Cm[e];if(typeof e!="string"&&e!=null)return e;throw new j(`Unknown metric ${e}`)}function Gf(e){if(Vr(e!==null,`Unknown LossOrMetricFn ${e}`),typeof e=="string")return e;{let t;for(let n of Object.keys(Sm))if(Sm[n]===e){t=n;break}if(t!==void 0)return t;for(let n of Object.keys(Cm))if(Cm[n]===e){t=n;break}return t!==void 0?t:e.name}}function MU(e){let t={Adagrad:()=>Fi.adagrad(.01),Adadelta:()=>Fi.adadelta(1,.95,hn()),Adam:()=>Fi.adam(.001,.9,.999,hn()),Adamax:()=>Fi.adamax(.002,.9,.999,hn(),0),RMSProp:()=>Fi.rmsprop(.001,.9,0,hn()),SGD:()=>Fi.sgd(.01)};if(t.adagrad=t.Adagrad,t.adadelta=t.Adadelta,t.adam=t.Adam,t.adamax=t.Adamax,t.rmsprop=t.RMSProp,t.sgd=t.SGD,e in t)return t[e]();throw new j(`Unknown Optimizer ${e}`)}var Gv=1*1024*1024;function Hv(e,t,n=!1){if(e==null||typeof e!="object"||Object.getPrototypeOf(e)!==Object.prototype||!M3(e))throw new Error("User-defined metadata is expected to be a JSON object, but is not.");if(n){let s=JSON.stringify(e);s.length>Gv&&console.warn(`User-defined metadata of model "${t}" is too large in size (length=${s.length} when serialized). It is not recommended to store such large objects in user-defined metadata. Please make sure its serialized length is <= ${Gv}.`)}}function M3(e){if(e===null)return!0;if(typeof e=="object")if(Object.getPrototypeOf(e)===Object.prototype){let t=Object.keys(e);for(let n of t)if(typeof n!="string"||!M3(e[n]))return!1;return!0}else if(Array.isArray(e)){for(let t of e)if(!M3(t))return!1;return!0}else return!1;else{let t=typeof e;return t==="string"||t==="number"||t==="boolean"}}function zU(e,t,n,s=console.log){let r=BU(e),a=["Layer (type)","Input Shape","Output shape","Param #"];r?(t=t||90,n=n||[.32,.61,.89,1]):(t=t||115,n=n||[.24,.48,.7,.8,1]),n[n.length-1]<=1&&(n=n.map(c=>Math.floor(t*c)));let o;if(!r){a.push("Receives inputs"),o=[];for(let c in e.nodesByDepth)o.push(...e.nodesByDepth[c])}s("_".repeat(t)),Tm(a,n,s),s("=".repeat(t));let i=e.layers;for(let c=0;c<i.length;++c)r?WU(i[c],n,s):VU(i[c],n,o,s),s((c===i.length-1?"=":"_").repeat(t));e.checkTrainableWeightsConsistency();let l=LU(e),u=bm(e.nonTrainableWeights);s(`Total params: ${l+u}`),s(`Trainable params: ${l}`),s(`Non-trainable params: ${u}`),s("_".repeat(t))}function LU(e){let t;return e.collectedTrainableWeights!=null?t=bm(e.collectedTrainableWeights):t=bm(e.trainableWeights),t}function BU(e){let t=!0,n=[],s=[];for(let r in e.nodesByDepth)n.push(e.nodesByDepth[r]);for(let r of n){if(r.length>1||r.length===1&&r[0].inboundLayers.length>1){t=!1;break}s.push(...r)}if(t)for(let r of e.layers){let a=!1;for(let o of r.inboundNodes)if(s.indexOf(o)!==-1)if(a){t=!1;break}else a=!0;if(!t)break}return t}function Tm(e,t,n=console.log){let s="";for(let r=0;r<e.length;++r)r>0&&(s=s.slice(0,s.length-1)+" "),s+=e[r],s=s.slice(0,t[r]),s+=" ".repeat(t[r]-s.length);n(s)}function WU(e,t,n){let s,r;try{r=e.inboundNodes.map(l=>JSON.stringify(l.inputShapes)).join(",")}catch(l){r="multiple"}try{s=JSON.stringify(e.outputShape)}catch(l){s="multiple"}let a=e.name,o=e.getClassName(),i=[`${a} (${o})`,r,s,e.countParams().toString()];Tm(i,t,n)}function VU(e,t,n,s){let r,a;try{a=e.inboundNodes.map(p=>JSON.stringify(p.inputShapes)).join(",")}catch(p){a="multiple"}try{r=JSON.stringify(e.outputShape)}catch(p){r="multiple"}let o=[];for(let p of e.inboundNodes)if(!(n!=null&&n.length>0&&n.indexOf(p)===-1))for(let d=0;d<p.inboundLayers.length;++d){let h=p.inboundLayers[d].name,f=p.nodeIndices[d],m=p.tensorIndices[d];o.push(`${h}[${f}][${m}]`)}let i=e.name,l=e.getClassName(),u=o.length===0?"":o[0],c=[`${i} (${l})`,a,r,e.countParams().toString(),u];Tm(c,t,s);for(let p=1;p<o.length;++p)Tm(["","","","",o[p]],t,s)}function B8(e,t,n){return(e==="inboundNodes"||e==="outputLayers"||e==="inputLayers")&&t===0&&typeof n=="string"}function Ep(e,t){if(e===null)return null;if(typeof e=="string")return Li(e);if(typeof e=="number"||typeof e=="boolean")return e;if(e instanceof Array){let n=[],s=e.length;for(let r=0;r<s;++r){let a=e[r];B8(t,r,a)?n.push(a):n.push(Ep(a,t))}return n}else{let n={};for(let s of Object.keys(e)){let r=e[s];if(s==="name"&&typeof r=="string")n[s]=r;else{let a=Li(s);n[a]=Ep(r,a)}}return n}}function z3(e,t){if(e==null)return null;if(typeof e=="string")return ua(e);if(typeof e=="number"||typeof e=="boolean")return e;if(e instanceof Array){let n=[],s=e.length;for(let r=0;r<s;++r){let a=e[r];B8(t,r,a)?n.push(a):n.push(z3(a,t))}return n}else{let n={};for(let s of Object.keys(e)){let r=e[s],a=ua(s);(s==="name"||s==="className")&&typeof r=="string"?n[a]=r:n[a]=z3(r,s)}return n}}var m5="3.19.0",Wr=class extends ut{constructor(e){if(super({}),this.containerNodes=new Set,this.name=e.name,this.name==null){let y=this.getClassName().toLowerCase();this.name=J0(y)}if(this.supportsMasking=!1,this.trainable_=!0,Array.isArray(e.inputs)?this.inputs=e.inputs.slice():this.inputs=[e.inputs],Array.isArray(e.outputs)?this.outputs=e.outputs.slice():this.outputs=[e.outputs],Xa(this.inputs).length!==this.inputs.length)throw new j(`The list of inputs passed to the model is redundant. All inputs should only appear once. Found: ${this.inputs.map(y=>y.name)}`);Xa(this.outputs).length!==this.outputs.length&&console.warn(`The list of outputs passed to the model is redundant. All outputs should only appear once. Found: ${this.outputs.map(y=>y.name)}`),this.inputLayers=[],this.inputLayersNodeIndices=[],this.inputLayersTensorIndices=[],this.outputLayers=[],this.outputLayersNodeIndices=[],this.outputLayersTensorIndices=[],this.layers=[],this.internalContainerRefs=[];for(let y of this.outputs){let x=y.sourceLayer,A=y.nodeIndex,b=y.tensorIndex;this.outputLayers.push(x),this.outputLayersNodeIndices.push(A),this.outputLayersTensorIndices.push(b)}for(let y of this.inputs){let x=y.sourceLayer,A=y.nodeIndex,b=y.tensorIndex;Vr(A===0,"input layer has >1 nodes"),Vr(b===0,"input layer has >1 tensors"),this.inputLayers.push(x),this.inputLayersNodeIndices.push(A),this.inputLayersTensorIndices.push(b)}this.inputNames=[],this.outputNames=[],this.feedInputShapes=[],this.feedInputNames=[],this.feedOutputNames=[];for(let y=0;y<this.inputLayers.length;y++){let x=this.inputLayers[y];if(!(x instanceof Yc))throw new TypeError(`Input layers to a LayersModel must be InputLayer objects. Received inputs: ${e.inputs}. Input ${y} (0-based) originates from layer type ${x.getClassName()}.`);this.inputNames.push(x.name),this.feedInputShapes.push(x.batchInputShape),this.feedInputNames.push(x.name)}for(let y of this.outputLayers)this.outputNames.push(y.name);this.internalInputShapes=this.inputs.map(y=>y.shape),this.internalOutputShapes=this.outputs.map(y=>y.shape);let t={},n={},s={},r={},a={},o=[],i=(y,x,A,b,w,k)=>{(b==null||w==null||k==null)&&(b=y.sourceLayer,w=y.nodeIndex,k=y.tensorIndex);let S=b.inboundNodes[w];if(A.indexOf(S)!==-1)throw new Cr(`The tensor ${y.name} at layer "${b.name}" is part of a cycle.`);if(x.indexOf(S)!==-1)return;this.containerNodes.add(Wr.nodeKey(b,w)),b.id in a||(a[b.id]=Object.keys(a).length),A.indexOf(S)===-1&&A.push(S);let E=S.inboundLayers.length;for(let R=0;R<E;R++){let $=S.inputTensors[R],_=S.inboundLayers[R],P=S.nodeIndices[R],C=S.tensorIndices[R];i($,x,A,_,P,C)}for(x.push(S);A.indexOf(S)>=0;)A.splice(A.indexOf(S),1);o.push(S)},l=[],u=[];for(let y of this.outputs)i(y,l,u);let c=o.slice().reverse();for(let y of c){n[y.id]=y,y.id in t||(t[y.id]=0);let x=t[y.id],A=s[y.outboundLayer.id]==null?0:s[y.outboundLayer.id];x=Math.max(x,A),s[y.outboundLayer.id]=x,r[y.outboundLayer.id]=y.outboundLayer,t[y.id]=x;for(let b=0;b<y.inboundLayers.length;b++){let w=y.inboundLayers[b],k=y.nodeIndices[b],S=w.inboundNodes[k],E=t[S.id]==null?0:t[S.id];t[S.id]=Math.max(x+1,E),n[S.id]=S}}let p={};for(let y in t){let x=t[y];x in p||(p[x]=[]),p[x].push(n[y])}let d={};for(let y in s){let x=s[y];x in d||(d[x]=[]),d[x].push(r[y])}let h=Object.keys(d).map(y=>parseInt(y,10)).sort(Wf);this.layers=[];for(let y of h){let x=d[y];x.sort((A,b)=>{let w=a[A.id],k=a[b.id];return w<k?-1:w>k?1:0});for(let A of x)A instanceof Wr&&this.internalContainerRefs.push(A),this.layers.push(A)}this.layersByDepth=d,h=Object.keys(p).map(y=>parseInt(y,10)).sort(Wf);let f=this.inputs.slice(),m=[];for(let y of h)for(let x of p[y]){let A=x.outboundLayer;if(A!=null){for(let b of x.inputTensors)if(f.indexOf(b)===-1)throw new Cr(`Graph disconnected: cannot obtain value for tensor ${b} at layer "${A.name}". The following previous layers were accessed without issue: ${m}`);for(let b of x.outputTensors)f.push(b);m.push(A.name)}}this.nodesByDepth=p;let g=this.layers.map(y=>y.name);for(let y of g){let x=g.filter(A=>A===y).length;if(x!==1)throw new Cr(`The name "${y}" is used ${x} times in the model. All layer names should be unique. Layer names: `+JSON.stringify(g))}this.outboundNodes=[],this.inboundNodes=[],new l2({outboundLayer:this,inboundLayers:[],nodeIndices:[],tensorIndices:[],inputTensors:this.inputs,outputTensors:this.outputs,inputMasks:this.inputs.map(y=>null),outputMasks:this.outputs.map(y=>null),inputShapes:this.inputs.map(y=>y.shape),outputShapes:this.outputs.map(y=>y.shape)}),this.built=!0,this._refCount=1}assertNotDisposed(){if(this._refCount===0)throw new Error(`Container '${this.name}' is already disposed.`)}dispose(){this.assertNotDisposed();let e={refCountAfterDispose:null,numDisposedVariables:0};if(--this._refCount===0){for(let t of this.layers)e.numDisposedVariables+=t.dispose().numDisposedVariables;for(let t of this.internalContainerRefs)e.numDisposedVariables+=t.dispose().numDisposedVariables}return e.refCountAfterDispose=this._refCount,e}get trainable(){return this.trainable_}set trainable(e){this.layers.forEach(t=>{t._trainableWeights.forEach(n=>n.trainable=e)}),this.trainable_=e}get trainableWeights(){if(this._trainableWeights.length>0)throw new j("Container instance unexpectedly contains _trainableWeights.The trainable weights of a Container are a union of the trainable weights of its consituent Layers. Its own _trainableWeights must remain an empty Array.");if(!this.trainable)return[];let e=[];for(let t of this.layers)e=e.concat(t.trainableWeights);return e}get nonTrainableWeights(){let e=[];for(let t of this.layers)e.push(...t.nonTrainableWeights);if(!this.trainable){let t=[];for(let n of this.layers)t.push(...n.trainableWeights);return t.concat(e)}return e}get weights(){return this.trainableWeights.concat(this.nonTrainableWeights)}loadWeights(e,t=!0){let n={},s=0;for(let a of this.layers)for(let o of a.weights){if(n[o.originalName]!=null)throw new j(`Duplicate weight name: ${o.originalName}`);n[o.originalName]=o,s++}let r=[];for(let a in e){let o=a;if(n[a]==null){let i=a.split("/");o=i.slice(0,-2).concat([i[i.length-1]]).join("/")}if(n[o]!=null)r.push([n[o],e[a]]);else if(t)throw new j(`Provided weight data has no target variable: ${a}`);delete n[o]}if(t){let a=[];for(let o in n)a.push(o);if(a.length>0)throw new j(`${a.length} of ${s} weights are not set: ${a}`)}a5(r)}updatedConfig(){let e=this.getConfig(),t={};return t.className=this.getClassName(),t.config=e,t.kerasVersion=`tfjs-layers ${m5}`,t.backend="TensorFlow.js",t}toJSON(e,t=!0){let n=z3(this.updatedConfig());return t?JSON.stringify(n):n}call(e,t){return Y(()=>{e=_t(e);let n=new Vi;for(let s=0;s<this.inputs.length;++s)n.add(this.inputs[s],e[s]);return ap(this.outputs,n,t)})}computeMask(e,t){return Y(()=>{e=_t(e);let n;return t==null?n=el(null,e.length):n=_t(t),this.runInternalGraph(e,n)[1]})}computeOutputShape(e){let t=xm(e);if(t.length!==this.inputLayers.length)throw new j(`Invalid inputShape argument ${e}: model has ${this.inputLayers.length} tensor inputs.`);let n={};for(let o=0;o<t.length;o++){let i=this.inputLayers[o],l=t[o],u=i.name+"_0_0";n[u]=l}let s=Object.keys(this.nodesByDepth).map(o=>parseInt(o,10)).sort(Wf);if(s.length>1)for(let o of s){let i=this.nodesByDepth[o];for(let l of i){let u=l.outboundLayer;if(this.inputLayers.map(f=>f.id).indexOf(u.id)!==-1)continue;let c=[];for(let f=0;f<l.inboundLayers.length;f++){let m=l.inboundLayers[f],g=l.nodeIndices[f],y=l.tensorIndices[f],x=`${m.name}_${g}_${y}`,A=n[x];c.push(A)}let p=u.computeOutputShape(ps(c)),d=xm(p),h=u.inboundNodes.indexOf(l);for(let f=0;f<d.length;f++){let m=`${u.name}_${h}_${f}`;n[m]=d[f]}}}let r=[],a=[];for(let o=0;o<this.outputLayers.length;o++){let i=this.outputLayers[o],l=this.outputLayersNodeIndices[o],u=this.outputLayersTensorIndices[o],c=`${i.name}_${l}_${u}`;a.push(c)}for(let o=0;o<a.length;o++){let i=a[o];Vr(i in n),r.push(n[i])}return ps(r)}runInternalGraph(e,t){t==null&&(t=el(null,e.length));let n={};for(let i=0;i<this.inputs.length;++i){let l=this.inputs[i],u=e[i],c=t[i];n[l.id]=[u,c]}let s=Object.keys(this.nodesByDepth).map(i=>parseInt(i,10)).sort(Wf);for(let i of s){let l=this.nodesByDepth[i];for(let u of l){let c=u.outboundLayer,p=u.inputTensors,d=u.outputTensors,h=new Array;for(let f of p)f.id in n&&h.push(n[f.id]);if(h.length===p.length){let f={},m,g,y,x;if(u.callArgs!=null&&(f=u.callArgs),h.length===1){let[A,b]=h[0];f.mask==null&&(f.mask=b),y=_t(c.call(A,f)),x=_t(c.computeMask(A,b)),m=[A],g=[b]}else m=h.map(A=>A[0]),g=h.map(A=>A[1]),f.mask==null&&(f.mask=g),y=_t(c.call(m,f)),x=_t(c.computeMask(m,g));if(c.activityRegularizer)throw new qe("LayersModel invocation with concrete Tensor value(s) in the presence of activity regularizer(s) is not supported yet.");for(let A=0;A<d.length;++A){let b=d[A],w=y[A],k=x[A];n[b.id]=[w,k]}}}}let r=[],a=[],o=[];for(let i of this.outputs){Vr(i.id in n,`Could not compute output ${i.name} : ${i.id}`);let[l,u]=n[i.id];o.push(l.shape),r.push(l),a.push(u)}return[r,a,o]}buildNodeConversionMap(e){let t={},n;for(let s of this.layers){n=s instanceof Wr?1:0;for(let r=0;r<s.inboundNodes.length;r++){let a=Wr.nodeKey(s,r);this.containerNodes.has(a)&&(t[a]=n,n+=1)}}return t}getLayer(e,t){if(t!=null){if(this.layers.length<=t)throw new j(`Was asked to retrieve layer at index ${t}, but model only has ${this.layers.length} layer(s).`);return this.layers[t]}else if(e==null)throw new j("Provide either a layer name or layer index");for(let n of this.layers)if(n.name===e)return n;throw new j(`No such layer: ${e}`)}calculateLosses(){return Y(()=>{let e=[];for(let t of this.layers)for(let n=0;n<t.inboundNodes.length;++n){let s=Wr.nodeKey(t,n);this.containerNodes.has(s)&&e.push(...t.calculateLosses())}return e})}getConfig(){let e={name:this.name},t=this.buildNodeConversionMap(this.layers),n=[];for(let a of this.layers){let o=a.getClassName(),i=a.getConfig(),l=[];for(let c=0;c<a.inboundNodes.length;c++){let p=a.inboundNodes[c],d=Wr.nodeKey(a,c),h={};if(this.containerNodes.has(d)){if(p.callArgs)try{JSON.stringify(p.callArgs),h=p.callArgs}catch(f){console.warn(`Layer ${a.name} was passed non-serializable keyword arguments: ${p.callArgs}. They will not be included in the serialized model (and thus will be missing at deserialization time).`),h={}}if(p.inboundLayers.length>0){let f=[];for(let m=0;m<p.inboundLayers.length;m++){let g=p.inboundLayers[m],y=p.nodeIndices[m],x=p.tensorIndices[m],A=Wr.nodeKey(g,y),b=t[A];b==null&&(b=0),f.push([g.name,b,x,h])}l.push(f)}}}let u={};u.name=a.name,u.className=o,u.config=i,u.inboundNodes=l,n.push(u)}e.layers=n;let s=[];for(let a=0;a<this.inputLayers.length;a++){let o=this.inputLayers[a],i=this.inputLayersNodeIndices[a],l=Wr.nodeKey(o,i);if(!this.containerNodes.has(l))continue;let u=t[l];u==null&&(u=0);let c=this.inputLayersTensorIndices[a];s.push([o.name,u,c])}e.inputLayers=s;let r=[];for(let a=0;a<this.outputLayers.length;a++){let o=this.outputLayers[a],i=this.outputLayersNodeIndices[a],l=Wr.nodeKey(o,i);if(!this.containerNodes.has(l))continue;let u=t[l];u==null&&(u=0);let c=this.outputLayersTensorIndices[a];r.push([o.name,u,c])}return e.outputLayers=r,e}static fromConfig(e,t,n={},s=!1){let r={},a={};function o(m,g){m.name in a?a[m.name].push(g):a[m.name]=[g]}function i(m,g){let y=[],x;for(let A of g){let b=A[0],w=A[1],k=A[2];if(x=A[3]==null?{}:A[3],!(b in r)){o(m,g);return}let S=r[b];if(S.inboundNodes.length<=w){o(m,g);return}let E=S.inboundNodes[w];y.push(E.outputTensors[k])}y.length>0&&m.apply(ps(y),x)}function l(m){let g=m.name,y=Er(m,t.customObjects!=null?t.customObjects:{});y.setFastWeightInitDuringBuild(s),r[g]=y,m.inboundNodes.forEach(A=>{if(!(A instanceof Array))throw new j(`Corrupted configuration, expected array for nodeData: ${A}`);o(y,A)})}let u=t.name,c=t.layers;for(let m of c)l(m);for(;!yV(a);)for(let m of c){let g=r[m.name];if(g.name in a){let y=a[g.name];delete a[g.name];for(let x of y)i(g,x)}}let p=[],d=[],h=t.inputLayers;for(let m of h){let g=m[0],y=m[1],x=m[2];Vr(g in r);let b=r[g].inboundNodes[y].outputTensors;p.push(b[x])}let f=t.outputLayers;for(let m of f){let g=m[0],y=m[1],x=m[2];Vr(g in r);let b=r[g].inboundNodes[y].outputTensors;d.push(b[x])}return new e({inputs:p,outputs:d,name:u})}get stateful(){if(this._stateful)throw new j("Container instance unexpectedly has _stateful = true. The statefulness of a Container is determined by the Layers it contains. Its _stateful property must remain the default false.");for(let e of this.layers)if(e.stateful)return!0;return!1}resetStates(){Y(()=>{this.layers.forEach(e=>{e.stateful&&e.resetStates()})})}};function UU(e,t,n){let s=t.length;if(e==null||Array.isArray(e)&&e.length===0)return t.map(r=>null);if(s===1)return Array.isArray(e)&&e.length===1?e:typeof e=="object"&&t[0]in e?[e[t[0]]]:[e];if(Array.isArray(e)){if(e.length!==s)throw new Error(`Provided ${n} is an array of ${e.length} element(s), but the model has ${s} outputs. Make sure a set of weights is provided for each model output.`);return e}else if(typeof e=="object"&&Object.keys(e).length>0&&typeof e[Object.keys(e)[0]]=="object"){let r=[];return t.forEach(a=>{a in e?r.push(e[a]):r.push(null)}),r}else throw new Error(`The model has multiple (${s}) outputs, so ${n} must be either an array with ${s} elements or an object with ${t} keys. Provided ${n} not understood: ${JSON.stringify(e)}`)}function W8(e,t){return UU(e,t,"classWeight")}async function V8(e,t,n,s){if(t!=null||s!=null)throw new Error("Support sampleWeight is not implemented yet");if(n!=null){let r=Y(()=>{if(e.shape.length===1)return On(e);if(e.shape.length===2){if(e.shape[1]>1)return Es(e,1);if(e.shape[1]===1)return U(e,[e.shape[0]]);throw new Error(`Encountered unexpected last-dimension size (${e.shape[1]}) during handling of class weights. The size is expected to be >= 1.`)}else throw new Error(`Unexpected rank of target (y) tensor (${e.rank}) during handling of class weights. The rank is expected to be 1 or 2.`)}),a=Array.from(await r.data());ee(r);let o=[];return a.forEach(i=>{if(n[i]==null)throw new Error(`classWeight must contain all classes in the training data. The class ${i} exists in the data but not in classWeight`);o.push(n[i])}),Ft(o,"float32")}else return null}function GU(e,t){return L(e,t)}var HU=32;function U8(e,t){let n,s,r=t;n=r.xs,s=r.ys,v.assert(n!=null&&s!=null,()=>`A Dataset iterator for fitDataset() is expected to generate objects of the form \`{xs: xVal, ys: yVal}\`, where the two values may be \`tf.Tensor\`, an array of Tensors, or a map of string to Tensor. The provided Dataset instead generates ${t}`);let a=jv("input",e.inputNames,n),o=jv("output",e.outputNames,s),i=a[0].shape[0];v.assert(a.length===e.inputs.length,()=>`LayersModel has ${e.inputs.length} inputs, but the dataset provides ${a.length} inputs. (Expected input keys: ${JSON.stringify(e.inputNames)})`),v.assert(o.length===e.outputs.length,()=>`LayersModel has ${e.outputs.length} outputs, but the dataset provides ${o.length} outputs. (Expected output keys: ${JSON.stringify(e.outputNames)})`);for(let l=0;l<a.length;l++)v.assert(a[l].shape[0]===i,()=>`Batch size mismatch: input ${e.inputNames[l]} has ${a[l].shape[0]}; expected ${i} based on input ${e.inputNames[0]}.`);for(let l=0;l<o.length;l++)v.assert(o[l].shape[0]===i,()=>`Batch size mismatch: output ${e.outputNames[l]} has ${o[l].shape[0]}; expected ${i} based on input ${e.inputNames[0]}.`);return{xs:a,ys:o}}function jv(e,t,n){if(n instanceof nt)return[n];if(Array.isArray(n))return v.assert(n.length===t.length,()=>`Received an array of ${n.length} Tensors, but expected ${t.length} to match the ${e} keys ${t}.`),n;{let s=[];for(let r of t){if(n[r]==null)throw new j(`The feature data generated by the dataset lacks the required ${e} key '${r}'.`);s.push(n[r])}return s}}function jU(e){if(e.length===3)throw new qe("Validation with sample weights is not implemented yet.");return{xs:e[0],ys:e[1]}}async function qU(e,t,n){let s=n.batchesPerEpoch!=null;if(v.assert(e.optimizer!=null,()=>"You must compile a model before training/testing. Use LayersModel.compile(modelCompileConfig)."),v.assert(n!=null,()=>"For fitDataset(), the 2nd argument (config) is required, but it is not provided in this call."),v.assert(n.epochs!=null&&n.epochs>0&&Number.isInteger(n.epochs),()=>`For fitDataset(), config.epochs is expected to be a positive integer, but got ${n.epochs}`),v.assert(!s||n.batchesPerEpoch>0&&Number.isInteger(n.batchesPerEpoch),()=>`For fitDataset(), config.batchesPerEpoch is expected to be a positive integer if specified, but got ${n.batchesPerEpoch}`),v.assert(n.validationSplit==null,()=>"`validationSplit` is not supported by `fitDataset()`. Use validationData instead."),e.isTraining)throw new Error("Cannot start training because another fit() call is ongoing.");e.isTraining=!0;try{let r=n.validationData!=null,a,o;if(r)if(qv(n.validationData))v.assert(n.validationBatches==null||n.validationBatches>0&&Number.isInteger(n.validationBatches),()=>`For fitDataset() with dataset-based validation, config.validationBatches is expected not to be provided, or to be a positive integer, but got ${n.validationBatches}`);else{let g=jU(n.validationData);a=g.xs,o=g.ys}let i=e.makeTrainFunction(),l=e.getDedupedMetricsNames(),u;r?u=l.slice().concat(l.map(g=>"val_"+g)):u=l.slice();let c=$8(n.callbacks,n.yieldEvery),p=n.verbose==null?1:n.verbose,{callbackList:d,history:h}=P8(c,p,n.epochs,null,null,XU(t,n),null,r,u);d.setModel(e),e.history=h,await d.onTrainBegin(),e.stopTraining_=!1;let f=n.initialEpoch==null?0:n.initialEpoch,m=await t.iterator();for(;f<n.epochs;){let g={};await d.onEpochBegin(f);let y=0,x=0;for(s||(m=await t.iterator());!s||y<n.batchesPerEpoch;){let A=await m.next();if(s&&A.done){console.warn(`You provided \`batchesPerEpoch\` as ${n.batchesPerEpoch}, but your dataset iterator ran out of data after ${y} batches; interrupting training. Make sure that your dataset can generate at least \`batchesPerEpoch * epochs\` batches (in this case, ${n.batchesPerEpoch*n.epochs} batches). You may need to use the repeat() function when building your dataset.`);break}if(A.value!=null){let{xs:b,ys:w}=U8(e,A.value),k={};k.batch=x,k.size=b[0].shape[0],await d.onBatchBegin(x,k);let S=[];if(n.classWeight!=null){let $=W8(n.classWeight,e.outputNames);for(let _=0;_<$.length;++_)S.push(await V8(w[_],null,$[_]))}let E=b.concat(w).concat(S),R=i(E);ee(E);for(let $=0;$<l.length;++$){let _=l[$],P=R[$];k[_]=P,bn(P)}await d.onBatchEnd(x,k),E8(k),x++,y++}if(s?y>=n.batchesPerEpoch:A.done){if(r){let b;qv(n.validationData)?b=_t(await e.evaluateDataset(n.validationData,{batches:n.validationBatches})):b=_t(e.evaluate(a,o,{batchSize:n.validationBatchSize==null?HU:n.validationBatchSize,verbose:0}));for(let w=0;w<e.metricsNames.length;++w)g[`val_${e.metricsNames[w]}`]=b[w]}break}if(e.stopTraining_)break}if(await d.onEpochEnd(f,g),f++,e.stopTraining_)break}return await d.onTrainEnd(),await e.history.syncData(),e.history}finally{e.isTraining=!1}}function XU(e,t){let n=null;return t.batchesPerEpoch!=null?n=t.batchesPerEpoch:Number.isFinite(e.size)&&(n=e.size),n}function qv(e){return typeof e.iterator=="function"}function KU(e){return typeof e.next=="function"}async function ZU(e,t,n){n=n||{};let s=n.batches!=null,r=e.testFunction,a=[];if(n.verbose>0)throw new qe("Verbose mode is not implemented yet.");v.assert(!s||n.batches>0&&Number.isInteger(n.batches),()=>`Test loop expects \`batches\` to be a positive integer, but received ${JSON.stringify(n.batches)}`);let o=KU(t)?t:await t.iterator(),i=0,l=0;for(;!s||l<n.batches;){let u=await o.next();if(a=Y(()=>{if(u.value){let{xs:c,ys:p}=U8(e,u.value),d=c.concat(p),h=Y(()=>r(d));if(ee(d),l===0)for(let m=0;m<h.length;++m)a.push(Ce(0));let f=d[0].shape[0];for(let m=0;m<h.length;++m){let g=h[m],y=a[m];a[m]=Y(()=>ce(a[m],L(f,g))),l>0&&ee(y)}ee(h),i+=f,++l}return a}),u.done){s&&console.warn(`Your dataset iterator ran out of data during evaluateDataset(). Interrupting evalution. Make sure that your dataset can generate at least \`batches\` batches (in this case, ${n.batches} batches). You may need to use the repeat() function when building your dataset.`);break}}for(let u=0;u<a.length;++u){let c=a[u];a[u]=he(a[u],i),ee(c)}return ps(a)}function L3(e){v.assert(e>0&&Number.isInteger(e),()=>`batchSize is required to be a positive integer, but got ${e}`)}function op(e,t,n){return e==null?[null]:Array.isArray(e)?e.map(s=>ji(s,t,n-t)):ji(e,t,n-t)}function g5(e,t){return Y(()=>e==null?null:Array.isArray(e)?e.map(n=>g5(n,t)):v8(e,t.dtype==="int32"?t:ye(t,"int32")))}function B3(e,t){let n=[],s=0,r=null;for(;s<e;)r=s+t,r>=e&&(r=e),n.push([s,r]),s=r;return n}async function YU(e,t,n,s,r,a,o,i,l,u,c,p,d,h,f){r==null&&(r=32),a==null&&(a=1),c==null&&(c=!0),d==null&&(d=0);let m=!1;if(l!=null&&u!=null&&(m=!0),f!=null&&(m=!0,h==null))throw new j("Can only use `validationSteps` when doing step-wise training, i.e., `stepsPerEpoch` must be set.");let g=e.checkNumSamples(n,r,h,"steps_per_epoch"),y;g!=null&&(y=_r(0,g)),o==null&&(o=1);let{callbackList:x,history:A}=P8(i,o,a,d,g,h,r,m,p);x.setModel(e),e.history=A,await x.onTrainBegin(),e.stopTraining_=!1;for(let b=d;b<a;++b){await x.onEpochBegin(b);let w={};if(h!=null)throw new qe("stepsPerEpoch mode is not implemented yet.");{if(c==="batch")throw new qe("batch shuffling is not implemneted yet");c&&v.shuffle(y);let k=Ft(y),S=B3(g,r);for(let E=0;E<S.length;++E){let R={};if(await x.onBatchBegin(E,R),Y(()=>{let $=S[E][0],_=S[E][1],P=ji(k,$,_-$);R.batch=E,R.size=_-$;let C=g5(n,P),F=t(C);for(let G=0;G<s.length;++G){let K=s[G],z=F[G];R[K]=z,bn(z)}if(E===S.length-1&&m){let G=e.testLoop(l,u,r);for(let K=0;K<s.length;++K){let z=s[K],Z=G[K];bn(Z),w["val_"+z]=Z}}}),await x.onBatchEnd(E,R),E8(R),e.stopTraining_)break}k.dispose()}if(await x.onEpochEnd(b,w),e.stopTraining_)break}return await x.onTrainEnd(),await e.history.syncData(),e.history}async function JU(e,t,n,s={}){if(e.isTraining)throw new Error("Cannot start training because another fit() call is ongoing.");e.isTraining=!0;let r,a,o,i,l,u,c,p,d;try{let h=s.batchSize==null?32:s.batchSize;L3(h);let f=!1,m=await e.standardizeUserData(t,n,s.sampleWeight,s.classWeight,f,h);r=m[0],a=m[1],d=m[2];let g=!1,y;if(s.validationData!=null&&s.validationData.length>0){if(g=!0,s.validationData.length===2)l=s.validationData[0],u=s.validationData[1];else throw s.validationData.length===3?new qe("validationData including sample weights is not supported yet."):new j(`When passing validation data, it must contain 2 (valX, valY) or 3 (valX, valY, valSampleWeight) items; ${s.validationData} is invalid.`);let R=!0,$=await e.standardizeUserData(l,u,null,null,R,h);c=$[0],p=$[1],y=c.concat(p)}else if(s.validationSplit!=null&&s.validationSplit>0&&s.validationSplit<1){g=!0;let R=Math.floor(r[0].shape[0]*(1-s.validationSplit)),$=r[0].shape[0];c=op(r,R,$),o=r,r=op(r,0,R),p=op(a,R,$),i=a,a=op(a,0,R),y=c.concat(p)}else s.validationSteps!=null&&(g=!0);let x=r.concat(a).concat(d);e.checkTrainableWeightsConsistency();let A=e.makeTrainFunction(),b=e.getDedupedMetricsNames(),w,k;g?(e.makeTestFunction(),w=e.testFunction,k=b.slice().concat(b.map(R=>"val_"+R))):(w=null,y=[],k=b.slice());let S=$8(s.callbacks,s.yieldEvery);return await YU(e,A,x,b,h,s.epochs,s.verbose,S,w,y,s.shuffle,k,s.initialEpoch,null,null)}finally{e.isTraining=!1,Sr(r,t),Sr(a,n),Sr(o,t),Sr(i,n),Sr(c,l),Sr(p,u),d!=null&&ee(d)}}function G8(e){let t=[];e instanceof nt&&(e=[e]);for(let n=0;n<e.length;++n){let s=e[n];if(s.rank===1)t.push(kh(s,1));else{if(s.rank===0)throw new Error("Expected tensor to be at least 1D, but received a 0D tensor (scalar).");t.push(s)}}return t}function Sr(e,t){if(e==null)return;let n=[];if(t instanceof nt)n.push(t.id);else if(Array.isArray(t))t.forEach(r=>n.push(r.id));else if(t!=null)for(let r in t){let a=t[r];n.push(a.id)}let s=[];if(e instanceof nt)n.indexOf(e.id)===-1&&s.push(e);else if(Array.isArray(e))e.forEach(r=>{n.indexOf(r.id)===-1&&s.push(r)});else if(e!=null)for(let r in e){let a=e[r];n.indexOf(a.id)===-1&&s.push(a)}s.forEach(r=>{r.isDisposed||r.dispose()})}function QU(e){return e instanceof nt}function W3(e){return Array.isArray(e)}function Xv(e){return!QU(e)&&!W3(e)}function Kv(e,t,n,s=!0,r=""){if(t==null||t.length===0){if(e!=null){let o=!1;if(W3(e)&&e.length>0)o=!0;else if(Xv(e)){for(let i in e)if(e.hasOwnProperty(i)){o=!0;break}}else o=!0;if(o)throw new j(`Error when checking model ${r} expected no data, but got ${e}`)}return[]}if(e==null)return t.map(o=>null);let a;if(Xv(e)){e=e,a=[];for(let o of t){if(e[o]==null)throw new j(`No data provided for "${o}". Need data for each key in: ${t}`);a.push(e[o])}}else if(W3(e)){if(e=e,e.length!==t.length)throw new j(`Error when checking model ${r}: the Array of Tensors that you are passing to your model is not the size the model expected. Expected to see ${t.length} Tensor(s), but instead got the following list of Tensor(s): ${e}`);a=e}else{if(e=e,t.length>1)throw new j(`The model ${r} expects ${t.length} Tensor(s), but only received one Tensor. Found: Tensor with shape ${e.shape}`);a=[e]}if(a=G8(a),n!=null)for(let o=0;o<t.length;++o){if(n[o]==null)continue;let i=a[o];if(i.shape.length!==n[o].length)throw new j(`Error when checking ${r}: expected ${t[o]} to have ${n[o].length} dimension(s). but got array with shape ${i.shape}`);for(let l=0;l<n[o].length;++l){if(l===0&&!s)continue;let u=i.shape[l],c=n[o][l];if(c!=null&&c>=0&&u!==c)throw new j(`${r} expected a batch of elements where each example has shape [${n[o].slice(1,n[o].length)}] (i.e.,tensor shape [*,${n[o].slice(1,n[o].length)}]) but the ${r} received an input with ${i.shape[0]} examples, each with shape [${i.shape.slice(1,i.shape.length)}] (tensor shape [${i.shape}])`)}}return a}function eG(e,t,n){let s=Xa(e.map(a=>a.shape[0]));s.sort();let r=Xa(t.map(a=>a.shape[0]));if(r.sort(),s.length>1)throw new j(`All input Tensors (x) should have the same number of samples. Got array shapes: ${JSON.stringify(e.map(a=>a.shape))}`);if(r.length>1)throw new j(`All target Tensors (y) should have the same number of samples. Got array shapes: ${JSON.stringify(t.map(a=>a.shape))}`);if(s.length>0&&r.length>0&&!v.arraysEqual(s,r))throw new j(`Input Tensors should have the same number of samples as target Tensors. Found ${s[0]} input sample(s) and ${r[0]} target sample(s).`)}function tG(e,t,n){let s=[tu,c2,Np];for(let r=0;r<e.length;++r){let a=e[r],o=t[r],i=n[r];if(o!=null){if(o===Np&&a.shape[a.shape.length-1]===1)throw new j(`You are passing a target array of shape ${a.shape} while using a loss 'categorical_crossentropy'. 'categorical_crossentropy'expects targets to be binary matrices (1s and 0s) of shape [samples, classes].`);if(s.indexOf(o)!==-1){let l=a.shape.slice(1),u=i.slice(1);for(let c=0;c<l.length;++c){let p=l[c],d=u[c];if(d!=null&&p!==d)throw new j(`A target Tensor with shape ${a.shape} was passed for an output of shape ${i}, while using a loss function that expects targets to have the same shape as the output.`)}}}}}function Zv(e,t,n,s=!0,r=""){let a;if(Array.isArray(e)){if(e.length!==t.length)throw new j(`Error when checking model ${r}: the Array of Tensors that you are passing to your model is not the size the the model expected. Expected to see ${t.length} Tensor(s), but instead got ${e.length} Tensors(s).`);a=e}else{if(t.length>1)throw new j(`The model expects ${t.length} ${r} Tensors, but only received one Tensor. Found: array with shape ${JSON.stringify(e.shape)}.`);a=[e]}if(n!=null)for(let o=0;o<t.length;++o){if(n[o]==null)continue;let i=a[o];if(i.shape.length!==n[o].length)throw new j(`Error when checking ${r}: expected ${t[o]} to have ${n[o].length} dimension(s), but got array with shape ${JSON.stringify(i.shape)}`);for(let l=0;l<n[o].length;++l){if(l===0&&!s)continue;let u=i.shape[l],c=n[o][l];if(c!=null&&c!==u)throw new j(`Error when checking ${r}: expected ${t[o]} to have shape ${JSON.stringify(n[o])} but got array with shape ${JSON.stringify(i.shape)}.`)}}}function nG(e,t){if(e==null||Array.isArray(e)&&e.length===0)return t.map(s=>[]);let n;if(typeof e=="string"||typeof e=="function")n=[e];else if(Array.isArray(e)||typeof e=="object")n=e;else throw new TypeError(`Type of metrics argument not understood. Expected an string,function, Array, or Object, found: ${e}`);if(Array.isArray(n))return t.map(s=>n);{let s=[];for(let r of t){let a=n.hasOwnProperty(r)?n[r]:[];Array.isArray(a)||(a=[a]),s.push(a)}return s}}var sG="layers-model",ha=class extends Wr{constructor(e){super(e),this.isTraining=!1}summary(e,t,n=console.log){if(!this.built)throw new j("This model has never been called, thus its weights have not been created yet. So no summary can be displayed. Build the model first (e.g., by calling it on some test data).");zU(this,e,t,n)}compile(e){if(e.loss==null&&(e.loss=[]),this.loss=e.loss,typeof e.optimizer=="string")this.optimizer_=MU(e.optimizer),this.isOptimizerOwned=!0;else{if(!(e.optimizer instanceof ka))throw new j("User-defined optimizer must be an instance of tf.Optimizer.");this.optimizer_=e.optimizer,this.isOptimizerOwned=!1}let t=[];if(!Array.isArray(e.loss)&&typeof e.loss!="string"&&typeof e.loss!="function"){e.loss=e.loss;for(let a in e.loss)if(this.outputNames.indexOf(a)===-1)throw new j(`Unknown entry in loss dictionary: "${a}". Only expected the following keys: ${this.outputNames}`);for(let a of this.outputNames)e.loss[a]==null&&console.warn(`Output "${a}" is missing from loss dictionary. We assume this was done on purpose, and we will not be expecting data to be passed to ${a} during training`),t.push(c3(e.loss[a]))}else if(Array.isArray(e.loss)){if(e.loss.length!==this.outputs.length)throw new j(`When passing an Array as loss, it should have one entry per model output. The model has ${this.outputs.length} output(s), but you passed loss=${e.loss}.`);t=e.loss.map(o=>c3(o))}else{let a=c3(e.loss);this.outputs.forEach(o=>{t.push(a)})}this.lossFunctions=t,this.feedOutputNames=[],this.feedOutputShapes=[],this.feedLossFns=[];for(let a=0;a<this.outputs.length;++a){let o=this.internalOutputShapes[a],i=this.outputNames[a];this.feedOutputNames.push(i),this.feedOutputShapes.push(o),this.feedLossFns.push(this.lossFunctions[a])}let n=[];this.metrics=e.metrics,this.metricsNames=["loss"],this.metricsTensors=[],Hi("loss",()=>{for(let a=0;a<this.outputs.length;++a){if(n.indexOf(a)!==-1)continue;let o=this.lossFunctions[a];this.outputs.length>1&&(this.metricsTensors.push([o,a]),this.metricsNames.push(this.outputNames[a]+"_loss"))}});let s=nG(e.metrics,this.outputNames),r=(a,o,i)=>{this.outputNames.length>1&&(o=this.outputNames[a]+"_"+o),this.metricsNames.push(o),this.metricsTensors.push([i,a])};Hi("metric",()=>{for(let a=0;a<this.outputs.length;++a){if(n.indexOf(a)!==-1)continue;let o=s[a];(l=>{let u="",c,p,d;for(let h of l){if(typeof h=="string"&&["accuracy","acc","crossentropy","ce"].indexOf(h)!==-1){let m=this.internalOutputShapes[a];m[m.length-1]===1||this.lossFunctions[a]===c2?["accuracy","acc"].indexOf(h)!==-1?p=p5:["crossentropy","ce"].indexOf(h)!==-1&&(p=M8):this.lossFunctions[a]===Im?["accuracy","acc"].indexOf(h)!==-1?p=z8:["crossentropy","ce"].indexOf(h)!==-1&&(p=L8):["accuracy","acc"].indexOf(h)!==-1?p=h5:["crossentropy","ce"].indexOf(h)!==-1&&(p=f5);let g;["accuracy","acc"].indexOf(h)!==-1?g="acc":["crossentropy","ce"].indexOf(h)!==-1&&(g="ce"),d=p,c=u+g}else d=OU(h),c=u+Gf(h);let f;Hi(c,()=>{f=d}),r(a,c,f)}})(o)}}),this.collectedTrainableWeights=this.trainableWeights}checkTrainableWeightsConsistency(){this.collectedTrainableWeights!=null&&this.trainableWeights.length!==this.collectedTrainableWeights.length&&console.warn("Discrepancy between trainableweights and collected trainable weights. Did you set `model.trainable` without calling `model.compile()` afterwards?")}evaluate(e,t,n={}){let s=n.batchSize==null?32:n.batchSize;L3(s);let r=!0,a=this.standardizeUserDataXY(e,t,r,s);try{let o=a[0].concat(a[1]);this.makeTestFunction();let i=this.testFunction,l=this.testLoop(i,o,s,n.verbose,n.steps);return ps(l)}finally{Sr(a[0],e),Sr(a[1],t)}}async evaluateDataset(e,t){return this.makeTestFunction(),ZU(this,e,t)}checkNumSamples(e,t,n,s="steps"){let r;if(n!=null){if(r=null,t!=null)throw new j(`If ${s} is set, batchSize must be null or undefined.Got batchSize = ${t}`)}else if(e!=null)Array.isArray(e)?r=e[0].shape[0]:r=e.shape[0];else throw new j(`Either the input data should have a defined shape, or ${s} shoud be specified.`);return r}execute(e,t){if(Array.isArray(t)&&t.length===0)throw new j("`outputs` is an empty Array, which is not allowed.");let n=Array.isArray(t),s=n?t:[t],r=this.retrieveSymbolicTensors(s),a=new Vi;if(e instanceof nt&&(e=[e]),Array.isArray(e)){if(e.length!==this.inputs.length)throw new j(`The number of inputs provided (${e.length}) does not match the number of inputs of this model (${this.inputs.length}).`);for(let i=0;i<this.inputs.length;++i)a.add(this.inputs[i],e[i])}else for(let i of this.inputs){let l=e[i.name];if(l==null)throw new j(`No value is provided for the model's input ${i.name}`);a.add(i,l)}let o=ap(r,a);return n?o:o[0]}retrieveSymbolicTensors(e){let t=el(null,e.length),n=e.length;for(let s of this.layers){let r=Array.isArray(s.output)?s.output:[s.output],a=r.map(o=>o.name);for(let o=0;o<e.length;++o){let i=a.indexOf(e[o]);if(i!==-1&&(t[o]=r[i],n--),n===0)break}if(n===0)break}if(n>0){let s=[];throw t.forEach((r,a)=>{r==null&&s.push(e[a])}),new j(`Cannot find SymbolicTensors for output name(s): ${JSON.stringify(s)}`)}return t}predictLoop(e,t=32,n=!1){return Y(()=>{let s=this.checkNumSamples(e);if(n)throw new qe("Verbose predictLoop() is not implemented yet.");let r=B3(s,t),a=this.outputs.map(o=>[]);for(let o=0;o<r.length;++o)Y(()=>{let l=r[o][0],u=r[o][1],c=op(e,l,u),p=[];if(Array.isArray(c))for(let h=0;h<c.length;++h)p.push({key:this.inputs[h],value:c[h]});else p.push({key:this.inputs[0],value:c});let d=new Vi(p);return ap(this.outputs,d)}).forEach((l,u)=>a[u].push(l));return ps(a.map(o=>Ct(o,0)))})}predict(e,t={}){let n=G8(e);Zv(n,this.inputNames,this.feedInputShapes,!1);try{let s=t.batchSize==null?32:t.batchSize;return L3(s),this.predictLoop(n,s)}finally{Sr(n,e)}}predictOnBatch(e){Zv(e,this.inputNames,this.feedInputShapes,!0);let t=(Array.isArray(e)?e[0]:e).shape[0];return this.predictLoop(e,t)}standardizeUserDataXY(e,t,n=!0,s){if(this.optimizer_==null)throw new Cr("You must compile a model before training/testing. Use LayersModel.compile(modelCompileArgs).");let r=[];for(let a=0;a<this.feedOutputShapes.length;++a){let o=this.feedOutputShapes[a];this.feedLossFns[a]===Im?r.push(o.slice(0,o.length-1).concat([1])):r.push(o)}if(e=Kv(e,this.feedInputNames,this.feedInputShapes,!1,"input"),t=Kv(t,this.feedOutputNames,r,!1,"target"),eG(e,t,null),tG(t,this.feedLossFns,this.feedOutputShapes),this.stateful&&s!=null&&s>0&&e[0].shape[0]%s!==0)throw new j(`In a stateful network, you should only pass inputs with a number of samples that is divisible by the batch size ${s}. Found: ${e[0].shape[0]} sample(s).`);return[e,t]}async standardizeUserData(e,t,n,s,r=!0,a){let[o,i]=this.standardizeUserDataXY(e,t,r,a);if(n!=null)throw new Error("sample weight is not supported yet.");let l=null;if(s!=null){let u=W8(s,this.outputNames);l=[];for(let c=0;c<u.length;++c)l.push(await V8(i[c],null,u[c]))}return[o,i,l]}testLoop(e,t,n,s=0,r){return Y(()=>{let a=this.checkNumSamples(t,n,r,"steps"),o=[];if(s>0)throw new qe("Verbose mode is not implemented yet.");if(r!=null)throw new qe("steps mode in testLoop() is not implemented yet");{let i=B3(a,n),l=Ft(_r(0,a));for(let u=0;u<i.length;++u){let c=i[u][0],p=i[u][1],d=ji(l,c,p-c),h=g5(t,d),f=e(h);if(u===0)for(let m=0;m<f.length;++m)o.push(Ce(0));for(let m=0;m<f.length;++m){let g=f[m];o[m]=ce(o[m],L(p-c,g))}}for(let u=0;u<o.length;++u)o[u]=he(o[u],a)}return o})}getDedupedMetricsNames(){let e=this.metricsNames,t=[];for(let n=0;n<e.length;++n){let s=e[n],r=s;Pv(e,s)>1&&(r+=`_${Pv(e.slice(0,n),s)}`),t.push(r)}return t}makeTrainFunction(){return e=>{let t=[],n=e.slice(0,this.inputs.length),s=e.slice(this.inputs.length,this.inputs.length+this.outputs.length),r=e.slice(this.inputs.length+this.outputs.length,this.inputs.length+this.outputs.length*2),a=[],o=()=>{let c=[];for(let f=0;f<this.inputs.length;++f)c.push({key:this.inputs[f],value:n[f]});let p=new Vi(c),d=ap(this.outputs,p,{training:!0}),h;for(let f=0;f<this.lossFunctions.length;++f){let m=this.lossFunctions[f],g=m(s[f],d[f]);r[f]!=null&&(g=GU(g,r[f]));let y=Bt(g);t.push(y),f===0?h=g:h=ce(h,g)}for(let f=0;f<this.metricsTensors.length;++f){let m;if(this.outputs.length>1&&f<this.outputs.length)m=t[f];else{let g=this.metricsTensors[f][0],y=this.metricsTensors[f][1];m=Bt(g(s[y],d[y]))}bn(m),a.push(m)}return h=Bt(h),this.calculateLosses().forEach(f=>{h=ce(h,f)}),h},i=this.collectedTrainableWeights.map(c=>c.read()),l=!0;return[this.optimizer_.minimize(o,l,i)].concat(a)}}makeTestFunction(){this.testFunction=e=>Y(()=>{let t=[],n,s=e.slice(0,this.inputs.length),r=e.slice(this.inputs.length,this.inputs.length+this.outputs.length),a=[];for(let l=0;l<this.inputs.length;++l)a.push({key:this.inputs[l],value:s[l]});let o=new Vi(a),i=ap(this.outputs,o);for(let l=0;l<this.lossFunctions.length;++l){let u=this.lossFunctions[l],c=Bt(u(r[l],i[l]));l===0?n=c:n=ce(n,c),t.push(n)}for(let l=0;l<this.metricsTensors.length;++l){let u=this.metricsTensors[l][0],c=this.metricsTensors[l][1],p=Bt(u(r[c],i[c]));t.push(p)}return t})}async fit(e,t,n={}){return JU(this,e,t,n)}async fitDataset(e,t){return qU(this,e,t)}async trainOnBatch(e,t){let n=await this.standardizeUserData(e,t),s=n[0],r=n[1],o=this.makeTrainFunction()(s.concat(r)),i=[];for(let l of o){let u=await l.data();i.push(u[0])}return ee(o),Sr(n[0],e),Sr(n[1],t),ps(i)}getNamedWeights(e){let t=[],n=e!=null&&e.trainableOnly,s=n?this.trainableWeights:this.weights,r=this.getWeights(n);for(let a=0;a<s.length;++a)n&&!s[a].trainable||t.push({name:s[a].originalName,tensor:r[a]});return t}set stopTraining(e){this.stopTraining_=e}get stopTraining(){return this.stopTraining_}get optimizer(){return this.optimizer_}set optimizer(e){this.optimizer_!==e&&(this.optimizer_=e,this.isOptimizerOwned=!1)}dispose(){let e=super.dispose();if(e.refCountAfterDispose===0&&this.optimizer!=null&&this.isOptimizerOwned){let t=ym().numTensors;this.optimizer_.dispose(),e.numDisposedVariables+=t-ym().numTensors}return e}getLossIdentifiers(){let e;if(typeof this.loss=="string")e=ua(this.loss);else if(Array.isArray(this.loss)){for(let t of this.loss)if(typeof t!="string")throw new Error("Serialization of non-string loss is not supported.");e=this.loss.map(t=>ua(t))}else{let t=Object.keys(this.loss);e={};let n=this.loss;for(let s of t)if(typeof n[s]=="string")e[s]=ua(n[s]);else throw new Error("Serialization of non-string loss is not supported.")}return e}getMetricIdentifiers(){if(typeof this.metrics=="string"||typeof this.metrics=="function")return[ua(Gf(this.metrics))];if(Array.isArray(this.metrics))return this.metrics.map(e=>ua(Gf(e)));{let e={};for(let t in this.metrics)e[t]=ua(Gf(this.metrics[t]));return e}}getTrainingConfig(){return{loss:this.getLossIdentifiers(),metrics:this.getMetricIdentifiers(),optimizer_config:{class_name:this.optimizer.getClassName(),config:this.optimizer.getConfig()}}}loadTrainingConfig(e){if(e.weighted_metrics!=null)throw new Error("Loading weight_metrics is not supported yet.");if(e.loss_weights!=null)throw new Error("Loading loss_weights is not supported yet.");if(e.sample_weight_mode!=null)throw new Error("Loading sample_weight_mode is not supported yet.");let t=Ep(e.optimizer_config),n=Er(t),s;if(typeof e.loss=="string")s=Li(e.loss);else if(Array.isArray(e.loss))s=e.loss.map(a=>Li(a));else if(e.loss!=null){s={};for(let a in e.loss)s[a]=Li(e.loss[a])}let r;if(Array.isArray(e.metrics))r=e.metrics.map(a=>Li(a));else if(e.metrics!=null){r={};for(let a in e.metrics)r[a]=Li(e.metrics[a])}this.compile({loss:s,metrics:r,optimizer:n})}async save(e,t){if(typeof e=="string"){let l=Ts.getSaveHandlers(e);if(l.length===0)throw new j(`Cannot find any save handlers for URL '${e}'`);if(l.length>1)throw new j(`Found more than one (${l.length}) save handlers for URL '${e}'`);e=l[0]}if(e.save==null)throw new j("LayersModel.save() cannot proceed because the IOHandler provided does not have the `save` attribute defined.");let n=await Ts.encodeWeights(this.getNamedWeights(t)),s=!1,r=null,o={modelTopology:this.toJSON(r,s),format:sG,generatedBy:`TensorFlow.js tfjs-layers v${m5}`,convertedBy:null};if((t==null?!1:t.includeOptimizer)&&this.optimizer!=null){o.trainingConfig=this.getTrainingConfig();let l="optimizer",{data:u,specs:c}=await Ts.encodeWeights(await this.optimizer.getWeights(),l);n.specs.push(...c),n.data=Ts.concatenateArrayBuffers([n.data,u])}return this.userDefinedMetadata!=null&&(Hv(this.userDefinedMetadata,this.name,!0),o.userDefinedMetadata=this.userDefinedMetadata),o.weightData=n.data,o.weightSpecs=n.specs,e.save(o)}setUserDefinedMetadata(e){Hv(e,this.name),this.userDefinedMetadata=e}getUserDefinedMetadata(){return this.userDefinedMetadata}};ha.className="Model";de.registerClass(ha);var H8=class extends ha{};H8.className="Functional";de.registerClass(H8);async function rG(e,t){"modelTopology"in e||(e={modelTopology:e}),e=e;let n=e.modelTopology;n.model_config!=null&&(n=n.model_config);let s=Ep(n),r=Er(s,t);if(e.weightsManifest!=null){let a=await Ts.loadWeights(e.weightsManifest,e.pathPrefix,r.weights.map(i=>i.originalName)),o={};for(let i of r.weights)o[i.originalName]=a[i.originalName];r.loadWeights(o),ee(a)}return r}async function aG(e,t){if(t==null&&(t={}),typeof e=="string"){let n=Ts.getLoadHandlers(e,t);if(n.length===0)n.push(Ts.browserHTTPRequest(e,t));else if(n.length>1)throw new j(`Found more than one (${n.length}) load handlers for URL '${e}'`);e=n[0]}return oG(e,void 0,t)}async function oG(e,t,n){if(n==null&&(n={}),e.load==null)throw new j("Cannot proceed with model loading because the IOHandler provided does not have the `load` method implemented.");let s=await e.load(),r=s.modelTopology;r.model_config!=null&&(r=r.model_config);let a=n.strict==null?!0:n.strict,o=s.weightData!=null&&s.weightSpecs!=null&&a,i=Er(Ep(r),t,o),l=s.trainingConfig;if(l!=null&&i.loadTrainingConfig(l),s.userDefinedMetadata!=null&&i.setUserDefinedMetadata(s.userDefinedMetadata),s.weightData!=null){if(s.weightSpecs==null)throw new j("LayersModel artifacts contains weight data, but not weight specs. Therefore loading of weights cannot proceed.");let{modelWeights:u,optimizerWeights:c}=iG(s.weightData,s.weightSpecs);i.loadWeights(u,a),i.optimizer!=null&&c.length>0&&await i.optimizer.setWeights(c),ee(u),ee(c.map(p=>p.tensor))}return i}function iG(e,t){let n=Ts.decodeWeights(e,t),s={},r=[];return t.forEach(a=>{a.group==="optimizer"?r.push({name:a.name,tensor:n[a.name]}):s[a.name]=n[a.name]}),{modelWeights:s,optimizerWeights:r}}var sc=class extends ha{constructor(e){if(super({inputs:[],outputs:[]}),e=e||{},this.trainable=!0,this.built=!1,this.name=e.name!=null?e.name:J0("sequential_"),e.layers!=null)for(let t of e.layers)this.add(t)}checkShape(e){if(e.inboundNodes[0].outputTensors[0].shape.some(n=>n<0))throw new j(`Negative dimension size caused by adding layer ${e.name} with input shape [${e.inboundNodes[0].inputTensors[0].shape}]`)}add(e){let t=e instanceof sc||e instanceof ha,n;if(t){if(n=e,n.outputs.length!==1)throw new j("All layers in a Sequential model should have a single output tensor. For multi-output layers, use the functional API.");if(n.inputs.length!==1)throw new j("All layers in a Sequential model should have a single input tensor. For multi-input layers, use the functional API.")}if(this.outputs.length===0){if(e.inboundNodes.length===0){if(e.batchInputShape==null)throw new j("The first layer in a Sequential model must get an `inputShape` or `batchInputShape` argument.");let s=S8({batchShape:e.batchInputShape,dtype:e.dtype,name:e.name+"_input"});e.apply(s)}if(t)this.outputs=n.outputs,this.inputs=n.inputs;else{if(e.inboundNodes.length!==1)throw new j(`A layer added to a Sequential model must not already be connected somewhere else. LayersModel received layer ${e.name} which has ${e.inboundNodes.length} pre-existing inbound connections.`);if(e.inboundNodes[0].outputTensors.length!==1)throw new j("All layers in a Sequential model should have a single output tensor. For multi-output layers, use the functional API.");this.checkShape(e),this.outputs=[e.inboundNodes[0].outputTensors[0]],this.inputs=I8(this.outputs[0])}this.inboundNodes=[],new l2({outboundLayer:this,inboundLayers:[],nodeIndices:[],tensorIndices:[],inputTensors:this.inputs,outputTensors:this.outputs,inputMasks:el(null,this.inputs.length),outputMasks:[null],inputShapes:this.inputs.map(s=>s.shape),outputShapes:this.outputs[0].shape})}else{let s=e.apply(this.outputs[0]);if(Array.isArray(s))throw new TypeError("All layers in a Sequential model should have a single output tensor. For multi-output layers, use the functional API.");this.checkShape(e),this.outputs=[s],this.inboundNodes[0].outputTensors=this.outputs,this.inboundNodes[0].outputShapes=[this.outputs[0].shape]}this.layers.push(e),this.built=!1}pop(){if(this.layers.length===0)throw new TypeError("There are no layers in the model.");if(this.layers.pop(),this.layers.length===0)this.outputs=[],this.inboundNodes=[],this.outboundNodes=[];else{let e=this.layers.length-1;this.layers[e].outboundNodes=[],this.outputs=[this.layers[e].output],this.inboundNodes[0].outputTensors=this.outputs,this.inboundNodes[0].outputShapes=[this.outputs[0].shape]}}call(e,t){return this.model==null&&this.build(),this.model.call(e,t)}build(e){if(xt(e),this.inputs.length===0||this.outputs.length===0)throw new TypeError("Sequential model cannot be built: model is empty. Add some layers first.");this.model=new ha({inputs:this.inputs,outputs:this.outputs[0],name:this.name+"_model"}),this.model.trainable=this.trainable,this.supportsMasking=this.model.supportsMasking,this.inputLayers=this.model.inputLayers,this.inputLayersNodeIndices=this.model.inputLayersNodeIndices,this.inputLayersTensorIndices=this.model.inputLayersTensorIndices,this.outputLayers=this.model.outputLayers,this.outputLayersNodeIndices=this.model.outputLayersNodeIndices,this.outputLayersTensorIndices=this.model.outputLayersTensorIndices,this.nodesByDepth=this.model.nodesByDepth,this.containerNodes=this.model.containerNodes,this.outputNames=this.model.outputNames,this.inputNames=this.model.inputNames,this.built=!0}countParams(){return this.built||this.build(),super.countParams()}summary(e,t,n=console.log){this.built||this.build(),super.summary(e,t,n)}setWeights(e){this.model==null&&this.build(),this.model.setWeights(e)}evaluate(e,t,n={}){if(!this.built)throw new Cr("The model needs to be compiled before being used.");return this.model.evaluate(e,t,n)}async evaluateDataset(e,t){if(!this.built)throw new Cr("The model needs to be compiled before being used.");return this.model.evaluateDataset(e,t)}predict(e,t={}){return this.model==null&&this.build(),this.model.predict(e,t)}predictOnBatch(e){return this.model==null&&this.build(),this.model.predictOnBatch(e)}compile(e){this.build(),this.model.compile(e),this.optimizer_=this.model.optimizer,this.isOptimizerOwned=this.model.isOptimizerOwned,this.loss=this.model.loss,this.metrics=this.model.metrics,this.metricsTensors=this.model.metricsTensors,this.metricsNames=this.model.metricsNames}get optimizer(){return this.model==null?void 0:this.model.optimizer}set optimizer(e){this.model.optimizer=e}async fit(e,t,n={}){if(!this.built)throw new Cr("The model needs to be compiled before being used.");return this.model.fit(e,t,n)}async fitDataset(e,t){if(!this.built)throw new Cr("The model needs to be compiled before being used.");return this.model.fitDataset(e,t)}async trainOnBatch(e,t){return this.model.trainOnBatch(e,t)}static fromConfig(e,t,n={},s=!1){let r,a={};if(t instanceof Array){if(t[0].className==null||t[0].className==="Merge")throw new j("Legacy serialization format not supported yet.");r=t}else v.assert(t.layers!=null,()=>"When the config data for a Sequential model is not an Array, it must be an Object that contains the 'layers' field."),r=t.layers,delete t.layers,a=t;let o=new e(a);if(!(o instanceof sc))throw new qe(`Sequential.fromConfig called on non-Sequential input: ${o}`);for(let i of r){let u=Er(i,void 0,s);s&&u.setFastWeightInitDuringBuild(!0),o.add(u)}return o}set stopTraining(e){if(this.model==null)throw new j("Cannot set the stopTraining property of a sequential model before it is compiled.");this.model.stopTraining=e}get stopTraining(){if(this.model==null)throw new j("Cannot get the stopTraining property of a sequential model before it is compiled.");return this.model.stopTraining}getConfig(){let e=[];for(let t of this.layers){let n={};n.className=t.getClassName(),n.config=t.getConfig(),e.push(n)}return{name:this.name,layers:e}}};sc.className="Sequential";de.registerClass(sc);function lG(e){return new ha(e)}function uG(e){return new sc(e)}function cG(e,t){return t==null&&(t={}),aG(e,t)}function j8(e){return S8(e)}function dG(e,t){ar.registerCallbackConstructor(e,t)}var As=class extends de.Serializable{getConfig(){return{}}},q8=class extends As{apply(e,t=1){return DV(e,t)}};q8.className="elu";de.registerClass(q8);var X8=class extends As{apply(e){return D0(e)}};X8.className="selu";de.registerClass(X8);var K8=class extends As{apply(e){return Pr(e)}};K8.className="relu";de.registerClass(K8);var Z8=class extends As{apply(e){return Y(()=>Xc(6,Pr(e)))}};Z8.className="relu6";de.registerClass(Z8);var Y8=class extends As{apply(e){return e}};Y8.className="linear";de.registerClass(Y8);var J8=class extends As{apply(e){return Cn(e)}};J8.className="sigmoid";de.registerClass(J8);var Q8=class extends As{apply(e){return PV(e)}};Q8.className="hardSigmoid";de.registerClass(Q8);var ek=class extends As{apply(e){return Zl(e)}};ek.className="softplus";de.registerClass(ek);var tk=class extends As{apply(e){return $V(e)}};tk.className="softsign";de.registerClass(tk);var nk=class extends As{apply(e){return Yi(e)}};nk.className="tanh";de.registerClass(nk);var y5=class extends As{apply(e,t=-1){return Jl(e,t)}};y5.className="softmax";de.registerClass(y5);var sk=class extends As{apply(e,t=-1){return I0(e,t)}};sk.className="logSoftmax";de.registerClass(sk);var rk=class extends As{apply(e,t=1){return Y(()=>L(Cn(L(e,t)),e))}};rk.className="swish";de.registerClass(rk);var ak=class extends As{apply(e){return Y(()=>L(e,Yi(Zl(e))))}};ak.className="mish";de.registerClass(ak);function ro(e){return e.getClassName()}function d3(e,t={}){return wh(e,de.SerializationMap.getMap().classNameMap,t,"activation")}function ao(e){if(e==null){let t={};return t.className="linear",t.config={},d3(t)}if(typeof e=="string"){let t={};return t.className=e,t.config={},d3(t)}else return e instanceof As?e:d3(e)}function A5(e){if(e!=null&&typeof e!="object")throw new Error(`Argument to L1L2 regularizer's constructor is expected to be an object, but received: ${e}`)}var ok=class extends de.Serializable{},Th=class extends ok{constructor(e){super(),A5(e),this.l1=e==null||e.l1==null?.01:e.l1,this.l2=e==null||e.l2==null?.01:e.l2,this.hasL1=this.l1!==0,this.hasL2=this.l2!==0}apply(e){return Y(()=>{let t=Wt([1]);return this.hasL1&&(t=ce(t,ke(L(this.l1,tn(e))))),this.hasL2&&(t=ce(t,ke(L(this.l2,Ih(e))))),U(t,[])})}getConfig(){return{l1:this.l1,l2:this.l2}}static fromConfig(e,t){return new e({l1:t.l1,l2:t.l2})}};Th.className="L1L2";de.registerClass(Th);function pG(e){return A5(e),new Th({l1:e!=null?e.l1:null,l2:0})}function hG(e){return A5(e),new Th({l2:e!=null?e.l2:null,l1:0})}var Yv={l1l2:"L1L2"};function St(e){return KA(e)}function Jv(e,t={}){return wh(e,de.SerializationMap.getMap().classNameMap,t,"regularizer")}function Mt(e){if(e==null)return null;if(typeof e=="string"){let n={className:e in Yv?Yv[e]:e,config:{}};return Jv(n)}else return e instanceof ok?e:Jv(e)}var x5=class extends ut{constructor(e){super(e==null?{}:e),this.supportsMasking=!0,e!=null&&(this.maxValue=e.maxValue)}call(e,t){e=Xe(e);let n=Pr(e);return this.maxValue!=null&&(n=fs(n,0,this.maxValue)),n}computeOutputShape(e){return e}getConfig(){let e={maxValue:this.maxValue},t=super.getConfig();return Object.assign(e,t),e}};x5.className="ReLU";de.registerClass(x5);var b5=class extends ut{constructor(e){super(e==null?{}:e),this.DEFAULT_ALPHA=.3,e==null&&(e={}),this.alpha=e.alpha==null?this.DEFAULT_ALPHA:e.alpha}call(e,t){let n=Xe(e);return dh(n,this.alpha)}computeOutputShape(e){return e}getConfig(){let e={alpha:this.alpha},t=super.getConfig();return Object.assign(e,t),e}};b5.className="LeakyReLU";de.registerClass(b5);var v5=class extends ut{constructor(e){if(super(e==null?{}:e),this.DEFAULT_ALPHA_INITIALIZER="zeros",e==null&&(e={}),this.supportsMasking=!0,this.alphaInitializer=Ot(e.alphaInitializer||this.DEFAULT_ALPHA_INITIALIZER),this.alphaRegularizer=Mt(e.alphaRegularizer),this.alphaConstraint=yn(e.alphaConstraint),e.sharedAxes==null)this.sharedAxes=null;else if(Array.isArray(e.sharedAxes))this.sharedAxes=e.sharedAxes;else if(typeof e.sharedAxes=="number")this.sharedAxes=[e.sharedAxes];else throw new j(`Expected sharedAxes to be a number or an array of numbers, but got ${e.sharedAxes}`)}build(e){e=xt(e);let t=e.slice(1);if(this.sharedAxes!=null)for(let s of this.sharedAxes)t[s-1]=1;this.alpha=this.addWeight("alpha",t,"float32",this.alphaInitializer,this.alphaRegularizer,!0,this.alphaConstraint);let n={};if(this.sharedAxes!=null)for(let s=1;s<e.length;++s)n[s]=e[s];this.inputSpec=[new rn({ndim:e.length,axes:n})],this.built=!0}call(e,t){return e=Xe(e),yh(e,this.alpha.read())}getConfig(){let e={alphaInitializer:Vt(this.alphaInitializer),alphaRegularizer:St(this.alphaRegularizer),alphaConstraint:gn(this.alphaConstraint),sharedAxes:this.sharedAxes},t=super.getConfig();return Object.assign(e,t),e}};v5.className="PReLU";de.registerClass(v5);var w5=class extends ut{constructor(e){if(super(e==null?{}:e),this.DEFAULT_ALPHA=1,e==null&&(e={}),e.alpha!=null&&e.alpha!==this.DEFAULT_ALPHA)throw new qe(`Non-default alpha value (${e.alpha}) is not supported by the ELU layer yet.`);this.alpha=e.alpha==null?this.DEFAULT_ALPHA:e.alpha}call(e,t){let n=Xe(e);return Uc(n)}computeOutputShape(e){return e}getConfig(){let e={alpha:this.alpha},t=super.getConfig();return Object.assign(e,t),e}};w5.className="ELU";de.registerClass(w5);var k5=class extends ut{constructor(e){super(e==null?{}:e),this.DEFAULT_THETA=1,e==null&&(e={}),this.theta=e.theta==null?this.DEFAULT_THETA:e.theta}call(e,t){let n=Xe(e);return L(n,ye(ys(n,this.theta),"float32"))}computeOutputShape(e){return e}getConfig(){let e={theta:this.theta},t=super.getConfig();return Object.assign(e,t),e}};k5.className="ThresholdedReLU";de.registerClass(k5);var I5=class extends ut{constructor(e){super(e==null?{}:e),this.DEFAULT_AXIS=1,e==null&&(e={}),this.softmax=new y5().apply,this.axis=e.axis==null?this.DEFAULT_AXIS:e.axis}call(e,t){let n=Xe(e);return this.softmax(n,this.axis)}computeOutputShape(e){return e}getConfig(){let e={axis:this.axis},t=super.getConfig();return Object.assign(e,t),e}};I5.className="Softmax";de.registerClass(I5);function Hu(e,t,n){if(typeof e=="number")return el(e,t);if(e.length!==t)throw new j(`The ${n} argument must be an integer or tuple of ${t} integers. Received: ${e.length} elements.`);for(let s=0;s<t;++s){let r=e[s];if(!NV(r))throw new j(`The ${n} argument must be an integer or tuple of ${t} integers. Received: ${JSON.stringify(e)} including a non-integer number ${r}`)}return e}function Rr(e,t,n,s,r=1){if(e==null)return e;let a=t+(t-1)*(r-1),o;return n==="same"?o=e:o=e-a+1,Math.floor((o+s-1)/s)}function Ur(e,t,n,s){if(e==null)return null;if(s==="valid")e=e*t+so([n-t,0]);else if(s==="same")e=e*t;else throw new j(`Unsupport padding mode: ${s}.`);return e}function S5(e,t){return Y(()=>(Yt(t),t==="channelsFirst"?et(e,[0,2,3,1]):e))}function ik(e,t){return Y(()=>(Yt(t),t==="channelsFirst"?et(e,[0,2,3,4,1]):e))}function fG(e,t,n,s=1,r="valid",a,o=1){return Y(()=>{if(a==null&&(a=Dr()),Yt(a),e.shape.length!==3)throw new j(`The input of a conv1dWithBias operation should be 3, but is ${e.shape.length} instead.`);if(t.shape.length!==3)throw new j(`The kernel for a conv1dWithBias operation should be 3, but is ${t.shape.length} instead`);if(n!=null&&n.shape.length!==1)throw new j(`The bias for a conv1dWithBias operation should be 1, but is ${t.shape.length} instead`);if(a==="channelsFirst"&&(e=et(e,[0,2,1])),r==="causal")throw new qe("The support for CAUSAL padding mode in conv1dWithBias is not implemented yet.");let i=y0(e,t,s,r==="same"?"same":"valid","NWC",o);return n!=null&&(i=Fr(i,n)),i})}function Qv(e,t,n,s=[1,1],r="valid",a,o,i=null){return Y(()=>{if(a==null&&(a=Dr()),Yt(a),e.rank!==3&&e.rank!==4)throw new j(`conv2dWithBiasActivation expects input to be of rank 3 or 4, but received ${e.rank}.`);if(t.rank!==3&&t.rank!==4)throw new j(`conv2dWithBiasActivation expects kernel to be of rank 3 or 4, but received ${e.rank}.`);let l=S5(e,a);if(r==="causal")throw new qe("The support for CAUSAL padding mode in conv1dWithBias is not implemented yet.");return l=ec.conv2d({x:l,filter:t,strides:s,pad:r==="same"?"same":"valid",dilations:o,dataFormat:"NHWC",bias:n,activation:i}),a==="channelsFirst"&&(l=et(l,[0,3,1,2])),l})}function mG(e,t,n,s=[1,1,1],r="valid",a,o){return Y(()=>{if(a==null&&(a=Dr()),Yt(a),e.rank!==4&&e.rank!==5)throw new j(`conv3dWithBias expects input to be of rank 4 or 5, but received ${e.rank}.`);if(t.rank!==4&&t.rank!==5)throw new j(`conv3dWithBias expects kernel to be of rank 4 or 5, but received ${e.rank}.`);let i=ik(e,a);if(r==="causal")throw new qe("The support for CAUSAL padding mode in conv3dWithBias is not implemented yet.");return i=dA(i,t,s,r==="same"?"same":"valid","NDHWC",o),n!=null&&(i=Fr(i,n)),a==="channelsFirst"&&(i=et(i,[0,4,1,2,3])),i})}var C5=class extends ut{constructor(e,t){if(super(t),this.bias=null,this.DEFAULT_KERNEL_INITIALIZER="glorotNormal",this.DEFAULT_BIAS_INITIALIZER="zeros",C5.verifyArgs(t),this.rank=e,vn(this.rank,"rank"),this.rank!==1&&this.rank!==2&&this.rank!==3)throw new qe(`Convolution layer for rank other than 1, 2, or 3 (${this.rank}) is not implemented yet.`);if(this.kernelSize=Hu(t.kernelSize,e,"kernelSize"),this.strides=Hu(t.strides==null?1:t.strides,e,"strides"),this.padding=t.padding==null?"valid":t.padding,er(this.padding),this.dataFormat=t.dataFormat==null?"channelsLast":t.dataFormat,Yt(this.dataFormat),this.activation=ao(t.activation),this.useBias=t.useBias==null?!0:t.useBias,this.biasInitializer=Ot(t.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.biasConstraint=yn(t.biasConstraint),this.biasRegularizer=Mt(t.biasRegularizer),this.activityRegularizer=Mt(t.activityRegularizer),this.dilationRate=Hu(t.dilationRate==null?1:t.dilationRate,e,"dilationRate"),this.rank===1&&Array.isArray(this.dilationRate)&&this.dilationRate.length!==1)throw new j(`dilationRate must be a number or an array of a single number for 1D convolution, but received ${JSON.stringify(this.dilationRate)}`);if(this.rank===2){if(typeof this.dilationRate=="number")this.dilationRate=[this.dilationRate,this.dilationRate];else if(this.dilationRate.length!==2)throw new j(`dilationRate must be a number or array of two numbers for 2D convolution, but received ${JSON.stringify(this.dilationRate)}`)}else if(this.rank===3){if(typeof this.dilationRate=="number")this.dilationRate=[this.dilationRate,this.dilationRate,this.dilationRate];else if(this.dilationRate.length!==3)throw new j(`dilationRate must be a number or array of three numbers for 3D convolution, but received ${JSON.stringify(this.dilationRate)}`)}}static verifyArgs(e){if(Vr("kernelSize"in e,"required key 'kernelSize' not in config"),typeof e.kernelSize!="number"&&!ZA(e.kernelSize,"number",1,3))throw new j(`BaseConv expects config.kernelSize to be number or number[] with length 1, 2, or 3, but received ${JSON.stringify(e.kernelSize)}.`)}getConfig(){let e={kernelSize:this.kernelSize,strides:this.strides,padding:this.padding,dataFormat:this.dataFormat,dilationRate:this.dilationRate,activation:ro(this.activation),useBias:this.useBias,biasInitializer:Vt(this.biasInitializer),biasRegularizer:St(this.biasRegularizer),activityRegularizer:St(this.activityRegularizer),biasConstraint:gn(this.biasConstraint)},t=super.getConfig();return Object.assign(e,t),e}},Nh=class extends C5{constructor(e,t){super(e,t),this.kernel=null,Nh.verifyArgs(t),this.filters=t.filters,vn(this.filters,"filters"),this.kernelInitializer=Ot(t.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.kernelConstraint=yn(t.kernelConstraint),this.kernelRegularizer=Mt(t.kernelRegularizer)}build(e){e=xt(e);let t=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[t]==null)throw new j(`The channel dimension of the input should be defined. Found ${e[t]}`);let n=e[t],s=this.kernelSize.concat([n,this.filters]);this.kernel=this.addWeight("kernel",s,null,this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.useBias&&(this.bias=this.addWeight("bias",[this.filters],null,this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint)),this.inputSpec=[{ndim:this.rank+2,axes:{[t]:n}}],this.built=!0}call(e,t){return Y(()=>{e=Xe(e);let n,s=this.bias==null?null:this.bias.read(),r=m8(this.activation.getClassName());if(r!=null&&this.rank===2)n=Qv(e,this.kernel.read(),s,this.strides,this.padding,this.dataFormat,this.dilationRate,r);else{if(this.rank===1)n=fG(e,this.kernel.read(),s,this.strides[0],this.padding,this.dataFormat,this.dilationRate[0]);else if(this.rank===2)n=Qv(e,this.kernel.read(),s,this.strides,this.padding,this.dataFormat,this.dilationRate);else if(this.rank===3)n=mG(e,this.kernel.read(),s,this.strides,this.padding,this.dataFormat,this.dilationRate);else throw new qe("convolutions greater than 3D are not implemented yet.");this.activation!=null&&(n=this.activation.apply(n))}return n})}computeOutputShape(e){e=xt(e);let t=[],n=this.dataFormat==="channelsLast"?e.slice(1,e.length-1):e.slice(2);for(let r=0;r<n.length;++r){let a=Rr(n[r],this.kernelSize[r],this.padding,this.strides[r],typeof this.dilationRate=="number"?this.dilationRate:this.dilationRate[r]);t.push(a)}let s=[e[0]];return this.dataFormat==="channelsLast"?(s=s.concat(t),s.push(this.filters)):(s.push(this.filters),s=s.concat(t)),s}getConfig(){let e={filters:this.filters,kernelInitializer:Vt(this.kernelInitializer),kernelRegularizer:St(this.kernelRegularizer),kernelConstraint:gn(this.kernelConstraint)},t=super.getConfig();return Object.assign(e,t),e}static verifyArgs(e){if(!("filters"in e)||typeof e.filters!="number"||e.filters<1)throw new j(`Convolution layer expected config.filters to be a 'number' > 0 but got ${JSON.stringify(e.filters)}`)}},Eh=class extends Nh{constructor(e){super(2,e),Eh.verifyArgs(e)}getConfig(){let e=super.getConfig();return delete e.rank,e}static verifyArgs(e){if(typeof e.kernelSize!="number"&&!ZA(e.kernelSize,"number",1,2))throw new j(`Conv2D expects config.kernelSize to be number or number[] with length 1 or 2, but received ${JSON.stringify(e.kernelSize)}.`)}};Eh.className="Conv2D";de.registerClass(Eh);var Rh=class extends Nh{constructor(e){super(3,e),Rh.verifyArgs(e)}getConfig(){let e=super.getConfig();return delete e.rank,e}static verifyArgs(e){if(typeof e.kernelSize!="number"&&!(Array.isArray(e.kernelSize)&&(e.kernelSize.length===1||e.kernelSize.length===3)))throw new j(`Conv3D expects config.kernelSize to be number or [number, number, number], but received ${JSON.stringify(e.kernelSize)}.`)}};Rh.className="Conv3D";de.registerClass(Rh);var T5=class extends Eh{constructor(e){if(super(e),this.inputSpec=[new rn({ndim:4})],this.padding!=="same"&&this.padding!=="valid")throw new j(`Conv2DTranspose currently supports only padding modes 'same' and 'valid', but received padding mode ${this.padding}`)}build(e){if(e=xt(e),e.length!==4)throw new j("Input should have rank 4; Received input shape: "+JSON.stringify(e));let t=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[t]==null)throw new j("The channel dimension of the inputs should be defined. Found `None`.");let n=e[t],s=this.kernelSize.concat([this.filters,n]);this.kernel=this.addWeight("kernel",s,"float32",this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.useBias&&(this.bias=this.addWeight("bias",[this.filters],"float32",this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint)),this.inputSpec=[new rn({ndim:4,axes:{[t]:n}})],this.built=!0}call(e,t){return Y(()=>{let n=Xe(e);if(n.shape.length!==4)throw new j(`Conv2DTranspose.call() expects input tensor to be rank-4, but received a tensor of rank-${n.shape.length}`);let s=n.shape,r=s[0],a,o;this.dataFormat==="channelsFirst"?(a=2,o=3):(a=1,o=2);let i=s[a],l=s[o],u=this.kernelSize[0],c=this.kernelSize[1],p=this.strides[0],d=this.strides[1],h=Ur(i,p,u,this.padding),f=Ur(l,d,c,this.padding),m=[r,h,f,this.filters];this.dataFormat!=="channelsLast"&&(n=et(n,[0,2,3,1]));let g=A0(n,this.kernel.read(),m,this.strides,this.padding);return this.dataFormat!=="channelsLast"&&(g=et(g,[0,3,1,2])),this.bias!=null&&(g=Fr(g,this.bias.read(),this.dataFormat)),this.activation!=null&&(g=this.activation.apply(g)),g})}computeOutputShape(e){e=xt(e);let t=e.slice(),n,s,r;this.dataFormat==="channelsFirst"?(n=1,s=2,r=3):(n=3,s=1,r=2);let a=this.kernelSize[0],o=this.kernelSize[1],i=this.strides[0],l=this.strides[1];return t[n]=this.filters,t[s]=Ur(t[s],i,a,this.padding),t[r]=Ur(t[r],l,o,this.padding),t}getConfig(){let e=super.getConfig();return delete e.dilationRate,e}};T5.className="Conv2DTranspose";de.registerClass(T5);var N5=class extends Rh{constructor(e){if(super(e),this.inputSpec=[new rn({ndim:5})],this.padding!=="same"&&this.padding!=="valid")throw new j(`Conv3DTranspose currently supports only padding modes 'same' and 'valid', but received padding mode ${this.padding}`)}build(e){if(e=xt(e),e.length!==5)throw new j("Input should have rank 5; Received input shape: "+JSON.stringify(e));let t=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[t]==null)throw new j("The channel dimension of the inputs should be defined. Found `None`.");let n=e[t],s=this.kernelSize.concat([this.filters,n]);this.kernel=this.addWeight("kernel",s,"float32",this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.useBias&&(this.bias=this.addWeight("bias",[this.filters],"float32",this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint)),this.inputSpec=[new rn({ndim:5,axes:{[t]:n}})],this.built=!0}call(e,t){return Y(()=>{let n=Xe(e);if(n.shape.length!==5)throw new j(`Conv3DTranspose.call() expects input tensor to be rank-4, but received a tensor of rank-${n.shape.length}`);let s=n.shape,r=s[0],a,o,i;this.dataFormat==="channelsFirst"?(i=2,a=3,o=4):(i=1,a=2,o=3);let l=s[i],u=s[a],c=s[o],p=this.kernelSize[0],d=this.kernelSize[1],h=this.kernelSize[2],f=this.strides[0],m=this.strides[1],g=this.strides[2],y=Ur(l,f,p,this.padding),x=Ur(u,m,d,this.padding),A=Ur(c,g,h,this.padding),b=[r,y,x,A,this.filters];this.dataFormat!=="channelsLast"&&(n=et(n,[0,2,3,4,1]));let w=pA(n,this.kernel.read(),b,this.strides,this.padding);return this.dataFormat!=="channelsLast"&&(w=et(w,[0,4,1,2,3])),this.bias!==null&&(w=Fr(w,this.bias.read(),this.dataFormat)),this.activation!==null&&(w=this.activation.apply(w)),w})}computeOutputShape(e){e=xt(e);let t=e.slice(),n,s,r,a;this.dataFormat==="channelsFirst"?(n=1,s=2,r=3,a=4):(n=4,s=1,r=2,a=3);let o=this.kernelSize[0],i=this.kernelSize[1],l=this.kernelSize[2],u=this.strides[0],c=this.strides[1],p=this.strides[2];return t[n]=this.filters,t[s]=Ur(t[s],u,o,this.padding),t[r]=Ur(t[r],c,i,this.padding),t[a]=Ur(t[a],p,l,this.padding),t}getConfig(){let e=super.getConfig();return delete e.dilationRate,e}};N5.className="Conv3DTranspose";de.registerClass(N5);var lk=class extends Nh{constructor(e,t){if(super(e,t),this.DEFAULT_DEPTHWISE_INITIALIZER="glorotUniform",this.DEFAULT_POINTWISE_INITIALIZER="glorotUniform",this.depthwiseKernel=null,this.pointwiseKernel=null,t.filters==null)throw new j("The `filters` configuration field is required by SeparableConv, but is unspecified.");if(t.kernelInitializer!=null||t.kernelRegularizer!=null||t.kernelConstraint!=null)throw new j("Fields kernelInitializer, kernelRegularizer and kernelConstraint are invalid for SeparableConv2D. Use depthwiseInitializer, depthwiseRegularizer, depthwiseConstraint, pointwiseInitializer, pointwiseRegularizer and pointwiseConstraint instead.");if(t.padding!=null&&t.padding!=="same"&&t.padding!=="valid")throw new j(`SeparableConv${this.rank}D supports only padding modes: 'same' and 'valid', but received ${JSON.stringify(t.padding)}`);this.depthMultiplier=t.depthMultiplier==null?1:t.depthMultiplier,this.depthwiseInitializer=Ot(t.depthwiseInitializer||this.DEFAULT_DEPTHWISE_INITIALIZER),this.depthwiseRegularizer=Mt(t.depthwiseRegularizer),this.depthwiseConstraint=yn(t.depthwiseConstraint),this.pointwiseInitializer=Ot(t.depthwiseInitializer||this.DEFAULT_POINTWISE_INITIALIZER),this.pointwiseRegularizer=Mt(t.pointwiseRegularizer),this.pointwiseConstraint=yn(t.pointwiseConstraint)}build(e){if(e=xt(e),e.length<this.rank+2)throw new j(`Inputs to SeparableConv${this.rank}D should have rank ${this.rank+2}, but received input shape: ${JSON.stringify(e)}`);let t=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[t]==null||e[t]<0)throw new j(`The channel dimension of the inputs should be defined, but found ${JSON.stringify(e[t])}`);let n=e[t],s=this.kernelSize.concat([n,this.depthMultiplier]),r=[];for(let o=0;o<this.rank;++o)r.push(1);r.push(n*this.depthMultiplier,this.filters);let a=!0;this.depthwiseKernel=this.addWeight("depthwise_kernel",s,"float32",this.depthwiseInitializer,this.depthwiseRegularizer,a,this.depthwiseConstraint),this.pointwiseKernel=this.addWeight("pointwise_kernel",r,"float32",this.pointwiseInitializer,this.pointwiseRegularizer,a,this.pointwiseConstraint),this.useBias?this.bias=this.addWeight("bias",[this.filters],"float32",this.biasInitializer,this.biasRegularizer,a,this.biasConstraint):this.bias=null,this.inputSpec=[new rn({ndim:this.rank+2,axes:{[t]:n}})],this.built=!0}call(e,t){return Y(()=>{e=Xe(e);let n;if(this.rank===1)throw new qe("1D separable convolution is not implemented yet.");return this.rank===2&&(this.dataFormat==="channelsFirst"&&(e=et(e,[0,2,3,1])),n=$0(e,this.depthwiseKernel.read(),this.pointwiseKernel.read(),this.strides,this.padding,this.dilationRate,"NHWC")),this.useBias&&(n=Fr(n,this.bias.read(),this.dataFormat)),this.activation!=null&&(n=this.activation.apply(n)),this.dataFormat==="channelsFirst"&&(n=et(n,[0,3,1,2])),n})}getConfig(){let e=super.getConfig();return delete e.rank,delete e.kernelInitializer,delete e.kernelRegularizer,delete e.kernelConstraint,e.depthwiseInitializer=Vt(this.depthwiseInitializer),e.pointwiseInitializer=Vt(this.pointwiseInitializer),e.depthwiseRegularizer=St(this.depthwiseRegularizer),e.pointwiseRegularizer=St(this.pointwiseRegularizer),e.depthwiseConstraint=gn(this.depthwiseConstraint),e.pointwiseConstraint=gn(this.pointwiseConstraint),e}};lk.className="SeparableConv";var E5=class extends lk{constructor(e){super(2,e)}};E5.className="SeparableConv2D";de.registerClass(E5);var d2=class extends Nh{constructor(e){super(1,e),d2.verifyArgs(e),this.inputSpec=[{ndim:3}]}getConfig(){let e=super.getConfig();return delete e.rank,delete e.dataFormat,e}static verifyArgs(e){if(typeof e.kernelSize!="number"&&!ZA(e.kernelSize,"number",1,1))throw new j(`Conv1D expects config.kernelSize to be number or number[] with length 1, but received ${JSON.stringify(e.kernelSize)}.`)}};d2.className="Conv1D";de.registerClass(d2);var R5=class extends ut{constructor(e){super(e),typeof e.cropping=="number"?this.cropping=[[e.cropping,e.cropping],[e.cropping,e.cropping]]:typeof e.cropping[0]=="number"?this.cropping=[[e.cropping[0],e.cropping[0]],[e.cropping[1],e.cropping[1]]]:this.cropping=e.cropping,this.dataFormat=e.dataFormat===void 0?"channelsLast":e.dataFormat,this.inputSpec=[{ndim:4}]}computeOutputShape(e){return this.dataFormat==="channelsFirst"?[e[0],e[1],e[2]-this.cropping[0][0]-this.cropping[0][1],e[3]-this.cropping[1][0]-this.cropping[1][1]]:[e[0],e[1]-this.cropping[0][0]-this.cropping[0][1],e[2]-this.cropping[1][0]-this.cropping[1][1],e[3]]}call(e,t){return Y(()=>{if(e=Xe(e),this.dataFormat==="channelsLast"){let n=Uf(e,this.cropping[0][0],e.shape[1]-this.cropping[0][0]-this.cropping[0][1],2);return Uf(n,this.cropping[1][0],e.shape[2]-this.cropping[1][1]-this.cropping[1][0],3)}else{let n=Uf(e,this.cropping[0][0],e.shape[2]-this.cropping[0][0]-this.cropping[0][1],3);return Uf(n,this.cropping[1][0],e.shape[3]-this.cropping[1][1]-this.cropping[1][0],4)}})}getConfig(){let e={cropping:this.cropping,dataFormat:this.dataFormat},t=super.getConfig();return Object.assign(e,t),e}};R5.className="Cropping2D";de.registerClass(R5);var _5=class extends ut{constructor(e){super(e),this.DEFAULT_SIZE=[2,2],this.inputSpec=[{ndim:4}],this.size=e.size==null?this.DEFAULT_SIZE:e.size,this.dataFormat=e.dataFormat==null?"channelsLast":e.dataFormat,Yt(this.dataFormat),this.interpolation=e.interpolation==null?"nearest":e.interpolation,SV(this.interpolation)}computeOutputShape(e){if(this.dataFormat==="channelsFirst"){let t=e[2]==null?null:this.size[0]*e[2],n=e[3]==null?null:this.size[1]*e[3];return[e[0],e[1],t,n]}else{let t=e[1]==null?null:this.size[0]*e[1],n=e[2]==null?null:this.size[1]*e[2];return[e[0],t,n,e[3]]}}call(e,t){return Y(()=>{let n=Xe(e),s=n.shape;if(this.dataFormat==="channelsFirst"){n=et(n,[0,2,3,1]);let r=this.size[0]*s[2],a=this.size[1]*s[3],o=this.interpolation==="nearest"?Se.resizeNearestNeighbor(n,[r,a]):Se.resizeBilinear(n,[r,a]);return et(o,[0,3,1,2])}else{let r=this.size[0]*s[1],a=this.size[1]*s[2];return this.interpolation==="nearest"?Se.resizeNearestNeighbor(n,[r,a]):Se.resizeBilinear(n,[r,a])}})}getConfig(){let e={size:this.size,dataFormat:this.dataFormat,interpolation:this.interpolation},t=super.getConfig();return Object.assign(e,t),e}};_5.className="UpSampling2D";de.registerClass(_5);function gG(e,t,n=[1,1],s="valid",r,a){return Y(()=>{r==null&&(r=Dr()),Yt(r);let o=S5(e,r);if(e.rank!==4)throw new j(`Input for depthwiseConv2d is required to be 4-D, but is instead ${e.rank}-D`);if(t.rank!==4)throw new j(`depthwiseKernel is required to be 4-D, but is instead ${t.rank}-D`);return o=Vc(o,t,n,s==="same"?"same":"valid","NHWC",a),r==="channelsFirst"&&(o=et(o,[0,3,1,2])),o})}var D5=class extends C5{constructor(e){super(2,e),this.depthwiseKernel=null,this.depthMultiplier=e.depthMultiplier==null?1:e.depthMultiplier,this.depthwiseInitializer=Ot(e.depthwiseInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.depthwiseConstraint=yn(e.depthwiseConstraint),this.depthwiseRegularizer=Mt(e.depthwiseRegularizer)}build(e){if(e=xt(e),e.length<4)throw new j(`Inputs to DepthwiseConv2D should have rank 4. Received input shape: ${JSON.stringify(e)}.`);let t=this.dataFormat==="channelsFirst"?1:3;if(e[t]==null||e[t]<0)throw new j(`The channel dimension of the inputs to DepthwiseConv2D should be defined, but is not (${e[t]}).`);let n=e[t],s=[this.kernelSize[0],this.kernelSize[1],n,this.depthMultiplier];this.depthwiseKernel=this.addWeight("depthwise_kernel",s,null,this.depthwiseInitializer,this.depthwiseRegularizer,!0,this.depthwiseConstraint),this.useBias?this.bias=this.addWeight("bias",[n*this.depthMultiplier],null,this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint):this.bias=null,this.built=!0}call(e,t){return Y(()=>{e=Xe(e);let n=gG(e,this.depthwiseKernel.read(),this.strides,this.padding,this.dataFormat,null);return this.useBias&&(n=Fr(n,this.bias.read(),this.dataFormat)),this.activation!=null&&(n=this.activation.apply(n)),n})}computeOutputShape(e){e=xt(e);let t=this.dataFormat==="channelsFirst"?e[2]:e[1],n=this.dataFormat==="channelsFirst"?e[3]:e[2],s=this.dataFormat==="channelsFirst"?e[1]*this.depthMultiplier:e[3]*this.depthMultiplier,r=Rr(t,this.kernelSize[0],this.padding,this.strides[0]),a=Rr(n,this.kernelSize[1],this.padding,this.strides[1]);return this.dataFormat==="channelsFirst"?[e[0],s,r,a]:[e[0],r,a,s]}getConfig(){let e=super.getConfig();return e.depthMultiplier=this.depthMultiplier,e.depthwiseInitializer=Vt(this.depthwiseInitializer),e.depthwiseRegularizer=St(this.depthwiseRegularizer),e.depthwiseConstraint=gn(this.depthwiseRegularizer),e}};D5.className="DepthwiseConv2D";de.registerClass(D5);function uk(e,t,n,s){if(Array.isArray(e)){if(t!=null||n!=null)throw new j("When inputs is an array, neither initialState or constants should be provided");s!=null&&(n=e.slice(e.length-s,e.length),e=e.slice(0,e.length-s)),e.length>1&&(t=e.slice(1,e.length)),e=e[0]}function r(a){return a==null||Array.isArray(a)?a:[a]}return t=r(t),n=r(n),{inputs:e,initialState:t,constants:n}}function ck(e,t,n,s=!1,r,a,o=!1,i=!1){return Y(()=>{let l=t.shape.length;if(l<3)throw new j(`Input should be at least 3D, but is ${l}D.`);let u=[1,0].concat(_r(2,l));if(t=et(t,u),a!=null)throw new qe("The rnn() functoin of the deeplearn.js backend does not support constants yet.");o&&console.warn("Backend rnn(): the unroll = true option is not applicable to the imperative deeplearn.js backend."),r!=null&&(r=ye(ye(r,"bool"),"float32"),r.rank===l-1&&(r=Kt(r,-1)),r=et(r,u)),s&&(t=Zs(t,0),r!=null&&(r=Zs(r,0)));let c=[],p,d=n,h=t.shape[0],f=En(t),m;r!=null&&(m=En(r));for(let y=0;y<h;++y){let x=f[y],A=Y(()=>e(x,d));if(r==null)p=A[0],d=A[1];else{let b=Y(()=>{let w=m[y],k=fe($s(w),w),S=ce(L(A[0],w),L(d[0],k)),E=d.map((R,$)=>ce(L(A[1][$],w),L(R,k)));return{output:S,newStates:E}});p=b.output,d=b.newStates}i&&c.push(p)}let g;return i&&(g=on(c,1)),[p,g,d]})}var Qr=class extends ut{constructor(e){super(e);let t;if(e.cell==null)throw new j("cell property is missing for the constructor of RNN.");if(Array.isArray(e.cell)?t=new f2({cells:e.cell}):t=e.cell,t.stateSize==null)throw new j("The RNN cell should have an attribute `stateSize` (tuple of integers, one integer per RNN state).");this.cell=t,this.returnSequences=e.returnSequences==null?!1:e.returnSequences,this.returnState=e.returnState==null?!1:e.returnState,this.goBackwards=e.goBackwards==null?!1:e.goBackwards,this._stateful=e.stateful==null?!1:e.stateful,this.unroll=e.unroll==null?!1:e.unroll,this.supportsMasking=!0,this.inputSpec=[new rn({ndim:3})],this.stateSpec=null,this.states_=null,this.numConstants=null,this.keptStates=[]}getStates(){if(this.states_==null){let e=Array.isArray(this.cell.stateSize)?this.cell.stateSize.length:1;return _r(0,e).map(t=>null)}else return this.states_}setStates(e){this.states_=e}computeOutputShape(e){F3(e)&&(e=e[0]),e=e;let t=this.cell.stateSize;Array.isArray(t)||(t=[t]);let n=t[0],s;if(this.returnSequences?s=[e[0],e[1],n]:s=[e[0],n],this.returnState){let r=[];for(let a of t)r.push([e[0],a]);return[s].concat(r)}else return s}computeMask(e,t){return Y(()=>{Array.isArray(t)&&(t=t[0]);let n=this.returnSequences?t:null;if(this.returnState){let s=this.states.map(r=>null);return[n].concat(s)}else return n})}get states(){if(this.states_==null){let e=Array.isArray(this.cell.stateSize)?this.cell.stateSize.length:1,t=[];for(let n=0;n<e;++n)t.push(null);return t}else return this.states_}set states(e){this.states_=e}build(e){if(this.numConstants!=null)throw new qe("Constants support is not implemented in RNN yet.");F3(e)&&(e=e[0]),e=e;let n=this.stateful?e[0]:null,s=e.slice(2);this.inputSpec[0]=new rn({shape:[n,null,...s]});let r=[e[0]].concat(e.slice(2));this.cell.build(r);let a;if(Array.isArray(this.cell.stateSize)?a=this.cell.stateSize:a=[this.cell.stateSize],this.stateSpec!=null){if(!v.arraysEqual(this.stateSpec.map(o=>o.shape[o.shape.length-1]),a))throw new j(`An initialState was passed that is not compatible with cell.stateSize. Received stateSpec=${this.stateSpec}; However cell.stateSize is ${this.cell.stateSize}`)}else this.stateSpec=a.map(o=>new rn({shape:[null,o]}));this.stateful&&this.resetStates()}resetStates(e,t=!1){Y(()=>{if(!this.stateful)throw new la("Cannot call resetStates() on an RNN Layer that is not stateful.");let n=this.inputSpec[0].shape[0];if(n==null)throw new j("If an RNN is stateful, it needs to know its batch size. Specify the batch size of your input tensors: \n- If using a Sequential model, specify the batch size by passing a `batchInputShape` option to your first layer.\n- If using the functional API, specify the batch size by passing a `batchShape` option to your Input layer.");if(this.states_==null)Array.isArray(this.cell.stateSize)?this.states_=this.cell.stateSize.map(s=>Wt([n,s])):this.states_=[Wt([n,this.cell.stateSize])];else if(e==null)ee(this.states_),this.keptStates!=null&&(ee(this.keptStates),this.keptStates=[]),Array.isArray(this.cell.stateSize)?this.states_=this.cell.stateSize.map(s=>Wt([n,s])):this.states_[0]=Wt([n,this.cell.stateSize]);else{if(Array.isArray(e)||(e=[e]),e.length!==this.states_.length)throw new j(`Layer ${this.name} expects ${this.states_.length} state(s), but it received ${e.length} state value(s). Input received: ${e}`);t===!0?this.keptStates.push(this.states_.slice()):ee(this.states_);for(let s=0;s<this.states_.length;++s){let r=e[s],a=Array.isArray(this.cell.stateSize)?this.cell.stateSize[s]:this.cell.stateSize,o=[n,a];if(!v.arraysEqual(r.shape,o))throw new j(`State ${s} is incompatible with layer ${this.name}: expected shape=${o}, received shape=${r.shape}`);this.states_[s]=r}}this.states_=this.states_.map(s=>bn(s.clone()))})}apply(e,t){let n=t==null?null:t.initialState,s=t==null?null:t.constants;t==null&&(t={});let r=uk(e,n,s,this.numConstants);e=r.inputs,n=r.initialState,s=r.constants;let a=[],o=[];if(n!=null){t.initialState=n,a=a.concat(n),this.stateSpec=[];for(let l of n)this.stateSpec.push(new rn({shape:l.shape}));o=o.concat(this.stateSpec)}if(s!=null&&(t.constants=s,a=a.concat(s),this.numConstants=s.length),a[0]instanceof Tr){let l=[e].concat(a),u=this.inputSpec.concat(o),c=this.inputSpec;this.inputSpec=u;let p=super.apply(l,t);return this.inputSpec=c,p}else return super.apply(e,t)}call(e,t){return Y(()=>{let n=t==null?null:t.mask,s=t==null?null:t.training,r=t==null?null:t.initialState;e=Xe(e),r==null&&(this.stateful?r=this.states_:r=this.getInitialState(e));let a=Array.isArray(this.cell.stateSize)?this.cell.stateSize.length:1;if(r.length!==a)throw new j(`RNN Layer has ${a} state(s) but was passed ${r.length} initial state(s).`);this.unroll&&console.warn("Ignoring unroll = true for RNN layer, due to imperative backend.");let o={training:s},l=ck((h,f)=>{let m=this.cell.call([h].concat(f),o);return[m[0],m.slice(1)]},e,r,this.goBackwards,n,null,this.unroll,this.returnSequences),u=l[0],c=l[1],p=l[2];this.stateful&&this.resetStates(p,s);let d=this.returnSequences?c:u;return this.returnState?[d].concat(p):d})}getInitialState(e){return Y(()=>{let t=Wt(e.shape);return t=ke(t,[1,2]),t=kh(t),Array.isArray(this.cell.stateSize)?this.cell.stateSize.map(n=>n>1?$3(t,[1,n]):t):this.cell.stateSize>1?[$3(t,[1,this.cell.stateSize])]:[t]})}get trainableWeights(){return this.trainable?this.cell.trainableWeights:[]}get nonTrainableWeights(){return this.trainable?this.cell.nonTrainableWeights:this.cell.weights}setFastWeightInitDuringBuild(e){super.setFastWeightInitDuringBuild(e),this.cell!=null&&this.cell.setFastWeightInitDuringBuild(e)}getConfig(){let e=super.getConfig(),t={returnSequences:this.returnSequences,returnState:this.returnState,goBackwards:this.goBackwards,stateful:this.stateful,unroll:this.unroll};this.numConstants!=null&&(t.numConstants=this.numConstants);let n=this.cell.getConfig();return this.getClassName()===Qr.className&&(t.cell={className:this.cell.getClassName(),config:n}),Object.assign({},n,e,t)}static fromConfig(e,t,n={}){let s=t.cell,r=Er(s,n);return new e(Object.assign(t,{cell:r}))}};Qr.className="RNN";de.registerClass(Qr);var _h=class extends ut{},p2=class extends _h{constructor(e){super(e),this.DEFAULT_ACTIVATION="tanh",this.DEFAULT_KERNEL_INITIALIZER="glorotNormal",this.DEFAULT_RECURRENT_INITIALIZER="orthogonal",this.DEFAULT_BIAS_INITIALIZER="zeros",this.units=e.units,vn(this.units,"units"),this.activation=ao(e.activation==null?this.DEFAULT_ACTIVATION:e.activation),this.useBias=e.useBias==null?!0:e.useBias,this.kernelInitializer=Ot(e.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.recurrentInitializer=Ot(e.recurrentInitializer||this.DEFAULT_RECURRENT_INITIALIZER),this.biasInitializer=Ot(e.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.kernelRegularizer=Mt(e.kernelRegularizer),this.recurrentRegularizer=Mt(e.recurrentRegularizer),this.biasRegularizer=Mt(e.biasRegularizer),this.kernelConstraint=yn(e.kernelConstraint),this.recurrentConstraint=yn(e.recurrentConstraint),this.biasConstraint=yn(e.biasConstraint),this.dropout=tc([1,so([0,e.dropout==null?0:e.dropout])]),this.recurrentDropout=tc([1,so([0,e.recurrentDropout==null?0:e.recurrentDropout])]),this.dropoutFunc=e.dropoutFunc,this.stateSize=this.units,this.dropoutMask=null,this.recurrentDropoutMask=null}build(e){e=xt(e),this.kernel=this.addWeight("kernel",[e[e.length-1],this.units],null,this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.recurrentKernel=this.addWeight("recurrent_kernel",[this.units,this.units],null,this.recurrentInitializer,this.recurrentRegularizer,!0,this.recurrentConstraint),this.useBias?this.bias=this.addWeight("bias",[this.units],null,this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint):this.bias=null,this.built=!0}call(e,t){return Y(()=>{if(e=e,e.length!==2)throw new j(`SimpleRNNCell expects 2 input Tensors, got ${e.length}.`);let n=e[1];e=e[0];let s=t.training==null?!1:t.training;0<this.dropout&&this.dropout<1&&this.dropoutMask==null&&(this.dropoutMask=oo({ones:()=>$s(e),rate:this.dropout,training:s,dropoutFunc:this.dropoutFunc})),0<this.recurrentDropout&&this.recurrentDropout<1&&this.recurrentDropoutMask==null&&(this.recurrentDropoutMask=oo({ones:()=>$s(n),rate:this.recurrentDropout,training:s,dropoutFunc:this.dropoutFunc}));let r,a=this.dropoutMask,o=this.recurrentDropoutMask;a!=null?r=jr(L(e,a),this.kernel.read()):r=jr(e,this.kernel.read()),this.bias!=null&&(r=Fr(r,this.bias.read())),o!=null&&(n=L(n,o));let i=ce(r,jr(n,this.recurrentKernel.read()));return this.activation!=null&&(i=this.activation.apply(i)),[i,i]})}getConfig(){let e=super.getConfig(),t={units:this.units,activation:ro(this.activation),useBias:this.useBias,kernelInitializer:Vt(this.kernelInitializer),recurrentInitializer:Vt(this.recurrentInitializer),biasInitializer:Vt(this.biasInitializer),kernelRegularizer:St(this.kernelRegularizer),recurrentRegularizer:St(this.recurrentRegularizer),biasRegularizer:St(this.biasRegularizer),activityRegularizer:St(this.activityRegularizer),kernelConstraint:gn(this.kernelConstraint),recurrentConstraint:gn(this.recurrentConstraint),biasConstraint:gn(this.biasConstraint),dropout:this.dropout,recurrentDropout:this.recurrentDropout};return Object.assign({},e,t)}};p2.className="SimpleRNNCell";de.registerClass(p2);var $5=class extends Qr{constructor(e){e.cell=new p2(e),super(e)}call(e,t){return Y(()=>{this.cell.dropoutMask!=null&&(ee(this.cell.dropoutMask),this.cell.dropoutMask=null),this.cell.recurrentDropoutMask!=null&&(ee(this.cell.recurrentDropoutMask),this.cell.recurrentDropoutMask=null);let n=t==null?null:t.mask,s=t==null?null:t.training,r=t==null?null:t.initialState;return super.call(e,{mask:n,training:s,initialState:r})})}static fromConfig(e,t){return new e(t)}};$5.className="SimpleRNN";de.registerClass($5);var h2=class extends _h{constructor(e){if(super(e),this.DEFAULT_ACTIVATION="tanh",this.DEFAULT_RECURRENT_ACTIVATION="hardSigmoid",this.DEFAULT_KERNEL_INITIALIZER="glorotNormal",this.DEFAULT_RECURRENT_INITIALIZER="orthogonal",this.DEFAULT_BIAS_INITIALIZER="zeros",e.resetAfter)throw new j("GRUCell does not support reset_after parameter set to true.");this.units=e.units,vn(this.units,"units"),this.activation=ao(e.activation===void 0?this.DEFAULT_ACTIVATION:e.activation),this.recurrentActivation=ao(e.recurrentActivation===void 0?this.DEFAULT_RECURRENT_ACTIVATION:e.recurrentActivation),this.useBias=e.useBias==null?!0:e.useBias,this.kernelInitializer=Ot(e.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.recurrentInitializer=Ot(e.recurrentInitializer||this.DEFAULT_RECURRENT_INITIALIZER),this.biasInitializer=Ot(e.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.kernelRegularizer=Mt(e.kernelRegularizer),this.recurrentRegularizer=Mt(e.recurrentRegularizer),this.biasRegularizer=Mt(e.biasRegularizer),this.kernelConstraint=yn(e.kernelConstraint),this.recurrentConstraint=yn(e.recurrentConstraint),this.biasConstraint=yn(e.biasConstraint),this.dropout=tc([1,so([0,e.dropout==null?0:e.dropout])]),this.recurrentDropout=tc([1,so([0,e.recurrentDropout==null?0:e.recurrentDropout])]),this.dropoutFunc=e.dropoutFunc,this.implementation=e.implementation,this.stateSize=this.units,this.dropoutMask=null,this.recurrentDropoutMask=null}build(e){e=xt(e);let t=e[e.length-1];this.kernel=this.addWeight("kernel",[t,this.units*3],null,this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.recurrentKernel=this.addWeight("recurrent_kernel",[this.units,this.units*3],null,this.recurrentInitializer,this.recurrentRegularizer,!0,this.recurrentConstraint),this.useBias?this.bias=this.addWeight("bias",[this.units*3],null,this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint):this.bias=null,this.built=!0}call(e,t){return Y(()=>{if(e=e,e.length!==2)throw new j(`GRUCell expects 2 input Tensors (inputs, h, c), got ${e.length}.`);let n=t.training==null?!1:t.training,s=e[1];e=e[0],0<this.dropout&&this.dropout<1&&this.dropoutMask==null&&(this.dropoutMask=oo({ones:()=>$s(e),rate:this.dropout,training:n,count:3,dropoutFunc:this.dropoutFunc})),0<this.recurrentDropout&&this.recurrentDropout<1&&this.recurrentDropoutMask==null&&(this.recurrentDropoutMask=oo({ones:()=>$s(s),rate:this.recurrentDropout,training:n,count:3,dropoutFunc:this.dropoutFunc}));let r=this.dropoutMask,a=this.recurrentDropoutMask,o,i,l;0<this.dropout&&this.dropout<1&&(e=L(e,r[0]));let u=jr(e,this.kernel.read());this.useBias&&(u=Fr(u,this.bias.read())),0<this.recurrentDropout&&this.recurrentDropout<1&&(s=L(s,a[0]));let c=this.recurrentKernel.read(),[p,d]=Zt(c,[2*this.units,this.units],c.rank-1),h=jr(s,p),[f,m,g]=Zt(u,3,u.rank-1),[y,x]=Zt(h,2,h.rank-1);o=this.recurrentActivation.apply(ce(f,y)),i=this.recurrentActivation.apply(ce(m,x));let A=jr(L(i,s),d);l=this.activation.apply(ce(g,A));let b=ce(L(o,s),L(ce(1,Dt(o)),l));return[b,b]})}getConfig(){let e=super.getConfig(),t={units:this.units,activation:ro(this.activation),recurrentActivation:ro(this.recurrentActivation),useBias:this.useBias,kernelInitializer:Vt(this.kernelInitializer),recurrentInitializer:Vt(this.recurrentInitializer),biasInitializer:Vt(this.biasInitializer),kernelRegularizer:St(this.kernelRegularizer),recurrentRegularizer:St(this.recurrentRegularizer),biasRegularizer:St(this.biasRegularizer),activityRegularizer:St(this.activityRegularizer),kernelConstraint:gn(this.kernelConstraint),recurrentConstraint:gn(this.recurrentConstraint),biasConstraint:gn(this.biasConstraint),dropout:this.dropout,recurrentDropout:this.recurrentDropout,implementation:this.implementation,resetAfter:!1};return Object.assign({},e,t)}};h2.className="GRUCell";de.registerClass(h2);var P5=class extends Qr{constructor(e){e.implementation===0&&console.warn("`implementation=0` has been deprecated, and now defaults to `implementation=1`. Please update your layer call."),e.cell=new h2(e),super(e)}call(e,t){return Y(()=>{this.cell.dropoutMask!=null&&(ee(this.cell.dropoutMask),this.cell.dropoutMask=null),this.cell.recurrentDropoutMask!=null&&(ee(this.cell.recurrentDropoutMask),this.cell.recurrentDropoutMask=null);let n=t==null?null:t.mask,s=t==null?null:t.training,r=t==null?null:t.initialState;return super.call(e,{mask:n,training:s,initialState:r})})}static fromConfig(e,t){return t.implmentation===0&&(t.implementation=1),new e(t)}};P5.className="GRU";de.registerClass(P5);var Dh=class extends _h{constructor(e){super(e),this.DEFAULT_ACTIVATION="tanh",this.DEFAULT_RECURRENT_ACTIVATION="hardSigmoid",this.DEFAULT_KERNEL_INITIALIZER="glorotNormal",this.DEFAULT_RECURRENT_INITIALIZER="orthogonal",this.DEFAULT_BIAS_INITIALIZER="zeros",this.units=e.units,vn(this.units,"units"),this.activation=ao(e.activation===void 0?this.DEFAULT_ACTIVATION:e.activation),this.recurrentActivation=ao(e.recurrentActivation===void 0?this.DEFAULT_RECURRENT_ACTIVATION:e.recurrentActivation),this.useBias=e.useBias==null?!0:e.useBias,this.kernelInitializer=Ot(e.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.recurrentInitializer=Ot(e.recurrentInitializer||this.DEFAULT_RECURRENT_INITIALIZER),this.biasInitializer=Ot(e.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.unitForgetBias=e.unitForgetBias,this.kernelRegularizer=Mt(e.kernelRegularizer),this.recurrentRegularizer=Mt(e.recurrentRegularizer),this.biasRegularizer=Mt(e.biasRegularizer),this.kernelConstraint=yn(e.kernelConstraint),this.recurrentConstraint=yn(e.recurrentConstraint),this.biasConstraint=yn(e.biasConstraint),this.dropout=tc([1,so([0,e.dropout==null?0:e.dropout])]),this.recurrentDropout=tc([1,so([0,e.recurrentDropout==null?0:e.recurrentDropout])]),this.dropoutFunc=e.dropoutFunc,this.implementation=e.implementation,this.stateSize=[this.units,this.units],this.dropoutMask=null,this.recurrentDropoutMask=null}build(e){var t;e=xt(e);let n=e[e.length-1];this.kernel=this.addWeight("kernel",[n,this.units*4],null,this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.recurrentKernel=this.addWeight("recurrent_kernel",[this.units,this.units*4],null,this.recurrentInitializer,this.recurrentRegularizer,!0,this.recurrentConstraint);let s;if(this.useBias){if(this.unitForgetBias){let r=this.biasInitializer,a=this.units;s=new(t=class extends hr{apply(i,l){let u=r.apply([a]),c=new t2().apply([a]),p=r.apply([a*2]);return Ov(Ov(u,c),p)}},t.className="CustomInit",t)}else s=this.biasInitializer;this.bias=this.addWeight("bias",[this.units*4],null,s,this.biasRegularizer,!0,this.biasConstraint)}else this.bias=null;this.built=!0}call(e,t){return Y(()=>{let n=t.training==null?!1:t.training;if(e=e,e.length!==3)throw new j(`LSTMCell expects 3 input Tensors (inputs, h, c), got ${e.length}.`);let s=e[1],r=e[2];e=e[0],0<this.dropout&&this.dropout<1&&this.dropoutMask==null&&(this.dropoutMask=oo({ones:()=>$s(e),rate:this.dropout,training:n,count:4,dropoutFunc:this.dropoutFunc})),0<this.recurrentDropout&&this.recurrentDropout<1&&this.recurrentDropoutMask==null&&(this.recurrentDropoutMask=oo({ones:()=>$s(s),rate:this.recurrentDropout,training:n,count:4,dropoutFunc:this.dropoutFunc}));let a=this.dropoutMask,o=this.recurrentDropoutMask,i,l,u,c;0<this.dropout&&this.dropout<1&&(e=L(e,a[0]));let p=jr(e,this.kernel.read());0<this.recurrentDropout&&this.recurrentDropout<1&&(s=L(s,o[0])),p=ce(p,jr(s,this.recurrentKernel.read())),this.useBias&&(p=Fr(p,this.bias.read()));let[d,h,f,m]=Zt(p,4,p.rank-1);i=this.recurrentActivation.apply(d),l=this.recurrentActivation.apply(h),u=ce(L(l,r),L(i,this.activation.apply(f))),c=this.recurrentActivation.apply(m);let g=L(c,this.activation.apply(u));return[g,g,u]})}getConfig(){let e=super.getConfig(),t={units:this.units,activation:ro(this.activation),recurrentActivation:ro(this.recurrentActivation),useBias:this.useBias,kernelInitializer:Vt(this.kernelInitializer),recurrentInitializer:Vt(this.recurrentInitializer),biasInitializer:Vt(this.biasInitializer),unitForgetBias:this.unitForgetBias,kernelRegularizer:St(this.kernelRegularizer),recurrentRegularizer:St(this.recurrentRegularizer),biasRegularizer:St(this.biasRegularizer),activityRegularizer:St(this.activityRegularizer),kernelConstraint:gn(this.kernelConstraint),recurrentConstraint:gn(this.recurrentConstraint),biasConstraint:gn(this.biasConstraint),dropout:this.dropout,recurrentDropout:this.recurrentDropout,implementation:this.implementation};return Object.assign({},e,t)}};Dh.className="LSTMCell";de.registerClass(Dh);var F5=class extends Qr{constructor(e){e.implementation===0&&console.warn("`implementation=0` has been deprecated, and now defaults to `implementation=1`. Please update your layer call."),e.cell=new Dh(e),super(e)}call(e,t){return Y(()=>{this.cell.dropoutMask!=null&&(ee(this.cell.dropoutMask),this.cell.dropoutMask=null),this.cell.recurrentDropoutMask!=null&&(ee(this.cell.recurrentDropoutMask),this.cell.recurrentDropoutMask=null);let n=t==null?null:t.mask,s=t==null?null:t.training,r=t==null?null:t.initialState;return super.call(e,{mask:n,training:s,initialState:r})})}static fromConfig(e,t){return t.implmentation===0&&(t.implementation=1),new e(t)}};F5.className="LSTM";de.registerClass(F5);var f2=class extends _h{constructor(e){super(e),this.cells=e.cells}get stateSize(){let e=[];for(let t of this.cells.slice().reverse())Array.isArray(t.stateSize)?e.push(...t.stateSize):e.push(t.stateSize);return e}call(e,t){return Y(()=>{e=e;let n=e.slice(1),s=[];for(let o of this.cells.slice().reverse())Array.isArray(o.stateSize)?s.push(n.splice(0,o.stateSize.length)):s.push(n.splice(0,1));s.reverse();let r=[],a;for(let o=0;o<this.cells.length;++o){let i=this.cells[o];n=s[o],o===0?a=[e[0]].concat(n):a=[a[0]].concat(n),a=i.call(a,t),r.push(a.slice(1))}n=[];for(let o of r.slice().reverse())n.push(...o);return[a[0]].concat(n)})}build(e){F3(e)&&(e=e[0]),e=e;let t;this.cells.forEach((n,s)=>{Hi(`RNNCell_${s}`,()=>{n.build(e),Array.isArray(n.stateSize)?t=n.stateSize[0]:t=n.stateSize,e=[e[0],t]})}),this.built=!0}getConfig(){let e=super.getConfig(),t=r=>({className:r.getClassName(),config:r.getConfig()}),s={cells:this.cells.map(t)};return Object.assign({},e,s)}static fromConfig(e,t,n={}){let s=[];for(let r of t.cells)s.push(Er(r,n));return new e({cells:s})}get trainableWeights(){if(!this.trainable)return[];let e=[];for(let t of this.cells)e.push(...t.trainableWeights);return e}get nonTrainableWeights(){let e=[];for(let t of this.cells)e.push(...t.nonTrainableWeights);if(!this.trainable){let t=[];for(let n of this.cells)t.push(...n.trainableWeights);return t.concat(e)}return e}getWeights(){let e=[];for(let t of this.cells)e.push(...t.weights);return O3(e)}setWeights(e){let t=[];for(let n of this.cells){let s=n.weights.length,r=e.splice(s);for(let a=0;a<n.weights.length;++a)t.push([n.weights[a],r[a]])}a5(t)}};f2.className="StackedRNNCells";de.registerClass(f2);function oo(e){let{ones:t,rate:n,training:s=!1,count:r=1,dropoutFunc:a}=e,o=()=>a!=null?a(t(),n):w8(t(),n),i=()=>Sh(o,t,s);return!r||r<=1?bn(i().clone()):Array(r).fill(void 0).map(i).map(u=>bn(u.clone()))}var yG=function(e,t){var n={};for(var s in e)Object.prototype.hasOwnProperty.call(e,s)&&t.indexOf(s)<0&&(n[s]=e[s]);if(e!=null&&typeof Object.getOwnPropertySymbols=="function")for(var r=0,s=Object.getOwnPropertySymbols(e);r<s.length;r++)t.indexOf(s[r])<0&&Object.prototype.propertyIsEnumerable.call(e,s[r])&&(n[s[r]]=e[s[r]]);return n},dk=class extends Qr{constructor(e){if(e.unroll)throw new qe("Unrolling is not possible with convolutional RNNs.");if(Array.isArray(e.cell))throw new qe("It is not possible at the moment to stack convolutional cells.");super(e),this.inputSpec=[new rn({ndim:5})]}call(e,t){return Y(()=>{if(this.cell.dropoutMask!=null&&(ee(this.cell.dropoutMask),this.cell.dropoutMask=null),this.cell.recurrentDropoutMask!=null&&(ee(this.cell.recurrentDropoutMask),this.cell.recurrentDropoutMask=null),t&&t.constants)throw new j("ConvRNN2D cell does not support constants");let n=t==null?null:t.mask,s=t==null?null:t.training,r=t==null?null:t.initialState;return super.call(e,{mask:n,training:s,initialState:r})})}computeOutputShape(e){let t=this.computeSingleOutputShape(e);return this.returnSequences||(t=[t[0],...t.slice(2)]),this.returnState&&(t=[t,...Array(2).fill([e[0],...t.slice(-3)])]),t}getInitialState(e){return Y(()=>{let{stateSize:t}=this.cell,n=e.shape,s=this.computeSingleOutputShape(n),r=[s[0],...s.slice(2)],a=Wt(r);return Array.isArray(t)?Array(t.length).fill(a):[a]})}resetStates(e,t=!1){Y(()=>{if(!this.stateful)throw new la("Cannot call resetStates() on an RNN Layer that is not stateful.");let n=this.inputSpec[0].shape,s=this.computeSingleOutputShape(n),r=[s[0],...s.slice(2)];if(n[0]==null)throw new j("If an RNN is stateful, it needs to know its batch size. Specify the batch size of your input tensors: \n- If using a Sequential model, specify the batch size by passing a `batchInputShape` option to your first layer.\n- If using the functional API, specify the batch size by passing a `batchShape` option to your Input layer.");if(this.getStates()==null)Array.isArray(this.cell.stateSize)?this.states_=this.cell.stateSize.map(()=>Wt(r)):this.states_=[Wt(r)];else if(e==null)ee(this.states_),this.keptStates!=null&&(ee(this.keptStates),this.keptStates=[]),Array.isArray(this.cell.stateSize)?this.states_=this.cell.stateSize.map(()=>Wt(r)):this.states_[0]=Wt(r);else{if(Array.isArray(e)||(e=[e]),e.length!==this.states_.length)throw new j(`Layer ${this.name} expects ${this.states_.length} state(s), but it received ${e.length} state value(s). Input received: ${e}`);t?this.keptStates.push(this.states_.slice()):ee(this.states_);for(let o=0;o<this.states_.length;++o){let i=e[o],l=r;if(!v.arraysEqual(i.shape,l))throw new j(`State ${o} is incompatible with layer ${this.name}: expected shape=${l}, received shape=${i.shape}`);this.states_[o]=i}}this.states_=this.states_.map(o=>bn(o.clone()))})}computeSingleOutputShape(e){let{dataFormat:t,filters:n,kernelSize:s,padding:r,strides:a,dilationRate:o}=this.cell,i=t==="channelsFirst",l=e[i?3:2],u=e[i?4:3],c=Rr(l,s[0],r,a[0],o[0]),p=Rr(u,s[1],r,a[1],o[1]);return[...e.slice(0,2),...i?[n,c,p]:[c,p,n]]}};dk.className="ConvRNN2D";var m2=class extends Dh{constructor(e){let{filters:t,kernelSize:n,strides:s,padding:r,dataFormat:a,dilationRate:o}=e;super(Object.assign({},e,{units:t})),this.filters=t,vn(this.filters,"filters"),this.kernelSize=Hu(n,2,"kernelSize"),this.kernelSize.forEach(i=>vn(i,"kernelSize")),this.strides=Hu(s||1,2,"strides"),this.strides.forEach(i=>vn(i,"strides")),this.padding=r||"valid",er(this.padding),this.dataFormat=a||"channelsLast",Yt(this.dataFormat),this.dilationRate=Hu(o||1,2,"dilationRate"),this.dilationRate.forEach(i=>vn(i,"dilationRate"))}build(e){var t;e=xt(e);let n=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[n]==null)throw new j(`The channel dimension of the input should be defined. Found ${e[n]}`);let s=e[n],r=4,a=this.kernelSize.concat([s,this.filters*r]);this.kernel=this.addWeight("kernel",a,null,this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint);let o=this.kernelSize.concat([this.filters,this.filters*r]);if(this.recurrentKernel=this.addWeight("recurrent_kernel",o,null,this.recurrentInitializer,this.recurrentRegularizer,!0,this.recurrentConstraint),this.useBias){let i;if(this.unitForgetBias){let l=this.biasInitializer,u=this.filters;i=new(t=class extends hr{apply(p,d){let h=l.apply([u]),f=Ns([u]),m=l.apply([u*2]);return YA([h,f,m])}},t.className="CustomInit",t)}else i=this.biasInitializer;this.bias=this.addWeight("bias",[this.filters*r],null,i,this.biasRegularizer,!0,this.biasConstraint)}this.built=!0}call(e,t){return Y(()=>{if(e.length!==3)throw new j(`ConvLSTM2DCell expects 3 input Tensors (inputs, h, c), got ${e.length}.`);let n=t.training||!1,s=e[0],r=e[1],a=e[2],o=4;0<this.dropout&&this.dropout<1&&this.dropoutMask==null&&(this.dropoutMask=oo({ones:()=>$s(s),rate:this.dropout,training:n,count:o,dropoutFunc:this.dropoutFunc}));let i=this.dropoutMask,l=(J,te,B)=>!te||!te[B]?J:L(te[B],J),u=l(s,i,0),c=l(s,i,1),p=l(s,i,2),d=l(s,i,3);0<this.recurrentDropout&&this.recurrentDropout<1&&this.recurrentDropoutMask==null&&(this.recurrentDropoutMask=oo({ones:()=>$s(r),rate:this.recurrentDropout,training:n,count:o,dropoutFunc:this.dropoutFunc}));let h=this.recurrentDropoutMask,f=l(r,h,0),m=l(r,h,1),g=l(r,h,2),y=l(r,h,3),x=3,[A,b,w,k]=Zt(this.kernel.read(),o,x),[S,E,R,$]=this.useBias?Zt(this.bias.read(),o):[null,null,null,null];u=this.inputConv(u,A,S,this.padding),c=this.inputConv(c,b,E,this.padding),p=this.inputConv(p,w,R,this.padding),d=this.inputConv(d,k,$,this.padding);let[_,P,C,F]=Zt(this.recurrentKernel.read(),o,x);f=this.recurrentConv(f,_),m=this.recurrentConv(m,P),g=this.recurrentConv(g,C),y=this.recurrentConv(y,F);let G=this.recurrentActivation.apply(ce(u,f)),K=this.recurrentActivation.apply(ce(c,m)),z=ce(L(K,a),L(G,this.activation.apply(ce(p,g)))),Z=L(this.recurrentActivation.apply(ce(d,y)),this.activation.apply(z));return[Z,Z,z]})}getConfig(){let e=super.getConfig(),{units:t}=e,n=yG(e,["units"]),s={filters:this.filters,kernelSize:this.kernelSize,padding:this.padding,dataFormat:this.dataFormat,dilationRate:this.dilationRate,strides:this.strides};return Object.assign({},n,s)}inputConv(e,t,n,s){let r=ma(e,t,this.strides,s||"valid",this.dataFormat==="channelsFirst"?"NCHW":"NHWC",this.dilationRate);return n?Fr(r,n,this.dataFormat):r}recurrentConv(e,t){return ma(e,t,1,"same",this.dataFormat==="channelsFirst"?"NCHW":"NHWC")}};m2.className="ConvLSTM2DCell";de.registerClass(m2);var O5=class extends dk{constructor(e){let t=new m2(e);super(Object.assign({},e,{cell:t}))}static fromConfig(e,t){return new e(t)}};O5.className="ConvLSTM2D";de.registerClass(O5);var g2=class extends ut{constructor(e){super(e),this.rate=Math.max(Math.min(e.rate,1),0),this.noiseShape=e.noiseShape,this.seed=e.seed,this.supportsMasking=!0}getNoiseShape(e){if(this.noiseShape==null)return this.noiseShape;let t=e.shape,n=[];for(let s=0;s<this.noiseShape.length;++s)n.push(this.noiseShape[s]==null?t[s]:this.noiseShape[s]);return n}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e);if(0<this.rate&&this.rate<1){let s=t.training==null?!1:t.training,r=this.getNoiseShape(n);return Sh(()=>w8(n,this.rate,r,this.seed),()=>n,s)}return e})}getConfig(){let e={rate:this.rate,noiseShape:this.noiseShape,seed:this.seed},t=super.getConfig();return Object.assign(e,t),e}dispose(){return super.dispose()}};g2.className="Dropout";de.registerClass(g2);var M5=class extends g2{constructor(e){super(e),this.inputSpec=[{ndim:3}]}getNoiseShape(e){let t=e.shape;return[t[0],1,t[2]]}};M5.className="SpatialDropout1D";de.registerClass(M5);var z5=class extends ut{constructor(e){if(super(e),this.activation=null,this.useBias=!0,this.kernel=null,this.bias=null,this.DEFAULT_KERNEL_INITIALIZER="glorotNormal",this.DEFAULT_BIAS_INITIALIZER="zeros",e.batchInputShape==null&&e.inputShape==null&&e.inputDim!=null){let t=null;e.batchSize!=null&&(t=e.batchSize),this.batchInputShape=[t,e.inputDim]}this.units=e.units,vn(this.units,"units"),this.activation=ao(e.activation),e.useBias!=null&&(this.useBias=e.useBias),this.kernelInitializer=Ot(e.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.biasInitializer=Ot(e.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.kernelConstraint=yn(e.kernelConstraint),this.biasConstraint=yn(e.biasConstraint),this.kernelRegularizer=Mt(e.kernelRegularizer),this.biasRegularizer=Mt(e.biasRegularizer),this.activityRegularizer=Mt(e.activityRegularizer),this.supportsMasking=!0,this.inputSpec=[{minNDim:2}]}build(e){e=xt(e);let t=e[e.length-1];this.kernel==null&&(this.kernel=this.addWeight("kernel",[t,this.units],null,this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.useBias&&(this.bias=this.addWeight("bias",[this.units],null,this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint))),this.inputSpec=[{minNDim:2,axes:{[-1]:t}}],this.built=!0}computeOutputShape(e){e=xt(e);let t=e.slice();return t[t.length-1]=this.units,t}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e),s=m8(this.activation.getClassName()),r;return s!=null?r=jr(n,this.kernel.read(),s,this.bias?this.bias.read():null):(r=jr(n,this.kernel.read()),this.bias!=null&&(r=Fr(r,this.bias.read())),this.activation!=null&&(r=this.activation.apply(r))),r})}getConfig(){let e={units:this.units,activation:ro(this.activation),useBias:this.useBias,kernelInitializer:Vt(this.kernelInitializer),biasInitializer:Vt(this.biasInitializer),kernelRegularizer:St(this.kernelRegularizer),biasRegularizer:St(this.biasRegularizer),activityRegularizer:St(this.activityRegularizer),kernelConstraint:gn(this.kernelConstraint),biasConstraint:gn(this.biasConstraint)},t=super.getConfig();return Object.assign(e,t),e}};z5.className="Dense";de.registerClass(z5);var L5=class extends ut{constructor(e){e=e||{},super(e),this.inputSpec=[{minNDim:3}],this.dataFormat=e.dataFormat}computeOutputShape(e){e=xt(e);for(let t of e.slice(1))if(t==null)throw new j(`The shape of the input to "Flatten" is not fully defined (got ${e.slice(1)}). Make sure to pass a complete "input_shape" or "batch_input_shape" argument to the first layer in your model.`);return[e[0],Ka(e,1)]}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e);if(this.dataFormat==="channelsFirst"&&n.rank>1){let s=[0];for(let r=2;r<n.rank;++r)s.push(r);s.push(1),n=et(n,s)}return _V(n)})}getConfig(){let e={};this.dataFormat!=null&&(e.dataFormat=this.dataFormat);let t=super.getConfig();return Object.assign(e,t),e}};L5.className="Flatten";de.registerClass(L5);var B5=class extends ut{constructor(e){super(e),this.supportsMasking=!0,this.activation=ao(e.activation)}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e);return this.activation.apply(n)})}getConfig(){let e={activation:ro(this.activation)},t=super.getConfig();return Object.assign(e,t),e}};B5.className="Activation";de.registerClass(B5);var W5=class extends ut{constructor(e){super(e),this.n=e.n,this.inputSpec=[{ndim:2}]}computeOutputShape(e){return[e[0],this.n,e[1]]}call(e,t){return Y(()=>(e=Xe(e),EV(e,this.n)))}getConfig(){let e={n:this.n},t=super.getConfig();return Object.assign(e,t),e}};W5.className="RepeatVector";de.registerClass(W5);var V5=class extends ut{constructor(e){super(e),this.targetShape=e.targetShape;for(let t=0;t<this.targetShape.length;++t)this.isUnknown(this.targetShape[t])&&(this.targetShape[t]=null)}isUnknown(e){return e<0||e==null}fixUnknownDimension(e,t){let n="Total size of new array must be unchanged.",s=t.slice(),r=1,a=null;for(let i=0;i<s.length;++i){let l=s[i];if(this.isUnknown(l))if(a===null)a=i;else throw new j("Can only specifiy one unknown dimension.");else r*=l}let o=Ka(e);if(a!==null){if(r===0||o%r!==0)throw new j(n);s[a]=o/r}else if(o!==r)throw new j(n);return s}computeOutputShape(e){let t=!1;for(let n=0;n<e.length;++n)if(this.isUnknown(e[n])){t=!0;break}return t?e.slice(0,1).concat(this.targetShape):e.slice(0,1).concat(this.fixUnknownDimension(e.slice(1),this.targetShape))}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e),s=n.shape,r=s.slice(0,1).concat(this.fixUnknownDimension(s.slice(1),this.targetShape));return U(n,r)})}getConfig(){let e={targetShape:this.targetShape},t=super.getConfig();return Object.assign(e,t),e}};V5.className="Reshape";de.registerClass(V5);var U5=class extends ut{constructor(e){if(super(e),e.dims==null)throw new Error("Required configuration field `dims` is missing during Permute constructor call.");if(!Array.isArray(e.dims))throw new Error(`Permute constructor requires \`dims\` to be an Array, but received ${e.dims} instead.`);let t=_r(1,e.dims.length+1);if(!v.arraysEqual(e.dims.slice().sort(),t))throw new Error("Invalid permutation `dims`: "+JSON.stringify(e.dims)+" `dims` must contain consecutive integers starting from 1.");this.dims=e.dims,this.dimsIncludingBatch=[0].concat(this.dims),this.inputSpec=[new rn({ndim:this.dims.length+1})]}computeOutputShape(e){e=xt(e);let t=e.slice();return this.dims.forEach((n,s)=>{t[s+1]=e[n]}),t}call(e,t){return et(Xe(e),this.dimsIncludingBatch)}getConfig(){let e={dims:this.dims},t=super.getConfig();return Object.assign(e,t),e}};U5.className="Permute";de.registerClass(U5);var G5=class extends ut{constructor(e){super(e==null?{}:e),this.supportsMasking=!0,e!=null?this.maskValue=e.maskValue==null?0:e.maskValue:this.maskValue=0}computeOutputShape(e){return e}getConfig(){let e=super.getConfig(),t={maskValue:this.maskValue};return Object.assign(t,e),t}computeMask(e,t){let n=Xe(e),s=-1;return Cp(Qi(n,this.maskValue),s)}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e),s=-1,r=!0,a=Cp(Qi(n,this.maskValue),s,r);return L(n,ye(a,n.dtype))})}};G5.className="Masking";de.registerClass(G5);var H5=class extends ut{constructor(e){if(super(e),this.embeddings=null,this.DEFAULT_EMBEDDINGS_INITIALIZER="randomUniform",e.batchInputShape==null&&e.inputShape==null){let t=null;e.batchSize!=null&&(t=e.batchSize),e.inputLength==null?this.batchInputShape=[t,null]:this.batchInputShape=[t].concat(_t(e.inputLength))}this.inputDim=e.inputDim,vn(this.inputDim,"inputDim"),this.outputDim=e.outputDim,vn(this.outputDim,"outputDim"),this.embeddingsInitializer=Ot(e.embeddingsInitializer||this.DEFAULT_EMBEDDINGS_INITIALIZER),this.embeddingsRegularizer=Mt(e.embeddingsRegularizer),this.activityRegularizer=Mt(e.activityRegularizer),this.embeddingsConstraint=yn(e.embeddingsConstraint),this.maskZero=e.maskZero,this.supportsMasking=e.maskZero,this.inputLength=e.inputLength}build(e){this.embeddings=this.addWeight("embeddings",[this.inputDim,this.outputDim],this.dtype,this.embeddingsInitializer,this.embeddingsRegularizer,!0,this.embeddingsConstraint),this.built=!0}warnOnIncompatibleInputShape(e){}computeMask(e,t){return Y(()=>this.maskZero?(e=Xe(e),Qi(e,it(e))):null)}computeOutputShape(e){if(e=xt(e),this.inputLength==null)return[...e,this.outputDim];let t=_t(this.inputLength);if(t.length!==e.length-1)throw new j(`"inputLength" is ${this.inputLength}, but received input shape has shape ${e}`);{let n=0;for(let s=0;s<t.length;++s){let r=t[s],a=e[s+1];if(r!=null&&a!=null&&r!==a)throw new j(`"inputLength" is ${this.inputLength}, but received input shape has shape ${e}`);r==null&&(t[n]=a),n++}}return[e[0],...t,this.outputDim]}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e);n.dtype!=="int32"&&(n=Q0(n,"int32"));let s=v8(this.embeddings.read(),U(n,[n.size]));return U(s,xt(this.computeOutputShape(n.shape)))})}getConfig(){let e={inputDim:this.inputDim,outputDim:this.outputDim,embeddingsInitializer:Vt(this.embeddingsInitializer),embeddingsRegularizer:St(this.embeddingsRegularizer),activityRegularizer:St(this.activityRegularizer),embeddingsConstraint:gn(this.embeddingsConstraint),maskZero:this.maskZero,inputLength:this.inputLength},t=super.getConfig();return Object.assign(e,t),e}};H5.className="Embedding";de.registerClass(H5);var nu=class extends ut{constructor(e){super(e||{}),this.supportsMasking=!0}mergeFunction(e){throw new qe}computeElementwiseOpOutputShape(e,t){if(e==null||t==null)return null;if(e.length<t.length)return this.computeElementwiseOpOutputShape(t,e);if(t.length===0)return e;let n=e.slice(0,e.length-t.length);for(let s=0;s<t.length;++s){let r=e[e.length-t.length+s],a=t[s];if(r==null||a==null||r<0||a<0)n.push(null);else if(r===1)n.push(a);else if(a===1)n.push(r);else{if(r!==a)throw new j("Operands could not be broadcast together with shapes "+JSON.stringify(e)+" "+JSON.stringify(t));n.push(r)}}return n}build(e){if(Array.isArray(e)&&!Array.isArray(e[0])&&(e=[xt(e)]),e=e,e.length<2)throw new j(`A merge layer should be called on an Array of at least 2 inputs. Got ${e.length} input(s).`);let t=[];for(let r of e)r!=null&&r[0]!==null&&t.push(r[0]);if(t=Xa(t),t.length>1)throw new j(`Can not merge tensors with different batch sizes. Got tensors with shapes: ${JSON.stringify(e)}.`);let n=e[0]==null?null:e[0].slice(1);for(let r=1;r<e.length;++r){let a=e[r]==null?null:e[r].slice(1);n=this.computeElementwiseOpOutputShape(n,a)}let s=e.map(r=>r.length);e.indexOf(null)===-1&&Xa(s).length===1?this.reshapeRequired=!1:this.reshapeRequired=!0}call(e,t){return Y(()=>{if(e=e,this.reshapeRequired){let n=[],s=e.map(r=>r.rank);if(s.indexOf(null)===-1){let r=so(s);for(let a of e){let o=a.rank;for(let i=0;i<r-o;++i)a=kh(a,1);n.push(a)}return this.mergeFunction(n)}else{let r=!1;for(let i of e){let l=i.rank;if(l==null){let u=i.shape,c=u[0],p=u.slice(1).concat([c]),d=U(i,[c].concat(Ka(u.slice(1))));d=et(d,[1,0]),d=U(d,p),n.push(d),r=!0}else if(l>1){let u=_r(1,l).concat([0]);n.push(et(i,u)),r=!0}else n.push(i)}let a=this.mergeFunction(n),o=a.rank;if(r){if(o==null){let i=a.shape,l=i.length,u=i[l-1],c=[u].concat(i.slice(0,i.length-1));a=U(et(U(a,[-1,u]),[1,0]),c)}else if(o>1){let i=[o-1].concat(_r(0,o-1));a=et(a,i)}}return a}}else return this.mergeFunction(e)})}computeOutputShape(e){e=e;let t;e[0]==null?t=null:t=e[0].slice(1);for(let s=1;s<e.length;++s){let r=e[s]==null?null:e[s].slice(1);t=this.computeElementwiseOpOutputShape(t,r)}let n=[];for(let s of e)s!=null&&s[0]!==null&&n.push(s[0]);return n=Xa(n),n.length===1?t=n.concat(t):t=[null].concat(t),t}computeMask(e,t){return Y(()=>{if(t==null)return null;if(!Array.isArray(t))throw new j("`mask` should be an Array");if(!Array.isArray(e))throw new j("`inputs` should be an Array");if(t.length!==e.length)throw new j(`The Array 'inputs' and 'mask' are expected to have the same length, but have different lengths (${e.length} vs ${t.length})`);if(t.every(s=>s==null))return null;t=t.map(s=>s==null?s:Kt(s,0));let n=t[0];for(let s=1;s<t.length-1;++s)n=ur(n,t[s]);return n})}},j5=class extends nu{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0].clone();for(let n=1;n<e.length;++n)t=ce(t,e[n]);return t})}};j5.className="Add";de.registerClass(j5);var q5=class extends nu{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0].clone();for(let n=1;n<e.length;++n)t=L(t,e[n]);return t})}};q5.className="Multiply";de.registerClass(q5);var X5=class extends nu{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0].clone();for(let n=1;n<e.length;++n)t=ce(t,e[n]);return L(1/e.length,t)})}};X5.className="Average";de.registerClass(X5);var K5=class extends nu{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0];for(let n=1;n<e.length;++n)t=Jr(t,e[n]);return t})}};K5.className="Maximum";de.registerClass(K5);var Z5=class extends nu{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0];for(let n=1;n<e.length;++n)t=Xc(t,e[n]);return t})}};Z5.className="Minimum";de.registerClass(Z5);var Y5=class extends nu{constructor(e){super(e),this.DEFAULT_AXIS=-1,e==null&&(e={}),this.axis=e.axis==null?this.DEFAULT_AXIS:e.axis,this.supportsMasking=!0,this.reshapeRequired=!1}build(e){if(!(Array.isArray(e)&&Array.isArray(e[0]))||e.length===1)throw new j("A `Concatenate` layer should be called on a list of at least 2 inputs");e=e;let t=!0;for(let s of e)if(s!=null){t=!1;break}if(t)return;let n=[];for(let s=0;s<e.length;++s){let r=e[s].slice();r.splice(this.axis,1);let a=!1;for(let o of n)if(v.arraysEqual(o,r)){a=!0;break}a||n.push(r)}if(n.length>1)throw new j("A `Concatenate` layer requires inputs with matching shapes except for the concat axis. Got input shapes: "+JSON.stringify(e))}mergeFunction(e){return Y(()=>YA(e,this.axis))}computeOutputShape(e){if(!(Array.isArray(e)&&Array.isArray(e[0])))throw new j("A `Concatenate` layer should be called on a list of inputs.");let t=e,n=t[0].slice(),s=this.axis<0?n.length+this.axis:this.axis;for(let r of t.slice(1)){if(n[s]==null||r[s]==null){n[s]=null;break}n[s]+=r[s]}return n}computeMask(e,t){if(t==null)return null;if(!Array.isArray(t))throw new j("`mask` should be an array for Concatenate");if(!Array.isArray(e))throw new j("`inputs` should be an array for Concatenate");if(t.length!==e.length)throw new j(`Mismatch in the length of mask (${t.length}) and the legnth of inputs (${e.length})`);return Y(()=>{let n=!0;if(t.forEach(a=>{if(a!=null){n=!1;return}}),n)return null;let s=[];for(let a=0;a<e.length;++a)t[a]==null?s.push(ye($s(e[a]),"bool")):t[a].rank<e[a].rank?s.push(Kt(t[a],-1)):s.push(t[a]);let r=Ct(s,this.axis);return g0(r,-1,!1)})}getConfig(){let e={axis:this.axis},t=super.getConfig();return Object.assign(e,t),e}};Y5.className="Concatenate";de.registerClass(Y5);function ep(e,t){for(;e<0;)e+=t;return e}function AG(e,t,n){if(e.shape.length>3||t.shape.length>3)throw new qe("batchDot is not implemented for tensors of 4D or higher rank yet");if(v.assert(e.shape.length>=2,()=>`batchDot requires the rank of x to be >= 2, but got ${e.shape.length}`),v.assert(e.shape.length>=2,()=>`batchDot requires the rank of y to be >= 2, but got ${t.shape.length}`),typeof n=="number"&&(n=[n,n]),e.dtype==="complex64"||t.dtype==="complex64")throw new qe("batchDot is not implemented for complex64-type Tensors yet.");let s=e.shape.length,r=t.shape.length;n==null&&(n=[s-1,r-2]);let a=n;return Y(()=>{let o;if(s>r){o=s-r;let l=[];for(let u=0;u<o;++u)l.push(1);t=U(t,t.shape.concat(l))}else if(r>s){o=r-s;let l=[];for(let u=0;u<o;++u)l.push(1);e=U(e,e.shape.concat(l))}else o=0;let i;if(e.shape.length===2&&t.shape.length===2)a[0]===a[1]?i=ke(L(e,t),a[0]):i=ke(L(et(e,[1,0]),t),a[1]);else{let l=a[0]!==e.shape.length-1,u=a[1]===t.shape.length-1;i=Qe(e,t,l,u)}if(o>0){let l;s>r?l=s+r-3:l=s-1;let u=[];for(let c=l;c<l+o;++c)u.push(c);i=st(i,u)}return i.shape.length===1&&(i=Kt(i,1)),i})}var J5=class extends nu{constructor(e){super(e),this.axes=e.axes,this.normalize=e.normalize==null?!1:e.normalize,this.supportsMasking=!0,this.reshapeRequired=!1}build(e){v.assert(Array.isArray(e)&&e.length===2&&Array.isArray(e[0])&&Array.isArray(e[1]),()=>"A `Dot` layer should be called on a list of exactly 2 inputs.");let t=e[0],n=e[1];if(t.length>3||n.length>3)throw new qe("Dot layer does not support tensors of 4D or higher rank yet.");let s=this.interpretAxes(t,n);if(t[s[0]]!==n[s[1]])throw new j(`Dimension incompatibility: ${t[s[0]]} !== ${n[s[1]]}`)}mergeFunction(e){if(e.length!==2)throw new j(`A \`Dot\` layer must be called on exactly 2 inputs, but received ${e.length} input(s).`);let t=e[0],n=e[1],s;return Array.isArray(this.axes)?s=this.axes.map((r,a)=>ep(r,e[a].shape.length)):s=[ep(this.axes,t.shape.length),ep(this.axes,n.shape.length)],this.normalize&&(t=km(t,s[0]),n=km(n,s[1])),AG(t,n,s)}interpretAxes(e,t){let n;return Array.isArray(this.axes)?n=this.axes:n=[ep(this.axes,e.length),ep(this.axes,t.length)],n}computeOutputShape(e){v.assert(Array.isArray(e)&&e.length===2&&Array.isArray(e[0])&&Array.isArray(e[1]),()=>"A `Dot` layer should be called on a list of exactly 2 inputs.");let t=e[0].slice(),n=e[1].slice();if(t.length>3||n.length>3)throw new qe("Dot layer does not support tensors of 4D or higher rank yet.");let s=this.interpretAxes(t,n);t.splice(s[0],1),n.splice(s[1],1),n.splice(0,1);let r=t.concat(n);return r.length===1&&r.push(1),r}computeMask(e,t){return null}getConfig(){let e={axes:this.axes,normalize:this.normalize},t=super.getConfig();return Object.assign(e,t),e}};J5.className="Dot";de.registerClass(J5);var Q5=class extends ut{constructor(e){super(e),this.supportsMasking=!0,this.stddev=e.stddev}computeOutputShape(e){return e}getConfig(){let e=super.getConfig(),t={stddev:this.stddev};return Object.assign(t,e),t}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e);return Sh(()=>ce(e2(n.shape,0,this.stddev),n),()=>n,t.training||!1)})}};Q5.className="GaussianNoise";de.registerClass(Q5);var ex=class extends ut{constructor(e){super(e),this.supportsMasking=!0,this.rate=e.rate}computeOutputShape(e){return e}getConfig(){let e=super.getConfig(),t={rate:this.rate};return Object.assign(t,e),t}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=Xe(e);return this.rate>0&&this.rate<1?Sh(()=>{let r=Math.sqrt(this.rate/(1-this.rate));return L(n,e2(n.shape,1,r))},()=>n,t.training||!1):n})}};ex.className="GaussianDropout";de.registerClass(ex);var tx=class extends ut{constructor(e){super(e),this.supportsMasking=!0,this.rate=e.rate,this.noiseShape=e.noiseShape}_getNoiseShape(e){return this.noiseShape||Xe(e).shape}computeOutputShape(e){return e}getConfig(){let e=super.getConfig(),t={rate:this.rate};return Object.assign(t,e),t}call(e,t){return Y(()=>{if(this.rate<1&&this.rate>0){let n=this._getNoiseShape(e);return Sh(()=>{let r=Xe(e),a=1.6732632423543772,o=1.0507009873554805,i=-a*o,l=li(Kc(n),this.rate);l=Q0(l,"float32");let u=((1-this.rate)*(1+this.rate*i**2))**-.5,c=-u*i*this.rate,p=ce(L(r,l),L(ce(l,-1),i));return ce(L(p,u),c)},()=>Xe(e),t.training||!1)}return e})}};tx.className="AlphaDropout";de.registerClass(tx);function Rp(e,t,n,s,r,a=.001){let o;if(e.rank===2)o=nA(e,t,n,s,r,a);else if(e.rank===3)o=sA(e,t,n,s,r,a);else if(e.rank===4)o=rA(e,t,n,s,r,a);else throw new qe(`batchNormalization is not implemented for array of rank ${e.rank} yet`);return o}function xG(e,t,n,s,r=.001){return Y(()=>{let a=mh(e,s),o=a.mean,i=a.variance;return[Rp(e,o,i,n,t,r),o,i]})}function bG(e,t,n,s,r=.001){return Y(()=>{let a=mh(e,s),o=a.mean,i=a.variance,l=[];for(let f of _r(0,e.rank))s.indexOf(f)!==-1?l.push(1):l.push(e.shape[f]);let u=U(o,l),c=U(i,l),p=t==null?null:U(t,l),d=n==null?null:U(n,l);return[Rp(e,u,c,d,p,r),o,i]})}function vG(e,t,n,s,r=.001){return v.arraysEqual(s.slice().sort(),_r(0,e.rank-1))?xG(e,t,n,s,r):bG(e,t,n,s,r)}var nx=class extends ut{constructor(e){e==null&&(e={}),super(e),this.supportsMasking=!0,this.axis=e.axis==null?-1:e.axis,this.momentum=e.momentum==null?.99:e.momentum,this.epsilon=e.epsilon==null?.001:e.epsilon,this.center=e.center==null?!0:e.center,this.scale=e.scale==null?!0:e.scale,this.betaInitializer=Ot(e.betaInitializer||"zeros"),this.gammaInitializer=Ot(e.gammaInitializer||"ones"),this.movingMeanInitializer=Ot(e.movingMeanInitializer||"zeros"),this.movingVarianceInitializer=Ot(e.movingVarianceInitializer||"ones"),this.betaConstraint=yn(e.betaConstraint),this.gammaConstraint=yn(e.gammaConstraint),this.betaRegularizer=Mt(e.betaRegularizer),this.gammaRegularizer=Mt(e.gammaRegularizer)}build(e){e=xt(e);let t=this.axis>=0?this.axis:this.axis+e.length,n=e[t];if(n==null)throw new j(`Axis ${t} of input tensor should have a defined dimension but the layer received an input with shape ${JSON.stringify(e)}.`);this.inputSpec=[new rn({ndim:e.length,axes:{[t]:n}})];let s=[n];this.scale&&(this.gamma=this.addWeight("gamma",s,null,this.gammaInitializer,this.gammaRegularizer,!0,this.gammaConstraint)),this.center&&(this.beta=this.addWeight("beta",s,null,this.betaInitializer,this.betaRegularizer,!0,this.betaConstraint)),this.movingMean=this.addWeight("moving_mean",s,null,this.movingMeanInitializer,null,!1),this.movingVariance=this.addWeight("moving_variance",s,null,this.movingVarianceInitializer,null,!1),this.built=!0}call(e,t){return Y(()=>{let n=t.training==null?!1:t.training,s=Xe(e),r=s.shape,a=r.length,o=_r(0,a),i=this.axis>=0?this.axis:this.axis+a;o.splice(i,1);let l=el(1,a);l[i]=r[i];let u=o.slice();u.sort();let c=!v.arraysEqual(u,_r(0,a).slice(0,a-1)),p=()=>{if(c){let y=U(this.movingMean.read(),l),x=U(this.movingVariance.read(),l),A=this.center?U(this.beta.read(),l):null,b=this.scale?U(this.gamma.read(),l):null;return Rp(s,y,x,A,b,this.epsilon)}else return Rp(s,this.movingMean.read(),this.movingVariance.read(),this.beta==null?null:this.beta.read(),this.gamma==null?null:this.gamma.read(),this.epsilon)};if(!n)return p();let[d,h,f]=vG(s,this.gamma.read(),this.beta.read(),o,this.epsilon),m=(y,x,A)=>{Y(()=>{let b=1-A,w=y.read(),k=L(fe(w,x),b);y.write(fe(w,k))})};return(()=>{m(this.movingMean,h,this.momentum),m(this.movingVariance,f,this.momentum)})(),d})}getConfig(){let e={axis:this.axis,momentum:this.momentum,epsilon:this.epsilon,center:this.center,scale:this.scale,betaInitializer:Vt(this.betaInitializer),gammaInitializer:Vt(this.gammaInitializer),movingMeanInitializer:Vt(this.movingMeanInitializer),movingVarianceInitializer:Vt(this.movingVarianceInitializer),betaRegularizer:St(this.betaRegularizer),gammaRegularizer:St(this.gammaRegularizer),betaConstraint:gn(this.betaConstraint),gammaConstraint:gn(this.gammaConstraint)},t=super.getConfig();return Object.assign(e,t),e}};nx.className="BatchNormalization";de.registerClass(nx);var sx=class extends ut{constructor(e){if(e==null&&(e={}),super(e),this.axis=e.axis==null?-1:e.axis,typeof this.axis=="number"){if(!Number.isInteger(this.axis))throw new Error(`Expected axis to be an integer, but received ${this.axis}`)}else if(Array.isArray(this.axis)){for(let t of this.axis)if(!Number.isInteger(t))throw new Error(`Expected axis to be an array of integers, but received ${JSON.stringify(this.axis)}`)}else throw new Error(`Expected axis to be an integer or an array of integers, but received ${JSON.stringify(this.axis)}`);this.epsilon=e.epsilon==null?.001:e.epsilon,this.center=e.center==null?!0:e.center,this.scale=e.scale==null?!0:e.scale,this.betaInitializer=Ot(e.betaInitializer||"zeros"),this.gammaInitializer=Ot(e.gammaInitializer||"ones"),this.betaRegularizer=Mt(e.betaRegularizer),this.gammaRegularizer=Mt(e.gammaRegularizer),this.supportsMasking=!0}build(e){e=xt(e);let t=e.length;typeof this.axis=="number"&&(this.axis=[this.axis]);for(let r=0;r<this.axis.length;++r)this.axis[r]<0&&(this.axis[r]+=t);for(let r of this.axis)if(r<0||r>=t)throw new Error(`Invalid axis: ${r}`);if(this.axis.length!==Xa(this.axis).length)throw new Error(`Found duplicate axes in: ${this.axis}`);let n=this.axis.map(r=>e[r]),s=!0;this.scale?this.gamma=this.addWeight("gamma",n,"float32",this.gammaInitializer,this.gammaRegularizer,s):this.gamma=null,this.center?this.beta=this.addWeight("beta",n,"float32",this.betaInitializer,this.betaRegularizer,s):this.beta=null,this.built=!0}call(e,t){let n=Xe(e),s=n.shape,r=s.length;return Y(()=>{let{mean:o,variance:i}=mh(n,this.axis,!0),l=el(1,r);for(let f of this.axis)l[f]=s[f];let u=f=>f!=null&&f.shape.length!==r?U(f,l):f,c=this.scale?u(this.gamma.read()):null,p=this.center?u(this.beta.read()):null,d=[],h=[];for(let f=0;f<r;++f)this.axis.indexOf(f)!==-1?(d.push(s[f]),h.push(1)):(d.push(1),h.push(s[f]));return o=qs(o,d),i=qs(i,d),c!=null&&(c=qs(c,h)),p!=null&&(p=qs(p,h)),Rp(n,o,i,p,c,this.epsilon)})}getConfig(){let e={axis:this.axis,epsilon:this.epsilon,center:this.center,scale:this.scale,betaInitializer:Vt(this.betaInitializer),gammaInitializer:Vt(this.gammaInitializer),betaRegularizer:St(this.betaRegularizer),gammaRegularizer:St(this.gammaRegularizer)},t=super.getConfig();return Object.assign(e,t),e}};sx.className="LayerNormalization";de.registerClass(sx);function wG(e,t,n){return Y(()=>{if(e.rank!==4)throw new j(`temporalPadding expects input tensor to be 4-D, but received a ${e.rank}-D tensor.`);if(t==null&&(t=[[1,1],[1,1]]),t.length!==2||t[0].length!==2||t[1].length!==2)throw new j("spatial2dPadding expects `padding` to be an Array of two Arrays, each of which is an Array of two integers.");if(n==null&&(n=Dr()),n!=="channelsLast"&&n!=="channelsFirst")throw new j(`Unknown data format: ${n}. Supported data formats are 'channelsLast' and 'channelsFirst.`);let s;return n==="channelsFirst"?s=[[0,0],[0,0],t[0],t[1]]:s=[[0,0],t[0],t[1],[0,0]],Qs(e,s)})}var rx=class extends ut{constructor(e){if(e==null&&(e={}),super(e),this.dataFormat=e.dataFormat==null?Dr():e.dataFormat,e.padding==null)this.padding=[[1,1],[1,1]];else if(typeof e.padding=="number")this.padding=[[e.padding,e.padding],[e.padding,e.padding]];else{if(e.padding=e.padding,e.padding.length!==2)throw new j(`ZeroPadding2D expects padding to be a length-2 array, but received a length-${e.padding.length} array.`);let t,n;if(typeof e.padding[0]=="number")t=[e.padding[0],e.padding[0]],n=[e.padding[1],e.padding[1]];else{if(e.padding=e.padding,e.padding[0].length!==2)throw new j(`ZeroPadding2D expects height padding to be a length-2 array, but received a length-${e.padding[0].length} array.`);if(t=e.padding[0],e.padding[1].length!==2)throw new j(`ZeroPadding2D expects width padding to be a length-2 array, but received a length-${e.padding[1].length} array.`);n=e.padding[1]}this.padding=[t,n]}this.inputSpec=[new rn({ndim:4})]}computeOutputShape(e){e=xt(e);let t,n;return this.dataFormat==="channelsFirst"?(e[2]!=null&&e[2]>=0?t=e[2]+this.padding[0][0]+this.padding[0][1]:t=null,e[3]!=null&&e[3]>=0?n=e[3]+this.padding[1][0]+this.padding[1][1]:n=null,[e[0],e[1],t,n]):(e[1]!=null&&e[1]>=0?t=e[1]+this.padding[0][0]+this.padding[0][1]:t=null,e[2]!=null&&e[2]>=0?n=e[2]+this.padding[1][0]+this.padding[1][1]:n=null,[e[0],t,n,e[3]])}call(e,t){return Y(()=>wG(Xe(e),this.padding,this.dataFormat))}getConfig(){let e={padding:this.padding,dataFormat:this.dataFormat},t=super.getConfig();return Object.assign(e,t),e}};rx.className="ZeroPadding2D";de.registerClass(rx);function y2(e,t,n,s,r,a){return Y(()=>{Yt(r),y8(a),er(s),n==null&&(n=[1,1]),s==null&&(s="valid"),r==null&&(r=Dr()),a==null&&(a="max"),e=S5(e,r);let o,i=s==="same"?"same":"valid";return a==="max"?o=fh(e,t,n,i):o=lh(e,t,n,i),r==="channelsFirst"&&(o=et(o,[0,3,1,2])),o})}function pk(e,t,n,s,r,a){return Y(()=>{Yt(r),y8(a),er(s),n==null&&(n=[1,1,1]),s==null&&(s="valid"),r==null&&(r=Dr()),a==null&&(a="max"),e=ik(e,r);let o,i=s==="same"?"same":"valid";return a==="max"?o=NA(e,t,n,i):o=tA(e,t,n,i),r==="channelsFirst"&&(o=et(o,[0,4,1,2,3])),o})}var hk=class extends ut{constructor(e){if(e.poolSize==null&&(e.poolSize=2),super(e),typeof e.poolSize=="number")this.poolSize=[e.poolSize];else if(Array.isArray(e.poolSize)&&e.poolSize.length===1&&typeof e.poolSize[0]=="number")this.poolSize=e.poolSize;else throw new j(`poolSize for 1D convolutional layer must be a number or an Array of a single number, but received ${JSON.stringify(e.poolSize)}`);if(vn(this.poolSize,"poolSize"),e.strides==null)this.strides=this.poolSize;else if(typeof e.strides=="number")this.strides=[e.strides];else if(Array.isArray(e.strides)&&e.strides.length===1&&typeof e.strides[0]=="number")this.strides=e.strides;else throw new j(`strides for 1D convolutional layer must be a number or an Array of a single number, but received ${JSON.stringify(e.strides)}`);vn(this.strides,"strides"),this.padding=e.padding==null?"valid":e.padding,er(this.padding),this.inputSpec=[new rn({ndim:3})]}computeOutputShape(e){e=xt(e);let t=Rr(e[1],this.poolSize[0],this.padding,this.strides[0]);return[e[0],t,e[2]]}call(e,t){return Y(()=>{this.invokeCallHook(e,t),e=kh(Xe(e),2);let n=this.poolingFunction(Xe(e),[this.poolSize[0],1],[this.strides[0],1],this.padding,"channelsLast");return st(n,[2])})}getConfig(){let e={poolSize:this.poolSize,padding:this.padding,strides:this.strides},t=super.getConfig();return Object.assign(e,t),e}},ax=class extends hk{constructor(e){super(e)}poolingFunction(e,t,n,s,r){return Yt(r),er(s),y2(e,t,n,s,r,"max")}};ax.className="MaxPooling1D";de.registerClass(ax);var ox=class extends hk{constructor(e){super(e)}poolingFunction(e,t,n,s,r){return Yt(r),er(s),y2(e,t,n,s,r,"avg")}};ox.className="AveragePooling1D";de.registerClass(ox);var fk=class extends ut{constructor(e){if(e.poolSize==null&&(e.poolSize=[2,2]),super(e),this.poolSize=Array.isArray(e.poolSize)?e.poolSize:[e.poolSize,e.poolSize],e.strides==null)this.strides=this.poolSize;else if(Array.isArray(e.strides)){if(e.strides.length!==2)throw new j(`If the strides property of a 2D pooling layer is an Array, it is expected to have a length of 2, but received length ${e.strides.length}.`);this.strides=e.strides}else this.strides=[e.strides,e.strides];vn(this.poolSize,"poolSize"),vn(this.strides,"strides"),this.padding=e.padding==null?"valid":e.padding,this.dataFormat=e.dataFormat==null?"channelsLast":e.dataFormat,Yt(this.dataFormat),er(this.padding),this.inputSpec=[new rn({ndim:4})]}computeOutputShape(e){e=xt(e);let t=this.dataFormat==="channelsFirst"?e[2]:e[1],n=this.dataFormat==="channelsFirst"?e[3]:e[2];return t=Rr(t,this.poolSize[0],this.padding,this.strides[0]),n=Rr(n,this.poolSize[1],this.padding,this.strides[1]),this.dataFormat==="channelsFirst"?[e[0],e[1],t,n]:[e[0],t,n,e[3]]}call(e,t){return Y(()=>(this.invokeCallHook(e,t),this.poolingFunction(Xe(e),this.poolSize,this.strides,this.padding,this.dataFormat)))}getConfig(){let e={poolSize:this.poolSize,padding:this.padding,strides:this.strides,dataFormat:this.dataFormat},t=super.getConfig();return Object.assign(e,t),e}},ix=class extends fk{constructor(e){super(e)}poolingFunction(e,t,n,s,r){return Yt(r),er(s),y2(e,t,n,s,r,"max")}};ix.className="MaxPooling2D";de.registerClass(ix);var lx=class extends fk{constructor(e){super(e)}poolingFunction(e,t,n,s,r){return Yt(r),er(s),y2(e,t,n,s,r,"avg")}};lx.className="AveragePooling2D";de.registerClass(lx);var mk=class extends ut{constructor(e){if(e.poolSize==null&&(e.poolSize=[2,2,2]),super(e),this.poolSize=Array.isArray(e.poolSize)?e.poolSize:[e.poolSize,e.poolSize,e.poolSize],e.strides==null)this.strides=this.poolSize;else if(Array.isArray(e.strides)){if(e.strides.length!==3)throw new j(`If the strides property of a 3D pooling layer is an Array, it is expected to have a length of 3, but received length ${e.strides.length}.`);this.strides=e.strides}else this.strides=[e.strides,e.strides,e.strides];vn(this.poolSize,"poolSize"),vn(this.strides,"strides"),this.padding=e.padding==null?"valid":e.padding,this.dataFormat=e.dataFormat==null?"channelsLast":e.dataFormat,Yt(this.dataFormat),er(this.padding),this.inputSpec=[new rn({ndim:5})]}computeOutputShape(e){e=xt(e);let t=this.dataFormat==="channelsFirst"?e[2]:e[1],n=this.dataFormat==="channelsFirst"?e[3]:e[2],s=this.dataFormat==="channelsFirst"?e[4]:e[3];return t=Rr(t,this.poolSize[0],this.padding,this.strides[0]),n=Rr(n,this.poolSize[1],this.padding,this.strides[1]),s=Rr(s,this.poolSize[2],this.padding,this.strides[2]),this.dataFormat==="channelsFirst"?[e[0],e[1],t,n,s]:[e[0],t,n,s,e[4]]}call(e,t){return Y(()=>(this.invokeCallHook(e,t),this.poolingFunction(Xe(e),this.poolSize,this.strides,this.padding,this.dataFormat)))}getConfig(){let e={poolSize:this.poolSize,padding:this.padding,strides:this.strides,dataFormat:this.dataFormat},t=super.getConfig();return Object.assign(e,t),e}},ux=class extends mk{constructor(e){super(e)}poolingFunction(e,t,n,s,r){return Yt(r),er(s),pk(e,t,n,s,r,"max")}};ux.className="MaxPooling3D";de.registerClass(ux);var cx=class extends mk{constructor(e){super(e)}poolingFunction(e,t,n,s,r){return Yt(r),er(s),pk(e,t,n,s,r,"avg")}};cx.className="AveragePooling3D";de.registerClass(cx);var gk=class extends ut{constructor(e){super(e),this.inputSpec=[new rn({ndim:3})]}computeOutputShape(e){return[e[0],e[2]]}call(e,t){throw new qe}},dx=class extends gk{constructor(e){super(e||{})}call(e,t){return Y(()=>{let n=Xe(e);return Bt(n,1)})}};dx.className="GlobalAveragePooling1D";de.registerClass(dx);var px=class extends gk{constructor(e){super(e||{})}call(e,t){return Y(()=>{let n=Xe(e);return mn(n,1)})}};px.className="GlobalMaxPooling1D";de.registerClass(px);var yk=class extends ut{constructor(e){super(e),this.dataFormat=e.dataFormat==null?"channelsLast":e.dataFormat,Yt(this.dataFormat),this.inputSpec=[new rn({ndim:4})]}computeOutputShape(e){return e=e,this.dataFormat==="channelsLast"?[e[0],e[3]]:[e[0],e[1]]}call(e,t){throw new qe}getConfig(){let e={dataFormat:this.dataFormat},t=super.getConfig();return Object.assign(e,t),e}},hx=class extends yk{call(e,t){return Y(()=>{let n=Xe(e);return this.dataFormat==="channelsLast"?Bt(n,[1,2]):Bt(n,[2,3])})}};hx.className="GlobalAveragePooling2D";de.registerClass(hx);var fx=class extends yk{call(e,t){return Y(()=>{let n=Xe(e);return this.dataFormat==="channelsLast"?mn(n,[1,2]):mn(n,[2,3])})}};fx.className="GlobalMaxPooling2D";de.registerClass(fx);var Ak=class extends ut{constructor(e){super(e),this.layer=e.layer}build(e){this.built=!0}get trainable(){return this.layer!=null?this.layer.trainable:!1}set trainable(e){this.layer!=null&&(this.layer.trainable=e)}get trainableWeights(){return this.layer.trainableWeights}get nonTrainableWeights(){return this.layer.nonTrainableWeights}get updates(){return this.layer._updates}get losses(){return this.layer.losses}getWeights(){return this.layer.getWeights()}setWeights(e){this.layer.setWeights(e)}getConfig(){let e={layer:{className:this.layer.getClassName(),config:this.layer.getConfig()}},t=super.getConfig();return Object.assign(e,t),e}setFastWeightInitDuringBuild(e){super.setFastWeightInitDuringBuild(e),this.layer!=null&&this.layer.setFastWeightInitDuringBuild(e)}static fromConfig(e,t,n={}){let s=t.layer,r=Er(s,n);delete t.layer;let a={layer:r};return Object.assign(a,t),new e(a)}},mx=class extends Ak{constructor(e){super(e),this.supportsMasking=!0}build(e){if(e=xt(e),e.length<3)throw new j(`TimeDistributed layer expects an input shape >= 3D, but received input shape ${JSON.stringify(e)}`);this.inputSpec=[{shape:e}];let t=[e[0]].concat(e.slice(2));this.layer.built||(this.layer.build(t),this.layer.built=!0),super.build(e)}computeOutputShape(e){e=xt(e);let t=[e[0]].concat(e.slice(2)),n=this.layer.computeOutputShape(t),s=e[1];return[n[0],s].concat(n.slice(1))}call(e,t){return Y(()=>(e=Xe(e),ck((a,o)=>[Xe(this.layer.call(a,t)),[]],e,[],!1,null,null,!1,!0)[1]))}};mx.className="TimeDistributed";de.registerClass(mx);function kG(e){eu(IV,"BidirectionalMergeMode",e)}var IG="concat",gx=class extends Ak{constructor(e){super(e);let t=e.layer.getConfig(),n={};n.className=e.layer.getClassName(),n.config=t,this.forwardLayer=Er(n),t.goBackwards=t.goBackwards!==!0;let s={};if(s.className=e.layer.getClassName(),s.config=t,this.backwardLayer=Er(s),this.forwardLayer.name="forward_"+this.forwardLayer.name,this.backwardLayer.name="backward_"+this.backwardLayer.name,this.mergeMode=e.mergeMode===void 0?IG:e.mergeMode,kG(this.mergeMode),e.weights)throw new qe("weights support is not implemented for Bidirectional layer yet.");this._stateful=e.layer.stateful,this.returnSequences=e.layer.returnSequences,this.returnState=e.layer.returnState,this.supportsMasking=!0,this._trainable=!0,this.inputSpec=e.layer.inputSpec,this.numConstants=null}get trainable(){return this._trainable}set trainable(e){this._trainable=e,this.forwardLayer!=null&&(this.forwardLayer.trainable=e),this.backwardLayer!=null&&(this.backwardLayer.trainable=e)}getWeights(){return this.forwardLayer.getWeights().concat(this.backwardLayer.getWeights())}setWeights(e){let t=e.length,n=Math.floor(t/2);this.forwardLayer.setWeights(e.slice(0,n)),this.backwardLayer.setWeights(e.slice(n))}computeOutputShape(e){let t=this.forwardLayer.computeOutputShape(e);Array.isArray(t)&&Array.isArray(t[0])||(t=[t]),t=t;let n,s,r;return this.returnState&&(r=t.slice(1)),n=t[0],n=n,this.mergeMode==="concat"?(n[n.length-1]*=2,s=[n]):this.mergeMode==null?s=[n,n.slice()]:s=[n],this.returnState?this.mergeMode==null?s.concat(r).concat(r.slice()):[n].concat(r).concat(r.slice()):ps(s)}apply(e,t){let n=t==null?null:t.initialState,s=t==null?null:t.constants;t==null&&(t={});let r=uk(e,n,s,this.numConstants);if(e=r.inputs,n=r.initialState,s=r.constants,Array.isArray(e)&&(n=e.slice(1),e=e[0]),(n==null||n.length===0)&&s==null)return super.apply(e,t);let a=[],o=[];if(n!=null){let l=n.length;if(l%2>0)throw new j("When passing `initialState` to a Bidrectional RNN, the state should be an Array containing the states of the underlying RNNs.");t.initialState=n,a.push(...n);let u=n.map(c=>new rn({shape:c.shape}));this.forwardLayer.stateSpec=u.slice(0,l/2),this.backwardLayer.stateSpec=u.slice(l/2),o.push(...u)}if(s!=null)throw new qe("Support for constants in Bidirectional layers is not implemented yet.");let i=a[0]instanceof Tr;for(let l of a)if(l instanceof Tr!==i)throw new j("The initial state of a Bidirectional layer cannot be specified as a mix of symbolic and non-symbolic tensors");if(i){let l=[e].concat(a),u=this.inputSpec.concat(o),c=this.inputSpec;this.inputSpec=u;let p=super.apply(l,t);return this.inputSpec=c,p}else return super.apply(e,t)}call(e,t){return Y(()=>{let n=t.initialState,s,r;if(n==null)s=this.forwardLayer.call(e,t),r=this.backwardLayer.call(e,t);else{let i=n.slice(0,n.length/2),l=n.slice(n.length/2);s=this.forwardLayer.call(e,Object.assign(t,{initialState:i})),r=this.backwardLayer.call(e,Object.assign(t,{initialState:l}))}let a;this.returnState&&(Array.isArray(s)&&(a=s.slice(1).concat(r.slice(1))),s=s[0],r=r[0]),this.returnSequences&&(r=Zs(r,1));let o;return this.mergeMode==="concat"?o=YA([s,r]):this.mergeMode==="sum"?o=ce(s,r):this.mergeMode==="ave"?o=L(.5,ce(s,r)):this.mergeMode==="mul"?o=L(s,r):this.mergeMode==null&&(o=[s,r]),this.returnState?this.mergeMode==null?o.concat(a):[o].concat(a):o})}resetStates(e){this.forwardLayer.resetStates(),this.backwardLayer.resetStates()}build(e){Hi(this.forwardLayer.name,()=>{this.forwardLayer.build(e)}),Hi(this.backwardLayer.name,()=>{this.backwardLayer.build(e)}),this.built=!0}computeMask(e,t){Array.isArray(t)&&(t=t[0]);let n;if(this.returnSequences?this.mergeMode==null?n=[t,t]:n=t:this.mergeMode==null?n=[null,null]:n=null,this.returnState){let r=this.forwardLayer.states.map(a=>null);return Array.isArray(n)?n.concat(r).concat(r):[n].concat(r).concat(r)}else return n}get trainableWeights(){return this.forwardLayer.trainableWeights.concat(this.backwardLayer.trainableWeights)}get nonTrainableWeights(){return this.forwardLayer.nonTrainableWeights.concat(this.backwardLayer.nonTrainableWeights)}setFastWeightInitDuringBuild(e){super.setFastWeightInitDuringBuild(e),this.forwardLayer!=null&&this.forwardLayer.setFastWeightInitDuringBuild(e),this.backwardLayer!=null&&this.backwardLayer.setFastWeightInitDuringBuild(e)}getConfig(){let e={mergeMode:this.mergeMode},t=super.getConfig();return Object.assign(e,t),e}static fromConfig(e,t){let n=Er(t.layer);if(delete t.layer,t.numConstants!=null)throw new qe("Deserialization of a Bidirectional layer with numConstants present is not supported yet.");let s=t;return s.layer=n,new e(s)}};gx.className="Bidirectional";de.registerClass(gx);function SG(e){return new Yc(e)}function CG(e){return new w5(e)}function TG(e){return new x5(e)}function NG(e){return new b5(e)}function EG(e){return new v5(e)}function RG(e){return new I5(e)}function _G(e){return new k5(e)}function DG(e){return new d2(e)}function $G(e){return new Eh(e)}function PG(e){return new T5(e)}function FG(e){return new Rh(e)}function OG(e){return new N5(e)}function MG(e){return new E5(e)}function zG(e){return new R5(e)}function LG(e){return new _5(e)}function BG(e){return new D5(e)}function WG(e){return new B5(e)}function VG(e){return new z5(e)}function UG(e){return new g2(e)}function GG(e){return new M5(e)}function HG(e){return new L5(e)}function jG(e){return new W5(e)}function qG(e){return new V5(e)}function XG(e){return new U5(e)}function KG(e){return new H5(e)}function ZG(e){return new j5(e)}function YG(e){return new X5(e)}function JG(e){return new Y5(e)}function QG(e){return new K5(e)}function eH(e){return new Z5(e)}function tH(e){return new q5(e)}function nH(e){return new J5(e)}function sH(e){return new nx(e)}function rH(e){return new sx(e)}function aH(e){return new rx(e)}function yx(e){return new ox(e)}function oH(e){return yx(e)}function iH(e){return yx(e)}function Ax(e){return new lx(e)}function lH(e){return Ax(e)}function uH(e){return Ax(e)}function xx(e){return new cx(e)}function cH(e){return xx(e)}function dH(e){return xx(e)}function pH(e){return new dx(e)}function hH(e){return new hx(e)}function xk(e){return new px(e)}function bk(e){return new fx(e)}function vk(e){return new ax(e)}function wk(e){return new ix(e)}function fH(e){return new ux(e)}function mH(e){return new P5(e)}function gH(e){return new h2(e)}function yH(e){return new F5(e)}function AH(e){return new Dh(e)}function xH(e){return new $5(e)}function bH(e){return new p2(e)}function vH(e){return new O5(e)}function wH(e){return new m2(e)}function kH(e){return new Qr(e)}function IH(e){return new f2(e)}function SH(e){return new gx(e)}function CH(e){return new mx(e)}var TH=xk,NH=bk,EH=vk,RH=wk;function _H(e){return new Q5(e)}function DH(e){return new ex(e)}function $H(e){return new tx(e)}function PH(e){return new G5(e)}var kk={};Ve(kk,{MAPE:()=>HH,MSE:()=>XH,binaryAccuracy:()=>FH,binaryCrossentropy:()=>OH,categoricalAccuracy:()=>zH,categoricalCrossentropy:()=>LH,cosineProximity:()=>VH,mape:()=>jH,meanAbsoluteError:()=>UH,meanAbsolutePercentageError:()=>GH,meanSquaredError:()=>qH,mse:()=>KH,precision:()=>BH,recall:()=>WH,sparseCategoricalAccuracy:()=>MH});function FH(e,t){return p5(e,t)}function OH(e,t){return M8(e,t)}function MH(e,t){return z8(e,t)}function zH(e,t){return h5(e,t)}function LH(e,t){return f5(e,t)}function BH(e,t){return O8(e,t)}function WH(e,t){return NU(e,t)}function VH(e,t){return d5(e,t)}function UH(e,t){return u2(e,t)}function GH(e,t){return Jc(e,t)}function HH(e,t){return Jc(e,t)}function jH(e,t){return Jc(e,t)}function qH(e,t){return tu(e,t)}function XH(e,t){return tu(e,t)}function KH(e,t){return tu(e,t)}var Ik={};Ve(Ik,{modelFromJSON:()=>rG});var Sk={};Ve(Sk,{l1:()=>YH,l1l2:()=>ZH,l2:()=>JH});function ZH(e){return new Th(e)}function YH(e){return pG(e)}function JH(e){return hG(e)}var Ck=class extends nc{constructor(){super(...arguments),this.model=null}setModel(e){if(!(e instanceof ha))throw new Error("model must be a LayersModel, not some other Container");this.model=e}};function Hf(e,t){return e<t}function e7(e,t){return e>t}var Tk=class extends Ck{constructor(e){if(super(),e==null&&(e={}),e.restoreBestWeights)throw new qe("restoreBestWeights = True is not implemented in EarlyStopping yet.");this.monitor=e.monitor||"val_loss",this.minDelta=Math.abs(e.minDelta||0),this.patience=e.patience||0,this.verbose=e.verbose||0,this.mode=e.mode||"auto",this.baseline=e.baseline,["auto","min","max"].indexOf(this.mode)===-1&&(console.warn(`EarlyStopping mode '${this.mode}' is invalid. Falling back to mode 'auto'.`),this.mode="auto"),this.mode==="min"?this.monitorFunc=Hf:this.mode==="max"?this.monitorFunc=e7:this.monitor.indexOf("acc")!==-1?this.monitorFunc=e7:this.monitorFunc=Hf,this.monitorFunc===Hf&&(this.minDelta*=-1)}async onTrainBegin(e){this.wait=0,this.stoppedEpoch=0,this.baseline!=null?this.best=this.baseline:this.best=this.monitorFunc===Hf?1/0:-1/0}async onEpochEnd(e,t){await Va(t);let n=this.getMonitorValue(t);n!=null&&(this.monitorFunc(n-this.minDelta,this.best)?(this.best=n,this.wait=0):(this.wait++,this.wait>=this.patience&&(this.stoppedEpoch=e,this.model.stopTraining=!0)))}async onTrainEnd(e){this.stoppedEpoch>0&&this.verbose&&console.log(`Epoch ${this.stoppedEpoch}: early stopping.`)}getMonitorValue(e){e==null&&(e={});let t=e[this.monitor];return t==null&&console.warn(`Metric for EarlyStopping ${this.monitor} is not available. Available metrics are: ${Object.keys(e)}`),t}};function QH(e){return new Tk(e)}var ej={earlyStopping:QH},tj=q();tj.registerFlag("KEEP_INTERMEDIATE_TENSORS",()=>!1,e=>{e&&console.warn("Keep intermediate tensors is ON. This will print the values of all intermediate tensors during model inference. Not all models support this mode. For details, check e2e/benchmarks/ model_config.js. This significantly impacts performance.")});var rr;(function(e){e[e.DT_INVALID=0]="DT_INVALID",e[e.DT_FLOAT=1]="DT_FLOAT",e[e.DT_DOUBLE=2]="DT_DOUBLE",e[e.DT_INT32=3]="DT_INT32",e[e.DT_UINT8=4]="DT_UINT8",e[e.DT_INT16=5]="DT_INT16",e[e.DT_INT8=6]="DT_INT8",e[e.DT_STRING=7]="DT_STRING",e[e.DT_COMPLEX64=8]="DT_COMPLEX64",e[e.DT_INT64=9]="DT_INT64",e[e.DT_BOOL=10]="DT_BOOL",e[e.DT_QINT8=11]="DT_QINT8",e[e.DT_QUINT8=12]="DT_QUINT8",e[e.DT_QINT32=13]="DT_QINT32",e[e.DT_BFLOAT16=14]="DT_BFLOAT16",e[e.DT_QINT16=15]="DT_QINT16",e[e.DT_QUINT16=16]="DT_QUINT16",e[e.DT_UINT16=17]="DT_UINT16",e[e.DT_COMPLEX128=18]="DT_COMPLEX128",e[e.DT_HALF=19]="DT_HALF",e[e.DT_RESOURCE=20]="DT_RESOURCE",e[e.DT_VARIANT=21]="DT_VARIANT",e[e.DT_UINT32=22]="DT_UINT32",e[e.DT_UINT64=23]="DT_UINT64",e[e.DT_FLOAT_REF=101]="DT_FLOAT_REF",e[e.DT_DOUBLE_REF=102]="DT_DOUBLE_REF",e[e.DT_INT32_REF=103]="DT_INT32_REF",e[e.DT_UINT8_REF=104]="DT_UINT8_REF",e[e.DT_INT16_REF=105]="DT_INT16_REF",e[e.DT_INT8_REF=106]="DT_INT8_REF",e[e.DT_STRING_REF=107]="DT_STRING_REF",e[e.DT_COMPLEX64_REF=108]="DT_COMPLEX64_REF",e[e.DT_INT64_REF=109]="DT_INT64_REF",e[e.DT_BOOL_REF=110]="DT_BOOL_REF",e[e.DT_QINT8_REF=111]="DT_QINT8_REF",e[e.DT_QUINT8_REF=112]="DT_QUINT8_REF",e[e.DT_QINT32_REF=113]="DT_QINT32_REF",e[e.DT_BFLOAT16_REF=114]="DT_BFLOAT16_REF",e[e.DT_QINT16_REF=115]="DT_QINT16_REF",e[e.DT_QUINT16_REF=116]="DT_QUINT16_REF",e[e.DT_UINT16_REF=117]="DT_UINT16_REF",e[e.DT_COMPLEX128_REF=118]="DT_COMPLEX128_REF",e[e.DT_HALF_REF=119]="DT_HALF_REF",e[e.DT_RESOURCE_REF=120]="DT_RESOURCE_REF",e[e.DT_VARIANT_REF=121]="DT_VARIANT_REF",e[e.DT_UINT32_REF=122]="DT_UINT32_REF",e[e.DT_UINT64_REF=123]="DT_UINT64_REF"})(rr||(rr={}));var t7;(function(e){let t;(function(n){n[n.LEGACY=0]="LEGACY",n[n.V1=1]="V1",n[n.V2=2]="V2"})(t=e.CheckpointFormatVersion||(e.CheckpointFormatVersion={}))})(t7||(t7={}));var bx={};function nj(e,t){let n={tfOpName:e,category:"custom",inputs:[],attrs:[],customExecutor:t};bx[e]=n}function Nk(e){return bx[e]}function sj(e){delete bx[e]}function I(e,t,n,s,r){let a=t.inputParams[e];if(a&&a.inputIndexStart!==void 0){let i=a.inputIndexStart,l=a.inputIndexEnd===0?void 0:a.inputIndexEnd===void 0?i+1:a.inputIndexEnd;if(a.type==="tensor")return Qn(t.inputNames[a.inputIndexStart],n,s,r);if(a.type==="tensors")return t.inputNames.slice(i,l).map(d=>Qn(d,n,s,r));let u=Qn(t.inputNames.slice(i)[0],n,s,r),c=u.dataSync();return a.type==="number"?c[0]:v.toNestedArray(u.shape,c)}let o=t.attrParams[e];return o&&o.value}function Qn(e,t,n,s){let[r,a]=Ss(e);if(s!=null){let i=s.getHashTableHandleByName(r);if(i!=null)return i}let o=n.currentContextIds.find(i=>!!t[Nm(r,i)]);return o!==void 0?t[Nm(r,o)][a]:void 0}function rj(e,t,n){return t[Nm(e,n.currentContextId)]}function Gr(e,t){let[n,s,r]=Ss(e);return[Nm(n,t&&t.currentContextId),s,r]}function Nm(e,t){return t?`${e}-${t}`:e}function Ss(e){let t=e.split(":");if(t.length===1)return[e,0,void 0];let n=t[0],s=t.length===3?t[1]:void 0,r=Number(t[t.length-1]);return[n,r,s]}function tm(e,t,n){let s=I("pad",e,t,n);if(s==="explicit"){s=I("explicitPaddings",e,t,n);let r=[[0,0],[0,0],[0,0],[0,0]];for(let a=0;a<4;a++)r[a][0]=s[a*2],r[a][1]=s[a*2+1];return r}return s}function ca(e){return e.kept?e:On(e)}var Ek={};Ve(Ek,{json:()=>aj});var aj=[{tfOpName:"Add",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"AddV2",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"AddN",category:"arithmetic",inputs:[{start:0,end:0,name:"tensors",type:"tensors"}]},{tfOpName:"BiasAdd",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0}]},{tfOpName:"Sub",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"RealDiv",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Div",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"DivNoNan",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"FloorDiv",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Mul",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Maximum",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Minimum",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Pow",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"SquaredDifference",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Mod",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"FloorMod",category:"arithmetic",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]}],Rk={};Ve(Rk,{json:()=>oj});var oj=[{tfOpName:"Abs",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Acos",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Asin",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Atan",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Atan2",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"y",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Ceil",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"ClipByValue",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"clipValueMin",type:"number"},{start:2,name:"clipValueMax",type:"number"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Complex",category:"basic_math",inputs:[{start:0,name:"real",type:"tensor"},{start:1,name:"imag",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"ComplexAbs",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Cos",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Cosh",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Elu",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Exp",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Floor",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Log",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Imag",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"Tout",name:"outputType",type:"dtype",notSupported:!0}]},{tfOpName:"Neg",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Real",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"Tout",name:"outputType",type:"dtype",notSupported:!0}]},{tfOpName:"Prelu",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"alpha",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Relu",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Relu6",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Selu",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Sigmoid",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Sin",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Sinh",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Sqrt",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Rsqrt",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Square",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Tan",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Tanh",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Sign",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Round",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Expm1",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Log1p",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Reciprocal",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Softplus",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Asinh",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Acosh",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Atanh",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Erf",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Prod",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axes",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool",notSupported:!0},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"LeakyRelu",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"alpha",name:"alpha",type:"number",defaultValue:.2},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"IsNan",category:"basic_math",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]}],_k={};Ve(_k,{json:()=>ij});var ij=[{tfOpName:"EmptyTensorList",category:"control",inputs:[{start:0,name:"elementShape",type:"shape"},{start:1,name:"maxNumElements",type:"number"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"LoopCond",category:"control",inputs:[{start:0,name:"pred",type:"tensor"}]},{tfOpName:"Switch",category:"control",inputs:[{start:0,name:"data",type:"tensor"},{start:1,name:"pred",type:"tensor"}]},{tfOpName:"Merge",category:"control",inputs:[{start:0,end:0,name:"tensors",type:"tensors"}]},{tfOpName:"Enter",category:"control",inputs:[{start:0,name:"tensor",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"frame_name",name:"frameName",type:"string"},{tfName:"is_constant",name:"isConstant",type:"bool"}]},{tfOpName:"Exit",category:"control",inputs:[{start:0,name:"tensor",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"NextIteration",category:"control",inputs:[{start:0,name:"tensor",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"TensorArrayV3",category:"control",inputs:[{start:0,name:"size",type:"number"}],attrs:[{tfName:"dtype",name:"dtype",type:"dtype"},{tfName:"element_shape",name:"elementShape",type:"shape"},{tfName:"dynamic_size",name:"dynamicSize",type:"bool"},{tfName:"clear_after_read",name:"clearAfterRead",type:"bool"},{tfName:"identical_element_shapes",name:"identicalElementShapes",type:"bool"},{tfName:"tensor_array_name",name:"name",type:"string"}]},{tfOpName:"TensorArrayWriteV3",category:"control",inputs:[{start:0,name:"tensorArrayId",type:"tensor"},{start:1,name:"index",type:"number"},{start:2,name:"tensor",type:"tensor"},{start:3,name:"flowIn",type:"number"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"TensorArrayReadV3",category:"control",inputs:[{start:0,name:"tensorArrayId",type:"tensor"},{start:1,name:"index",type:"number"},{start:2,name:"flowIn",type:"number"}],attrs:[{tfName:"dtype",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"TensorArrayGatherV3",category:"control",inputs:[{start:0,name:"tensorArrayId",type:"tensor"},{start:1,name:"indices",type:"number[]"},{start:2,name:"flowIn",type:"number"}],attrs:[{tfName:"dtype",name:"dtype",type:"dtype"},{tfName:"element_shape",name:"elementShape",type:"shape"}]},{tfOpName:"TensorArrayScatterV3",category:"control",inputs:[{start:0,name:"tensorArrayId",type:"tensor"},{start:1,name:"indices",type:"number[]"},{start:2,name:"tensor",type:"tensor"},{start:3,name:"flowIn",type:"number"}],attrs:[{tfName:"T",name:"dtype",type:"dtype"}]},{tfOpName:"TensorArrayConcatV3",category:"control",inputs:[{start:0,name:"tensorArrayId",type:"tensor"},{start:1,name:"flowIn",type:"number"}],attrs:[{tfName:"dtype",name:"dtype",type:"dtype"},{tfName:"element_shape_except0",name:"elementShapeExcept0",type:"shape",notSupported:!0}]},{tfOpName:"TensorArraySplitV3",category:"control",inputs:[{start:0,name:"tensorArrayId",type:"tensor"},{start:1,name:"tensor",type:"tensor"},{start:2,name:"lengths",type:"number[]"},{start:3,name:"flowIn",type:"number"}],attrs:[{tfName:"T",name:"dtype",type:"dtype"}]},{tfOpName:"TensorArraySizeV3",category:"control",inputs:[{start:0,name:"tensorArrayId",type:"tensor"},{start:1,name:"flowIn",type:"number"}]},{tfOpName:"TensorArrayCloseV3",category:"control",inputs:[{start:0,name:"tensorArrayId",type:"tensor"}]},{tfOpName:"StatelessIf",category:"control",inputs:[{start:0,name:"cond",type:"tensor"},{start:1,end:0,name:"args",type:"tensors"}],attrs:[{tfName:"then_branch",name:"thenBranch",type:"func"},{tfName:"else_branch",name:"elseBranch",type:"func"}]},{tfOpName:"If",category:"control",inputs:[{start:0,name:"cond",type:"tensor"},{start:1,end:0,name:"args",type:"tensors"}],attrs:[{tfName:"then_branch",name:"thenBranch",type:"func"},{tfName:"else_branch",name:"elseBranch",type:"func"}]},{tfOpName:"StatelessWhile",category:"control",inputs:[{start:0,end:0,name:"args",type:"tensors"}],attrs:[{tfName:"cond",name:"cond",type:"func"},{tfName:"body",name:"body",type:"func"}]},{tfOpName:"While",category:"control",inputs:[{start:0,end:0,name:"args",type:"tensors"}],attrs:[{tfName:"cond",name:"cond",type:"func"},{tfName:"body",name:"body",type:"func"}]},{tfOpName:"TensorListScatter",category:"control",inputs:[{start:0,name:"tensor",type:"tensor"},{start:1,name:"indices",type:"number[]"},{start:2,name:"elementShape",type:"shape"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListScatterV2",category:"control",inputs:[{start:0,name:"tensor",type:"tensor"},{start:1,name:"indices",type:"number[]"},{start:2,name:"elementShape",type:"shape"},{start:3,name:"numElements",type:"number"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListGather",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"},{start:1,name:"indices",type:"number[]"},{start:2,name:"elementShape",type:"shape"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListGetItem",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"},{start:1,name:"index",type:"number"},{start:2,name:"elementShape",type:"shape"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListSetItem",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"},{start:1,name:"index",type:"number"},{start:2,name:"tensor",type:"tensor"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListReserve",category:"control",inputs:[{start:0,name:"elementShape",type:"shape"},{start:1,name:"numElements",type:"number"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListFromTensor",category:"control",inputs:[{start:0,name:"tensor",type:"tensor"},{start:1,name:"elementShape",type:"shape"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListStack",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"},{start:1,name:"elementShape",type:"shape"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"},{tfName:"num_elements",name:"numElements",type:"dtype"}]},{tfOpName:"TensorListSplit",category:"control",inputs:[{start:0,name:"tensor",type:"tensor"},{start:1,name:"elementShape",type:"shape"},{start:2,name:"lengths",type:"number[]"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListConcat",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"}],attrs:[{tfName:"element_shape",name:"elementShape",type:"shape"},{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListConcatV2",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"}],attrs:[{tfName:"element_shape",name:"elementShape",type:"shape"},{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListPopBack",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"},{start:1,name:"elementShape",type:"shape"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListPushBack",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"},{start:1,name:"tensor",type:"tensor"}],attrs:[{tfName:"element_dtype",name:"elementDType",type:"dtype"}]},{tfOpName:"TensorListLength",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"}]},{tfOpName:"TensorListResize",category:"control",inputs:[{start:0,name:"tensorListId",type:"tensor"},{start:1,name:"size",type:"number"}]}],Dk={};Ve(Dk,{json:()=>lj});var lj=[{tfOpName:"AvgPool",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0},{tfName:"ksize",name:"kernelSize",type:"number[]"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"MaxPool",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0},{tfName:"ksize",name:"kernelSize",type:"number[]"},{tfName:"explicit_paddings",name:"explicitPaddings",type:"number[]",defaultValue:[],notSupported:!0},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"MaxPoolWithArgmax",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"ksize",name:"kernelSize",type:"number[]"},{tfName:"include_batch_in_index",name:"includeBatchInIndex",type:"bool"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"AvgPool3D",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0},{tfName:"ksize",name:"kernelSize",type:"number[]"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"MaxPool3D",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0},{tfName:"ksize",name:"kernelSize",type:"number[]"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Conv1D",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"filter",type:"tensor"}],attrs:[{tfName:"stride",name:"stride",type:"number"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",defaultValue:"NWC"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"dilation",name:"dilation",type:"number",defaultValue:1}]},{tfOpName:"Conv2D",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"filter",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"useCudnnOnGpu",name:"useCudnnOnGpu",type:"bool"},{tfName:"data_format",name:"dataFormat",type:"string",defaultValue:"NHWC"},{tfName:"explicit_paddings",name:"explicitPaddings",type:"number[]",defaultValue:[]},{tfName:"dilations",name:"dilations",type:"number[]"}]},{tfOpName:"_FusedConv2D",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"filter",type:"tensor"},{start:2,end:0,name:"args",type:"tensors"}],attrs:[{tfName:"num_args",name:"numArgs",type:"number"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"explicit_paddings",name:"explicitPaddings",type:"number[]",defaultValue:[]},{tfName:"use_cudnn_on_gpu",name:"useCudnnOnGpu",type:"bool",defaultValue:!0},{tfName:"data_format",name:"dataFormat",type:"string",defaultValue:"NHWC"},{tfName:"dilations",name:"dilations",type:"number[]",defaultValue:[1,1,1,1]},{tfName:"fused_ops",name:"fusedOps",type:"string[]",defaultValue:[]},{tfName:"epsilon",name:"epsilon",type:"number",defaultValue:1e-4},{tfName:"leakyrelu_alpha",name:"leakyreluAlpha",type:"number",defaultValue:.2}]},{tfOpName:"Conv2DBackpropInput",category:"convolution",inputs:[{start:2,name:"x",type:"tensor"},{start:1,name:"filter",type:"tensor"},{start:0,name:"outputShape",type:"number[]"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0},{tfName:"explicit_paddings",name:"explicitPaddings",type:"number[]",defaultValue:[]},{tfName:"dilations",name:"dilations",type:"number[]",notSupported:!0}]},{tfOpName:"DepthwiseConv2d",category:"convolution",inputs:[{start:0,name:"input",type:"tensor"},{start:1,name:"filter",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",defaultValue:"NHWC"},{tfName:"explicit_paddings",name:"explicitPaddings",type:"number[]",defaultValue:[]},{tfName:"dilations",name:"dilations",type:"number[]"}]},{tfOpName:"DepthwiseConv2dNative",category:"convolution",inputs:[{start:0,name:"input",type:"tensor"},{start:1,name:"filter",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",defaultValue:"NHWC"},{tfName:"explicit_paddings",name:"explicitPaddings",type:"number[]",defaultValue:[]},{tfName:"dilations",name:"dilations",type:"number[]"}]},{tfOpName:"FusedDepthwiseConv2dNative",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"filter",type:"tensor"},{start:2,end:0,name:"args",type:"tensors"}],attrs:[{tfName:"num_args",name:"numArgs",type:"number"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",defaultValue:"NHWC"},{tfName:"dilations",name:"dilations",type:"number[]",defaultValue:[1,1,1,1]},{tfName:"fused_ops",name:"fusedOps",type:"string[]",defaultValue:[]},{tfName:"explicit_paddings",name:"explicitPaddings",type:"number[]",defaultValue:[]}]},{tfOpName:"Conv3D",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"filter",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"padding",name:"pad",type:"string"},{tfName:"data_format",name:"dataFormat",type:"string",defaultValue:"NHWC"},{tfName:"dilations",name:"dilations",type:"number[]"}]},{tfOpName:"Dilation2D",category:"convolution",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"filter",type:"tensor"}],attrs:[{tfName:"strides",name:"strides",type:"number[]"},{tfName:"rates",name:"dilations",type:"number[]"},{tfName:"padding",name:"pad",type:"string"}]}],$k={};Ve($k,{json:()=>uj});var uj=[{tfOpName:"Fill",category:"creation",inputs:[{start:0,name:"shape",type:"number[]"},{start:1,name:"value",type:"number"}],attrs:[{tfName:"T",name:"dtype",type:"dtype"}]},{tfOpName:"LinSpace",category:"creation",inputs:[{start:0,name:"start",type:"number"},{start:1,name:"stop",type:"number"},{start:2,name:"num",type:"number"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"OneHot",category:"creation",inputs:[{start:0,name:"indices",type:"tensor"},{start:1,name:"depth",type:"number"},{start:2,name:"onValue",type:"number",defaultValue:1},{start:3,name:"offValue",type:"number",defaultValue:0}],attrs:[{tfName:"axis",name:"axis",type:"number",notSupported:!0},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Ones",category:"creation",inputs:[{start:0,name:"shape",type:"number[]"}],attrs:[{tfName:"T",name:"dtype",type:"dtype"}]},{tfOpName:"OnesLike",category:"creation",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"dtype",name:"dtype",type:"dtype"}]},{tfOpName:"RandomStandardNormal",category:"creation",inputs:[{start:0,name:"shape",type:"number[]"}],attrs:[{tfName:"seed",name:"seed",type:"number",defaultValue:0},{tfName:"seed2",name:"seed2",type:"number",defaultValue:0,notSupported:!0},{tfName:"dtype",name:"dtype",type:"dtype"},{tfName:"T",name:"T",type:"number",notSupported:!0}]},{tfOpName:"RandomUniform",category:"creation",inputs:[{start:0,name:"shape",type:"number[]"}],attrs:[{tfName:"minval",name:"minval",type:"number",defaultValue:0},{tfName:"maxval",name:"maxval",type:"number",defaultValue:1},{tfName:"dtype",name:"dtype",type:"dtype"},{tfName:"seed",name:"seed",type:"number",defaultValue:0},{tfName:"seed2",name:"seed2",type:"number",defaultValue:0,notSupported:!0},{tfName:"T",name:"T",type:"number",notSupported:!0}]},{tfOpName:"Range",category:"creation",inputs:[{start:0,name:"start",type:"number"},{start:1,name:"stop",type:"number"},{start:2,name:"step",type:"number",defaultValue:0}],attrs:[{tfName:"Tidx",name:"dtype",type:"dtype"}]},{tfOpName:"TruncatedNormal",category:"creation",inputs:[{start:0,name:"shape",type:"number[]"}],attrs:[{tfName:"means",name:"mean",type:"number",defaultValue:0},{tfName:"stddev",name:"stdDev",type:"number",defaultValue:1},{tfName:"seed",name:"seed",type:"number"},{tfName:"seed2",name:"seed2",type:"number",defaultValue:0,notSupported:!0},{tfName:"dtype",name:"dtype",type:"dtype"},{tfName:"T",name:"T",type:"number",notSupported:!0}]},{tfOpName:"Zeros",category:"creation",inputs:[{start:0,name:"shape",type:"number[]"}],attrs:[{tfName:"T",name:"dtype",type:"dtype"}]},{tfOpName:"ZerosLike",category:"creation",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype"}]},{tfOpName:"Multinomial",category:"creation",inputs:[{start:0,name:"logits",type:"tensor"},{start:1,name:"numSamples",type:"number"}],attrs:[{tfName:"seed",name:"seed",type:"number"},{tfName:"seed2",name:"seed2",type:"number"},{tfName:"T",name:"dtype",type:"dtype"},{tfName:"output_dtype",name:"output_dtype",type:"dtype"}]}],Pk={};Ve(Pk,{json:()=>cj});var cj=[{tfOpName:"NonMaxSuppressionV2",category:"dynamic",inputs:[{start:0,name:"boxes",type:"tensor"},{start:1,name:"scores",type:"tensor"},{start:2,name:"maxOutputSize",type:"number"},{start:3,name:"iouThreshold",type:"number"}]},{tfOpName:"NonMaxSuppressionV3",category:"dynamic",inputs:[{start:0,name:"boxes",type:"tensor"},{start:1,name:"scores",type:"tensor"},{start:2,name:"maxOutputSize",type:"number"},{start:3,name:"iouThreshold",type:"number"},{start:4,name:"scoreThreshold",type:"number"}]},{tfOpName:"NonMaxSuppressionV4",category:"dynamic",inputs:[{start:0,name:"boxes",type:"tensor"},{start:1,name:"scores",type:"tensor"},{start:2,name:"maxOutputSize",type:"number"},{start:3,name:"iouThreshold",type:"number"},{start:4,name:"scoreThreshold",type:"number"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0},{tfName:"T_threshold",name:"threshold",type:"dtype",notSupported:!0},{tfName:"pad_to_max_output_size",name:"padToMaxOutputSize",type:"bool"}]},{tfOpName:"NonMaxSuppressionV5",category:"dynamic",inputs:[{start:0,name:"boxes",type:"tensor"},{start:1,name:"scores",type:"tensor"},{start:2,name:"maxOutputSize",type:"number"},{start:3,name:"iouThreshold",type:"number"},{start:4,name:"scoreThreshold",type:"number"},{start:5,name:"softNmsSigma",type:"number"}]},{tfOpName:"Where",category:"dynamic",inputs:[{start:0,name:"condition",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"ListDiff",category:"dynamic",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"y",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]}],Fk={};Ve(Fk,{json:()=>dj});var dj=[{tfOpName:"LowerBound",category:"evaluation",inputs:[{start:0,name:"sortedSequence",type:"tensor"},{start:1,name:"values",type:"tensor"}]},{tfOpName:"TopKV2",category:"evaluation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"k",type:"number"}],attrs:[{tfName:"sorted",name:"sorted",type:"bool"}]},{tfOpName:"UpperBound",category:"evaluation",inputs:[{start:0,name:"sortedSequence",type:"tensor"},{start:1,name:"values",type:"tensor"}]},{tfOpName:"Unique",category:"evaluation",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"UniqueV2",category:"evaluation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number"}]}],Ok={};Ve(Ok,{json:()=>pj});var pj=[{tfOpName:"PlaceholderWithDefault",category:"graph",inputs:[{start:0,name:"default",type:"tensor"}],attrs:[{tfName:"shape",name:"shape",type:"shape"},{tfName:"dtype",name:"dtype",type:"dtype"}]},{tfOpName:"Placeholder",category:"graph",attrs:[{tfName:"shape",name:"shape",type:"shape"},{tfName:"dtype",name:"dtype",type:"dtype"}]},{tfOpName:"Const",category:"graph"},{tfOpName:"Identity",category:"graph",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"IdentityN",category:"graph",inputs:[{start:0,end:0,name:"x",type:"tensors"}]},{tfOpName:"Snapshot",category:"graph",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"Rank",category:"graph",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"Size",category:"graph",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"Shape",category:"graph",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"ShapeN",category:"graph",inputs:[{start:0,end:0,name:"x",type:"tensors"}]},{tfOpName:"Print",category:"graph",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"data",type:"tensors"}],attrs:[{tfName:"message",name:"message",type:"string"},{tfName:"first_n",name:"firstN",type:"number",notSupported:!0},{tfName:"summarize",name:"summarize",type:"number",defaultValue:3}]},{tfOpName:"NoOp",category:"graph",inputs:[]},{tfOpName:"StopGradient",category:"graph",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"FakeQuantWithMinMaxVars",category:"graph",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"min",name:"min",type:"number"},{tfName:"max",name:"max",type:"number"}]}],Mk={};Ve(Mk,{json:()=>hj});var hj=[{tfOpName:"HashTable",category:"hash_table",inputs:[],attrs:[{tfName:"shared_name",name:"sharedName",type:"string"},{tfName:"use_node_name_sharing",name:"useNodeNameSharing",type:"bool"},{tfName:"key_dtype",name:"keyDType",type:"dtype"},{tfName:"value_dtype",name:"valueDType",type:"dtype"}]},{tfOpName:"HashTableV2",category:"hash_table",inputs:[],attrs:[{tfName:"shared_name",name:"sharedName",type:"string"},{tfName:"use_node_name_sharing",name:"useNodeNameSharing",type:"bool"},{tfName:"key_dtype",name:"keyDType",type:"dtype"},{tfName:"value_dtype",name:"valueDType",type:"dtype"}]},{tfOpName:"LookupTableImport",category:"hash_table",inputs:[{start:0,name:"tableHandle",type:"tensor"},{start:1,name:"keys",type:"tensor"},{start:2,name:"values",type:"tensor"}],attrs:[{tfName:"Tin",name:"tIn",type:"dtype",notSupported:!0},{tfName:"Tout",name:"tOut",type:"dtype",notSupported:!0}]},{tfOpName:"LookupTableImportV2",category:"hash_table",inputs:[{start:0,name:"tableHandle",type:"tensor"},{start:1,name:"keys",type:"tensor"},{start:2,name:"values",type:"tensor"}],attrs:[{tfName:"Tin",name:"tIn",type:"dtype",notSupported:!0},{tfName:"Tout",name:"tOut",type:"dtype",notSupported:!0}]},{tfOpName:"LookupTableFind",category:"hash_table",inputs:[{start:0,name:"tableHandle",type:"tensor"},{start:1,name:"keys",type:"tensor"},{start:2,name:"defaultValue",type:"tensor"}],attrs:[{tfName:"Tin",name:"tIn",type:"dtype",notSupported:!0},{tfName:"Tout",name:"tOut",type:"dtype",notSupported:!0}]},{tfOpName:"LookupTableFindV2",category:"hash_table",inputs:[{start:0,name:"tableHandle",type:"tensor"},{start:1,name:"keys",type:"tensor"},{start:2,name:"defaultValue",type:"tensor"}],attrs:[{tfName:"Tin",name:"tIn",type:"dtype",notSupported:!0},{tfName:"Tout",name:"tOut",type:"dtype",notSupported:!0}]},{tfOpName:"LookupTableSize",category:"hash_table",inputs:[{start:0,name:"tableHandle",type:"tensor"}]},{tfOpName:"LookupTableSizeV2",category:"hash_table",inputs:[{start:0,name:"tableHandle",type:"tensor"}]}],zk={};Ve(zk,{json:()=>fj});var fj=[{tfOpName:"ResizeBilinear",category:"image",inputs:[{start:0,name:"images",type:"tensor"},{start:1,name:"size",type:"number[]"}],attrs:[{tfName:"align_corners",name:"alignCorners",type:"bool"},{tfName:"half_pixel_centers",name:"halfPixelCenters",type:"bool"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"ResizeNearestNeighbor",category:"image",inputs:[{start:0,name:"images",type:"tensor"},{start:1,name:"size",type:"number[]"}],attrs:[{tfName:"align_corners",name:"alignCorners",type:"bool"},{tfName:"half_pixel_centers",name:"halfPixelCenters",type:"bool"},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"CropAndResize",category:"image",inputs:[{start:0,name:"image",type:"tensor"},{start:1,name:"boxes",type:"tensor"},{start:2,name:"boxInd",type:"tensor"},{start:3,name:"cropSize",type:"number[]"}],attrs:[{tfName:"method",name:"method",type:"string"},{tfName:"extrapolation_value",name:"extrapolationValue",type:"number"}]},{tfOpName:"ImageProjectiveTransformV3",category:"image",inputs:[{start:0,name:"images",type:"tensor"},{start:1,name:"transforms",type:"tensor"},{start:2,name:"outputShape",type:"number[]"},{start:3,name:"fillValue",type:"number"}],attrs:[{tfName:"interpolation",name:"interpolation",type:"string"},{tfName:"fill_mode",name:"fillMode",type:"string"}]}],Lk={};Ve(Lk,{json:()=>mj});var mj=[{tfOpName:"Equal",category:"logical",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"NotEqual",category:"logical",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Greater",category:"logical",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"GreaterEqual",category:"logical",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Less",category:"logical",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"LessEqual",category:"logical",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"LogicalAnd",category:"logical",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"LogicalNot",category:"logical",inputs:[{start:0,name:"a",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"LogicalOr",category:"logical",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Select",category:"logical",inputs:[{start:0,name:"condition",type:"tensor"},{start:1,name:"a",type:"tensor"},{start:2,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"SelectV2",category:"logical",inputs:[{start:0,name:"condition",type:"tensor"},{start:1,name:"a",type:"tensor"},{start:2,name:"b",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]}],Bk={};Ve(Bk,{json:()=>gj});var gj=[{tfOpName:"_FusedMatMul",category:"matrices",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"},{start:2,end:0,name:"args",type:"tensors"}],attrs:[{tfName:"num_args",name:"numArgs",type:"number"},{tfName:"fused_ops",name:"fusedOps",type:"string[]",defaultValue:[]},{tfName:"epsilon",name:"epsilon",type:"number",defaultValue:1e-4},{tfName:"transpose_a",name:"transposeA",type:"bool",defaultValue:!1},{tfName:"transpose_b",name:"transposeB",type:"bool",defaultValue:!1},{tfName:"leakyrelu_alpha",name:"leakyreluAlpha",type:"number",defaultValue:.2},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"MatMul",category:"matrices",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"transpose_a",name:"transposeA",type:"bool",defaultValue:!1},{tfName:"transpose_b",name:"transposeB",type:"bool",defaultValue:!1},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"BatchMatMul",category:"matrices",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"adj_x",name:"transposeA",type:"bool",defaultValue:!1},{tfName:"adj_y",name:"transposeB",type:"bool",defaultValue:!1},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"BatchMatMulV2",category:"matrices",inputs:[{start:0,name:"a",type:"tensor"},{start:1,name:"b",type:"tensor"}],attrs:[{tfName:"adj_x",name:"transposeA",type:"bool",defaultValue:!1},{tfName:"adj_y",name:"transposeB",type:"bool",defaultValue:!1},{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Transpose",category:"matrices",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"perm",type:"number[]"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"Einsum",category:"matrices",inputs:[{start:0,end:0,name:"tensors",type:"tensors"}],attrs:[{tfName:"equation",name:"equation",type:"string"},{tfName:"N",name:"n",type:"number",defaultValue:2},{tfName:"T",name:"dtype",type:"dtype"}]}],Wk={};Ve(Wk,{json:()=>yj});var yj=[{tfOpName:"EuclideanNorm",category:"normalization",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool",defaultValue:!1}]},{tfOpName:"FusedBatchNorm",category:"normalization",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"scale",type:"tensor"},{start:2,name:"offset",type:"tensor"},{start:3,name:"mean",type:"tensor"},{start:4,name:"variance",type:"tensor"}],attrs:[{tfName:"epsilon",name:"epsilon",type:"number",defaultValue:.001},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0}]},{tfOpName:"FusedBatchNormV2",category:"normalization",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"scale",type:"tensor"},{start:2,name:"offset",type:"tensor"},{start:3,name:"mean",type:"tensor"},{start:4,name:"variance",type:"tensor"}],attrs:[{tfName:"epsilon",name:"epsilon",type:"number",defaultValue:.001},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0}]},{tfOpName:"FusedBatchNormV3",category:"normalization",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"scale",type:"tensor"},{start:2,name:"offset",type:"tensor"},{start:3,name:"mean",type:"tensor"},{start:4,name:"variance",type:"tensor"}],attrs:[{tfName:"epsilon",name:"epsilon",type:"number",defaultValue:.001},{tfName:"data_format",name:"dataFormat",type:"string",notSupported:!0}]},{tfOpName:"LRN",category:"normalization",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"depth_radius",name:"radius",type:"number",defaultValue:5},{tfName:"bias",name:"bias",type:"number",defaultValue:1},{tfName:"alpha",name:"alpha",type:"number",defaultValue:1},{tfName:"beta",name:"beta",type:"number",defaultValue:.5}]},{tfOpName:"Softmax",category:"normalization",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"LogSoftmax",category:"normalization",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"SparseToDense",category:"normalization",inputs:[{start:0,name:"sparseIndices",type:"tensor"},{start:1,name:"outputShape",type:"number[]"},{start:2,name:"sparseValues",type:"tensor"},{start:3,name:"defaultValue",type:"tensor"}],attrs:[{tfName:"validate_indices",name:"validateIndices",type:"bool",defaultValue:!0,notSupported:!0}]}],Vk={};Ve(Vk,{json:()=>Aj});var Aj=[{tfOpName:"Bincount",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"size",type:"number"},{start:2,name:"weights",type:"tensor"}]},{tfOpName:"DenseBincount",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"size",type:"number"},{start:2,name:"weights",type:"tensor"}],attrs:[{tfName:"binary_output",name:"binaryOutput",type:"bool"}]},{tfOpName:"Max",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool"}]},{tfOpName:"Mean",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool"}]},{tfOpName:"Min",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool"}]},{tfOpName:"Sum",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool"}]},{tfOpName:"All",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool"}]},{tfOpName:"Any",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool"}]},{tfOpName:"ArgMax",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number"}]},{tfOpName:"ArgMin",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number"}]},{tfOpName:"Prod",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}],attrs:[{tfName:"keep_dims",name:"keepDims",type:"bool"}]},{tfOpName:"Cumprod",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number"}],attrs:[{tfName:"exclusive",name:"exclusive",type:"bool"},{tfName:"reverse",name:"reverse",type:"bool"}]},{tfOpName:"Cumsum",category:"reduction",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number"}],attrs:[{tfName:"exclusive",name:"exclusive",type:"bool"},{tfName:"reverse",name:"reverse",type:"bool"}]}],Uk={};Ve(Uk,{json:()=>xj});var xj=[{tfOpName:"ConcatV2",category:"slice_join",inputs:[{start:0,end:-1,name:"tensors",type:"tensors"},{start:-1,name:"axis",type:"number"}],attrs:[{tfName:"N",name:"n",type:"number",defaultValue:2}]},{tfOpName:"Concat",category:"slice_join",inputs:[{start:1,end:0,name:"tensors",type:"tensors"},{start:0,name:"axis",type:"number"}],attrs:[{tfName:"N",name:"n",type:"number",defaultValue:2}]},{tfOpName:"GatherV2",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"indices",type:"tensor"},{start:2,name:"axis",type:"number",defaultValue:0}],attrs:[{tfName:"batch_dims",name:"batchDims",type:"number",defaultValue:0}]},{tfOpName:"Gather",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"indices",type:"tensor"}],attrs:[{tfName:"validate_indices",name:"validateIndices",type:"bool",notSupported:!0}]},{tfOpName:"Reverse",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"dims",type:"bool[]"}]},{tfOpName:"ReverseV2",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number[]"}]},{tfOpName:"Slice",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"begin",type:"number[]"},{start:2,name:"size",type:"number[]"}]},{tfOpName:"StridedSlice",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"begin",type:"number[]"},{start:2,name:"end",type:"number[]"},{start:3,name:"strides",type:"number[]"}],attrs:[{tfName:"begin_mask",name:"beginMask",type:"number",defaultValue:0},{tfName:"end_mask",name:"endMask",type:"number",defaultValue:0},{tfName:"new_axis_mask",name:"newAxisMask",type:"number",defaultValue:0},{tfName:"ellipsis_mask",name:"ellipsisMask",type:"number",defaultValue:0},{tfName:"shrink_axis_mask",name:"shrinkAxisMask",type:"number",defaultValue:0}]},{tfOpName:"Pack",category:"slice_join",inputs:[{start:0,end:0,name:"tensors",type:"tensors"}],attrs:[{tfName:"axis",name:"axis",type:"number",defaultValue:0}]},{tfOpName:"Unpack",category:"slice_join",inputs:[{start:0,name:"tensor",type:"tensor"}],attrs:[{tfName:"axis",name:"axis",type:"number",defaultValue:0},{tfName:"num",name:"num",type:"number",defaultValue:0,notSupported:!0}]},{tfOpName:"Tile",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"reps",type:"number[]"}]},{tfOpName:"Split",category:"slice_join",inputs:[{start:0,name:"axis",type:"number",defaultValue:0},{start:1,name:"x",type:"tensor"}],attrs:[{tfName:"num_split",name:"numOrSizeSplits",type:"number",defaultValue:1}]},{tfOpName:"SplitV",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"numOrSizeSplits",type:"number[]"},{start:2,name:"axis",type:"number",defaultValue:0}]},{tfOpName:"ScatterNd",category:"slice_join",inputs:[{start:0,name:"indices",type:"tensor"},{start:1,name:"values",type:"tensor"},{start:2,name:"shape",type:"number[]"}]},{tfOpName:"GatherNd",category:"slice_join",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"indices",type:"tensor"}]},{tfOpName:"SparseToDense",category:"slice_join",inputs:[{start:0,name:"sparseIndices",type:"tensor"},{start:1,name:"outputShape",type:"number[]"},{start:2,name:"sparseValues",type:"tensor"},{start:3,name:"defaultValue",type:"tensor"}],attrs:[{tfName:"validate_indices",name:"validateIndices",type:"bool",defaultValue:!1,notSupported:!0}]}],Gk={};Ve(Gk,{json:()=>bj});var bj=[{tfOpName:"SparseFillEmptyRows",category:"sparse",inputs:[{start:0,name:"indices",type:"tensor"},{start:1,name:"values",type:"tensor"},{start:2,name:"denseShape",type:"tensor"},{start:3,name:"defaultValue",type:"tensor"}]},{tfOpName:"SparseReshape",category:"sparse",inputs:[{start:0,name:"inputIndices",type:"tensor"},{start:1,name:"inputShape",type:"tensor"},{start:2,name:"newShape",type:"tensor"}],attrs:[{tfName:"T",name:"dtype",type:"dtype",notSupported:!0}]},{tfOpName:"SparseSegmentMean",category:"sparse",inputs:[{start:0,name:"data",type:"tensor"},{start:1,name:"indices",type:"tensor"},{start:2,name:"segmentIds",type:"tensor"}]},{tfOpName:"SparseSegmentSum",category:"sparse",inputs:[{start:0,name:"data",type:"tensor"},{start:1,name:"indices",type:"tensor"},{start:2,name:"segmentIds",type:"tensor"}]}],Hk={};Ve(Hk,{json:()=>vj});var vj=[{tfOpName:"FFT",category:"spectral",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"IFFT",category:"spectral",inputs:[{start:0,name:"x",type:"tensor"}]},{tfOpName:"RFFT",category:"spectral",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"fft_length",type:"number",notSupported:!0}]},{tfOpName:"IRFFT",category:"spectral",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"fft_length",type:"number",notSupported:!0}]}],jk={};Ve(jk,{json:()=>wj});var wj=[{tfOpName:"StringNGrams",category:"string",inputs:[{start:0,name:"data",type:"tensor"},{start:1,name:"dataSplits",type:"tensor"}],attrs:[{tfName:"separator",name:"separator",type:"string"},{tfName:"ngram_widths",name:"nGramWidths",type:"number[]"},{tfName:"left_pad",name:"leftPad",type:"string"},{tfName:"right_pad",name:"rightPad",type:"string"},{tfName:"pad_width",name:"padWidth",type:"number"},{tfName:"preserve_short_sequences",name:"preserveShortSequences",type:"bool"}],outputs:["ngrams","ngrams_splits"]},{tfOpName:"StringSplit",category:"string",inputs:[{start:0,name:"input",type:"tensor"},{start:1,name:"delimiter",type:"tensor"}],attrs:[{tfName:"skip_empty",name:"skipEmpty",type:"bool"}],outputs:["indices","values","shape"]},{tfOpName:"StringToHashBucketFast",category:"string",inputs:[{start:0,name:"input",type:"tensor"}],attrs:[{tfName:"num_buckets",name:"numBuckets",type:"number"}]}],qk={};Ve(qk,{json:()=>kj});var kj=[{tfOpName:"Cast",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"SrcT",name:"sdtype",type:"dtype",notSupported:!0},{tfName:"DstT",name:"dtype",type:"dtype"}]},{tfOpName:"ExpandDims",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"axis",type:"number"}]},{tfOpName:"MirrorPad",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"padding",type:"number[]"}],attrs:[{tfName:"mode",name:"mode",type:"string"}]},{tfOpName:"Pad",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"padding",type:"number[]"}],attrs:[{tfName:"constant_value",name:"constantValue",type:"number",defaultValue:0}]},{tfOpName:"PadV2",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"padding",type:"number[]"},{start:2,name:"constantValue",type:"number",defaultValue:0}]},{tfOpName:"Reshape",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"shape",type:"number[]"}]},{tfOpName:"Squeeze",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"axis",tfDeprecatedName:"squeeze_dims",name:"axis",type:"number[]"}]},{tfOpName:"SpaceToBatchND",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"blockShape",type:"number[]"},{start:2,name:"paddings",type:"number[]"}]},{tfOpName:"BatchToSpaceND",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"blockShape",type:"number[]"},{start:2,name:"crops",type:"number[]"}]},{tfOpName:"DepthToSpace",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"}],attrs:[{tfName:"block_size",name:"blockSize",type:"number"},{tfName:"data_format",name:"dataFormat",type:"string"}]},{tfOpName:"BroadcastTo",category:"transformation",inputs:[{start:0,name:"x",type:"tensor"},{start:1,name:"shape",type:"number[]"}],attrs:[]},{tfOpName:"BroadcastArgs",category:"transformation",inputs:[{start:0,name:"s0",type:"tensor"},{start:1,name:"s1",type:"tensor"}],attrs:[]}],n7=class{static get Instance(){return this._instance||(this._instance=new this)}constructor(){let e=[Ek,Rk,_k,Dk,$k,Pk,Fk,Ok,Mk,zk,Lk,Bk,Wk,Vk,Uk,Gk,Hk,jk,qk],t=[].concat(...e.map(n=>n.json));this.opMappers=t.reduce((n,s)=>(n[s.tfOpName]=s,n),{})}transformGraph(e,t={}){let n=e.node,s=[],r=[],a=[],o=n.reduce((f,m)=>(f[m.name]=this.mapNode(m),m.op.startsWith("Placeholder")?s.push(f[m.name]):m.op==="Const"?r.push(f[m.name]):(m.input==null||m.input.length===0)&&a.push(f[m.name]),f),{}),i=[],l=[],u={},c={};t!=null&&(u=this.mapSignatureEntries(t.inputs),c=this.mapSignatureEntries(t.outputs));let p=Object.keys(o);p.forEach(f=>{let m=o[f];m.inputNames.forEach((g,y)=>{let[x,,A]=Gr(g),b=o[x];if(b.outputs!=null){let w=b.outputs.indexOf(A);if(w!==-1){let k=`${x}:${w}`;m.inputNames[y]=k}}m.inputs.push(b),b.children.push(m)})}),Object.keys(c).length===0?p.forEach(f=>{let m=o[f];m.children.length===0&&l.push(m)}):Object.keys(c).forEach(f=>{let[m]=Gr(f),g=o[m];g!=null&&(g.signatureKey=c[f],l.push(g))}),Object.keys(u).length>0?Object.keys(u).forEach(f=>{let[m]=Gr(f),g=o[m];g&&(g.signatureKey=u[f],i.push(g))}):i=s;let d={};e.library!=null&&e.library.function!=null&&(d=e.library.function.reduce((f,m)=>(f[m.signature.name]=this.mapFunction(m),f),{}));let h={nodes:o,inputs:i,outputs:l,weights:r,placeholders:s,signature:t,functions:d};return a.length>0&&(h.initNodes=a),h}mapSignatureEntries(e){return Object.keys(e||{}).reduce((t,n)=>(t[e[n].name]=n,t),{})}mapNode(e){let t=Nk(e.op)||this.opMappers[e.op]||{};e.attr==null&&(e.attr={});let n={name:e.name,op:e.op,category:t.category,inputNames:(e.input||[]).map(s=>s.startsWith("^")?s.slice(1):s),inputs:[],children:[],inputParams:{},attrParams:{},rawAttrs:e.attr,outputs:t.outputs};return t.inputs!=null&&(n.inputParams=t.inputs.reduce((s,r)=>(s[r.name]={type:r.type,inputIndexStart:r.start,inputIndexEnd:r.end},s),{})),t.attrs!=null&&(n.attrParams=t.attrs.reduce((s,r)=>{let a=r.type,o;switch(r.type){case"string":o=V3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=V3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"string[]":o=K3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=K3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"number":o=G3(e.attr,r.tfName,r.defaultValue||0),o===void 0&&!!r.tfDeprecatedName&&(o=G3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"number[]":o=X3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=X3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"bool":o=U3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=U3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"bool[]":o=Y3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=Y3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"shape":o=q3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=q3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"shape[]":o=Z3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=Z3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"dtype":o=H3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=H3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"dtype[]":o=j3(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=j3(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"func":o=s7(e.attr,r.tfName,r.defaultValue),o===void 0&&!!r.tfDeprecatedName&&(o=s7(e.attr,r.tfDeprecatedName,r.defaultValue));break;case"tensor":case"tensors":break;default:throw new Error(`Unsupported param type: ${r.type} for op: ${e.op}`)}return s[r.name]={value:o,type:a},s},{})),n}mapFunction(e){let t=e.nodeDef,n=[],s=[],r={};t!=null&&(r=t.reduce((c,p)=>(c[p.name]=this.mapNode(p),p.op==="Const"&&s.push(c[p.name]),c),{}));let a=[],o=[];e.signature.inputArg.forEach(c=>{let[p]=Gr(c.name),d={name:p,op:"Placeholder",inputs:[],inputNames:[],category:"graph",inputParams:{},attrParams:{dtype:{value:vx(c.type),type:"dtype"}},children:[]};d.signatureKey=c.name,a.push(d),r[p]=d}),Object.keys(r).forEach(c=>{let p=r[c];p.inputNames.forEach((d,h)=>{let[f,,m]=Gr(d),g=r[f];if(g.outputs!=null){let y=g.outputs.indexOf(m);if(y!==-1){let x=`${f}:${y}`;p.inputNames[h]=x}}p.inputs.push(g),g.children.push(p)})});let l=e.ret;e.signature.outputArg.forEach(c=>{let[p,d]=Gr(l[c.name]),h=r[p];h!=null&&(h.defaultOutput=d,o.push(h))});let u=this.mapArgsToSignature(e);return{nodes:r,inputs:a,outputs:o,weights:s,placeholders:n,signature:u}}mapArgsToSignature(e){return{methodName:e.signature.name,inputs:e.signature.inputArg.reduce((t,n)=>(t[n.name]=this.mapArgToTensorInfo(n),t),{}),outputs:e.signature.outputArg.reduce((t,n)=>(t[n.name]=this.mapArgToTensorInfo(n,e.ret),t),{})}}mapArgToTensorInfo(e,t){let n=e.name;return t!=null&&(n=t[n]),{name:n,dtype:e.type}}};function Ij(e){let t=q().global;if(typeof t.atob!="undefined")return t.atob(e);if(typeof Buffer!="undefined")return new Buffer(e,"base64").toString();throw new Error("Unable to decode base64 in this environment. Missing built-in atob() or Buffer()")}function Xk(e,t){let n=Array.isArray(e)?String.fromCharCode.apply(null,e):Ij(e);return t?n:n.toLowerCase()}function V3(e,t,n,s=!1){let r=e[t];return r!=null?Xk(r.s,s):n}function U3(e,t,n){let s=e[t];return s?s.b:n}function G3(e,t,n){let s=e[t]||{},r=s.i!=null?s.i:s.f!=null?s.f:n;return typeof r=="number"?r:parseInt(r,10)}function vx(e){switch(typeof e=="string"&&(e=rr[e]),e){case rr.DT_FLOAT:case rr.DT_HALF:return"float32";case rr.DT_INT32:case rr.DT_INT64:case rr.DT_INT8:case rr.DT_UINT8:return"int32";case rr.DT_BOOL:return"bool";case rr.DT_DOUBLE:return"float32";case rr.DT_STRING:return"string";default:return null}}function s7(e,t,n){let s=e[t];return s&&s.func?s.func.name:n}function H3(e,t,n){let s=e[t];return s&&s.type?vx(s.type):n}function j3(e,t,n){let s=e[t];return s&&s.list&&s.list.type?s.list.type.map(r=>vx(r)):n}function Kk(e){if(!e.unknownRank)return e.dim!=null?e.dim.map(t=>typeof t.size=="number"?t.size:parseInt(t.size,10)):[]}function q3(e,t,n){let s=e[t];return s&&s.shape?Kk(s.shape):n}function X3(e,t,n){let s=e[t];return s?((s.list.f&&s.list.f.length?s.list.f:s.list.i)||[]).map(r=>typeof r=="number"?r:parseInt(r,10)):n}function K3(e,t,n,s=!1){let r=e[t];return r&&r.list&&r.list.s?r.list.s.map(a=>Xk(a,s)):n}function Z3(e,t,n){let s=e[t];return s&&s.list&&s.list.shape?s.list.shape.map(r=>Kk(r)):n}function Y3(e,t,n){let s=e[t];return s&&s.list&&s.list.b?s.list.b:n}var Sj=class{constructor(e,t,n){this.node=e,this.tensorMap=t,this.context=n,this.inputs=[],this.attrs={},this.inputs=e.inputNames.map(s=>this.getInput(s)),e.rawAttrs!=null&&(this.attrs=Object.keys(e.rawAttrs).reduce((s,r)=>(s[r]=this.getAttr(r),s),{}))}getInput(e){return Qn(e,this.tensorMap,this.context)}getAttr(e,t){let n=this.node.rawAttrs[e];if(n.tensor!=null)return Qn(e,this.tensorMap,this.context);if(n.i!=null||n.f!=null)return G3(this.node.rawAttrs,e,t);if(n.s!=null)return V3(this.node.rawAttrs,e,t);if(n.b!=null)return U3(this.node.rawAttrs,e,t);if(n.shape!=null)return q3(this.node.rawAttrs,e,t);if(n.type!=null)return H3(this.node.rawAttrs,e,t);if(n.list!=null){if(n.list.i!=null||n.list.f!=null)return X3(this.node.rawAttrs,e,t);if(n.list.s!=null)return K3(this.node.rawAttrs,e,t);if(n.list.shape!=null)return Z3(this.node.rawAttrs,e,t);if(n.list.b!=null)return Y3(this.node.rawAttrs,e,t);if(n.list.type!=null)return j3(this.node.rawAttrs,e,t)}return t}},Rn={};Ve(Rn,{OP_SCOPE_SUFFIX:()=>Ty,abs:()=>tn,acos:()=>jy,acosh:()=>qy,add:()=>ce,addN:()=>m0,all:()=>g0,any:()=>Cp,argMax:()=>Es,argMin:()=>Xy,asin:()=>Ky,asinh:()=>Zy,atan:()=>Yy,atan2:()=>Jy,atanh:()=>Qy,avgPool:()=>lh,avgPool3d:()=>tA,basicLSTMCell:()=>ow,batchNorm:()=>Wc,batchNorm2d:()=>nA,batchNorm3d:()=>sA,batchNorm4d:()=>rA,batchToSpaceND:()=>uh,bincount:()=>aA,booleanMaskAsync:()=>Ww,broadcastArgs:()=>iw,broadcastTo:()=>Gu,buffer:()=>Le,cast:()=>ye,ceil:()=>oA,clipByValue:()=>fs,clone:()=>On,complex:()=>fa,concat:()=>Ct,concat1d:()=>iA,concat2d:()=>Kl,concat3d:()=>lA,concat4d:()=>uA,conv1d:()=>y0,conv2d:()=>ma,conv2dTranspose:()=>A0,conv3d:()=>dA,conv3dTranspose:()=>pA,cos:()=>ch,cosh:()=>x0,cosineWindow:()=>W0,cumprod:()=>Tp,cumsum:()=>b0,denseBincount:()=>uw,depthToSpace:()=>hA,depthwiseConv2d:()=>Vc,diag:()=>cw,dilation2d:()=>fA,div:()=>he,divNoNan:()=>mA,dot:()=>gA,dropout:()=>VA,einsum:()=>dw,elu:()=>Uc,enclosingPowerOfTwo:()=>UA,equal:()=>Rs,erf:()=>yA,euclideanNorm:()=>bA,exp:()=>_s,expandDims:()=>Kt,expm1:()=>vA,eye:()=>v0,fft:()=>xh,fill:()=>Hc,floor:()=>jc,floorDiv:()=>Bc,fused:()=>ec,gather:()=>qc,gatherND:()=>Hw,greater:()=>ys,greaterEqual:()=>li,ifft:()=>Qu,imag:()=>oh,image:()=>Se,inTopKAsync:()=>jw,irfft:()=>M0,isFinite:()=>wA,isInf:()=>kA,isNaN:()=>IA,leakyRelu:()=>dh,less:()=>w0,lessEqual:()=>ui,linalg:()=>jA,linspace:()=>gw,localResponseNormalization:()=>SA,log:()=>Ds,log1p:()=>ph,logSigmoid:()=>CA,logSoftmax:()=>I0,logSumExp:()=>S0,logicalAnd:()=>ur,logicalNot:()=>hh,logicalOr:()=>C0,logicalXor:()=>TA,losses:()=>r8,lowerBound:()=>Aw,matMul:()=>Qe,max:()=>mn,maxPool:()=>fh,maxPool3d:()=>NA,maxPoolWithArgmax:()=>xw,maximum:()=>Jr,mean:()=>Bt,meshgrid:()=>bw,min:()=>ga,minimum:()=>Xc,mirrorPad:()=>EA,mod:()=>Yl,moments:()=>mh,movingAverage:()=>Vw,mul:()=>L,multiRNNCell:()=>vw,multinomial:()=>ww,neg:()=>Dt,norm:()=>Gc,notEqual:()=>Qi,oneHot:()=>Zu,ones:()=>Ns,onesLike:()=>$s,op:()=>V,outerProduct:()=>kw,pad:()=>Qs,pad1d:()=>Iw,pad2d:()=>Sw,pad3d:()=>Cw,pad4d:()=>Tw,pool:()=>RA,pow:()=>ya,prelu:()=>yh,print:()=>_y,prod:()=>_A,rand:()=>Nw,randomGamma:()=>Ew,randomNormal:()=>N0,randomStandardNormal:()=>Rw,randomUniform:()=>Kc,range:()=>Ju,real:()=>Yu,reciprocal:()=>PA,relu:()=>Pr,relu6:()=>E0,reshape:()=>U,reverse:()=>Zs,reverse1d:()=>_w,reverse2d:()=>Dw,reverse3d:()=>$w,reverse4d:()=>Pw,rfft:()=>bh,round:()=>R0,rsqrt:()=>_0,scalar:()=>Ce,scatterND:()=>Uw,searchSorted:()=>T0,selu:()=>D0,separableConv2d:()=>$0,setdiff1dAsync:()=>Fw,sigmoid:()=>Cn,sign:()=>FA,signal:()=>s8,sin:()=>P0,sinh:()=>F0,slice:()=>Oe,slice1d:()=>Ah,slice2d:()=>O0,slice3d:()=>ci,slice4d:()=>no,softmax:()=>Jl,softplus:()=>Zl,spaceToBatchND:()=>gh,sparse:()=>a8,sparseToDense:()=>Gw,spectral:()=>n8,split:()=>Zt,sqrt:()=>Nn,square:()=>vt,squaredDifference:()=>z0,squeeze:()=>st,stack:()=>on,step:()=>Ql,stridedSlice:()=>OA,string:()=>o8,sub:()=>fe,sum:()=>ke,tan:()=>MA,tanh:()=>Yi,tensor:()=>ct,tensor1d:()=>Ft,tensor2d:()=>ir,tensor3d:()=>zy,tensor4d:()=>Ow,tensor5d:()=>Mw,tensor6d:()=>zw,tile:()=>qs,topk:()=>zA,transpose:()=>et,truncatedNormal:()=>L0,unique:()=>LA,unsortedSegmentSum:()=>B0,unstack:()=>En,upperBound:()=>Lw,variable:()=>BA,where:()=>zn,whereAsync:()=>WA,zeros:()=>Wt,zerosLike:()=>it});var Cj=(e,t,n,s=Rn)=>{switch(e.op){case"BiasAdd":case"AddV2":case"Add":return[s.add(I("a",e,t,n),I("b",e,t,n))];case"AddN":return[s.addN(I("tensors",e,t,n))];case"FloorMod":case"Mod":return[s.mod(I("a",e,t,n),I("b",e,t,n))];case"Mul":return[s.mul(I("a",e,t,n),I("b",e,t,n))];case"RealDiv":case"Div":return[s.div(I("a",e,t,n),I("b",e,t,n))];case"DivNoNan":return[s.divNoNan(I("a",e,t,n),I("b",e,t,n))];case"FloorDiv":return[s.floorDiv(I("a",e,t,n),I("b",e,t,n))];case"Sub":return[s.sub(I("a",e,t,n),I("b",e,t,n))];case"Minimum":return[s.minimum(I("a",e,t,n),I("b",e,t,n))];case"Maximum":return[s.maximum(I("a",e,t,n),I("b",e,t,n))];case"Pow":return[s.pow(I("a",e,t,n),I("b",e,t,n))];case"SquaredDifference":return[s.squaredDifference(I("a",e,t,n),I("b",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Tj=(e,t,n,s=Rn)=>{switch(e.op){case"Abs":case"ComplexAbs":return[s.abs(I("x",e,t,n))];case"Acos":return[s.acos(I("x",e,t,n))];case"Acosh":return[s.acosh(I("x",e,t,n))];case"Asin":return[s.asin(I("x",e,t,n))];case"Asinh":return[s.asinh(I("x",e,t,n))];case"Atan":return[s.atan(I("x",e,t,n))];case"Atan2":return[s.atan2(I("x",e,t,n),I("y",e,t,n))];case"Atanh":return[s.atanh(I("x",e,t,n))];case"Ceil":return[s.ceil(I("x",e,t,n))];case"Complex":return[s.complex(I("real",e,t,n),I("imag",e,t,n))];case"Cos":return[s.cos(I("x",e,t,n))];case"Cosh":return[s.cosh(I("x",e,t,n))];case"Elu":return[s.elu(I("x",e,t,n))];case"Erf":return[s.erf(I("x",e,t,n))];case"Exp":return[s.exp(I("x",e,t,n))];case"Expm1":return[s.expm1(I("x",e,t,n))];case"Floor":return[s.floor(I("x",e,t,n))];case"Log":return[s.log(I("x",e,t,n))];case"Log1p":return[s.log1p(I("x",e,t,n))];case"Imag":return[s.imag(I("x",e,t,n))];case"Neg":return[s.neg(I("x",e,t,n))];case"Reciprocal":return[s.reciprocal(I("x",e,t,n))];case"Real":return[s.real(I("x",e,t,n))];case"Relu":return[s.relu(I("x",e,t,n))];case"Round":return[s.round(I("x",e,t,n))];case"Selu":return[s.selu(I("x",e,t,n))];case"Sigmoid":return[s.sigmoid(I("x",e,t,n))];case"Sin":return[s.sin(I("x",e,t,n))];case"Sign":return[s.sign(I("x",e,t,n))];case"Sinh":return[s.sinh(I("x",e,t,n))];case"Softplus":return[s.softplus(I("x",e,t,n))];case"Sqrt":return[s.sqrt(I("x",e,t,n))];case"Square":return[s.square(I("x",e,t,n))];case"Tanh":return[s.tanh(I("x",e,t,n))];case"Tan":return[s.tan(I("x",e,t,n))];case"ClipByValue":return[s.clipByValue(I("x",e,t,n),I("clipValueMin",e,t,n),I("clipValueMax",e,t,n))];case"Relu6":return[s.relu6(I("x",e,t,n))];case"Rsqrt":return[s.rsqrt(Qn(e.inputNames[0],t,n))];case"Prod":return[s.prod(I("x",e,t,n),I("axes",e,t,n))];case"LeakyRelu":return[s.leakyRelu(I("x",e,t,n),I("alpha",e,t,n))];case"Prelu":return[s.prelu(I("x",e,t,n),I("alpha",e,t,n))];case"IsNan":return[s.isNaN(Qn(e.inputNames[0],t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}};function or(e,t,n=""){if(!(typeof e=="number"||typeof t=="number")){v.assert(e.length===t.length,()=>n+` Shapes ${e} and ${t} must match`);for(let s=0;s<e.length;s++){let r=e[s],a=t[s];v.assert(r<0||a<0||r===a,()=>n+` Shapes ${e} and ${t} must match`)}}}function r7(e){return!(typeof e=="number"||e.some(t=>t<0))}function tp(e,t,n){let s=J3(e,n),r=!r7(s);if(r&&t.length===0)throw new Error(`Tried to calculate elements of an empty list with non-fully-defined elementShape: ${s}`);if(r&&t.forEach(a=>{s=J3(a.shape,s)}),!r7(s))throw new Error(`Non-fully-defined elementShape: ${s}`);return s}function J3(e,t){if(typeof e=="number")return t;if(typeof t=="number")return e;if(e.length!==t.length)throw new Error(`Incompatible ranks during merge: ${e} vs. ${t}`);let n=[];for(let s=0;s<e.length;++s){let r=e[s],a=t[s];if(r>=0&&a>=0&&r!==a)throw new Error(`Incompatible shape during merge: ${e} vs. ${t}`);n[s]=r>=0?r:a}return n}var Nj=class{constructor(e,t,n,s,r,a,o){this.name=e,this.dtype=t,this.maxSize=n,this.elementShape=s,this.identicalElementShapes=r,this.dynamicSize=a,this.clearAfterRead=o,this.tensors=[],this.closed_=!1,this.idTensor=Ce(0),bn(this.idTensor)}get id(){return this.idTensor.id}get closed(){return this.closed_}clearAndClose(e){this.tensors.forEach(t=>{(e==null||!e.has(t.tensor.id))&&t.tensor.dispose()}),this.tensors=[],this.closed_=!0,this.idTensor.dispose()}size(){return this.tensors.length}read(e){if(this.closed_)throw new Error(`TensorArray ${this.name} has already been closed.`);if(e<0||e>=this.size())throw new Error(`Tried to read from index ${e}, but array size is: ${this.size()}`);let t=this.tensors[e];if(t.cleared)throw new Error(`TensorArray ${this.name}: Could not read index ${e} twice because it was cleared after a previous read (perhaps try setting clear_after_read = false?).`);return this.clearAfterRead&&(t.cleared=!0),t.read=!0,t.tensor}readMany(e){return e.map(t=>this.read(t))}write(e,t){if(this.closed_)throw new Error(`TensorArray ${this.name} has already been closed.`);if(e<0||!this.dynamicSize&&e>=this.maxSize)throw new Error(`Tried to write to index ${e}, but array is not resizeable and size is: ${this.maxSize}`);let n=this.tensors[e]||{};if(t.dtype!==this.dtype)throw new Error(`TensorArray ${this.name}: Could not write to TensorArray index ${e},
|
|
because the value dtype is ${t.dtype}, but TensorArray dtype is ${this.dtype}.`);if(this.size()===0&&(this.elementShape==null||this.elementShape.length===0)&&(this.elementShape=t.shape),or(this.elementShape,t.shape,`TensorArray ${this.name}: Could not write to TensorArray index ${e}.`),n.read)throw new Error(`TensorArray ${this.name}: Could not write to TensorArray index ${e}, because it has already been read.`);if(n.written)throw new Error(`TensorArray ${this.name}: Could not write to TensorArray index ${e}, because it has already been written.`);n.tensor=t,bn(t),n.written=!0,this.tensors[e]=n}writeMany(e,t){if(e.length!==t.length)throw new Error(`TensorArray ${this.name}: could not write multiple tensors,because the index size: ${e.length} is not the same as tensors size: ${t.length}.`);e.forEach((n,s)=>this.write(n,t[s]))}gather(e,t){if(!!t&&t!==this.dtype)throw new Error(`TensorArray dtype is ${this.dtype} but gather requested dtype ${t}`);if(e)e=e.slice(0,this.size());else{e=[];for(let s=0;s<this.size();s++)e.push(s)}if(e.length===0)return ct([],[0].concat(this.elementShape));let n=this.readMany(e);return or(this.elementShape,n[0].shape,"TensorArray shape mismatch: "),on(n,0)}concat(e){if(!!e&&e!==this.dtype)throw new Error(`TensorArray dtype is ${this.dtype} but concat requested dtype ${e}`);if(this.size()===0)return ct([],[0].concat(this.elementShape));let t=[];for(let s=0;s<this.size();s++)t.push(s);let n=this.readMany(t);return or(this.elementShape,n[0].shape,`TensorArray shape mismatch: tensor array shape (${this.elementShape}) vs first tensor shape (${n[0].shape})`),Ct(n,0)}scatter(e,t){if(t.dtype!==this.dtype)throw new Error(`TensorArray dtype is ${this.dtype} but tensor has dtype ${t.dtype}`);if(e.length!==t.shape[0])throw new Error(`Expected len(indices) == tensor.shape[0], but saw: ${e.length} vs. ${t.shape[0]}`);let n=Math.max(...e);if(!this.dynamicSize&&n>=this.maxSize)throw new Error(`Max index must be < array size (${n} vs. ${this.maxSize})`);this.writeMany(e,En(t,0))}split(e,t){if(t.dtype!==this.dtype)throw new Error(`TensorArray dtype is ${this.dtype} but tensor has dtype ${t.dtype}`);let n=0,s=e.map(i=>(n+=i,n));if(n!==t.shape[0])throw new Error(`Expected sum of lengths to be equal to
|
|
tensor.shape[0], but sum of lengths is
|
|
${n}, and tensor's shape is: ${t.shape}`);if(!this.dynamicSize&&e.length!==this.maxSize)throw new Error(`TensorArray's size is not equal to the size of lengths (${this.maxSize} vs. ${e.length}), and the TensorArray is not marked as dynamically resizeable`);let r=n===0?0:t.size/n,a=[];Y(()=>{t=U(t,[1,n,r]);for(let i=0;i<e.length;++i){let l=i===0?0:s[i-1],u=[0,l,0],c=[1,e[i],r];a[i]=U(Oe(t,u,c),this.elementShape)}return a});let o=[];for(let i=0;i<e.length;i++)o[i]=i;this.writeMany(o,a)}},rc=class{constructor(e,t,n,s=-1){this.tensors=e,this.elementShape=t,this.elementDtype=n,e!=null&&e.forEach(r=>{if(n!==r.dtype)throw new Error(`Invalid data types; op elements ${n}, but list elements ${r.dtype}`);or(t,r.shape,"TensorList shape mismatch: "),bn(r)}),this.idTensor=Ce(0),this.maxNumElements=s,bn(this.idTensor)}get id(){return this.idTensor.id}copy(){return new rc([...this.tensors],this.elementShape,this.elementDtype)}clearAndClose(e){this.tensors.forEach(t=>{(e==null||!e.has(t.id))&&t.dispose()}),this.tensors.length=0,this.idTensor.dispose()}size(){return this.tensors.length}stack(e,t,n=-1){if(t!==this.elementDtype)throw new Error(`Invalid data types; op elements ${t}, but list elements ${this.elementDtype}`);if(n!==-1&&this.tensors.length!==n)throw new Error(`Operation expected a list with ${n} elements but got a list with ${this.tensors.length} elements.`);or(e,this.elementShape,"TensorList shape mismatch: ");let s=tp(this.elementShape,this.tensors,e);return Y(()=>{let r=this.tensors.map(a=>U(a,s));return on(r,0)})}popBack(e,t){if(t!==this.elementDtype)throw new Error(`Invalid data types; op elements ${t}, but list elements ${this.elementDtype}`);if(this.size()===0)throw new Error("Trying to pop from an empty list.");let n=tp(this.elementShape,this.tensors,e),s=this.tensors.pop();return or(s.shape,e,"TensorList shape mismatch: "),U(s,n)}pushBack(e){if(e.dtype!==this.elementDtype)throw new Error(`Invalid data types; op elements ${e.dtype}, but list elements ${this.elementDtype}`);if(or(e.shape,this.elementShape,"TensorList shape mismatch: "),this.maxNumElements===this.size())throw new Error("Trying to push element into a full list.");bn(e),this.tensors.push(e)}resize(e){if(e<0)throw new Error(`TensorListResize expects size to be non-negative. Got: ${e}`);if(this.maxNumElements!==-1&&e>this.maxNumElements)throw new Error(`TensorListResize input size ${e} is greater maxNumElement ${this.maxNumElements}.`);let t=new rc([],this.elementShape,this.elementDtype,this.maxNumElements);t.tensors.length=e;for(let n=0;n<Math.min(this.tensors.length,e);++n)t.tensors[n]=this.tensors[n];return t}getItem(e,t,n){if(n!==this.elementDtype)throw new Error(`Invalid data types; op elements ${n}, but list elements ${this.elementDtype}`);if(e<0||e>this.tensors.length)throw new Error(`Trying to access element ${e} in a list with ${this.tensors.length} elements.`);if(this.tensors[e]==null)throw new Error(`element at index ${e} is null.`);or(this.tensors[e].shape,t,"TensorList shape mismatch: ");let s=tp(this.elementShape,this.tensors,t);return U(this.tensors[e],s)}setItem(e,t){if(t.dtype!==this.elementDtype)throw new Error(`Invalid data types; op elements ${t.dtype}, but list elements ${this.elementDtype}`);if(e<0||this.maxNumElements!==-1&&e>=this.maxNumElements)throw new Error(`Trying to set element ${e} in a list with max ${this.maxNumElements} elements.`);or(this.elementShape,t.shape,"TensorList shape mismatch: "),bn(t),this.tensors[e]=t}gather(e,t,n){if(t!==this.elementDtype)throw new Error(`Invalid data types; op elements ${t}, but list elements ${this.elementDtype}`);or(this.elementShape,n,"TensorList shape mismatch: "),e=e.slice(0,this.size());let s=tp(this.elementShape,this.tensors,n);return e.length===0?ct([],[0].concat(s)):Y(()=>{let r=e.map(a=>U(this.tensors[a],s));return on(r,0)})}concat(e,t){if(!!e&&e!==this.elementDtype)throw new Error(`TensorList dtype is ${this.elementDtype} but concat requested dtype ${e}`);or(this.elementShape,t,"TensorList shape mismatch: ");let n=tp(this.elementShape,this.tensors,t);return this.size()===0?ct([],[0].concat(n)):Y(()=>{let s=this.tensors.map(r=>U(r,n));return Ct(s,0)})}};function Ej(e,t,n){let s=e.dtype;if(e.shape.length<1)throw new Error(`Tensor must be at least a vector, but saw shape: ${e.shape}`);if(e.dtype!==n)throw new Error(`Invalid data types; op elements ${e.dtype}, but list elements ${n}`);let r=e.shape.slice(1);or(r,t,"TensorList shape mismatch: ");let a=En(e);return new rc(a,t,s)}function Rj(e,t,n,s){return new rc([],e,t,s)}function _j(e,t,n,s){if(t.length!==e.shape[0])throw new Error(`Expected len(indices) == tensor.shape[0], but saw: ${t.length} vs. ${e.shape[0]}`);let r=Math.max(...t);if(s!=null&&s!==-1&&r>=s)throw new Error(`Max index must be < array size (${r} vs. ${s})`);let a=new rc([],n,e.dtype,s),o=En(e,0);return t.forEach((i,l)=>{a.setItem(i,o[l])}),a}function Dj(e,t,n){let s=0,r=t.map(c=>(s+=c,s));if(s!==e.shape[0])throw new Error(`Expected sum of lengths to be equal to
|
|
tensor.shape[0], but sum of lengths is
|
|
${s}, and tensor's shape is: ${e.shape}`);let a=e.shape.slice(1),o=J3(a,n),i=s===0?0:e.size/s,l=Y(()=>{let c=[];e=U(e,[1,s,i]);for(let p=0;p<t.length;++p){let d=p===0?0:r[p-1],h=[0,d,0],f=[1,t[p],i];c[p]=U(Oe(e,h,f),o)}return e.dispose(),c}),u=new rc([],n,e.dtype,t.length);for(let c=0;c<l.length;c++)u.setItem(c,l[c]);return u}var $j=async(e,t,n)=>{switch(e.op){case"If":case"StatelessIf":{let s=I("thenBranch",e,t,n),r=I("elseBranch",e,t,n),a=I("cond",e,t,n),o=I("args",e,t,n);return(await a.data())[0]?n.functionMap[s].executeFunctionAsync(o,n.tensorArrayMap,n.tensorListMap):n.functionMap[r].executeFunctionAsync(o,n.tensorArrayMap,n.tensorListMap)}case"While":case"StatelessWhile":{let s=I("body",e,t,n),r=I("cond",e,t,n),a=I("args",e,t,n),o=await n.functionMap[r].executeFunctionAsync(a,n.tensorArrayMap,n.tensorListMap),i=a.map(c=>c.id),l=await o[0].data();o.forEach(c=>{!c.kept&&i.indexOf(c.id)===-1&&c.dispose()});let u=a;for(;l[0];){let c=u;u=await n.functionMap[s].executeFunctionAsync(u,n.tensorArrayMap,n.tensorListMap);let p=u.map(h=>h.id);c.forEach(h=>{!h.kept&&i.indexOf(h.id)===-1&&p.indexOf(h.id)===-1&&h.dispose()});let d=await n.functionMap[r].executeFunctionAsync(u,n.tensorArrayMap,n.tensorListMap);l=await d[0].data(),d.forEach(h=>{!h.kept&&i.indexOf(h.id)===-1&&p.indexOf(h.id)===-1&&h.dispose()})}return u}case"LoopCond":{let s=I("pred",e,t,n);return[ca(s)]}case"Switch":{let s=I("pred",e,t,n),r=I("data",e,t,n);return r.kept||(r=ca(r)),(await s.data())[0]?[void 0,r]:[r,void 0]}case"Merge":{let s=e.inputNames.find(r=>Qn(r,t,n)!==void 0);if(s){let r=Qn(s,t,n);return[ca(r)]}return}case"Enter":{let s=I("frameName",e,t,n),r=I("tensor",e,t,n);return n.enterFrame(s),[ca(r)]}case"Exit":{let s=I("tensor",e,t,n);return n.exitFrame(),[ca(s)]}case"NextIteration":{let s=I("tensor",e,t,n);return n.nextIteration(),[ca(s)]}case"TensorArrayV3":{let s=I("size",e,t,n),r=I("dtype",e,t,n),a=I("elementShape",e,t,n),o=I("dynamicSize",e,t,n),i=I("clearAfterRead",e,t,n),l=I("identicalElementShapes",e,t,n),u=I("name",e,t,n),c=new Nj(u,r,s,a,l,o,i);return n.addTensorArray(c),[c.idTensor,Ce(1)]}case"TensorArrayWriteV3":{let s=I("tensorArrayId",e,t,n),r=I("index",e,t,n),a=I("tensor",e,t,n),o=n.getTensorArray(s.id);return o.write(r,a),[o.idTensor]}case"TensorArrayReadV3":{let s=I("tensorArrayId",e,t,n),r=I("index",e,t,n);return[n.getTensorArray(s.id).read(r)]}case"TensorArrayGatherV3":{let s=I("tensorArrayId",e,t,n),r=I("indices",e,t,n),a=I("dtype",e,t,n);return[n.getTensorArray(s.id).gather(r,a)]}case"TensorArrayScatterV3":{let s=I("tensorArrayId",e,t,n),r=I("indices",e,t,n),a=I("tensor",e,t,n),o=n.getTensorArray(s.id);return o.scatter(r,a),[o.idTensor]}case"TensorArrayConcatV3":{let s=I("tensorArrayId",e,t,n),r=n.getTensorArray(s.id),a=I("dtype",e,t,n);return[r.concat(a)]}case"TensorArraySplitV3":{let s=I("tensorArrayId",e,t,n),r=I("tensor",e,t,n),a=I("lengths",e,t,n),o=n.getTensorArray(s.id);return o.split(a,r),[o.idTensor]}case"TensorArraySizeV3":{let s=I("tensorArrayId",e,t,n),r=n.getTensorArray(s.id);return[Ce(r.size(),"int32")]}case"TensorArrayCloseV3":{let s=I("tensorArrayId",e,t,n),r=n.getTensorArray(s.id);return r.clearAndClose(),[r.idTensor]}case"TensorListSetItem":{let s=I("tensorListId",e,t,n),r=I("index",e,t,n),a=I("tensor",e,t,n),o=n.getTensorList(s.id);return o.setItem(r,a),[o.idTensor]}case"TensorListGetItem":{let s=I("tensorListId",e,t,n),r=I("index",e,t,n),a=I("elementShape",e,t,n),o=I("elementDType",e,t,n);return[n.getTensorList(s.id).getItem(r,a,o)]}case"TensorListScatterV2":case"TensorListScatter":{let s=I("indices",e,t,n),r=I("tensor",e,t,n),a=I("elementShape",e,t,n),o=I("numElements",e,t,n),i=_j(r,s,a,o);return n.addTensorList(i),[i.idTensor]}case"TensorListReserve":case"EmptyTensorList":{let s=I("elementShape",e,t,n),r=I("elementDType",e,t,n),a;e.op==="TensorListReserve"?a="numElements":a="maxNumElements";let o=I(a,e,t,n),i=e.op==="TensorListReserve"?-1:o,l=Rj(s,r,o,i);return n.addTensorList(l),[l.idTensor]}case"TensorListGather":{let s=I("tensorListId",e,t,n),r=I("indices",e,t,n),a=I("elementShape",e,t,n),o=I("elementDType",e,t,n);return[n.getTensorList(s.id).gather(r,o,a)]}case"TensorListStack":{let s=I("tensorListId",e,t,n),r=I("elementShape",e,t,n),a=I("elementDType",e,t,n),o=I("numElements",e,t,n);return[n.getTensorList(s.id).stack(r,a,o)]}case"TensorListFromTensor":{let s=I("tensor",e,t,n),r=I("elementShape",e,t,n),a=I("elementDType",e,t,n),o=Ej(s,r,a);return n.addTensorList(o),[o.idTensor]}case"TensorListConcat":case"TensorListConcatV2":{let s=I("tensorListId",e,t,n),r=n.getTensorList(s.id),a=I("dtype",e,t,n),o=I("elementShape",e,t,n);return[r.concat(a,o)]}case"TensorListPushBack":{let s=I("tensorListId",e,t,n),r=I("tensor",e,t,n),a=n.getTensorList(s.id);return a.pushBack(r),[a.idTensor]}case"TensorListPopBack":{let s=I("tensorListId",e,t,n),r=I("elementShape",e,t,n),a=I("elementDType",e,t,n);return[n.getTensorList(s.id).popBack(r,a)]}case"TensorListSplit":{let s=I("tensor",e,t,n),r=I("elementShape",e,t,n),a=I("lengths",e,t,n),o=Dj(s,a,r);return n.addTensorList(o),[o.idTensor]}case"TensorListLength":{let s=I("tensorListId",e,t,n),r=n.getTensorList(s.id);return[Ce(r.size(),"int32")]}case"TensorListResize":{let s=I("tensorListId",e,t,n),r=I("size",e,t,n),o=n.getTensorList(s.id).resize(r);return n.addTensorList(o),[o.idTensor]}default:throw TypeError(`Node type ${e.op} is not implemented`)}};function a7(e,t,n){let[s,r]=I("fusedOps",e,t,n),a=s==="biasadd",o=!a,i=r==="prelu",l=s==="fusedbatchnorm",u=I("numArgs",e,t,n);if(a){if(i&&u!==2)throw new Error("FusedConv2d and DepthwiseConv2d with BiasAdd and Prelu must have two extra arguments: bias and alpha.");if(!i&&a&&u!==1)throw new Error("FusedConv2d and DepthwiseConv2d with BiasAdd must have one extra argument: bias.")}if(l)throw new Error("FusedConv2d and DepthwiseConv2d with FusedBatchNorm is not supported");let c=I("strides",e,t,n),p=tm(e,t,n),d=I("dataFormat",e,t,n).toUpperCase(),h=I("dilations",e,t,n),[f,m]=I("args",e,t,n);o&&(m=f,f=void 0);let g=I("leakyreluAlpha",e,t,n);return{stride:c,pad:p,dataFormat:d,dilations:h,biasArg:f,preluArg:m,activationFunc:r,leakyreluAlpha:g}}var Pj=(e,t,n,s=Rn)=>{switch(e.op){case"Conv1D":{let r=I("stride",e,t,n),a=I("pad",e,t,n),o=I("dataFormat",e,t,n).toUpperCase(),i=I("dilation",e,t,n);return[s.conv1d(I("x",e,t,n),I("filter",e,t,n),r,a,o,i)]}case"Conv2D":{let r=I("strides",e,t,n),a=tm(e,t,n),o=I("dataFormat",e,t,n).toUpperCase(),i=I("dilations",e,t,n);return[s.conv2d(I("x",e,t,n),I("filter",e,t,n),[r[1],r[2]],a,o,[i[1],i[2]])]}case"_FusedConv2D":{let{stride:r,pad:a,dataFormat:o,dilations:i,biasArg:l,preluArg:u,activationFunc:c,leakyreluAlpha:p}=a7(e,t,n);return[s.fused.conv2d({x:I("x",e,t,n),filter:I("filter",e,t,n),strides:[r[1],r[2]],pad:a,dataFormat:o,dilations:[i[1],i[2]],bias:l,activation:c,preluActivationWeights:u,leakyreluAlpha:p})]}case"FusedDepthwiseConv2dNative":{let{stride:r,pad:a,dataFormat:o,dilations:i,biasArg:l,preluArg:u,activationFunc:c,leakyreluAlpha:p}=a7(e,t,n);return[s.fused.depthwiseConv2d({x:I("x",e,t,n),filter:I("filter",e,t,n),strides:[r[1],r[2]],pad:a,dataFormat:o,dilations:[i[1],i[2]],bias:l,activation:c,preluActivationWeights:u,leakyreluAlpha:p})]}case"Conv2DBackpropInput":case"Conv2dTranspose":{let r=I("outputShape",e,t,n),a=I("strides",e,t,n),o=tm(e,t,n);return[s.conv2dTranspose(I("x",e,t,n),I("filter",e,t,n),r,[a[1],a[2]],o)]}case"DepthwiseConv2dNative":case"DepthwiseConv2d":{let r=I("strides",e,t,n),a=tm(e,t,n),o=I("dilations",e,t,n),i=I("dataFormat",e,t,n).toUpperCase();return[s.depthwiseConv2d(I("input",e,t,n),I("filter",e,t,n),[r[1],r[2]],a,i,[o[1],o[2]])]}case"Conv3D":{let r=I("strides",e,t,n),a=I("pad",e,t,n),o=I("dataFormat",e,t,n).toUpperCase(),i=I("dilations",e,t,n);return[s.conv3d(I("x",e,t,n),I("filter",e,t,n),[r[1],r[2],r[3]],a,o,[i[1],i[2],i[3]])]}case"AvgPool":{let r=I("strides",e,t,n),a=I("pad",e,t,n),o=I("kernelSize",e,t,n);return[s.avgPool(I("x",e,t,n),[o[1],o[2]],[r[1],r[2]],a)]}case"MaxPool":{let r=I("strides",e,t,n),a=I("pad",e,t,n),o=I("kernelSize",e,t,n);return[s.maxPool(I("x",e,t,n),[o[1],o[2]],[r[1],r[2]],a)]}case"MaxPoolWithArgmax":{let r=I("strides",e,t,n),a=I("pad",e,t,n),o=I("kernelSize",e,t,n),i=I("includeBatchInIndex",e,t,n),{result:l,indexes:u}=s.maxPoolWithArgmax(I("x",e,t,n),[o[1],o[2]],[r[1],r[2]],a,i);return[l,u]}case"AvgPool3D":{let r=I("strides",e,t,n),a=I("pad",e,t,n),o=I("kernelSize",e,t,n);return[s.avgPool3d(I("x",e,t,n),[o[1],o[2],o[3]],[r[1],r[2],r[3]],a)]}case"MaxPool3D":{let r=I("strides",e,t,n),a=I("pad",e,t,n),o=I("kernelSize",e,t,n);return[s.maxPool3d(I("x",e,t,n),[o[1],o[2],o[3]],[r[1],r[2],r[3]],a)]}case"Dilation2D":{let r=I("strides",e,t,n),a=I("pad",e,t,n),o=I("dilations",e,t,n),i=r[1],l=r[2],u=o[1],c=o[2];return[s.dilation2d(I("x",e,t,n),I("filter",e,t,n),[i,l],a,[u,c],"NHWC")]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Fj=(e,t,n,s=Rn)=>{switch(e.op){case"Fill":{let r=I("shape",e,t,n),a=I("dtype",e,t,n),o=I("value",e,t,n);return[s.fill(r,o,a)]}case"LinSpace":{let r=I("start",e,t,n),a=I("stop",e,t,n),o=I("num",e,t,n);return[s.linspace(r,a,o)]}case"Multinomial":{let r=I("logits",e,t,n),a=I("numSamples",e,t,n),o=I("seed",e,t,n);return[s.multinomial(r,a,o)]}case"OneHot":{let r=I("indices",e,t,n),a=I("depth",e,t,n),o=I("onValue",e,t,n),i=I("offValue",e,t,n);return[s.oneHot(r,a,o,i)]}case"Ones":return[s.ones(I("shape",e,t,n),I("dtype",e,t,n))];case"OnesLike":return[s.onesLike(I("x",e,t,n))];case"RandomStandardNormal":return[s.randomStandardNormal(I("shape",e,t,n),I("dtype",e,t,n),I("seed",e,t,n))];case"RandomUniform":return[s.randomUniform(I("shape",e,t,n),I("minval",e,t,n),I("maxval",e,t,n),I("dtype",e,t,n))];case"Range":{let r=I("start",e,t,n),a=I("stop",e,t,n),o=I("step",e,t,n);return[s.range(r,a,o,I("dtype",e,t,n))]}case"TruncatedNormal":{let r=I("shape",e,t,n),a=I("mean",e,t,n),o=I("stdDev",e,t,n),i=I("seed",e,t,n);return[s.truncatedNormal(r,a,o,I("dtype",e,t,n),i)]}case"Zeros":return[s.zeros(I("shape",e,t,n),I("dtype",e,t,n))];case"ZerosLike":return[s.zerosLike(I("x",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}};function p3(e,t,n){let s=I("boxes",e,t,n),r=I("scores",e,t,n),a=I("maxOutputSize",e,t,n),o=I("iouThreshold",e,t,n),i=I("scoreThreshold",e,t,n),l=I("softNmsSigma",e,t,n);return{boxes:s,scores:r,maxOutputSize:a,iouThreshold:o,scoreThreshold:i,softNmsSigma:l}}var Oj=async(e,t,n,s,r=Rn)=>{switch(e.op){case"NonMaxSuppressionV5":{let{boxes:a,scores:o,maxOutputSize:i,iouThreshold:l,scoreThreshold:u,softNmsSigma:c}=p3(e,t,n),p=await r.image.nonMaxSuppressionWithScoreAsync(a,o,i,l,u,c);return[p.selectedIndices,p.selectedScores]}case"NonMaxSuppressionV4":{let{boxes:a,scores:o,maxOutputSize:i,iouThreshold:l,scoreThreshold:u}=p3(e,t,n),c=I("padToMaxOutputSize",e,t,n),p=await r.image.nonMaxSuppressionPaddedAsync(a,o,i,l,u,c);return[p.selectedIndices,p.validOutputs]}case"NonMaxSuppressionV3":case"NonMaxSuppressionV2":{let{boxes:a,scores:o,maxOutputSize:i,iouThreshold:l,scoreThreshold:u}=p3(e,t,n);return[await r.image.nonMaxSuppressionAsync(a,o,i,l,u)]}case"Where":{let a=r.cast(I("condition",e,t,n),"bool"),o=[await r.whereAsync(a)];return a.dispose(),o}case"ListDiff":return r.setdiff1dAsync(I("x",e,t,n),I("y",e,t,n));default:throw TypeError(`Node type ${e.op} is not implemented`)}},Mj=(e,t,n,s=Rn)=>{switch(e.op){case"LowerBound":{let r=I("sortedSequence",e,t,n),a=I("values",e,t,n);return[s.lowerBound(r,a)]}case"TopKV2":{let r=I("x",e,t,n),a=I("k",e,t,n),o=I("sorted",e,t,n),i=s.topk(r,a,o);return[i.values,i.indices]}case"UpperBound":{let r=I("sortedSequence",e,t,n),a=I("values",e,t,n);return[s.upperBound(r,a)]}case"Unique":{let r=I("x",e,t,n),a=s.unique(r);return[a.values,a.indices]}case"UniqueV2":{let r=I("x",e,t,n),a=I("axis",e,t,n),o=s.unique(r,a);return[o.values,o.indices]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},zj=(e,t,n,s=Rn)=>{switch(e.op){case"Const":return t[e.name];case"PlaceholderWithDefault":let r=I("default",e,t,n);return[Qn(e.name,t,n)||r];case"Placeholder":return[Qn(e.name,t,n)];case"Identity":case"StopGradient":case"FakeQuantWithMinMaxVars":{let c=I("x",e,t,n);return[ca(c)]}case"IdentityN":return I("x",e,t,n).map(c=>ca(c));case"Snapshot":let a=I("x",e,t,n);return[ca(a)];case"Shape":return[s.tensor1d(I("x",e,t,n).shape,"int32")];case"ShapeN":return I("x",e,t,n).map(c=>s.tensor1d(c.shape));case"Size":return[s.scalar(I("x",e,t,n).size,"int32")];case"Rank":return[s.scalar(I("x",e,t,n).rank,"int32")];case"NoOp":return[s.scalar(1)];case"Print":let o=I("x",e,t,n),i=I("data",e,t,n),l=I("message",e,t,n),u=I("summarize",e,t,n);console.warn("The graph has a tf.print() operation,usually used for debugging, which slows down performance."),console.log(l);for(let c=0;c<i.length;c++)console.log(Array.prototype.slice.call(i[c].dataSync()).slice(0,u));return[o];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Lj=class{constructor(e,t){this.keyDType=e,this.valueDType=t,this.handle=Ce(0),this.tensorMap=new Map,bn(this.handle)}get id(){return this.handle.id}clearAndClose(){this.tensorMap.forEach(e=>e.dispose()),this.tensorMap.clear(),this.handle.dispose()}size(){return this.tensorMap.size}tensorSize(){return Ce(this.size(),"int32")}async import(e,t){this.checkKeyAndValueTensor(e,t);let n=await e.data();return this.tensorMap.forEach(s=>s.dispose()),this.tensorMap.clear(),Y(()=>{let s=En(t),r=n.length,a=s.length;v.assert(r===a,()=>`The number of elements doesn't match, keys has ${r} elements, the values has ${a} elements.`);for(let o=0;o<r;o++){let i=n[o],l=s[o];bn(l),this.tensorMap.set(i,l)}return this.handle})}async find(e,t){this.checkKeyAndValueTensor(e,t);let n=await e.data();return Y(()=>{let s=[];for(let r=0;r<n.length;r++){let a=n[r],o=this.findWithDefault(a,t);s.push(o)}return on(s)})}findWithDefault(e,t){let n=this.tensorMap.get(e);return n!=null?n:t}checkKeyAndValueTensor(e,t){if(e.dtype!==this.keyDType)throw new Error(`Expect key dtype ${this.keyDType}, but got ${e.dtype}`);if(t.dtype!==this.valueDType)throw new Error(`Expect value dtype ${this.valueDType}, but got ${t.dtype}`)}},Bj=async(e,t,n,s)=>{switch(e.op){case"HashTable":case"HashTableV2":{let r=I("keyDType",e,t,n),a=I("valueDType",e,t,n),o=new Lj(r,a);return s.addHashTable(e.name,o),[o.handle]}case"LookupTableImport":case"LookupTableImportV2":{let r=I("tableHandle",e,t,n,s),a=I("keys",e,t,n),o=I("values",e,t,n);return[await s.getHashTableById(r.id).import(a,o)]}case"LookupTableFind":case"LookupTableFindV2":{let r=I("tableHandle",e,t,n,s),a=I("keys",e,t,n),o=I("defaultValue",e,t,n);return[await s.getHashTableById(r.id).find(a,o)]}case"LookupTableSize":case"LookupTableSizeV2":{let r=I("tableHandle",e,t,n,s);return[s.getHashTableById(r.id).tensorSize()]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Wj=(e,t,n,s=Rn)=>{switch(e.op){case"ResizeBilinear":{let r=I("images",e,t,n),a=I("size",e,t,n),o=I("alignCorners",e,t,n),i=I("halfPixelCenters",e,t,n);return[s.image.resizeBilinear(r,[a[0],a[1]],o,i)]}case"ResizeNearestNeighbor":{let r=I("images",e,t,n),a=I("size",e,t,n),o=I("alignCorners",e,t,n),i=I("halfPixelCenters",e,t,n);return[s.image.resizeNearestNeighbor(r,[a[0],a[1]],o,i)]}case"CropAndResize":{let r=I("image",e,t,n),a=I("boxes",e,t,n),o=I("boxInd",e,t,n),i=I("cropSize",e,t,n),l=I("method",e,t,n),u=I("extrapolationValue",e,t,n);return[s.image.cropAndResize(r,a,o,i,l,u)]}case"ImageProjectiveTransformV3":{let r=I("images",e,t,n),a=I("transforms",e,t,n),o=I("outputShape",e,t,n),i=I("fillValue",e,t,n),l=I("interpolation",e,t,n),u=I("fillMode",e,t,n);return[s.image.transform(r,a,l.toLowerCase(),u.toLowerCase(),i,o)]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Vj=(e,t,n,s=Rn)=>{switch(e.op){case"Equal":return[s.equal(I("a",e,t,n),I("b",e,t,n))];case"NotEqual":return[s.notEqual(I("a",e,t,n),I("b",e,t,n))];case"Greater":return[s.greater(I("a",e,t,n),I("b",e,t,n))];case"GreaterEqual":return[s.greaterEqual(I("a",e,t,n),I("b",e,t,n))];case"Less":return[s.less(I("a",e,t,n),I("b",e,t,n))];case"LessEqual":return[s.lessEqual(I("a",e,t,n),I("b",e,t,n))];case"LogicalAnd":return[s.logicalAnd(I("a",e,t,n),I("b",e,t,n))];case"LogicalNot":return[s.logicalNot(I("a",e,t,n))];case"LogicalOr":return[s.logicalOr(I("a",e,t,n),I("b",e,t,n))];case"Select":case"SelectV2":return[s.where(I("condition",e,t,n),I("a",e,t,n),I("b",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Uj=(e,t,n,s=Rn)=>{switch(e.op){case"BatchMatMul":case"BatchMatMulV2":case"MatMul":return[s.matMul(I("a",e,t,n),I("b",e,t,n),I("transposeA",e,t,n),I("transposeB",e,t,n))];case"Einsum":return[s.einsum(I("equation",e,t,n),...I("tensors",e,t,n))];case"Transpose":return[s.transpose(I("x",e,t,n),I("perm",e,t,n))];case"_FusedMatMul":let[r,a]=I("fusedOps",e,t,n),o=r==="biasadd",i=a==="prelu",l=I("numArgs",e,t,n),u=I("leakyreluAlpha",e,t,n);if(o){if(i&&l!==2)throw new Error("Fused MatMul with BiasAdd and Prelu must have two extra arguments: bias and alpha.");if(!i&&l!==1)throw new Error("Fused MatMul with BiasAdd must have one extra argument: bias.")}let[c,p]=I("args",e,t,n);return[s.fused.matMul({a:I("a",e,t,n),b:I("b",e,t,n),transposeA:I("transposeA",e,t,n),transposeB:I("transposeB",e,t,n),bias:c,activation:a,preluActivationWeights:p,leakyreluAlpha:u})];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Gj=(e,t,n,s=Rn)=>{switch(e.op){case"EuclideanNorm":return[s.euclideanNorm(I("x",e,t,n),I("axis",e,t,n),I("keepDims",e,t,n))];case"FusedBatchNorm":case"FusedBatchNormV2":return[s.batchNorm(I("x",e,t,n),I("mean",e,t,n),I("variance",e,t,n),I("offset",e,t,n),I("scale",e,t,n),I("epsilon",e,t,n))];case"FusedBatchNormV3":return[s.batchNorm(I("x",e,t,n),I("mean",e,t,n),I("variance",e,t,n),I("offset",e,t,n),I("scale",e,t,n),I("epsilon",e,t,n))];case"LRN":return[s.localResponseNormalization(I("x",e,t,n),I("radius",e,t,n),I("bias",e,t,n),I("alpha",e,t,n),I("beta",e,t,n))];case"Softmax":return[s.softmax(I("x",e,t,n))];case"LogSoftmax":return[s.logSoftmax(I("x",e,t,n))];case"SparseToDense":return[s.sparseToDense(I("sparseIndices",e,t,n),I("outputShape",e,t,n),I("sparseValues",e,t,n),I("defaultValue",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Hj=(e,t,n,s=Rn)=>{switch(e.op){case"Max":{let i=I("axis",e,t,n),l=I("keepDims",e,t,n);return[s.max(I("x",e,t,n),i,l)]}case"Mean":{let i=I("axis",e,t,n),l=I("keepDims",e,t,n);return[s.mean(I("x",e,t,n),i,l)]}case"Min":{let i=I("axis",e,t,n),l=I("keepDims",e,t,n);return[s.min(I("x",e,t,n),i,l)]}case"Sum":{let i=I("axis",e,t,n),l=I("keepDims",e,t,n);return[s.sum(I("x",e,t,n),i,l)]}case"All":{let i=I("axis",e,t,n),l=I("keepDims",e,t,n);return[s.all(I("x",e,t,n),i,l)]}case"Any":{let i=I("axis",e,t,n),l=I("keepDims",e,t,n);return[s.any(I("x",e,t,n),i,l)]}case"ArgMax":{let i=I("axis",e,t,n);return[s.argMax(I("x",e,t,n),i)]}case"ArgMin":{let i=I("axis",e,t,n);return[s.argMin(I("x",e,t,n),i)]}case"Prod":{let i=I("axis",e,t,n),l=I("keepDims",e,t,n);return[s.prod(I("x",e,t,n),i,l)]}case"Cumprod":{let i=I("axis",e,t,n),l=I("exclusive",e,t,n),u=I("reverse",e,t,n);return[s.cumprod(I("x",e,t,n),i,l,u)]}case"Cumsum":{let i=I("axis",e,t,n),l=I("exclusive",e,t,n),u=I("reverse",e,t,n);return[s.cumsum(I("x",e,t,n),i,l,u)]}case"Bincount":let r=I("x",e,t,n),a=I("weights",e,t,n),o=I("size",e,t,n);return[s.bincount(r,a,o)];case"DenseBincount":{let i=I("x",e,t,n),l=I("weights",e,t,n),u=I("size",e,t,n),c=I("binaryOutput",e,t,n);return[s.denseBincount(i,l,u,c)]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},jj=(e,t,n,s=Rn)=>{switch(e.op){case"ConcatV2":case"Concat":{let r=I("n",e,t,n),a=I("axis",e,t,n),o=I("tensors",e,t,n);return o=o.slice(0,r),[s.concat(o,a)]}case"Gather":{let r=I("x",e,t,n),a=I("indices",e,t,n);return[s.gather(r,s.cast(a,"int32"),0)]}case"GatherV2":{let r=I("axis",e,t,n),a=I("batchDims",e,t,n),o=I("x",e,t,n),i=I("indices",e,t,n);return[s.gather(o,s.cast(i,"int32"),r,a)]}case"Reverse":{let r=I("dims",e,t,n),a=[];for(let i=0;i<r.length;i++)r[i]&&a.push(i);let o=I("x",e,t,n);return[s.reverse(o,a)]}case"ReverseV2":{let r=I("axis",e,t,n),a=I("x",e,t,n);return[s.reverse(a,r)]}case"Slice":{let r=I("begin",e,t,n),a=I("size",e,t,n);return[s.slice(I("x",e,t,n),r,a)]}case"StridedSlice":{let r=I("begin",e,t,n),a=I("end",e,t,n),o=I("strides",e,t,n),i=I("beginMask",e,t,n),l=I("endMask",e,t,n),u=I("ellipsisMask",e,t,n),c=I("newAxisMask",e,t,n),p=I("shrinkAxisMask",e,t,n),d=I("x",e,t,n);return[s.stridedSlice(d,r,a,o,i,l,u,c,p)]}case"Pack":return Y(()=>{let r=I("axis",e,t,n),a=I("tensors",e,t,n),o=a[0].shape,i=s.squeeze(a[0]).shape,l=a.map(u=>{let c=v.arraysEqual(u.shape,o);if(!c&&!v.arraysEqual(s.squeeze(u).shape,i))throw new Error("the input tensors shape does not match");return c?u:s.reshape(u,o)});return[s.stack(l,r)]});case"Unpack":{let r=I("axis",e,t,n),a=I("tensor",e,t,n);return s.unstack(a,r)}case"Tile":{let r=I("reps",e,t,n);return[s.tile(I("x",e,t,n),r)]}case"Split":case"SplitV":{let r=I("axis",e,t,n),a=I("numOrSizeSplits",e,t,n),o=I("x",e,t,n);return s.split(o,a,r)}case"ScatterNd":{let r=I("indices",e,t,n),a=I("values",e,t,n),o=I("shape",e,t,n);return[s.scatterND(r,a,o)]}case"GatherNd":{let r=I("x",e,t,n),a=I("indices",e,t,n);return[s.gatherND(r,a)]}case"SparseToDense":{let r=I("sparseIndices",e,t,n),a=I("outputShape",e,t,n),o=I("sparseValues",e,t,n),i=I("defaultValue",e,t,n);return[s.sparseToDense(r,o,a,o.dtype===i.dtype?i:s.cast(i,o.dtype))]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},qj=(e,t,n,s=Rn)=>{switch(e.op){case"SparseFillEmptyRows":{let{outputIndices:r,outputValues:a,emptyRowIndicator:o,reverseIndexMap:i}=s.sparse.sparseFillEmptyRows(I("indices",e,t,n),I("values",e,t,n),I("denseShape",e,t,n),I("defaultValue",e,t,n));return[r,a,o,i]}case"SparseReshape":{let{outputIndices:r,outputShape:a}=s.sparse.sparseReshape(I("inputIndices",e,t,n),I("inputShape",e,t,n),I("newShape",e,t,n));return[r,a]}case"SparseSegmentMean":return[s.sparse.sparseSegmentMean(I("data",e,t,n),I("indices",e,t,n),I("segmentIds",e,t,n))];case"SparseSegmentSum":return[s.sparse.sparseSegmentSum(I("data",e,t,n),I("indices",e,t,n),I("segmentIds",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Xj=(e,t,n,s=Rn)=>{switch(e.op){case"FFT":return[s.fft(I("x",e,t,n))];case"IFFT":return[s.ifft(I("x",e,t,n))];case"RFFT":return[s.rfft(I("x",e,t,n))];case"IRFFT":return[s.irfft(I("x",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Kj=(e,t,n,s=Rn)=>{switch(e.op){case"StringNGrams":{let{nGrams:r,nGramsSplits:a}=s.string.stringNGrams(I("data",e,t,n),I("dataSplits",e,t,n),I("separator",e,t,n),I("nGramWidths",e,t,n),I("leftPad",e,t,n),I("rightPad",e,t,n),I("padWidth",e,t,n),I("preserveShortSequences",e,t,n));return[r,a]}case"StringSplit":{let{indices:r,values:a,shape:o}=s.string.stringSplit(I("input",e,t,n),I("delimiter",e,t,n),I("skipEmpty",e,t,n));return[r,a,o]}case"StringToHashBucketFast":return[s.string.stringToHashBucketFast(I("input",e,t,n),I("numBuckets",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Zj=(e,t,n,s=Rn)=>{switch(e.op){case"Cast":return[s.cast(I("x",e,t,n),I("dtype",e,t,n))];case"ExpandDims":{let r=I("axis",e,t,n);return[s.expandDims(I("x",e,t,n),r)]}case"Squeeze":{let r=I("axis",e,t,n);return[s.squeeze(I("x",e,t,n),r)]}case"Reshape":return[s.reshape(I("x",e,t,n),I("shape",e,t,n))];case"MirrorPad":return[s.mirrorPad(I("x",e,t,n),I("padding",e,t,n),I("mode",e,t,n))];case"PadV2":case"Pad":return[s.pad(I("x",e,t,n),I("padding",e,t,n),I("constantValue",e,t,n))];case"SpaceToBatchND":{let r=I("blockShape",e,t,n),a=I("paddings",e,t,n);return[s.spaceToBatchND(I("x",e,t,n),r,a)]}case"BatchToSpaceND":{let r=I("blockShape",e,t,n),a=I("crops",e,t,n);return[s.batchToSpaceND(I("x",e,t,n),r,a)]}case"DepthToSpace":{let r=I("blockSize",e,t,n),a=I("dataFormat",e,t,n).toUpperCase();return[s.depthToSpace(I("x",e,t,n),r,a)]}case"BroadcastTo":return[s.broadcastTo(I("x",e,t,n),I("shape",e,t,n))];case"BroadcastArgs":return[s.broadcastArgs(I("s0",e,t,n),I("s1",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}};function o7(e,t,n,s,r=Y){let a=((o,i,l)=>{switch(o.category){case"arithmetic":return r(()=>Cj(o,i,l));case"basic_math":return r(()=>Tj(o,i,l));case"control":return $j(o,i,l);case"convolution":return r(()=>Pj(o,i,l));case"creation":return r(()=>Fj(o,i,l));case"dynamic":return Oj(o,i,l);case"evaluation":return r(()=>Mj(o,i,l));case"image":return r(()=>Wj(o,i,l));case"graph":return r(()=>zj(o,i,l));case"logical":return r(()=>Vj(o,i,l));case"matrices":return r(()=>Uj(o,i,l));case"normalization":return r(()=>Gj(o,i,l));case"reduction":return r(()=>Hj(o,i,l));case"slice_join":return r(()=>jj(o,i,l));case"sparse":return r(()=>qj(o,i,l));case"spectral":return r(()=>Xj(o,i,l));case"string":return r(()=>Kj(o,i,l));case"transformation":return r(()=>Zj(o,i,l));case"hash_table":return Bj(o,i,l,s);case"custom":let u=Nk(o.op);if(u&&u.customExecutor)return u.customExecutor(new Sj(o,i,l));throw TypeError(`Custom op ${o.op} is not registered.`);default:throw TypeError(`Unknown op '${o.op}'. File an issue at https://github.com/tensorflow/tfjs/issues so we can add it, or register a custom execution with tf.registerOp()`)}})(e,t,n);return v.isPromise(a)?a.then(o=>[].concat(o)):[].concat(a)}var i7=class{constructor(e={},t={},n={},s={}){this.weightMap=e,this.tensorArrayMap=t,this.tensorListMap=n,this.functionMap=s,this.rootContext={id:0,frameName:"",iterationId:0},this.contexts=[this.rootContext],this.lastId=0,this.generateCurrentContextIds()}newFrame(e,t){return{id:e,frameName:t,iterationId:0}}set currentContext(e){this.contexts!==e&&(this.contexts=e,this.generateCurrentContextIds())}get currentContext(){return this.contexts}get currentContextId(){return this._currentContextIds[0]}get currentContextIds(){return this._currentContextIds}generateCurrentContextIds(){let e=[];for(let t=0;t<this.contexts.length-1;t++){let n=this.contexts.slice(0,this.contexts.length-t);e.push(this.contextIdforContexts(n))}e.push(""),this._currentContextIds=e}contextIdforContexts(e){return e?e.map(t=>t.id===0&&t.iterationId===0?"":`${t.frameName}-${t.iterationId}`).join("/"):""}enterFrame(e){this.contexts&&(this.lastId++,this.contexts=this.contexts.slice(),this.contexts.push(this.newFrame(this.lastId,e)),this._currentContextIds.unshift(this.contextIdforContexts(this.contexts)))}exitFrame(){if(this.contexts&&this.contexts.length>1)this.contexts=this.contexts.slice(),this.contexts.splice(-1),this.currentContextIds.shift();else throw new Error("Cannot exit frame, the context is empty")}nextIteration(){if(this.contexts&&this.contexts.length>0){this.contexts=this.contexts.slice(),this.lastId++;let e=Object.assign({},this.contexts[this.contexts.length-1]);e.iterationId+=1,e.id=this.lastId,this.contexts.splice(-1,1,e),this._currentContextIds.splice(0,1,this.contextIdforContexts(this.contexts))}else throw new Error("Cannot increase frame iteration, the context is empty")}getWeight(e){return this.weightMap[e]}addTensorArray(e){this.tensorArrayMap[e.id]=e}getTensorArray(e){return this.tensorArrayMap[e]}addTensorList(e){this.tensorListMap[e.id]=e}getTensorList(e){return this.tensorListMap[e]}dispose(e){for(let t in this.tensorArrayMap)this.tensorArrayMap[t].clearAndClose(e);for(let t in this.tensorListMap)this.tensorListMap[t].clearAndClose(e)}};function l7(e,t,n,s){let r=new Set,a=[],o=null,i=null,l=new Set,u=Object.keys(e).map(d=>Ss(d)[0]),c=[];s!=null&&(c=s.map(d=>Ss(d.name)[0]));let p=[...t];for(;p.length>0;){let d=p.pop();if((Zk(d)||tq(d)||nq(d))&&o==null&&(o=d,i=o.children.map(h=>h.name).filter(h=>r.has(h))),r.add(d.name),n[d.name]==null&&u.indexOf(d.name)===-1&&c.indexOf(d.name)===-1){if(d.inputs.length===0){a.push(d.name);continue}d.inputs.forEach(h=>{l.has(h.name)||(l.add(h.name),p.push(h))})}}return{inputs:e,outputs:t,usedNodes:r,missingInputs:a,dynamicNode:o,syncInputs:i}}function Yj(e,t,n){let{usedNodes:s,inputs:r}=n,a=[],o=Object.keys(r).map(c=>Ss(c)[0]).map(c=>e.nodes[c]),i=e.initNodes;o.forEach(c=>{s.has(c.name)&&a.push(c)}),e.weights.forEach(c=>{s.has(c.name)&&a.push(c)}),i!=null&&i.forEach(c=>{s.has(c.name)&&a.push(c)});let l=new Set,u=[];for(;a.length>0;){let c=a.pop();l.add(c.name),t[c.name]||u.push(c),c.children.forEach(p=>{!l.has(p.name)&&s.has(p.name)&&p.inputs.every(d=>l.has(d.name))&&a.push(p)})}return u}var Jj=["Switch","Merge","Enter","Exit","NextIteration","StatelessIf","StatelessWhile","if","While"],Qj=["NonMaxSuppressionV2","NonMaxSuppressionV3","NonMaxSuppressionV5","Where"],eq=["HashTable","HashTableV2","LookupTableImport","LookupTableImportV2","LookupTableFind","LookupTableFindV2","LookupTableSize","LookupTableSizeV2"];function Zk(e){return Jj.indexOf(e.op)>=0}function tq(e){return Qj.indexOf(e.op)>=0}function nq(e){return eq.indexOf(e.op)>=0}var Q3=class{constructor(e,t){this.graph=e,this.parent=t,this.compiledMap=new Map,this._weightMap={},this.SEPERATOR=",",this._functions={},this._functionExecutorMap={},this.intermediateTensors={},this.keepTensorForDebug=!1,this._outputs=e.outputs,this._inputs=e.inputs,this._initNodes=e.initNodes,this._signature=e.signature,this._functions=e.functions,e.functions!=null&&Object.keys(e.functions).forEach(n=>{this._functionExecutorMap[n]=new Q3(e.functions[n],this)})}get weightIds(){return this.parent?this.parent.weightIds:this._weightIds}get functionExecutorMap(){return this.parent?this.parent.functionExecutorMap:this._functionExecutorMap}get weightMap(){return this.parent?this.parent.weightMap:this._weightMap}set weightMap(e){let t=Object.keys(e).map(n=>e[n].map(s=>s.id));this._weightIds=[].concat(...t),this._weightMap=e}set resourceManager(e){this._resourceManager=e}get inputs(){return this._inputs.map(e=>({name:e.name,shape:e.attrParams.shape?e.attrParams.shape.value:void 0,dtype:e.attrParams.dtype?e.attrParams.dtype.value:void 0}))}get outputs(){return this._outputs.map(e=>({name:e.name,shape:e.attrParams.shape?e.attrParams.shape.value:void 0,dtype:e.attrParams.dtype?e.attrParams.dtype.value:void 0}))}get inputNodes(){return this._inputs.map(e=>e.signatureKey||e.name)}get outputNodes(){return this._outputs.map(e=>{let t=e.signatureKey||e.name;return e.defaultOutput?`${t}:${e.defaultOutput}`:t})}get functions(){return Object.keys(this._functions).reduce((e,t)=>(e[t]=this._functions[t].signature,e),{})}getCompilationKey(e,t){let n=e.map(r=>r.name).sort(),s=t.map(r=>r.name).sort();return n.join(this.SEPERATOR)+"--"+s.join(this.SEPERATOR)}compile(e,t){let n=l7(e,t,this.weightMap,this._initNodes),{missingInputs:s,dynamicNode:r,syncInputs:a}=n;if(r!=null)throw new Error(`This execution contains the node '${r.name}', which has the dynamic op '${r.op}'. Please use model.executeAsync() instead. Alternatively, to avoid the dynamic ops, specify the inputs [${a}]`);if(s.length>0){let o=t.map(l=>l.name),i=Object.keys(e);throw new Error(`Cannot compute the outputs [${o}] from the provided inputs [${i}]. Missing the following inputs: [${s}]`)}return Yj(this.graph,this.weightMap,n)}execute(e,t){e=this.mapInputs(e);let n=Object.keys(e).sort();this.checkInputs(e),this.checkInputShapeAndType(e),t=this.mapOutputs(t),this.checkOutputs(t);let s=n.map(c=>this.graph.nodes[Ss(c)[0]]),r=t.map(c=>Ss(c)[0]),a=r.map(c=>this.graph.nodes[c]);this.resetIntermediateTensors(),a.length===0&&(a=this._outputs);let o=this.getCompilationKey(s,a),i=this.compiledMap.get(o);i==null&&(i=this.compile(e,a),this.compiledMap.set(o,i));let l={},u={};return Y(()=>{let c=new i7(this.weightMap,l,u,this.functionExecutorMap),p=Object.assign({},this.weightMap);Object.keys(e).forEach(f=>{let[m,g]=Ss(f),y=[];y[g]=e[f],p[m]=y});let d=this.getFrozenTensorIds(p),h={};for(let f=0;f<i.length;f++){let m=i[f];if(!p[m.name]){let g=o7(m,p,c,this._resourceManager);if(v.isPromise(g))throw new Error(`The execution of the op '${m.op}' returned a promise. Please use model.executeAsync() instead.`);p[m.name]=g,this.checkTensorForDisposal(m.name,m,p,c,d,r,h)}}return this.parent==null&&c.dispose(d),t.map(f=>Qn(f,p,c))})}getFrozenTensorIds(e){let t=[].concat.apply([],Object.keys(e).map(n=>e[n]).map(n=>n.map(s=>s.id)));return new Set(t)}checkTensorForDisposal(e,t,n,s,r,a,o){t.category==="control"||a.indexOf(e)!==-1||(n[e].forEach(i=>{i!=null&&(o[i.id]=(o[i.id]||0)+t.children.length)}),t.inputs.forEach(i=>{if(i.category!=="control"){let l=rj(i.name,n,s);l!=null&&l.forEach(u=>{if(u&&!u.kept&&!r.has(u.id)){let c=o[u.id];if(c===1){if(!this.keepTensorForDebug)u.dispose();else{let[p,d]=Gr(t.name,s);this.intermediateTensors[p]?this.intermediateTensors[p][d]=u:(this.intermediateTensors[p]=[],this.intermediateTensors[p][d]=u)}delete o[u.id]}else c!=null&&o[u.id]--}})}}))}async executeAsync(e,t){return this._executeAsync(e,t)}disposeIntermediateTensors(){!this.intermediateTensors||(Object.keys(this.intermediateTensors).forEach(e=>this.intermediateTensors[e].forEach(t=>t.dispose())),this.disposeTensorsMap())}disposeTensorsMap(){!this.tensorsMap||Object.keys(this.tensorsMap).forEach(e=>{this.tensorsMap[e].forEach(n=>{n&&!n.kept&&!n.isDisposed&&!this.keepIds.has(n.id)&&n.dispose()})})}getIntermediateTensors(){return this.tensorsMap}resetIntermediateTensors(){for(let e in this.intermediateTensors)this.intermediateTensors[e].forEach(t=>t.dispose()),delete this.intermediateTensors[e]}async _executeAsync(e,t,n=!1,s={},r={}){n||(e=this.mapInputs(e),this.checkInputs(e),this.checkInputShapeAndType(e),t=this.mapOutputs(t),this.checkOutputs(t));try{this.keepTensorForDebug=q().getBool("KEEP_INTERMEDIATE_TENSORS")}catch(u){console.warn(u.message)}this.resetIntermediateTensors();let a=new i7(this.weightMap,s,r,this.functionExecutorMap);this.tensorsMap=await this.executeWithControlFlow(e,a,t,n);let o=t.map(u=>Qn(u,this.tensorsMap,a)),i=o.map(u=>u.id),l=Object.keys(e).map(u=>e[u].id);return this.keepIds=new Set([...i,...l,...this.weightIds]),this.keepTensorForDebug||this.disposeTensorsMap(),this.parent==null&&a.dispose(this.keepIds),o}async executeFunctionAsync(e,t,n){let s=e.reduce((r,a,o)=>(r[this.inputs[o].name]=a,r),{});return this._executeAsync(s,this.outputNodes,!0,t,n)}async executeWithControlFlow(e,t,n,s){let r=Object.keys(e),a=r.map(x=>this.graph.nodes[Ss(x)[0]]),o=n.map(x=>Ss(x)[0]),i=o.map(x=>this.graph.nodes[x]);i.length===0&&(i=this._outputs);let{usedNodes:l,missingInputs:u,dynamicNode:c,syncInputs:p}=l7(e,i,this.weightMap,this._initNodes),d=[...a,...this.graph.weights,...this._initNodes||[]].map(x=>({node:x,contexts:t.currentContext})),h=Object.assign({},this.weightMap);Object.keys(e).forEach(x=>{let[A,b]=Ss(x),w=[];w[b]=e[x],h[A]=w});let f={},m=this.getFrozenTensorIds(h),g={};for(;d.length>0;){let x=this.processStack(a,d,t,h,g,m,o,f,l);await Promise.all(x)}c==null&&!s&&console.warn("This model execution did not contain any nodes with control flow or dynamic output shapes. You can use model.execute() instead.");let y=i.filter(x=>!Zk(x)&&!Qn(x.name,h,t)).map(x=>x.name);if(y.length>0){let x="";throw c!=null&&(x=`Alternatively, to avoid the dynamic ops, use model.execute() and specify the inputs [${p}]`),new Error(`Cannot compute the outputs [${y}] from the provided inputs [${r}]. Consider providing the following inputs: [${u}]. ${x}`)}return h}processStack(e,t,n,s,r,a,o,i,l){let u=[];for(;t.length>0;){let c=t.pop();n.currentContext=c.contexts;let p="";if(c.node.op==="Enter"&&I("isConstant",c.node,s,n)&&([p]=Gr(c.node.name,n)),s[c.node.name]==null){let d=o7(c.node,s,n,this._resourceManager);p||([p]=Gr(c.node.name,n));let h=n.currentContext;v.isPromise(d)?u.push(d.then(f=>(s[p]=f,n.currentContext=h,this.checkTensorForDisposal(p,c.node,s,n,a,o,i),this.processChildNodes(c.node,t,n,s,r,l),f))):(s[p]=d,this.checkTensorForDisposal(p,c.node,s,n,a,o,i),this.processChildNodes(c.node,t,n,s,r,l))}else this.processChildNodes(c.node,t,n,s,r,l)}return u}processChildNodes(e,t,n,s,r,a){e.children.forEach(o=>{let[i]=Gr(o.name,n);r[i]||!a.has(o.name)||(o.op==="Merge"?o.inputNames.some(l=>!!Qn(l,s,n))&&(r[i]=!0,t.push({contexts:n.currentContext,node:o})):o.inputNames.every(l=>!!Qn(l,s,n))&&(r[i]=!0,t.push({contexts:n.currentContext,node:o})))})}dispose(){Object.keys(this.weightMap).forEach(e=>this.weightMap[e].forEach(t=>t.dispose()))}checkInputShapeAndType(e){Object.keys(e).forEach(t=>{let n=e[t],[s]=Ss(t),r=this.graph.nodes[s];if(r.attrParams.shape&&r.attrParams.shape.value){let a=r.attrParams.shape.value,o=a.length===n.shape.length&&n.shape.every((i,l)=>a[l]===-1||a[l]===i);v.assert(o,()=>`The shape of dict['${r.name}'] provided in model.execute(dict) must be [${a}], but was [${n.shape}]`)}r.attrParams.dtype&&r.attrParams.dtype.value&&v.assert(n.dtype===r.attrParams.dtype.value,()=>`The dtype of dict['${r.name}'] provided in model.execute(dict) must be ${r.attrParams.dtype.value}, but was ${n.dtype}`)})}mapInputs(e){let t={};for(let n in e)if(this._signature!=null&&this._signature.inputs!=null&&this._signature.inputs[n]!=null){let s=this._signature.inputs[n];t[s.name]=e[n]}else t[n]=e[n];return t}checkInputs(e){let t=Object.keys(e).filter(n=>{let[s]=Ss(n);return this.graph.nodes[s]==null});if(t.length>0)throw new Error(`The dict provided in model.execute(dict) has keys: [${t}] that are not part of graph`)}mapOutputs(e){return e.map(t=>this._signature!=null&&this._signature.outputs!=null&&this._signature.outputs[t]!=null?this._signature.outputs[t].name:t,{})}checkOutputs(e){e.forEach(t=>{let[n]=Ss(t);if(!this.graph.nodes[n])throw new Error(`The output '${t}' is not found in the graph`)})}},sq=class{constructor(e={},t={}){this.hashTableNameToHandle=e,this.hashTableMap=t}addHashTable(e,t){this.hashTableNameToHandle[e]=t.handle,this.hashTableMap[t.id]=t}getHashTableHandleByName(e){return this.hashTableNameToHandle[e]}getHashTableById(e){return this.hashTableMap[e]}dispose(){for(let e in this.hashTableMap)this.hashTableMap[e].clearAndClose(),delete this.hashTableMap[e];for(let e in this.hashTableNameToHandle)this.hashTableNameToHandle[e].dispose(),delete this.hashTableNameToHandle[e]}},rq="?tfjs-format=file",aq="model.json",$h=class{constructor(e,t={},n=Ts){this.modelUrl=e,this.loadOptions=t,this.version="n/a",this.io=n,t==null&&(this.loadOptions={}),this.resourceManager=new sq}get modelVersion(){return this.version}get inputNodes(){return this.executor.inputNodes}get outputNodes(){return this.executor.outputNodes}get inputs(){return this.executor.inputs}get outputs(){return this.executor.outputs}get weights(){return this.executor.weightMap}get metadata(){return this.artifacts.userDefinedMetadata}get modelSignature(){return this.signature}get modelStructuredOutputKeys(){return this.structuredOutputKeys}findIOHandler(){let e=this.modelUrl;if(e.load!=null)this.handler=e;else if(this.loadOptions.requestInit!=null)this.handler=this.io.browserHTTPRequest(e,this.loadOptions);else{let t=this.io.getLoadHandlers(e,this.loadOptions);if(t.length===0)t.push(this.io.browserHTTPRequest(e,this.loadOptions));else if(t.length>1)throw new Error(`Found more than one (${t.length}) load handlers for URL '${[e]}'`);this.handler=t[0]}}load(){if(this.findIOHandler(),this.handler.load==null)throw new Error("Cannot proceed with model loading because the IOHandler provided does not have the `load` method implemented.");let e=this.handler.load();return v.isPromise(e)?e.then(t=>this.loadSync(t)):this.loadSync(e)}loadSync(e){this.artifacts=e;let t=this.artifacts.modelTopology,n=this.artifacts.signature;if(this.artifacts.userDefinedMetadata!=null){let r=this.artifacts.userDefinedMetadata;r.signature!=null&&(n=r.signature),r.structuredOutputKeys!=null&&(this.structuredOutputKeys=r.structuredOutputKeys)}this.signature=n,this.version=`${t.versions.producer}.${t.versions.minConsumer}`;let s=this.io.decodeWeights(this.artifacts.weightData,this.artifacts.weightSpecs);if(this.executor=new Q3(n7.Instance.transformGraph(t,this.signature)),this.executor.weightMap=this.convertTensorMapToTensorsMap(s),this.executor.resourceManager=this.resourceManager,e.modelInitializer!=null&&e.modelInitializer.node!=null){let r=n7.Instance.transformGraph(e.modelInitializer);this.initializer=new Q3(r),this.initializer.weightMap=this.executor.weightMap,this.initializer.resourceManager=this.resourceManager,this.initializer.executeAsync({},[])}return!0}async save(e,t){if(typeof e=="string"){let n=this.io.getSaveHandlers(e);if(n.length===0)throw new Error(`Cannot find any save handlers for URL '${e}'`);if(n.length>1)throw new Error(`Found more than one (${n.length}) save handlers for URL '${e}'`);e=n[0]}if(e.save==null)throw new Error("GraphModel.save() cannot proceed because the IOHandler provided does not have the `save` attribute defined.");return e.save(this.artifacts)}predict(e,t){let n=this.execute(e,this.outputNodes);if(this.structuredOutputKeys){let s=n instanceof nt?[n]:n,r={};return s.forEach((a,o)=>r[this.structuredOutputKeys[o]]=a),r}return n}normalizeInputs(e){if(!(e instanceof nt)&&!Array.isArray(e))return e;if(e=Array.isArray(e)?e:[e],e.length!==this.inputNodes.length)throw new Error(`Input tensor count mismatch,the graph model has ${this.inputNodes.length} placeholders, while there are ${e.length} input tensors.`);return this.inputNodes.reduce((t,n,s)=>(t[n]=e[s],t),{})}normalizeOutputs(e){return e=e||this.outputNodes,Array.isArray(e)?e:[e]}execute(e,t){e=this.normalizeInputs(e),t=this.normalizeOutputs(t);let n=this.executor.execute(e,t);return n.length>1?n:n[0]}async executeAsync(e,t){e=this.normalizeInputs(e),t=this.normalizeOutputs(t);let n=await this.executor.executeAsync(e,t);return n.length>1?n:n[0]}getIntermediateTensors(){return this.executor.getIntermediateTensors()}disposeIntermediateTensors(){this.executor.disposeIntermediateTensors()}convertTensorMapToTensorsMap(e){return Object.keys(e).reduce((t,n)=>(t[n]=[e[n]],t),{})}dispose(){this.executor.dispose(),this.initializer&&this.initializer.dispose(),this.resourceManager.dispose()}};async function oq(e,t={},n=Ts){if(e==null)throw new Error("modelUrl in loadGraphModel() cannot be null. Please provide a url or an IOHandler that loads the model");t==null&&(t={}),t.fromTFHub&&typeof e=="string"&&(e=lq(e));let s=new $h(e,t,n);return await s.load(),s}function iq(e){if(e==null)throw new Error("modelUrl in loadGraphModelSync() cannot be null. Please provide a url or an IOHandler that loads the model");if(!e.load)throw new Error(`modelUrl IO Handler ${e} has no load function`);let t=new $h(e);return t.load(),t}function lq(e){return e.endsWith("/")||(e=e+"/"),`${e}${aq}${rq}`}var uq="3.19.0",Yk={};Ve(Yk,{CSVDataset:()=>oI,Dataset:()=>Qc,FileDataSource:()=>hI,TextLineDataset:()=>aI,URLDataSource:()=>fI,array:()=>Dq,csv:()=>Uq,func:()=>Gq,generator:()=>Hq,microphone:()=>qq,version_data:()=>Xq,webcam:()=>jq,zip:()=>$q});var cq=uo(Wm()),dq=uo(Wm());function pq(e,t){return Em(e,t)}function Em(e,t,n=new Map,s=new Set){if(e==null)return null;if(typeof Blob=="function"&&e instanceof Blob)return e.slice();if(s.has(e))throw new Error("Circular references are not supported.");if(n.has(e))return n.get(e);let r=t(e);if(r.recurse&&r.value!==null)throw new Error("A deep map function may not return both a value and recurse=true.");if(r.recurse)if(ac(e)){let a=Array.isArray(e)?[]:{};s.add(e);for(let o in e){let i=e[o],l=Em(i,t,n,s);a[o]=l}return s.delete(e),e.__proto__&&(a.__proto__=e.__proto__),a}else throw new Error(`Can't recurse into non-iterable type: ${e}`);else return n.set(e,r.value),r.value}function hq(e,t=Qk){return Jk(e,t)}function Jk(e,t,n=new Set){let s=e[0];if(n.has(s))throw new Error("Circular references are not supported.");let r=t(e);if(r.recurse&&r.value!==null)throw new Error("A deep zip function may not return both a value and recurse=true.");if(r.recurse)if(ac(s)){let a=Array.isArray(s)?[]:{};n.add(s);for(let o in s){let i=e.map(u=>u[o]),l=Jk(i,t,n);a[o]=l}return n.delete(s),a}else throw new Error(`Can't recurse into non-iterable type: ${s}`);else return r.value}function Qk(e){return e===null?null:ac(e[0])?{value:null,recurse:!0}:{value:e,recurse:!1}}async function eI(e,t){let n=new Map;Em(e,t,n);for(let r of Array.from(n.keys())){let a=n.get(r);if(v.isPromise(a)){let o=await a;n.set(r,o)}}return Em(e,t,n)}function ac(e){let t=!1;if(q().get("IS_BROWSER"))t=e instanceof TextDecoder;else{let{StringDecoder:n}=e6();t=e instanceof n}return e!=null&&!ArrayBuffer.isView(e)&&(Array.isArray(e)||typeof e=="object"&&!(e instanceof nt)&&!(e instanceof Promise)&&!t)}function fq(e){return e==null||mq(e)||Array.isArray(e)||typeof e=="object"&&e instanceof nt||v.isTypedArray(e)}function mq(e){return e===null||typeof e!="object"&&typeof e!="function"}function gq(e){return pq(e,yq)}function yq(e){return e instanceof nt?{value:e.clone(),recurse:!1}:ac(e)?{value:null,recurse:!0}:{value:e,recurse:!1}}var tI=class{constructor(e){if(this.capacity=e,this.begin=0,this.end=0,e==null)throw new RangeError("Can't create a ring buffer of unknown capacity.");if(e<1)throw new RangeError("Can't create ring buffer of capacity < 1.");this.data=new Array(e),this.doubledCapacity=2*e}wrap(e){for(;e<0;)e+=this.doubledCapacity;return e%this.doubledCapacity}get(e){if(e<0)throw new RangeError("Can't get item at a negative index.");return this.data[e%this.capacity]}set(e,t){if(e<0)throw new RangeError("Can't set item at a negative index.");this.data[e%this.capacity]=t}length(){let e=this.end-this.begin;return e<0&&(e=this.doubledCapacity+e),e}isFull(){return this.length()===this.capacity}isEmpty(){return this.length()===0}push(e){if(this.isFull())throw new RangeError("Ring buffer is full.");this.set(this.end,e),this.end=this.wrap(this.end+1)}pushAll(e){for(let t of e)this.push(t)}pop(){if(this.isEmpty())throw new RangeError("Ring buffer is empty.");this.end=this.wrap(this.end-1);let e=this.get(this.end);return this.set(this.end,void 0),e}unshift(e){if(this.isFull())throw new RangeError("Ring buffer is full.");this.begin=this.wrap(this.begin-1),this.set(this.begin,e)}shift(){if(this.isEmpty())throw new RangeError("Ring buffer is empty.");let e=this.get(this.begin);return this.set(this.begin,void 0),this.begin=this.wrap(this.begin+1),e}shuffleExcise(e){if(this.isEmpty())throw new RangeError("Ring buffer is empty.");let t=this.wrap(this.begin+e),n=this.get(t);return this.set(t,this.pop()),n}},wx=class extends tI{constructor(){super(wx.INITIAL_CAPACITY)}isFull(){return!1}push(e){super.isFull()&&this.expand(),super.push(e)}unshift(e){super.isFull()&&this.expand(),super.unshift(e)}expand(){let e=this.capacity*2,t=new Array(e),n=this.length();for(let s=0;s<n;s++)t[s]=this.get(this.wrap(this.begin+s));this.data=t,this.capacity=e,this.doubledCapacity=2*this.capacity,this.begin=0,this.end=n}};wx.INITIAL_CAPACITY=32;function nI(e){return new bq(e)}function kx(e){return new vq(e)}function Aq(e,t){return new sI(e,t)}function xq(e,t=ja.FAIL){return new Rq(e,t)}var wn=class{async toArray(){let e=[],t=await this.next();for(;!t.done;)e.push(t.value),t=await this.next();return e}async toArrayForTest(){let e=this.prefetch(100),t=[],n=await e.next();for(;!n.done;)t.push(n.value),n=await e.next();return t}async resolveFully(){let e=await this.next();for(;!e.done;)e=await this.next()}async resolveWhile(e){let t=await this.next(),n=e(t.value);for(;!t.done&&n;)t=await this.next(),n=e(t.value)}handleErrors(e){return new Nq(this,e)}filter(e){return new Cq(this,e)}map(e){return new Tq(this,e)}mapAsync(e){return new u7(this,e)}serialMapAsync(e){return new u7(this,e).serial()}flatmap(e){return new Eq(this,e)}async forEachAsync(e){return this.map(e).resolveFully()}async serialForEach(e){return this.serialMapAsync(e).resolveWhile(t=>t===!0)}rowMajorBatch(e,t=!0){return new Sq(this,e,t)}columnMajorBatch(e,t=!0,n=Qk){return this.rowMajorBatch(e,t).map(r=>hq(r,n))}concatenate(e,t){return new sI(nI([this,e]),t)}take(e){return e<0||e==null?this:new Iq(this,e)}skip(e){return e<0||e==null?this:new kq(this,e)}prefetch(e){return new rI(this,e)}shuffle(e,t){return new _q(this,e,t)}serial(){return new wq(this)}},bq=class extends wn{constructor(e){super(),this.items=e,this.trav=0}summary(){return`Array of ${this.items.length} items`}async next(){if(this.trav>=this.items.length)return{value:null,done:!0};let e=this.items[this.trav];return this.trav++,{value:gq(e),done:!1}}},vq=class extends wn{constructor(e){super(),this.nextFn=e}summary(){return"Function call"}async next(){try{return this.nextFn()}catch(e){throw e.message=`Error thrown while iterating through a dataset: ${e.message}`,e}}},wq=class extends wn{constructor(e){super(),this.upstream=e,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${this.upstream.summary()} -> Serial`}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){return this.upstream.next()}},kq=class extends wn{constructor(e,t){super(),this.upstream=e,this.maxCount=t,this.count=0,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${this.upstream.summary()} -> Skip`}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){for(;this.count++<this.maxCount;){let e=await this.upstream.next();if(e.done)return e;ee(e.value)}return this.upstream.next()}},Iq=class extends wn{constructor(e,t){super(),this.upstream=e,this.maxCount=t,this.count=0}summary(){return`${this.upstream.summary()} -> Take`}async next(){return this.count++>=this.maxCount?{value:null,done:!0}:this.upstream.next()}},Sq=class extends wn{constructor(e,t,n=!0){super(),this.upstream=e,this.batchSize=t,this.enableSmallLastBatch=n,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${this.upstream.summary()} -> RowMajorBatch`}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){let e=[];for(;e.length<this.batchSize;){let t=await this.upstream.next();if(t.done)return this.enableSmallLastBatch&&e.length>0?{value:e,done:!1}:{value:null,done:!0};e.push(t.value)}return{value:e,done:!1}}},Cq=class extends wn{constructor(e,t){super(),this.upstream=e,this.predicate=t,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${this.upstream.summary()} -> Filter`}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){for(;;){let e=await this.upstream.next();if(e.done||this.predicate(e.value))return e;ee(e.value)}}},Tq=class extends wn{constructor(e,t){super(),this.upstream=e,this.transform=t}summary(){return`${this.upstream.summary()} -> Map`}async next(){let e=await this.upstream.next();if(e.done)return{value:null,done:!0};let t=Nr.getTensorsInContainer(e.value),n=this.transform(e.value),s=Nr.getTensorsInContainer(n);for(let r of t)Nr.isTensorInList(r,s)||r.dispose();return{value:n,done:!1}}},Nq=class extends wn{constructor(e,t){super(),this.upstream=e,this.handler=t,this.count=0,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${this.upstream.summary()} -> handleErrors`}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){for(;;)try{return await this.upstream.next()}catch(e){if(!this.handler(e))return{value:null,done:!0}}}},u7=class extends wn{constructor(e,t){super(),this.upstream=e,this.transform=t}summary(){return`${this.upstream.summary()} -> AsyncMap`}async next(){let e=await this.upstream.next();if(e.done)return{value:null,done:!0};let t=Nr.getTensorsInContainer(e.value),n=await this.transform(e.value),s=Nr.getTensorsInContainer(n);for(let r of t)Nr.isTensorInList(r,s)||r.dispose();return{value:n,done:!1}}},Ix=class extends wn{constructor(){super(),this.outputQueue=new wx,this.lastRead=Promise.resolve({value:null,done:!1})}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){for(;this.outputQueue.length()===0;)if(!await this.pump())return{value:null,done:!0};return{value:this.outputQueue.shift(),done:!1}}},Eq=class extends Ix{constructor(e,t){super(),this.upstream=e,this.transform=t}summary(){return`${this.upstream.summary()} -> Flatmap`}async pump(){let e=await this.upstream.next();if(e.done)return!1;let t=Nr.getTensorsInContainer(e.value),n=this.transform(e.value),s=Nr.getTensorsInContainer(n);this.outputQueue.pushAll(n);for(let r of t)Nr.isTensorInList(r,s)||r.dispose();return!0}},sI=class extends wn{constructor(e,t){super(),this.baseErrorHandler=t,this.lastRead=null,this.iterator=null,this.moreIterators=e}summary(){return"TODO: fill in upstream of chained summaries -> Chained"}async next(){return this.lastRead=this.readFromChain(this.lastRead),this.lastRead}async readFromChain(e){if(await e,this.iterator==null){let n=await this.moreIterators.next();if(n.done)return{value:null,done:!0};this.iterator=n.value,this.baseErrorHandler!=null&&(this.iterator=this.iterator.handleErrors(this.baseErrorHandler))}let t=await this.iterator.next();return t.done?(this.iterator=null,this.readFromChain(e)):t}},ja;(function(e){e[e.FAIL=0]="FAIL",e[e.SHORTEST=1]="SHORTEST",e[e.LONGEST=2]="LONGEST"})(ja||(ja={}));var Rq=class extends wn{constructor(e,t=ja.FAIL){super(),this.iterators=e,this.mismatchMode=t,this.count=0,this.currentPromise=null}summary(){return"{TODO: fill in upstream of zip summaries} -> Zip"}async nextState(e){await e;let t=0,n=0;function s(a){return a instanceof wn?{value:a.next().then(i=>(t++,i.done&&n++,i.value)),recurse:!1}:{value:null,recurse:!0}}let r=await eI(this.iterators,s);if(t===n)return{value:null,done:!0};if(n>0)switch(this.mismatchMode){case ja.FAIL:throw new Error(`Zipped streams should have the same length. Mismatched at element ${this.count}.`);case ja.SHORTEST:return{value:null,done:!0};case ja.LONGEST:default:}return this.count++,{value:r,done:!1}}async next(){return this.currentPromise=this.nextState(this.currentPromise),this.currentPromise}},rI=class extends wn{constructor(e,t){super(),this.upstream=e,this.bufferSize=t,this.buffer=new tI(t)}summary(){return`${this.upstream.summary()} -> Prefetch`}refill(){for(;!this.buffer.isFull();){let e=this.upstream.next();this.buffer.push(e)}}next(){return this.refill(),this.buffer.shift()}},_q=class extends rI{constructor(e,t,n){super(e,t),this.upstream=e,this.windowSize=t,this.upstreamExhausted=!1,this.random=dq.alea(n||v.now().toString()),this.lastRead=Promise.resolve({value:null,done:!1})}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}randomInt(e){return Math.floor(this.random()*e)}chooseIndex(){return this.randomInt(this.buffer.length())}async serialNext(){for(this.upstreamExhausted||this.refill();!this.buffer.isEmpty();){let e=this.chooseIndex(),t=await this.buffer.shuffleExcise(e);if(t.done)this.upstreamExhausted=!0;else return this.refill(),t}return{value:null,done:!0}}},Qc=class{constructor(){this.size=null}batch(e,t=!0){let n=this;v.assert(e>0,()=>`batchSize needs to be positive, but it is
|
|
${e}`);let s;return this.size===1/0||this.size==null?s=this.size:t?s=Math.ceil(this.size/e):s=Math.floor(this.size/e),Is(async()=>(await n.iterator()).columnMajorBatch(e,t,Pq),s)}concatenate(e){let t=this,n;return this.size===1/0||e.size===1/0?n=1/0:this.size!=null&&e.size!=null?n=this.size+e.size:n=null,Is(async()=>(await t.iterator()).concatenate(await e.iterator()),n)}filter(e){let t=this,n;return this.size===1/0?n=1/0:n=null,Is(async()=>(await t.iterator()).filter(s=>Y(()=>e(s))),n)}async forEachAsync(e){return(await this.iterator()).forEachAsync(e)}map(e){let t=this;return Is(async()=>(await t.iterator()).map(n=>Y(()=>e(n))),this.size)}mapAsync(e){let t=this;return Is(async()=>(await t.iterator()).mapAsync(e),this.size)}prefetch(e){if(e==null)throw new RangeError("`Dataset.prefetch()` requires bufferSize to be specified.");let t=this;return Is(async()=>(await t.iterator()).prefetch(e),this.size)}repeat(e){let t=this,n;return this.size!=null&&e>0?n=this.size*e:e===0?n=0:this.size!=null&&(e===void 0||e<0)?n=1/0:n=null,Is(async()=>{let s=kx(async()=>({value:await t.iterator(),done:!1}));return Aq(s.take(e))},n)}skip(e){let t=this,n;return this.size!=null&&e>=0&&this.size>=e?n=this.size-e:this.size!=null&&(this.size<e||e===void 0||e<0)?n=0:n=null,Is(async()=>(await t.iterator()).skip(e),n)}shuffle(e,t,n=!0){if(e==null||e<0)throw this.size==null?new RangeError("`Dataset.shuffle()` requires bufferSize to be specified."):new RangeError(`\`Dataset.shuffle()\` requires bufferSize to be specified. If your data fits in main memory (for regular JS objects), and/or GPU memory (for \`tf.Tensor\`s), consider setting bufferSize to the dataset size (${this.size} elements)`);let s=this,r=cq.alea(t||v.now().toString());return Is(async()=>{let a=r.int32();return n&&(a+=r.int32()),(await s.iterator()).shuffle(e,a.toString())},this.size)}take(e){let t=this,n;return this.size!=null&&this.size>e?n=e:this.size!=null&&this.size<=e?n=this.size:n=null,Is(async()=>(await t.iterator()).take(e),n)}async toArray(){if(this.size===1/0)throw new Error("Can not convert infinite data stream to array.");return(await this.iterator()).toArray()}async toArrayForTest(){if(this.size===1/0)throw new Error("Can not convert infinite data stream to array.");return(await this.iterator()).toArrayForTest()}};Qc.MAX_BUFFER_SIZE=1e4;function Is(e,t=null){return new class extends Qc{constructor(){super(...arguments),this.size=t}async iterator(){return e()}}}function Dq(e){return Is(async()=>nI(e),e.length)}function $q(e){if(!ac(e))throw new Error("The argument to zip() must be an object or array.");let t;if(Array.isArray(e))for(let n=0;n<e.length;n++)t=t==null?e[n].size:Math.min(t,e[n].size);else if(e instanceof Object)for(let n in e)t=t==null?e[n].size:Math.min(t,e[n].size);return Is(async()=>{let n=await eI(e,s=>{if(s instanceof Qc)return{value:s.iterator(),recurse:!1};if(ac(s))return{value:null,recurse:!0};throw new Error("Leaves of the structure passed to zip() must be Datasets, not primitives.")});return xq(n,ja.SHORTEST)},t)}function Pq(e){if(e===null)return null;let t=e[0];return fq(t)?{value:Fq(e),recurse:!1}:{value:null,recurse:!0}}function Fq(e){if(e.length===0)throw new Error("Can't make a batch of zero elements.");return e[0]instanceof nt?on(e):ct(e)}var aI=class extends Qc{constructor(e){super(),this.input=e}async iterator(){return(await this.input.iterator()).decodeUTF8().split(`
|
|
`).map(s=>(s.endsWith("\r")&&(s=s.slice(0,-1)),s))}},jf='"',np=Symbol("out"),c7=Symbol("field"),qf=Symbol("quote"),h3=Symbol("quoteafterquote"),d7=Symbol("quoteinquote"),oI=class extends Qc{constructor(e,t){super(),this.input=e,this.hasHeader=!0,this.fullColumnNames=null,this.columnNamesValidated=!1,this.columnConfigs=null,this.configuredColumnsOnly=!1,this.delimiter=",",this.delimWhitespace=!1,this.base=new aI(e),t||(t={}),this.hasHeader=t.hasHeader!==!1,this.fullColumnNames=t.columnNames,this.columnConfigs=t.columnConfigs,this.configuredColumnsOnly=t.configuredColumnsOnly,t.delimWhitespace?(v.assert(t.delimiter==null,()=>"Delimiter should not be provided when delimWhitespace is true."),this.delimWhitespace=!0,this.delimiter=" "):this.delimiter=t.delimiter?t.delimiter:","}async columnNames(){return this.columnNamesValidated||await this.setColumnNames(),this.configuredColumnsOnly?Object.keys(this.columnConfigs):this.fullColumnNames}async setColumnNames(){let e=await this.maybeReadHeaderLine();if(!this.fullColumnNames&&!e)throw new Error("Column names must be provided if there is no header line.");this.fullColumnNames&&e&&v.assert(e.length===this.fullColumnNames.length,()=>"The length of provided columnNames ("+this.fullColumnNames.length.toString()+") does not match the length of the header line read from file ("+e.length.toString()+")."),this.fullColumnNames||(this.fullColumnNames=e);let t=this.fullColumnNames.reduce((s,r)=>(s[r]=s[r]+1||1,s),{}),n=Object.keys(t).filter(s=>t[s]>1);if(v.assert(n.length===0,()=>"Duplicate column names found: "+n.toString()),this.columnConfigs){for(let s of Object.keys(this.columnConfigs))if(this.fullColumnNames.indexOf(s)===-1)throw new Error('The key "'+s+'" provided in columnConfigs does not match any of the column names ('+this.fullColumnNames.toString()+").")}this.columnNamesValidated=!0}async maybeReadHeaderLine(){if(this.hasHeader){let t=await(await this.base.iterator()).next();if(t.done)throw new Error("No data was found for CSV parsing.");let n=t.value;return this.parseRow(n,!1)}else return null}async iterator(){this.columnNamesValidated||await this.setColumnNames();let e=await this.base.iterator();return this.hasHeader&&(e=e.skip(1)),e.map(t=>this.makeDataElement(t))}makeDataElement(e){let t=this.parseRow(e),n={},s={};for(let r=0;r<this.fullColumnNames.length;r++){let a=this.fullColumnNames[r],o=this.columnConfigs?this.columnConfigs[a]:null;if(!(this.configuredColumnsOnly&&!o)){let i=t[r],l=null;if(i==="")if(o&&o.default!==void 0)l=o.default;else{if(o&&(o.required||o.isLabel))throw new Error(`Required column ${a} is empty in this line: ${e}`);l=void 0}else{let u=Number(i);if(isNaN(u))o&&o.dtype==="bool"?l=this.getBoolean(i):l=i;else if(!o||!o.dtype)l=u;else switch(o.dtype){case"float32":l=u;break;case"int32":l=Math.floor(u);break;case"bool":l=this.getBoolean(i);break;default:l=u}}o&&o.isLabel?s[a]=l:n[a]=l}}return Object.keys(s).length===0?n:{xs:n,ys:s}}getBoolean(e){return e==="1"||e.toLowerCase()==="true"?1:0}parseRow(e,t=!0){let n=[],s=0,r=e.length,a=np;for(let o=0;o<r;o++)switch(a){case np:switch(e.charAt(o)){case jf:s=o+1,a=qf;break;case this.delimiter:if(s=o+1,this.delimiter===" "&&this.delimWhitespace)break;n.push(""),a=np;break;default:a=c7,s=o;break}break;case c7:switch(e.charAt(o)){case this.delimiter:n.push(e.substring(s,o)),a=np,s=o+1;break;default:}break;case qf:switch(e.charAt(o)){case jf:a=h3;break;default:}break;case h3:switch(e.charAt(o)){case this.delimiter:n.push(e.substring(s,o-1)),a=np,s=o+1;break;case jf:a=qf;break;default:a=d7;break}break;case d7:switch(e.charAt(o)){case jf:a=qf;break;default:}break;default:}if(a===h3?n.push(e.substring(s,r-1)):n.push(e.substring(s)),t&&n.length!==this.fullColumnNames.length)throw new Error(`Invalid row in csv file. Should have ${this.fullColumnNames.length} elements in a row, but got ${n}`);return n}},iI=class extends wn{constructor(e){super(),this.microphoneConfig=e,this.isClosed=!1,this.fftSize=e.fftSize||1024;let t=Math.log2(this.fftSize);if(this.fftSize<0||t<4||t>14||!Number.isInteger(t))throw new Error(`Invalid fftSize: it must be a power of 2 between 2 to 4 and 2 to 14, but got ${this.fftSize}`);if(this.numFrames=e.numFramesPerSpectrogram||43,this.sampleRateHz=e.sampleRateHz,this.columnTruncateLength=e.columnTruncateLength||this.fftSize,this.audioTrackConstraints=e.audioTrackConstraints,this.smoothingTimeConstant=e.smoothingTimeConstant||0,this.includeSpectrogram=e.includeSpectrogram!==!1,this.includeWaveform=e.includeWaveform===!0,!this.includeSpectrogram&&!this.includeWaveform)throw new Error("Both includeSpectrogram and includeWaveform are false. At least one type of data should be returned.")}summary(){return"microphone"}static async create(e={}){if(!q().get("IS_BROWSER"))throw new Error("microphone API is only supported in browser environment.");let t=new iI(e);return await t.start(),t}async start(){try{this.stream=await navigator.mediaDevices.getUserMedia({audio:this.audioTrackConstraints==null?!0:this.audioTrackConstraints,video:!1})}catch(n){throw new Error(`Error thrown while initializing video stream: ${n.message}`)}if(!this.stream)throw new Error("Could not obtain audio from microphone.");let e=window.AudioContext||window.webkitAudioContext;if(this.audioContext=new e,!this.sampleRateHz)this.sampleRateHz=this.audioContext.sampleRate;else if(this.audioContext.sampleRate!==this.sampleRateHz)throw new Error(`Mismatch in sampling rate: Expected: ${this.sampleRateHz}; Actual: ${this.audioContext.sampleRate}`);let t=this.audioContext.createMediaStreamSource(this.stream);this.analyser=this.audioContext.createAnalyser(),this.analyser.fftSize=this.fftSize*2,this.analyser.smoothingTimeConstant=this.smoothingTimeConstant,t.connect(this.analyser),this.freqData=new Float32Array(this.fftSize),this.timeData=new Float32Array(this.fftSize)}async next(){if(this.isClosed)return{value:null,done:!0};let e,t,n=await this.getAudioData();if(this.includeSpectrogram){let s=this.flattenQueue(n.freqDataQueue);e=this.getTensorFromAudioDataArray(s,[this.numFrames,this.columnTruncateLength,1])}if(this.includeWaveform){let s=this.flattenQueue(n.timeDataQueue);t=this.getTensorFromAudioDataArray(s,[this.numFrames*this.fftSize,1])}return{value:{spectrogram:e,waveform:t},done:!1}}async capture(){return(await this.next()).value}async getAudioData(){let e=[],t=[],n=0;return new Promise(s=>{let r=setInterval(()=>{this.includeSpectrogram&&(this.analyser.getFloatFrequencyData(this.freqData),this.freqData[0]===-1/0&&s({freqDataQueue:e,timeDataQueue:t}),e.push(this.freqData.slice(0,this.columnTruncateLength))),this.includeWaveform&&(this.analyser.getFloatTimeDomainData(this.timeData),t.push(this.timeData.slice())),++n===this.numFrames&&(clearInterval(r),s({freqDataQueue:e,timeDataQueue:t}))},this.fftSize/this.sampleRateHz*1e3)})}stop(){this.isClosed||(this.isClosed=!0,this.analyser.disconnect(),this.audioContext.close(),this.stream!=null&&this.stream.getTracks().length>0&&this.stream.getTracks()[0].stop())}toArray(){throw new Error("Can not convert infinite audio stream to array.")}getSampleRate(){return this.sampleRateHz}flattenQueue(e){let t=e[0].length,n=new Float32Array(e.length*t);return e.forEach((s,r)=>n.set(s,r*t)),n}getTensorFromAudioDataArray(e,t){let n=new Float32Array(v.sizeFromShape(t));return n.set(e,n.length-e.length),ct(n,t)}},lI=class extends wn{constructor(e,t){if(super(),this.webcamVideoElement=e,this.webcamConfig=t,this.isClosed=!0,this.resize=!1,this.needToResize())if(this.resize=!0,this.cropSize=[this.webcamConfig.resizeHeight,this.webcamConfig.resizeWidth],this.cropBoxInd=Ft([0],"int32"),this.webcamConfig.centerCrop){let n=this.webcamConfig.resizeWidth*1/this.webcamVideoElement.width,s=this.webcamConfig.resizeHeight*1/this.webcamVideoElement.height,r=(1-n)/2,a=(1-s)/2,o=r+n,i=s+a;this.cropBox=ir([a,r,i,o],[1,4])}else this.cropBox=ir([0,0,1,1],[1,4])}summary(){return"webcam"}static async create(e,t={}){if(!q().get("IS_BROWSER"))throw new Error("tf.data.webcam is only supported in browser environment.");if(!e){if(e=document.createElement("video"),!t.resizeWidth||!t.resizeHeight)throw new Error("Please provide webcam video element, or resizeWidth and resizeHeight to create a hidden video element.");e.width=t.resizeWidth,e.height=t.resizeHeight}let n=new lI(e,t);return await n.start(),n}async start(){this.webcamConfig.facingMode&&v.assert(this.webcamConfig.facingMode==="user"||this.webcamConfig.facingMode==="environment",()=>`Invalid webcam facing mode: ${this.webcamConfig.facingMode}. Please provide 'user' or 'environment'`);try{this.stream=await navigator.mediaDevices.getUserMedia({video:{deviceId:this.webcamConfig.deviceId,facingMode:this.webcamConfig.facingMode?this.webcamConfig.facingMode:"user",width:this.webcamVideoElement.width,height:this.webcamVideoElement.height}})}catch(e){throw e.message=`Error thrown while initializing video stream: ${e.message}`,e}if(!this.stream)throw new Error("Could not obtain video from webcam.");try{this.webcamVideoElement.srcObject=this.stream}catch(e){console.log(e),this.webcamVideoElement.src=window.URL.createObjectURL(this.stream)}return this.webcamVideoElement.play(),this.isClosed=!1,new Promise(e=>{this.webcamVideoElement.onloadedmetadata=()=>{e()}})}async next(){if(this.isClosed)return{value:null,done:!0};let e;try{e=Js.fromPixels(this.webcamVideoElement)}catch(t){throw new Error(`Error thrown converting video to pixels: ${JSON.stringify(t)}`)}if(this.resize)try{return{value:this.cropAndResizeFrame(e),done:!1}}catch(t){throw new Error(`Error thrown cropping the video: ${t.message}`)}finally{e.dispose()}else return{value:e,done:!1}}needToResize(){return!!(this.webcamConfig.resizeWidth&&this.webcamConfig.resizeHeight&&(this.webcamVideoElement.width!==this.webcamConfig.resizeWidth||this.webcamVideoElement.height!==this.webcamConfig.resizeHeight))}cropAndResizeFrame(e){return Y(()=>{let t=Kt(ye(e,"float32"),0),n;n=Se.cropAndResize(t,this.cropBox,this.cropBoxInd,this.cropSize,"bilinear");let s=n.shape;return U(n,s.slice(1))})}async capture(){return(await this.next()).value}stop(){this.stream.getTracks().forEach(t=>t.stop());try{this.webcamVideoElement.srcObject=null}catch(t){console.log(t),this.webcamVideoElement.src=null}this.isClosed=!0}toArray(){throw new Error("Can not convert infinite video stream to array.")}},uI=class{},cI=class extends wn{split(e){return new Oq(this,e)}},Oq=class extends cI{constructor(e,t){super(),this.upstream=e,this.impl=new Mq(e,t)}summary(){return this.impl.summary()}async next(){return this.impl.next()}},Mq=class extends Ix{constructor(e,t){super(),this.upstream=e,this.separator=t,this.carryover=""}summary(){return`${this.upstream.summary()} -> Split('${this.separator}')`}async pump(){let e=await this.upstream.next();if(e.done)return this.carryover===""?!1:(this.outputQueue.push(this.carryover),this.carryover="",!0);let t=e.value.split(this.separator);t[0]=this.carryover+t[0];for(let n of t.slice(0,-1))this.outputQueue.push(n);return this.carryover=t[t.length-1],!0}},zq=class extends wn{decodeUTF8(){return new Lq(this)}},Lq=class extends cI{constructor(e){super(),this.upstream=e,this.impl=new Bq(e)}summary(){return this.impl.summary()}async next(){return this.impl.next()}},Bq=class extends Ix{constructor(e){if(super(),this.upstream=e,q().get("IS_BROWSER"))this.decoder=new TextDecoder("utf-8");else{let{StringDecoder:t}=e6();this.decoder=new t("utf8")}}summary(){return`${this.upstream.summary()} -> Utf8`}async pump(){let e=await this.upstream.next(),t;if(e.done)return!1;t=e.value;let n;return q().get("IS_BROWSER")?n=this.decoder.decode(t,{stream:!0}):n=this.decoder.write(Buffer.from(t.buffer)),this.outputQueue.push(n),!0}},dI=class extends zq{constructor(e,t={}){super(),this.file=e,this.options=t,v.assert(e instanceof Uint8Array||(q().get("IS_BROWSER")?e instanceof File||e instanceof Blob:!1),()=>"FileChunkIterator only supports File, Blob and Uint8Array right now."),this.offset=t.offset||0,this.chunkSize=t.chunkSize||1024*1024}summary(){return`FileChunks ${this.file}`}async next(){return this.offset>=(this.file instanceof Uint8Array?this.file.byteLength:this.file.size)?{value:null,done:!0}:{value:await new Promise((t,n)=>{let s=this.offset+this.chunkSize;if(this.file instanceof Uint8Array)t(new Uint8Array(this.file.slice(this.offset,s)));else{let r=new FileReader;r.onload=o=>{let i=r.result;if(i instanceof ArrayBuffer&&(i=new Uint8Array(i)),!(i instanceof Uint8Array))return n(new TypeError("FileReader returned unknown type."));t(i)},r.onabort=o=>n(new Error("Aborted")),r.onerror=o=>n(new Error(o.type));let a=this.file.slice(this.offset,s);r.readAsArrayBuffer(a)}this.offset=s}),done:!1}}};async function Wq(e,t={},n){let s,r;typeof e=="string"?s=e:(s=e.url,r=Vq(e));let a=await(n||v.fetch)(s,r);if(a.ok){let o=new Uint8Array(await a.arrayBuffer());return new dI(o,t)}else throw new Error(a.statusText)}var Vq=e=>({method:e.method,headers:e.headers,body:e.body,mode:e.mode,credentials:e.credentials,cache:e.cache,redirect:e.redirect,referrer:e.referrer,integrity:e.integrity});function pI(e){return typeof e=="string"&&e.slice(0,7)==="file://"}var hI=class extends uI{constructor(e,t={}){super(),this.input=e,this.options=t}async iterator(){if(pI(this.input)&&q().get("IS_NODE")){let e=xy();this.input=e.readFileSync(this.input.slice(7))}return new dI(this.input,this.options)}},fI=class extends uI{constructor(e,t={}){super(),this.url=e,this.fileOptions=t}async iterator(){return pI(this.url)?new hI(this.url,this.fileOptions).iterator():Wq(this.url,this.fileOptions)}};function Uq(e,t={}){return new oI(new fI(e),t)}function Gq(e){let t=kx(e);return Is(async()=>t)}function Hq(e){return Is(async()=>{let t=await e();return kx(()=>t.next())})}async function jq(e,t){return lI.create(e,t)}async function qq(e){return iI.create(e)}var Xq="3.19.0";function Te(e,t){Array.isArray(e)||(e=[e]),e.forEach(n=>{n!=null&&v.assert(n.dtype!=="complex64",()=>`${t} does not support complex64 tensors in the CPU backend.`)})}var Kq=pr.whereImpl,Sx=class extends cc{constructor(){super(),this.blockSize=48,this.firstUse=!0,this.data=new Mp(this,sn())}nextDataId(){return Sx.nextDataId++}write(e,t,n){this.firstUse&&(this.firstUse=!1,q().get("IS_NODE")&&T.warn(`
|
|
============================
|
|
Hi, looks like you are running TensorFlow.js in Node.js. To speed things up dramatically, install our node backend, visit https://github.com/tensorflow/tfjs-node for more details.
|
|
============================`));let s={id:this.nextDataId()};return this.data.set(s,{values:e,dtype:n,refCount:1}),s}makeTensorInfo(e,t,n){let s;if(t==="string"&&n!=null&&n.length>0&&v.isString(n[0])){let r=n.map(a=>v.encodeString(a));s=this.write(r,e,t)}else s=this.write(n,e,t);return{dataId:s,shape:e,dtype:t}}refCount(e){return this.data.has(e)?this.data.get(e).refCount:0}incRef(e){let t=this.data.get(e);t.refCount++}decRef(e){if(this.data.has(e)){let t=this.data.get(e);t.refCount--}}move(e,t,n,s,r){this.data.set(e,{values:t,dtype:s,refCount:r})}numDataIds(){return this.data.numDataIds()}async read(e){return this.readSync(e)}readSync(e){let{dtype:t,complexTensorInfos:n}=this.data.get(e);if(t==="complex64"){let s=this.readSync(n.real.dataId),r=this.readSync(n.imag.dataId);return T.mergeRealAndImagArrays(s,r)}return this.data.get(e).values}bufferSync(e){let t=this.readSync(e.dataId);if(e.dtype==="string")try{let n=t.map(s=>v.decodeString(s));return Le(e.shape,e.dtype,n)}catch(n){throw new Error("Failed to decode encoded string bytes into utf-8")}return Le(e.shape,e.dtype,t)}makeOutput(e,t,n){return sn().makeTensorFromTensorInfo(this.makeTensorInfo(t,n,e),this)}disposeData(e,t=!1){if(this.data.has(e)){if(this.data.get(e).refCount--,!t&&this.data.get(e).refCount>0)return!1;let{complexTensorInfos:n}=this.data.get(e);n!=null&&(this.disposeData(n.real.dataId,!0),this.disposeData(n.imag.dataId,!0)),this.data.delete(e)}return!0}disposeIntermediateTensorInfo(e){this.disposeData(e.dataId)}async time(e){let t=v.now();return e(),{kernelMs:v.now()-t}}memory(){return{unreliable:!0,reasons:["The reported memory is an upper bound. Due to automatic garbage collection, the true allocated memory may be less."]}}where(e){Te([e],"where");let t=this.readSync(e.dataId);return Kq(e.shape,t)}dispose(){}floatPrecision(){return 32}epsilon(){return super.epsilon()}};Sx.nextDataId=0;var Cx={};Ve(Cx,{addImpl:()=>gI,bincountImpl:()=>Nx,bincountReduceImpl:()=>yI,ceilImpl:()=>AI,concatImpl:()=>Ex,equalImpl:()=>xI,expImpl:()=>vI,expm1Impl:()=>kI,floorImpl:()=>II,gatherNdImpl:()=>SI,gatherV2Impl:()=>CI,greaterEqualImpl:()=>NI,greaterImpl:()=>TI,lessEqualImpl:()=>RI,lessImpl:()=>EI,linSpaceImpl:()=>_I,logImpl:()=>DI,maxImpl:()=>$I,maximumImpl:()=>PI,minimumImpl:()=>FI,multiplyImpl:()=>Rx,negImpl:()=>OI,notEqualImpl:()=>MI,prodImpl:()=>zI,rangeImpl:()=>Dx,rsqrtImpl:()=>LI,scatterImpl:()=>Bu,sigmoidImpl:()=>MX,simpleAbsImpl:()=>mI,sliceImpl:()=>_m,sparseFillEmptyRowsImpl:()=>WI,sparseReshapeImpl:()=>VI,sparseSegmentReductionImpl:()=>$x,sqrtImpl:()=>BX,squaredDifferenceImpl:()=>UI,stridedSliceImpl:()=>GI,stringNGramsImpl:()=>Px,stringSplitImpl:()=>Fx,stringToHashBucketFastImpl:()=>Ox,subImpl:()=>HI,tileImpl:()=>jI,topKImpl:()=>XI,transposeImpl:()=>_x,uniqueImpl:()=>KI});function mI(e){let t=new Float32Array(e.length);for(let n=0;n<e.length;++n)t[n]=Math.abs(e[n]);return t}var Zq=e=>{let{x:t}=e.inputs,n=e.backend;Te(t,"abs");let s=new Float32Array(v.sizeFromShape(t.shape)),r=n.data.get(t.dataId).values;return s=mI(r),n.makeOutput(s,t.shape,t.dtype)},Yq={kernelName:il,backendName:"cpu",kernelFunc:Zq};function un(e){return(t,n,s,r,a)=>{let o=T.assertAndGetBroadcastShape(t,n),i=o.length,l=v.computeStrides(o),u=v.sizeFromShape(o),c=v.getTypedArrayFromDType(a,u),p=t.length,d=n.length,h=v.computeStrides(t),f=v.computeStrides(n),m=T.getBroadcastDims(t,o),g=T.getBroadcastDims(n,o);if(m.length+g.length===0)for(let y=0;y<c.length;++y)c[y]=e(s[y%s.length],r[y%r.length]);else for(let y=0;y<c.length;++y){let x=v.indexToLoc(y,i,l),A=x.slice(-p);m.forEach(S=>A[S]=0);let b=v.locToIndex(A,p,h),w=x.slice(-d);g.forEach(S=>w[S]=0);let k=v.locToIndex(w,d,f);c[y]=e(s[b],r[k])}return[c,o]}}function Cs(e){let{inputs:t,backend:n}=e,{real:s,imag:r}=t,a=n.data.get(s.dataId).values,o=n.data.get(r.dataId).values,i=n.makeTensorInfo(s.shape,"complex64"),l=n.data.get(i.dataId);return l.complexTensorInfos={real:n.makeTensorInfo(s.shape,"float32",a),imag:n.makeTensorInfo(r.shape,"float32",o)},i}var Jq={kernelName:Lp,backendName:"cpu",kernelFunc:Cs};function Rm(e,t,n="float32"){if(n==="complex64"){let r=Rm(e,t,"float32"),a=Rm(e,t,"float32");return Cs({inputs:{real:r,imag:a},backend:e})}let s=v.makeZerosTypedArray(v.sizeFromShape(t),n);return e.makeTensorInfo(t,n,s)}function Zr(e){let{inputs:t,backend:n}=e,{x:s}=t;return n.incRef(s.dataId),{dataId:s.dataId,shape:s.shape,dtype:s.dtype}}var Qq={kernelName:_o,backendName:"cpu",kernelFunc:Zr};function tl(e){let{inputs:t,backend:n}=e,{input:s}=t,r=n.data.get(s.dataId).complexTensorInfos.real,a=n.data.get(r.dataId).values;return n.makeTensorInfo(r.shape,r.dtype,a)}var eX={kernelName:qp,backendName:"cpu",kernelFunc:tl};function io(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{dtype:a}=s;if(a==="complex64"){if(r.dtype==="complex64")return Zr({inputs:{x:r},backend:n});let o=Rm(n,r.shape,r.dtype),i=io({inputs:{x:r},backend:n,attrs:{dtype:"float32"}}),l=Cs({inputs:{real:i,imag:o},backend:n});return n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}if(r.dtype==="complex64"){let o=tl({inputs:{input:r},backend:n}),i=io({inputs:{x:o},backend:n,attrs:{dtype:a}});return n.disposeIntermediateTensorInfo(o),i}if(!v.hasEncodingLoss(r.dtype,a)){let o=Zr({inputs:{x:r},backend:n});return{dataId:o.dataId,shape:o.shape,dtype:a}}if(a==="int32"){let o=n.data.get(r.dataId).values,i=Int32Array.from(o);return n.makeTensorInfo(r.shape,"int32",i)}if(a==="bool"){let o=n.data.get(r.dataId).values,i=v.toTypedArray([0],r.dtype),[l,u]=un((c,p)=>c!==p?1:0)(r.shape,[],o,i,"bool");return n.makeTensorInfo(u,"bool",l)}throw new Error(`Error in Cast: failed to cast ${r.dtype} to ${a}`)}var tX={kernelName:go,backendName:"cpu",kernelFunc:io};function kn(e,t,n,s){return n==null?({inputs:r,backend:a})=>{let{a:o,b:i}=r,l=a;Te([o,i],e);let u=l.data.get(o.dataId).values,c=l.data.get(i.dataId).values,p=o.dtype==="string"?T.fromUint8ToStringArray(u):u,d=o.dtype==="string"?T.fromUint8ToStringArray(c):c,h=s||o.dtype,[f,m]=t(o.shape,i.shape,p,d,h);return l.makeTensorInfo(m,h,f)}:({inputs:r,backend:a})=>{let{a:o,b:i}=r,l=a;if(o.dtype==="complex64"||i.dtype==="complex64"){let u=io({inputs:{x:o},backend:l,attrs:{dtype:"complex64"}}),c=l.data.get(u.dataId),p=c.complexTensorInfos.real,d=c.complexTensorInfos.imag,h=l.data.get(p.dataId).values,f=l.data.get(d.dataId).values,m=io({inputs:{x:i},backend:l,attrs:{dtype:"complex64"}}),g=l.data.get(m.dataId),y=g.complexTensorInfos.real,x=g.complexTensorInfos.imag,A=l.data.get(y.dataId).values,b=l.data.get(x.dataId).values,[w,k,S]=n(o.shape,i.shape,h,f,A,b),E=l.makeTensorInfo(S,"float32",w),R=l.makeTensorInfo(S,"float32",k),$=Cs({inputs:{real:E,imag:R},backend:l});return l.disposeIntermediateTensorInfo(u),l.disposeIntermediateTensorInfo(m),l.disposeIntermediateTensorInfo(E),l.disposeIntermediateTensorInfo(R),$}else{let u=l.data.get(o.dataId).values,c=l.data.get(i.dataId).values,p=s||o.dtype,[d,h]=t(o.shape,i.shape,u,c,p);return l.makeTensorInfo(h,p,d)}}}function Tx(e){return(t,n,s,r,a,o)=>{let i=T.assertAndGetBroadcastShape(t,n),l=v.sizeFromShape(i),u=i.length,c=v.computeStrides(i),p=v.getTypedArrayFromDType("float32",l),d=v.getTypedArrayFromDType("float32",l),h=T.getBroadcastDims(t,i),f=T.getBroadcastDims(n,i),m=T.mergeRealAndImagArrays(s,r),g=T.mergeRealAndImagArrays(a,o),y=t.length,x=v.computeStrides(t),A=n.length,b=v.computeStrides(n);if(h.length+f.length===0)for(let w=0;w<p.length;w++){let k=w%m.length,S=w%g.length,E=e(m[k*2],m[k*2+1],g[S*2],g[S*2+1]);p[w]=E.real,d[w]=E.imag}else for(let w=0;w<p.length;w++){let k=v.indexToLoc(w,u,c),S=k.slice(-y);h.forEach(P=>S[P]=0);let E=v.locToIndex(S,y,x),R=k.slice(-A);f.forEach(P=>R[P]=0);let $=v.locToIndex(R,A,b),_=e(m[E*2],m[E*2+1],g[$*2],g[$*2+1]);p[w]=_.real,d[w]=_.imag}return[p,d,i]}}var gI=un((e,t)=>e+t),nX=Tx((e,t,n,s)=>({real:e+n,imag:t+s})),oc=kn(xa,gI,nX),sX={kernelName:xa,backendName:"cpu",kernelFunc:oc};function Nx(e,t,n,s,r){let a=v.sizeFromShape(s),o=v.makeZerosTypedArray(r,n);for(let i=0;i<e.length;i++){let l=e[i];if(l<0)throw new Error("Input x must be non-negative!");l>=r||(a>0?o[l]+=t[i]:o[l]+=1)}return o}function yI(e,t,n,s=!1){let r=e.shape[0],a=e.shape[1],o=Le([r,n],t.dtype);for(let i=0;i<r;i++)for(let l=0;l<a;l++){let u=e.get(i,l);if(u<0)throw new Error("Input x must be non-negative!");u>=n||(s?o.set(1,i,u):t.size>0?o.set(o.get(i,u)+t.get(i,l),i,u):o.set(o.get(i,u)+1,i,u))}return o}function di(e){return(t,n,s)=>{let r=v.getTypedArrayFromDType(n,t.length);for(let a=0;a<t.length;++a)r[a]=e(t[a],s);return r}}function bt(e,t,n){return({inputs:s,attrs:r,backend:a})=>{let{x:o}=s;if(Te(o,e),o.dtype==="string"||n==="string")throw new Error("unaryKernelFunc does not support string input/output");let i=a,l=i.data.get(o.dataId).values,u=v.sizeFromShape(o.shape),c=n||o.dtype,p=v.getArrayFromDType(c,u);for(let d=0;d<u;++d)p[d]=t(l[d],r);return i.makeTensorInfo(o.shape,c,p)}}function ed(e,t,n){return({inputs:s,attrs:r,backend:a})=>{let{x:o}=s;if(Te(o,e),o.dtype==="string"||n==="string")throw new Error("unaryKernelFunc does not support string input/output");let i=a,l=i.data.get(o.dataId).values,u=n||o.dtype,c=t(l,u,r);return i.makeTensorInfo(o.shape,u,c)}}var AI=di(e=>Math.ceil(e)),rX=ed(yo,AI),aX={kernelName:yo,backendName:"cpu",kernelFunc:rX};function Ex(e,t,n,s){let r=v.getArrayFromDType(n,v.sizeFromShape(t));if(s&&n!=="string"){let a=0;e.forEach(o=>{let i=v.sizeFromShape(o.shape);r.set(o.vals,a),a+=i})}else{let a=0;e.forEach(o=>{let i=n==="string"?T.fromUint8ToStringArray(o.vals):o.vals,l=0;for(let u=0;u<o.shape[0];++u){let c=u*t[1]+a;for(let p=0;p<o.shape[1];++p)r[c+p]=i[l++]}a+=o.shape[1]})}return r}var xI=un((e,t)=>e===t?1:0),bI=kn(hl,xI,null,"bool"),oX={kernelName:hl,backendName:"cpu",kernelFunc:bI},vI=di(e=>Math.exp(e)),wI=ed(Co,vI,"float32"),iX={kernelName:Co,backendName:"cpu",kernelFunc:wI},kI=di(e=>Math.expm1(e)),lX=ed(ml,kI),uX={kernelName:ml,backendName:"cpu",kernelFunc:lX},II=di(e=>Math.floor(e)),cX=ed(To,II),dX={kernelName:To,backendName:"cpu",kernelFunc:cX};function SI(e,t,n,s,r,a,o,i,l){let u=Le([s,a],n);for(let c=0;c<s;c++){let p=[],d=0;for(let h=0;h<r;h++){let f=e[c*r+h];d+=f*o[h],p.push(f)}if(d<0||d>=l/a)throw new Error(`Invalid indices: ${p} does not index into ${i}`);for(let h=0;h<a;h++)u.values[c*a+h]=t.get(...t.indexToLoc(d*a+h))}return u}function CI(e,t,n){let s=Le(n,e.dtype);for(let r=0;r<s.size;++r){let o=s.indexToLoc(r).slice(),i=o[0],l=o[2],u=t.locToIndex([i,l]);o[2]=t.values[u];let c=e.locToIndex(o);0<=c&&c<e.values.length&&(s.values[r]=e.values[c])}return s}var TI=un((e,t)=>e>t?1:0),pX=kn(xl,TI,null,"bool"),hX={kernelName:xl,backendName:"cpu",kernelFunc:pX},NI=un((e,t)=>e>=t?1:0),fX=kn(Ro,NI,null,"bool"),mX={kernelName:Ro,backendName:"cpu",kernelFunc:fX},EI=un((e,t)=>e<t?1:0),gX=kn(bl,EI,null,"bool"),yX={kernelName:bl,backendName:"cpu",kernelFunc:gX},RI=un((e,t)=>e<=t?1:0),AX=kn(vl,RI,null,"bool"),xX={kernelName:vl,backendName:"cpu",kernelFunc:AX};function _I(e,t,n){let s=(t-e)/(n-1),r=v.makeZerosTypedArray(n,"float32");r[0]=e;for(let a=1;a<r.length;a++)r[a]=r[a-1]+s;return r}var DI=di(e=>Math.log(e)),bX=ed($o,DI),vX={kernelName:$o,backendName:"cpu",kernelFunc:bX};function $I(e,t,n,s){let r=v.getTypedArrayFromDType(s,v.sizeFromShape(n));for(let a=0;a<r.length;++a){let o=a*t,i=e[o];for(let l=0;l<t;++l){let u=e[o+l];(Number.isNaN(u)||u>i)&&(i=u)}r[a]=i}return r}var PI=un((e,t)=>Math.max(e,t)),wX=kn(Fo,PI),kX={kernelName:Fo,backendName:"cpu",kernelFunc:wX},FI=un((e,t)=>Math.min(e,t)),IX=kn(Lo,FI),SX={kernelName:Lo,backendName:"cpu",kernelFunc:IX},Rx=un((e,t)=>e*t),CX=Tx((e,t,n,s)=>({real:e*n-t*s,imag:e*s+t*n})),A2=kn(Wo,Rx,CX),TX={kernelName:Wo,backendName:"cpu",kernelFunc:A2};function OI(e,t,n){let s=v.createScalarValue(-1,n);return Rx([],t,s,e,n)}function NX(e){let{inputs:t,backend:n}=e,{x:s}=t;Te(s,"neg");let r=n.data.get(s.dataId).values,[a,o]=OI(r,s.shape,s.dtype);return n.makeTensorInfo(o,s.dtype,a)}var EX={kernelName:Il,backendName:"cpu",kernelFunc:NX},MI=un((e,t)=>e!==t?1:0),RX=kn(Sl,MI,null,"bool"),_X={kernelName:Sl,backendName:"cpu",kernelFunc:RX};function _x(e,t,n,s,r){let a=t.length,o=v.sizeFromShape(t),i=v.computeStrides(t),l=v.computeStrides(r),u=v.getTypedArrayFromDType(n,v.sizeFromShape(r));for(let c=0;c<o;++c){let p=v.indexToLoc(c,a,i),d=new Array(p.length);for(let f=0;f<d.length;f++)d[f]=p[s[f]];let h=v.locToIndex(d,a,l);u[h]=e[c]}return u}function gs(e){let{inputs:t,attrs:n,backend:s}=e,{x:r}=t,{perm:a}=n;Te(r,"transpose");let o=r.shape.length,i=new Array(o);for(let p=0;p<i.length;p++)i[p]=r.shape[a[p]];let l=s.data.get(r.dataId).values,u=_x(l,r.shape,r.dtype,a,i);return{dataId:s.write(u,i,r.dtype),shape:i,dtype:r.dtype}}var DX={kernelName:Hr,backendName:"cpu",kernelFunc:gs};function zI(e,t,n,s){let[r,a]=T.computeOutAndReduceShapes(e,s),o=Mn(t,"int32"),i=v.makeZerosTypedArray(v.sizeFromShape(r),o),l=v.sizeFromShape(a);for(let u=0;u<i.length;++u){let c=u*l,p=1;for(let d=0;d<l;++d)p*=n[c+d];i[u]=p}return{outVals:i,outShape:r,outDtype:o}}function $X(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;Te(r,"prod");let i=r.shape.length,l=v.parseAxisParam(a,r.shape),u=T.getAxesPermutation(l,i),c=l,p=r,d=[];u!=null&&(p=gs({inputs:{x:r},backend:n,attrs:{perm:u}}),d.push(p),c=T.getInnerMostAxes(c.length,i));let h=n.data.get(p.dataId).values,{outVals:f,outShape:m,outDtype:g}=zI(p.shape,p.dtype,h,c),y=m;return o&&(y=T.expandShapeToKeepDim(m,l)),d.forEach(x=>n.disposeIntermediateTensorInfo(x)),n.makeTensorInfo(y,g,f)}var PX={kernelName:Ho,backendName:"cpu",kernelFunc:$X};function Dx(e,t,n,s){let r=e===t,a=e<t&&n<0,o=t<e&&n>1;if(r||a||o)return v.makeZerosTypedArray(0,s);let i=Math.abs(Math.ceil((t-e)/n)),l=v.makeZerosTypedArray(i,s);t<e&&n===1&&(n=-1),l[0]=e;for(let u=1;u<l.length;u++)l[u]=l[u-1]+n;return l}var LI=di(e=>1/Math.sqrt(e)),FX=ed(Zo,LI),OX={kernelName:Zo,backendName:"cpu",kernelFunc:FX};function Bu(e,t,n,s,r,a,o,i,l,u){let c=[s/r,r],p=e.values,d=t.values;if(s===0)return Le(n,t.dtype);let h=Le(c,t.dtype);typeof l=="string"||typeof l=="number"?h.values.fill(l):typeof l=="boolean"&&h.values.fill(+l);for(let f=0;f<a;f++){let m=[],g=0;for(let y=0;y<o;y++){let x=p[f*o+y];m.push(x),g+=x*i[y]}if(g<0||g>=s/r)throw new Error(`Invalid indices: ${m} does not index into ${n}`);for(let y=0;y<r;y++)u?h.values[g*r+y]+=d[f*r+y]:h.values[g*r+y]=t.rank===0?d[0]:d[f*r+y]}return h}var MX=di(e=>1/(1+Math.exp(-e))),BI=bt(Jo,e=>1/(1+Math.exp(-e))),zX={kernelName:Jo,backendName:"cpu",kernelFunc:BI};function _m(e,t,n,s,r){let a=Ut.isSliceContinous(s,t,n),o=v.sizeFromShape(n),i=v.computeStrides(s);if(a){let p=Ut.computeFlatOffset(t,i);return r==="string"?e.slice(p,p+o):e.subarray(p,p+o)}let l=r==="string"?T.fromUint8ToStringArray(e):e,u=Le(s,r,l),c=Le(n,r);for(let p=0;p<c.size;++p){let d=c.indexToLoc(p),h=d.map((f,m)=>f+t[m]);c.set(u.get(...h),...d)}return r==="string"?T.fromStringArrayToUint8(c.values):c.values}function nl(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{begin:a,size:o}=s;Te(r,"slice");let[i,l]=Ut.parseSliceParams(r,a,o);Ut.assertParamsValid(r,i,l);let u=n.data.get(r.dataId).values,c=_m(u,i,l,r.shape,r.dtype);return n.makeTensorInfo(l,r.dtype,c)}var LX={kernelName:Ol,backendName:"cpu",kernelFunc:nl};function WI(e,t,n,s,r,a,o){let i=t[0],l=a[0],u=new Array(l),c=new Array(i),p=t[1];if(l===0){if(i!==0)throw new Error(T.getSparseFillEmptyRowsIndicesDenseShapeMismatch(i));let g=v.getArrayFromDType(n,0),y=v.getArrayFromDType(r,0);return[g,[0,p],y,u,c]}let d=!0,h=0,f=new Array(l).fill(0);for(let g=0;g<i;++g){let y=e[g*p];if(y<0)throw new Error(T.getSparseFillEmptyRowsNegativeIndexErrorMessage(g,y));if(y>=l)throw new Error(T.getSparseFillEmptyRowsOutOfRangeIndexErrorMessage(g,y,l));++f[y],d=d&&y>=h,h=y}let m=!0;for(let g=0;g<l;++g){let y=f[g]===0;u[g]=y,m=m&&!y,f[g]=Math.max(f[g],1),g>0&&(f[g]+=f[g-1])}if(m&&d){let g=e,y=s;for(let x=0;x<i;++x)c[x]=x;return[g,[i,p],y,u,c]}else{let g=f[l-1],y=v.getArrayFromDType(n,g*p),x=v.getArrayFromDType(r,g),A=new Array(l).fill(0);for(let b=0;b<i;++b){let w=e[b*p],k=A[w],S=(w===0?0:f[w-1])+k;A[w]++;for(let E=0;E<p;++E)y[S*p+E]=e[b*p+E];x[S]=s[b],c[b]=S}for(let b=0;b<l;++b)if(A[b]===0){let k=b===0?0:f[b-1];y[k*p+0]=b;for(let S=1;S<p;++S)y[k*p+S]=0;x[k]=o}return[y,[g,p],x,u,c]}}function VI(e,t,n,s,r){let a=v.sizeFromShape(s),o=t[0],i=r.length,l=[],u=1,c=-1;for(let g=0;g<i;++g){let y=r[g];if(y===-1){if(c!==-1)throw new Error(T.getSparseReshapeMultipleNegativeOneOutputDimErrorMessage(c,g));c=g,l.push(1)}else{if(y<0)throw new Error(T.getSparseReshapeNegativeOutputDimErrorMessage(g,y));u*=y,l.push(y)}}if(c!==-1){if(u<=0)throw new Error(T.getSparseReshapeEmptyTensorZeroOutputDimErrorMessage());let g=Math.trunc(a/u);if(u*g!==a)throw new Error(T.getSparseReshapeInputOutputMultipleErrorMessage(s,l));l[c]=g}if(v.sizeFromShape(l)!==a)throw new Error(T.getSparseReshapeInputOutputMismatchErrorMessage(s,l));let d=s.length,h=[];if(d>0){h[d-1]=1;for(let g=d-2;g>=0;--g)h[g]=h[g+1]*s[g+1]}let f=[];if(i>0){f[i-1]=1;for(let g=i-2;g>=0;--g)f[g]=f[g+1]*l[g+1]}let m=v.getArrayFromDType(n,o*i);for(let g=0;g<o;++g){let y=0;for(let x=0;x<d;++x)y+=e[g*d+x]*h[x];for(let x=0;x<i;++x)m[g*i+x]=Math.trunc(y/f[x]),y%=f[x]}return[m,[o,i],l]}function $x(e,t,n,s,r,a=!1,o=0){let i=s.length,l=[t[0],e.length/t[0]],u=l[1],p=i>0?r[i-1]+1:0;if(p<0)throw new Error(T.getSparseSegmentReductionNegativeSegmentIdsErrorMessage());let d=t.slice();d[0]=p;let h=d.reduce((A,b)=>A*b,1),f=v.getArrayFromDType(n,h);if(i===0)return p>0&&f.fill(o),[f,d];if(p<=0)throw new Error(T.getSparseSegmentReductionNegativeSegmentIdsErrorMessage());let m=0,g=1,y=0,x=r[m];for(;;){let A=0;if(g<i){if(A=r[g],x===A){++g;continue}if(x>=A)throw new Error(T.getSparseSegmentReductionNonIncreasingSegmentIdsErrorMessage())}if(x<0||x>=p)throw new Error(T.getSparseSegmentReductionSegmentIdOutOfRangeErrorMessage(x,p));x>y&&f.fill(o,y*u,x*u);for(let b=m;b<g;++b){let w=s[b];if(w<0||w>=l[0])throw new Error(T.getSparseSegmentReductionIndicesOutOfRangeErrorMessage(b,s[b],l[0]));for(let k=0;k<u;k++)f[x*u+k]+=e[w*u+k]}if(a)for(let b=0;b<u;b++)f[x*u+b]/=g-m;if(m=g,++g,y=x+1,x=A,g>i)break}return y<p&&f.fill(o,y*u,p*u),[f,d]}var BX=di(e=>Math.sqrt(e)),WX=bt(Qo,e=>Math.sqrt(e)),VX={kernelName:Qo,backendName:"cpu",kernelFunc:WX},UI=un((e,t)=>{let n=e-t;return n*n}),UX=kn(ni,UI),GX={kernelName:ni,backendName:"cpu",kernelFunc:UX};function GI(e,t,n,s){let r=Le(e,t.dtype);for(let a=0;a<r.size;a++){let o=r.indexToLoc(a),i=new Array(o.length);for(let l=0;l<i.length;l++)i[l]=o[l]*n[l]+s[l];r.set(t.get(...i),...o)}return r}var HX=class{constructor(e,t,n,s,r,a){this.separator=v.encodeString(e),this.nGramWidths=t,this.leftPad=v.encodeString(n),this.rightPad=v.encodeString(s),this.padWidth=r,this.preserveShort=a}getPadWidth(e){return Math.min(this.padWidth<0?e-1:this.padWidth,e-1)}getNumNGrams(e,t){let n=this.getPadWidth(t);return Math.max(0,e+2*n-t+1)}createNGrams(e,t,n,s,r,a){for(let o=0;o<r;++o){let i=this.getPadWidth(a),l=Math.max(0,i-o),u=Math.max(0,i-(r-(o+1))),c=a-(l+u),p=t+(l>0?0:o-i),d=0;d+=l*this.leftPad.length;for(let y=0;y<c;++y)d+=e[p+y].length;d+=u*this.rightPad.length,d+=(l+u+c-1)*this.separator.length,n[s+o]=new Uint8Array(d);let f=n[s+o],m=0,g=y=>y.forEach(x=>f[m++]=x);for(let y=0;y<l;++y)g(this.leftPad),g(this.separator);for(let y=0;y<c-1;++y)g(e[p+y]),g(this.separator);if(c>0){g(e[p+c-1]);for(let y=0;y<u;++y)g(this.separator),g(this.rightPad)}else{for(let y=0;y<u-1;++y)g(this.rightPad),g(this.separator);g(this.rightPad)}}}compute(e,t){let n=e.length,s=t.length;if(s>0){let i=t[0];if(i!==0)throw new Error(`First split value must be 0, got ${i}`);for(let l=1;l<s;++l){let u=t[l]>=i;if(u=u&&t[l]<=n,!u)throw new Error(`Invalid split value ${t[l]}, must be in [${i}, ${n}]`);i=t[l]}if(i!==n)throw new Error(`Last split value must be data size. Expected ${n}, got ${i}`)}let r=s-1,a=v.getArrayFromDType("int32",s);if(n===0||s===0){let i=new Array(n);for(let l=0;l<=r;++l)a[l]=0;return[i,a]}a[0]=0;for(let i=1;i<=r;++i){let l=t[i]-t[i-1],u=0;this.nGramWidths.forEach(c=>{u+=this.getNumNGrams(l,c)}),this.preserveShort&&l>0&&u===0&&(u=1),a[i]=a[i-1]+u}let o=new Array(a[r]);for(let i=0;i<r;++i){let l=t[i],u=a[i];if(this.nGramWidths.forEach(c=>{let p=t[i+1]-t[i],d=this.getNumNGrams(p,c);this.createNGrams(e,l,o,u,d,c),u+=d}),this.preserveShort&&u===a[i]){let c=t[i+1]-t[i];if(c===0)continue;let p=c+2*this.padWidth,d=1;this.createNGrams(e,l,o,u,d,p)}}return[o,a]}};function Px(e,t,n,s,r,a,o,i){return new HX(n,s,r,a,o,i).compute(e,t)}function jX(e,t,n,s){if(!e.length)return;if(t.length===0){for(let a=0;a<e.length;++a)s.push(e.subarray(a,a+1));return}if(t.length===1){let a=t[0],o=e.indexOf(a);for(;o!==-1;){let i=e.subarray(0,o);(!n||i.length!==0)&&s.push(i),e=e.subarray(o+1),o=e.indexOf(a)}(!n||e.length!==0)&&s.push(e);return}let r=0;for(let a=0;a<e.length+1;a++)if(a===e.length||t.indexOf(e[a])!==-1){let o=e.subarray(r,a);(!n||o.length!==0)&&s.push(o),r=a+1}}function Fx(e,t,n){let s=e.length,r=[],a=0,o=0,i=new Array(s);for(let d=0;d<s;++d){let h=r.length;jX(e[d],t,n,r);let f=r.length-h;i[d]=f,a+=f,o=Math.max(o,f)}let l=v.getArrayFromDType("int32",a*2),u=new Array(a),c=[s,o],p=0;for(let d=0;d<s;++d)for(let h=0;h<i[d];++h)l[p*2]=d,l[p*2+1]=h,u[p]=r[p],++p;return[l,u,c]}function Ox(e,t){let n=v.getArrayFromDType("int32",e.length);for(let s=0;s<e.length;++s)n[s]=v.fingerPrint64(e[s]).modulo(t).getLowBitsUnsigned();return n}var HI=un((e,t)=>e-t),qX=Tx((e,t,n,s)=>({real:e-n,imag:t-s})),Mx=kn(si,HI,qX),XX={kernelName:si,backendName:"cpu",kernelFunc:Mx};function jI(e,t){let n=new Array(e.rank);for(let r=0;r<n.length;r++)n[r]=e.shape[r]*t[r];let s=Le(n,e.dtype);for(let r=0;r<s.values.length;++r){let a=s.indexToLoc(r),o=new Array(e.rank);for(let l=0;l<o.length;l++)o[l]=a[l]%e.shape[l];let i=e.locToIndex(o);s.values[r]=e.values[i]}return s}var ip=(e,t)=>{let n=t.value-e.value;return n===0?e.index-t.index:n};function qI(e,t,n=0,s=e.length-1){for(;s>n;){if(s-n>600){let i=s-n+1,l=t-n+1,u=Math.log(i),c=.5*Math.exp(2*u/3),p=.5*Math.sqrt(u*c*(i-c)/i)*Math.sign(l-i/2),d=Math.max(n,Math.floor(t-l*c/i+p)),h=Math.min(s,Math.floor(t+(i-l)*c/i+p));qI(e,t,d,h)}let r=e[t],a=n,o=s;for(v.swap(e,n,t),ip(e[s],r)>0&&v.swap(e,n,s);a<o;){for(v.swap(e,a,o),a++,o--;ip(e[a],r)<0;)a=a+1;for(;ip(e[o],r)>0;)o=o-1}ip(e[n],r)===0?v.swap(e,n,o):(o=o+1,v.swap(e,o,s)),o<=t&&(n=o+1),t<=o&&(s=o-1)}}function XI(e,t,n,s,r){let a=t[t.length-1],[o,i]=[e.length/a,a],l=v.getTypedArrayFromDType(n,o*s),u=v.getTypedArrayFromDType("int32",o*s);for(let p=0;p<o;p++){let d=p*i,h=e.subarray(d,d+i),f=new Array(h.length);h.forEach((x,A)=>f[A]={value:x,index:A}),s<f.length&&(qI(f,s),f=f.slice(0,s)),r&&f.sort(ip);let m=p*s,g=l.subarray(m,m+s),y=u.subarray(m,m+s);for(let x=0;x<s;x++)g[x]=f[x].value,y[x]=f[x].index}let c=t.slice();return c[c.length-1]=s,[Le(c,n,l),Le(c,"int32",u)]}function KI(e,t,n,s){let r=v.parseAxisParam(t,n)[0],a=[1,n[0],1];for(let f=0;f<r;f++)a[0]*=n[f];a[1]=n[r];for(let f=r+1;f<n.length;f++)a[2]*=n[f];let o={},i=new Int32Array(n[r]),l=new fn(a,s,e),u=[],c=a[0]===1&&a[2]===1;for(let f=0;f<n[r];f++){let m;if(c)m=e[f].toString();else{let g=[];for(let y=0;y<a[0];y++)for(let x=0;x<a[2];x++)g.push(l.get(y,f,x));m=g.join(",")}if(o[m]!==void 0)i[f]=o[m];else{let g=Object.keys(o).length;o[m]=g,i[f]=g,u.push(f)}}let p=a.slice();p[1]=Object.keys(o).length;let d=new fn(p,s);u.forEach((f,m)=>{for(let g=0;g<a[0];g++)for(let y=0;y<a[2];y++)d.set(l.get(g,f,y),g,m,y)});let h=n.slice();return h[r]=p[1],{outputValues:d.values,outputShape:h,indices:i}}ql("cpu",()=>new Sx,1);var ZI=bt(So,e=>e>=0?e:Math.exp(e)-1),KX={kernelName:So,backendName:"cpu",kernelFunc:ZI};function YI(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{alpha:a}=s;Te([r],"leakyRelu");let o=v.sizeFromShape(r.shape),i=n.data.get(r.dataId).values,l=v.getTypedArrayFromDType("float32",o);for(let u=0;u<i.length;u++)l[u]=i[u]<0?a*i[u]:i[u];return n.makeTensorInfo(r.shape,"float32",l)}var ZX={kernelName:Do,backendName:"cpu",kernelFunc:YI},YX=un((e,t)=>e<0?t*e:e);function JI(e){let{inputs:t,backend:n}=e,{x:s,alpha:r}=t;Te([s,r],"prelu");let a=n.data.get(s.dataId).values,o=n.data.get(r.dataId).values,[i,l]=YX(s.shape,r.shape,a,o,"float32");return n.makeTensorInfo(l,"float32",i)}var JX={kernelName:Go,backendName:"cpu",kernelFunc:JI},QI=bt(jo,e=>Math.max(0,e)),QX={kernelName:jo,backendName:"cpu",kernelFunc:QI},eS=bt(Ko,e=>Math.min(Math.max(0,e),6)),eK={kernelName:Ko,backendName:"cpu",kernelFunc:eS};function Dm(e,t,n,s,r){if(n==="linear")return Zr({inputs:{x:t},backend:e});if(n==="relu")return QI({inputs:{x:t},backend:e});if(n==="elu")return ZI({inputs:{x:t},backend:e});if(n==="relu6")return eS({inputs:{x:t},backend:e});if(n==="prelu")return JI({inputs:{x:t,alpha:s},backend:e});if(n==="leakyrelu")return YI({inputs:{x:t},backend:e,attrs:{alpha:r}});if(n==="sigmoid")return BI({inputs:{x:t},backend:e});throw new Error(`Activation ${n} has not been implemented for the CPU backend.`)}function Rt(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{shape:a}=s,o=v.sizeFromShape(r.shape),i=v.inferFromImplicitShape(a,o),l=v.sizeFromShape(i);v.assert(o===l,()=>`The new shape (${i}) has ${l} elements and the old shape (${r.shape}) has ${o} elements. The new shape and old shape must have the same number of elements.`),n.incRef(r.dataId);let u=n.data.get(r.dataId);if(u.complexTensorInfos!=null){let c=u.complexTensorInfos.real,p=u.complexTensorInfos.imag;c.shape=i,p.shape=i}return{dataId:r.dataId,shape:i,dtype:r.dtype}}var tK={kernelName:_l,backendName:"cpu",kernelFunc:Rt};function tS(e){let{inputs:t,backend:n,attrs:s}=e,{a:r,b:a}=t,{transposeA:o,transposeB:i}=s;Te([r,a],"matMul");let l=r.shape.length,u=a.shape.length,c=o?r.shape[l-2]:r.shape[l-1],p=i?a.shape[u-1]:a.shape[u-2],d=o?r.shape[l-1]:r.shape[l-2],h=i?a.shape[u-2]:a.shape[u-1],f=r.shape.slice(0,-2),m=a.shape.slice(0,-2),g=v.sizeFromShape(f),y=v.sizeFromShape(m),A=Xl.assertAndGetBroadcastShape(r.shape.slice(0,-2),a.shape.slice(0,-2)).concat([d,h]);v.assert(c===p,()=>`Error in matMul: inner shapes (${c}) and (${p}) of Tensors with shapes ${r.shape} and ${a.shape} and transposeA=${o} and transposeB=${i} must match.`);let b=o?[g,c,d]:[g,d,c],w=i?[y,h,p]:[y,p,h],k=Rt({inputs:{x:r},backend:n,attrs:{shape:b}}),S=Rt({inputs:{x:a},backend:n,attrs:{shape:w}}),E=o?k.shape[1]:k.shape[2],R=o?k.shape[2]:k.shape[1],$=i?S.shape[1]:S.shape[2],_=Math.max(g,y),P=n.data.get(k.dataId).values,C=n.data.get(S.dataId).values,F=v.computeStrides(k.shape),G=v.computeStrides(S.shape),[K,z,Z]=o?[F[0],1,F[1]]:[F[0],F[1],1],[J,te,B]=i?[1,G[1],G[0]]:[G[1],1,G[0]],oe=R*$,Q=Le([_,R,$],k.dtype),ae=Q.values,ie=n.blockSize;for(let me=0;me<_;me++)for(let we=0;we<R;we+=ie)for(let Re=0;Re<$;Re+=ie)for(let _e=0;_e<E;_e+=ie){let Be=Math.min(we+ie,R),He=Math.min(Re+ie,$),ot=Math.min(_e+ie,E);for(let pt=we;pt<Be;pt++)for(let ht=Re;ht<He;ht++){let At=0;for(let $e=_e;$e<ot;$e++){let Tt=Math.min(me,g-1)*K,It=Math.min(me,y-1)*B,Gn=P[Tt+pt*z+$e*Z],Qt=C[$e*J+ht*te+It];At+=Gn*Qt}ae[me*oe+(pt*$+ht)]+=At}}return n.disposeIntermediateTensorInfo(k),n.disposeIntermediateTensorInfo(S),n.makeTensorInfo(A,Q.dtype,Q.values)}var nK={kernelName:mo,backendName:"cpu",kernelFunc:tS};function sK(e){let{inputs:t,backend:n,attrs:s}=e,{a:r,b:a,bias:o,preluActivationWeights:i}=t,{transposeA:l,transposeB:u,activation:c,leakyreluAlpha:p}=s,d,h,f,m=[];d=tS({inputs:{a:r,b:a},attrs:{transposeA:l,transposeB:u},backend:n}),o&&(h=oc({inputs:{a:d,b:o},backend:n}),m.push(d),d=h),c&&(f=Dm(n,d,c,i,p),m.push(d),d=f);for(let y of m)n.disposeIntermediateTensorInfo(y);return d}var rK={kernelName:Ja,backendName:"cpu",kernelFunc:sK},aK=bt(pc,e=>Math.acos(e)),oK={kernelName:pc,backendName:"cpu",kernelFunc:aK},iK=bt(hc,e=>Math.acosh(e)),lK={kernelName:hc,backendName:"cpu",kernelFunc:iK};function uK(e){let{inputs:t,backend:n}=e,s=t;Te(t,"addN");let r=s.map(i=>n.data.get(i.dataId).values),a=Le(s[0].shape,s[0].dtype),o=a.values;for(let i=0;i<s.length;i++){let l=r[i];for(let u=0;u<o.length;u++)o[u]+=l[u]}return n.makeTensorInfo(a.shape,a.dtype,a.values)}var cK={kernelName:po,backendName:"cpu",kernelFunc:uK};function dK(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;Te(r,"all");let i=v.parseAxisParam(a,r.shape),l=i,u=T.getAxesPermutation(l,r.shape.length),c=r;u!=null&&(c=gs({inputs:{x:r},backend:n,attrs:{perm:u}}),l=T.getInnerMostAxes(l.length,r.shape.length)),T.assertAxesAreInnerMostDims("all",l,c.shape.length);let[p,d]=T.computeOutAndReduceShapes(c.shape,l),h=v.sizeFromShape(d),f=v.makeZerosTypedArray(v.sizeFromShape(p),c.dtype),m=n.data.get(c.dataId).values;for(let y=0;y<f.length;++y){let x=y*h,A=m[x];for(let b=0;b<h;++b){let w=m[x+b];A=A&&w}f[y]=A}u!=null&&n.disposeIntermediateTensorInfo(c);let g=n.makeTensorInfo(p,c.dtype,f);if(o){let y=T.expandShapeToKeepDim(p,i),x=Rt({inputs:{x:g},backend:n,attrs:{shape:y}});return n.disposeIntermediateTensorInfo(g),x}return g}var pK={kernelName:fc,backendName:"cpu",kernelFunc:dK};function hK(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;Te(r,"any");let i=v.parseAxisParam(a,r.shape),l=i,u=T.getAxesPermutation(l,r.shape.length),c=r;u!=null&&(c=gs({inputs:{x:r},backend:n,attrs:{perm:u}}),l=T.getInnerMostAxes(l.length,r.shape.length)),T.assertAxesAreInnerMostDims("any",l,c.shape.length);let[p,d]=T.computeOutAndReduceShapes(c.shape,l),h=v.sizeFromShape(d),f=v.makeZerosTypedArray(v.sizeFromShape(p),c.dtype),m=n.data.get(c.dataId).values;for(let y=0;y<f.length;++y){let x=y*h,A=m[x];for(let b=0;b<h;++b){let w=m[x+b];A=A||w}f[y]=A}u!=null&&n.disposeIntermediateTensorInfo(c);let g=n.makeTensorInfo(p,c.dtype,f);if(o){let y=T.expandShapeToKeepDim(p,i),x=Rt({inputs:{x:g},backend:n,attrs:{shape:y}});return n.disposeIntermediateTensorInfo(g),x}return g}var fK={kernelName:mc,backendName:"cpu",kernelFunc:hK};function mK(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a}=s;Te(r,"argMax");let o=v.parseAxisParam(a,r.shape),i=T.getAxesPermutation(o,r.shape.length),l=r,u=[];i!=null&&(l=gs({inputs:{x:r},backend:n,attrs:{perm:i}}),u.push(l),o=T.getInnerMostAxes(o.length,l.shape.length)),o=[o[0]],T.assertAxesAreInnerMostDims("argMax",o,l.shape.length);let[c,p]=T.computeOutAndReduceShapes(l.shape,o),d=v.sizeFromShape(c),h=v.makeZerosTypedArray(d,"int32"),f=v.sizeFromShape(p),m=n.data.get(l.dataId).values;for(let g=0;g<h.length;++g){let y=g*f,x=m[y],A=0;for(let b=0;b<f;++b){let w=m[y+b];w>x&&(x=w,A=b)}h[g]=A}return u.forEach(g=>n.disposeIntermediateTensorInfo(g)),n.makeTensorInfo(c,"int32",h)}var gK={kernelName:ho,backendName:"cpu",kernelFunc:mK};function yK(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a}=s;Te(r,"argMin");let o=v.parseAxisParam(a,r.shape),i=T.getAxesPermutation(o,r.shape.length),l=r,u=[];i!=null&&(l=gs({inputs:{x:r},backend:n,attrs:{perm:i}}),u.push(l),o=T.getInnerMostAxes(o.length,l.shape.length)),o=[o[0]],T.assertAxesAreInnerMostDims("argMin",o,l.shape.length);let[c,p]=T.computeOutAndReduceShapes(l.shape,o),d=v.sizeFromShape(c),h=v.makeZerosTypedArray(d,"int32"),f=v.sizeFromShape(p),m=n.data.get(l.dataId).values;for(let g=0;g<h.length;++g){let y=g*f,x=m[y],A=0;for(let b=0;b<f;++b){let w=m[y+b];w<x&&(x=w,A=b)}h[g]=A}return u.forEach(g=>n.disposeIntermediateTensorInfo(g)),n.makeTensorInfo(c,"int32",h)}var AK={kernelName:gc,backendName:"cpu",kernelFunc:yK},xK=bt(yc,e=>Math.asin(e)),bK={kernelName:yc,backendName:"cpu",kernelFunc:xK},vK=bt(Ac,e=>Math.asinh(e)),wK={kernelName:Ac,backendName:"cpu",kernelFunc:vK},kK=bt(xc,e=>Math.atan(e)),IK={kernelName:xc,backendName:"cpu",kernelFunc:kK},SK=un((e,t)=>Math.atan2(e,t)),CK=kn(vc,SK),TK={kernelName:vc,backendName:"cpu",kernelFunc:CK},NK=bt(bc,e=>Math.atanh(e)),EK={kernelName:bc,backendName:"cpu",kernelFunc:NK};function zx(e,t,n,s,r,a){let o=r.strideHeight,i=r.strideWidth,l=r.dilationHeight,u=r.dilationWidth,c=r.effectiveFilterHeight,p=r.effectiveFilterWidth,d=r.padInfo.top,h=r.padInfo.left,f=a==="max"?Number.NEGATIVE_INFINITY:Number.POSITIVE_INFINITY,m=Le(r.outShape,n),g=m.values,y=r.outShape[1]*r.outShape[2]*r.outShape[3],x=r.outShape[2]*r.outShape[3],A=r.outShape[3];for(let b=0;b<r.batchSize;++b){let w=b*y,k=b*s[0];for(let S=0;S<r.inChannels;++S)for(let E=0;E<r.outHeight;++E){let R=E*o-d,$=Math.max(0,R),_=Math.min(r.inHeight,c+R),P=w+E*x;for(let C=0;C<r.outWidth;++C){let F=C*i-h,G=Math.max(0,F),K=Math.min(r.inWidth,p+F),z=f,Z=0,J=0;for(let B=$;B<_;B+=l){let oe=k+B*s[1];for(let Q=G;Q<K;Q+=u){let ae=oe+Q*s[2],ie=e[ae+S];a==="max"&&ie>z?z=ie:a==="avg"&&(Z+=ie,J++)}if(isNaN(z))break}let te=P+C*A+S;g[te]=a==="avg"?Z/J:z}}}return m}function nS(e,t,n,s,r=!1,a=!1){let o=Le(s.outShape,"int32"),i=s.strideHeight,l=s.strideWidth,u=s.dilationHeight,c=s.dilationWidth,p=s.effectiveFilterHeight,d=s.effectiveFilterWidth,h=s.padInfo.top,f=s.padInfo.left,m=Le(t,n,e);for(let g=0;g<s.batchSize;++g)for(let y=0;y<s.inChannels;++y)for(let x=0;x<s.outHeight;++x){let A=x*i-h,b=A;for(;b<0;)b+=u;let w=Math.min(s.inHeight,p+A);for(let k=0;k<s.outWidth;++k){let S=k*l-f,E=S;for(;E<0;)E+=c;let R=Math.min(s.inWidth,d+S),$=Number.NEGATIVE_INFINITY,_=-1;for(let P=b;P<w;P+=u){let C=P-A;for(let F=E;F<R;F+=c){let G=F-S,K=m.get(g,P,F,y);K>$&&($=K,r?_=a?((g*s.inHeight+P)*s.inWidth+F)*s.inChannels+y:(P*s.inWidth+F)*s.inChannels+y:_=C*d+G)}}o.set(_,g,x,k,y)}}return o}function sS(e,t,n,s,r,a){let o=r.strideDepth,i=r.strideHeight,l=r.strideWidth,u=r.dilationDepth,c=r.dilationHeight,p=r.dilationWidth,d=r.effectiveFilterDepth,h=r.effectiveFilterHeight,f=r.effectiveFilterWidth,m=r.padInfo.front,g=r.padInfo.top,y=r.padInfo.left,x=a==="max"?Number.NEGATIVE_INFINITY:Number.POSITIVE_INFINITY,A=Le(r.outShape,n),b=A.values,w=r.outShape[1]*r.outShape[2]*r.outShape[3]*r.outShape[4],k=r.outShape[2]*r.outShape[3]*r.outShape[4],S=r.outShape[3]*r.outShape[4],E=r.outShape[4];for(let R=0;R<r.batchSize;++R){let $=R*w,_=R*s[0];for(let P=0;P<r.inChannels;++P)for(let C=0;C<r.outDepth;++C){let F=C*o-m,G=F;for(;G<0;)G+=u;let K=Math.min(r.inDepth,d+F),z=$+C*k;for(let Z=0;Z<r.outHeight;++Z){let J=Z*i-g,te=J;for(;te<0;)te+=c;let B=Math.min(r.inHeight,h+J),oe=z+Z*S;for(let Q=0;Q<r.outWidth;++Q){let ae=Q*l-y,ie=ae;for(;ie<0;)ie+=p;let me=Math.min(r.inWidth,f+ae),we=oe+Q*E,Re=x,_e=0,Be=0;for(let ot=G;ot<K;ot+=u){let pt=_+ot*s[1];for(let ht=te;ht<B;ht+=c){let At=pt+ht*s[2];for(let $e=ie;$e<me;$e+=p){let Tt=At+$e*s[3],It=e[Tt+P];if(a==="max"&&It>Re?Re=It:a==="avg"&&(_e+=It,Be++),isNaN(Re))break}if(isNaN(Re))break}if(isNaN(Re))break}let He=we+P;b[He]=a==="avg"?_e/Be:Re}}}}return A}function RK(e,t){let n=Le(t.outShape,"int32"),s=t.strideDepth,r=t.strideHeight,a=t.strideWidth,o=t.dilationDepth,i=t.dilationHeight,l=t.dilationWidth,u=t.effectiveFilterDepth,c=t.effectiveFilterHeight,p=t.effectiveFilterWidth,d=t.padInfo.front,h=t.padInfo.top,f=t.padInfo.left;for(let m=0;m<t.batchSize;++m)for(let g=0;g<t.inChannels;++g)for(let y=0;y<t.outDepth;++y){let x=y*s-d,A=x;for(;A<0;)A+=o;let b=Math.min(t.inDepth,u+x);for(let w=0;w<t.outHeight;++w){let k=w*r-h,S=k;for(;S<0;)S+=i;let E=Math.min(t.inHeight,c+k);for(let R=0;R<t.outWidth;++R){let $=R*a-f,_=$;for(;_<0;)_+=l;let P=Math.min(t.inWidth,p+$),C=Number.NEGATIVE_INFINITY,F=-1;for(let G=A;G<b;G+=o){let K=G-x;for(let z=S;z<E;z+=i){let Z=z-k;for(let J=_;J<P;J+=l){let te=J-$,B=e.get(m,G,z,J,g);B>=C&&(C=B,F=K*c*p+Z*c+te)}}}n.set(F,m,y,w,R,g)}}}return n}function _K(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t;Te(r,"avgPool");let{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=s,u=1;v.assert(T.eitherStridesOrDilationsAreOne(o,u),()=>`Error in avgPool: Either strides or dilations must be 1. Got strides ${o} and dilations '${u}'`);let c=T.computePool2DInfo(r.shape,a,o,u,i,l),p;if(c.filterWidth===1&&c.filterHeight===1&&v.arraysEqual(c.inShape,c.outShape))p=Zr({inputs:{x:r},backend:n});else{let d=n.data.get(r.dataId).values,h=v.computeStrides(r.shape),f=zx(d,r.shape,r.dtype,h,c,"avg");p=n.makeTensorInfo(c.outShape,r.dtype,f.values)}return p}var DK={kernelName:fo,backendName:"cpu",kernelFunc:_K};function $K(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l,dataFormat:u}=s;Te(r,"avgPool3d");let c=T.computePool3DInfo(r.shape,a,o,1,i,l,u),p=n.data.get(r.dataId).values,d=sS(p,r.shape,r.dtype,v.computeStrides(r.shape),c,"avg");return n.makeTensorInfo(d.shape,"float32",d.values)}var PK={kernelName:zp,backendName:"cpu",kernelFunc:$K};function FK(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,input:a}=t,{filterSize:o,strides:i,pad:l,dimRoundingMode:u}=s;Te([r,a],"avgPool3DGrad");let c=T.computePool3DInfo(a.shape,o,i,1,l,u),p=c.strideDepth,d=c.strideHeight,h=c.strideWidth,f=c.filterDepth,m=c.filterHeight,g=c.filterWidth,y=c.dilationDepth,x=c.dilationHeight,A=c.dilationWidth,b=c.effectiveFilterDepth,w=c.effectiveFilterHeight,k=c.effectiveFilterWidth,S=b-1-c.padInfo.front,E=k-1-c.padInfo.left,R=w-1-c.padInfo.top,$=Le(a.shape,"float32"),_=1/(f*m*g),P=n.bufferSync(r);for(let C=0;C<c.batchSize;++C)for(let F=0;F<c.inChannels;++F)for(let G=0;G<c.inDepth;++G)for(let K=0;K<c.inHeight;++K)for(let z=0;z<c.inWidth;++z){let Z=G-S,J=K-R,te=z-E,B=0;for(let oe=0;oe<b;oe+=y){let Q=(Z+oe)/p;if(!(Q<0||Q>=c.outDepth||Math.floor(Q)!==Q))for(let ae=0;ae<w;ae+=x){let ie=(J+ae)/d;if(!(ie<0||ie>=c.outHeight||Math.floor(ie)!==ie))for(let me=0;me<k;me+=A){let we=(te+me)/h;if(we<0||we>=c.outWidth||Math.floor(we)!==we)continue;B+=P.get(C,Q,ie,we,F)}}}$.set(B*_,C,G,K,z,F)}return n.makeTensorInfo($.shape,$.dtype,$.values)}var OK={kernelName:Hm,backendName:"cpu",kernelFunc:FK};function MK(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,input:a}=t,o=a;Te([r,a],"avgPoolGrad");let{filterSize:i,strides:l,pad:u}=s,c=T.computePool2DInfo(o.shape,i,l,1,u),p=c.strideHeight,d=c.strideWidth,h=c.filterHeight,f=c.filterWidth,m=c.dilationHeight,g=c.dilationWidth,y=c.effectiveFilterHeight,x=c.effectiveFilterWidth,A=x-1-c.padInfo.left,b=y-1-c.padInfo.top,w=Le(o.shape,"float32"),k=1/(h*f),S=n.data.get(r.dataId).values,E=Le(r.shape,"float32",S);for(let R=0;R<c.batchSize;++R)for(let $=0;$<c.inChannels;++$)for(let _=0;_<c.inHeight;++_)for(let P=0;P<c.inWidth;++P){let C=_-b,F=P-A,G=0;for(let K=0;K<y;K+=m){let z=(C+K)/p;if(!(z<0||z>=c.outHeight||Math.floor(z)!==z))for(let Z=0;Z<x;Z+=g){let J=(F+Z)/d;if(J<0||J>=c.outWidth||Math.floor(J)!==J)continue;G+=E.get(R,z,J,$)}}w.set(G*k,R,_,P,$)}return n.makeTensorInfo(w.shape,w.dtype,w.values)}var zK={kernelName:Gm,backendName:"cpu",kernelFunc:MK};function LK(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,scale:a,offset:o,mean:i,variance:l}=t;v.assert(i.shape.length===l.shape.length,()=>"Batch normalization gradient requires mean and variance to have equal ranks."),v.assert(o==null||i.shape.length===o.shape.length,()=>"Batch normalization gradient requires mean and offset to have equal ranks."),v.assert(a==null||i.shape.length===a.shape.length,()=>"Batch normalization gradient requires mean and scale to have equal ranks."),Te([r,i,l,a,o],"batchNorm");let{varianceEpsilon:u}=s;u==null&&(u=.001);let c=n.data.get(r.dataId).values,p=n.data.get(i.dataId).values,d=n.data.get(l.dataId).values,h=a?n.data.get(a.dataId).values:new Float32Array([1]),f=o?n.data.get(o.dataId).values:new Float32Array([0]),m=new Float32Array(c.length),g=f.length,y=h.length,x=d.length,A=p.length,b=0,w=0,k=0,S=0;for(let E=0;E<c.length;++E)m[E]=f[b++]+(c[E]-p[w++])*h[k++]/Math.sqrt(d[S++]+u),b>=g&&(b=0),w>=A&&(w=0),k>=y&&(k=0),S>=x&&(S=0);return n.makeTensorInfo(r.shape,r.dtype,m)}var BK={kernelName:Eo,backendName:"cpu",kernelFunc:LK};function WK(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockShape:a,crops:o}=s;Te([r],"batchToSpaceND");let i=a.reduce((y,x)=>y*x),l=T.getReshaped(r.shape,a,i),u=T.getPermuted(l.length,a.length),c=T.getReshapedPermuted(r.shape,a,i),p=T.getSliceBeginCoords(o,a.length),d=T.getSliceSize(c,o,a.length),h=Rt({inputs:{x:r},backend:n,attrs:{shape:l}}),f=gs({inputs:{x:h},backend:n,attrs:{perm:u}}),m=Rt({inputs:{x:f},backend:n,attrs:{shape:c}}),g=nl({inputs:{x:m},backend:n,attrs:{begin:p,size:d}});return n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(f),n.disposeIntermediateTensorInfo(m),g}var VK={kernelName:ll,backendName:"cpu",kernelFunc:WK};function UK(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,weights:a}=t,{size:o}=s,i=n.data.get(r.dataId).values,l=n.data.get(a.dataId).values,u=Nx(i,l,a.dtype,a.shape,o);return n.makeTensorInfo([o],a.dtype,u)}var GK={kernelName:jm,backendName:"cpu",kernelFunc:UK};function HK(e){let{inputs:t,backend:n}=e,{s0:s,s1:r}=t,a=n.data.get(s.dataId).values,o=n.data.get(r.dataId).values,i=T.assertAndGetBroadcastShape(Array.from(a),Array.from(o));return n.makeTensorInfo([i.length],"int32",Int32Array.from(i))}var jK={kernelName:qm,backendName:"cpu",kernelFunc:HK},qK=bt(ba,(e,t)=>{let n=t;return e>n.clipValueMax?n.clipValueMax:e<n.clipValueMin?n.clipValueMin:e}),XK={kernelName:ba,backendName:"cpu",kernelFunc:qK},KK=e=>{let{x:t}=e.inputs,n=e.backend,s=new Float32Array(v.sizeFromShape(t.shape)),r=n.data.get(t.dataId),a=r.complexTensorInfos.real,o=r.complexTensorInfos.imag,i=n.data.get(a.dataId).values,l=n.data.get(o.dataId).values;for(let u=0;u<i.length;u++){let c=i[u],p=l[u];s[u]=Math.hypot(c,p)}return n.makeOutput(s,t.shape,"float32")},ZK={kernelName:Bp,backendName:"cpu",kernelFunc:KK};function ic(e){let{inputs:t,backend:n}=e,{input:s}=t,r=n.data.get(s.dataId).complexTensorInfos.imag,a=n.data.get(r.dataId).values;return n.makeTensorInfo(r.shape,r.dtype,a)}var YK={kernelName:Gp,backendName:"cpu",kernelFunc:ic};function lc(e){let{inputs:t,backend:n,attrs:s}=e,{axis:r}=s,a=v.parseAxisParam(r,t[0].shape)[0],o=T.computeOutShape(t.map(m=>m.shape),a);if(v.sizeFromShape(o)===0)return n.makeTensorInfo(o,t[0].dtype,[]);let i=t.filter(m=>v.sizeFromShape(m.shape)>0);if(i.length===1)return Zr({inputs:{x:i[0]},backend:n});let l=i.map(m=>m.shape);if(T.assertParamsConsistent(l,a),i[0].dtype==="complex64"){let m=i.map(b=>tl({inputs:{input:b},backend:n})),g=i.map(b=>ic({inputs:{input:b},backend:n})),y=lc({inputs:m,backend:n,attrs:{axis:a}}),x=lc({inputs:g,backend:n,attrs:{axis:a}}),A=Cs({inputs:{real:y,imag:x},backend:n});return m.forEach(b=>n.disposeIntermediateTensorInfo(b)),g.forEach(b=>n.disposeIntermediateTensorInfo(b)),n.disposeIntermediateTensorInfo(y),n.disposeIntermediateTensorInfo(x),A}let u=i.map(m=>{let g=v.sizeFromShape(m.shape.slice(a));return Rt({inputs:{x:m},backend:n,attrs:{shape:[-1,g]}})}),c=u.map(m=>({vals:n.data.get(m.dataId).values,shape:m.shape}));o=T.computeOutShape(u.map(m=>m.shape),1);let p=u[0].shape[0]===1,d=Ex(c,o,t[0].dtype,p),h=T.computeOutShape(i.map(m=>m.shape),a),f=n.makeTensorInfo(h,t[0].dtype,d);return u.forEach(m=>n.disposeIntermediateTensorInfo(m)),f}var JK={kernelName:ul,backendName:"cpu",kernelFunc:lc};function rS(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dataFormat:l,dilations:u,dimRoundingMode:c}=s;Te([r,a],"conv2d");let p=T.convertConv2DDataFormat(l),d=T.computeConv2DInfo(r.shape,a.shape,o,u,i,c,!1,p),h=d.filterHeight,f=d.filterWidth,m=d.dilationHeight,g=d.dilationWidth,y=d.padInfo.left,x=d.padInfo.top,A=d.dataFormat==="channelsLast",b=new fn(d.outShape,r.dtype),w=v.computeStrides(r.shape),k=v.computeStrides(a.shape),S=w[0],E=A?w[1]:w[2],R=A?w[2]:1,$=A?1:w[1],_=b.strides[0],P=A?b.strides[1]:b.strides[2],C=A?b.strides[2]:1,F=A?1:b.strides[1],G=n.data.get(r.dataId).values,K=n.data.get(a.dataId).values,z=b.values;for(let Z=0;Z<d.batchSize;++Z){let J=Z*S,te=Z*_;for(let B=0;B<d.outHeight;++B){let oe=te+B*P,Q=B*d.strideHeight-x;for(let ae=0;ae<h;++ae){let ie=Q+ae*m;if(ie<0||ie>=d.inHeight)continue;let me=ae*k[0],we=J+ie*E;for(let Re=0;Re<d.outWidth;++Re){let _e=oe+Re*C,Be=Re*d.strideWidth-y;for(let He=0;He<f;++He){let ot=Be+He*g;if(ot<0||ot>=d.inWidth)continue;let pt=me+He*k[1],ht=we+ot*R,At=pt;for(let $e=0;$e<d.inChannels;++$e){let Tt=G[ht+$e*$];for(let It=0;It<d.outChannels;++It)z[_e+It*F]+=Tt*K[At+It];At+=d.outChannels}}}}}}return n.makeTensorInfo(b.shape,b.dtype,z)}var QK={kernelName:Ao,backendName:"cpu",kernelFunc:rS};function eZ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,dy:a}=t,{strides:o,pad:i,dataFormat:l,dimRoundingMode:u,filterShape:c}=s;Te([r,a],"conv2dBackpropFilter");let p=T.convertConv2DDataFormat(l),d=T.computeConv2DInfo(r.shape,c,o,1,i,u,!1,p),{strideHeight:h,strideWidth:f,filterHeight:m,filterWidth:g}=d,y=d.dataFormat==="channelsLast",x=new fn(d.filterShape,"float32"),A=d.padInfo.left,b=d.padInfo.top,w=n.data.get(r.dataId).values,k=n.data.get(a.dataId).values,S=new fn(r.shape,r.dtype,w),E=new fn(a.shape,a.dtype,k);for(let R=0;R<m;++R){let $=Math.max(0,Math.ceil((b-R)/h)),_=Math.min(d.outHeight,(d.inHeight+b-R)/h);for(let P=0;P<g;++P){let C=Math.max(0,Math.ceil((A-P)/f)),F=Math.min(d.outWidth,(d.inWidth+A-P)/f);for(let G=0;G<d.inChannels;++G)for(let K=0;K<d.outChannels;++K){let z=0;for(let Z=0;Z<d.batchSize;++Z)for(let J=$;J<_;++J){let te=R+J*h-b;for(let B=C;B<F;++B){let oe=P+B*f-A;y?z+=S.get(Z,te,oe,G)*E.get(Z,J,B,K):z+=S.get(Z,G,te,oe)*E.get(Z,K,J,B)}}x.set(z,R,P,G,K)}}}return n.makeTensorInfo(x.shape,x.dtype,x.values)}var tZ={kernelName:Xm,backendName:"cpu",kernelFunc:eZ};function nZ(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,filter:a}=t,{inputShape:o,strides:i,pad:l,dataFormat:u,dimRoundingMode:c}=s;Te([r,a],"conv2dBackpropInput");let p=v.computeStrides(a.shape),d=v.computeStrides(r.shape),h=T.convertConv2DDataFormat(u),f=T.computeConv2DInfo(o,a.shape,i,1,l,c,!1,h),m=new fn(f.inShape,"float32"),g=m.values,y=n.data.get(r.dataId).values,x=n.data.get(a.dataId).values,[A,b,w]=p,{batchSize:k,filterHeight:S,filterWidth:E,inChannels:R,inHeight:$,inWidth:_,outChannels:P,outHeight:C,outWidth:F,strideHeight:G,strideWidth:K}=f;h=f.dataFormat;let z=S-1-f.padInfo.top,Z=E-1-f.padInfo.left,J=h==="channelsLast",te=m.strides[0],B=J?m.strides[1]:m.strides[2],oe=J?m.strides[2]:1,Q=J?1:m.strides[1],ae=d[0],ie=J?d[1]:d[2],me=J?d[2]:1,we=J?1:d[1];for(let Re=0;Re<k;++Re)for(let _e=0;_e<R;++_e)for(let Be=0;Be<$;++Be){let He=Be-z,ot=Math.max(0,Math.ceil(He/G)),pt=Math.min(C,(S+He)/G);for(let ht=0;ht<_;++ht){let At=ht-Z,$e=Math.max(0,Math.ceil(At/K)),Tt=Math.min(F,(E+At)/K),It=0;for(let Qt=ot;Qt<pt;++Qt){let bs=Qt*G-He;for(let pn=$e;pn<Tt;++pn){let Hn=pn*K-At,vs=ae*Re+ie*Qt+me*pn,ws=A*(S-1-bs)+b*(E-1-Hn)+w*_e;for(let Pn=0;Pn<P;++Pn){let Us=y[vs+we*Pn],jn=x[ws+Pn];It+=Us*jn}}}let Gn=te*Re+B*Be+oe*ht+Q*_e;g[Gn]=It}}return n.makeTensorInfo(m.shape,m.dtype,m.values)}var sZ={kernelName:xo,backendName:"cpu",kernelFunc:nZ};function rZ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dilations:l}=s;Te([r,a],"conv3d");let u=T.computeConv3DInfo(r.shape,a.shape,o,l,i),{filterDepth:c,filterHeight:p,filterWidth:d,dilationDepth:h,dilationHeight:f,dilationWidth:m,padInfo:g}=u,y=g.front,x=g.left,A=g.top,b=new fn(u.outShape,r.dtype),w=n.data.get(r.dataId).values,k=n.data.get(a.dataId).values,S=b.values,E=v.computeStrides(r.shape),R=v.computeStrides(a.shape);for(let $=0;$<u.batchSize;++$){let _=$*E[0],P=$*b.strides[0];for(let C=0;C<u.outDepth;++C){let F=P+C*b.strides[1],G=C*u.strideDepth-y;for(let K=0;K<c;++K){let z=G+K*h;if(z<0||z>=u.inDepth)continue;let Z=K*R[0],J=_+z*E[1];for(let te=0;te<u.outHeight;++te){let B=F+te*b.strides[2],oe=te*u.strideHeight-A;for(let Q=0;Q<p;++Q){let ae=oe+Q*f;if(ae<0||ae>=u.inHeight)continue;let ie=Z+Q*R[1],me=J+ae*E[2];for(let we=0;we<u.outWidth;++we){let Re=B+we*u.outChannels,_e=we*u.strideWidth-x;for(let Be=0;Be<d;++Be){let He=_e+Be*m;if(He<0||He>=u.inWidth)continue;let ot=ie+Be*R[2],pt=me+He*u.inChannels,ht=ot;for(let At=0;At<u.inChannels;++At){let $e=w[pt+At];for(let Tt=0;Tt<u.outChannels;++Tt)S[Re+Tt]+=$e*k[ht+Tt];ht+=u.outChannels}}}}}}}}return n.makeTensorInfo(b.shape,b.dtype,b.values)}var aZ={kernelName:Wp,backendName:"cpu",kernelFunc:rZ};function oZ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,dy:a}=t,{strides:o,pad:i,filterShape:l}=s;Te([r,a],"conv3dBackpropFilterV2");let u=v.computeStrides(r.shape),c=v.computeStrides(a.shape),p=T.computeConv3DInfo(r.shape,l,o,1,i),d=p.strideDepth,h=p.strideHeight,f=p.strideWidth,m=p.filterDepth,g=p.filterHeight,y=p.filterWidth,x=new fn(p.filterShape,"float32"),A=x.values,[b,w,k,S]=x.strides,E=n.data.get(a.dataId).values,[R,$,_,P]=c,C=n.data.get(r.dataId).values,[F,G,K,z]=u,Z=p.padInfo.front,J=p.padInfo.left,te=p.padInfo.top;for(let B=0;B<m;++B){let oe=Math.max(0,Math.ceil((Z-B)/d)),Q=Math.min(p.outDepth,(p.inDepth+Z-B)/d),ae=B*b;for(let ie=0;ie<g;++ie){let me=Math.max(0,Math.ceil((te-ie)/h)),we=Math.min(p.outHeight,(p.inHeight+te-ie)/h),Re=ie*w+ae;for(let _e=0;_e<y;++_e){let Be=Math.max(0,Math.ceil((J-_e)/f)),He=Math.min(p.outWidth,(p.inWidth+J-_e)/f),ot=_e*k+Re;for(let pt=0;pt<p.inChannels;++pt){let ht=pt*S+ot;for(let At=0;At<p.outChannels;++At){let $e=0;for(let Tt=0;Tt<p.batchSize;++Tt){let It=Tt*F,Gn=Tt*R;for(let Qt=oe;Qt<Q;++Qt){let pn=(B+Qt*d-Z)*G+It,Hn=Qt*$+Gn;for(let vs=me;vs<we;++vs){let Pn=(ie+vs*h-te)*K+pn,Us=vs*_+Hn;for(let jn=Be;jn<He;++jn){let aa=(_e+jn*f-J)*z+Pn,xu=jn*P+Us;$e+=C[aa+pt]*E[xu+At]}}}}A[ht+At]=$e}}}}}return n.makeTensorInfo(x.shape,x.dtype,x.values)}var iZ={kernelName:Km,backendName:"cpu",kernelFunc:oZ};function lZ(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,filter:a}=t,{pad:o,strides:i,inputShape:l}=s;Te([r],"conv3dBackpropInputV2");let u=v.computeStrides(r.shape),c=v.computeStrides(a.shape),p=T.computeConv3DInfo(l,a.shape,i,1,o),d=new fn(p.inShape,"float32"),h=d.values,[f,m,g,y]=d.strides,x=n.data.get(r.dataId).values,[A,b,w,k]=u,S=n.data.get(a.dataId).values,[E,R,$,_]=c,{batchSize:P,filterDepth:C,filterHeight:F,filterWidth:G,inChannels:K,inDepth:z,inHeight:Z,inWidth:J,outChannels:te,outDepth:B,outHeight:oe,outWidth:Q,strideDepth:ae,strideHeight:ie,strideWidth:me}=p,we=C-1-p.padInfo.front,Re=F-1-p.padInfo.top,_e=G-1-p.padInfo.left;for(let Be=0;Be<P;++Be)for(let He=0;He<K;++He)for(let ot=0;ot<z;++ot){let pt=ot-we,ht=Math.max(0,Math.ceil(pt/ae)),At=Math.min(B,(C+pt)/ae);for(let $e=0;$e<Z;++$e){let Tt=$e-Re,It=Math.max(0,Math.ceil(Tt/ie)),Gn=Math.min(oe,(F+Tt)/ie);for(let Qt=0;Qt<J;++Qt){let bs=Qt-_e,pn=Math.max(0,Math.ceil(bs/me)),Hn=Math.min(Q,(G+bs)/me),vs=0;for(let ws=ht;ws<At;++ws){let Pn=ws*ae-pt;for(let Us=It;Us<Gn;++Us){let jn=Us*ie-Tt;for(let ra=pn;ra<Hn;++ra){let aa=ra*me-bs,xu=A*Be+b*ws+w*Us+k*ra,Oa=E*(C-1-Pn)+R*(F-1-jn)+$*(G-1-aa)+_*He;for(let oa=0;oa<te;++oa){let Md=x[xu+oa],bu=S[Oa+oa];vs+=Md*bu}}}}h[f*Be+m*ot+g*$e+y*Qt+He]=vs}}}return n.makeTensorInfo(d.shape,d.dtype,d.values)}var uZ={kernelName:Zm,backendName:"cpu",kernelFunc:lZ},cZ=bt(bo,e=>Math.cos(e)),dZ={kernelName:bo,backendName:"cpu",kernelFunc:cZ},pZ=bt(vo,e=>Math.cosh(e)),hZ={kernelName:vo,backendName:"cpu",kernelFunc:pZ};function fZ(e){let{inputs:t,backend:n,attrs:s}=e,{image:r,boxes:a,boxInd:o}=t,{cropSize:i,method:l,extrapolationValue:u}=s,[c,p,d,h]=r.shape,f=a.shape[0],[m,g]=i,y=Le([f,m,g,h],"float32"),x=n.data.get(a.dataId).values,A=n.data.get(o.dataId).values,b=n.data.get(r.dataId).values,w=v.computeStrides(r.shape),k=v.computeStrides(y.shape);for(let S=0;S<f;S++){let E=S*4,R=x[E],$=x[E+1],_=x[E+2],P=x[E+3],C=A[S];if(C>=c)continue;let F=m>1?(_-R)*(p-1)/(m-1):0,G=g>1?(P-$)*(d-1)/(g-1):0;for(let K=0;K<m;K++){let z=m>1?R*(p-1)+K*F:.5*(R+_)*(p-1);if(z<0||z>p-1){for(let Z=0;Z<g;Z++)for(let J=0;J<h;J++){let te=J+Z*k[2]+K*k[1]+S*k[0];y.values[te]=u}continue}if(l==="bilinear"){let Z=Math.floor(z),J=Math.ceil(z),te=z-Z;for(let B=0;B<g;B++){let oe=g>1?$*(d-1)+B*G:.5*($+P)*(d-1);if(oe<0||oe>d-1){for(let me=0;me<h;me++){let we=me+B*k[2]+K*k[1]+S*k[0];y.values[we]=u}continue}let Q=Math.floor(oe),ae=Math.ceil(oe),ie=oe-Q;for(let me=0;me<h;me++){let we=me+Q*w[2]+Z*w[1]+C*w[0],Re=b[we];we=me+ae*w[2]+Z*w[1]+C*w[0];let _e=b[we];we=me+Q*w[2]+J*w[1]+C*w[0];let Be=b[we];we=me+ae*w[2]+J*w[1]+C*w[0];let He=b[we],ot=Re+(_e-Re)*ie,pt=Be+(He-Be)*ie;we=me+B*k[2]+K*k[1]+S*k[0],y.values[we]=ot+(pt-ot)*te}}}else for(let Z=0;Z<g;++Z){let J=g>1?$*(d-1)+Z*G:.5*($+P)*(d-1);if(J<0||J>d-1){for(let oe=0;oe<h;oe++){let Q=oe+Z*k[2]+K*k[1]+S*k[0];y.values[Q]=u}continue}let te=Math.round(J),B=Math.round(z);for(let oe=0;oe<h;oe++){let Q=oe+te*w[2]+B*w[1]+C*w[0],ae=oe+Z*k[2]+K*k[1]+S*k[0];y.values[ae]=b[Q]}}}}return n.makeTensorInfo(y.shape,y.dtype,y.values)}var mZ={kernelName:dl,backendName:"cpu",kernelFunc:fZ};function gZ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,exclusive:o,reverse:i}=s;Te(r,"cumprod");let l=T.getAxesPermutation([a],r.shape.length),u=r;l!=null&&(u=gs({inputs:{x:r},backend:n,attrs:{perm:l}}));let c=T.getInnerMostAxes(1,r.shape.length)[0];if(c!==u.shape.length-1)throw new Error(`backend.cumprod in CPU expects an inner-most axis=${u.shape.length-1} but got axis=${c}`);let p=Mn(u.dtype,"int32"),d=v.makeOnesTypedArray(v.sizeFromShape(u.shape),p),h=n.data.get(u.dataId).values,f=u.shape[u.shape.length-1],m=i?(y,x)=>y+f-x-1:(y,x)=>y+x;for(let y=0;y<h.length;y+=f)for(let x=0;x<f;x++){let A=m(y,x);if(x===0)d[A]=o?1:h[A];else{let b=m(y,x-1);d[A]=o?h[b]*d[b]:h[A]*d[b]}}let g=n.makeTensorInfo(u.shape,p,d);if(l!=null){let y=T.getUndoAxesPermutation(l),x=gs({inputs:{x:g},backend:n,attrs:{perm:y}});return n.disposeIntermediateTensorInfo(g),n.disposeIntermediateTensorInfo(u),x}return g}var yZ={kernelName:cl,backendName:"cpu",kernelFunc:gZ};function AZ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,exclusive:o,reverse:i}=s;Te(r,"cumsum");let l=T.getAxesPermutation([a],r.shape.length),u=r;l!=null&&(u=gs({inputs:{x:r},backend:n,attrs:{perm:l}}));let c=T.getInnerMostAxes(1,r.shape.length)[0];if(c!==u.shape.length-1)throw new Error(`backend.cumsum in CPU expects an inner-most axis=${u.shape.length-1} but got axis=${c}`);let p=Mn(u.dtype,"int32"),d=v.makeZerosTypedArray(v.sizeFromShape(u.shape),p),h=n.data.get(u.dataId).values,f=u.shape[u.shape.length-1],m=i?(y,x)=>y+f-x-1:(y,x)=>y+x;for(let y=0;y<h.length;y+=f)for(let x=0;x<f;x++){let A=m(y,x);if(x===0)d[A]=o?0:h[A];else{let b=m(y,x-1);d[A]=o?h[b]+d[b]:h[A]+d[b]}}let g=n.makeTensorInfo(u.shape,p,d);if(l!=null){let y=T.getUndoAxesPermutation(l),x=gs({inputs:{x:g},backend:n,attrs:{perm:y}});return n.disposeIntermediateTensorInfo(g),n.disposeIntermediateTensorInfo(u),x}return g}var xZ={kernelName:wo,backendName:"cpu",kernelFunc:AZ};function bZ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,weights:a}=t,{size:o,binaryOutput:i}=s;if(r.shape.length===1){let l=n.data.get(r.dataId).values,u=n.data.get(a.dataId).values,c=Nx(l,u,a.dtype,a.shape,o);return n.makeTensorInfo([o],a.dtype,c)}else if(r.shape.length===2){let l=n.bufferSync(r),u=n.bufferSync(a),c=yI(l,u,o,i);return n.makeTensorInfo(c.shape,a.dtype,c.values)}throw new Error(`Error in denseBincount: input must be at most rank 2, but got rank${r.shape.length}.`)}var vZ={kernelName:Ym,backendName:"cpu",kernelFunc:bZ};function wZ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockSize:a,dataFormat:o}=s;v.assert(o==="NHWC",()=>`Only NHWC dataFormat supported on CPU for depthToSpace. Got ${o}`);let i=r.shape[0],l=r.shape[1],u=r.shape[2],c=r.shape[3],p=l*a,d=u*a,h=c/(a*a),f=n.data.get(r.dataId).values,m=new Float32Array(i*p*d*h),g=0;for(let y=0;y<i;++y)for(let x=0;x<p;++x){let A=Math.floor(x/a),b=x%a;for(let w=0;w<d;++w){let k=Math.floor(w/a),S=w%a,E=(b*a+S)*h;for(let R=0;R<h;++R){let _=R+E+c*(k+u*(A+l*y));m[g++]=f[_]}}}return n.makeTensorInfo([i,p,d,h],r.dtype,m)}var kZ={kernelName:pl,backendName:"cpu",kernelFunc:wZ};function aS(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dilations:l,dimRoundingMode:u}=s;Te([r,a],"depthwiseConv2DNative");let c=v.computeStrides(r.shape),p=v.computeStrides(a.shape),d=l;d==null&&(d=[1,1]),v.assert(T.eitherStridesOrDilationsAreOne(o,d),()=>`Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${o} and dilations '${d}'`);let h=T.computeConv2DInfo(r.shape,a.shape,o,d,i,u,!0),{filterHeight:f,filterWidth:m,dilationHeight:g,dilationWidth:y,padInfo:x}=h,A=x.left,b=x.top,w=h.outChannels/h.inChannels,k=new fn(h.outShape,r.dtype),S=n.data.get(r.dataId).values,E=n.data.get(a.dataId).values,R=k.values;for(let $=0;$<h.batchSize;++$){let _=$*c[0],P=$*k.strides[0];for(let C=0;C<h.outHeight;++C){let F=P+C*k.strides[1],G=C*h.strideHeight-b;for(let K=0;K<f;++K){let z=G+K*g;if(z<0||z>=h.inHeight)continue;let Z=K*p[0],J=_+z*c[1];for(let te=0;te<h.outWidth;++te){let B=F+te*k.strides[2],oe=te*h.strideWidth-A;for(let Q=0;Q<m;++Q){let ae=oe+Q*y;if(ae<0||ae>=h.inWidth)continue;let ie=Z+Q*p[1],me=J+ae*h.inChannels,we=B,Re=ie;for(let _e=0;_e<h.inChannels;++_e){let Be=S[me+_e];for(let He=0;He<w;++He)R[we+He]+=Be*E[Re+He];we+=w,Re+=w}}}}}}return n.makeTensorInfo(k.shape,k.dtype,k.values)}var IZ={kernelName:ko,backendName:"cpu",kernelFunc:aS};function SZ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,dy:a}=t,{strides:o,dilations:i,pad:l,dimRoundingMode:u,filterShape:c}=s;Te([r,a],"depthwiseConv2dNativeBackpropFilter");let p=T.computeConv2DInfo(r.shape,c,o,i,l,u,!0),{strideHeight:d,strideWidth:h,filterHeight:f,filterWidth:m}=p,g=new fn(p.filterShape,"float32"),y=p.padInfo.left,x=p.padInfo.top,A=p.outChannels/p.inChannels,b=n.data.get(r.dataId).values,w=new fn(r.shape,r.dtype,b),k=n.data.get(a.dataId).values,S=new fn(a.shape,a.dtype,k);for(let E=0;E<f;++E){let R=Math.max(0,Math.ceil((x-E)/d)),$=Math.min(p.outHeight,(p.inHeight+x-E)/d);for(let _=0;_<m;++_){let P=Math.max(0,Math.ceil((y-_)/h)),C=Math.min(p.outWidth,(p.inWidth+y-_)/h);for(let F=0;F<p.outChannels;++F){let G=Math.trunc(F/A),K=F%A,z=0;for(let Z=0;Z<p.batchSize;++Z)for(let J=R;J<$;++J){let te=E+J*d-x;for(let B=P;B<C;++B){let oe=_+B*h-y;z+=w.get(Z,te,oe,G)*S.get(Z,J,B,F)}}g.set(z,E,_,G,K)}}}return n.makeTensorInfo(g.shape,g.dtype,g.values)}var CZ={kernelName:Jm,backendName:"cpu",kernelFunc:SZ};function TZ(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,filter:a}=t,{strides:o,dilations:i,pad:l,dimRoundingMode:u,inputShape:c}=s;Te([r,a],"depthwiseConv2DNativeBackpropInput");let p=v.computeStrides(r.shape),d=v.computeStrides(a.shape),h=T.computeConv2DInfo(c,a.shape,o,i,l,u,!0),f=new fn(h.inShape,"float32"),m=f.values,[g,y,x]=f.strides,A=n.data.get(r.dataId).values,[b,w,k]=p,S=n.data.get(a.dataId).values,[E,R,$]=d,{batchSize:_,filterHeight:P,filterWidth:C,inChannels:F,inHeight:G,inWidth:K,outChannels:z,outHeight:Z,outWidth:J,strideHeight:te,strideWidth:B}=h,oe=P-1-h.padInfo.top,Q=C-1-h.padInfo.left,ae=z/F;for(let ie=0;ie<_;++ie)for(let me=0;me<F;++me)for(let we=0;we<G;++we){let Re=we-oe,_e=Math.max(0,Math.ceil(Re/te)),Be=Math.min(Z,(P+Re)/te);for(let He=0;He<K;++He){let ot=He-Q,pt=Math.max(0,Math.ceil(ot/B)),ht=Math.min(J,(C+ot)/B),At=0;for(let $e=_e;$e<Be;++$e){let Tt=$e*te-Re;for(let It=pt;It<ht;++It){let Gn=It*B-ot,Qt=b*ie+w*$e+k*It,bs=E*(P-1-Tt)+R*(C-1-Gn)+$*me;for(let pn=0;pn<ae;++pn){let Hn=me*ae+pn,vs=A[Qt+Hn],ws=S[bs+pn];At+=vs*ws}}}m[g*ie+y*we+x*He+me]=At}}return n.makeTensorInfo(f.shape,f.dtype,f.values)}var NZ={kernelName:Qm,backendName:"cpu",kernelFunc:TZ};function EZ(e){let{inputs:t,backend:n}=e,{x:s}=t,r=v.sizeFromShape(s.shape),a=n.data.get(s.dataId).values,o=Le([r,r],s.dtype),i=o.values;for(let u=0;u<a.length;u++)i[u*r+u]=a[u];let l=[...s.shape,...s.shape];return n.makeTensorInfo(l,o.dtype,o.values)}var RZ={kernelName:e0,backendName:"cpu",kernelFunc:EZ},_Z={kernelName:Vp,backendName:"cpu",kernelFunc:({inputs:e,backend:t,attrs:n})=>{let{x:s,filter:r}=e,{strides:a,pad:o,dilations:i}=n,l=t,u=l.data.get(s.dataId).values,c=s.shape.length,p=l.data.get(r.dataId).values,d=r.shape.length,{batchSize:h,inHeight:f,inWidth:m,inChannels:g,outHeight:y,outWidth:x,padInfo:A,strideHeight:b,strideWidth:w,filterHeight:k,filterWidth:S,dilationHeight:E,dilationWidth:R,outShape:$}=T.computeDilation2DInfo(s.shape,r.shape,a,o,"NHWC",i),_=v.sizeFromShape($),P=$.length,C=v.getArrayFromDType(s.dtype,_);for(let G=0;G<h;++G)for(let K=0;K<y;++K){let z=K*b-A.top;for(let Z=0;Z<x;++Z){let J=Z*w-A.left;for(let te=0;te<g;++te){let B=Number.MIN_SAFE_INTEGER;for(let Q=0;Q<k;++Q){let ae=z+Q*E;if(ae>=0&&ae<f)for(let ie=0;ie<S;++ie){let me=J+ie*R;if(me>=0&&me<m){let we=v.locToIndex([G,ae,me,te],c,v.computeStrides(s.shape)),Re=v.locToIndex([Q,ie,te],d,v.computeStrides(r.shape)),_e=u[we]+p[Re];_e>B&&(B=_e)}}}let oe=v.locToIndex([G,K,Z,te],P,v.computeStrides($));C[oe]=B}}}return{dataId:l.write(v.toTypedArray(C,s.dtype),$,s.dtype),shape:$,dtype:s.dtype}}},DZ={kernelName:hm,backendName:"cpu",kernelFunc:({inputs:e,backend:t,attrs:n})=>{let{x:s,filter:r,dy:a}=e,{strides:o,pad:i,dilations:l}=n,u=t,c=v.toNestedArray(s.shape,u.data.get(s.dataId).values),p=v.toNestedArray(r.shape,u.data.get(r.dataId).values),{batchSize:d,inHeight:h,inWidth:f,inChannels:m,outHeight:g,outWidth:y,padInfo:x,strideHeight:A,strideWidth:b,filterHeight:w,filterWidth:k,dilationHeight:S,dilationWidth:E,outShape:R}=T.computeDilation2DInfo(s.shape,r.shape,o,i,"NHWC",l);v.assert(a.rank===R.length,()=>`Error in ${hm}, dy must have the same rank as output ${R.length}, but got ${a.rank}`);let $=v.toNestedArray(R,u.data.get(a.dataId).values),_=v.makeZerosNestedTypedArray(r.shape,r.dtype);for(let C=0;C<d;++C)for(let F=0;F<g;++F){let G=F*A-x.top;for(let K=0;K<y;++K){let z=K*b-x.left;for(let Z=0;Z<m;++Z){let J=Number.MIN_SAFE_INTEGER,te=0,B=0;for(let oe=0;oe<w;++oe){let Q=G+oe*S;if(Q>=0&&Q<h)for(let ae=0;ae<k;++ae){let ie=z+ae*E;if(ie>=0&&ie<f){let me=c[C][Q][ie][Z]+p[oe][ae][Z];me>J&&(J=me,te=oe,B=ae)}}}_[te][B][Z]+=$[C][F][K][Z]}}}return{dataId:u.write(v.toTypedArray(_,s.dtype),r.shape,r.dtype),shape:r.shape,dtype:r.dtype}}},$Z={kernelName:pm,backendName:"cpu",kernelFunc:({inputs:e,backend:t,attrs:n})=>{let{x:s,filter:r,dy:a}=e,{strides:o,pad:i,dilations:l}=n,u=t,c=v.toNestedArray(s.shape,u.data.get(s.dataId).values),p=v.toNestedArray(r.shape,u.data.get(r.dataId).values),{batchSize:d,inHeight:h,inWidth:f,inChannels:m,outHeight:g,outWidth:y,padInfo:x,strideHeight:A,strideWidth:b,filterHeight:w,filterWidth:k,dilationHeight:S,dilationWidth:E,outShape:R}=T.computeDilation2DInfo(s.shape,r.shape,o,i,"NHWC",l);v.assert(a.rank===R.length,()=>`Error in ${pm}, dy must have the same rank as output ${R.length}, but got ${a.rank}`);let $=v.toNestedArray(R,u.data.get(a.dataId).values),_=v.makeZerosNestedTypedArray(s.shape,s.dtype);for(let C=0;C<d;++C)for(let F=0;F<g;++F){let G=F*A-x.top;for(let K=0;K<y;++K){let z=K*b-x.left;for(let Z=0;Z<m;++Z){let J=Number.MIN_SAFE_INTEGER,te=G<0?0:G,B=z<0?0:z;for(let oe=0;oe<w;++oe){let Q=G+oe*S;if(Q>=0&&Q<h)for(let ae=0;ae<k;++ae){let ie=z+ae*E;if(ie>=0&&ie<f){let me=c[C][Q][ie][Z]+p[oe][ae][Z];me>J&&(J=me,te=Q,B=ie)}}}_[C][te][B][Z]+=$[C][F][K][Z]}}}return{dataId:u.write(v.toTypedArray(_,s.dtype),s.shape,s.dtype),shape:s.shape,dtype:s.dtype}}};function Ph(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;Te(r,"sum");let i;r.dtype==="bool"?i=io({inputs:{x:r},backend:n,attrs:{dtype:"int32"}}):i=Zr({inputs:{x:r},backend:n});let l=i.shape.length,u=v.parseAxisParam(a,i.shape),c=T.getAxesPermutation(u,l),p=u,d=i;c!=null&&(d=gs({inputs:{x:i},backend:n,attrs:{perm:c}}),p=T.getInnerMostAxes(p.length,l)),T.assertAxesAreInnerMostDims("sum",p,d.shape.length);let[h,f]=T.computeOutAndReduceShapes(d.shape,p),m=T.upcastType(d.dtype,"int32"),g=Rm(n,h,m),y=v.sizeFromShape(f),x=n.data.get(g.dataId).values,A=n.data.get(d.dataId).values;for(let b=0;b<x.length;++b){let w=b*y,k=0;for(let S=0;S<y;++S)k+=A[w+S];x[b]=k}if(o){let b=T.expandShapeToKeepDim(g.shape,u),w=g;g=Rt({inputs:{x:g},backend:n,attrs:{shape:b}}),n.disposeIntermediateTensorInfo(w)}return n.disposeIntermediateTensorInfo(i),c!=null&&n.disposeIntermediateTensorInfo(d),g}var PZ={kernelName:ei,backendName:"cpu",kernelFunc:Ph};function FZ(e){let{inputs:t,backend:n,attrs:s}=e,{equation:r}=s,a=t,{allDims:o,summedDims:i,idDims:l}=T.decodeEinsumEquation(r,a.length);T.checkEinsumDimSizes(o.length,l,a);let{path:u,steps:c}=T.getEinsumComputePath(i,l),p=c.length,d=null,h=o.length,f=[];for(let m=0;m<p;++m){for(let g of c[m]){let{permutationIndices:y,expandDims:x}=T.getEinsumPermutation(h,l[g]),A;T.isIdentityPermutation(y)?A=a[g]:(A=gs({inputs:{x:a[g]},backend:n,attrs:{perm:y}}),f.push(A));let b=A.shape.slice();for(let w=0;w<x.length;++w)b.splice(x[w],0,1);v.arraysEqual(A.shape,b)||(A=Rt({inputs:{x:A},backend:n,attrs:{shape:b}}),f.push(A)),d===null?d=A:(d=A2({inputs:{a:A,b:d},backend:n}),f.push(d))}m<p-1&&(u[m]>=0&&(d=Ph({inputs:{x:d},backend:n,attrs:{axis:u[m]-(o.length-h),keepDims:!1}}),f.push(d)),h--)}for(let m of f)m!==d&&n.disposeIntermediateTensorInfo(m);return d}var OZ={kernelName:Up,backendName:"cpu",kernelFunc:FZ};function MZ(e){let{inputs:t,backend:n}=e,{dy:s,y:r}=t;Te([s,r],"eluGrad");let a=new Float32Array(v.sizeFromShape(r.shape)),o=n.data.get(r.dataId).values,i=n.data.get(s.dataId).values;for(let l=0;l<o.length;++l){let u=o[l];u>=1?a[l]=i[l]:a[l]=i[l]*(u+1)}return n.makeTensorInfo(r.shape,"float32",a)}var zZ={kernelName:t0,backendName:"cpu",kernelFunc:MZ},LZ=T.ERF_P,BZ=T.ERF_A1,WZ=T.ERF_A2,VZ=T.ERF_A3,UZ=T.ERF_A4,GZ=T.ERF_A5,HZ=bt(wc,e=>{let t=Math.sign(e),n=Math.abs(e),s=1/(1+LZ*n);return t*(1-((((GZ*s+UZ)*s+VZ)*s+WZ)*s+BZ)*s*Math.exp(-n*n))}),jZ={kernelName:wc,backendName:"cpu",kernelFunc:HZ};function $m(e){let{inputs:t,backend:n,attrs:s}=e,{input:r}=t,{dim:a}=s,o=r.shape.length,i=r.shape.slice(),l=a;return a<0&&(v.assert(-(o+1)<=a,()=>`Axis must be in the interval [${-(o+1)}, ${o}]`),l=o+a+1),i.splice(l,0,1),Rt({inputs:{x:r},backend:n,attrs:{shape:i}})}var qZ={kernelName:fl,backendName:"cpu",kernelFunc:$m},XZ=un((e,t)=>e/t),Lx=kn(Io,XZ),ey={kernelName:Io,backendName:"cpu",kernelFunc:Lx};function oS(e,t,n){let s=e.shape,r=s[0],a=s[1],o=n.data.get(e.dataId),i=o.complexTensorInfos.real,l=o.complexTensorInfos.imag,u=[r,a],c=v.sizeFromShape(u),p=v.getTypedArrayFromDType("float32",c),d=v.getTypedArrayFromDType("float32",c);for(let g=0;g<r;g++){let y=nl({inputs:{x:i},backend:n,attrs:{begin:[g,0],size:[1,a]}}),x=nl({inputs:{x:l},backend:n,attrs:{begin:[g,0],size:[1,a]}}),A=Cs({inputs:{real:y,imag:x},backend:n}),{real:b,imag:w}=KZ(A,t,n),k=T.mergeRealAndImagArrays(b,w);for(let S=0;S<a;S++){let E=T.getComplexWithIndex(k,S);p[g*a+S]=E.real,d[g*a+S]=E.imag}n.disposeIntermediateTensorInfo(y),n.disposeIntermediateTensorInfo(x),n.disposeIntermediateTensorInfo(A)}let h=n.makeTensorInfo(u,"float32",p),f=n.makeTensorInfo(u,"float32",d),m=Cs({inputs:{real:h,imag:f},backend:n});return n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(f),m}function KZ(e,t,n){let s=v.sizeFromShape(e.shape),r=n.data.get(e.dataId),a=n.data.get(r.complexTensorInfos.real.dataId).values,o=n.data.get(r.complexTensorInfos.imag.dataId).values;if(ZZ(s)){let i=ty(a,o,s,t,n),l=[e.shape[0],e.shape[1]];if(t){let u=n.makeTensorInfo(l,"float32",i.real),c=n.makeTensorInfo(l,"float32",i.imag),p=n.makeTensorInfo([],"float32",v.createScalarValue(s,"float32")),d=Zr({inputs:{x:p},backend:n}),h=ey.kernelFunc({inputs:{a:u,b:p},backend:n}),f=ey.kernelFunc({inputs:{a:c,b:d},backend:n}),m=n.data.get(h.dataId).values,g=n.data.get(f.dataId).values;return n.disposeIntermediateTensorInfo(u),n.disposeIntermediateTensorInfo(c),n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(f),{real:m,imag:g}}return i}else{let i=T.mergeRealAndImagArrays(a,o),l=YZ(i,s,t);return T.splitRealAndImagArrays(l)}}function ZZ(e){return(e&e-1)===0}function ty(e,t,n,s,r){if(n===1)return{real:e,imag:t};let a=T.mergeRealAndImagArrays(e,t),o=n/2,i=T.complexWithEvenIndex(a),l=i.real,u=i.imag,c=[l.length],p=r.makeTensorInfo(c,"float32",l),d=r.makeTensorInfo(c,"float32",u),h=Cs({inputs:{real:p,imag:d},backend:r}),f=T.complexWithOddIndex(a),m=f.real,g=f.imag,y=[m.length],x=r.makeTensorInfo(y,"float32",m),A=r.makeTensorInfo(y,"float32",g),b=Cs({inputs:{real:x,imag:A},backend:r}),w=ty(l,u,o,s,r),k=w.real,S=w.imag,E=[k.length],R=r.makeTensorInfo(E,"float32",k),$=r.makeTensorInfo(E,"float32",S),_=Cs({inputs:{real:R,imag:$},backend:r}),P=ty(m,g,o,s,r),C=P.real,F=P.imag,G=[C.length],K=r.makeTensorInfo(G,"float32",C),z=r.makeTensorInfo(G,"float32",F),Z=Cs({inputs:{real:K,imag:z},backend:r}),J=T.exponents(n,s),te=[J.real.length],B=r.makeTensorInfo(te,"float32",J.real),oe=r.makeTensorInfo(te,"float32",J.imag),Q=Cs({inputs:{real:B,imag:oe},backend:r}),ae=A2({inputs:{a:Q,b:Z},backend:r}),ie=oc({inputs:{a:_,b:ae},backend:r}),me=Mx({inputs:{a:_,b:ae},backend:r}),we=tl({inputs:{input:ie},backend:r}),Re=tl({inputs:{input:me},backend:r}),_e=ic({inputs:{input:ie},backend:r}),Be=ic({inputs:{input:me},backend:r}),He=lc({inputs:[we,Re],backend:r,attrs:{axis:0}}),ot=lc({inputs:[_e,Be],backend:r,attrs:{axis:0}}),pt=r.data.get(He.dataId).values,ht=r.data.get(ot.dataId).values;return r.disposeIntermediateTensorInfo(p),r.disposeIntermediateTensorInfo(d),r.disposeIntermediateTensorInfo(h),r.disposeIntermediateTensorInfo(x),r.disposeIntermediateTensorInfo(A),r.disposeIntermediateTensorInfo(b),r.disposeIntermediateTensorInfo(R),r.disposeIntermediateTensorInfo($),r.disposeIntermediateTensorInfo(_),r.disposeIntermediateTensorInfo(K),r.disposeIntermediateTensorInfo(z),r.disposeIntermediateTensorInfo(Z),r.disposeIntermediateTensorInfo(B),r.disposeIntermediateTensorInfo(oe),r.disposeIntermediateTensorInfo(Q),r.disposeIntermediateTensorInfo(ae),r.disposeIntermediateTensorInfo(ie),r.disposeIntermediateTensorInfo(me),r.disposeIntermediateTensorInfo(we),r.disposeIntermediateTensorInfo(_e),r.disposeIntermediateTensorInfo(Re),r.disposeIntermediateTensorInfo(Be),r.disposeIntermediateTensorInfo(He),r.disposeIntermediateTensorInfo(ot),{real:pt,imag:ht}}function YZ(e,t,n){let s=new Float32Array(t*2);for(let r=0;r<t;r++){let a=0,o=0;for(let i=0;i<t;i++){let l=T.exponent(r*i,t,n),u=T.getComplexWithIndex(e,i);a+=u.real*l.real-u.imag*l.imag,o+=u.real*l.imag+u.imag*l.real}n&&(a/=t,o/=t),T.assignToTypedArray(s,a,o,r)}return s}function JZ(e){let{inputs:t,backend:n}=e,{input:s}=t,r=v.sizeFromShape(s.shape),a=s.shape[s.shape.length-1],o=r/a,i=Rt({inputs:{x:s},backend:n,attrs:{shape:[o,a]}}),l=oS(i,!1,n),u=Rt({inputs:{x:l},backend:n,attrs:{shape:s.shape}});return n.disposeIntermediateTensorInfo(i),n.disposeIntermediateTensorInfo(l),u}var QZ={kernelName:n0,backendName:"cpu",kernelFunc:JZ};function Bx(e){let{backend:t,attrs:n}=e,{shape:s,value:r,dtype:a}=n,o=a||v.inferDtype(r),i=v.getArrayFromDType(o,v.sizeFromShape(s));return tY(i,r,o),t.makeTensorInfo(s,o,i)}var eY={kernelName:kc,backendName:"cpu",kernelFunc:Bx};function tY(e,t,n){e.fill(t)}var nY={kernelName:gl,backendName:"cpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{image:s}=e,r=n,a=v.getTypedArrayFromDType(s.dtype,v.sizeFromShape(s.shape)),[o,i,l,u]=s.shape,c=r.data.get(s.dataId).values;for(let d=0;d<o;d++){let h=d*l*i*u;for(let f=0;f<i;f++){let m=f*(l*u);for(let g=0;g<l;g++){let y=g*u;for(let x=0;x<u;x++){let A=Math.round(l-g-1),b=h+m+y+x,w=c[b];if(A>=0&&A<l){let k=A*u,S=h+m+k+x;w=c[S]}a[b]=w}}}}return{dataId:r.write(a,s.shape,s.dtype),shape:s.shape,dtype:s.dtype}}},sY=un((e,t)=>Math.floor(e/t)),rY=kn(No,sY,null,"int32"),aY={kernelName:No,backendName:"cpu",kernelFunc:rY};function oY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dataFormat:c,dilations:p,dimRoundingMode:d,activation:h,leakyreluAlpha:f}=s,m=rS({inputs:{x:r,filter:a},backend:n,attrs:{strides:l,pad:u,dataFormat:c,dilations:p,dimRoundingMode:d}});if(o){let g=m;if(c==="NCHW"&&o.shape.length===1&&o.shape[0]!==1){let y=Rt({inputs:{x:o},backend:n,attrs:{shape:[o.shape[0],1,1]}});m=oc({inputs:{a:m,b:y},backend:n}),n.disposeIntermediateTensorInfo(y)}else m=oc({inputs:{a:m,b:o},backend:n});n.disposeIntermediateTensorInfo(g)}if(h){let g=m;if(c==="NCHW"&&h==="prelu"&&i.shape.length===1&&i.shape[0]!==1){let y=Rt({inputs:{x:i},backend:n,attrs:{shape:[i.shape[0],1,1]}});m=Dm(n,m,h,y,f),n.disposeIntermediateTensorInfo(y)}else m=Dm(n,m,h,i,f);n.disposeIntermediateTensorInfo(g)}return m}var iY={kernelName:Qa,backendName:"cpu",kernelFunc:oY};function lY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dataFormat:c,dilations:p,dimRoundingMode:d,activation:h,leakyreluAlpha:f}=s,m=aS({inputs:{x:r,filter:a},backend:n,attrs:{strides:l,pad:u,dataFormat:c,dilations:p,dimRoundingMode:d}});if(o){let g=m;m=oc({inputs:{a:m,b:o},backend:n}),n.disposeIntermediateTensorInfo(g)}if(h){let g=m;m=Dm(n,m,h,i,f),n.disposeIntermediateTensorInfo(g)}return m}var uY={kernelName:eo,backendName:"cpu",kernelFunc:lY};function cY(e){let{inputs:t,backend:n}=e,{params:s,indices:r}=t,a=v.sizeFromShape(s.shape),o=r.shape,i=o[o.length-1],[l,u,c,p]=T.prepareAndValidate(s,r);if(u===0)return n.makeTensorInfo(l,s.dtype,[]);let d=n.data.get(r.dataId).values,h=n.bufferSync(s),f=SI(d,h,s.dtype,u,i,c,p,s.shape,a);return n.makeTensorInfo(l,s.dtype,f.values)}var dY={kernelName:Al,backendName:"cpu",kernelFunc:cY};function pY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,indices:a}=t,{axis:o,batchDims:i}=s;Te([r,a],"gatherV2");let l=v.parseAxisParam(o,r.shape)[0],u=n.data.get(a.dataId).values,c=r.shape[l];for(let b=0;b<u.length;++b){let w=u[b];v.assert(w<=c-1&&w>=0,()=>`GatherV2: the index value ${w} is not in [0, ${c-1}]`)}let p=i;i==null&&(p=0);let d=v.sizeFromShape(a.shape),h=T.segment_util.collectGatherOpShapeInfo(r,a,l,p),f=Rt({inputs:{x:r},backend:n,attrs:{shape:[h.batchSize,h.outerSize,h.dimSize,h.sliceSize]}}),m=Rt({inputs:{x:a},backend:n,attrs:{shape:[h.batchSize,d/h.batchSize]}}),g=[h.batchSize,h.outerSize,d/h.batchSize,h.sliceSize],y=n.bufferSync(m),x=n.bufferSync(f),A=CI(x,y,g);return n.disposeIntermediateTensorInfo(f),n.disposeIntermediateTensorInfo(m),n.makeTensorInfo(h.outputShape,A.dtype,A.values)}var hY={kernelName:yl,backendName:"cpu",kernelFunc:pY};function fY(e){let{inputs:t,backend:n}=e,{input:s}=t,r=v.sizeFromShape(s.shape),a=s.shape[s.shape.length-1],o=r/a,i=Rt({inputs:{x:s},backend:n,attrs:{shape:[o,a]}}),l=oS(i,!0,n),u=Rt({inputs:{x:l},backend:n,attrs:{shape:s.shape}});return n.disposeIntermediateTensorInfo(i),n.disposeIntermediateTensorInfo(l),u}var mY={kernelName:s0,backendName:"cpu",kernelFunc:fY},gY=bt(Ic,e=>Number.isFinite(e)?1:0,"bool"),yY={kernelName:Ic,backendName:"cpu",kernelFunc:gY},AY=bt(Sc,e=>Math.abs(e)===1/0?1:0,"bool"),xY={kernelName:Sc,backendName:"cpu",kernelFunc:AY},bY=bt(Cc,e=>Number.isNaN(e)?1:0,"bool"),vY={kernelName:Cc,backendName:"cpu",kernelFunc:bY};function wY(e){let{backend:t,attrs:n}=e,{start:s,stop:r,num:a}=n,o=_I(s,r,a);return t.makeTensorInfo([o.length],"float32",o)}var kY={kernelName:r0,backendName:"cpu",kernelFunc:wY},IY=bt(Tc,e=>Math.log1p(e)),SY={kernelName:Tc,backendName:"cpu",kernelFunc:IY},CY=un((e,t)=>e&&t),TY=kn(wl,CY,null,"bool"),NY={kernelName:wl,backendName:"cpu",kernelFunc:TY},EY=bt(kl,e=>e?0:1,"bool"),RY={kernelName:kl,backendName:"cpu",kernelFunc:EY},_Y=un((e,t)=>e||t),DY=kn(Nc,_Y,null,"bool"),$Y={kernelName:Nc,backendName:"cpu",kernelFunc:DY};function PY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{depthRadius:a,bias:o,alpha:i,beta:l}=s;Te(r,"LRN");let u=r.shape[3],c=u-1,p=n.data.get(r.dataId).values,d=v.sizeFromShape(r.shape),h=new Float32Array(d);function f(m){let g=m%u,y=m-g+Math.max(0,g-a),x=m-g+Math.min(g+a,c),A=0;for(;y<=x;y++){let b=p[y];A+=b*b}return A}for(let m=0;m<d;m++){let g=f(m),y=p[m]*Math.pow(o+i*g,-l);h[m]=y}return n.makeTensorInfo(r.shape,r.dtype,h)}var FY={kernelName:Hp,backendName:"cpu",kernelFunc:PY};function OY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,y:a,dy:o}=t,{depthRadius:i,bias:l,alpha:u,beta:c}=s;Te(o,"LRNGrad");let p=v.sizeFromShape(o.shape),d=o.shape[3],h=n.data.get(o.dataId).values,f=n.data.get(r.dataId).values,m=n.data.get(a.dataId).values,g=new Float32Array(p),y=p;for(let x=0;x<y;x++){let A=x%d,b=x-A+Math.max(0,A-i),w=x-A+Math.min(d,A+i+1),k=0;for(let S=b;S<w;S++)k+=Math.pow(f[S],2);k=u*k+l;for(let S=b;S<w;S++){let E=-2*u*c*f[S]*m[x]/k;x===S&&(E+=Math.pow(k,-c)),E*=h[x],g[S]+=E}}return n.makeTensorInfo(o.shape,r.dtype,g)}var MY={kernelName:a0,backendName:"cpu",kernelFunc:OY};function iS(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{reductionIndices:a,keepDims:o}=s,i=n,l=r.shape,u=l.length,c=v.parseAxisParam(a,l),p=c,d=T.getAxesPermutation(p,u),h=i.data.get(r.dataId).values;if(d!=null){let b=new Array(u);for(let w=0;w<b.length;w++)b[w]=l[d[w]];h=_x(h,l,r.dtype,d,b),p=T.getInnerMostAxes(p.length,u),l=b}Te(r,"max"),T.assertAxesAreInnerMostDims("max",p,u);let[f,m]=T.computeOutAndReduceShapes(l,p),g=v.sizeFromShape(m),y=$I(h,g,f,r.dtype),x=i.write(y,f,r.dtype),A=f;return o&&(A=T.expandShapeToKeepDim(f,c)),{dataId:x,shape:A,dtype:r.dtype}}var zY={kernelName:Po,backendName:"cpu",kernelFunc:iS};function LY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t;Te(r,"maxPool");let{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=s,u=1;v.assert(T.eitherStridesOrDilationsAreOne(o,u),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${o} and dilations '${u}'`);let c=T.computePool2DInfo(r.shape,a,o,u,i,l),p;if(c.filterWidth===1&&c.filterHeight===1&&v.arraysEqual(c.inShape,c.outShape))p=Zr({inputs:{x:r},backend:n});else{let d=n.data.get(r.dataId).values,h=v.computeStrides(r.shape),f=zx(d,r.shape,r.dtype,h,c,"max");p=n.makeTensorInfo(c.outShape,r.dtype,f.values)}return p}var BY={kernelName:Oo,backendName:"cpu",kernelFunc:LY};function WY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l,dataFormat:u}=s;Te(r,"maxPool3d");let c=T.computePool3DInfo(r.shape,a,o,1,i,l,u),p=n.data.get(r.dataId).values,d=sS(p,r.shape,r.dtype,v.computeStrides(r.shape),c,"max");return n.makeTensorInfo(d.shape,"float32",d.values)}var VY={kernelName:jp,backendName:"cpu",kernelFunc:WY};function UY(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,input:a}=t,{filterSize:o,strides:i,pad:l,dimRoundingMode:u}=s;Te([r,a],"maxPool3DGrad");let c=T.computePool3DInfo(a.shape,o,i,1,l,u),p=n.bufferSync(a),d=RK(p,c),h=c.strideDepth,f=c.strideHeight,m=c.strideWidth,g=c.dilationDepth,y=c.dilationHeight,x=c.dilationWidth,A=c.effectiveFilterDepth,b=c.effectiveFilterHeight,w=c.effectiveFilterWidth,k=A-1-c.padInfo.front,S=w-1-c.padInfo.left,E=b-1-c.padInfo.top,R=Le(a.shape,"float32"),$=n.bufferSync(r);for(let _=0;_<c.batchSize;++_)for(let P=0;P<c.inChannels;++P)for(let C=0;C<c.inDepth;++C)for(let F=0;F<c.inHeight;++F)for(let G=0;G<c.inWidth;++G){let K=C-k,z=F-E,Z=G-S,J=0;for(let te=0;te<A;te+=g){let B=(K+te)/h;if(!(B<0||B>=c.outDepth||Math.floor(B)!==B))for(let oe=0;oe<b;oe+=y){let Q=(z+oe)/f;if(!(Q<0||Q>=c.outHeight||Math.floor(Q)!==Q))for(let ae=0;ae<w;ae+=x){let ie=(Z+ae)/m;if(ie<0||ie>=c.outWidth||Math.floor(ie)!==ie)continue;let me=A*b*w-1-d.get(_,B,Q,ie,P),we=te*b*w+oe*w+ae,Re=me===we?1:0;if(Re===0)continue;J+=$.get(_,B,Q,ie,P)*Re}}}R.set(J,_,C,F,G,P)}return n.makeTensorInfo(R.shape,R.dtype,R.values)}var GY={kernelName:i0,backendName:"cpu",kernelFunc:UY};function HY(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,input:a,output:o}=t,i=a;Te([a,o],"maxPoolGrad");let{filterSize:l,strides:u,pad:c,dimRoundingMode:p}=s,d=T.computePool2DInfo(i.shape,l,u,1,c,p),h=n.data.get(i.dataId).values,f=Le(d.outShape,i.dtype,nS(h,i.shape,i.dtype,d).values),m=d.strideHeight,g=d.strideWidth,y=d.dilationHeight,x=d.dilationWidth,A=d.effectiveFilterHeight,b=d.effectiveFilterWidth,w=b-1-d.padInfo.left,k=A-1-d.padInfo.top,S=Le(i.shape,"float32"),E=n.data.get(r.dataId).values,R=Le(r.shape,"float32",E);for(let $=0;$<d.batchSize;++$)for(let _=0;_<d.inChannels;++_)for(let P=0;P<d.inHeight;++P)for(let C=0;C<d.inWidth;++C){let F=P-k,G=C-w,K=0;for(let z=0;z<A;z+=y){let Z=(F+z)/m;if(!(Z<0||Z>=d.outHeight||Math.floor(Z)!==Z))for(let J=0;J<b;J+=x){let te=(G+J)/g;if(te<0||te>=d.outWidth||Math.floor(te)!==te)continue;let B=A*b-1-f.get($,Z,te,_),oe=z*b+J,Q=B===oe?1:0;if(Q===0)continue;K+=R.get($,Z,te,_)*Q}}S.set(K,$,P,C,_)}return n.makeTensorInfo(S.shape,S.dtype,S.values)}var jY={kernelName:o0,backendName:"cpu",kernelFunc:HY};function qY(e,t,n,s,r){let a=v.computeStrides(t),o=zx(e,t,n,a,r,"max"),i=nS(e,t,n,r,!0,s);return[o.values,i.values]}var XY={kernelName:l0,backendName:"cpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{x:s}=e,{filterSize:r,strides:a,pad:o,includeBatchInIndex:i}=t,l=n;Te(s,"MaxPoolWithArgmax");let u=l.data.get(s.dataId).values,c=T.computePool2DInfo(s.shape,r,a,[1,1],o),[p,d]=qY(u,s.shape,s.dtype,i,c),h=l.write(p,c.outShape,s.dtype),f=l.write(d,c.outShape,s.dtype);return[{dataId:h,shape:c.outShape,dtype:s.dtype},{dataId:f,shape:c.outShape,dtype:"int32"}]}};function KY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s,i=v.parseAxisParam(a,r.shape),u=T.computeOutAndReduceShapes(r.shape,i)[1],c=v.sizeFromShape(u),p=[],d=n.makeTensorInfo([],"float32",new Float32Array([c]));p.push(d);let h=io({inputs:{x:r},backend:n,attrs:{dtype:"float32"}});p.push(h);let f=Lx({inputs:{a:h,b:d},backend:n});p.push(f);let m=Ph({inputs:{x:f},backend:n,attrs:{axis:a,keepDims:o}});return p.forEach(g=>n.disposeIntermediateTensorInfo(g)),m}var ZY={kernelName:Mo,backendName:"cpu",kernelFunc:KY};function YY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;Te(r,"min");let i=v.parseAxisParam(a,r.shape),l=i,u=T.getAxesPermutation(l,r.shape.length),c=r;u!=null&&(c=gs({inputs:{x:r},backend:n,attrs:{perm:u}}),l=T.getInnerMostAxes(l.length,r.shape.length)),T.assertAxesAreInnerMostDims("min",l,c.shape.length);let[p,d]=T.computeOutAndReduceShapes(c.shape,l),h=v.sizeFromShape(d),f=v.makeZerosTypedArray(v.sizeFromShape(p),c.dtype),m=n.data.get(c.dataId).values;for(let y=0;y<f.length;++y){let x=y*h,A=m[x];for(let b=0;b<h;++b){let w=m[x+b];(Number.isNaN(w)||w<A)&&(A=w)}f[y]=A}u!=null&&n.disposeIntermediateTensorInfo(c);let g=n.makeTensorInfo(p,c.dtype,f);if(o){let y=T.expandShapeToKeepDim(p,i),x=Rt({inputs:{x:g},backend:n,attrs:{shape:y}});return n.disposeIntermediateTensorInfo(g),x}return g}var JY={kernelName:zo,backendName:"cpu",kernelFunc:YY};function QY(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{paddings:a,mode:o}=s;Te(r,"mirrorPad");let i=a.map((A,b)=>A[0]+r.shape[b]+A[1]),l=a.map(A=>A[0]),u=a.map((A,b)=>A[0]+r.shape[b]),c=o==="reflect"?0:1,p=n.data.get(r.dataId).values,d=r.shape.length,h=v.computeStrides(r.shape),f=v.sizeFromShape(i),m=i.length,g=v.computeStrides(i),y=v.getTypedArrayFromDType(r.dtype,f);for(let A=0;A<f;A++){let b=v.indexToLoc(A,m,g);for(let k=0;k<m;k++)b[k]<l[k]?b[k]=l[k]*2-b[k]-c:b[k]>=u[k]&&(b[k]=(u[k]-1)*2-b[k]+c);b=b.map((k,S)=>k-l[S]);let w=v.locToIndex(b,d,h);y[A]=p[w]}return{dataId:n.write(y,i,r.dtype),shape:i,dtype:r.dtype}}var eJ={kernelName:Bo,backendName:"cpu",kernelFunc:QY},tJ=un((e,t)=>{let n=e%t;return e<0&&t<0||e>=0&&t>=0?n:(n+t)%t}),nJ=kn(Ec,tJ),sJ={kernelName:Ec,backendName:"cpu",kernelFunc:nJ},rJ=uo(Wm());function lS(e){let{inputs:t,backend:n,attrs:s}=e,{logits:r}=t,{dim:a}=s,o=r.shape.length,i=a;if(i===-1&&(i=o-1),i!==o-1)throw Error(`Softmax along a non-last dimension is not yet supported. Logits was rank ${o} and dim was ${i}`);let l=v.parseAxisParam([i],r.shape),u=iS({inputs:{x:r},backend:n,attrs:{reductionIndices:l,keepDims:!1}}),c=T.expandShapeToKeepDim(u.shape,l),p=Rt({inputs:{x:u},backend:n,attrs:{shape:c}}),d=Mx({inputs:{a:r,b:p},backend:n}),h=wI({inputs:{x:d},backend:n}),f=Ph({inputs:{x:h},backend:n,attrs:{axis:l,keepDims:!1}}),m=Rt({inputs:{x:f},backend:n,attrs:{shape:c}}),g=Lx({inputs:{a:h,b:m},backend:n});return n.disposeIntermediateTensorInfo(u),n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(f),n.disposeIntermediateTensorInfo(m),g}var aJ={kernelName:ti,backendName:"cpu",kernelFunc:lS};function oJ(e){let{inputs:t,backend:n,attrs:s}=e,{logits:r}=t,{numSamples:a,seed:o,normalized:i}=s;Te(r,"multinomial");let l=i?r:lS({inputs:{logits:r},backend:n,attrs:{dim:-1}}),u=l.shape[0],c=l.shape[1],p=n.data.get(l.dataId).values,d=[u,a],h=v.makeZerosTypedArray(v.sizeFromShape(d),"int32");for(let f=0;f<u;++f){let m=f*c,g=new Float32Array(c-1);g[0]=p[m];for(let A=1;A<g.length;++A)g[A]=g[A-1]+p[m+A];let y=rJ.alea(o.toString()),x=f*a;for(let A=0;A<a;++A){let b=y();h[x+A]=g.length;for(let w=0;w<g.length;w++)if(b<g[w]){h[x+A]=w;break}}}return i||n.disposeIntermediateTensorInfo(l),n.makeTensorInfo(d,"int32",h)}var iJ={kernelName:u0,backendName:"cpu",kernelFunc:oJ},lJ=pr.nonMaxSuppressionV3Impl;function uJ(e){let{inputs:t,backend:n,attrs:s}=e,{boxes:r,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l}=s;Te(r,"NonMaxSuppression");let u=n.data.get(r.dataId).values,c=n.data.get(a.dataId).values,{selectedIndices:p}=lJ(u,c,o,i,l);return n.makeTensorInfo([p.length],"int32",new Int32Array(p))}var cJ={kernelName:Cl,backendName:"cpu",kernelFunc:uJ},dJ=pr.nonMaxSuppressionV4Impl;function pJ(e){let{inputs:t,backend:n,attrs:s}=e,{boxes:r,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,padToMaxOutputSize:u}=s;Te(r,"NonMaxSuppressionPadded");let c=n.data.get(r.dataId).values,p=n.data.get(a.dataId).values,{selectedIndices:d,validOutputs:h}=dJ(c,p,o,i,l,u);return[n.makeTensorInfo([d.length],"int32",new Int32Array(d)),n.makeTensorInfo([],"int32",new Int32Array([h]))]}var hJ={kernelName:Rc,backendName:"cpu",kernelFunc:pJ},fJ=pr.nonMaxSuppressionV5Impl;function mJ(e){let{inputs:t,backend:n,attrs:s}=e,{boxes:r,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,softNmsSigma:u}=s;Te(r,"NonMaxSuppressionWithScore");let c=n.data.get(r.dataId).values,p=n.data.get(a.dataId).values,d=o,h=i,f=l,m=u,{selectedIndices:g,selectedScores:y}=fJ(c,p,d,h,f,m);return[n.makeTensorInfo([g.length],"int32",new Int32Array(g)),n.makeTensorInfo([y.length],"float32",new Float32Array(y))]}var gJ={kernelName:Tl,backendName:"cpu",kernelFunc:mJ};function yJ(e){let{inputs:t,backend:n,attrs:s}=e,{indices:r}=t,{depth:a,onValue:o,offValue:i}=s;Te(r,"oneHot");let l=v.sizeFromShape(r.shape),u=new Float32Array(l*a);u.fill(i);let c=n.data.get(r.dataId).values;for(let p=0;p<l;++p)c[p]>=0&&c[p]<a&&(u[p*a+c[p]]=o);return n.makeTensorInfo([...r.shape,a],"int32",u)}var AJ={kernelName:El,backendName:"cpu",kernelFunc:yJ};function Pm(e){let{inputs:t,backend:n}=e,{x:s}=t;if(s.dtype==="string")throw new Error("zerosLike is not supported for string tensors");if(s.dtype==="complex64"){let r=tl({inputs:{input:s},backend:n}),a=Pm({inputs:{x:r},backend:n}),o=ic({inputs:{input:s},backend:n}),i=Pm({inputs:{x:o},backend:n}),l=Cs({inputs:{real:a,imag:i},backend:n});return n.disposeIntermediateTensorInfo(r),n.disposeIntermediateTensorInfo(a),n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}else return Bx({backend:n,attrs:{shape:s.shape,value:0,dtype:s.dtype}})}var xJ={kernelName:Hl,backendName:"cpu",kernelFunc:Pm};function uS(e){let{inputs:t,backend:n}=e,{x:s}=t;if(s.dtype==="string")throw new Error("onesLike is not supported for string tensors");if(s.dtype==="complex64"){let r=tl({inputs:{input:s},backend:n}),a=uS({inputs:{x:r},backend:n}),o=ic({inputs:{input:s},backend:n}),i=Pm({inputs:{x:o},backend:n}),l=Cs({inputs:{real:a,imag:i},backend:n});return n.disposeIntermediateTensorInfo(r),n.disposeIntermediateTensorInfo(a),n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}else return Bx({backend:n,attrs:{shape:s.shape,value:1,dtype:s.dtype}})}var bJ={kernelName:Nl,backendName:"cpu",kernelFunc:uS};function cS(e){let{inputs:t,backend:n,attrs:s}=e,{axis:r}=s;if(t.length===1)return $m({inputs:{input:t[0]},backend:n,attrs:{dim:r}});let a=t[0].shape,o=t[0].dtype;t.forEach(c=>{v.assertShapesMatch(a,c.shape,"All tensors passed to stack must have matching shapes"),v.assert(o===c.dtype,()=>"All tensors passed to stack must have matching dtypes")});let i=[],l=t.map(c=>{let p=$m({inputs:{input:c},backend:n,attrs:{dim:r}});return i.push(p),p}),u=lc({inputs:l,backend:n,attrs:{axis:r}});return i.forEach(c=>n.disposeIntermediateTensorInfo(c)),u}var vJ={kernelName:Rl,backendName:"cpu",kernelFunc:cS};function wJ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{paddings:a,constantValue:o}=s;Te(r,"pad");let i=a.map((x,A)=>x[0]+r.shape[A]+x[1]),l=a.map(x=>x[0]),u=n.data.get(r.dataId).values,c=v.sizeFromShape(r.shape),p=r.shape.length,d=v.computeStrides(r.shape),h=v.sizeFromShape(i),f=i.length,m=v.computeStrides(i),g=v.getTypedArrayFromDType(r.dtype,h);o!==0&&g.fill(o);for(let x=0;x<c;x++){let b=v.indexToLoc(x,p,d).map((k,S)=>k+l[S]),w=v.locToIndex(b,f,m);g[w]=u[x]}return{dataId:n.write(g,i,r.dtype),shape:i,dtype:r.dtype}}var dS={kernelName:Vo,backendName:"cpu",kernelFunc:wJ},kJ=un((e,t)=>Math.pow(e,t)),IJ=kn(Uo,kJ),SJ={kernelName:Uo,backendName:"cpu",kernelFunc:IJ};function CJ(e){let{backend:t,attrs:n}=e,{start:s,stop:r,dtype:a,step:o}=n,i=Dx(s,r,o,a);return t.makeTensorInfo([i.length],a,i)}var TJ={kernelName:_c,backendName:"cpu",kernelFunc:CJ},NJ=bt(Dc,e=>1/e),EJ={kernelName:Dc,backendName:"cpu",kernelFunc:NJ};function RJ(e){let{inputs:t,backend:n,attrs:s}=e,{images:r}=t,{alignCorners:a,halfPixelCenters:o,size:i}=s;Te(r,"resizeBilinear");let l=v.computeStrides(r.shape),[u,c]=i,[p,d,h,f]=r.shape,m=n.data.get(r.dataId).values,g=new Float32Array(v.sizeFromShape([p,u,c,f])),y=[a&&u>1?d-1:d,a&&c>1?h-1:h],x=[a&&u>1?u-1:u,a&&c>1?c-1:c],A=0,b=y[0]/x[0],w=y[1]/x[1];for(let k=0;k<p;k++)for(let S=0;S<u;S++){let E;o?E=b*(S+.5)-.5:E=b*S;let R=Math.max(0,Math.floor(E)),$=E-R,_=Math.min(d-1,Math.ceil(E)),P=k*l[0]+R*l[1],C=k*l[0]+_*l[1];for(let F=0;F<c;F++){let G;o?G=w*(F+.5)-.5:G=w*F;let K=Math.max(0,Math.floor(G)),z=G-K,Z=Math.min(h-1,Math.ceil(G)),J=P+K*l[2],te=C+K*l[2],B=P+Z*l[2],oe=C+Z*l[2];for(let Q=0;Q<f;Q++){let ae=m[J+Q],ie=m[te+Q],me=m[B+Q],we=m[oe+Q],Re=ae+(me-ae)*z,_e=ie+(we-ie)*z,Be=Re+(_e-Re)*$;g[A++]=Be}}}return n.makeTensorInfo([p,u,c,f],"float32",g)}var _J={kernelName:Xo,backendName:"cpu",kernelFunc:RJ};function DJ(e){let{inputs:t,backend:n,attrs:s}=e,{images:r,dy:a}=t,{alignCorners:o}=s;Te([a,r],"resizeBilinearGrad");let i=v.computeStrides(r.shape),[l,u,c,p]=r.shape,[,d,h]=a.shape,f=new Float32Array(l*u*c*p),m=[o&&d>1?u-1:u,o&&h>1?c-1:c],g=[o&&d>1?d-1:d,o&&h>1?h-1:h],y=m[0]/g[0],x=m[1]/g[1],A=n.data.get(a.dataId).values,b=0;for(let w=0;w<l;w++){let k=w*i[0];for(let S=0;S<d;S++){let E=S*y,R=Math.floor(E),$=Math.min(Math.ceil(E),u-1),_=k+R*i[1],P=k+$*i[1],C=E-R,F=1-C;for(let G=0;G<h;G++){let K=G*x,z=Math.floor(K),Z=Math.min(Math.ceil(K),c-1),J=K-z,te=1-J,B=_+z*i[2],oe=_+Z*i[2],Q=P+z*i[2],ae=P+Z*i[2],ie=F*te,me=F*J,we=C*te,Re=C*J;for(let _e=0;_e<p;_e++){let Be=A[b++];f[B+_e]+=Be*ie,f[oe+_e]+=Be*me,f[Q+_e]+=Be*we,f[ae+_e]+=Be*Re}}}}return n.makeTensorInfo([l,c,u,p],"float32",f)}var $J={kernelName:d0,backendName:"cpu",kernelFunc:DJ};function PJ(e){let{inputs:t,backend:n,attrs:s}=e,{images:r}=t,{alignCorners:a,halfPixelCenters:o,size:i}=s;Te(r,"resizeNearestNeighbor");let l=v.computeStrides(r.shape),[u,c]=i,[p,d,h,f]=r.shape,m=n.data.get(r.dataId).values,g=new Float32Array(p*u*c*f),y=[a&&u>1?d-1:d,a&&c>1?h-1:h],x=[a&&u>1?u-1:u,a&&c>1?c-1:c],A=y[0]/x[0],b=y[1]/x[1],w=0;for(let k=0;k<p;k++){let S=k*l[0];for(let E=0;E<u;E++){let R=o?A*(E+.5):A*E,$=Math.min(d-1,a?Math.round(R):Math.floor(R));o&&($=Math.max(0,$));let _=S+$*l[1];for(let P=0;P<c;P++){let C=o?b*(P+.5):b*P,F=Math.min(h-1,a?Math.round(C):Math.floor(C));o&&(F=Math.max(0,F));let G=_+F*l[2];for(let K=0;K<f;K++){let z=m[G+K];g[w++]=z}}}}return n.makeTensorInfo([p,u,c,f],r.dtype,g)}var FJ={kernelName:qo,backendName:"cpu",kernelFunc:PJ};function OJ(e){let{inputs:t,backend:n,attrs:s}=e,{images:r,dy:a}=t,{alignCorners:o}=s;Te([a,r],"resizeNearestNeighborGrad");let i=v.computeStrides(r.shape),l=v.computeStrides(a.shape),[u,c,p,d]=r.shape,[,h,f]=a.shape,m=new Float32Array(u*c*p*d),g=n.data.get(a.dataId).values,y=[o&&h>1?c-1:c,o&&f>1?p-1:p],x=[o&&h>1?h-1:h,o&&f>1?f-1:f],A=y[0]/x[0],b=y[1]/x[1],w=1/A,k=1/b,S=Math.ceil(w)*2+2,E=Math.ceil(k)*2+2;for(let R=0;R<u;R++){let $=R*i[0];for(let _=0;_<c;_++){let P=$+_*i[1],C=Math.floor(_*w),F=Math.floor(C-S/2);for(let G=0;G<p;G++){let K=P+G*i[2],z=Math.floor(G*k),Z=Math.floor(z-E/2);for(let J=0;J<d;J++){let te=0;for(let B=0;B<S;B++){let oe=B+F;if(oe<0||oe>=h)continue;let Q=$+oe*l[1],ae=oe*A,ie=Math.min(c-1,o?Math.round(ae):Math.floor(ae));if(_===ie)for(let me=0;me<E;me++){let we=me+Z;if(we<0||we>=f)continue;let Re=Q+we*l[2],_e=we*b,Be=Math.min(p-1,o?Math.round(_e):Math.floor(_e));G===Be&&(te+=g[Re+J])}}m[K+J]=te}}}}return n.makeTensorInfo(r.shape,r.dtype,m)}var MJ={kernelName:c0,backendName:"cpu",kernelFunc:OJ};function zJ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{dims:a}=s;Te(r,"reverse");let o=r.shape.length,i=v.parseAxisParam(a,r.shape);if(o===0)return Zr({inputs:{x:r},backend:n});let l=new fn(r.shape,r.dtype),u=n.bufferSync(r);for(let c=0;c<l.size;c++){let p=l.indexToLoc(c),d=p.slice();i.forEach(h=>d[h]=r.shape[h]-1-d[h]),l.set(u.get(...d),...p)}return n.makeTensorInfo(l.shape,l.dtype,l.values)}var LJ={kernelName:Dl,backendName:"cpu",kernelFunc:zJ},BJ={kernelName:jl,backendName:"cpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{image:s}=e,{radians:r,fillValue:a,center:o}=t,i=n,l=v.getTypedArrayFromDType(s.dtype,v.sizeFromShape(s.shape)),[u,c,p,d]=s.shape,[h,f]=T.getImageCenter(o,c,p),m=255,g=Math.sin(r),y=Math.cos(r),x=i.data.get(s.dataId).values;for(let b=0;b<u;b++){let w=b*p*c*d;for(let k=0;k<c;k++){let S=k*(p*d);for(let E=0;E<p;E++){let R=E*d;for(let $=0;$<d;$++){let _=[u,k,E,$],P=_[2],C=_[1],F=(P-h)*y-(C-f)*g,G=(P-h)*g+(C-f)*y;F=Math.round(F+h),G=Math.round(G+f);let K=a;if(typeof a!="number"&&($===3?K=m:K=a[$]),F>=0&&F<p&&G>=0&&G<c){let Z=G*(p*d),J=F*d,te=w+Z+J+$;K=x[te]}let z=w+S+R+$;l[z]=K}}}}return{dataId:i.write(l,s.shape,s.dtype),shape:s.shape,dtype:s.dtype}}},WJ=bt($l,e=>{let t=Math.floor(e);return e-t<.5?Math.floor(e):e-t>.5?Math.ceil(e):t%2===0?t:t+1}),VJ={kernelName:$l,backendName:"cpu",kernelFunc:WJ};function UJ(e){let{inputs:t,backend:n,attrs:s}=e,{indices:r,updates:a}=t,{shape:o}=s,{sliceRank:i,numUpdates:l,sliceSize:u,strides:c,outputSize:p}=T.calculateShapes(a,r,o),d=!0,h=n.bufferSync(r),f=n.bufferSync(a),m=Bu(h,f,o,p,u,l,i,c,0,d);return n.makeTensorInfo(o,m.dtype,m.values)}var GJ={kernelName:Pl,backendName:"cpu",kernelFunc:UJ};function HJ(e,t){let n=0,s=e.length,r=0;for(;n<s;)r=Math.floor((n+s)/2),e[r]<t?n=r+1:s=r;return s}function jJ(e,t){let n=0,s=e.length,r=0;for(;n<s;)r=Math.floor((n+s)/2),e[r]<=t?n=r+1:s=r;return s}function qJ(e,t,n,s,r,a){let o=v.getArrayFromDType("int32",n*r);for(let i=0;i<n;++i){let l=e.slice(i*s,(i+1)*s),u=i*r;for(let c=0;c<r;++c)o[u+c]=a==="left"?HJ(l,t[c+u]):jJ(l,t[c+u])}return o}function XJ(e){let{inputs:t,backend:n,attrs:s}=e,{sortedSequence:r,values:a}=t,{side:o}=s,i=n.data.get(r.dataId).values,l=n.data.get(a.dataId).values,u=qJ(i,l,r.shape[0],r.shape[1],a.shape[1],o);return n.makeTensorInfo(a.shape,"int32",u)}var KJ={kernelName:p0,backendName:"cpu",kernelFunc:XJ};function ZJ(e){let{inputs:t,backend:n}=e,{condition:s,t:r,e:a}=t;Te([s,r,a],"select");let o=s.shape.length,i=n.data.get(s.dataId).values,l=n.data.get(r.dataId).values,u=n.data.get(a.dataId).values,c=Mn(r.dtype,a.dtype),p=v.makeZerosTypedArray(v.sizeFromShape(r.shape),c),d=0,h=o===0||o>1||r.shape.length===1?1:v.sizeFromShape(r.shape.slice(1));for(let f=0;f<i.length;f++)for(let m=0;m<h;m++)i[f]===1?p[d++]=l[f]:p[d++]=u[f];return n.makeTensorInfo(r.shape,c,p)}var YJ={kernelName:Fl,backendName:"cpu",kernelFunc:ZJ},JJ=T.SELU_SCALEALPHA,QJ=T.SELU_SCALE,eQ=bt($c,e=>e>=0?QJ*e:JJ*(Math.exp(e)-1)),tQ={kernelName:$c,backendName:"cpu",kernelFunc:eQ},nQ=bt(Pc,e=>e<0?-1:e>0?1:0),sQ={kernelName:Pc,backendName:"cpu",kernelFunc:nQ},rQ=bt(Yo,e=>Math.sin(e)),aQ={kernelName:Yo,backendName:"cpu",kernelFunc:rQ},oQ=bt(Ml,e=>Math.sinh(e)),iQ={kernelName:Ml,backendName:"cpu",kernelFunc:oQ},lQ=11920928955078125e-23,p7=Math.log(lQ)+2,uQ=bt(Fc,e=>{let t=e>-p7,n=e<p7,s=Math.exp(e),r;return n?r=s:t?r=e:r=Math.log(1+s),r}),cQ={kernelName:Fc,backendName:"cpu",kernelFunc:uQ};function dQ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockShape:a,paddings:o}=s;Te([r],"spaceToBatchND");let i=v.sizeFromShape(a),l=[[0,0]];l.push(...o);for(let k=1+a.length;k<r.shape.length;++k)l.push([0,0]);let u=dS.kernelFunc({inputs:{x:r},backend:n,attrs:{paddings:l,constantValue:0}}),c=T.getReshaped(u.shape,a,i,!1),p=T.getPermuted(c.length,a.length,!1),d=T.getReshapedPermuted(u.shape,a,i,!1),m=Rt({inputs:{x:u},backend:n,attrs:{shape:c}}),x=gs({inputs:{x:m},backend:n,attrs:{perm:p}}),w=Rt({inputs:{x},backend:n,attrs:{shape:d}});return n.disposeIntermediateTensorInfo(u),n.disposeIntermediateTensorInfo(m),n.disposeIntermediateTensorInfo(x),w}var pQ={kernelName:zl,backendName:"cpu",kernelFunc:dQ};function hQ(e){let{inputs:t,backend:n}=e,{indices:s,values:r,denseShape:a,defaultValue:o}=t;if(a.shape.length!==1)throw new Error(`Dense shape must be a vector, saw:
|
|
${a.shape}`);if(s.shape.length!==2)throw new Error(`Indices must be a matrix, saw:
|
|
${s.shape}`);if(r.shape.length!==1)throw new Error(`Values must be a vector, saw:
|
|
${r.shape}`);if(o.shape.length!==0)throw new Error(`Default value must be a scalar, saw:
|
|
${o.shape}`);let i=n.data.get(s.dataId).values,l=n.data.get(r.dataId).values,u=n.data.get(a.dataId).values,c=n.data.get(o.dataId).values[0],[p,d,h,f,m]=WI(i,s.shape,s.dtype,l,r.dtype,u,c);return[n.makeTensorInfo(d,s.dtype,p),n.makeTensorInfo([d[0]],r.dtype,h),n.makeTensorInfo([f.length],"bool",new Uint8Array(f.map(g=>Number(g)))),n.makeTensorInfo([m.length],s.dtype,new Int32Array(m))]}var fQ={kernelName:Xp,backendName:"cpu",kernelFunc:hQ};function mQ(e){let{inputs:t,backend:n}=e,{inputIndices:s,inputShape:r,newShape:a}=t;if(s.shape.length!==2)throw new Error(`Input indices should be a matrix but received shape
|
|
${s.shape}`);if(r.shape.length!==1)throw new Error(`Input shape should be a vector but received shape
|
|
${r.shape}`);if(a.shape.length!==1)throw new Error(`Target shape should be a vector but received shape ${a.shape}`);let o=Array.from(n.data.get(r.dataId).values),i=n.data.get(s.dataId).values,l=Array.from(n.data.get(a.dataId).values),[u,c,p]=VI(i,s.shape,s.dtype,o,l);return[n.makeTensorInfo(c,s.dtype,u),n.makeTensorInfo([p.length],a.dtype,new Int32Array(p))]}var gQ={kernelName:Oc,backendName:"cpu",kernelFunc:mQ};function yQ(e){let{inputs:t,backend:n}=e,{data:s,indices:r,segmentIds:a}=t;if(s.shape.length<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(r.shape.length!==1)throw new Error(`Indices should be a vector but received shape
|
|
${r.shape}`);if(a.shape.length!==1)throw new Error(`Segment ids should be a vector but received shape
|
|
${a.shape}`);if(r.shape[0]!==a.shape[0])throw new Error("segmentIds and indices should have same size.");let o=n.data.get(s.dataId).values,i=n.data.get(r.dataId).values,l=n.data.get(a.dataId).values,[u,c]=$x(o,s.shape,s.dtype,i,l,!0);return n.makeTensorInfo(c,s.dtype,u)}var AQ={kernelName:Kp,backendName:"cpu",kernelFunc:yQ};function xQ(e){let{inputs:t,backend:n}=e,{data:s,indices:r,segmentIds:a}=t;if(s.shape.length<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(r.shape.length!==1)throw new Error(`Indices should be a vector but received shape
|
|
${r.shape}`);if(a.shape.length!==1)throw new Error(`Segment ids should be a vector but received shape
|
|
${a.shape}`);if(r.shape[0]!==a.shape[0])throw new Error("segmentIds and indices should have same size.");let o=n.data.get(s.dataId).values,i=n.data.get(r.dataId).values,l=n.data.get(a.dataId).values,[u,c]=$x(o,s.shape,s.dtype,i,l);return n.makeTensorInfo(c,s.dtype,u)}var bQ={kernelName:Zp,backendName:"cpu",kernelFunc:xQ};function vQ(e){let{inputs:t,backend:n,attrs:s}=e,{sparseIndices:r,sparseValues:a,defaultValue:o}=t,{outputShape:i}=s,{sliceRank:l,numUpdates:u,sliceSize:c,strides:p,outputSize:d}=T.calculateShapes(a,r,i),h=!1,f=n.bufferSync(r),m;switch(a.dtype){case"bool":{let g=n.bufferSync(a),y=Boolean(n.data.get(o.dataId).values[0]);m=Bu(f,g,i,d,c,u,l,p,y,h);break}case"float32":{let g=n.bufferSync(a),y=n.data.get(o.dataId).values[0];m=Bu(f,g,i,d,c,u,l,p,y,h);break}case"int32":{let g=n.bufferSync(a),y=n.data.get(o.dataId).values[0];m=Bu(f,g,i,d,c,u,l,p,y,h);break}case"string":{let g=n.bufferSync(a),y=v.decodeString(n.data.get(o.dataId).values[0]);m=Bu(f,g,i,d,c,u,l,p,y,h);break}default:throw new Error(`Unsupported type ${a.dtype}`)}return n.makeTensorInfo(i,m.dtype,m.values)}var wQ={kernelName:Yp,backendName:"cpu",kernelFunc:vQ};function kQ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{numOrSizeSplits:a,axis:o}=s,i=v.parseAxisParam(o,r.shape)[0],l=T.prepareSplitSize(r,a,i),u=new Array(r.shape.length).fill(0),c=r.shape.slice();return l.map(p=>{let d=[...c];d[i]=p;let h=nl({inputs:{x:r},backend:n,attrs:{begin:u,size:d}});return u[i]+=p,h})}var IQ={kernelName:Ll,backendName:"cpu",kernelFunc:kQ},SQ={kernelName:Mc,backendName:"cpu",kernelFunc:({inputs:e,backend:t})=>{let{x:n}=e,s=t;Te(n,"square");let r=s.data.get(n.dataId).values,a=new Float32Array(r.length);for(let i=0;i<r.length;++i){let l=r[i];a[i]=l*l}return{dataId:s.write(a,n.shape,n.dtype),shape:n.shape,dtype:n.dtype}}},CQ=bt(ai,(e,t)=>{let n=t;return isNaN(e)?NaN:e>0?1:n.alpha}),TQ={kernelName:ai,backendName:"cpu",kernelFunc:CQ};function NQ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{begin:a,end:o,strides:i,beginMask:l,endMask:u,ellipsisMask:c,newAxisMask:p,shrinkAxisMask:d}=s;Te(r,"stridedSlice");let{finalShapeSparse:h,finalShape:f,isIdentity:m,sliceDim0:g,isSimpleSlice:y,begin:x,end:A,strides:b}=Ut.sliceInfo(r.shape,a,o,i,l,u,c,p,d),w;if(m)w=Rt({inputs:{x:r},backend:n,attrs:{shape:f}});else if(g||y){v.assert(r.shape.length>=1,()=>`Input must have rank at least 1, got: ${r.shape.length}`);let k=Ut.computeOutShape(x,A,b),S=nl({inputs:{x:r},backend:n,attrs:{begin:x,size:k}});w=Rt({inputs:{x:S},backend:n,attrs:{shape:f}}),n.disposeIntermediateTensorInfo(S)}else{let k=n.bufferSync(r),S=GI(h,k,b,x);w=n.makeTensorInfo(f,S.dtype,S.values)}return w}var EQ={kernelName:Bl,backendName:"cpu",kernelFunc:NQ};function RQ(e){let{inputs:t,backend:n,attrs:s}=e,{separator:r,nGramWidths:a,leftPad:o,rightPad:i,padWidth:l,preserveShortSequences:u}=s,{data:c,dataSplits:p}=t,d=n.data.get(c.dataId).values,h=n.data.get(p.dataId).values,[f,m]=Px(d,h,r,a,o,i,l,u);return[n.makeTensorInfo([f.length],"string",f),n.makeTensorInfo(p.shape,"int32",m)]}var _Q={kernelName:zc,backendName:"cpu",kernelFunc:RQ};function DQ(e){let{inputs:t,backend:n,attrs:s}=e,{skipEmpty:r}=s,{input:a,delimiter:o}=t;if(a.dtype!=="string")throw new Error("Input must be of datatype string");if(a.shape.length!==1)throw new Error(`Input must be a vector, got shape: ${a.shape}`);if(o.shape.length!==0)throw new Error(`Delimiter must be a scalar, got shape: ${o.shape}`);let i=n.data.get(a.dataId).values,l=n.data.get(o.dataId).values[0],[u,c,p]=Fx(i,l,r),d=c.length;return[n.makeTensorInfo([d,2],"int32",u),n.makeTensorInfo([d],"string",c),n.makeTensorInfo([2],"int32",new Int32Array(p))]}var $Q={kernelName:Jp,backendName:"cpu",kernelFunc:DQ};function PQ(e){let{inputs:t,backend:n,attrs:s}=e,{numBuckets:r}=s,{input:a}=t;if(a.dtype!=="string")throw new Error("Input must be of datatype string");if(r<=0)throw new Error("Number of buckets must be at least 1");let o=n.data.get(a.dataId).values,i=Ox(o,r);return n.makeTensorInfo(a.shape,"int32",i)}var FQ={kernelName:Qp,backendName:"cpu",kernelFunc:PQ},OQ=bt(Wl,e=>Math.tan(e)),MQ={kernelName:Wl,backendName:"cpu",kernelFunc:OQ},zQ=bt(ri,e=>Math.tanh(e)),LQ={kernelName:ri,backendName:"cpu",kernelFunc:zQ};function BQ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{reps:a}=s;Te(r,"tile");let o=jI(n.bufferSync(r),a);return n.makeTensorInfo(o.shape,o.dtype,o.values)}var WQ={kernelName:va,backendName:"cpu",kernelFunc:BQ};function VQ(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{k:a,sorted:o}=s;Te(r,"topk");let i=n.data.get(r.dataId).values,[l,u]=XI(i,r.shape,r.dtype,a,o);return[n.makeTensorInfo(l.shape,l.dtype,l.values),n.makeTensorInfo(u.shape,u.dtype,u.values)]}var UQ={kernelName:Vl,backendName:"cpu",kernelFunc:VQ};function GQ(e){let{inputs:t,attrs:n,backend:s}=e,{image:r,transforms:a}=t,{interpolation:o,fillMode:i,fillValue:l,outputShape:u}=n,[c,p,d,h]=r.shape,[f,m]=u!=null?u:[p,d],g=[c,f,m,h],y=v.computeStrides(r.shape),x=y[0],A=y[1],b=y[2],w=v.getTypedArrayFromDType(r.dtype,v.sizeFromShape(g));w.fill(l);let k=s.data.get(r.dataId).values,S=s.data.get(a.dataId).values;for(let R=0;R<c;++R){let $=a.shape[0]===1?S:S.subarray(R*8,R*8+8);for(let _=0;_<f;++_)for(let P=0;P<m;++P)for(let C=0;C<h;++C){let F,G=$[6]*P+$[7]*_+1;if(G===0)continue;let K=($[0]*P+$[1]*_+$[2])/G,z=($[3]*P+$[4]*_+$[5])/G,Z=h7(K,d,i),J=h7(z,p,i);switch(o){case"nearest":F=ZQ(k,p,d,x,A,b,R,J,Z,C,l);break;case"bilinear":F=YQ(k,p,d,x,A,b,R,J,Z,C,l);break;default:throw new Error(`Error in Transform: Expect 'nearest' or 'bilinear', but got ${o}`)}let te=R*x+_*A+P*b+C;w[te]=F}return s.makeTensorInfo(g,r.dtype,w)}return{dataId:s.write(w,g,r.dtype),shape:r.shape,dtype:r.dtype}}var HQ={kernelName:Ul,backendName:"cpu",kernelFunc:GQ};function h7(e,t,n){switch(n){case"reflect":return jQ(e,t);case"wrap":return qQ(e,t);case"nearest":return KQ(e,t);case"constant":default:return XQ(e,t)}}function jQ(e,t){let n=e;if(n<0)if(t<=1)n=0;else{let s=2*t;n<s&&(n=s*Math.trunc(-n/s)+n),n=n<-t?n+s:-n-1}else if(n>t-1)if(t<=1)n=0;else{let s=2*t;n-=s*Math.trunc(n/s),n>=t&&(n=s-n-1)}return v.clamp(0,n,t-1)}function qQ(e,t){let n=e;if(n<0)if(t<=1)n=0;else{let s=t-1;n+=t*(Math.trunc(-n/s)+1)}else if(n>t-1)if(t<=1)n=0;else{let s=t-1;n-=t*Math.trunc(n/s)}return v.clamp(0,n,t-1)}function XQ(e,t){return e}function KQ(e,t){return v.clamp(0,e,t-1)}function lp(e,t,n,s,r,a,o,i,l,u,c){let p=o*s+i*r+l*a+u;return 0<=i&&i<t&&0<=l&&l<n?e[p]:c}function ZQ(e,t,n,s,r,a,o,i,l,u,c){let p=Math.round(i),d=Math.round(l);return lp(e,t,n,s,r,a,o,p,d,u,c)}function YQ(e,t,n,s,r,a,o,i,l,u,c){let p=Math.floor(i),d=Math.floor(l),h=p+1,f=d+1,m=(f-l)*lp(e,t,n,s,r,a,o,p,d,u,c)+(l-d)*lp(e,t,n,s,r,a,o,p,f,u,c),g=(f-l)*lp(e,t,n,s,r,a,o,h,d,u,c)+(l-d)*lp(e,t,n,s,r,a,o,h,f,u,c);return(h-i)*m+(i-p)*g}function JQ(e){let{inputs:t,attrs:n,backend:s}=e,{axis:r}=n,{x:a}=t;Te(a,"unique");let o=s.data.get(a.dataId).values,{outputValues:i,outputShape:l,indices:u}=KI(o,r,a.shape,a.dtype);return[s.makeTensorInfo(l,a.dtype,i),s.makeTensorInfo([u.length],"int32",u)]}var QQ={kernelName:h0,backendName:"cpu",kernelFunc:JQ};function eee(e){let{inputs:t,backend:n,attrs:s}=e,{value:r}=t,{axis:a}=s;a<0&&(a+=r.shape.length);let o=r.shape.length,i=r.shape[a],l=new Array(o-1),u=0;for(let h=0;h<o;h++)h!==a&&(l[u++]=r.shape[h]);let c=new Array(o).fill(0),p=r.shape.slice();p[a]=1;let d=new Array(i);for(let h=0;h<d.length;h++){c[a]=h;let f=nl({inputs:{x:r},backend:n,attrs:{begin:c,size:p}});d[h]=Rt({inputs:{x:f},backend:n,attrs:{shape:l}}),n.disposeIntermediateTensorInfo(f)}return d}var tee={kernelName:Gl,backendName:"cpu",kernelFunc:eee};function nee(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,segmentIds:a}=t,{numSegments:o}=s;Te(r,"unsortedSegmentSum");let i=r.shape.length,l=a.shape.length,u=[],c=[],p=i-l,d=a;for(let f=0;f<p;++f){let m=$m({inputs:{input:d},backend:n,attrs:{dim:f+1}});d=m,c.push(m)}for(let f=0;f<o;++f){let m=v.createScalarValue(f,"int32"),g=n.makeTensorInfo([],"int32",m),y=bI({inputs:{a:g,b:d},backend:n}),x=io({inputs:{x:y},backend:n,attrs:{dtype:"float32"}}),A=A2({inputs:{a:x,b:r},backend:n}),b=Ph({inputs:{x:A},backend:n,attrs:{axis:0,keepDims:!1}});u.push(b),c.push(g),c.push(y),c.push(x),c.push(A),c.push(b)}let h=cS({inputs:u,backend:n,attrs:{axis:0}});return c.forEach(f=>n.disposeIntermediateTensorInfo(f)),h}var see={kernelName:eh,backendName:"cpu",kernelFunc:nee},ree=[rK,Yq,oK,lK,sX,cK,pK,fK,gK,AK,bK,wK,IK,TK,EK,DK,PK,OK,zK,nK,BK,VK,GK,jK,tX,aX,XK,Jq,ZK,JK,QK,tZ,sZ,aZ,iZ,uZ,dZ,hZ,mZ,yZ,xZ,vZ,kZ,IZ,CZ,NZ,RZ,_Z,DZ,$Z,OZ,KX,zZ,oX,jZ,iX,qZ,uX,QZ,eY,nY,dX,aY,iY,uY,dY,hY,hX,mX,Qq,mY,YK,yY,xY,vY,ZX,yX,xX,kY,vX,SY,NY,RY,$Y,FY,MY,zY,kX,BY,VY,GY,jY,XY,ZY,JY,SX,eJ,sJ,iJ,TX,EX,cJ,hJ,gJ,_X,AJ,bJ,vJ,dS,SJ,JX,PX,TJ,eX,ey,EJ,QX,eK,tK,_J,$J,FJ,MJ,LJ,BJ,VJ,OX,GJ,KJ,YJ,tQ,zX,sQ,aQ,iQ,LX,aJ,cQ,pQ,fQ,gQ,AQ,bQ,wQ,IQ,VX,SQ,GX,TQ,EQ,_Q,$Q,FQ,XX,PZ,MQ,LQ,WQ,UQ,HQ,DX,QQ,tee,see,xJ];for(let e of ree)dr(e);var pS={};Ve(pS,{assertNotComplex:()=>nd,bindCanvasToFramebuffer:()=>mee,bindColorTextureToFramebuffer:()=>sm,bindTextureToProgramUniformSampler:()=>TS,bindTextureUnit:()=>IS,bindVertexBufferToProgramAttribute:()=>ny,callAndCheck:()=>Ie,canBeRepresented:()=>hS,createFragmentShader:()=>gS,createFramebuffer:()=>kS,createProgram:()=>yS,createStaticIndexBuffer:()=>bS,createStaticVertexBuffer:()=>xS,createTexture:()=>vS,createVertexShader:()=>mS,getBatchDim:()=>sl,getExtensionOrThrow:()=>up,getFramebufferErrorMessage:()=>NS,getMaxTexturesInShader:()=>DS,getNumChannels:()=>hee,getProgramUniformLocation:()=>CS,getProgramUniformLocationOrThrow:()=>SS,getRowsCols:()=>rl,getShapeAs3D:()=>rm,getTextureShapeFromLogicalShape:()=>RS,getWebGLDisjointQueryTimerVersion:()=>$S,getWebGLErrorMessage:()=>fS,getWebGLMaxTextureSize:()=>_S,hasExtension:()=>Ks,isCapableOfRenderingToFloatTexture:()=>PS,isDownloadFloatTextureEnabled:()=>FS,isReshapeFree:()=>Dp,isWebGLFenceEnabled:()=>OS,isWebGLVersionEnabled:()=>ry,linkProgram:()=>AS,logShaderSourceAndInfoLog:()=>Vx,resetMaxTextureSize:()=>gee,resetMaxTexturesInShader:()=>yee,unbindColorTextureFromFramebuffer:()=>sy,unbindTextureUnit:()=>fee,validateFramebuffer:()=>cp,validateProgram:()=>nm,validateTextureSize:()=>wS});var Bi={},f3={alpha:!1,antialias:!1,premultipliedAlpha:!1,preserveDrawingBuffer:!1,depth:!1,stencil:!1,failIfMajorPerformanceCaveat:!0};function x2(e,t){Bi[e]=t}function $r(e,t){if(!(e in Bi)||t!=null){let s=oee(e,t);if(s!==null)Bi[e]=s;else return console.log("Could not get context for WebGL version",e),null}let n=Bi[e];return n==null||n.isContextLost()?(delete Bi[e],$r(e)):(n.disable(n.DEPTH_TEST),n.disable(n.STENCIL_TEST),n.disable(n.BLEND),n.disable(n.DITHER),n.disable(n.POLYGON_OFFSET_FILL),n.disable(n.SAMPLE_COVERAGE),n.enable(n.SCISSOR_TEST),n.enable(n.CULL_FACE),n.cullFace(n.BACK),Bi[e])}function aee(e){if(typeof OffscreenCanvas!="undefined"&&e===2)return new OffscreenCanvas(300,150);if(typeof document!="undefined")return document.createElement("canvas");throw new Error("Cannot create a canvas in this context")}function oee(e,t){if(e!==1&&e!==2)throw new Error("Cannot get WebGL rendering context, WebGL is disabled.");let n=t==null?aee(e):t;return n.addEventListener("webglcontextlost",s=>{s.preventDefault(),delete Bi[e]},!1),e===1?n.getContext("webgl",f3)||n.getContext("experimental-webgl",f3):n.getContext("webgl2",f3)}var _p;(function(e){e[e.DENSE=0]="DENSE",e[e.SHARED_BATCH=1]="SHARED_BATCH"})(_p||(_p={}));var Xs;(function(e){e[e.RENDER=0]="RENDER",e[e.UPLOAD=1]="UPLOAD",e[e.PIXELS=2]="PIXELS",e[e.DOWNLOAD=3]="DOWNLOAD"})(Xs||(Xs={}));var Sn;(function(e){e[e.UNPACKED_FLOAT16=0]="UNPACKED_FLOAT16",e[e.UNPACKED_FLOAT32=1]="UNPACKED_FLOAT32",e[e.PACKED_4X1_UNSIGNED_BYTE=2]="PACKED_4X1_UNSIGNED_BYTE",e[e.PACKED_2X2_FLOAT32=3]="PACKED_2X2_FLOAT32",e[e.PACKED_2X2_FLOAT16=4]="PACKED_2X2_FLOAT16"})(Sn||(Sn={}));function Fh(e,t){return[t,e]}function iee(e,t){return e*t}function Xf(e){let t=v.sizeFromShape(e),n=Math.ceil(t/4);return v.sizeToSquarishShape(n)}function td(e,t){return[Math.max(1,Math.ceil(t/2)),Math.max(1,Math.ceil(e/2))]}function lee(e,t){let[n,s]=td(e,t);return n*s*4}function Wx(e,t){let n=e,s,r,a,o,i,l,u,c,p,d;return q().getNumber("WEBGL_VERSION")===2?(s=n.R32F,r=n.R16F,a=n.RGBA16F,o=n.RGBA32F,i=n.RED,u=4,c=1,p=n.HALF_FLOAT,d=n.FLOAT,l=n.RGBA8):(s=e.RGBA,r=e.RGBA,a=e.RGBA,o=n.RGBA,i=e.RGBA,u=4,c=4,p=t!=null?t.HALF_FLOAT_OES:null,d=e.FLOAT,l=e.RGBA),{internalFormatFloat:s,internalFormatHalfFloat:r,internalFormatPackedHalfFloat:a,internalFormatPackedFloat:o,textureFormatFloat:i,downloadTextureFormat:l,downloadUnpackNumChannels:u,defaultNumChannels:c,textureTypeHalfFloat:p,textureTypeFloat:d}}function Ie(e,t){let n=t();return q().getBool("DEBUG")&&uee(e),n}function uee(e){let t=e.getError();if(t!==e.NO_ERROR)throw new Error("WebGL Error: "+fS(e,t))}var cee=596e-10,dee=65504;function hS(e){return!!(q().getBool("WEBGL_RENDER_FLOAT32_ENABLED")||e===0||cee<Math.abs(e)&&Math.abs(e)<dee)}function fS(e,t){switch(t){case e.NO_ERROR:return"NO_ERROR";case e.INVALID_ENUM:return"INVALID_ENUM";case e.INVALID_VALUE:return"INVALID_VALUE";case e.INVALID_OPERATION:return"INVALID_OPERATION";case e.INVALID_FRAMEBUFFER_OPERATION:return"INVALID_FRAMEBUFFER_OPERATION";case e.OUT_OF_MEMORY:return"OUT_OF_MEMORY";case e.CONTEXT_LOST_WEBGL:return"CONTEXT_LOST_WEBGL";default:return`Unknown error code ${t}`}}function up(e,t){return Ia(e,()=>e.getExtension(t),'Extension "'+t+'" not supported on this browser.')}function mS(e,t){let n=Ia(e,()=>e.createShader(e.VERTEX_SHADER),"Unable to create vertex WebGLShader.");if(Ie(e,()=>e.shaderSource(n,t)),Ie(e,()=>e.compileShader(n)),e.getShaderParameter(n,e.COMPILE_STATUS)===!1)throw console.log(e.getShaderInfoLog(n)),new Error("Failed to compile vertex shader.");return n}function gS(e,t){let n=Ia(e,()=>e.createShader(e.FRAGMENT_SHADER),"Unable to create fragment WebGLShader.");if(Ie(e,()=>e.shaderSource(n,t)),Ie(e,()=>e.compileShader(n)),q().get("ENGINE_COMPILE_ONLY"))return n;if(e.getShaderParameter(n,e.COMPILE_STATUS)===!1)throw Vx(t,e.getShaderInfoLog(n)),new Error("Failed to compile fragment shader.");return n}var pee=/ERROR: [0-9]+:([0-9]+):/g;function Vx(e,t){let n=pee.exec(t);if(n==null){console.log(`Couldn't parse line number in error: ${t}`),console.log(e);return}let s=+n[1],r=e.split(`
|
|
`),a=r.length.toString().length+2,o=r.map((p,d)=>v.rightPad((d+1).toString(),a)+p),i=0;for(let p=0;p<o.length;p++)i=Math.max(o[p].length,i);let l=o.slice(0,s-1),u=o.slice(s-1,s),c=o.slice(s);console.log(l.join(`
|
|
`)),console.log(t.split(`
|
|
`)[0]),console.log(`%c ${v.rightPad(u[0],i)}`,"border:1px solid red; background-color:#e3d2d2; color:#a61717"),console.log(c.join(`
|
|
`))}function yS(e){return Ia(e,()=>e.createProgram(),"Unable to create WebGLProgram.")}function AS(e,t){if(Ie(e,()=>e.linkProgram(t)),!q().get("ENGINE_COMPILE_ONLY")&&e.getProgramParameter(t,e.LINK_STATUS)===!1)throw console.log(e.getProgramInfoLog(t)),new Error("Failed to link vertex and fragment shaders.")}function nm(e,t){if(Ie(e,()=>e.validateProgram(t)),e.getProgramParameter(t,e.VALIDATE_STATUS)===!1)throw console.log(e.getProgramInfoLog(t)),new Error("Shader program validation failed.")}function xS(e,t){let n=Ia(e,()=>e.createBuffer(),"Unable to create WebGLBuffer");return Ie(e,()=>e.bindBuffer(e.ARRAY_BUFFER,n)),Ie(e,()=>e.bufferData(e.ARRAY_BUFFER,t,e.STATIC_DRAW)),n}function bS(e,t){let n=Ia(e,()=>e.createBuffer(),"Unable to create WebGLBuffer");return Ie(e,()=>e.bindBuffer(e.ELEMENT_ARRAY_BUFFER,n)),Ie(e,()=>e.bufferData(e.ELEMENT_ARRAY_BUFFER,t,e.STATIC_DRAW)),n}function hee(){return q().getNumber("WEBGL_VERSION")===2?1:4}function vS(e){return Ia(e,()=>e.createTexture(),"Unable to create WebGLTexture.")}function wS(e,t){let n=q().getNumber("WEBGL_MAX_TEXTURE_SIZE");if(e<=0||t<=0){let s=`[${e}x${t}]`;throw new Error("Requested texture size "+s+" is invalid.")}if(e>n||t>n){let s=`[${e}x${t}]`,r=`[${n}x${n}]`;throw new Error("Requested texture size "+s+" greater than WebGL maximum on this browser / GPU "+r+".")}}function kS(e){return Ia(e,()=>e.createFramebuffer(),"Unable to create WebGLFramebuffer.")}function ny(e,t,n,s,r,a,o){let i=e.getAttribLocation(t,n);return i===-1?!1:(Ie(e,()=>e.bindBuffer(e.ARRAY_BUFFER,s)),Ie(e,()=>e.vertexAttribPointer(i,r,e.FLOAT,!1,a,o)),Ie(e,()=>e.enableVertexAttribArray(i)),!0)}function IS(e,t,n){ES(e,n),Ie(e,()=>e.activeTexture(e.TEXTURE0+n)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,t))}function fee(e,t){ES(e,t),Ie(e,()=>e.activeTexture(e.TEXTURE0+t)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,null))}function SS(e,t,n){return Ia(e,()=>e.getUniformLocation(t,n),'uniform "'+n+'" not present in program.')}function CS(e,t,n){return e.getUniformLocation(t,n)}function TS(e,t,n,s){Ie(e,()=>IS(e,t,s)),Ie(e,()=>e.uniform1i(n,s))}function mee(e){Ie(e,()=>e.bindFramebuffer(e.FRAMEBUFFER,null)),Ie(e,()=>e.viewport(0,0,e.canvas.width,e.canvas.height)),Ie(e,()=>e.scissor(0,0,e.canvas.width,e.canvas.height))}function sm(e,t,n){Ie(e,()=>e.bindFramebuffer(e.FRAMEBUFFER,n)),Ie(e,()=>e.framebufferTexture2D(e.FRAMEBUFFER,e.COLOR_ATTACHMENT0,e.TEXTURE_2D,t,0))}function sy(e,t){Ie(e,()=>e.bindFramebuffer(e.FRAMEBUFFER,t)),Ie(e,()=>e.framebufferTexture2D(e.FRAMEBUFFER,e.COLOR_ATTACHMENT0,e.TEXTURE_2D,null,0))}function cp(e){let t=e.checkFramebufferStatus(e.FRAMEBUFFER);if(t!==e.FRAMEBUFFER_COMPLETE)throw new Error("Error binding framebuffer: "+NS(e,t))}function NS(e,t){switch(t){case e.FRAMEBUFFER_INCOMPLETE_ATTACHMENT:return"FRAMEBUFFER_INCOMPLETE_ATTACHMENT";case e.FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT:return"FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT";case e.FRAMEBUFFER_INCOMPLETE_DIMENSIONS:return"FRAMEBUFFER_INCOMPLETE_DIMENSIONS";case e.FRAMEBUFFER_UNSUPPORTED:return"FRAMEBUFFER_UNSUPPORTED";default:return`unknown error ${t}`}}function Ia(e,t,n){let s=Ie(e,()=>t());if(s==null)throw new Error(n);return s}function ES(e,t){let n=e.MAX_COMBINED_TEXTURE_IMAGE_UNITS-1,s=t+e.TEXTURE0;if(s<e.TEXTURE0||s>n){let r=`[gl.TEXTURE0, gl.TEXTURE${n}]`;throw new Error(`textureUnit must be in ${r}.`)}}function sl(e,t=2){return v.sizeFromShape(e.slice(0,e.length-t))}function rl(e){if(e.length===0)throw Error("Cannot get rows and columns of an empty shape array.");return[e.length>1?e[e.length-2]:1,e[e.length-1]]}function rm(e){let t=[1,1,1];return e.length===0||e.length===1&&e[0]===1||(t=[sl(e),...rl(e)]),t}function RS(e,t=!1){let n=q().getNumber("WEBGL_MAX_TEXTURE_SIZE");t&&(n=n*2,e=e.map((r,a)=>a>=e.length-2?v.nearestLargerEven(e[a]):e[a]),e.length===1&&(e=[2,e[0]])),e.length!==2&&(e=v.squeezeShape(e).newShape);let s=v.sizeFromShape(e);if(e.length<=1&&s<=n)return[1,s];if(e.length===2&&e[0]<=n&&e[1]<=n)return e;if(e.length===3&&e[0]*e[1]<=n&&e[2]<=n)return[e[0]*e[1],e[2]];if(e.length===3&&e[0]<=n&&e[1]*e[2]<=n)return[e[0],e[1]*e[2]];if(e.length===4&&e[0]*e[1]*e[2]<=n&&e[3]<=n)return[e[0]*e[1]*e[2],e[3]];if(e.length===4&&e[0]<=n&&e[1]*e[2]*e[3]<=n)return[e[0],e[1]*e[2]*e[3]];if(t){let r=sl(e),a=2,o=2;return e.length&&([a,o]=rl(e)),s=r*(a/2)*(o/2),v.sizeToSquarishShape(s).map(i=>i*2)}return v.sizeToSquarishShape(s)}function Kf(e){return e%2===0}function Dp(e,t){if(e=e.slice(-2),t=t.slice(-2),v.arraysEqual(e,t)||!e.length||!t.length||e[0]===0||e[1]===0||t[0]===0||t[1]===0)return!0;if(e.length!==t.length){let n=e.slice(-1)[0],s=t.slice(-1)[0];if(n===s||Kf(n)&&Kf(s)&&(e[0]===1||t[0]===1))return!0}return e[1]===t[1]&&Kf(e[0])&&Kf(t[0])}var am,om;function _S(e){if(am==null){let t=$r(e);am=t.getParameter(t.MAX_TEXTURE_SIZE)}return am}function gee(){am=null}function yee(){om=null}function DS(e){if(om==null){let t=$r(e);om=t.getParameter(t.MAX_TEXTURE_IMAGE_UNITS)}return Math.min(16,om)}function $S(e){if(e===0)return 0;let t,n=$r(e);return Ks(n,"EXT_disjoint_timer_query_webgl2")&&e===2?t=2:Ks(n,"EXT_disjoint_timer_query")?t=1:t=0,t}function Ks(e,t){return e.getExtension(t)!=null}function ry(e){try{if($r(e)!=null)return!0}catch(t){return console.log("Error when getting WebGL context: ",t),!1}return!1}function PS(e){if(e===0)return!1;let t=$r(e);if(e===1){if(!Ks(t,"OES_texture_float"))return!1}else if(!Ks(t,"EXT_color_buffer_float"))return!1;return ay(t)}function FS(e){if(e===0)return!1;let t=$r(e);if(e===1){if(!Ks(t,"OES_texture_float")||!Ks(t,"WEBGL_color_buffer_float"))return!1}else{if(Ks(t,"EXT_color_buffer_float"))return ay(t);let s="EXT_color_buffer_half_float";if(Ks(t,s)){let r=t.getExtension(s);return Aee(t,r)}return!1}return ay(t)}function ay(e){let t=Wx(e),n=e.createTexture();e.bindTexture(e.TEXTURE_2D,n);let s=1,r=1;e.texImage2D(e.TEXTURE_2D,0,t.internalFormatFloat,s,r,0,t.textureFormatFloat,t.textureTypeFloat,null);let a=e.createFramebuffer();e.bindFramebuffer(e.FRAMEBUFFER,a),e.framebufferTexture2D(e.FRAMEBUFFER,e.COLOR_ATTACHMENT0,e.TEXTURE_2D,n,0);let o=e.checkFramebufferStatus(e.FRAMEBUFFER)===e.FRAMEBUFFER_COMPLETE;return e.bindTexture(e.TEXTURE_2D,null),e.bindFramebuffer(e.FRAMEBUFFER,null),e.deleteTexture(n),e.deleteFramebuffer(a),o}function Aee(e,t){let n=Wx(e,t),s=e.createTexture();e.bindTexture(e.TEXTURE_2D,s);let r=1,a=1;e.texImage2D(e.TEXTURE_2D,0,n.internalFormatHalfFloat,r,a,0,n.textureFormatFloat,n.textureTypeHalfFloat,null);let o=e.createFramebuffer();e.bindFramebuffer(e.FRAMEBUFFER,o),e.framebufferTexture2D(e.FRAMEBUFFER,e.COLOR_ATTACHMENT0,e.TEXTURE_2D,s,0);let i=e.checkFramebufferStatus(e.FRAMEBUFFER)===e.FRAMEBUFFER_COMPLETE;return e.bindTexture(e.TEXTURE_2D,null),e.bindFramebuffer(e.FRAMEBUFFER,null),e.deleteTexture(s),e.deleteFramebuffer(o),i}function OS(e){return e!==2?!1:$r(e).fenceSync!=null}function nd(e,t){Array.isArray(e)||(e=[e]),e.forEach(n=>{n!=null&&v.assert(n.dtype!=="complex64",()=>`${t} does not support complex64 tensors in the WebGL backend.`)})}var Pe=q();Pe.registerFlag("HAS_WEBGL",()=>Pe.getNumber("WEBGL_VERSION")>0);Pe.registerFlag("WEBGL_VERSION",()=>ry(2)?2:ry(1)?1:0);Pe.registerFlag("WEBGL_CHECK_NUMERICAL_PROBLEMS",()=>!1);Pe.registerFlag("WEBGL_BUFFER_SUPPORTED",()=>Pe.get("WEBGL_VERSION")===2);Pe.registerFlag("WEBGL_CPU_FORWARD",()=>!0);Pe.registerFlag("WEBGL_FORCE_F16_TEXTURES",()=>!1);Pe.registerFlag("WEBGL_PACK",()=>Pe.getBool("HAS_WEBGL"));Pe.registerFlag("WEBGL_PACK_NORMALIZATION",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_PACK_CLIP",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_PACK_DEPTHWISECONV",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_PACK_BINARY_OPERATIONS",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_PACK_UNARY_OPERATIONS",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_PACK_ARRAY_OPERATIONS",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_PACK_IMAGE_OPERATIONS",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_PACK_REDUCE",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_LAZILY_UNPACK",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_CONV_IM2COL",()=>Pe.getBool("WEBGL_PACK"));Pe.registerFlag("WEBGL_MAX_TEXTURE_SIZE",()=>_S(Pe.getNumber("WEBGL_VERSION")));Pe.registerFlag("WEBGL_MAX_TEXTURES_IN_SHADER",()=>DS(Pe.getNumber("WEBGL_VERSION")));Pe.registerFlag("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION",()=>{let e=Pe.getNumber("WEBGL_VERSION");return e===0?0:$S(e)});Pe.registerFlag("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE",()=>Pe.getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION")>0&&!rh.isMobile());Pe.registerFlag("WEBGL_RENDER_FLOAT32_CAPABLE",()=>PS(Pe.getNumber("WEBGL_VERSION")));Pe.registerFlag("WEBGL_RENDER_FLOAT32_ENABLED",()=>Pe.getBool("WEBGL_FORCE_F16_TEXTURES")?!1:Pe.getBool("WEBGL_RENDER_FLOAT32_CAPABLE"));Pe.registerFlag("WEBGL_DOWNLOAD_FLOAT_ENABLED",()=>FS(Pe.getNumber("WEBGL_VERSION")));Pe.registerFlag("WEBGL_FENCE_API_ENABLED",()=>OS(Pe.getNumber("WEBGL_VERSION")));Pe.registerFlag("WEBGL_SIZE_UPLOAD_UNIFORM",()=>Pe.getBool("WEBGL_RENDER_FLOAT32_ENABLED")?4:0);Pe.registerFlag("WEBGL_DELETE_TEXTURE_THRESHOLD",()=>-1,e=>{if(e<0&&e!==-1)throw new Error(`WEBGL_DELETE_TEXTURE_THRESHOLD must be -1 (indicating never delete) or at least 0, but got ${e}.`)});Pe.registerFlag("WEBGL_FLUSH_THRESHOLD",()=>rh.isMobile()?1:-1,e=>{if(e<0&&e!==-1)throw new Error(`WEBGL_FLUSH_THRESHOLD must be -1 (indicating never manual flush) or at least 0, but got ${e}.`)});Pe.registerFlag("CPU_HANDOFF_SIZE_THRESHOLD",()=>128);Pe.registerFlag("WEBGL_USE_SHAPES_UNIFORMS",()=>!1);Pe.registerFlag("TOPK_LAST_DIM_CPU_HANDOFF_SIZE_THRESHOLD",()=>1e5);Pe.registerFlag("TOPK_K_CPU_HANDOFF_THRESHOLD",()=>128);function as(){let e,t,n,s,r,a,o,i,l,u;return q().getNumber("WEBGL_VERSION")===2?(e="#version 300 es",t="in",n="out",s="in",r="texture",a="outputColor",o="out vec4 outputColor;",i=`
|
|
bool isnan_custom(float val) {
|
|
uint floatToUint = floatBitsToUint(val);
|
|
return (floatToUint & 0x7fffffffu) > 0x7f800000u;
|
|
}
|
|
|
|
bvec4 isnan_custom(vec4 val) {
|
|
return bvec4(isnan_custom(val.x),
|
|
isnan_custom(val.y), isnan_custom(val.z), isnan_custom(val.w));
|
|
}
|
|
|
|
#define isnan(value) isnan_custom(value)
|
|
`,l="",u=`
|
|
#define round(value) newRound(value)
|
|
int newRound(float value) {
|
|
return int(floor(value + 0.5));
|
|
}
|
|
|
|
ivec4 newRound(vec4 value) {
|
|
return ivec4(floor(value + vec4(0.5)));
|
|
}
|
|
`):(e="",t="attribute",n="varying",s="varying",r="texture2D",a="gl_FragColor",o="",i=`
|
|
#define isnan(value) isnan_custom(value)
|
|
bool isnan_custom(float val) {
|
|
return (val > 0. || val < 1. || val == 0.) ? false : true;
|
|
}
|
|
bvec4 isnan_custom(vec4 val) {
|
|
return bvec4(isnan(val.x), isnan(val.y), isnan(val.z), isnan(val.w));
|
|
}
|
|
`,l=`
|
|
uniform float INFINITY;
|
|
|
|
bool isinf(float val) {
|
|
return abs(val) == INFINITY;
|
|
}
|
|
bvec4 isinf(vec4 val) {
|
|
return equal(abs(val), vec4(INFINITY));
|
|
}
|
|
`,u=`
|
|
int round(float value) {
|
|
return int(floor(value + 0.5));
|
|
}
|
|
|
|
ivec4 round(vec4 value) {
|
|
return ivec4(floor(value + vec4(0.5)));
|
|
}
|
|
`),{version:e,attribute:t,varyingVs:n,varyingFs:s,texture2D:r,output:a,defineOutput:o,defineSpecialNaN:i,defineSpecialInf:l,defineRound:u}}function su(e,t,n="index"){let s=v.computeStrides(t);return s.map((r,a)=>{let o=`int ${e[a]} = ${n} / ${r}`,i=a===s.length-1?`int ${e[a+1]} = ${n} - ${e[a]} * ${r}`:`index -= ${e[a]} * ${r}`;return`${o}; ${i};`}).join("")}function b2(e,t,n="index"){let s=v.computeStrides(t);return s.map((r,a)=>{let o=`int ${e[a]} = ${n} / outShapeStrides[${a}]`,i=a===s.length-1?`int ${e[a+1]} = ${n} - ${e[a]} * outShapeStrides[${a}]`:`index -= ${e[a]} * outShapeStrides[${a}]`;return`${o}; ${i};`}).join("")}function xee(e,t){let n=e.length,s=e.map(a=>`${t}[${a}]`),r=new Array(n-1);r[n-2]=s[n-1];for(let a=n-3;a>=0;--a)r[a]=`(${r[a+1]} * ${s[a+1]})`;return r}function bee(e,t,n="index"){let s=e.map((a,o)=>o),r=xee(s,t);return r.map((a,o)=>{let i=`int ${e[o]} = ${n} / ${r[o]}`,l=o===r.length-1?`int ${e[o+1]} = ${n} - ${e[o]} * ${r[o]}`:`index -= ${e[o]} * ${r[o]}`;return`${i}; ${l};`}).join("")}function Ux(e){let t=v.computeStrides(e).map(n=>n.toString());return`
|
|
int getFlatIndex(ivec3 coords) {
|
|
return coords.x * ${t[0]} + coords.y * ${t[1]} + coords.z;
|
|
}
|
|
`}function Gx(){return`
|
|
int getFlatIndex(ivec3 coords) {
|
|
return coords.x * outShapeStrides[0] + coords.y * outShapeStrides[1] + coords.z;
|
|
}
|
|
`}var MS=`
|
|
const float FLOAT_MAX = 1.70141184e38;
|
|
const float FLOAT_MIN = 1.17549435e-38;
|
|
|
|
lowp vec4 encode_float(highp float v) {
|
|
if (isnan(v)) {
|
|
return vec4(255, 255, 255, 255);
|
|
}
|
|
|
|
highp float av = abs(v);
|
|
|
|
if(av < FLOAT_MIN) {
|
|
return vec4(0.0, 0.0, 0.0, 0.0);
|
|
} else if(v > FLOAT_MAX) {
|
|
return vec4(0.0, 0.0, 128.0, 127.0) / 255.0;
|
|
} else if(v < -FLOAT_MAX) {
|
|
return vec4(0.0, 0.0, 128.0, 255.0) / 255.0;
|
|
}
|
|
|
|
highp vec4 c = vec4(0,0,0,0);
|
|
|
|
highp float e = floor(log2(av));
|
|
highp float m = exp2(fract(log2(av))) - 1.0;
|
|
|
|
c[2] = floor(128.0 * m);
|
|
m -= c[2] / 128.0;
|
|
c[1] = floor(32768.0 * m);
|
|
m -= c[1] / 32768.0;
|
|
c[0] = floor(8388608.0 * m);
|
|
|
|
highp float ebias = e + 127.0;
|
|
c[3] = floor(ebias / 2.0);
|
|
ebias -= c[3] * 2.0;
|
|
c[2] += floor(ebias) * 128.0;
|
|
|
|
c[3] += 128.0 * step(0.0, -v);
|
|
|
|
return c / 255.0;
|
|
}
|
|
`,{getBroadcastDims:zS}=T;function vee(e,t,n){let s=[];if(e.forEach(h=>{let f=v.sizeFromShape(h.shapeInfo.logicalShape);if(h.shapeInfo.isUniform?s.push(`uniform float ${h.name}${f>1?`[${f}]`:""};`):(s.push(`uniform sampler2D ${h.name};`),s.push(`uniform int offset${h.name};`)),n.enableShapeUniforms){let{uniformShape:m}=Hx(n.packedInputs,h.shapeInfo.logicalShape,h.shapeInfo.texShape);switch(m.length){case 1:s.push(`uniform int ${h.name}Shape;`);break;case 2:s.push(`uniform ivec2 ${h.name}Shape;`);break;case 3:s.push(`uniform ivec3 ${h.name}Shape;`);break;case 4:s.push(`uniform ivec4 ${h.name}Shape;`);break;default:break}s.push(`uniform ivec2 ${h.name}TexShape;`)}}),n.enableShapeUniforms){switch(t.logicalShape.length){case 1:s.push("uniform int outShape;");break;case 2:s.push("uniform ivec2 outShape;"),s.push("uniform int outShapeStrides;");break;case 3:s.push("uniform ivec3 outShape;"),s.push("uniform ivec2 outShapeStrides;");break;case 4:s.push("uniform ivec4 outShape;"),s.push("uniform ivec3 outShapeStrides;");break;default:break}s.push("uniform ivec2 outTexShape;")}n.customUniforms&&n.customUniforms.forEach(h=>{s.push(`uniform ${h.type} ${h.name}${h.arrayIndex?`[${h.arrayIndex}]`:""};`)});let r=s.join(`
|
|
`),a=e.map(h=>wee(h,t,n.packedInputs,n.enableShapeUniforms)).join(`
|
|
`),o=t.texShape,i=as(),l=See(i),u,c,p=Nee(i);return t.isPacked?(u=kee(t.logicalShape,o,n.enableShapeUniforms),c=Tee(i)):(u=Iee(t.logicalShape,o,n.enableShapeUniforms),c=Cee(i)),n.packedInputs&&(p+=Dee),[p,l,c,r,u,a,n.userCode].join(`
|
|
`)}function sd(e,t=!1){let n=e.shapeInfo.logicalShape;switch(n.length){case 0:return Gee(e,t);case 1:return jee(e,t);case 2:return Xee(e,t);case 3:return Zee(e,t);case 4:return Jee(e,t);case 5:return Qee(e);case 6:return ete(e);default:throw new Error(`${n.length}-D input sampling is not yet supported`)}}function LS(e,t){switch(e.shapeInfo.logicalShape.length){case 0:return Uee(e);case 1:return Hee(e,t);case 2:return qee(e,t);case 3:return Kee(e,t);default:return Yee(e,t)}}function wee(e,t,n=!1,s){let r="";n?r+=LS(e,s):r+=sd(e,s);let a=e.shapeInfo.logicalShape,o=t.logicalShape;return a.length<=o.length&&(n?r+=tte(e,t):r+=nte(e,t)),r}function kee(e,t,n){switch(e.length){case 0:return BS();case 1:return $ee(e,t,n);case 2:return Wee(e,t,n);case 3:return Fee(e,t,n);default:return Mee(e,t,n)}}function Iee(e,t,n){switch(e.length){case 0:return BS();case 1:return Pee(e,t,n);case 2:return Vee(e,t,n);case 3:return Oee(e,t,n);case 4:return zee(e,t,n);case 5:return Lee(e,t);case 6:return Bee(e,t);default:throw new Error(`${e.length}-D output sampling is not yet supported`)}}function See(e){return`
|
|
float sampleTexture(sampler2D textureSampler, vec2 uv) {
|
|
return ${e.texture2D}(textureSampler, uv).r;
|
|
}
|
|
`}function Cee(e){return`
|
|
void setOutput(float val) {
|
|
${e.output} = vec4(val, 0, 0, 0);
|
|
}
|
|
`}function Tee(e){return`
|
|
void setOutput(vec4 val) {
|
|
${e.output} = val;
|
|
}
|
|
`}function Nee(e){return`${e.version}
|
|
precision highp float;
|
|
precision highp int;
|
|
precision highp sampler2D;
|
|
${e.varyingFs} vec2 resultUV;
|
|
${e.defineOutput}
|
|
const vec2 halfCR = vec2(0.5, 0.5);
|
|
|
|
struct ivec5
|
|
{
|
|
int x;
|
|
int y;
|
|
int z;
|
|
int w;
|
|
int u;
|
|
};
|
|
|
|
struct ivec6
|
|
{
|
|
int x;
|
|
int y;
|
|
int z;
|
|
int w;
|
|
int u;
|
|
int v;
|
|
};
|
|
|
|
uniform float NAN;
|
|
${e.defineSpecialNaN}
|
|
${e.defineSpecialInf}
|
|
${e.defineRound}
|
|
|
|
int imod(int x, int y) {
|
|
return x - y * (x / y);
|
|
}
|
|
|
|
int idiv(int a, int b, float sign) {
|
|
int res = a / b;
|
|
int mod = imod(a, b);
|
|
if (sign < 0. && mod != 0) {
|
|
res -= 1;
|
|
}
|
|
return res;
|
|
}
|
|
|
|
//Based on the work of Dave Hoskins
|
|
//https://www.shadertoy.com/view/4djSRW
|
|
#define HASHSCALE1 443.8975
|
|
float random(float seed){
|
|
vec2 p = resultUV * seed;
|
|
vec3 p3 = fract(vec3(p.xyx) * HASHSCALE1);
|
|
p3 += dot(p3, p3.yzx + 19.19);
|
|
return fract((p3.x + p3.y) * p3.z);
|
|
}
|
|
|
|
${Eee}
|
|
${Ree}
|
|
${_ee}
|
|
`}var Eee=`
|
|
vec2 uvFromFlat(int texNumR, int texNumC, int index) {
|
|
int texR = index / texNumC;
|
|
int texC = index - texR * texNumC;
|
|
return (vec2(texC, texR) + halfCR) / vec2(texNumC, texNumR);
|
|
}
|
|
vec2 packedUVfrom1D(int texNumR, int texNumC, int index) {
|
|
int texelIndex = index / 2;
|
|
int texR = texelIndex / texNumC;
|
|
int texC = texelIndex - texR * texNumC;
|
|
return (vec2(texC, texR) + halfCR) / vec2(texNumC, texNumR);
|
|
}
|
|
`,Ree=`
|
|
vec2 packedUVfrom2D(int texelsInLogicalRow, int texNumR,
|
|
int texNumC, int row, int col) {
|
|
int texelIndex = (row / 2) * texelsInLogicalRow + (col / 2);
|
|
int texR = texelIndex / texNumC;
|
|
int texC = texelIndex - texR * texNumC;
|
|
return (vec2(texC, texR) + halfCR) / vec2(texNumC, texNumR);
|
|
}
|
|
`,_ee=`
|
|
vec2 packedUVfrom3D(int texNumR, int texNumC,
|
|
int texelsInBatch, int texelsInLogicalRow, int b,
|
|
int row, int col) {
|
|
int index = b * texelsInBatch + (row / 2) * texelsInLogicalRow + (col / 2);
|
|
int texR = index / texNumC;
|
|
int texC = index - texR * texNumC;
|
|
return (vec2(texC, texR) + halfCR) / vec2(texNumC, texNumR);
|
|
}
|
|
`,Dee=`
|
|
float getChannel(vec4 frag, vec2 innerDims) {
|
|
vec2 modCoord = mod(innerDims, 2.);
|
|
return modCoord.x == 0. ?
|
|
(modCoord.y == 0. ? frag.r : frag.g) :
|
|
(modCoord.y == 0. ? frag.b : frag.a);
|
|
}
|
|
float getChannel(vec4 frag, int dim) {
|
|
float modCoord = mod(float(dim), 2.);
|
|
return modCoord == 0. ? frag.r : frag.g;
|
|
}
|
|
`;function BS(){return`
|
|
int getOutputCoords() {
|
|
return 0;
|
|
}
|
|
`}function $ee(e,t,n){let s=[Math.ceil(t[0]/2),Math.ceil(t[1]/2)];return s[0]===1?n?`
|
|
int getOutputCoords() {
|
|
return 2 * int(resultUV.x * ceil(float(outTexShape[1]) / 2.0));
|
|
}
|
|
`:`
|
|
int getOutputCoords() {
|
|
return 2 * int(resultUV.x * ${s[1]}.0);
|
|
}
|
|
`:s[1]===1?n?`
|
|
int getOutputCoords() {
|
|
return 2 * int(resultUV.y * ceil(float(outTexShape[0]) / 2.0));
|
|
}
|
|
`:`
|
|
int getOutputCoords() {
|
|
return 2 * int(resultUV.y * ${s[0]}.0);
|
|
}
|
|
`:n?`
|
|
int getOutputCoords() {
|
|
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(packedTexShape[0], packedTexShape[1]));
|
|
return 2 * (resTexRC.x * packedTexShape[1] + resTexRC.y);
|
|
}
|
|
`:`
|
|
int getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${s[0]}, ${s[1]}));
|
|
return 2 * (resTexRC.x * ${s[1]} + resTexRC.y);
|
|
}
|
|
`}function Pee(e,t,n){return t[0]===1?n?`
|
|
int getOutputCoords() {
|
|
return int(resultUV.x * float(outTexShape[1]));
|
|
}
|
|
`:`
|
|
int getOutputCoords() {
|
|
return int(resultUV.x * ${t[1]}.0);
|
|
}
|
|
`:t[1]===1?n?`
|
|
int getOutputCoords() {
|
|
return int(resultUV.y * float(outTexShape[0]));
|
|
}
|
|
`:`
|
|
int getOutputCoords() {
|
|
return int(resultUV.y * ${t[0]}.0);
|
|
}
|
|
`:n?`
|
|
int getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(outTexShape[0], outTexShape[1]));
|
|
return resTexRC.x * outTexShape[1] + resTexRC.y;
|
|
}
|
|
`:`
|
|
int getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${t[0]}, ${t[1]}));
|
|
return resTexRC.x * ${t[1]} + resTexRC.y;
|
|
}
|
|
`}function Fee(e,t,n){if(n)return`
|
|
ivec3 getOutputCoords() {
|
|
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
|
|
int texelsInLogicalRow = int(ceil(float(outShape[2]) / 2.0));
|
|
int texelsInBatch = texelsInLogicalRow * int(ceil(float(outShape[1]) / 2.0));
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(packedTexShape[0], packedTexShape[1]));
|
|
int index = resTexRC.x * packedTexShape[1] + resTexRC.y;
|
|
|
|
int b = index / texelsInBatch;
|
|
index -= b * texelsInBatch;
|
|
|
|
int r = 2 * (index / texelsInLogicalRow);
|
|
int c = imod(index, texelsInLogicalRow) * 2;
|
|
|
|
return ivec3(b, r, c);
|
|
}
|
|
`;let s=[Math.ceil(t[0]/2),Math.ceil(t[1]/2)],r=Math.ceil(e[2]/2),a=r*Math.ceil(e[1]/2);return`
|
|
ivec3 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${s[0]}, ${s[1]}));
|
|
int index = resTexRC.x * ${s[1]} + resTexRC.y;
|
|
|
|
int b = index / ${a};
|
|
index -= b * ${a};
|
|
|
|
int r = 2 * (index / ${r});
|
|
int c = imod(index, ${r}) * 2;
|
|
|
|
return ivec3(b, r, c);
|
|
}
|
|
`}function Oee(e,t,n){if(n)return`
|
|
ivec3 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(outTexShape[0], outTexShape[1]));
|
|
int index = resTexRC.x * outTexShape[1] + resTexRC.y;
|
|
${b2(["r","c","d"],e)}
|
|
return ivec3(r, c, d);
|
|
}
|
|
`;let s=su(["r","c","d"],e);return`
|
|
ivec3 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${t[0]}, ${t[1]}));
|
|
int index = resTexRC.x * ${t[1]} + resTexRC.y;
|
|
${s}
|
|
return ivec3(r, c, d);
|
|
}
|
|
`}function Mee(e,t,n){if(n)return`
|
|
ivec4 getOutputCoords() {
|
|
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(packedTexShape[0], packedTexShape[1]));
|
|
int index = resTexRC.x * packedTexShape[1] + resTexRC.y;
|
|
|
|
int texelsInLogicalRow = int(ceil(float(outShape[3]) / 2.0));
|
|
int texelsInBatch = texelsInLogicalRow * int(ceil(float(outShape[2]) / 2.0));
|
|
int texelsInBatchN = texelsInBatch * outShape[1];
|
|
|
|
int b2 = index / texelsInBatchN;
|
|
index -= b2 * texelsInBatchN;
|
|
|
|
int b = index / texelsInBatch;
|
|
index -= b * texelsInBatch;
|
|
|
|
int r = 2 * (index / texelsInLogicalRow);
|
|
int c = imod(index, texelsInLogicalRow) * 2;
|
|
|
|
return ivec4(b2, b, r, c);
|
|
}
|
|
`;let s=[Math.ceil(t[0]/2),Math.ceil(t[1]/2)],r=Math.ceil(e[e.length-1]/2),a=r*Math.ceil(e[e.length-2]/2),o=a,i="",l="b, r, c";for(let u=2;u<e.length-1;u++)o*=e[e.length-u-1],i=`
|
|
int b${u} = index / ${o};
|
|
index -= b${u} * ${o};
|
|
`+i,l=`b${u}, `+l;return`
|
|
ivec${e.length} getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${s[0]}, ${s[1]}));
|
|
int index = resTexRC.x * ${s[1]} + resTexRC.y;
|
|
|
|
${i}
|
|
|
|
int b = index / ${a};
|
|
index -= b * ${a};
|
|
|
|
int r = 2 * (index / ${r});
|
|
int c = imod(index, ${r}) * 2;
|
|
|
|
return ivec${e.length}(${l});
|
|
}
|
|
`}function zee(e,t,n){if(n)return`
|
|
ivec4 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(outTexShape[0], outTexShape[1]));
|
|
int index = resTexRC.x * outTexShape[1] + resTexRC.y;
|
|
${b2(["r","c","d","d2"],e)}
|
|
return ivec4(r, c, d, d2);
|
|
}
|
|
`;let s=su(["r","c","d","d2"],e);return`
|
|
ivec4 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${t[0]}, ${t[1]}));
|
|
int index = resTexRC.x * ${t[1]} + resTexRC.y;
|
|
${s}
|
|
return ivec4(r, c, d, d2);
|
|
}
|
|
`}function Lee(e,t){let n=su(["r","c","d","d2","d3"],e);return`
|
|
ivec5 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx * vec2(${t[0]},
|
|
${t[1]}));
|
|
|
|
int index = resTexRC.x * ${t[1]} + resTexRC.y;
|
|
|
|
${n}
|
|
|
|
ivec5 outShape = ivec5(r, c, d, d2, d3);
|
|
return outShape;
|
|
}
|
|
`}function Bee(e,t){let n=su(["r","c","d","d2","d3","d4"],e);return`
|
|
ivec6 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${t[0]}, ${t[1]}));
|
|
int index = resTexRC.x * ${t[1]} + resTexRC.y;
|
|
|
|
${n}
|
|
|
|
ivec6 result = ivec6(r, c, d, d2, d3, d4);
|
|
return result;
|
|
}
|
|
`}function Wee(e,t,n){let s=[Math.ceil(t[0]/2),Math.ceil(t[1]/2)];if(v.arraysEqual(e,t))return n?`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
|
|
return 2 * ivec2(resultUV.yx * vec2(packedTexShape[0], packedTexShape[1]));
|
|
}
|
|
`:`
|
|
ivec2 getOutputCoords() {
|
|
return 2 * ivec2(resultUV.yx * vec2(${s[0]}, ${s[1]}));
|
|
}
|
|
`;let r=Math.ceil(e[1]/2);return n?`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
|
|
int texelsInLogicalRow = int(ceil(float(outShape[1]) / 2.0));
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(packedTexShape[0], packedTexShape[1]));
|
|
|
|
int index = resTexRC.x * packedTexShape[1] + resTexRC.y;
|
|
int r = 2 * (index / texelsInLogicalRow);
|
|
int c = imod(index, texelsInLogicalRow) * 2;
|
|
|
|
return ivec2(r, c);
|
|
}
|
|
`:`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${s[0]}, ${s[1]}));
|
|
|
|
int index = resTexRC.x * ${s[1]} + resTexRC.y;
|
|
int r = 2 * (index / ${r});
|
|
int c = imod(index, ${r}) * 2;
|
|
|
|
return ivec2(r, c);
|
|
}
|
|
`}function Vee(e,t,n){return v.arraysEqual(e,t)?n?`
|
|
ivec2 getOutputCoords() {
|
|
return ivec2(resultUV.yx * vec2(outTexShape[0], outTexShape[1]));
|
|
}
|
|
`:`
|
|
ivec2 getOutputCoords() {
|
|
return ivec2(resultUV.yx * vec2(${t[0]}, ${t[1]}));
|
|
}
|
|
`:e[1]===1?n?`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(outTexShape[0], outTexShape[1]));
|
|
int index = resTexRC.x * outTexShape[1] + resTexRC.y;
|
|
return ivec2(index, 0);
|
|
}
|
|
`:`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${t[0]}, ${t[1]}));
|
|
int index = resTexRC.x * ${t[1]} + resTexRC.y;
|
|
return ivec2(index, 0);
|
|
}
|
|
`:e[0]===1?n?`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(outTexShape[0], outTexShape[1]));
|
|
int index = resTexRC.x * outTexShape[1] + resTexRC.y;
|
|
return ivec2(0, index);
|
|
}
|
|
`:`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${t[0]}, ${t[1]}));
|
|
int index = resTexRC.x * ${t[1]} + resTexRC.y;
|
|
return ivec2(0, index);
|
|
}
|
|
`:n?`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(outTexShape[0], outTexShape[1]));
|
|
int index = resTexRC.x * outTexShape[1] + resTexRC.y;
|
|
int r = index / outShape[1];
|
|
int c = index - r * outShape[1];
|
|
return ivec2(r, c);
|
|
}
|
|
`:`
|
|
ivec2 getOutputCoords() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx *
|
|
vec2(${t[0]}, ${t[1]}));
|
|
int index = resTexRC.x * ${t[1]} + resTexRC.y;
|
|
int r = index / ${e[1]};
|
|
int c = index - r * ${e[1]};
|
|
return ivec2(r, c);
|
|
}
|
|
`}function ru(e){return`offset${e}`}function Uee(e){let t=e.name,n="get"+t.charAt(0).toUpperCase()+t.slice(1),s=as();return`
|
|
vec4 ${n}() {
|
|
return ${s.texture2D}(${t}, halfCR);
|
|
}
|
|
`}function Gee(e,t){let n=e.name,s="get"+n.charAt(0).toUpperCase()+n.slice(1);if(e.shapeInfo.isUniform)return`float ${s}() {return ${n};}`;let[r,a]=e.shapeInfo.texShape;if(r===1&&a===1)return`
|
|
float ${s}() {
|
|
return sampleTexture(${n}, halfCR);
|
|
}
|
|
`;let o=ru(n);if(t)return`
|
|
float ${s}() {
|
|
vec2 uv = uvFromFlat(${n}TexShape[0], ${n}TexShape[1], ${o});
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`;let[i,l]=e.shapeInfo.texShape;return`
|
|
float ${s}() {
|
|
vec2 uv = uvFromFlat(${i}, ${l}, ${o});
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`}function Hee(e,t){let n=e.name,s="get"+n.charAt(0).toUpperCase()+n.slice(1),r=e.shapeInfo.texShape,a=as();if(t)return`
|
|
vec4 ${s}(int index) {
|
|
ivec2 packedTexShape = ivec2(ceil(float(${n}TexShape[0]) / 2.0), ceil(float(${n}TexShape[1]) / 2.0));
|
|
vec2 uv = packedUVfrom1D(
|
|
packedTexShape[0], packedTexShape[1], index);
|
|
return ${a.texture2D}(${n}, uv);
|
|
}
|
|
`;let o=[Math.ceil(r[0]/2),Math.ceil(r[1]/2)];return`
|
|
vec4 ${s}(int index) {
|
|
vec2 uv = packedUVfrom1D(
|
|
${o[0]}, ${o[1]}, index);
|
|
return ${a.texture2D}(${n}, uv);
|
|
}
|
|
`}function jee(e,t){let n=e.name,s="get"+n.charAt(0).toUpperCase()+n.slice(1);if(e.shapeInfo.isUniform)return`
|
|
float ${s}(int index) {
|
|
${rd(e)}
|
|
}
|
|
`;let r=e.shapeInfo.texShape,a=r[0],o=r[1];if(o===1&&a===1)return`
|
|
float ${s}(int index) {
|
|
return sampleTexture(${n}, halfCR);
|
|
}
|
|
`;let i=ru(n);return o===1?t?`
|
|
float ${s}(int index) {
|
|
vec2 uv = vec2(0.5, (float(index + ${i}) + 0.5) / float(${n}TexShape[0]));
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`:`
|
|
float ${s}(int index) {
|
|
vec2 uv = vec2(0.5, (float(index + ${i}) + 0.5) / ${a}.0);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`:a===1?t?`
|
|
float ${s}(int index) {
|
|
vec2 uv = vec2((float(index + ${i}) + 0.5) / float(${n}TexShape[1]), 0.5);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`:`
|
|
float ${s}(int index) {
|
|
vec2 uv = vec2((float(index + ${i}) + 0.5) / ${o}.0, 0.5);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`:t?`
|
|
float ${s}(int index) {
|
|
vec2 uv = uvFromFlat(${n}TexShape[0], ${n}TexShape[1], index + ${i});
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`:`
|
|
float ${s}(int index) {
|
|
vec2 uv = uvFromFlat(${a}, ${o}, index + ${i});
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`}function qee(e,t){let n=e.shapeInfo.logicalShape,s=e.name,r="get"+s.charAt(0).toUpperCase()+s.slice(1),a=e.shapeInfo.texShape,o=a[0],i=a[1],l=as();if(a!=null&&v.arraysEqual(n,a))return t?`
|
|
vec4 ${r}(int row, int col) {
|
|
vec2 uv = (vec2(col, row) + halfCR) / vec2(${s}TexShape[1], ${s}TexShape[0]);
|
|
|
|
return ${l.texture2D}(${s}, uv);
|
|
}
|
|
`:`
|
|
vec4 ${r}(int row, int col) {
|
|
vec2 uv = (vec2(col, row) + halfCR) / vec2(${i}.0, ${o}.0);
|
|
|
|
return ${l.texture2D}(${s}, uv);
|
|
}
|
|
`;if(t)return`
|
|
vec4 ${r}(int row, int col) {
|
|
ivec2 packedTexShape = ivec2(ceil(float(${s}TexShape[0]) / 2.0), ceil(float(${s}TexShape[1]) / 2.0));
|
|
int valuesPerRow = int(ceil(float(${s}Shape[1]) / 2.0));
|
|
vec2 uv = packedUVfrom2D(valuesPerRow, packedTexShape[0], packedTexShape[1], row, col);
|
|
return ${l.texture2D}(${s}, uv);
|
|
}
|
|
`;let u=[Math.ceil(a[0]/2),Math.ceil(a[1]/2)],c=Math.ceil(n[1]/2);return`
|
|
vec4 ${r}(int row, int col) {
|
|
vec2 uv = packedUVfrom2D(${c}, ${u[0]}, ${u[1]}, row, col);
|
|
return ${l.texture2D}(${s}, uv);
|
|
}
|
|
`}function Xee(e,t){let n=e.shapeInfo.logicalShape,s=e.name,r="get"+s.charAt(0).toUpperCase()+s.slice(1),a=e.shapeInfo.texShape;if(a!=null&&v.arraysEqual(n,a)){if(t)return`
|
|
float ${r}(int row, int col) {
|
|
vec2 uv = (vec2(col, row) + halfCR) / vec2(${s}TexShape[1], ${s}TexShape[0]);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`;let d=a[0],h=a[1];return`
|
|
float ${r}(int row, int col) {
|
|
vec2 uv = (vec2(col, row) + halfCR) / vec2(${h}.0, ${d}.0);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`}let{newShape:o,keptDims:i}=v.squeezeShape(n),l=o;if(l.length<n.length){let d=ad(e,l),h=["row","col"];return`
|
|
${sd(d,t)}
|
|
float ${r}(int row, int col) {
|
|
return ${r}(${od(h,i)});
|
|
}
|
|
`}if(e.shapeInfo.isUniform)return`
|
|
float ${r}(int row, int col) {
|
|
int index = round(dot(vec2(row, col), vec2(${n[1]}, 1)));
|
|
${rd(e)}
|
|
}
|
|
`;let u=a[0],c=a[1],p=ru(s);return c===1?t?`
|
|
float ${r}(int row, int col) {
|
|
float index = dot(vec3(row, col, ${p}), vec3(${s}Shape[1], 1, 1));
|
|
vec2 uv = vec2(0.5, (index + 0.5) / float(${s}TexShape[0]));
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col) {
|
|
float index = dot(vec3(row, col, ${p}), vec3(${n[1]}, 1, 1));
|
|
vec2 uv = vec2(0.5, (index + 0.5) / ${u}.0);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:u===1?t?`
|
|
float ${r}(int row, int col) {
|
|
float index = dot(vec3(row, col, ${p}), vec3(${s}Shape[1], 1, 1));
|
|
vec2 uv = vec2((index + 0.5) / float(${s}TexShape[1]), 0.5);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col) {
|
|
float index = dot(vec3(row, col, ${p}), vec3(${n[1]}, 1, 1));
|
|
vec2 uv = vec2((index + 0.5) / ${c}.0, 0.5);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:t?`
|
|
float ${r}(int row, int col) {
|
|
// Explicitly use integer operations as dot() only works on floats.
|
|
int index = row * ${s}Shape[1] + col + ${p};
|
|
vec2 uv = uvFromFlat(${s}TexShape[0], ${s}TexShape[1], index);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col) {
|
|
// Explicitly use integer operations as dot() only works on floats.
|
|
int index = row * ${n[1]} + col + ${p};
|
|
vec2 uv = uvFromFlat(${u}, ${c}, index);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`}function Kee(e,t){let n=e.shapeInfo.logicalShape,s=e.name,r="get"+s.charAt(0).toUpperCase()+s.slice(1),a=e.shapeInfo.texShape,o=[Math.ceil(a[0]/2),Math.ceil(a[1]/2)];if(n[0]===1){let d=n.slice(1),h=[1,2],f=ad(e,d),m=["b","row","col"];return`
|
|
${LS(f,t)}
|
|
vec4 ${r}(int b, int row, int col) {
|
|
return ${r}(${od(m,h)});
|
|
}
|
|
`}let i=as();if(t)return`
|
|
vec4 ${r}(int b, int row, int col) {
|
|
ivec2 packedTexShape = ivec2(ceil(float(${s}TexShape[0]) / 2.0), ceil(float(${s}TexShape[1]) / 2.0));
|
|
int valuesPerRow = int(ceil(float(${s}Shape[2]) / 2.0));
|
|
int texelsInBatch = valuesPerRow * int(ceil(float(${s}Shape[1]) / 2.0));
|
|
vec2 uv = packedUVfrom3D(
|
|
packedTexShape[0], packedTexShape[1], texelsInBatch, valuesPerRow, b, row, col);
|
|
return ${i.texture2D}(${s}, uv);
|
|
}
|
|
`;let l=o[0],u=o[1],c=Math.ceil(n[2]/2),p=c*Math.ceil(n[1]/2);return`
|
|
vec4 ${r}(int b, int row, int col) {
|
|
vec2 uv = packedUVfrom3D(
|
|
${l}, ${u}, ${p}, ${c}, b, row, col);
|
|
return ${i.texture2D}(${s}, uv);
|
|
}
|
|
`}function Zee(e,t){let n=e.shapeInfo.logicalShape,s=e.name,r="get"+s.charAt(0).toUpperCase()+s.slice(1),a=n[1]*n[2],o=n[2],{newShape:i,keptDims:l}=v.squeezeShape(n),u=i;if(u.length<n.length){let m=ad(e,u),g=["row","col","depth"];return`
|
|
${sd(m,t)}
|
|
float ${r}(int row, int col, int depth) {
|
|
return ${r}(${od(g,l)});
|
|
}
|
|
`}if(e.shapeInfo.isUniform)return`
|
|
float ${r}(int row, int col, int depth) {
|
|
int index = round(dot(vec3(row, col, depth),
|
|
vec3(${a}, ${o}, 1)));
|
|
${rd(e)}
|
|
}
|
|
`;let c=e.shapeInfo.texShape,p=c[0],d=c[1],h=e.shapeInfo.flatOffset;if(d===a&&h==null)return t?`
|
|
float ${r}(int row, int col, int depth) {
|
|
int stride1 = ${s}Shape[2];
|
|
float texR = float(row);
|
|
float texC = dot(vec2(col, depth), vec2(stride1, 1));
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${s}TexShape[1], ${s}TexShape[0]);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col, int depth) {
|
|
float texR = float(row);
|
|
float texC = dot(vec2(col, depth), vec2(${o}, 1));
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${d}.0, ${p}.0);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`;if(d===o&&h==null)return t?`
|
|
float ${r}(int row, int col, int depth) {
|
|
float texR = dot(vec2(row, col), vec2(${s}Shape[1], 1));
|
|
float texC = float(depth);
|
|
vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${s}TexShape[1], ${s}TexShape[0]);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col, int depth) {
|
|
float texR = dot(vec2(row, col), vec2(${n[1]}, 1));
|
|
float texC = float(depth);
|
|
vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${d}.0, ${p}.0);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`;let f=ru(s);return t?`
|
|
float ${r}(int row, int col, int depth) {
|
|
// Explicitly use integer operations as dot() only works on floats.
|
|
int stride0 = ${s}Shape[1] * ${s}Shape[2];
|
|
int stride1 = ${s}Shape[2];
|
|
int index = row * ${a} + col * ${o} + depth + ${f};
|
|
vec2 uv = uvFromFlat(${s}TexShape[0], ${s}TexShape[1], index);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col, int depth) {
|
|
// Explicitly use integer operations as dot() only works on floats.
|
|
int index = row * ${a} + col * ${o} + depth + ${f};
|
|
vec2 uv = uvFromFlat(${p}, ${d}, index);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`}function Yee(e,t){let n=e.name,s="get"+n.charAt(0).toUpperCase()+n.slice(1),r=as();if(t)return`
|
|
vec4 ${s}(int b2, int b, int row, int col) {
|
|
int valuesPerRow = int(ceil(float(${n}Shape[3]) / 2.0));
|
|
int texelsInBatch = valuesPerRow * int(ceil(float(${n}Shape[2]) / 2.0));
|
|
int index = b * texelsInBatch + (row / 2) * valuesPerRow + (col / 2);
|
|
texelsInBatch *= ${n}Shape[1];
|
|
index = b2 * texelsInBatch + index;
|
|
ivec2 packedTexShape = ivec2(ceil(float(${n}TexShape[0]) / 2.0), ceil(float(${n}TexShape[1]) / 2.0));
|
|
int texR = index / packedTexShape[1];
|
|
int texC = index - texR * packedTexShape[1];
|
|
vec2 uv = (vec2(texC, texR) + halfCR) / vec2(packedTexShape[1], packedTexShape[0]); return ${r.texture2D}(${n}, uv);
|
|
}
|
|
`;let a=e.shapeInfo.logicalShape,o=a.length,i=e.shapeInfo.texShape,l=[Math.ceil(i[0]/2),Math.ceil(i[1]/2)],u=l[0],c=l[1],p=Math.ceil(a[o-1]/2),d=p*Math.ceil(a[o-2]/2),h="int b, int row, int col",f=`b * ${d} + (row / 2) * ${p} + (col / 2)`;for(let m=2;m<o-1;m++)h=`int b${m}, `+h,d*=a[o-m-1],f=`b${m} * ${d} + `+f;return`
|
|
vec4 ${s}(${h}) {
|
|
int index = ${f};
|
|
int texR = index / ${c};
|
|
int texC = index - texR * ${c};
|
|
vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${c}, ${u});
|
|
return ${r.texture2D}(${n}, uv);
|
|
}
|
|
`}function Jee(e,t){let n=e.shapeInfo.logicalShape,s=e.name,r="get"+s.charAt(0).toUpperCase()+s.slice(1),a=n[3],o=n[2]*a,i=n[1]*o,{newShape:l,keptDims:u}=v.squeezeShape(n);if(l.length<n.length){let x=ad(e,l),A=["row","col","depth","depth2"];return`
|
|
${sd(x,t)}
|
|
float ${r}(int row, int col, int depth, int depth2) {
|
|
return ${r}(${od(A,u)});
|
|
}
|
|
`}if(e.shapeInfo.isUniform)return`
|
|
float ${r}(int row, int col, int depth, int depth2) {
|
|
int index = round(dot(vec4(row, col, depth, depth2),
|
|
vec4(${i}, ${o}, ${a}, 1)));
|
|
${rd(e)}
|
|
}
|
|
`;let c=e.shapeInfo.flatOffset,p=e.shapeInfo.texShape,d=p[0],h=p[1],f=`int stride2 = ${s}Shape[3];`,m=`int stride1 = ${s}Shape[2] * stride2;`,g=`int stride0 = ${s}Shape[1] * stride1;`;if(h===i&&c==null)return t?`
|
|
float ${r}(int row, int col, int depth, int depth2) {
|
|
${f}
|
|
${m}
|
|
float texR = float(row);
|
|
float texC =
|
|
dot(vec3(col, depth, depth2),
|
|
vec3(stride1, stride2, 1));
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${s}TexShape[1], ${s}TexShape[0]);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col, int depth, int depth2) {
|
|
float texR = float(row);
|
|
float texC =
|
|
dot(vec3(col, depth, depth2),
|
|
vec3(${o}, ${a}, 1));
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${h}.0, ${d}.0);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`;if(h===a&&c==null)return t?`
|
|
float ${r}(int row, int col, int depth, int depth2) {
|
|
float texR = dot(vec3(row, col, depth),
|
|
vec3(${s}Shape[1] * ${s}Shape[2], ${s}Shape[2], 1));
|
|
float texC = float(depth2);
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${s}TexShape[1], ${s}TexShape[0]);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col, int depth, int depth2) {
|
|
float texR = dot(vec3(row, col, depth),
|
|
vec3(${n[1]*n[2]}, ${n[2]}, 1));
|
|
float texC = float(depth2);
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${h}.0, ${d}.0);
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`;let y=ru(s);return t?`
|
|
float ${r}(int row, int col, int depth, int depth2) {
|
|
// Explicitly use integer operations as dot() only works on floats.
|
|
${f}
|
|
${m}
|
|
${g}
|
|
int index = row * stride0 + col * stride1 +
|
|
depth * stride2 + depth2;
|
|
vec2 uv = uvFromFlat(${s}TexShape[0], ${s}TexShape[1], index + ${y});
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`:`
|
|
float ${r}(int row, int col, int depth, int depth2) {
|
|
// Explicitly use integer operations as dot() only works on floats.
|
|
int index = row * ${i} + col * ${o} +
|
|
depth * ${a} + depth2;
|
|
vec2 uv = uvFromFlat(${d}, ${h}, index + ${y});
|
|
return sampleTexture(${s}, uv);
|
|
}
|
|
`}function Qee(e){let t=e.shapeInfo.logicalShape,n=e.name,s="get"+n.charAt(0).toUpperCase()+n.slice(1),r=t[4],a=t[3]*r,o=t[2]*a,i=t[1]*o,{newShape:l,keptDims:u}=v.squeezeShape(t);if(l.length<t.length){let m=ad(e,l),g=["row","col","depth","depth2","depth3"];return`
|
|
${sd(m)}
|
|
float ${s}(int row, int col, int depth, int depth2, int depth3) {
|
|
return ${s}(${od(g,u)});
|
|
}
|
|
`}if(e.shapeInfo.isUniform)return`
|
|
float ${s}(int row, int col, int depth, int depth2, int depth3) {
|
|
float index = dot(
|
|
vec4(row, col, depth, depth2),
|
|
vec4(${i}, ${o}, ${a}, ${r})) +
|
|
depth3;
|
|
${rd(e)}
|
|
}
|
|
`;let c=e.shapeInfo.flatOffset,p=e.shapeInfo.texShape,d=p[0],h=p[1];if(h===i&&c==null)return`
|
|
float ${s}(int row, int col, int depth, int depth2, int depth3) {
|
|
int texR = row;
|
|
float texC = dot(vec4(col, depth, depth2, depth3),
|
|
vec4(${o}, ${a}, ${r}, 1));
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${h}.0, ${d}.0);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`;if(h===r&&c==null)return`
|
|
float ${s}(int row, int col, int depth, int depth2, int depth3) {
|
|
float texR = dot(
|
|
vec4(row, col, depth, depth2),
|
|
vec4(${t[1]*t[2]*t[3]},
|
|
${t[2]*t[3]}, ${t[3]}, 1));
|
|
int texC = depth3;
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${h}.0, ${d}.0);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`;let f=ru(n);return`
|
|
float ${s}(int row, int col, int depth, int depth2, int depth3) {
|
|
// Explicitly use integer operations as dot() only works on floats.
|
|
int index = row * ${i} + col * ${o} + depth * ${a} +
|
|
depth2 * ${r} + depth3 + ${f};
|
|
vec2 uv = uvFromFlat(${d}, ${h}, index);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`}function ete(e){let t=e.shapeInfo.logicalShape,n=e.name,s="get"+n.charAt(0).toUpperCase()+n.slice(1),{newShape:r,keptDims:a}=v.squeezeShape(t);if(r.length<t.length){let g=ad(e,r),y=["row","col","depth","depth2","depth3","depth4"];return`
|
|
${sd(g)}
|
|
float ${s}(int row, int col, int depth,
|
|
int depth2, int depth3, int depth4) {
|
|
return ${s}(${od(y,a)});
|
|
}
|
|
`}let o=t[5],i=t[4]*o,l=t[3]*i,u=t[2]*l,c=t[1]*u;if(e.shapeInfo.isUniform)return`
|
|
float ${s}(int row, int col, int depth,
|
|
int depth2, int depth3, int depth4) {
|
|
int index = round(dot(
|
|
vec4(row, col, depth, depth2),
|
|
vec4(${c}, ${u}, ${l}, ${i})) +
|
|
dot(
|
|
vec2(depth3, depth4),
|
|
vec2(${o}, 1)));
|
|
${rd(e)}
|
|
}
|
|
`;let p=e.shapeInfo.flatOffset,d=e.shapeInfo.texShape,h=d[0],f=d[1];if(f===c&&p==null)return`
|
|
float ${s}(int row, int col, int depth,
|
|
int depth2, int depth3, int depth4) {
|
|
int texR = row;
|
|
float texC = dot(vec4(col, depth, depth2, depth3),
|
|
vec4(${u}, ${l}, ${i}, ${o})) +
|
|
float(depth4);
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${f}.0, ${h}.0);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`;if(f===o&&p==null)return`
|
|
float ${s}(int row, int col, int depth,
|
|
int depth2, int depth3, int depth4) {
|
|
float texR = dot(vec4(row, col, depth, depth2),
|
|
vec4(${t[1]*t[2]*t[3]*t[4]},
|
|
${t[2]*t[3]*t[4]},
|
|
${t[3]*t[4]},
|
|
${t[4]})) + float(depth3);
|
|
int texC = depth4;
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${f}.0, ${h}.0);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`;let m=ru(n);return`
|
|
float ${s}(int row, int col, int depth,
|
|
int depth2, int depth3, int depth4) {
|
|
// Explicitly use integer operations as dot() only works on floats.
|
|
int index = row * ${c} + col * ${u} + depth * ${l} +
|
|
depth2 * ${i} + depth3 * ${o} + depth4 + ${m};
|
|
vec2 uv = uvFromFlat(${h}, ${f}, index);
|
|
return sampleTexture(${n}, uv);
|
|
}
|
|
`}function rd(e){let t=e.name,n=v.sizeFromShape(e.shapeInfo.logicalShape);return n<2?`return ${t};`:`
|
|
for (int i = 0; i < ${n}; i++) {
|
|
if (i == index) {
|
|
return ${t}[i];
|
|
}
|
|
}
|
|
`}function tte(e,t){let n=e.name,s=n.charAt(0).toUpperCase()+n.slice(1),r="get"+s+"AtOutCoords",a=e.shapeInfo.logicalShape.length,o=t.logicalShape.length,i=zS(e.shapeInfo.logicalShape,t.logicalShape),l=wt(o),u=o-a,c,p=["x","y","z","w","u","v"];a===0?c="":o<2&&i.length>=1?c="coords = 0;":c=i.map(x=>`coords.${p[x+u]} = 0;`).join(`
|
|
`);let d="";o<2&&a>0?d="coords":d=e.shapeInfo.logicalShape.map((x,A)=>`coords.${p[A+u]}`).join(", ");let h="return outputValue;",m=v.sizeFromShape(e.shapeInfo.logicalShape)===1,y=v.sizeFromShape(t.logicalShape)===1;if(a===1&&!m&&!y)h=`
|
|
return vec4(outputValue.xy, outputValue.xy);
|
|
`;else if(m&&!y)o===1?h=`
|
|
return vec4(outputValue.x, outputValue.x, 0., 0.);
|
|
`:h=`
|
|
return vec4(outputValue.x);
|
|
`;else if(i.length){let x=a-2,A=a-1;i.indexOf(x)>-1&&i.indexOf(A)>-1?h="return vec4(outputValue.x);":i.indexOf(x)>-1?h="return vec4(outputValue.x, outputValue.y, outputValue.x, outputValue.y);":i.indexOf(A)>-1&&(h="return vec4(outputValue.xx, outputValue.zz);")}return`
|
|
vec4 ${r}() {
|
|
${l} coords = getOutputCoords();
|
|
${c}
|
|
vec4 outputValue = get${s}(${d});
|
|
${h}
|
|
}
|
|
`}function nte(e,t){let n=e.name,s=n.charAt(0).toUpperCase()+n.slice(1),r="get"+s+"AtOutCoords",a=t.texShape,o=e.shapeInfo.texShape,i=e.shapeInfo.logicalShape.length,l=t.logicalShape.length;if(!e.shapeInfo.isUniform&&i===l&&e.shapeInfo.flatOffset==null&&v.arraysEqual(o,a))return`
|
|
float ${r}() {
|
|
return sampleTexture(${n}, resultUV);
|
|
}
|
|
`;let u=wt(l),c=zS(e.shapeInfo.logicalShape,t.logicalShape),p=l-i,d,h=["x","y","z","w","u","v"];i===0?d="":l<2&&c.length>=1?d="coords = 0;":d=c.map(m=>`coords.${h[m+p]} = 0;`).join(`
|
|
`);let f="";return l<2&&i>0?f="coords":f=e.shapeInfo.logicalShape.map((m,g)=>`coords.${h[g+p]}`).join(", "),`
|
|
float ${r}() {
|
|
${u} coords = getOutputCoords();
|
|
${d}
|
|
return get${s}(${f});
|
|
}
|
|
`}function wt(e){if(e<=1)return"int";if(e===2)return"ivec2";if(e===3)return"ivec3";if(e===4)return"ivec4";if(e===5)return"ivec5";if(e===6)return"ivec6";throw Error(`GPU for rank ${e} is not yet supported`)}function Hx(e,t,n){let{newShape:s,keptDims:r}=v.squeezeShape(t),a=t.length,o=e&&a===3&&t[0]===1,i=o?t.slice(1):s,l=!e&&a>1&&!v.arraysEqual(t,n)&&s.length<a||o;return{useSqueezeShape:l,uniformShape:l?i:t,keptDims:r}}function ad(e,t){let n=JSON.parse(JSON.stringify(e));return n.shapeInfo.logicalShape=t,n}function od(e,t){return t.map(n=>e[n]).join(", ")}function ste(e,t,n,s){let r=n.map((c,p)=>{let d={logicalShape:c.shape,texShape:c.isUniform?null:c.texData.texShape,isUniform:c.isUniform,isPacked:c.isUniform?!1:c.texData.isPacked,flatOffset:null};return c.texData!=null&&c.texData.slice!=null&&c.texData.slice.flatOffset>0&&(d.flatOffset=c.texData.slice.flatOffset),{name:t.variableNames[p],shapeInfo:d}}),a=r.map(c=>c.shapeInfo),o={logicalShape:s.shape,texShape:s.texData.texShape,isUniform:!1,isPacked:s.texData.isPacked,flatOffset:null},i=vee(r,o,t),l=gS(e.gl,i),u=e.createProgram(l);return q().get("ENGINE_COMPILE_ONLY")?{program:t,fragmentShader:l,source:i,webGLProgram:u,inShapeInfos:a,outShapeInfo:o,uniformLocations:null,customUniformLocations:null,infLoc:null,nanLoc:null,inShapesLocations:null,inTexShapesLocations:null,outShapeLocation:null,outShapeStridesLocation:null,outTexShapeLocation:null}:Object.assign({program:t,fragmentShader:l,source:i,webGLProgram:u,inShapeInfos:a,outShapeInfo:o},WS(e,t,u))}function WS(e,t,n){let s={},r={},a={},o=[],i,l,u,c=null,p=null;p=e.getUniformLocation(n,"NAN",!1),q().getNumber("WEBGL_VERSION")===1&&(c=e.getUniformLocation(n,"INFINITY",!1));let d=!1;for(let h=0;h<t.variableNames.length;h++){let f=t.variableNames[h];s[f]=e.getUniformLocation(n,f,d),s[`offset${f}`]=e.getUniformLocation(n,`offset${f}`,d),t.enableShapeUniforms&&(r[`${f}Shape`]=e.getUniformLocation(n,`${f}Shape`,d),a[`${f}TexShape`]=e.getUniformLocation(n,`${f}TexShape`,d))}return t.enableShapeUniforms&&(i=e.getUniformLocation(n,"outShape",d),u=e.getUniformLocation(n,"outShapeStrides",d),l=e.getUniformLocation(n,"outTexShape",d)),t.customUniforms&&t.customUniforms.forEach((h,f)=>{o[f]=e.getUniformLocation(n,h.name,d)}),{uniformLocations:s,customUniformLocations:o,infLoc:c,nanLoc:p,inShapesLocations:r,inTexShapesLocations:a,outShapeLocation:i,outShapeStridesLocation:u,outTexShapeLocation:l}}function f7(e,t){if(e.length!==t.length)throw Error(`Binary was compiled with ${e.length} inputs, but was executed with ${t.length} inputs`);e.forEach((n,s)=>{let r=n.logicalShape,a=t[s],o=a.shape;if(!v.arraysEqual(r,o))throw Error(`Binary was compiled with different shapes than the current args. Shapes ${r} and ${o} must match`);if(n.isUniform&&a.isUniform)return;let i=n.texShape,l=a.isUniform?null:a.texData.texShape;if(!v.arraysEqual(i,l))throw Error(`Binary was compiled with different texture shapes than the current args. Shape ${i} and ${l} must match`)})}function rte(e,t,n,s,r){t.program.enableShapeUniforms||(f7(t.inShapeInfos,n),f7([t.outShapeInfo],[s]));let a=s.texData.texture,o=s.texData.texShape;s.texData.isPacked?e.setOutputPackedMatrixTexture(a.texture,o[0],o[1]):e.setOutputMatrixTexture(a.texture,o[0],o[1]),e.setProgram(t.webGLProgram),q().getNumber("WEBGL_VERSION")===1&&t.infLoc!==null&&e.gl.uniform1f(t.infLoc,1/0),t.nanLoc!==null&&e.gl.uniform1f(t.nanLoc,NaN),n.forEach((l,u)=>{let c=t.program.variableNames[u],p=t.uniformLocations[c],d=t.uniformLocations[`offset${c}`],h=t.inShapesLocations[`${c}Shape`],f=t.inTexShapesLocations[`${c}TexShape`];if(h){let{uniformShape:m}=Hx(t.program.packedInputs,l.shape,l.texData.texShape);switch(m.length){case 1:e.gl.uniform1iv(h,new Int32Array(m));break;case 2:e.gl.uniform2iv(h,new Int32Array(m));break;case 3:e.gl.uniform3iv(h,new Int32Array(m));break;case 4:e.gl.uniform4iv(h,new Int32Array(m));break;default:break}}if(f&&e.gl.uniform2i(f,l.texData.texShape[0],l.texData.texShape[1]),p!=null){if(l.isUniform){if(v.sizeFromShape(l.shape)<2)e.gl.uniform1f(p,l.uniformValues[0]);else{let m=l.uniformValues;m instanceof Float32Array||(m=new Float32Array(m)),e.gl.uniform1fv(p,m)}return}l.texData.slice!=null&&d!=null&&e.gl.uniform1i(d,l.texData.slice.flatOffset),e.setInputMatrixTexture(l.texData.texture.texture,p,u)}});let i=t.outShapeLocation;if(i)switch(s.shape.length){case 1:e.gl.uniform1iv(i,new Int32Array(s.shape));break;case 2:e.gl.uniform2iv(i,new Int32Array(s.shape));break;case 3:e.gl.uniform3iv(i,new Int32Array(s.shape));break;case 4:e.gl.uniform4iv(i,new Int32Array(s.shape));break;default:break}if(t.outShapeStridesLocation){let l=v.computeStrides(s.shape);switch(s.shape.length){case 2:e.gl.uniform1iv(t.outShapeStridesLocation,new Int32Array(l));break;case 3:e.gl.uniform2iv(t.outShapeStridesLocation,new Int32Array(l));break;case 4:e.gl.uniform3iv(t.outShapeStridesLocation,new Int32Array(l));break;default:break}}t.outTexShapeLocation&&e.gl.uniform2i(t.outTexShapeLocation,s.texData.texShape[0],s.texData.texShape[1]),t.program.customUniforms&&r&&t.program.customUniforms.forEach((l,u)=>{let c=t.customUniformLocations[u],p=r[u];if(l.type==="float")e.gl.uniform1fv(c,p);else if(l.type==="vec2")e.gl.uniform2fv(c,p);else if(l.type==="vec3")e.gl.uniform3fv(c,p);else if(l.type==="vec4")e.gl.uniform4fv(c,p);else if(l.type==="int")e.gl.uniform1iv(c,p);else if(l.type==="ivec2")e.gl.uniform2iv(c,p);else if(l.type==="ivec3")e.gl.uniform3iv(c,p);else if(l.type==="ivec4")e.gl.uniform4iv(c,p);else throw Error(`uniform type ${l.type} is not supported yet.`)}),e.executeProgram()}function ate(e,t,n){let s="";t.concat(n).forEach(o=>{let i=o.texData!=null&&o.texData.slice!=null&&o.texData.slice.flatOffset>0;if(e.enableShapeUniforms&&!o.isUniform){let l=o.texData.texShape,{useSqueezeShape:u,uniformShape:c,keptDims:p}=Hx(e.packedInputs,o.shape,l),d="",h="",f="";if(c.length===1&&e.packedInputs){let w=[Math.ceil(l[0]/2),Math.ceil(l[1]/2)];d=`${w[0]>1}_${w[1]>1}`}else if(c.length===2&&!e.packedInputs)h=`${c[0]>1}_${c[1]>1}`;else if(c.length>2&&!e.packedInputs){let w=v.computeStrides(c);f=`${w[0]===l[1]}_${w[w.length-1]===l[1]}`}let m=o.shape.length,g=c.length===2&&v.arraysEqual(o.shape,l),y=v.sizeFromShape(o.shape)===1,x=T.getBroadcastDims(o.shape,n.shape),A=!e.packedInputs&&m===n.shape.length&&v.arraysEqual(l,n.texData.texShape),b=e.packedInputs||c.length>2?"":`${l[0]>1}_${l[1]>1}`;s+=`${m}_${A}_${u?p:""}_${c.length}_${y}_${x}_${g}_${d}_${h}_${f}_${b}_${i}`}else{let l=o.isUniform?"uniform":o.texData.texShape;s+=`${o.shape}_${l}_${i}`}});let r=e.userCode,a=e.constructor.name;return a+="_"+s+"_"+r+`${q().getNumber("WEBGL_VERSION")}`,a}function xs(e){return q().getBool("WEBGL_USE_SHAPES_UNIFORMS")&&e<=4}var ote=class{constructor(e){this.variableNames=["A"],this.packedInputs=!1,this.packedOutput=!0,this.outPackingScheme=_p.DENSE,this.customUniforms=[{name:"texShape",type:"ivec2"}];let t=as();this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length),this.userCode=`
|
|
ivec3 outCoordsFromFlatIndex(int index) {
|
|
${this.enableShapeUniforms?b2(["r","c","d"],e):su(["r","c","d"],e)}
|
|
return ivec3(r, c, d);
|
|
}
|
|
|
|
void main() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx * vec2(texShape[0], texShape[1]));
|
|
int index = 4 * (resTexRC.x * texShape[1] + resTexRC.y);
|
|
|
|
vec4 result = vec4(0.);
|
|
|
|
for (int i=0; i<4; i++) {
|
|
int flatIndex = index + i;
|
|
ivec3 rc = outCoordsFromFlatIndex(flatIndex);
|
|
result[i] = getA(rc.x, rc.y, rc.z);
|
|
}
|
|
|
|
${t.output} = result;
|
|
}
|
|
`}},ite=class{constructor(e){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.outPackingScheme=_p.DENSE,this.customUniforms=[{name:"texShape",type:"ivec2"}];let t=as();this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length),this.userCode=`
|
|
ivec3 outCoordsFromFlatIndex(int index) {
|
|
${this.enableShapeUniforms?b2(["r","c","d"],e):su(["r","c","d"],e)}
|
|
return ivec3(r, c, d);
|
|
}
|
|
|
|
void main() {
|
|
ivec2 resTexRC = ivec2(resultUV.yx * vec2(texShape[0], texShape[1]));
|
|
int index = 4 * (resTexRC.x * texShape[1] + resTexRC.y);
|
|
|
|
vec4 result = vec4(0.);
|
|
|
|
for (int i=0; i<4; i++) {
|
|
int flatIndex = index + i;
|
|
ivec3 rc = outCoordsFromFlatIndex(flatIndex);
|
|
result[i] = getChannel(getA(rc.x, rc.y, rc.z), vec2(rc.y, rc.z));
|
|
}
|
|
|
|
${t.output} = result;
|
|
}
|
|
`}},lte=class{constructor(e){this.variableNames=["A"],this.outTexUsage=Xs.DOWNLOAD;let t=as();this.outputShape=e,this.userCode=`
|
|
${MS}
|
|
|
|
void main() {
|
|
float x = getAAtOutCoords();
|
|
${t.output} = encode_float(x);
|
|
}
|
|
`}},ute=class{constructor(e){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!1,this.outTexUsage=Xs.DOWNLOAD;let t=as();this.outputShape=e,this.userCode=`
|
|
${MS}
|
|
|
|
void main() {
|
|
ivec3 coords = getOutputCoords();
|
|
float x = getChannel(getAAtOutCoords(), vec2(coords.y, coords.z));
|
|
${t.output} = encode_float(x);
|
|
}
|
|
`}},cte=class{constructor(e,t=!1){this.variableNames=["A"],this.customUniforms=[{name:"texShape",type:"ivec2"}];let n=as();this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length);let s="result";t&&(s="floor(result * 255. + 0.5)"),this.userCode=`
|
|
${this.enableShapeUniforms?Gx():Ux(e)}
|
|
|
|
void main() {
|
|
ivec3 coords = getOutputCoords();
|
|
|
|
int flatIndex = getFlatIndex(coords);
|
|
int offset = imod(flatIndex, 4);
|
|
|
|
flatIndex = idiv(flatIndex, 4, 1.);
|
|
|
|
int r = flatIndex / texShape[1];
|
|
int c = imod(flatIndex, texShape[1]);
|
|
vec2 uv = (vec2(c, r) + halfCR) / vec2(texShape[1], texShape[0]);
|
|
vec4 values = ${n.texture2D}(A, uv);
|
|
|
|
float result;
|
|
|
|
if(offset == 0) {
|
|
result = values[0];
|
|
} else if(offset == 1) {
|
|
result = values[1];
|
|
} else if(offset == 2) {
|
|
result = values[2];
|
|
} else {
|
|
result = values[3];
|
|
}
|
|
|
|
${n.output} = vec4(${s}, 0., 0., 0.);
|
|
}
|
|
`}},dte=class{constructor(e,t=!1){this.variableNames=["A"],this.packedInputs=!1,this.packedOutput=!0,this.customUniforms=[{name:"texShape",type:"ivec2"}];let n=as();this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length);let s="",r="result";t&&(r="floor(result * 255. + 0.5)");for(let a=0;a<=1;a++)for(let o=0;o<=1;o++){let i=a*2+o;s+=`
|
|
localCoords = coords;
|
|
if(localCoords[2] + ${o} < ${this.enableShapeUniforms?"outShape[2]":`${e[2]}`}) {
|
|
localCoords[2] += ${o};
|
|
if (localCoords[1] + ${a} < ${this.enableShapeUniforms?"outShape[1]":`${e[1]}`}) {
|
|
localCoords[1] += ${a};
|
|
|
|
flatIndex = getFlatIndex(localCoords);
|
|
offset = imod(flatIndex, 4);
|
|
|
|
flatIndex = idiv(flatIndex, 4, 1.);
|
|
|
|
int r = flatIndex / texShape[1];
|
|
int c = imod(flatIndex, texShape[1]);
|
|
vec2 uv = (vec2(c, r) + halfCR) / vec2(texShape[1], texShape[0]);
|
|
values = ${n.texture2D}(A, uv);
|
|
|
|
if (offset == 0) {
|
|
result[${i}] = values[0];
|
|
} else if (offset == 1) {
|
|
result[${i}] = values[1];
|
|
} else if (offset == 2) {
|
|
result[${i}] = values[2];
|
|
} else {
|
|
result[${i}] = values[3];
|
|
}
|
|
}
|
|
}
|
|
`}this.userCode=`
|
|
${this.enableShapeUniforms?Gx():Ux(e)}
|
|
|
|
void main() {
|
|
ivec3 coords = getOutputCoords();
|
|
|
|
vec4 result = vec4(0.);
|
|
int flatIndex, r, c, offset;
|
|
ivec3 localCoords;
|
|
vec2 uv;
|
|
vec4 values;
|
|
|
|
${s}
|
|
|
|
${n.output} = ${r};
|
|
}
|
|
`}},VS={};Ve(VS,{bindVertexProgramAttributeStreams:()=>YS,createBufferFromOutputTexture:()=>e9,createFloat16MatrixTexture:()=>qS,createFloat16PackedMatrixTexture:()=>ZS,createFloat32MatrixTexture:()=>jS,createIndexBuffer:()=>HS,createPackedMatrixTexture:()=>KS,createUnsignedBytesMatrixTexture:()=>XS,createVertexBuffer:()=>GS,createVertexShader:()=>US,downloadByteEncodedFloatMatrixFromOutputTexture:()=>n9,downloadFloat32MatrixFromBuffer:()=>t9,downloadMatrixFromPackedOutputTexture:()=>r9,downloadPackedMatrixFromBuffer:()=>s9,getInternalFormatForFloat16MatrixTexture:()=>qx,getInternalFormatForFloat16PackedMatrixTexture:()=>Zx,getInternalFormatForFloat32MatrixTexture:()=>jx,getInternalFormatForPackedMatrixTexture:()=>Kx,getInternalFormatForUnsignedBytesMatrixTexture:()=>Xx,uploadDenseMatrixToTexture:()=>JS,uploadPixelDataToTexture:()=>QS});function US(e){let t=as(),n=`${t.version}
|
|
precision highp float;
|
|
${t.attribute} vec3 clipSpacePos;
|
|
${t.attribute} vec2 uv;
|
|
${t.varyingVs} vec2 resultUV;
|
|
|
|
void main() {
|
|
gl_Position = vec4(clipSpacePos, 1);
|
|
resultUV = uv;
|
|
}`;return mS(e,n)}function GS(e){let t=new Float32Array([-1,1,0,0,1,-1,-1,0,0,0,1,1,0,1,1,1,-1,0,1,0]);return xS(e,t)}function HS(e){let t=new Uint16Array([0,1,2,2,1,3]);return bS(e,t)}function Oh(e,t,n,s,r,a){wS(t,n);let o=vS(e),i=e.TEXTURE_2D;return Ie(e,()=>e.bindTexture(i,o)),Ie(e,()=>e.texParameteri(i,e.TEXTURE_WRAP_S,e.CLAMP_TO_EDGE)),Ie(e,()=>e.texParameteri(i,e.TEXTURE_WRAP_T,e.CLAMP_TO_EDGE)),Ie(e,()=>e.texParameteri(i,e.TEXTURE_MIN_FILTER,e.NEAREST)),Ie(e,()=>e.texParameteri(i,e.TEXTURE_MAG_FILTER,e.NEAREST)),q().getNumber("WEBGL_VERSION")===1?Ie(e,()=>e.texImage2D(i,0,s,t,n,0,r,a,null)):Ie(e,()=>e.texStorage2D(i,1,s,t,n)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,null)),{texture:o,texShape:[n,t]}}function jx(e){return e.internalFormatFloat}function jS(e,t,n,s){let[r,a]=Fh(t,n);return Oh(e,r,a,jx(s),s.textureFormatFloat,e.FLOAT)}function qx(e){return e.internalFormatHalfFloat}function qS(e,t,n,s){let[r,a]=Fh(t,n);return Oh(e,r,a,qx(s),s.textureFormatFloat,s.textureTypeHalfFloat)}function Xx(e){return e.downloadTextureFormat}function XS(e,t,n,s){let[r,a]=Fh(t,n);return Oh(e,r,a,Xx(s),e.RGBA,e.UNSIGNED_BYTE)}function Kx(e){return e.internalFormatPackedFloat}function KS(e,t,n,s){let[r,a]=td(t,n);return Oh(e,r,a,Kx(s),e.RGBA,e.FLOAT)}function Zx(e){return e.internalFormatPackedHalfFloat}function ZS(e,t,n,s){let[r,a]=td(t,n);return Oh(e,r,a,Zx(s),e.RGBA,s.textureTypeHalfFloat)}function YS(e,t,n){return Ie(e,()=>e.bindBuffer(e.ARRAY_BUFFER,n)),ny(e,t,"clipSpacePos",n,3,20,0)&&ny(e,t,"uv",n,2,20,12)}function JS(e,t,n,s,r,a){Ie(e,()=>e.bindTexture(e.TEXTURE_2D,t));let o,i,l;r instanceof Uint8Array?(o=new Uint8Array(n*s*4),i=e.UNSIGNED_BYTE,l=e.RGBA):(o=new Float32Array(n*s*4),i=e.FLOAT,l=a.internalFormatPackedFloat),o.set(r),q().getNumber("WEBGL_VERSION")===2?Ie(e,()=>e.texSubImage2D(e.TEXTURE_2D,0,0,0,n,s,e.RGBA,i,o)):Ie(e,()=>e.texImage2D(e.TEXTURE_2D,0,l,n,s,0,e.RGBA,i,o)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,null))}function QS(e,t,n){Ie(e,()=>e.bindTexture(e.TEXTURE_2D,t)),n.data instanceof Uint8Array?q().getNumber("WEBGL_VERSION")===2?Ie(e,()=>e.texSubImage2D(e.TEXTURE_2D,0,0,0,n.width,n.height,e.RGBA,e.UNSIGNED_BYTE,n.data)):Ie(e,()=>e.texImage2D(e.TEXTURE_2D,0,e.RGBA,n.width,n.height,0,e.RGBA,e.UNSIGNED_BYTE,n.data)):q().getNumber("WEBGL_VERSION")===2?Ie(e,()=>e.texSubImage2D(e.TEXTURE_2D,0,0,0,e.RGBA,e.UNSIGNED_BYTE,n)):Ie(e,()=>e.texImage2D(e.TEXTURE_2D,0,e.RGBA,e.RGBA,e.UNSIGNED_BYTE,n)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,null))}function e9(e,t,n,s){let r=e.createBuffer();Ie(e,()=>e.bindBuffer(e.PIXEL_PACK_BUFFER,r));let i=4*4*t*n;return Ie(e,()=>e.bufferData(e.PIXEL_PACK_BUFFER,i,e.STREAM_READ)),Ie(e,()=>e.readPixels(0,0,n,t,e.RGBA,e.FLOAT,0)),Ie(e,()=>e.bindBuffer(e.PIXEL_PACK_BUFFER,null)),r}function t9(e,t,n){let s=e,r=new Float32Array(n);return s.bindBuffer(s.PIXEL_PACK_BUFFER,t),s.getBufferSubData(s.PIXEL_PACK_BUFFER,0,r),s.bindBuffer(s.PIXEL_PACK_BUFFER,null),r}function n9(e,t,n,s){let[r,a]=Fh(t,n),o=4,i=new Uint8Array(iee(t*n,o));return Ie(e,()=>e.readPixels(0,0,r,a,s.downloadTextureFormat,e.UNSIGNED_BYTE,i)),new Float32Array(i.buffer)}function s9(e,t,n,s,r,a,o,i){let l=e,u=new Float32Array(lee(a,o));return l.bindBuffer(l.PIXEL_PACK_BUFFER,t),l.getBufferSubData(l.PIXEL_PACK_BUFFER,0,u),l.bindBuffer(l.PIXEL_PACK_BUFFER,null),u}function r9(e,t,n){let s=new Float32Array(t*n*4);return Ie(e,()=>e.readPixels(0,0,n,t,e.RGBA,e.FLOAT,s)),s}var ju=class{constructor(e){this.outputTexture=null,this.program=null,this.disposed=!1,this.vertexAttrsAreBound=!1,this.itemsToPoll=[];let t=q().getNumber("WEBGL_VERSION");e!=null?(this.gl=e,x2(t,e)):this.gl=$r(t);let n="WEBGL_color_buffer_float",s="EXT_color_buffer_half_float";if(this.parallelCompilationExtension=this.gl.getExtension("KHR_parallel_shader_compile"),q().getNumber("WEBGL_VERSION")===1){let r="OES_texture_float",a="OES_texture_half_float";if(this.textureFloatExtension=up(this.gl,r),Ks(this.gl,a))this.textureHalfFloatExtension=up(this.gl,a);else if(q().get("WEBGL_FORCE_F16_TEXTURES"))throw new Error("GL context does not support half float textures, yet the environment flag WEBGL_FORCE_F16_TEXTURES is set to true.");if(this.colorBufferFloatExtension=this.gl.getExtension(n),Ks(this.gl,s))this.colorBufferHalfFloatExtension=up(this.gl,s);else if(q().get("WEBGL_FORCE_F16_TEXTURES"))throw new Error("GL context does not support color renderable half floats, yet the environment flag WEBGL_FORCE_F16_TEXTURES is set to true.")}else if(n="EXT_color_buffer_float",Ks(this.gl,n))this.colorBufferFloatExtension=this.gl.getExtension(n);else if(Ks(this.gl,s))this.colorBufferHalfFloatExtension=this.gl.getExtension(s);else throw new Error("GL context does not support color renderable floats");this.vertexBuffer=GS(this.gl),this.indexBuffer=HS(this.gl),this.framebuffer=kS(this.gl),this.textureConfig=Wx(this.gl,this.textureHalfFloatExtension)}get debug(){return q().getBool("DEBUG")}dispose(){if(this.disposed)return;this.program!=null&&console.warn("Disposing a GPGPUContext that still has a bound WebGLProgram. This is probably a resource leak, delete the program with GPGPUContext.deleteProgram before disposing."),this.outputTexture!=null&&console.warn("Disposing a GPGPUContext that still has a bound output matrix texture. This is probably a resource leak, delete the output matrix texture with GPGPUContext.deleteMatrixTexture before disposing.");let e=this.gl;Ie(e,()=>e.finish()),Ie(e,()=>e.bindFramebuffer(e.FRAMEBUFFER,null)),Ie(e,()=>e.deleteFramebuffer(this.framebuffer)),Ie(e,()=>e.bindBuffer(e.ARRAY_BUFFER,null)),Ie(e,()=>e.bindBuffer(e.ELEMENT_ARRAY_BUFFER,null)),Ie(e,()=>e.deleteBuffer(this.indexBuffer)),this.disposed=!0}createFloat32MatrixTexture(e,t){return this.throwIfDisposed(),jS(this.gl,e,t,this.textureConfig)}createFloat16MatrixTexture(e,t){return this.throwIfDisposed(),qS(this.gl,e,t,this.textureConfig)}createUnsignedBytesMatrixTexture(e,t){return this.throwIfDisposed(),XS(this.gl,e,t,this.textureConfig)}uploadPixelDataToTexture(e,t){this.throwIfDisposed(),QS(this.gl,e,t)}uploadDenseMatrixToTexture(e,t,n,s){this.throwIfDisposed(),JS(this.gl,e,t,n,s,this.textureConfig)}createFloat16PackedMatrixTexture(e,t){return this.throwIfDisposed(),ZS(this.gl,e,t,this.textureConfig)}createPackedMatrixTexture(e,t){return this.throwIfDisposed(),KS(this.gl,e,t,this.textureConfig)}deleteMatrixTexture(e){this.throwIfDisposed(),this.outputTexture===e&&(sy(this.gl,this.framebuffer),this.outputTexture=null),Ie(this.gl,()=>this.gl.deleteTexture(e))}downloadByteEncodedFloatMatrixFromOutputTexture(e,t,n){return this.downloadMatrixDriver(e,()=>n9(this.gl,t,n,this.textureConfig))}downloadPackedMatrixFromBuffer(e,t,n,s,r,a){return s9(this.gl,e,t,n,s,r,a,this.textureConfig)}downloadFloat32MatrixFromBuffer(e,t){return t9(this.gl,e,t)}createBufferFromTexture(e,t,n){this.bindTextureToFrameBuffer(e);let s=e9(this.gl,t,n,this.textureConfig);return this.unbindTextureToFrameBuffer(),s}createAndWaitForFence(){let e=this.createFence(this.gl);return this.pollFence(e)}createFence(e){let t,n;if(q().getBool("WEBGL_FENCE_API_ENABLED")){let s=e,r=s.fenceSync(s.SYNC_GPU_COMMANDS_COMPLETE,0);e.flush(),n=()=>{let a=s.clientWaitSync(r,0,0);return a===s.ALREADY_SIGNALED||a===s.CONDITION_SATISFIED},t=r}else q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION")>0?(t=this.beginQuery(),this.endQuery(),n=()=>this.isQueryAvailable(t,q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION"))):n=()=>!0;return{query:t,isFencePassed:n}}downloadMatrixFromPackedTexture(e,t,n){return this.downloadMatrixDriver(e,()=>r9(this.gl,t,n))}createProgram(e){this.throwIfDisposed();let t=this.gl;this.vertexShader==null&&(this.vertexShader=US(t));let n=yS(t);return Ie(t,()=>t.attachShader(n,this.vertexShader)),Ie(t,()=>t.attachShader(n,e)),AS(t,n),this.debug&&nm(t,n),this.vertexAttrsAreBound||(this.setProgram(n),this.vertexAttrsAreBound=YS(t,this.program,this.vertexBuffer)),n}deleteProgram(e){this.throwIfDisposed(),e===this.program&&(this.program=null),e!=null&&Ie(this.gl,()=>this.gl.deleteProgram(e))}setProgram(e){this.throwIfDisposed(),this.program=e,this.program!=null&&this.debug&&nm(this.gl,this.program),Ie(this.gl,()=>this.gl.useProgram(e))}getUniformLocation(e,t,n=!0){return this.throwIfDisposed(),n?SS(this.gl,e,t):CS(this.gl,e,t)}getAttributeLocation(e,t){return this.throwIfDisposed(),Ie(this.gl,()=>this.gl.getAttribLocation(e,t))}getUniformLocationNoThrow(e,t){return this.throwIfDisposed(),this.gl.getUniformLocation(e,t)}setInputMatrixTexture(e,t,n){this.throwIfDisposed(),this.throwIfNoProgram(),TS(this.gl,e,t,n)}setOutputMatrixTexture(e,t,n){this.setOutputMatrixTextureDriver(e,n,t)}setOutputPackedMatrixTexture(e,t,n){this.throwIfDisposed();let[s,r]=td(t,n);this.setOutputMatrixTextureDriver(e,s,r)}setOutputMatrixWriteRegion(e,t,n,s){this.setOutputMatrixWriteRegionDriver(n,e,s,t)}setOutputPackedMatrixWriteRegion(e,t,n,s){throw new Error("setOutputPackedMatrixWriteRegion not implemented.")}debugValidate(){this.program!=null&&nm(this.gl,this.program),cp(this.gl)}executeProgram(){this.throwIfDisposed(),this.throwIfNoProgram();let e=this.gl;this.debug&&this.debugValidate(),Ie(e,()=>e.drawElements(e.TRIANGLES,6,e.UNSIGNED_SHORT,0))}blockUntilAllProgramsCompleted(){this.throwIfDisposed(),Ie(this.gl,()=>this.gl.finish())}getQueryTimerExtension(){return this.disjointQueryTimerExtension==null&&(this.disjointQueryTimerExtension=up(this.gl,q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION")===2?"EXT_disjoint_timer_query_webgl2":"EXT_disjoint_timer_query")),this.disjointQueryTimerExtension}getQueryTimerExtensionWebGL2(){return this.getQueryTimerExtension()}getQueryTimerExtensionWebGL1(){return this.getQueryTimerExtension()}beginQuery(){if(q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION")===2){let n=this.gl,s=this.getQueryTimerExtensionWebGL2(),r=n.createQuery();return n.beginQuery(s.TIME_ELAPSED_EXT,r),r}let e=this.getQueryTimerExtensionWebGL1(),t=e.createQueryEXT();return e.beginQueryEXT(e.TIME_ELAPSED_EXT,t),t}endQuery(){if(q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION")===2){let t=this.gl,n=this.getQueryTimerExtensionWebGL2();t.endQuery(n.TIME_ELAPSED_EXT);return}let e=this.getQueryTimerExtensionWebGL1();e.endQueryEXT(e.TIME_ELAPSED_EXT)}async waitForQueryAndGetTime(e){return await v.repeatedTry(()=>this.disposed||this.isQueryAvailable(e,q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION"))),this.getQueryTime(e,q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION"))}getQueryTime(e,t){if(t===0)return null;if(t===2){let n=this.gl;return n.getQueryParameter(e,n.QUERY_RESULT)/1e6}else{let n=this.getQueryTimerExtensionWebGL1();return n.getQueryObjectEXT(e,n.QUERY_RESULT_EXT)/1e6}}isQueryAvailable(e,t){if(t===0)return!0;if(t===2){let n=this.gl,s=this.getQueryTimerExtensionWebGL2(),r=n.getQueryParameter(e,n.QUERY_RESULT_AVAILABLE);return this.disjoint==null&&(this.disjoint=this.gl.getParameter(s.GPU_DISJOINT_EXT)),r&&!this.disjoint}else{let n=this.getQueryTimerExtensionWebGL1(),s=n.getQueryObjectEXT(e,n.QUERY_RESULT_AVAILABLE_EXT);return this.disjoint==null&&(this.disjoint=this.gl.getParameter(n.GPU_DISJOINT_EXT)),s&&!this.disjoint}}pollFence(e){return new Promise(t=>{this.addItemToPoll(()=>e.isFencePassed(),()=>t())})}pollItems(){let e=pte(this.itemsToPoll.map(t=>t.isDoneFn));for(let t=0;t<=e;++t){let{resolveFn:n}=this.itemsToPoll[t];n()}this.itemsToPoll=this.itemsToPoll.slice(e+1)}addItemToPoll(e,t){this.itemsToPoll.push({isDoneFn:e,resolveFn:t}),!(this.itemsToPoll.length>1)&&v.repeatedTry(()=>(this.pollItems(),this.itemsToPoll.length===0))}bindTextureToFrameBuffer(e){this.throwIfDisposed(),sm(this.gl,e,this.framebuffer),this.debug&&cp(this.gl)}unbindTextureToFrameBuffer(){this.outputTexture!=null?(sm(this.gl,this.outputTexture,this.framebuffer),this.debug&&cp(this.gl)):sy(this.gl,this.framebuffer)}downloadMatrixDriver(e,t){this.bindTextureToFrameBuffer(e);let n=t();return this.unbindTextureToFrameBuffer(),n}setOutputMatrixTextureDriver(e,t,n){this.throwIfDisposed();let s=this.gl;sm(s,e,this.framebuffer),this.debug&&cp(s),this.outputTexture=e,Ie(s,()=>s.viewport(0,0,t,n)),Ie(s,()=>s.scissor(0,0,t,n))}setOutputMatrixWriteRegionDriver(e,t,n,s){this.throwIfDisposed(),Ie(this.gl,()=>this.gl.scissor(e,t,n,s))}throwIfDisposed(){if(this.disposed)throw new Error("Attempted to use disposed GPGPUContext.")}throwIfNoProgram(){if(this.program==null)throw new Error("No GPU program is currently set.")}};function pte(e){let t=0;for(;t<e.length&&e[t]();++t);return t-1}var{addImpl:hte,bincountImpl:a9,bincountReduceImpl:fte,ceilImpl:mte,concatImpl:gte,equalImpl:yte,expImpl:Ate,expm1Impl:xte,floorImpl:bte,gatherNdImpl:vte,gatherV2Impl:wte,greaterImpl:kte,greaterEqualImpl:Ite,lessImpl:Ste,lessEqualImpl:Cte,linSpaceImpl:Tte,logImpl:Nte,maxImpl:Ete,maximumImpl:Rte,minimumImpl:_te,multiplyImpl:Dte,negImpl:$te,notEqualImpl:Pte,prodImpl:Fte,rangeImpl:Ote,rsqrtImpl:Mte,scatterImpl:zte,sigmoidImpl:Lte,simpleAbsImpl:o9,sliceImpl:Bte,sparseFillEmptyRowsImpl:Wte,sparseReshapeImpl:Vte,sparseSegmentReductionImpl:i9,sqrtImpl:Ute,stridedSliceImpl:Gte,stringNGramsImpl:Hte,stringSplitImpl:jte,stringToHashBucketFastImpl:qte,subImpl:Xte,tileImpl:Kte,topKImpl:Zte,transposeImpl:Yx,uniqueImpl:Yte}=Cx;function l9(e,t){return["x","y","z","w","u","v"].slice(0,t).map(n=>`${e}.${n}`)}function es(e,t){return t===1?[e]:l9(e,t)}function Jte(e,t){if(e===1)return"rc";let n="";for(let s=0;s<e;s++)n+=t[s],s<e-1&&(n+=",");return n}var Qte=class{constructor(e){if(this.variableNames=["A"],this.packedInputs=!1,this.packedOutput=!0,this.outputShape=e,this.rank=e.length,this.enableShapeUniforms=xs(this.outputShape.length),this.rank===0)this.userCode=`
|
|
void main() {
|
|
setOutput(vec4(getA(), 0., 0., 0.));
|
|
}
|
|
`;else{let t=es("rc",this.rank),n=wt(this.rank),s=this.getOutOfBoundsCondition(t),r=this.getSetup(t),a=this.getOutput(t);this.userCode=`
|
|
void main() {
|
|
${n} rc = getOutputCoords();
|
|
|
|
if(${s}) {
|
|
setOutput(vec4(0));
|
|
} else {
|
|
${r}
|
|
|
|
setOutput(vec4(${a}));
|
|
}
|
|
}
|
|
`}}getSourceCoordsArr(e){let t=[];for(let n=0;n<=1;n++)for(let s=0;s<=1;s++){let r=`${n===0?"r":"rp1"}, ${s===0?"c":"cp1"}`;for(let a=2;a<this.rank;a++)r=`${e[e.length-1-a]},`+r;t.push(r)}return t}getOutOfBoundsCondition(e){if(this.rank===1)return`rc > ${this.enableShapeUniforms?"outShape":this.outputShape[0]}`;let t="";for(let n=this.rank-2;n<this.rank;n++)t+=`${e[n]} >= ${this.enableShapeUniforms?`outShape[${n}]`:this.outputShape[n]}`,n<this.rank-1&&(t+="||");return t}getSetup(e){if(this.rank===1)return"";let t=e.slice(-2),n=this.enableShapeUniforms?`outShape[${this.rank} - 1]`:this.outputShape[this.rank-1],s=this.enableShapeUniforms?`outShape[${this.rank} - 2]`:this.outputShape[this.rank-2];return`
|
|
int r = ${t[0]};
|
|
int c = ${t[1]};
|
|
int rp1 = r + 1;
|
|
int cp1 = c + 1;
|
|
|
|
bool cEdge = cp1 >= ${n};
|
|
bool rEdge = rp1 >= ${s};
|
|
`}getOutput(e){let t=this.getSourceCoordsArr(e);return this.rank===1?`getA(rc), (rc + 1 >= ${this.enableShapeUniforms?"outShape":this.outputShape[0]} ? 0. : getA(rc + 1)), 0, 0`:`getA(${t[0]}),
|
|
cEdge ? 0. : getA(${t[1]}),
|
|
rEdge ? 0. : getA(${t[2]}),
|
|
rEdge || cEdge ? 0. : getA(${t[3]})`}},u9=class{constructor(e,t){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.customUniforms=[{name:"inputShape",type:"ivec3"}],this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length);let n="";for(let s=0;s<4;s++){let r="thisRC = rc;";s%2===1&&(r+="thisRC.z += 1;"),s>1&&(r+="thisRC.y += 1;"),n+=`
|
|
${r}
|
|
${s>0?"if(thisRC.y < rows && thisRC.z < cols){":""}
|
|
int flatIndex = getFlatIndex(thisRC);
|
|
|
|
ivec3 inputRC = inputCoordsFromReshapedOutCoords(flatIndex);
|
|
vec2 inputRCInnerDims = vec2(float(inputRC.y),float(inputRC.z));
|
|
|
|
result[${s}] =
|
|
getChannel(getA(inputRC.x, inputRC.y, inputRC.z), inputRCInnerDims);
|
|
${s>0?"}":""}
|
|
`}this.userCode=`
|
|
${ene(t,this.enableShapeUniforms)}
|
|
${this.enableShapeUniforms?Gx():Ux(e)}
|
|
|
|
void main() {
|
|
ivec3 rc = getOutputCoords();
|
|
|
|
vec4 result = vec4(0.);
|
|
|
|
ivec3 thisRC;
|
|
int rows = ${this.enableShapeUniforms?"outShape[1]":e[1]};
|
|
int cols = ${this.enableShapeUniforms?"outShape[2]":e[2]};
|
|
|
|
${n}
|
|
|
|
setOutput(result);
|
|
}
|
|
`}};function ene(e,t){return`
|
|
ivec3 inputCoordsFromReshapedOutCoords(int index) {
|
|
${t?bee(["r","c","d"],"inputShape"):su(["r","c","d"],e)}
|
|
return ivec3(r, c, d);
|
|
}
|
|
`}var tne=class{constructor(e){this.gpgpu=e,this.numUsedTextures=0,this.numFreeTextures=0,this._numBytesAllocated=0,this._numBytesFree=0,this.freeTextures={},this.logEnabled=!1,this.usedTextures={}}acquireTexture(e,t,n){let s=g7(t,n),r=y7(e,s,n);r in this.freeTextures||(this.freeTextures[r]=[]),r in this.usedTextures||(this.usedTextures[r]=[]);let a=m7(e,s,this.gpgpu.gl,this.gpgpu.textureConfig,n);if(this.freeTextures[r].length>0){this.numFreeTextures--,this.numUsedTextures++,this._numBytesFree-=a,this.log();let i=this.freeTextures[r].shift();return this.usedTextures[r].push(i),i}let o;return s===Sn.PACKED_2X2_FLOAT32?o=this.gpgpu.createPackedMatrixTexture(e[0],e[1]):s===Sn.PACKED_2X2_FLOAT16?o=this.gpgpu.createFloat16PackedMatrixTexture(e[0],e[1]):s===Sn.UNPACKED_FLOAT32?o=this.gpgpu.createFloat32MatrixTexture(e[0],e[1]):s===Sn.UNPACKED_FLOAT16?o=this.gpgpu.createFloat16MatrixTexture(e[0],e[1]):s===Sn.PACKED_4X1_UNSIGNED_BYTE&&(o=this.gpgpu.createUnsignedBytesMatrixTexture(e[0],e[1])),this.usedTextures[r].push(o),this.numUsedTextures++,this._numBytesAllocated+=a,this.log(),o}releaseTexture(e,t,n,s){if(this.freeTextures==null)return;let r=g7(n,s),a=y7(t,r,s);a in this.freeTextures||(this.freeTextures[a]=[]);let o=m7(t,r,this.gpgpu.gl,this.gpgpu.textureConfig,s),i=q().get("WEBGL_DELETE_TEXTURE_THRESHOLD");i!==-1&&this._numBytesAllocated>i?(this.gpgpu.deleteMatrixTexture(e.texture),this._numBytesAllocated-=o):(this.freeTextures[a].push(e),this.numFreeTextures++,this._numBytesFree+=o),this.numUsedTextures--;let l=this.usedTextures[a],u=l.indexOf(e);if(u<0)throw new Error("Cannot release a texture that was never provided by this texture manager");l.splice(u,1),this.log()}log(){if(!this.logEnabled)return;let e=this.numFreeTextures+this.numUsedTextures;console.log("Free/Used",`${this.numFreeTextures} / ${this.numUsedTextures}`,`(${e})`);let t=this._numBytesFree/this._numBytesAllocated;console.log(`Bytes allocated: ${this._numBytesAllocated}`),console.log(`Bytes unused: ${this._numBytesFree} (${Math.round(100*t)}%)`)}get numBytesAllocated(){return this._numBytesAllocated}get numBytesFree(){return this._numBytesFree}getNumUsedTextures(){return this.numUsedTextures}getNumFreeTextures(){return this.numFreeTextures}dispose(){if(this.freeTextures!=null){for(let e in this.freeTextures)this.freeTextures[e].forEach(t=>{this.gpgpu.deleteMatrixTexture(t.texture)});for(let e in this.usedTextures)this.usedTextures[e].forEach(t=>{this.gpgpu.deleteMatrixTexture(t.texture)});this.freeTextures=null,this.usedTextures=null,this.numUsedTextures=0,this.numFreeTextures=0,this._numBytesAllocated=0,this._numBytesFree=0}}};function nne(e,t){let n=e;if(t===n.R32F)return 4;if(t===n.R16F)return 2;if(t===n.RGBA32F)return 16;if(t===e.RGBA)return 16;if(t===n.RGBA16F)return 8;if(t===n.RGBA8)return 4;throw new Error(`Unknown internal format ${t}`)}function m7(e,t,n,s,r){let a=sne(t,s),o;if(r){let[l,u]=td(e[0],e[1]);o=l*u}else{let[l,u]=Fh(e[0],e[1]);o=l*u}let i=nne(n,a);return o*i}function sne(e,t){switch(e){case Sn.PACKED_2X2_FLOAT32:return Kx(t);case Sn.PACKED_2X2_FLOAT16:return Zx(t);case Sn.UNPACKED_FLOAT32:return jx(t);case Sn.UNPACKED_FLOAT16:return qx(t);case Sn.PACKED_4X1_UNSIGNED_BYTE:return Xx(t);default:throw new Error(`Unknown physical texture type ${e}`)}}function rne(e){return q().getBool("WEBGL_RENDER_FLOAT32_ENABLED")?e?Sn.PACKED_2X2_FLOAT32:Sn.UNPACKED_FLOAT32:e?Sn.PACKED_2X2_FLOAT16:Sn.UNPACKED_FLOAT16}function g7(e,t){if(e===Xs.UPLOAD)return Sn.PACKED_2X2_FLOAT32;if(e===Xs.RENDER||e==null)return rne(t);if(e===Xs.DOWNLOAD||e===Xs.PIXELS)return Sn.PACKED_4X1_UNSIGNED_BYTE;throw new Error(`Unknown logical texture type ${e}`)}function y7(e,t,n){return`${e[0]}_${e[1]}_${t}_${n}`}var pa=class{constructor(e,t){this.variableNames=["A"],this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length),this.userCode=`
|
|
float unaryOperation(float x) {
|
|
${t}
|
|
}
|
|
|
|
void main() {
|
|
float x = getAAtOutCoords();
|
|
float y = unaryOperation(x);
|
|
|
|
setOutput(y);
|
|
}
|
|
`}},fr="if (isnan(x)) return x;",ane="return x;",A7="return abs(x);",one="return (x >= 0.0) ? x : (exp(x) - 1.0);",ine=fr+`
|
|
return (x < 0.0) ? 0.0 : x;
|
|
`,lne=fr+`
|
|
return (x < 0.0) ? 0.0 : min(6.0, x);
|
|
`,Pu="return x;",une="return 1.0 / (1.0 + exp(-1.0 * x));",cne="return x;",dne=`
|
|
vec4 result;
|
|
|
|
result.r = (x.r >= 0.0) ? x.r : (exp(x.r) - 1.0);
|
|
result.g = (x.g >= 0.0) ? x.g : (exp(x.g) - 1.0);
|
|
result.b = (x.b >= 0.0) ? x.b : (exp(x.b) - 1.0);
|
|
result.a = (x.a >= 0.0) ? x.a : (exp(x.a) - 1.0);
|
|
|
|
return result;
|
|
`,pne=`
|
|
vec4 result = x * vec4(greaterThanEqual(x, vec4(0.0)));
|
|
bvec4 isNaN = isnan(x);
|
|
|
|
result.r = isNaN.r ? x.r : result.r;
|
|
result.g = isNaN.g ? x.g : result.g;
|
|
result.b = isNaN.b ? x.b : result.b;
|
|
result.a = isNaN.a ? x.a : result.a;
|
|
|
|
return result;
|
|
`,hne=`
|
|
vec4 result = min(x, vec4(6.)) * vec4(greaterThanEqual(x, vec4(0.0)));
|
|
bvec4 isNaN = isnan(x);
|
|
|
|
result.r = isNaN.r ? x.r : result.r;
|
|
result.g = isNaN.g ? x.g : result.g;
|
|
result.b = isNaN.b ? x.b : result.b;
|
|
result.a = isNaN.a ? x.a : result.a;
|
|
|
|
return result;
|
|
`,fne="return 1.0 / (1.0 + exp(-1.0 * x));",Ui=class{constructor(e,t){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length),this.userCode=`
|
|
vec4 unaryOperation(vec4 x) {
|
|
${t}
|
|
}
|
|
|
|
void main() {
|
|
vec4 x = getAAtOutCoords();
|
|
vec4 y = unaryOperation(x);
|
|
|
|
setOutput(y);
|
|
}
|
|
`}},mne=class{constructor(e){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!1,this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length);let t=e.length,n=es("rc",t),s=wt(t),r=Jte(t,n),a=n.slice(-2),o=t<=1?"rc":`vec2(${a.join(",")})`;this.userCode=`
|
|
void main() {
|
|
${s} rc = getOutputCoords();
|
|
vec4 packedInput = getA(${r});
|
|
|
|
setOutput(getChannel(packedInput, ${o}));
|
|
}
|
|
`}},gne=pr.whereImpl,yne=1e-7,Ane=1e-4,Zf={};function xne(e){return e in Zf||(Zf[e]={}),Zf[e]}var bne=q().getNumber("CPU_HANDOFF_SIZE_THRESHOLD"),vne=600;function wne(){return q().global.screen==null?1024:q().global.screen.height*q().global.screen.width*window.devicePixelRatio*vne/1024/1024}var id=class extends cc{constructor(e){if(super(),this.pendingRead=new WeakMap,this.pendingDisposal=new WeakSet,this.dataRefCount=new WeakMap,this.numBytesInGPU=0,this.uploadWaitMs=0,this.downloadWaitMs=0,this.lastGlFlushTime=0,this.warnedAboutMemory=!1,this.pendingDeletes=0,this.disposed=!1,!q().getBool("HAS_WEBGL"))throw new Error("WebGL is not supported on this device");let t;if(e!=null){if(e instanceof ju)t=e;else{let n=$r(q().getNumber("WEBGL_VERSION"),e);t=new ju(n)}this.binaryCache={},this.gpgpuCreatedLocally=!1}else{let n=$r(q().getNumber("WEBGL_VERSION"));t=new ju(n),this.binaryCache=xne(q().getNumber("WEBGL_VERSION")),this.gpgpuCreatedLocally=!0}this.gpgpu=t,this.canvas=this.gpgpu.gl.canvas,this.textureManager=new tne(this.gpgpu),this.numMBBeforeWarning=wne(),this.texData=new Mp(this,sn())}nextDataId(){return id.nextDataId++}numDataIds(){return this.texData.numDataIds()-this.pendingDeletes}write(e,t,n){if((q().getBool("WEBGL_CHECK_NUMERICAL_PROBLEMS")||q().getBool("DEBUG"))&&this.checkNumericalProblems(e),n==="complex64"&&e!=null)throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");let s={id:this.nextDataId()};return this.texData.set(s,{shape:t,dtype:n,values:e,usage:Xs.UPLOAD,refCount:1}),s}refCount(e){return this.texData.has(e)?this.texData.get(e).refCount:0}incRef(e){let t=this.texData.get(e);t.refCount++}decRef(e){if(this.texData.has(e)){let t=this.texData.get(e);t.refCount--}}move(e,t,n,s,r){if(q().getBool("DEBUG")&&this.checkNumericalProblems(t),s==="complex64")throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");this.texData.set(e,{shape:n,dtype:s,values:t,usage:Xs.UPLOAD,refCount:r})}disposeIntermediateTensorInfo(e){this.disposeData(e.dataId)}readSync(e){let t=this.texData.get(e),{values:n,dtype:s,complexTensorInfos:r,slice:a,shape:o,isPacked:i}=t;if(a!=null){let p;i?p=new Ui(o,Pu):p=new pa(o,Pu);let d=this.runWebGLProgram(p,[{dataId:e,shape:o,dtype:s}],s),h=this.readSync(d.dataId);return this.disposeIntermediateTensorInfo(d),h}if(n!=null)return this.convertAndCacheOnCPU(e);if(s==="string")return n;let l=this.activeTimers!=null,u;l&&(u=v.now());let c;if(s==="complex64"){let p=this.readSync(r.real.dataId),d=this.readSync(r.imag.dataId);c=T.mergeRealAndImagArrays(p,d)}else c=this.getValuesFromTexture(e);return l&&(this.downloadWaitMs+=v.now()-u),this.convertAndCacheOnCPU(e,c)}async read(e){if(this.pendingRead.has(e)){let h=this.pendingRead.get(e);return new Promise(f=>h.push(f))}let t=this.texData.get(e),{values:n,shape:s,slice:r,dtype:a,complexTensorInfos:o,isPacked:i}=t;if(r!=null){let h;i?h=new Ui(s,Pu):h=new pa(s,Pu);let f=this.runWebGLProgram(h,[{dataId:e,shape:s,dtype:a}],a),m=this.read(f.dataId);return this.disposeIntermediateTensorInfo(f),m}if(n!=null)return this.convertAndCacheOnCPU(e);if(q().getBool("DEBUG")&&!q().getBool("WEBGL_DOWNLOAD_FLOAT_ENABLED")&&q().getNumber("WEBGL_VERSION")===2)throw new Error("tensor.data() with WEBGL_DOWNLOAD_FLOAT_ENABLED=false and WEBGL_VERSION=2 not yet supported.");let l=null,u;if(a!=="complex64"&&q().get("WEBGL_BUFFER_SUPPORTED")){u=this.decode(e);let h=this.texData.get(u.dataId);l=this.gpgpu.createBufferFromTexture(h.texture.texture,...Xf(s))}this.pendingRead.set(e,[]),a!=="complex64"&&await this.gpgpu.createAndWaitForFence();let c;if(a==="complex64"){let h=await Promise.all([this.read(o.real.dataId),this.read(o.imag.dataId)]),f=h[0],m=h[1];c=T.mergeRealAndImagArrays(f,m)}else if(l==null)c=this.getValuesFromTexture(e);else{let h=v.sizeFromShape(s);c=this.gpgpu.downloadFloat32MatrixFromBuffer(l,h)}if(u!=null&&this.disposeIntermediateTensorInfo(u),l!=null){let h=this.gpgpu.gl;Ie(h,()=>h.deleteBuffer(l))}let p=this.convertAndCacheOnCPU(e,c),d=this.pendingRead.get(e);return this.pendingRead.delete(e),d.forEach(h=>h(p)),this.pendingDisposal.has(e)&&(this.pendingDisposal.delete(e),this.disposeData(e)&&sn().removeDataId(e,this),this.pendingDeletes--),p}readToGPU(e,t={}){let n=this.texData.get(e),{values:s,shape:r,slice:a,dtype:o,isPacked:i,texture:l}=n;if(o==="complex64")throw new Error("Does not support reading texture for complex64 dtype.");if(a!=null){let d;i?d=new Ui(r,Pu):d=new pa(r,Pu);let h=this.runWebGLProgram(d,[{dataId:e,shape:r,dtype:o}],o),f=this.readToGPU(h,t);return this.disposeIntermediateTensorInfo(h),f}if(l==null)throw s!=null?new Error("Data is not on GPU but on CPU."):new Error("There is no data on GPU or CPU.");let u=this.decode(e,t.customTexShape),c=sn().makeTensorFromTensorInfo(u),p=this.texData.get(u.dataId);return Object.assign({tensorRef:c},p.texture)}bufferSync(e){let t=this.readSync(e.dataId);if(e.dtype==="string")try{let n=t.map(s=>v.decodeString(s));return Le(e.shape,e.dtype,n)}catch(n){throw new Error("Failed to decode encoded string bytes into utf-8")}return Le(e.shape,e.dtype,t)}checkNumericalProblems(e){if(e!=null)for(let t=0;t<e.length;t++){let n=e[t];if(!hS(n))throw q().getBool("WEBGL_RENDER_FLOAT32_CAPABLE")?Error(`The value ${n} cannot be represented with your current settings. Consider enabling float32 rendering: 'tf.env().set('WEBGL_RENDER_FLOAT32_ENABLED', true);'`):Error(`The value ${n} cannot be represented on this device.`)}}getValuesFromTexture(e){let{shape:t,dtype:n,isPacked:s}=this.texData.get(e),r=v.sizeFromShape(t);if(q().getBool("WEBGL_DOWNLOAD_FLOAT_ENABLED")){let p=this.decode(e),d=this.texData.get(p.dataId),h=this.gpgpu.downloadMatrixFromPackedTexture(d.texture.texture,...Xf(t)).subarray(0,r);return this.disposeIntermediateTensorInfo(p),h}let a=q().getBool("WEBGL_PACK")&&s===!0,o=a?rm(t):t,i=a?new ute(o):new lte(o),l=this.runWebGLProgram(i,[{shape:o,dtype:n,dataId:e}],"float32"),u=this.texData.get(l.dataId),c=this.gpgpu.downloadByteEncodedFloatMatrixFromOutputTexture(u.texture.texture,u.texShape[0],u.texShape[1]).subarray(0,r);return this.disposeIntermediateTensorInfo(l),c}timerAvailable(){return q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0}time(e){let t=this.activeTimers,n=[],s=!1;this.programTimersStack==null?(this.programTimersStack=n,s=!0):this.activeTimers.push(n),this.activeTimers=n,e();let r=v.flatten(this.activeTimers.map(i=>i.query)).filter(i=>i!=null),a=v.flatten(this.activeTimers.map(i=>i.name)).filter(i=>i!=null);this.activeTimers=t,s&&(this.programTimersStack=null);let o={uploadWaitMs:this.uploadWaitMs,downloadWaitMs:this.downloadWaitMs,kernelMs:null,wallMs:null};return(async()=>{if(q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0){let i=await Promise.all(r);o.kernelMs=v.sum(i),o.getExtraProfileInfo=()=>i.map((l,u)=>({name:a[u],ms:l})).map(l=>`${l.name}: ${l.ms}`).join(", ")}else o.kernelMs={error:"WebGL query timers are not supported in this environment."};return this.uploadWaitMs=0,this.downloadWaitMs=0,o})()}memory(){return{unreliable:!1,numBytesInGPU:this.numBytesInGPU,numBytesInGPUAllocated:this.textureManager.numBytesAllocated,numBytesInGPUFree:this.textureManager.numBytesFree}}startTimer(){return q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0?this.gpgpu.beginQuery():{startMs:v.now(),endMs:null}}endTimer(e){return q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0?(this.gpgpu.endQuery(),e):(e.endMs=v.now(),e)}async getQueryTime(e){if(q().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0)return this.gpgpu.waitForQueryAndGetTime(e);let t=e;return t.endMs-t.startMs}disposeData(e,t=!1){if(this.pendingDisposal.has(e))return!1;if(!this.texData.has(e))return!0;if(t?this.texData.get(e).refCount=0:this.texData.get(e).refCount--,!t&&this.texData.get(e).refCount>0)return!1;if(this.pendingRead.has(e))return this.pendingDisposal.add(e),this.pendingDeletes++,!1;this.releaseGPUData(e);let{complexTensorInfos:n}=this.texData.get(e);return n!=null&&(this.disposeData(n.real.dataId,t),this.disposeData(n.imag.dataId,t)),this.texData.delete(e),!0}releaseGPUData(e){let{texture:t,dtype:n,texShape:s,usage:r,isPacked:a,slice:o}=this.texData.get(e),i=o&&o.origDataId||e,l=this.dataRefCount.get(i);l>1?this.dataRefCount.set(i,l-1):(this.dataRefCount.delete(i),t!=null&&(this.numBytesInGPU-=this.computeBytes(s,n),this.textureManager.releaseTexture(t,s,r,a)));let u=this.texData.get(e);u.texture=null,u.texShape=null,u.isPacked=!1,u.slice=null}getTexture(e){return this.uploadToGPU(e),this.texData.get(e).texture.texture}getDataInfo(e){return this.texData.get(e)}shouldExecuteOnCPU(e,t=bne){return q().getBool("WEBGL_CPU_FORWARD")&&e.every(n=>this.texData.get(n.dataId).texture==null&&v.sizeFromShape(n.shape)<t)}getGPGPUContext(){return this.gpgpu}where(e){T.warn("tf.where() in webgl locks the UI thread. Call tf.whereAsync() instead");let t=e.dataSync();return gne(e.shape,t)}packedUnaryOp(e,t,n){let s=new Ui(e.shape,t),r=this.compileAndRun(s,[e],n);return sn().makeTensorFromTensorInfo(r)}abs(e){if(this.shouldExecuteOnCPU([e])&&e.dtype!=="complex64"){let s=o9(this.texData.get(e.dataId).values);return this.makeOutput(e.shape,e.dtype,s)}if(q().getBool("WEBGL_PACK_UNARY_OPERATIONS"))return this.packedUnaryOp(e,A7,e.dtype);let t=new pa(e.shape,A7),n=this.compileAndRun(t,[e]);return sn().makeTensorFromTensorInfo(n)}makeTensorInfo(e,t,n){let s;if(t==="string"&&n!=null&&n.length>0&&v.isString(n[0])){let r=n.map(a=>v.encodeString(a));s=this.write(r,e,t)}else s=this.write(n,e,t);return this.texData.get(s).usage=null,{dataId:s,shape:e,dtype:t}}makeOutput(e,t,n){return sn().makeTensorFromTensorInfo(this.makeTensorInfo(e,t,n),this)}unpackTensor(e){let t=new mne(e.shape);return this.runWebGLProgram(t,[e],e.dtype)}packTensor(e){let t=new Qte(e.shape),n=!0;return this.runWebGLProgram(t,[e],e.dtype,null,n)}packedReshape(e,t){let n=[sl(e.shape),...rl(e.shape)],s={dtype:e.dtype,shape:n,dataId:e.dataId},r=[sl(t),...rl(t)],a=new u9(r,n),o=!0,i=[n],l=this.runWebGLProgram(a,[s],e.dtype,i,o);return{dataId:l.dataId,shape:t,dtype:l.dtype}}decode(e,t){let n=this.texData.get(e),{isPacked:s,shape:r,dtype:a}=n;if(t!=null){let p=v.sizeFromShape(r),d=t[0]*t[1]*4;v.assert(p<=d,()=>"customTexShape is too small. Row * Column * 4 should be equal or larger than the size of the tensor data.")}let o=rm(r),i;s?i=new ite(o):i=new ote(o);let l=!0,u=[t!=null?t:Xf(o)],c=this.runWebGLProgram(i,[{shape:o,dtype:a,dataId:e}],a,u,l,t);return{dtype:a,shape:r,dataId:c.dataId}}runWebGLProgram(e,t,n,s,r=!1,a){let o=this.makeTensorInfo(e.outputShape,n),i=this.texData.get(o.dataId);if(e.packedOutput&&(i.isPacked=!0),e.outPackingScheme===_p.DENSE){let g=a!=null?a:Xf(e.outputShape);i.texShape=g.map(y=>y*2)}if(e.outTexUsage!=null&&(i.usage=e.outTexUsage),v.sizeFromShape(o.shape)===0)return i.values=v.getTypedArrayFromDType(o.dtype,0),o;let l=[],u=t.map(g=>{if(g.dtype==="complex64")throw new Error("GPGPUProgram does not support complex64 input. For complex64 dtypes, please separate the program into real and imaginary parts.");let y=this.texData.get(g.dataId);if(y.texture==null){if(!e.packedInputs&&v.sizeFromShape(g.shape)<=q().getNumber("WEBGL_SIZE_UPLOAD_UNIFORM"))return{shape:g.shape,texData:null,isUniform:!0,uniformValues:y.values};e.packedInputs&&(y.isPacked=!0,y.shape=g.shape)}if(this.uploadToGPU(g.dataId),!!y.isPacked!=!!e.packedInputs)g=y.isPacked?this.unpackTensor(g):this.packTensor(g),l.push(g),y=this.texData.get(g.dataId);else if(y.isPacked&&!Dp(y.shape,g.shape)){let x=g,A=g.shape;g.shape=y.shape,g=this.packedReshape(g,A),l.push(g),y=this.texData.get(g.dataId),x.shape=A}return{shape:g.shape,texData:y,isUniform:!1}});this.uploadToGPU(o.dataId);let c={shape:o.shape,texData:i,isUniform:!1},p=ate(e,u,c),d=this.getAndSaveBinary(p,()=>ste(this.gpgpu,e,u,c)),h=this.activeTimers!=null,f;h&&(f=this.startTimer()),q().get("ENGINE_COMPILE_ONLY")||rte(this.gpgpu,d,u,c,s),l.forEach(g=>this.disposeIntermediateTensorInfo(g)),h&&(f=this.endTimer(f),this.activeTimers.push({name:e.constructor.name,query:this.getQueryTime(f)}));let m=q().get("WEBGL_FLUSH_THRESHOLD");if(m>0){let g=v.now();g-this.lastGlFlushTime>m&&(this.gpgpu.gl.flush(),this.lastGlFlushTime=g)}if(!q().getBool("WEBGL_LAZILY_UNPACK")&&i.isPacked&&r===!1){let g=this.unpackTensor(o);return this.disposeIntermediateTensorInfo(o),g}return o}compileAndRun(e,t,n,s,r=!1){return n=n||t[0].dtype,this.runWebGLProgram(e,t,n,s,r)}getAndSaveBinary(e,t){return e in this.binaryCache||(this.binaryCache[e]=t()),this.binaryCache[e]}getTextureManager(){return this.textureManager}dispose(){this.disposed||(q().getBool("IS_TEST")||Object.keys(this.binaryCache).forEach(t=>{this.gpgpu.deleteProgram(this.binaryCache[t].webGLProgram),delete this.binaryCache[t]}),this.textureManager.dispose(),this.canvas!=null&&typeof HTMLCanvasElement!="undefined"&&this.canvas instanceof HTMLCanvasElement?this.canvas.remove():this.canvas=null,this.gpgpuCreatedLocally&&(this.gpgpu.program=null,this.gpgpu.dispose()),this.disposed=!0)}floatPrecision(){return this.floatPrecisionValue==null&&(this.floatPrecisionValue=Y(()=>{if(!q().get("WEBGL_RENDER_FLOAT32_ENABLED")){let e=q().getBool("DEBUG");q().set("DEBUG",!1);let t=this.abs(Ce(1e-8)).dataSync()[0];if(q().set("DEBUG",e),t>0)return 32}return 16})),this.floatPrecisionValue}epsilon(){return this.floatPrecision()===32?yne:Ane}uploadToGPU(e){let t=this.texData.get(e),{shape:n,dtype:s,values:r,texture:a,usage:o,isPacked:i}=t;if(a!=null)return;let l=this.activeTimers!=null,u;l&&(u=v.now());let c=t.texShape;if(c==null&&(c=RS(n,i),t.texShape=c),r!=null){let p=rm(n),d,h=c[1],f=c[0],m=r instanceof Uint8Array||r instanceof Uint8ClampedArray;(i||!m)&&([h,f]=td(c[0],c[1])),i?d=new dte(p,m):d=new cte(p,m);let g=m?[f,h]:c,y=this.makeTensorInfo(g,s),x=this.texData.get(y.dataId);m?x.usage=Xs.PIXELS:x.usage=Xs.UPLOAD,x.texShape=g,this.gpgpu.uploadDenseMatrixToTexture(this.getTexture(y.dataId),h,f,r);let A=[[f,h]],b=!0,w=this.runWebGLProgram(d,[y],s,A,b),k=this.texData.get(w.dataId);t.texShape=k.texShape,t.isPacked=k.isPacked,t.usage=k.usage,q().get("ENGINE_COMPILE_ONLY")?this.disposeData(w.dataId):(t.texture=k.texture,t.values=null,this.texData.delete(w.dataId)),this.disposeIntermediateTensorInfo(y),l&&(this.uploadWaitMs+=v.now()-u)}else{let p=this.acquireTexture(c,o,s,i);t.texture=p}}convertAndCacheOnCPU(e,t){let n=this.texData.get(e),{dtype:s}=n;return this.releaseGPUData(e),t!=null&&(n.values=kne(t,s)),n.values}acquireTexture(e,t,n,s){if(this.numBytesInGPU+=this.computeBytes(e,n),!this.warnedAboutMemory&&this.numBytesInGPU>this.numMBBeforeWarning*1024*1024){let r=(this.numBytesInGPU/1024/1024).toFixed(2);this.warnedAboutMemory=!0,console.warn(`High memory usage in GPU: ${r} MB, most likely due to a memory leak`)}return this.textureManager.acquireTexture(e,t,s)}computeBytes(e,t){return e[0]*e[1]*v.bytesPerElement(t)}checkCompileCompletion(){for(let[,e]of Object.entries(this.binaryCache))this.checkCompletion_(e)}async checkCompileCompletionAsync(){let e=[];if(this.gpgpu.parallelCompilationExtension){for(let[,t]of Object.entries(this.binaryCache))e.push(this.checkCompletionAsync_(t));return Promise.all(e)}else{for(let[,t]of Object.entries(this.binaryCache)){let n=new Promise(s=>{try{this.checkCompletion_(t),s(!0)}catch(r){throw r}});e.push(n)}return Promise.all(e)}}async checkCompletionAsync_(e){return this.gpgpu.gl.getProgramParameter(e.webGLProgram,this.gpgpu.parallelCompilationExtension.COMPLETION_STATUS_KHR)?this.checkCompletion_(e):(await qA(),this.checkCompletionAsync_(e))}checkCompletion_(e){if(this.gpgpu.gl.getProgramParameter(e.webGLProgram,this.gpgpu.gl.LINK_STATUS)===!1)throw console.log(this.gpgpu.gl.getProgramInfoLog(e.webGLProgram)),this.gpgpu.gl.getShaderParameter(e.fragmentShader,this.gpgpu.gl.COMPILE_STATUS)===!1?(Vx(e.source,this.gpgpu.gl.getShaderInfoLog(e.fragmentShader)),new Error("Failed to compile fragment shader.")):new Error("Failed to link vertex and fragment shaders.");return!0}getUniformLocations(){for(let[,e]of Object.entries(this.binaryCache)){let{uniformLocations:t,customUniformLocations:n,infLoc:s,nanLoc:r,inShapesLocations:a,inTexShapesLocations:o,outShapeLocation:i,outShapeStridesLocation:l,outTexShapeLocation:u}=WS(this.gpgpu,e.program,e.webGLProgram);e.uniformLocations=t,e.customUniformLocations=n,e.infLoc=s,e.nanLoc=r,e.inShapesLocations=a,e.inTexShapesLocations=o,e.outShapeLocation=i,e.outShapeStridesLocation=l,e.outTexShapeLocation=u}}};id.nextDataId=0;function kne(e,t){if(t==="float32"||t==="complex64")return e;if(t==="int32"||t==="bool"){let n=t==="int32"?new Int32Array(e.length):new Uint8Array(e.length);for(let s=0;s<n.length;++s)n[s]=Math.round(e[s]);return n}else throw new Error(`Unknown dtype ${t}`)}var Ine="3.19.0";function c9(){q().set("WEBGL_FORCE_F16_TEXTURES",!0)}rh.isBrowser()&&ql("webgl",()=>new id,2);var Sne={forceHalfFloat:c9},d9=`
|
|
if (isnan(a)) return a;
|
|
if (isnan(b)) return b;
|
|
`,uc=class{constructor(e,t,n){this.variableNames=["A","B"],this.outputShape=T.assertAndGetBroadcastShape(t,n),this.enableShapeUniforms=xs(this.outputShape.length),this.userCode=`
|
|
float binaryOperation(float a, float b) {
|
|
${e}
|
|
}
|
|
|
|
void main() {
|
|
float a = getAAtOutCoords();
|
|
float b = getBAtOutCoords();
|
|
setOutput(binaryOperation(a, b));
|
|
}
|
|
`}},v2=`
|
|
result.r = isNaN.r > 0. ? NAN : result.r;
|
|
result.g = isNaN.g > 0. ? NAN : result.g;
|
|
result.b = isNaN.b > 0. ? NAN : result.b;
|
|
result.a = isNaN.a > 0. ? NAN : result.a;
|
|
`,Mh=class{constructor(e,t,n,s=!1){this.variableNames=["A","B"],this.supportsBroadcasting=!0,this.packedInputs=!0,this.packedOutput=!0,this.outputShape=T.assertAndGetBroadcastShape(t,n);let r=this.outputShape.length;this.enableShapeUniforms=xs(r);let a="";if(s)if(r===0||v.sizeFromShape(this.outputShape)===1)a=`
|
|
result.y = 0.;
|
|
result.z = 0.;
|
|
result.w = 0.;
|
|
`;else if(a=`
|
|
${wt(r)} coords = getOutputCoords();
|
|
`,r===1)this.enableShapeUniforms?a+=`
|
|
result.y = (coords + 1) >= outShape ? 0. : result.y;
|
|
result.z = 0.;
|
|
result.w = 0.;
|
|
`:a+=`
|
|
result.y = (coords + 1) >= ${this.outputShape[0]} ? 0. : result.y;
|
|
result.z = 0.;
|
|
result.w = 0.;
|
|
`;else{let i=es("coords",r);this.enableShapeUniforms?a+=`
|
|
bool nextRowOutOfBounds =
|
|
(${i[r-2]} + 1) >= outShape[${r} - 2];
|
|
bool nextColOutOfBounds =
|
|
(${i[r-1]} + 1) >= outShape[${r} - 1];
|
|
result.y = nextColOutOfBounds ? 0. : result.y;
|
|
result.z = nextRowOutOfBounds ? 0. : result.z;
|
|
result.w = nextColOutOfBounds || nextRowOutOfBounds ? 0. : result.w;
|
|
`:a+=`
|
|
bool nextRowOutOfBounds =
|
|
(${i[r-2]} + 1) >= ${this.outputShape[r-2]};
|
|
bool nextColOutOfBounds =
|
|
(${i[r-1]} + 1) >= ${this.outputShape[r-1]};
|
|
result.y = nextColOutOfBounds ? 0. : result.y;
|
|
result.z = nextRowOutOfBounds ? 0. : result.z;
|
|
result.w = nextColOutOfBounds || nextRowOutOfBounds ? 0. : result.w;
|
|
`}this.userCode=`
|
|
vec4 binaryOperation(vec4 a, vec4 b) {
|
|
${e}
|
|
}
|
|
|
|
void main() {
|
|
vec4 a = getAAtOutCoords();
|
|
vec4 b = getBAtOutCoords();
|
|
|
|
vec4 result = binaryOperation(a, b);
|
|
${a}
|
|
|
|
setOutput(result);
|
|
}
|
|
`}};function Ps(e){let{inputs:t,backend:n}=e,{x:s}=t;return n.incRef(s.dataId),{dataId:s.dataId,shape:s.shape,dtype:s.dtype}}var Cne={kernelName:_o,backendName:"webgl",kernelFunc:Ps};function pi(e){let{inputs:t,backend:n}=e,{real:s,imag:r}=t,a=n.makeTensorInfo(s.shape,"complex64"),o=n.texData.get(a.dataId),i=Ps({inputs:{x:s},backend:n}),l=Ps({inputs:{x:r},backend:n});return o.complexTensorInfos={real:i,imag:l},a}var Tne={kernelName:Lp,backendName:"webgl",kernelFunc:pi},p9="return (a < 0.) ? b * a : a;",h9=`
|
|
vec4 aLessThanZero = vec4(lessThan(a, vec4(0.)));
|
|
return (aLessThanZero * (b * a)) + ((vec4(1.0) - aLessThanZero) * a);
|
|
`;function Nne(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{alpha:a}=s,o=n.makeTensorInfo([],"float32",v.createScalarValue(a,"float32")),i=q().getBool("WEBGL_PACK_BINARY_OPERATIONS")?new Mh(h9,r.shape,o.shape):new uc(p9,r.shape,o.shape),l=n.runWebGLProgram(i,[r,o],"float32");return n.disposeIntermediateTensorInfo(o),l}var Ene={kernelName:Do,backendName:"webgl",kernelFunc:Nne},f9="return (a < 0.) ? b * a : a;",m9=`
|
|
vec4 aLessThanZero = vec4(lessThan(a, vec4(0.)));
|
|
return (aLessThanZero * (b * a)) + ((vec4(1.0) - aLessThanZero) * a);
|
|
`;function Rne(e){let{inputs:t,backend:n}=e,{x:s,alpha:r}=t,a=q().getBool("WEBGL_PACK_BINARY_OPERATIONS")?new Mh(m9,s.shape,r.shape):new uc(f9,s.shape,r.shape);return n.runWebGLProgram(a,[s,r],"float32")}var _ne={kernelName:Go,backendName:"webgl",kernelFunc:Rne},ld="if (isnan(x)) return x;",Dne=`
|
|
if (isnan(a)) return a;
|
|
if (isnan(b)) return b;
|
|
`,$ne=`
|
|
result.r = isNaN.r > 0. ? NAN : result.r;
|
|
result.g = isNaN.g > 0. ? NAN : result.g;
|
|
result.b = isNaN.b > 0. ? NAN : result.b;
|
|
result.a = isNaN.a > 0. ? NAN : result.a;
|
|
`;function dt({opSnippet:e,packedOpSnippet:t,cpuKernelImpl:n,dtype:s}){return({inputs:r,backend:a})=>{let{x:o}=r,i=a,l=s||o.dtype;if(i.shouldExecuteOnCPU([o])&&n!=null){let p=i.texData.get(o.dataId),d=n(p.values,l);return i.makeTensorInfo(o.shape,l,d)}let u=q().getBool("WEBGL_PACK_UNARY_OPERATIONS")&&t!=null,c;return u?c=new Ui(o.shape,t):c=new pa(o.shape,e),i.runWebGLProgram(c,[o],l)}}function _n({opSnippet:e,packedOpSnippet:t,checkOutOfBounds:n=!1,supportsComplex:s=!1,cpuKernelImpl:r,dtype:a}){return({inputs:o,backend:i})=>{let{a:l,b:u}=o,c=i;if(s&&l.dtype==="complex64"){let f=c.texData.get(l.dataId),m=c.texData.get(u.dataId),[g,y]=[[f.complexTensorInfos.real,m.complexTensorInfos.real],[f.complexTensorInfos.imag,m.complexTensorInfos.imag]].map(A=>{let[b,w]=A,k={dataId:b.dataId,dtype:b.dtype,shape:l.shape},S={dataId:w.dataId,dtype:w.dtype,shape:u.shape},E=new uc(e,l.shape,u.shape);return c.runWebGLProgram(E,[k,S],Mn(b.dtype,w.dtype))}),x=pi({inputs:{real:g,imag:y},backend:c});return c.disposeIntermediateTensorInfo(g),c.disposeIntermediateTensorInfo(y),x}let p=a||Mn(l.dtype,u.dtype);if((l.dtype==="string"||u.dtype==="string"||c.shouldExecuteOnCPU([l,u]))&&r!=null){let f=c.texData.get(l.dataId).values,m=c.texData.get(u.dataId).values,g=l.dtype==="string"?T.fromUint8ToStringArray(f):f,y=l.dtype==="string"?T.fromUint8ToStringArray(m):m,[x,A]=r(l.shape,u.shape,g,y,p),b=c.makeTensorInfo(A,p),w=c.texData.get(b.dataId);return w.values=x,b}let d=q().getBool("WEBGL_PACK_BINARY_OPERATIONS")&&t!=null,h;return d?h=new Mh(t,l.shape,u.shape,n):h=new uc(e,l.shape,u.shape),c.runWebGLProgram(h,[l,u],p)}}function w2(e,t=!1){if(e==="linear")return t?cne:ane;if(e==="relu")return t?pne:ine;if(e==="elu")return t?dne:one;if(e==="relu6")return t?hne:lne;if(e==="prelu")return t?m9:f9;if(e==="leakyrelu")return t?h9:p9;if(e==="sigmoid")return t?fne:une;throw new Error(`Activation ${e} has not been implemented for the WebGL backend.`)}var g9=class{constructor(e,t,n,s=!1,r=!1,a=!1,o=null,i=!1,l=!1){this.variableNames=["matrixA","matrixB"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=n,this.enableShapeUniforms=xs(this.outputShape.length);let u=s?e[1]:e[2],c=Math.ceil(u/2),p=s?"i * 2, rc.y":"rc.y, i * 2",d=r?"rc.z, i * 2":"i * 2, rc.z",h=s?["a.xxyy","a.zzww"]:["a.xxzz","a.yyww"],f=r?["b.xzxz","b.ywyw"]:["b.xyxy","b.zwzw"],m="",g="";o&&(i?m=`vec4 activation(vec4 a) {
|
|
vec4 b = getPreluActivationWeightsAtOutCoords();
|
|
${o}
|
|
}`:l?m=`vec4 activation(vec4 a) {
|
|
vec4 b = getLeakyreluAlphaAtOutCoords();
|
|
${o}
|
|
}`:m=`vec4 activation(vec4 x) {
|
|
${o}
|
|
}`,g="result = activation(result);");let y=a?"result += getBiasAtOutCoords();":"";a&&this.variableNames.push("bias"),i&&this.variableNames.push("preluActivationWeights"),l&&this.variableNames.push("leakyreluAlpha");let x="rc.x",A="rc.x";e[0]<t[0]?x=`int(min(float(rc.x), ${e[0]-1}.))`:t[0]<e[0]&&(A=`int(min(float(rc.x), ${t[0]-1}.))`),this.userCode=`
|
|
${m}
|
|
// Don't use uniform for sharedDimensionPacked for performance.
|
|
const float sharedDimension = ${c}.0;
|
|
|
|
vec4 dot2x2ARowBCol(ivec3 rc) {
|
|
vec4 result = vec4(0);
|
|
for (int i = 0; i < ${c}; i++) {
|
|
int batchA = ${x};
|
|
int batchB = ${A};
|
|
vec4 a = getMatrixA(batchA, ${p});
|
|
vec4 b = getMatrixB(batchB, ${d});
|
|
|
|
// These swizzled products need to be separately added.
|
|
// See: https://github.com/tensorflow/tfjs/issues/1735
|
|
result += (${h[0]} * ${f[0]});
|
|
result += (${h[1]} * ${f[1]});
|
|
}
|
|
return result;
|
|
}
|
|
|
|
void main() {
|
|
ivec3 rc = getOutputCoords();
|
|
vec4 result = dot2x2ARowBCol(rc);
|
|
|
|
${y}
|
|
|
|
${g}
|
|
|
|
setOutput(result);
|
|
}
|
|
`}},x7={REAL:"return areal * breal - aimag * bimag;",IMAG:"return areal * bimag + aimag * breal;"},b7=class{constructor(e,t,n){this.variableNames=["AReal","AImag","BReal","BImag"],this.outputShape=T.assertAndGetBroadcastShape(t,n),this.userCode=`
|
|
float binaryOpComplex(
|
|
float areal, float aimag, float breal, float bimag) {
|
|
${e}
|
|
}
|
|
|
|
void main() {
|
|
float areal = getARealAtOutCoords();
|
|
float aimag = getAImagAtOutCoords();
|
|
float breal = getBRealAtOutCoords();
|
|
float bimag = getBImagAtOutCoords();
|
|
setOutput(binaryOpComplex(areal, aimag, breal, bimag));
|
|
}
|
|
`}},v7="return a * b;";function Jx(e){let{inputs:t,backend:n}=e,{a:s,b:r}=t,a=T.upcastType(s.dtype,r.dtype);if(s.dtype==="complex64"){let i=n.texData.get(s.dataId),l=n.texData.get(r.dataId),u=new b7(x7.REAL,s.shape,r.shape),c=new b7(x7.IMAG,s.shape,r.shape),p=[{dataId:i.complexTensorInfos.real.dataId,dtype:i.complexTensorInfos.real.dtype,shape:s.shape},{dataId:i.complexTensorInfos.imag.dataId,dtype:i.complexTensorInfos.imag.dtype,shape:s.shape},{dataId:l.complexTensorInfos.real.dataId,dtype:l.complexTensorInfos.real.dtype,shape:r.shape},{dataId:l.complexTensorInfos.imag.dataId,dtype:l.complexTensorInfos.imag.dtype,shape:r.shape}],d=n.runWebGLProgram(u,p,"float32"),h=n.runWebGLProgram(c,p,"float32"),f=pi({inputs:{real:d,imag:h},backend:n});return n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),f}if(n.shouldExecuteOnCPU([s,r])){let i=n.texData.get(s.dataId),l=n.texData.get(r.dataId),[u,c]=Dte(s.shape,r.shape,i.values,l.values,a),p=n.makeTensorInfo(c,a),d=n.texData.get(p.dataId);return d.values=u,p}let o;return q().getBool("WEBGL_PACK_BINARY_OPERATIONS")?o=new Mh(v7,s.shape,r.shape):o=new uc(v7,s.shape,r.shape),n.runWebGLProgram(o,[s,r],a)}var Pne={kernelName:Wo,backendName:"webgl",kernelFunc:Jx};function Fne(e,t,n){let s=[sl(e.shape),...rl(e.shape)],r={dtype:e.dtype,shape:s,dataId:e.dataId},a=[sl(t),...rl(t)],o=new u9(a,s),i=!0,l=[s],u=n.runWebGLProgram(o,[r],e.dtype,l,i);return{dataId:u.dataId,shape:t,dtype:u.dtype}}function be(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{shape:a}=s,o=n,i=v.sizeFromShape(r.shape),l=v.inferFromImplicitShape(a,i),u=v.sizeFromShape(l);v.assert(i===u,()=>`The new shape (${l}) has ${u} elements and the old shape (${r.shape}) has ${i} elements. The new shape and old shape must have the same number of elements.`);let c=o.texData.get(r.dataId);return c.isPacked&&!Dp(r.shape,l)&&!(c.texture!==null&&Dp(c.shape,l))?Fne(r,l,o):(o.incRef(r.dataId),{dataId:r.dataId,shape:l,dtype:r.dtype})}var One={kernelName:_l,backendName:"webgl",kernelFunc:be},w7=class{constructor(e,t){this.variableNames=["x"];let{windowSize:n,batchSize:s,inSize:r,outSize:a}=e;this.outputShape=[s,a];let o=Math.floor(n/4)*4,i=n%4,l="sumValue += dot(values, ones);";if(t!=null){let c=1/t;l=`sumValue += dot(values * ${v.isInt(c)?c.toPrecision(2):c}, ones);`}let u="";r%n>0&&(u=`
|
|
if (inIdx < 0 || inIdx >= ${r}) {
|
|
return 0.0;
|
|
}
|
|
`),this.userCode=`
|
|
const vec4 ones = vec4(1.0, 1.0, 1.0, 1.0);
|
|
|
|
float getValue(int batch, int inIdx) {
|
|
${u}
|
|
return getX(batch, inIdx);
|
|
}
|
|
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int outIdx = coords[1];
|
|
int inOffset = outIdx * ${n};
|
|
|
|
float sumValue = 0.0;
|
|
|
|
for (int i = 0; i < ${o}; i += 4) {
|
|
int inIdx = inOffset + i;
|
|
vec4 values = vec4(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1),
|
|
getValue(batch, inIdx + 2),
|
|
getValue(batch, inIdx + 3)
|
|
);
|
|
|
|
${l}
|
|
}
|
|
|
|
int inIdx = inOffset + ${o};
|
|
if (${i===1}) {
|
|
vec4 values = vec4(getValue(batch, inIdx), 0.0, 0.0, 0.0);
|
|
|
|
${l}
|
|
} else if (${i===2}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1), 0.0, 0.0);
|
|
|
|
${l}
|
|
} else if (${i===3}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1),
|
|
getValue(batch, inIdx + 2), 0.0);
|
|
|
|
${l}
|
|
}
|
|
setOutput(sumValue);
|
|
}
|
|
`}},Mne=class{constructor(e,t){this.variableNames=["x"];let{windowSize:n,batchSize:s,inSize:r,outSize:a}=e;this.outputShape=[s,a];let o="0.0",i="";t==="prod"?o="1.0":t==="min"?(o="1.0 / 1e-20",i="min"):t==="max"&&(o="-1.0 / 1e-20",i="max");let l=`${t}(${t}(${t}(minMaxValue[0], minMaxValue[1]), minMaxValue[2]), minMaxValue[3])`;t==="sum"?l="sumValue":t==="prod"?l="prodValue":t==="all"?l="allValue":t==="any"&&(l="anyValue");let u=Math.floor(n/4)*4,c=n%4,p=`
|
|
if (${t==="sum"}) {
|
|
sumValue += dot(values, ones);
|
|
} else if (${t==="prod"}) {
|
|
vec2 tmp = vec2(values[0], values[1]) * vec2(values[2], values[3]);
|
|
prodValue *= tmp[0] * tmp[1];
|
|
} else {
|
|
minMaxValue = ${i}(values, minMaxValue);
|
|
if (${t==="min"} || ${t==="max"}) {
|
|
minMaxValue = ${i}(values, minMaxValue);
|
|
bvec4 isNaN = isnan(values);
|
|
if (isNaN.r || isNaN.g || isNaN.b || isNaN.a) {
|
|
minMaxValue = vec4(NAN);
|
|
}
|
|
}
|
|
}
|
|
`,d="vec4";t==="all"?(o="1.0",p=`
|
|
bool reducedAllValue = all(values);
|
|
float floatedReducedAllValue = float(reducedAllValue);
|
|
allValue = float(allValue >= 1.0 && floatedReducedAllValue >= 1.0);
|
|
`,d="bvec4"):t==="any"&&(o="0.0",p=`
|
|
bool reducedAnyValue = any(values);
|
|
float floatedReducedAnyValue = float(reducedAnyValue);
|
|
anyValue = float(anyValue >= 1.0 || floatedReducedAnyValue >= 1.0);
|
|
`,d="bvec4");let h="";r%n>0&&(h=`
|
|
if (inIdx < 0 || inIdx >= ${r}) {
|
|
return initializationValue;
|
|
}
|
|
`),this.userCode=`
|
|
const float initializationValue = ${o};
|
|
const vec4 ones = vec4(1.0, 1.0, 1.0, 1.0);
|
|
|
|
float getValue(int batch, int inIdx) {
|
|
${h}
|
|
return getX(batch, inIdx);
|
|
}
|
|
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int outIdx = coords[1];
|
|
int inOffset = outIdx * ${n};
|
|
|
|
vec4 minMaxValue = vec4(${o});
|
|
float prodValue = 1.0;
|
|
float sumValue = 0.0;
|
|
float allValue = 1.0;
|
|
float anyValue = 0.0;
|
|
|
|
for (int i = 0; i < ${u}; i += 4) {
|
|
int inIdx = inOffset + i;
|
|
${d} values = ${d}(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1),
|
|
getValue(batch, inIdx + 2),
|
|
getValue(batch, inIdx + 3)
|
|
);
|
|
|
|
${p}
|
|
}
|
|
|
|
int inIdx = inOffset + ${u};
|
|
if (${c===1}) {
|
|
${d} values = ${d}(
|
|
getValue(batch, inIdx),
|
|
initializationValue,
|
|
initializationValue,
|
|
initializationValue
|
|
);
|
|
|
|
${p}
|
|
} else if (${c===2}) {
|
|
${d} values = ${d}(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1),
|
|
initializationValue,
|
|
initializationValue
|
|
);
|
|
|
|
${p}
|
|
} else if (${c===3}) {
|
|
${d} values = ${d}(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1),
|
|
getValue(batch, inIdx + 2),
|
|
initializationValue
|
|
);
|
|
|
|
${p}
|
|
}
|
|
setOutput(${l});
|
|
}
|
|
`}};function zne(e){let t=[];for(;t.length===0||t[t.length-1].outSize!==1;){let n=t.length?t[t.length-1].outSize:e[1],s=T.computeOptimalWindowSize(n);t.push({inSize:n,windowSize:s,outSize:Math.ceil(n/s)})}return t}function au(e,t,n,s){let r=zne(e.shape),a=e;for(let o=0;o<r.length;o++){let{inSize:i,windowSize:l,outSize:u}=r[o],c,p;n==="mean"?c=o===0?new w7({windowSize:l,inSize:i,batchSize:e.shape[0],outSize:u},i):new w7({windowSize:l,inSize:i,batchSize:e.shape[0],outSize:u}):c=new Mne({windowSize:l,inSize:i,batchSize:e.shape[0],outSize:u},n),p=a,a=s.runWebGLProgram(c,[a],t),p.dataId!==e.dataId&&s.disposeIntermediateTensorInfo(p)}return a}var Lne=class{constructor(e,t){this.variableNames=["A"];let n=new Array(e.length);for(let a=0;a<n.length;a++)n[a]=e[t[a]];this.outputShape=n,this.rank=n.length;let s=wt(this.rank),r=Bne(t);this.userCode=`
|
|
void main() {
|
|
${s} resRC = getOutputCoords();
|
|
setOutput(getA(${r}));
|
|
}
|
|
`}};function Bne(e){let t=e.length;if(t>6)throw Error(`Transpose for rank ${t} is not yet supported`);let n=["resRC.x","resRC.y","resRC.z","resRC.w","resRC.u","resRC.v"],s=new Array(t);for(let r=0;r<e.length;r++)s[e[r]]=n[r];return s.join()}var Wne=class{constructor(e,t){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0;let n=new Array(e.length);for(let u=0;u<n.length;u++)n[u]=e[t[u]];if(this.outputShape=n,this.rank=n.length,this.rank>6)throw Error(`Packed transpose for rank ${this.rank} is not yet supported.`);let s=wt(this.rank),r=l9("rc",this.rank),a=new Array(this.rank);for(let u=0;u<t.length;u++)a[t[u]]=r[u];let o=`vec2(${a.slice(-2).join()})`,i=`++${r[this.rank-1]} < ${n[this.rank-1]}`,l=`getChannel(getA(${a.join()}), ${o})`;this.userCode=`
|
|
void main() {
|
|
${s} rc = getOutputCoords();
|
|
vec4 result = vec4(0.);
|
|
result[0] = ${l};
|
|
if(${i}) {
|
|
result[1] = ${l};
|
|
}
|
|
--${r[this.rank-1]};
|
|
if(++${r[this.rank-2]} < ${n[this.rank-2]}) {
|
|
result[2] = ${l};
|
|
if(${i}) {
|
|
result[3] = ${l};
|
|
}
|
|
}
|
|
setOutput(result);
|
|
}
|
|
`}};function k2(e,t,n){let s=q().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new Wne(e.shape,t):new Lne(e.shape,t);return n.runWebGLProgram(s,[e],e.dtype)}function Vne(e,t,n,s){let r=t,a=e.shape.length,o=v.parseAxisParam(r,e.shape),i=o,l=T.getAxesPermutation(i,a),u=l!=null,c=e;u&&(c=k2(e,l,s),i=T.getInnerMostAxes(i.length,a)),T.assertAxesAreInnerMostDims("sum",i,a);let[p,d]=T.computeOutAndReduceShapes(c.shape,i),h=p;n&&(h=T.expandShapeToKeepDim(p,o));let f=v.sizeFromShape(d),g=v.sizeFromShape(e.shape)/f,y=be({inputs:{x:c},attrs:{shape:[g,f]},backend:s}),x=sh(e.dtype),A=au(y,x,"sum",s),b=be({inputs:{x:A},attrs:{shape:h},backend:s});return s.disposeIntermediateTensorInfo(y),s.disposeIntermediateTensorInfo(A),u&&s.disposeIntermediateTensorInfo(c),b}function I2(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;return Vne(r,a,o,n)}var Une={kernelName:ei,backendName:"webgl",kernelFunc:I2};function ts(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{perm:a}=s,o=n,i=r.shape.length,l=new Array(i);for(let c=0;c<l.length;c++)l[c]=r.shape[a[c]];let u;if(o.shouldExecuteOnCPU([r])){let p=o.texData.get(r.dataId).values,d=Yx(p,r.shape,r.dtype,a,l);u=o.makeTensorInfo(l,r.dtype);let h=o.texData.get(u.dataId);h.values=d}else u=k2(r,a,o);return u}var Gne={kernelName:Hr,backendName:"webgl",kernelFunc:ts},y9=1e3;function Fm({a:e,b:t,transposeA:n,transposeB:s,backend:r,bias:a=null,preluActivationWeights:o=null,leakyreluAlpha:i=0,activation:l=null}){let u=e.shape.length,c=t.shape.length,p=n?e.shape[u-2]:e.shape[u-1],d=s?t.shape[c-1]:t.shape[c-2],h=n?e.shape[u-1]:e.shape[u-2],f=s?t.shape[c-2]:t.shape[c-1],m=e.shape.slice(0,-2),g=t.shape.slice(0,-2),y=v.sizeFromShape(m),x=v.sizeFromShape(g),b=Xl.assertAndGetBroadcastShape(e.shape.slice(0,-2),t.shape.slice(0,-2)).concat([h,f]);v.assert(p===d,()=>`Error in matMul: inner shapes (${p}) and (${d}) of Tensors with shapes ${e.shape} and ${t.shape} and transposeA=${n} and transposeB=${s} must match.`);let w=n?[y,p,h]:[y,h,p],k=s?[x,f,d]:[x,d,f],S=be({inputs:{x:e},backend:r,attrs:{shape:w}}),E=be({inputs:{x:t},backend:r,attrs:{shape:k}}),R=[S,E],$=Math.max(y,x),_=n?S.shape[1]:S.shape[2],P=a!=null,C=o!=null,F=l==="leakyrelu",G=l!=null?w2(l,!0):null,K=P||C||F||G!=null,z;if((h===1||f===1)&&_>y9&&K===!1){let J=S,te=E;n&&(J=ts({inputs:{x:S},backend:r,attrs:{perm:[0,2,1]}}),R.push(J)),s&&(te=ts({inputs:{x:E},backend:r,attrs:{perm:[0,2,1]}}),R.push(te));let B=f!==1,oe=f===1,Q=J;B&&(Q=be({inputs:{x:J},backend:r,attrs:{shape:[$,_,1]}}),R.push(Q));let ae=f===1?2:1,ie=te;oe&&(ie=be({inputs:{x:te},backend:r,attrs:{shape:[$,1,_]}}),R.push(ie));let me=Jx({inputs:{a:Q,b:ie},backend:r});z=I2({inputs:{x:me},backend:r,attrs:{axis:ae,keepDims:!0}}),R.push(me)}else{let J=Mn(e.dtype,t.dtype),te=new g9(w,k,[$,h,f],n,s,P,G,C,F),B=[S,E];if(a!=null&&B.push(a),C&&B.push(o),F){let oe=r.makeTensorInfo([],"float32",v.createScalarValue(i,"float32"));B.push(oe),R.push(oe)}z=r.runWebGLProgram(te,B,J)}let Z=be({inputs:{x:z},backend:r,attrs:{shape:b}});R.push(z);for(let J of R)r.disposeIntermediateTensorInfo(J);return Z}function Hne(e){let{inputs:t,backend:n,attrs:s}=e,{a:r,b:a,bias:o,preluActivationWeights:i}=t,{transposeA:l,transposeB:u,activation:c,leakyreluAlpha:p}=s;return Fm({a:r,b:a,transposeA:l,transposeB:u,backend:n,bias:o,preluActivationWeights:i,leakyreluAlpha:p,activation:c})}var jne={kernelName:Ja,backendName:"webgl",kernelFunc:Hne},k7="return abs(x);";function qne(e){let{inputs:t,backend:n}=e,{x:s}=t;if(n.shouldExecuteOnCPU([s])&&s.dtype!=="complex64"){let a=n.texData.get(s.dataId),o=o9(a.values);return n.makeTensorInfo(s.shape,s.dtype,o)}let r;return q().getBool("WEBGL_PACK_UNARY_OPERATIONS")?r=new Ui(s.shape,k7):r=new pa(s.shape,k7),n.runWebGLProgram(r,[s],s.dtype)}var Xne={kernelName:il,backendName:"webgl",kernelFunc:qne},Kne=fr+`
|
|
if (abs(x) > 1.) {
|
|
return NAN;
|
|
}
|
|
return acos(x);
|
|
`,Zne=dt({opSnippet:Kne}),Yne={kernelName:pc,backendName:"webgl",kernelFunc:Zne},Jne=fr+`
|
|
if (x < 1.0) return NAN;
|
|
return log(x + sqrt(x * x - 1.0));`,Qne=dt({opSnippet:Jne}),ese={kernelName:hc,backendName:"webgl",kernelFunc:Qne},I7="return a + b;",tse=_n({opSnippet:I7,packedOpSnippet:I7,supportsComplex:!0,cpuKernelImpl:hte}),nse={kernelName:xa,backendName:"webgl",kernelFunc:tse},sse=class{constructor(e,t){this.outputShape=[],this.outputShape=e,this.variableNames=t.map((r,a)=>`T${a}`);let n=[];this.variableNames.forEach(r=>{n.push(`float v${r} = get${r}AtOutCoords();`)});let s=this.variableNames.map(r=>`v${r}`).join(" + ");this.userCode=`
|
|
void main() {
|
|
${n.join(`
|
|
`)}
|
|
|
|
float result = ${s};
|
|
setOutput(result);
|
|
}
|
|
`}},rse=class{constructor(e,t){this.outputShape=[],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e,this.variableNames=t.map((r,a)=>`T${a}`);let n=[];this.variableNames.forEach(r=>{n.push(`vec4 v${r} = get${r}AtOutCoords();`)});let s=this.variableNames.map(r=>`v${r}`).join(" + ");this.userCode=`
|
|
void main() {
|
|
${n.join(`
|
|
`)}
|
|
|
|
vec4 result = ${s};
|
|
setOutput(result);
|
|
}
|
|
`}};function im(e){let{inputs:t,backend:n}=e,s=t;if(s.length===1)return Ps({inputs:{x:s[0]},backend:n});if(s.length>q().get("WEBGL_MAX_TEXTURES_IN_SHADER")){let l=Math.floor(s.length/2),u=im({inputs:s.slice(0,l),backend:n}),c=im({inputs:s.slice(l),backend:n});return im({inputs:[u,c],backend:n})}let r=s.map(l=>l.dtype).reduce((l,u)=>Mn(l,u)),a=s.map(l=>l.shape),i=q().getBool("WEBGL_PACK")?new rse(s[0].shape,a):new sse(s[0].shape,a);return n.runWebGLProgram(i,s,r)}var ase={kernelName:po,backendName:"webgl",kernelFunc:im};function ose(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s,i=r.shape.length,l=v.parseAxisParam(a,r.shape),u=l,c=T.getAxesPermutation(u,i),p=r;c!=null&&(p=ts({inputs:{x:r},backend:n,attrs:{perm:c}}),u=T.getInnerMostAxes(u.length,i)),T.assertAxesAreInnerMostDims("all",u,i);let[d,h]=T.computeOutAndReduceShapes(p.shape,u),f=v.sizeFromShape(h),m=be({inputs:{x:p},backend:n,attrs:{shape:[-1,f]}}),g=au(m,m.dtype,"all",n),y;if(o){let x=T.expandShapeToKeepDim(d,l);y=be({inputs:{x:g},backend:n,attrs:{shape:x}})}else y=be({inputs:{x:g},backend:n,attrs:{shape:d}});return n.disposeIntermediateTensorInfo(m),n.disposeIntermediateTensorInfo(g),c!=null&&n.disposeIntermediateTensorInfo(p),y}var ise={kernelName:fc,backendName:"webgl",kernelFunc:ose};function lse(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s,i=r.shape.length,l=v.parseAxisParam(a,r.shape),u=l,c=T.getAxesPermutation(u,i),p=r;c!=null&&(p=ts({inputs:{x:r},backend:n,attrs:{perm:c}}),u=T.getInnerMostAxes(u.length,i)),T.assertAxesAreInnerMostDims("any",u,i);let[d,h]=T.computeOutAndReduceShapes(p.shape,u),f=v.sizeFromShape(h),m=be({inputs:{x:p},backend:n,attrs:{shape:[-1,f]}}),g=au(m,m.dtype,"any",n),y;if(o){let x=T.expandShapeToKeepDim(d,l);y=be({inputs:{x:g},backend:n,attrs:{shape:x}})}else y=be({inputs:{x:g},backend:n,attrs:{shape:d}});return n.disposeIntermediateTensorInfo(m),n.disposeIntermediateTensorInfo(g),c!=null&&n.disposeIntermediateTensorInfo(p),y}var use={kernelName:mc,backendName:"webgl",kernelFunc:lse},cse=class{constructor(e,t,n){this.variableNames=["A"];let{windowSize:s,batchSize:r,outSize:a}=e;n||this.variableNames.push("bestIndicesA"),this.outputShape=[r,a];let o=t==="max"?">":"<",i=n?"inOffset + i;":"round(getBestIndicesA(batch, inOffset + i));";this.userCode=`
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int outIdx = coords[1];
|
|
int inOffset = outIdx * ${s};
|
|
|
|
int bestIndex = inOffset;
|
|
float bestValue = getA(batch, bestIndex);
|
|
|
|
for (int i = 0; i < ${s}; i++) {
|
|
int inIdx = ${i};
|
|
float candidate = getA(batch, inIdx);
|
|
if (candidate ${o} bestValue) {
|
|
bestValue = candidate;
|
|
bestIndex = inIdx;
|
|
}
|
|
}
|
|
setOutput(float(bestIndex));
|
|
}
|
|
`}},dse=class{constructor(e,t,n,s){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,v.assert(e.length>2,()=>`Packed arg${n.charAt(0).toUpperCase()+n.slice(1)} supports only inputs with rank above 2.`);let r=e[e.length-1],a=Math.ceil(r/t);this.outputShape=e.slice(0,-1),a>1&&this.outputShape.push(a),s||this.variableNames.push("bestIndicesA");let o=this.outputShape,i=o.length,l=wt(i),u=es("coords",i),c,p;if(a===1){p=i+1;let S=wt(p);c=`
|
|
${S} sourceLocR = ${S}(${u.join()}, 0);
|
|
++${u[i-1]};
|
|
${S} sourceLocG = ${S}(${u.join()}, 0);
|
|
++${u[i-2]};
|
|
${S} sourceLocA = ${S}(${u.join()}, 0);
|
|
--${u[i-1]};
|
|
${S} sourceLocB = ${S}(${u.join()}, 0);
|
|
--${u[i-2]};`}else p=i,c=`
|
|
${l} sourceLocR = coords;
|
|
++${u[i-1]};
|
|
${l} sourceLocG = coords;
|
|
++${u[i-2]};
|
|
${l} sourceLocA = coords;
|
|
--${u[i-1]};
|
|
${l} sourceLocB = coords;
|
|
--${u[i-2]};`;let d=["x","y","z","w","u","v"].slice(0,p),h="."+d[p-1],f=d.map(S=>"int "+S),m=es("sourceLocR",p-1).concat("inIdx.r"),g=es("sourceLocG",p-1).concat("inIdx.g"),y=es("sourceLocB",p-1).concat("inIdx.b"),x=es("sourceLocA",p-1).concat("inIdx.a"),A=n==="max"?"greaterThan":"lessThan",b=s?"":`
|
|
inIdx = round(vec4(getBestIndicesAChannel(${m.join()}),
|
|
getBestIndicesAChannel(${g.join()}),
|
|
getBestIndicesAChannel(${y.join()}),
|
|
getBestIndicesAChannel(${x.join()})));`,w=`vec4(
|
|
getAChannel(${m.join()}),
|
|
hasNextCol ? getAChannel(${g.join()}) : 0.,
|
|
hasNextRow ? getAChannel(${y.join()}) : 0.,
|
|
hasNextRow && hasNextCol ? getAChannel(${x.join()}) : 0.)`,k=s?"":`
|
|
float getBestIndicesAChannel(${f.join()}) {
|
|
return getChannel(getBestIndicesA(${d.join()}),
|
|
vec2(${d.slice(-2).join()}));
|
|
}`;this.userCode=`
|
|
float getAChannel(${f.join()}) {
|
|
return getChannel(getA(${d.join()}),
|
|
vec2(${d.slice(-2).join()}));
|
|
}
|
|
${k}
|
|
void main() {
|
|
${l} coords = getOutputCoords();
|
|
bool hasNextCol = ${u[i-1]} < ${o[i-1]-1};
|
|
bool hasNextRow = ${u[i-2]} < ${o[i-2]-1};
|
|
${c}
|
|
ivec4 srcIdx = ivec4(sourceLocR${h}, sourceLocG${h},
|
|
sourceLocB${h}, sourceLocA${h}) * ${t};
|
|
ivec4 inIdx = srcIdx;
|
|
vec4 bestIndex = vec4(inIdx);
|
|
vec4 bestValue = ${w};
|
|
|
|
for (int i = 0; i < ${t}; i++) {
|
|
inIdx = srcIdx;
|
|
${b}
|
|
vec4 candidate = ${w};
|
|
bvec4 nan = isnan(candidate);
|
|
bvec4 replace = bvec4(
|
|
vec4(${A}(candidate, bestValue)) * (vec4(1.0) - vec4(nan)));
|
|
|
|
bestValue = vec4(replace.x ? candidate.x : bestValue.x,
|
|
replace.y ? candidate.y : bestValue.y,
|
|
replace.z ? candidate.z : bestValue.z,
|
|
replace.w ? candidate.w : bestValue.w);
|
|
bestIndex = mix(bestIndex, vec4(inIdx), vec4(replace));
|
|
srcIdx++;
|
|
}
|
|
setOutput(bestIndex);
|
|
}
|
|
`}};function A9(e,t,n,s=null){let r=t.shape[0],a=t.shape[1];s!=null&&(r=s.shape[0],a=s.shape[1]);let o=T.computeOptimalWindowSize(a),i={windowSize:o,inSize:a,batchSize:r,outSize:Math.ceil(a/o)},l=new cse(i,n,s==null),u=[t];s!=null&&u.push(s);let c=e.runWebGLProgram(l,u,"int32");if(c.shape[1]===1)return c;let p=A9(e,t,n,c);return e.disposeIntermediateTensorInfo(c),p}function x9(e,t,n,s=null){let r=s!=null?s.shape:t.shape,a=r[r.length-1],o=T.computeOptimalWindowSize(a),i=new dse(r,o,n,s==null),l=s==null?[t]:[t,s],u=e.runWebGLProgram(i,l,"int32");if(u.shape.length===t.shape.length){let c=x9(e,t,n,u);return e.disposeIntermediateTensorInfo(u),c}return u}function b9(e,t,n,s){let r=[n];if(T.assertAxesAreInnerMostDims("arg"+s.charAt(0).toUpperCase()+s.slice(1),r,t.shape.length),!q().getBool("WEBGL_PACK_REDUCE")||t.shape.length<=2){let a=[],o=e.texData.get(t.dataId),i=o!==null&&o.isPacked,l=t;i&&(l=e.unpackTensor(t),a.push(l));let[u,c]=T.computeOutAndReduceShapes(l.shape,r),p=v.sizeFromShape(c),d=be({inputs:{x:l},backend:e,attrs:{shape:[-1,p]}});a.push(d);let h=A9(e,d,s);a.push(h);let f=be({inputs:{x:h},backend:e,attrs:{shape:u}});return a.forEach(m=>e.disposeIntermediateTensorInfo(m)),f}return x9(e,t,s)}function pse(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a}=s,o=v.parseAxisParam(a,r.shape),i=T.getAxesPermutation(o,r.shape.length),l=r,u=[];i!=null&&(l=ts({inputs:{x:r},backend:n,attrs:{perm:i}}),u.push(l),o=T.getInnerMostAxes(o.length,l.shape.length)),T.assertAxesAreInnerMostDims("argMax",[o[0]],l.shape.length);let c=b9(n,l,o[0],"max");return u.forEach(p=>n.disposeIntermediateTensorInfo(p)),c}var hse={kernelName:ho,backendName:"webgl",kernelFunc:pse};function fse(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a}=s,o=v.parseAxisParam(a,r.shape),i=T.getAxesPermutation(o,r.shape.length),l=r,u=[];i!=null&&(l=ts({inputs:{x:r},backend:n,attrs:{perm:i}}),u.push(l),o=T.getInnerMostAxes(o.length,l.shape.length)),T.assertAxesAreInnerMostDims("argMin",[o[0]],l.shape.length);let c=b9(n,l,o[0],"min");return u.forEach(p=>n.disposeIntermediateTensorInfo(p)),c}var mse={kernelName:gc,backendName:"webgl",kernelFunc:fse},gse=fr+`
|
|
if (abs(x) > 1.) {
|
|
return NAN;
|
|
}
|
|
return asin(x);
|
|
`,yse=dt({opSnippet:gse}),Ase={kernelName:yc,backendName:"webgl",kernelFunc:yse},xse=fr+"return log(x + sqrt(x * x + 1.0));",bse=dt({opSnippet:xse}),vse={kernelName:Ac,backendName:"webgl",kernelFunc:bse},wse=fr+`
|
|
return atan(x);
|
|
`,kse=dt({opSnippet:wse}),Ise={kernelName:xc,backendName:"webgl",kernelFunc:kse},Sse=Dne+`
|
|
return atan(a, b);
|
|
`,Cse=`
|
|
vec4 result = atan(a, b);
|
|
vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
|
|
`+$ne+`
|
|
return result;
|
|
`,Tse=_n({opSnippet:Sse,packedOpSnippet:Cse}),Nse={kernelName:vc,backendName:"webgl",kernelFunc:Tse},Ese=fr+`
|
|
if ((x < -1.0) || (x > 1.0)) return NAN;
|
|
return (log(1.0 + x) - log(1.0 - x)) / 2.0;`,Rse=dt({opSnippet:Ese}),_se={kernelName:bc,backendName:"webgl",kernelFunc:Rse},$p=class{constructor(e,t,n,s=!1,r=!1){if(this.variableNames=["x"],t==="avg"&&n)throw new Error("Cannot compute positions for average pool.");let a=e.filterWidth,o=e.strideHeight,i=e.strideWidth,l=e.dilationHeight,u=e.dilationWidth,c=e.effectiveFilterHeight,p=e.effectiveFilterWidth,d=e.padInfo.top,h=e.padInfo.left;this.outputShape=e.outShape;let f=t==="avg",m=`((batch * ${e.inHeight} + xR) * ${e.inWidth} + xC) * ${e.inChannels} + d`,g=`(xR * ${e.inWidth} + xC) * ${e.inChannels} + d`,y="0.0";if(f||(y="-1.0 / 1e-20"),n){let S=">=";this.userCode=`
|
|
const ivec2 strides = ivec2(${o}, ${i});
|
|
const ivec2 pads = ivec2(${d}, ${h});
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int d = coords[3];
|
|
|
|
ivec2 xRCCorner = coords.yz * strides - pads;
|
|
int xRCorner = xRCCorner.x;
|
|
int xCCorner = xRCCorner.y;
|
|
|
|
// max/min x(?, ?, d) to get y(yR, yC, d).
|
|
// ? = to be determined
|
|
float minMaxValue = 0.0;
|
|
float minMaxValueFound = 0.0;
|
|
int minMaxPosition = 0;
|
|
float avgValue = 0.0;
|
|
|
|
for (int wR = 0; wR < ${c};
|
|
wR += ${l}) {
|
|
int xR = xRCorner + wR;
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wC = 0; wC < ${p};
|
|
wC += ${u}) {
|
|
int xC = xCCorner + wC;
|
|
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
continue;
|
|
}
|
|
|
|
float value = getX(batch, xR, xC, d);
|
|
|
|
// If a min / max value has already been found, use it. If not,
|
|
// use the current value.
|
|
float currMinMaxValue = mix(
|
|
value, minMaxValue, minMaxValueFound);
|
|
if (value ${S} currMinMaxValue) {
|
|
minMaxValue = value;
|
|
minMaxValueFound = 1.0;
|
|
minMaxPosition = ${s?r?m:g:`wR * ${p} + wC`};
|
|
}
|
|
}
|
|
}
|
|
setOutput(float(minMaxPosition));
|
|
}
|
|
`;return}let x="max",A=`${t}(${t}(${t}(minMaxValue[0], minMaxValue[1]), minMaxValue[2]), minMaxValue[3])`;t==="avg"&&(A="avgValue / count");let b=Math.floor(a/4)*4,w=a%4,k=`
|
|
if (${f}) {
|
|
avgValue += dot(values, ones);
|
|
} else {
|
|
minMaxValue = ${x}(values, minMaxValue);
|
|
}
|
|
`;this.userCode=`
|
|
const ivec2 strides = ivec2(${o}, ${i});
|
|
const ivec2 pads = ivec2(${d}, ${h});
|
|
const float initializationValue = ${y};
|
|
const vec4 ones = vec4(1.0, 1.0, 1.0, 1.0);
|
|
|
|
float count = 0.0;
|
|
|
|
float getValue(int batch, int xR, int xC, int d) {
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
return initializationValue;
|
|
}
|
|
count += 1.0;
|
|
return getX(batch, xR, xC, d);
|
|
}
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int d = coords[3];
|
|
|
|
ivec2 xRCCorner = coords.yz * strides - pads;
|
|
int xRCorner = xRCCorner.x;
|
|
int xCCorner = xRCCorner.y;
|
|
|
|
// max/min x(?, ?, d) to get y(yR, yC, d).
|
|
// ? = to be determined
|
|
vec4 minMaxValue = vec4(${y});
|
|
float avgValue = 0.0;
|
|
count = 0.0;
|
|
|
|
for (int wR = 0; wR < ${c};
|
|
wR += ${l}) {
|
|
int xR = xRCorner + wR;
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wC = 0; wC < ${b}; wC += 4) {
|
|
int xC = xCCorner + wC * ${u};
|
|
|
|
vec4 values = vec4(
|
|
getValue(batch, xR, xC, d),
|
|
getValue(batch, xR, xC + ${u}, d),
|
|
getValue(batch, xR, xC + 2 * ${u}, d),
|
|
getValue(batch, xR, xC + 3 * ${u}, d)
|
|
);
|
|
|
|
${k}
|
|
}
|
|
|
|
int xC = xCCorner + ${b};
|
|
if (${w===1}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, xR, xC, d),
|
|
initializationValue,
|
|
initializationValue,
|
|
initializationValue
|
|
);
|
|
|
|
${k}
|
|
} else if (${w===2}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, xR, xC, d),
|
|
getValue(batch, xR, xC + ${u}, d),
|
|
initializationValue,
|
|
initializationValue
|
|
);
|
|
|
|
${k}
|
|
} else if (${w===3}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, xR, xC, d),
|
|
getValue(batch, xR, xC + ${u}, d),
|
|
getValue(batch, xR, xC + 2 * ${u}, d),
|
|
initializationValue
|
|
);
|
|
|
|
${k}
|
|
}
|
|
}
|
|
setOutput(${A});
|
|
}
|
|
`}},Qx=class{constructor(e,t,n,s=!1,r=!1){if(this.variableNames=["x"],t==="avg"&&n)throw new Error("Cannot compute positions for average pool.");let a=e.filterWidth,o=e.strideDepth,i=e.strideHeight,l=e.strideWidth,u=e.dilationDepth,c=e.dilationHeight,p=e.dilationWidth,d=e.effectiveFilterDepth,h=e.effectiveFilterHeight,f=e.effectiveFilterWidth,m=e.padInfo.front,g=e.padInfo.top,y=e.padInfo.left;this.outputShape=e.outShape;let x=t==="avg",A="0.0";if(x||(A="-1.0 / 1e-20"),n){let R=">=";this.userCode=`
|
|
const ivec3 strides =
|
|
ivec3(${o}, ${i}, ${l});
|
|
const ivec3 pads = ivec3(${m}, ${g}, ${y});
|
|
|
|
void main() {
|
|
ivec5 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
int ch = coords.u;
|
|
|
|
ivec3 xCorner = ivec3(coords.y, coords.z, coords.w) * strides - pads;
|
|
int xDCorner = xCorner.x;
|
|
int xRCorner = xCorner.y;
|
|
int xCCorner = xCorner.z;
|
|
|
|
// max/min x(?, ?, ?, ch) to get y(yD, yR, yC, ch).
|
|
// ? = to be determined
|
|
float minMaxValue = 0.0;
|
|
float minMaxValueFound = 0.0;
|
|
int minMaxPosition = 0;
|
|
|
|
for (int wD = 0; wD < ${d};
|
|
wD += ${u}) {
|
|
int xD = xDCorner + wD;
|
|
|
|
if (xD < 0 || xD >= ${e.inDepth}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wR = 0; wR < ${h};
|
|
wR += ${c}) {
|
|
int xR = xRCorner + wR;
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wC = 0; wC < ${f};
|
|
wC += ${p}) {
|
|
int xC = xCCorner + wC;
|
|
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
continue;
|
|
}
|
|
|
|
float value = getX(batch, xD, xR, xC, ch);
|
|
|
|
// If a min / max value has already been found, use it. If not,
|
|
// use the current value.
|
|
float currMinMaxValue = mix(
|
|
value, minMaxValue, minMaxValueFound);
|
|
if (value ${R} currMinMaxValue) {
|
|
minMaxValue = value;
|
|
minMaxValueFound = 1.0;
|
|
minMaxPosition = ${s?r?`(((batch * ${e.inDepth} + xD) * ${e.inHeight} + xR) * ${e.inWidth} + xC) * ${e.inChannels} + ch`:`((xD * ${e.inHeight} + xR) * ${e.inWidth} + xC) * ${e.inChannels} + ch`:`wD * ${h} * ${f} +
|
|
wR * ${f} + wC`};
|
|
}
|
|
}
|
|
}
|
|
}
|
|
setOutput(float(minMaxPosition));
|
|
}
|
|
`;return}let b="max",w=`${t}(${t}(${t}(minMaxValue[0], minMaxValue[1]), minMaxValue[2]), minMaxValue[3])`;t==="avg"&&(w="avgValue / count");let k=Math.floor(a/4)*4,S=a%4,E=`
|
|
if (${x}) {
|
|
avgValue += dot(values, ones);
|
|
} else {
|
|
minMaxValue = ${b}(values, minMaxValue);
|
|
}
|
|
`;this.userCode=`
|
|
const ivec3 strides =
|
|
ivec3(${o}, ${i}, ${l});
|
|
const ivec3 pads = ivec3(${m}, ${g}, ${y});
|
|
const float initializationValue = ${A};
|
|
const vec4 ones = vec4(1.0, 1.0, 1.0, 1.0);
|
|
|
|
float count = 0.0;
|
|
|
|
float getValue(int batch, int xD, int xR, int xC, int ch) {
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
return initializationValue;
|
|
}
|
|
count += 1.0;
|
|
return getX(batch, xD, xR, xC, ch);
|
|
}
|
|
|
|
void main() {
|
|
ivec5 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
int ch = coords.u;
|
|
|
|
ivec3 xCorner = ivec3(coords.y, coords.z, coords.w) * strides - pads;
|
|
int xDCorner = xCorner.x;
|
|
int xRCorner = xCorner.y;
|
|
int xCCorner = xCorner.z;
|
|
|
|
// max/min x(?, ?, ?, d) to get y(yD, yR, yC, ch).
|
|
// ? = to be determined
|
|
vec4 minMaxValue = vec4(${A});
|
|
float avgValue = 0.0;
|
|
count = 0.0;
|
|
|
|
for (int wD = 0; wD < ${d};
|
|
wD += ${u}) {
|
|
int xD = xDCorner + wD;
|
|
|
|
if (xD < 0 || xD >= ${e.inDepth}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wR = 0; wR < ${h};
|
|
wR += ${c}) {
|
|
int xR = xRCorner + wR;
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wC = 0; wC < ${k}; wC += 4) {
|
|
int xC = xCCorner + wC * ${p};
|
|
|
|
vec4 values = vec4(
|
|
getValue(batch, xD, xR, xC, ch),
|
|
getValue(batch, xD, xR, xC + ${p}, ch),
|
|
getValue(batch, xD, xR, xC + 2 * ${p}, ch),
|
|
getValue(batch, xD, xR, xC + 3 * ${p}, ch)
|
|
);
|
|
|
|
${E}
|
|
}
|
|
|
|
int xC = xCCorner + ${k};
|
|
if (${S===1}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, xD, xR, xC, ch),
|
|
initializationValue,
|
|
initializationValue,
|
|
initializationValue
|
|
);
|
|
|
|
${E}
|
|
} else if (${S===2}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, xD, xR, xC, ch),
|
|
getValue(batch, xD, xR, xC + ${p}, ch),
|
|
initializationValue,
|
|
initializationValue
|
|
);
|
|
|
|
${E}
|
|
} else if (${S===3}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, xD, xR, xC, ch),
|
|
getValue(batch, xD, xR, xC + ${p}, ch),
|
|
getValue(batch, xD, xR, xC + 2 * ${p}, ch),
|
|
initializationValue
|
|
);
|
|
|
|
${E}
|
|
}
|
|
}
|
|
setOutput(${w});
|
|
}
|
|
}
|
|
`}};function Dse(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t;nd(r,"avgPool");let{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=s,u=1;v.assert(T.eitherStridesOrDilationsAreOne(o,u),()=>`Error in avgPool: Either strides or dilations must be 1. Got strides ${o} and dilations '${u}'`);let c=T.computePool2DInfo(r.shape,a,o,u,i,l);if(c.filterWidth===1&&c.filterHeight===1&&v.arraysEqual(c.inShape,c.outShape))return Ps({inputs:{x:r},backend:n});let p=new $p(c,"avg",!1);return n.runWebGLProgram(p,[r],"float32")}var $se={kernelName:fo,backendName:"webgl",kernelFunc:Dse};function Pse(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l,dataFormat:u}=s,c=[1,1,1],p=T.computePool3DInfo(r.shape,a,o,c,i,l,u),d=new Qx(p,"avg",!1);return n.runWebGLProgram(d,[r],"float32")}var Fse={kernelName:zp,backendName:"webgl",kernelFunc:Pse},Ose=class{constructor(e){this.variableNames=["dy"],this.outputShape=e.inShape;let t=e.filterHeight,n=e.filterWidth,s=e.strideHeight,r=e.strideWidth,a=e.dilationHeight,o=e.dilationWidth,i=e.effectiveFilterHeight,l=e.effectiveFilterWidth,u=i-1-e.padInfo.top,c=l-1-e.padInfo.left,p=1/(t*n);this.userCode=`
|
|
const ivec2 pads = ivec2(${u}, ${c});
|
|
const float avgMultiplier = float(${p});
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int d = coords[3];
|
|
|
|
ivec2 dyRCCorner = coords.yz - pads;
|
|
int dyRCorner = dyRCCorner.x;
|
|
int dyCCorner = dyRCCorner.y;
|
|
|
|
// Convolve dy(?, ?, d) with pos mask(:, :, d) to get dx(xR, xC, d).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
float dotProd = 0.0;
|
|
for (int wR = 0; wR < ${i};
|
|
wR += ${a}) {
|
|
float dyR = float(dyRCorner + wR) / ${s}.0;
|
|
|
|
if (dyR < 0.0 || dyR >= ${e.outHeight}.0 || fract(dyR) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyR = int(dyR);
|
|
|
|
for (int wC = 0; wC < ${l};
|
|
wC+= ${o}) {
|
|
float dyC = float(dyCCorner + wC) / ${r}.0;
|
|
|
|
if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
|
|
fract(dyC) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyC = int(dyC);
|
|
|
|
float dyValue = getDy(b, idyR, idyC, d);
|
|
|
|
dotProd += dyValue * avgMultiplier;
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}},Mse=class{constructor(e){this.variableNames=["dy"],this.outputShape=e.inShape;let t=e.filterDepth,n=e.filterHeight,s=e.filterWidth,r=e.strideDepth,a=e.strideHeight,o=e.strideWidth,i=e.dilationDepth,l=e.dilationHeight,u=e.dilationWidth,c=e.effectiveFilterDepth,p=e.effectiveFilterHeight,d=e.effectiveFilterWidth,h=c-1-e.padInfo.front,f=p-1-e.padInfo.top,m=d-1-e.padInfo.left,g=1/(t*n*s);this.userCode=`
|
|
const ivec3 pads = ivec3(${h}, ${f}, ${m});
|
|
const float avgMultiplier = float(${g});
|
|
|
|
void main() {
|
|
ivec5 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
int ch = coords.u;
|
|
|
|
ivec3 dyCorner = ivec3(coords.y, coords.z, coords.w) - pads;
|
|
int dyDCorner = dyCorner.x;
|
|
int dyRCorner = dyCorner.y;
|
|
int dyCCorner = dyCorner.z;
|
|
|
|
// Convolve dy(?, ?, ?, d) with pos mask(:, :, :, ch) to get
|
|
// dx(xD, xR, xC, ch).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
float dotProd = 0.0;
|
|
|
|
for (int wD = 0; wD < ${c};
|
|
wD += ${i}) {
|
|
float dyD = float(dyDCorner + wD) / ${r}.0;
|
|
|
|
if (dyD < 0.0 || dyD >= ${e.outDepth}.0 || fract(dyD) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyD = int(dyD);
|
|
|
|
for (int wR = 0; wR < ${p};
|
|
wR += ${l}) {
|
|
float dyR = float(dyRCorner + wR) / ${a}.0;
|
|
|
|
if (dyR < 0.0 || dyR >= ${e.outHeight}.0 ||
|
|
fract(dyR) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyR = int(dyR);
|
|
|
|
for (int wC = 0; wC < ${d};
|
|
wC += ${u}) {
|
|
float dyC = float(dyCCorner + wC) / ${o}.0;
|
|
|
|
if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
|
|
fract(dyC) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyC = int(dyC);
|
|
|
|
float dyValue = getDy(batch, idyD, idyR, idyC, ch);
|
|
|
|
dotProd += dyValue * avgMultiplier;
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}};function zse(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,input:a}=t,o=a,{filterSize:i,strides:l,pad:u,dimRoundingMode:c}=s,p=[1,1,1],d=T.computePool3DInfo(o.shape,i,l,p,u,c),h=new Mse(d);return n.runWebGLProgram(h,[r],o.dtype)}var Lse={kernelName:Hm,backendName:"webgl",kernelFunc:zse};function Bse(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,input:a}=t,o=a;nd([r,a],"avgPoolGrad");let{filterSize:i,strides:l,pad:u}=s,c=T.computePool2DInfo(o.shape,i,l,1,u),p=new Ose(c);return n.runWebGLProgram(p,[r],o.dtype)}var Wse={kernelName:Gm,backendName:"webgl",kernelFunc:Bse};function Vse(e){let{inputs:t,backend:n,attrs:s}=e,{a:r,b:a}=t,{transposeA:o,transposeB:i}=s;return Fm({a:r,b:a,transposeA:o,transposeB:i,backend:n})}var Use={kernelName:mo,backendName:"webgl",kernelFunc:Vse},Gse=class{constructor(e,t,n,s,r,a){this.outputShape=[],this.variableNames=["x","mean","variance"],T.assertAndGetBroadcastShape(e,t),T.assertAndGetBroadcastShape(e,n);let o="0.0";s!=null&&(T.assertAndGetBroadcastShape(e,s),this.variableNames.push("offset"),o="getOffsetAtOutCoords()");let i="1.0";r!=null&&(T.assertAndGetBroadcastShape(e,r),this.variableNames.push("scale"),i="getScaleAtOutCoords()"),this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
float x = getXAtOutCoords();
|
|
float mean = getMeanAtOutCoords();
|
|
float variance = getVarianceAtOutCoords();
|
|
float offset = ${o};
|
|
float scale = ${i};
|
|
float inv = scale * inversesqrt(variance + float(${a}));
|
|
setOutput(dot(vec3(x, -mean, offset), vec3(inv, inv, 1)));
|
|
}
|
|
`}},Hse=class{constructor(e,t,n,s,r,a){this.packedInputs=!0,this.packedOutput=!0,this.variableNames=["x","mean","variance"],T.assertAndGetBroadcastShape(e,t),T.assertAndGetBroadcastShape(e,n);let o="vec4(0.0)";s!=null&&(T.assertAndGetBroadcastShape(e,s),this.variableNames.push("offset"),o="getOffsetAtOutCoords()");let i="vec4(1.0)";r!=null&&(T.assertAndGetBroadcastShape(e,r),this.variableNames.push("scale"),i="getScaleAtOutCoords()"),this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
vec4 offset = ${o};
|
|
vec4 scale = ${i};
|
|
|
|
vec4 x = getXAtOutCoords();
|
|
vec4 mean = getMeanAtOutCoords();
|
|
vec4 variance = getVarianceAtOutCoords();
|
|
|
|
vec4 inv = scale * inversesqrt(variance + vec4(${a}));
|
|
|
|
setOutput((x - mean) * inv + offset);
|
|
}
|
|
`}},jse=({inputs:e,backend:t,attrs:n})=>{let{x:s,mean:r,variance:a,offset:o,scale:i}=e;v.assert(r.shape.length===a.shape.length,()=>"Batch normalization gradient requires mean and variance to have equal ranks."),v.assert(o==null||r.shape.length===o.shape.length,()=>"Batch normalization gradient requires mean and offset to have equal ranks."),v.assert(i==null||r.shape.length===i.shape.length,()=>"Batch normalization gradient requires mean and scale to have equal ranks.");let{varianceEpsilon:l}=n;l==null&&(l=.001);let u=[s,r,a],c=null;o!=null&&(c=o.shape,u.push(o));let p=null;i!=null&&(p=i.shape,u.push(i));let d=q().getBool("WEBGL_PACK_NORMALIZATION")?new Hse(s.shape,r.shape,a.shape,c,p,l):new Gse(s.shape,r.shape,a.shape,c,p,l);return t.runWebGLProgram(d,u,u[0].dtype)},qse={kernelName:Eo,backendName:"webgl",kernelFunc:jse},Xse=class{constructor(e){this.variableNames=["source"],this.outputShape=e,this.rank=e.length;let t=wt(this.rank);this.customUniforms=[{name:"start",arrayIndex:this.rank,type:"int"}];let n=Kse(this.rank),s,r=e.map((a,o)=>`sourceLoc.${oy[o]} = start[${o}] + coords.${oy[o]};`);s=`
|
|
${t} sourceLoc;
|
|
${t} coords = getOutputCoords();
|
|
${r.join(`
|
|
`)}
|
|
`,this.userCode=`
|
|
void main() {
|
|
${s}
|
|
setOutput(getSource(${n}));
|
|
}
|
|
`}},oy=["x","y","z","w","u","v"];function Kse(e){if(e===1)return"sourceLoc";if(e<=6)return oy.slice(0,e).map(t=>"sourceLoc."+t).join(",");throw Error(`Slicing for rank ${e} is not yet supported`)}var Zse=class{constructor(e){this.variableNames=["source"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e,this.rank=e.length,this.customUniforms=[{name:"start",arrayIndex:this.rank,type:"int"}];let t=wt(this.rank),n=es("coords",this.rank),s=es("sourceLoc",this.rank),r=this.rank===1?"sourceLoc":`vec2(${s.slice(-2).join()})`,a=`getChannel(getSource(${s.join()}), ${r})`,o=`
|
|
result.x = ${a};
|
|
if (++${n[this.rank-1]} < ${e[this.rank-1]}) {
|
|
++${s[this.rank-1]};
|
|
result.y = ${a};
|
|
--${s[this.rank-1]};
|
|
}
|
|
`,i=this.rank===1?"":`
|
|
--${n[this.rank-1]};
|
|
if (++${n[this.rank-2]} < ${e[this.rank-2]}) {
|
|
++${s[this.rank-2]};
|
|
result.z = ${a};
|
|
if (++${n[this.rank-1]} < ${e[this.rank-1]}) {
|
|
++${s[this.rank-1]};
|
|
result.w = ${a};
|
|
}
|
|
}
|
|
`,l=this.rank<=4?`sourceLoc = coords +
|
|
${t}(${e.map((u,c)=>`start[${c}]`).join()});`:e.map((u,c)=>`${s[c]} = ${n[c]} + start[${c}];`).join(`
|
|
`);this.userCode=`
|
|
void main() {
|
|
${t} coords = getOutputCoords();
|
|
${t} sourceLoc;
|
|
${l}
|
|
vec4 result = vec4(0.);
|
|
${o}
|
|
${i}
|
|
setOutput(result);
|
|
}
|
|
`}};function Yse(e,t,n,s){let r=s.texData.get(e.dataId),a=s.makeTensorInfo(n,e.dtype),o=s.texData.get(a.dataId);Object.assign(o,r),o.refCount=1,o.shape=n,o.dtype=e.dtype;let i=Ut.computeFlatOffset(t,v.computeStrides(e.shape));r.slice&&(i+=r.slice.flatOffset),o.slice={flatOffset:i,origDataId:r.slice&&r.slice.origDataId||e.dataId};let l=s.dataRefCount.get(o.slice.origDataId)||1;return s.dataRefCount.set(o.slice.origDataId,l+1),a}function ud(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{begin:a,size:o}=s,[i,l]=Ut.parseSliceParams(r,a,o);if(Ut.assertParamsValid(r,i,l),v.sizeFromShape(l)===0)return n.makeTensorInfo(l,r.dtype,[]);if(n.shouldExecuteOnCPU([r])||r.dtype==="string"){let p=n.texData.get(r.dataId),d=Bte(p.values,i,l,r.shape,r.dtype);return n.makeTensorInfo(l,r.dtype,d)}let{isPacked:u}=n.texData.get(r.dataId),c=Ut.isSliceContinous(r.shape,i,l);if(u||!c){let p=q().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new Zse(l):new Xse(l),d=[i];return n.runWebGLProgram(p,[r],r.dtype,d)}return n.uploadToGPU(r.dataId),Yse(r,i,l,n)}var Jse={kernelName:Ol,backendName:"webgl",kernelFunc:ud},Qse=e=>{let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockShape:a,crops:o}=s;v.assert(r.shape.length<=4,()=>"batchToSpaceND for rank > 4 with a WebGL backend not implemented yet");let i=a.reduce((x,A)=>x*A),l=T.getReshaped(r.shape,a,i),u=T.getPermuted(l.length,a.length),c=T.getReshapedPermuted(r.shape,a,i),p=T.getSliceBeginCoords(o,a.length),d=T.getSliceSize(c,o,a.length),h=[],f=be({inputs:{x:r},backend:n,attrs:{shape:l}}),m=ts({inputs:{x:f},backend:n,attrs:{perm:u}}),g=be({inputs:{x:m},backend:n,attrs:{shape:c}}),y=ud({inputs:{x:g},backend:n,attrs:{begin:p,size:d}});return h.push(f),h.push(m),h.push(g),h.forEach(x=>n.disposeIntermediateTensorInfo(x)),y},ere={kernelName:ll,backendName:"webgl",kernelFunc:Qse};function tre(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,weights:a}=t,{size:o}=s,i=n.readSync(r.dataId),l=n.readSync(a.dataId),u=a9(i,l,a.dtype,a.shape,o);return n.makeTensorInfo([o],a.dtype,u)}var nre={kernelName:jm,backendName:"webgl",kernelFunc:tre};function sre(e){let{inputs:t,backend:n}=e,{s0:s,s1:r}=t,a=n.readSync(s.dataId),o=n.readSync(r.dataId),i=T.assertAndGetBroadcastShape(Array.from(a),Array.from(o));return n.makeTensorInfo([i.length],"int32",Int32Array.from(i))}var rre={kernelName:qm,backendName:"webgl",kernelFunc:sre},are="return float(a != b);",v9=_n({opSnippet:are,cpuKernelImpl:Pte,dtype:"bool"}),ore={kernelName:Sl,backendName:"webgl",kernelFunc:v9};function zh(e){let{inputs:t,backend:n}=e,{input:s}=t,r=n.texData.get(s.dataId);return Ps({inputs:{x:r.complexTensorInfos.real},backend:n})}var ire={kernelName:qp,backendName:"webgl",kernelFunc:zh},lre="return float(int(x));";function ure(e,t){let n=new pa(e.shape,lre),s=t.runWebGLProgram(n,[e],"int32");return{dataId:s.dataId,shape:s.shape,dtype:s.dtype}}function iy(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{dtype:a}=s;if(a==="complex64"){if(r.dtype==="complex64")return Ps({inputs:{x:r},backend:n});let o=Wt(r.shape),i=iy({inputs:{x:r},backend:n,attrs:{dtype:"float32"}}),l=pi({inputs:{real:i,imag:o},backend:n});return o.dispose(),n.disposeIntermediateTensorInfo(i),l}if(r.dtype==="complex64"){let o=zh({inputs:{input:r},backend:n}),i=iy({inputs:{x:o},backend:n,attrs:{dtype:a}});return n.disposeIntermediateTensorInfo(o),i}if(!v.hasEncodingLoss(r.dtype,a)){let o=Ps({inputs:{x:r},backend:n});return{dataId:o.dataId,shape:o.shape,dtype:a}}if(a==="int32")return ure(r,n);if(a==="bool"){let o=n.makeTensorInfo([],"bool",v.getTypedArrayFromDType("bool",1)),l=v9({inputs:{a:r,b:o},backend:n});return n.disposeIntermediateTensorInfo(o),l}throw new Error(`Error in Cast: failed to cast ${r.dtype} to ${a}`)}var cre={kernelName:go,backendName:"webgl",kernelFunc:iy},S7="return ceil(x);",dre=dt({opSnippet:S7,packedOpSnippet:S7,cpuKernelImpl:mte}),pre={kernelName:yo,backendName:"webgl",kernelFunc:dre},hre=class{constructor(e){this.variableNames=["A"],this.customUniforms=[{name:"minVal",type:"float"},{name:"maxVal",type:"float"}],this.outputShape=e,this.userCode=`
|
|
|
|
void main() {
|
|
float value = getAAtOutCoords();
|
|
if (isnan(value)) {
|
|
setOutput(value);
|
|
return;
|
|
}
|
|
|
|
setOutput(clamp(value, minVal, maxVal));
|
|
}
|
|
`}},fre=class{constructor(e){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.customUniforms=[{name:"minVal",type:"float"},{name:"maxVal",type:"float"}],this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
vec4 value = getAAtOutCoords();
|
|
|
|
if (any(isnan(value))) {
|
|
setOutput(value);
|
|
return;
|
|
}
|
|
|
|
setOutput(clamp(value, vec4(minVal), vec4(maxVal)));
|
|
}
|
|
`}};function mre(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{clipValueMin:a,clipValueMax:o}=s,i;q().getBool("WEBGL_PACK_CLIP")?i=new fre(r.shape):i=new hre(r.shape);let l=[[a],[o]];return n.runWebGLProgram(i,[r],r.dtype,l)}var gre={kernelName:ba,backendName:"webgl",kernelFunc:mre},yre=class{constructor(e){this.variableNames=["real","imag"],this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
float re = abs(getRealAtOutCoords());
|
|
float im = abs(getImagAtOutCoords());
|
|
float mx = max(re, im);
|
|
|
|
// sadly the length function in glsl is not underflow-safe
|
|
// (at least not on Intel GPUs). So the safe solution is
|
|
// to ensure underflow-safety in all cases.
|
|
setOutput(
|
|
mx == 0.0 ? 0.0 : mx * length(vec2(1, min(re, im)/mx))
|
|
);
|
|
}
|
|
`}};function C7(e,t){return{dataId:t.dataId,dtype:t.dtype,shape:e.shape}}function Are(e){let{inputs:t,backend:n}=e,{x:s}=t,r=n.texData.get(s.dataId),a=new yre(s.shape),o=[C7(s,r.complexTensorInfos.real),C7(s,r.complexTensorInfos.imag)];return n.runWebGLProgram(a,o,o[0].dtype)}var xre={kernelName:Bp,backendName:"webgl",kernelFunc:Are},bre=class{constructor(e){this.outputShape=[],this.outputShape=T.computeOutShape(e,1),this.variableNames=e.map((a,o)=>`T${o}`);let t=new Array(e.length-1);t[0]=e[0][1];for(let a=1;a<t.length;a++)t[a]=t[a-1]+e[a][1];let n=[`if (yC < ${t[0]}) setOutput(getT0(yR, yC));`];for(let a=1;a<t.length;a++){let o=t[a-1];n.push(`else if (yC < ${t[a]}) setOutput(getT${a}(yR, yC-${o}));`)}let s=t.length,r=t[t.length-1];n.push(`else setOutput(getT${s}(yR, yC-${r}));`),this.userCode=`
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int yR = coords.x;
|
|
int yC = coords.y;
|
|
|
|
${n.join(`
|
|
`)}
|
|
}
|
|
`}},vre=class{constructor(e,t){this.packedInputs=!0,this.packedOutput=!0,this.outputShape=[],this.outputShape=T.computeOutShape(e,t);let n=this.outputShape,s=n.length,r=wt(s),a=es("coords",s),o=["x","y","z","w","u","v"].slice(0,s);this.variableNames=e.map((f,m)=>`T${m}`);let i=new Array(e.length-1);i[0]=e[0][t];for(let f=1;f<i.length;f++)i[f]=i[f-1]+e[f][t];let l=o[t],u=o.slice(-2),c=o.join(),p=`if (${l} < ${i[0]}) {
|
|
return getChannel(
|
|
getT0(${c}), vec2(${u.join()}));
|
|
}`;for(let f=1;f<i.length;f++){let m=i[f-1];p+=`
|
|
if (${l} < ${i[f]} && ${l} >= ${i[f-1]}) {
|
|
return getChannel(
|
|
getT${f}(${Yf(o,l,m)}),
|
|
vec2(${Yf(u,l,m)}));
|
|
}`}let d=i.length,h=i[i.length-1];p+=`
|
|
return getChannel(
|
|
getT${d}(${Yf(o,l,h)}),
|
|
vec2(${Yf(u,l,h)}));`,this.userCode=`
|
|
float getValue(${o.map(f=>"int "+f)}) {
|
|
${p}
|
|
}
|
|
|
|
void main() {
|
|
${r} coords = getOutputCoords();
|
|
vec4 result = vec4(getValue(${a}), 0., 0., 0.);
|
|
|
|
${a[s-1]} = ${a[s-1]} + 1;
|
|
if (${a[s-1]} < ${n[s-1]}) {
|
|
result.g = getValue(${a});
|
|
}
|
|
|
|
${a[s-2]} = ${a[s-2]} + 1;
|
|
if (${a[s-2]} < ${n[s-2]}) {
|
|
result.a = getValue(${a});
|
|
}
|
|
|
|
${a[s-1]} = ${a[s-1]} - 1;
|
|
if (${a[s-2]} < ${n[s-2]} &&
|
|
${a[s-1]} < ${n[s-1]}) {
|
|
result.b = getValue(${a});
|
|
}
|
|
setOutput(result);
|
|
}
|
|
`}};function Yf(e,t,n){let s=e.indexOf(t);return e.map((a,o)=>o===s?`${a} - ${n}`:a).join()}function S2(e){let{inputs:t,backend:n}=e,{input:s}=t,r=n.texData.get(s.dataId);return Ps({inputs:{x:r.complexTensorInfos.imag},backend:n})}var wre={kernelName:Gp,backendName:"webgl",kernelFunc:S2};function dp(e,t,n){let s=e[0].dtype;if(s==="complex64"){let p=e.map(g=>zh({inputs:{input:g},backend:n})),d=e.map(g=>S2({inputs:{input:g},backend:n})),h=dp(p,t,n),f=dp(d,t,n),m=pi({inputs:{real:h,imag:f},backend:n});return p.forEach(g=>n.disposeIntermediateTensorInfo(g)),d.forEach(g=>n.disposeIntermediateTensorInfo(g)),n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(f),m}let r=n.shouldExecuteOnCPU(e);if(s==="string"&&(r=!0),r){let p=e.map(x=>{let A=v.sizeFromShape(x.shape.slice(t));return be({inputs:{x},backend:n,attrs:{shape:[-1,A]}})}),d=p.map(x=>({vals:n.readSync(x.dataId),shape:x.shape})),h=T.computeOutShape(p.map(x=>x.shape),1),f=p[0].shape[0]===1,m=gte(d,h,s,f),g=T.computeOutShape(e.map(x=>x.shape),t),y=n.makeTensorInfo(g,s,m);return p.forEach(x=>n.disposeIntermediateTensorInfo(x)),y}let a=q().getNumber("WEBGL_MAX_TEXTURES_IN_SHADER");if(e.length>a){let p=[];for(let h=0;h<e.length;h+=a){let f=e.slice(h,h+a);p.push(dp(f,t,n))}let d=dp(p,t,n);for(let h of p)n.disposeIntermediateTensorInfo(h);return d}if(q().getBool("WEBGL_PACK_ARRAY_OPERATIONS")&&e[0].shape.length>1){let p=new vre(e.map(d=>d.shape),t);return n.runWebGLProgram(p,e,s)}let{tensors2D:o,outShape:i}=kre(e,t,n),l=new bre(o.map(p=>p.shape)),u=n.runWebGLProgram(l,o,s);o.forEach(p=>n.disposeIntermediateTensorInfo(p));let c=be({inputs:{x:u},attrs:{shape:i},backend:n});return n.disposeIntermediateTensorInfo(u),c}function kre(e,t,n){let s=T.computeOutShape(e.map(a=>a.shape),t);return{tensors2D:e.map(a=>be({inputs:{x:a},attrs:{shape:[-1,v.sizeFromShape(a.shape.slice(t))]},backend:n})),outShape:s}}function w9(e){let{inputs:t,backend:n,attrs:s}=e,{axis:r}=s,a=v.parseAxisParam(r,t[0].shape)[0],o=T.computeOutShape(t.map(u=>u.shape),a);if(v.sizeFromShape(o)===0)return n.makeTensorInfo(o,t[0].dtype,[]);let i=t.filter(u=>v.sizeFromShape(u.shape)>0);if(i.length===1)return Ps({inputs:{x:i[0]},backend:n});let l=i.map(u=>u.shape);return T.assertParamsConsistent(l,a),dp(i,a,n)}var Ire={kernelName:ul,backendName:"webgl",kernelFunc:w9},k9=class{constructor(e,t=!1,n=null,s=!1,r=!1){this.variableNames=["x","W"],this.outputShape=e.outShape;let a=e.padInfo.top,o=e.padInfo.left,i=e.strideHeight,l=e.strideWidth,u=e.dilationHeight,c=e.dilationWidth,p=e.filterHeight,d=e.filterWidth,h=Math.floor(e.inChannels/4)*4,f=e.inChannels%4,m=e.dataFormat==="channelsLast",g=m?1:2,y=m?2:3,x=m?3:1,A="",b="";n&&(s?A=`float activation(float a) {
|
|
float b = getPreluActivationWeightsAtOutCoords();
|
|
${n}
|
|
}`:r?A=`float activation(float a) {
|
|
float b = getLeakyreluAlphaAtOutCoords();
|
|
${n}
|
|
}`:A=`
|
|
float activation(float x) {
|
|
${n}
|
|
}
|
|
`,b="result = activation(result);");let w=t?"result += getBiasAtOutCoords();":"";t&&this.variableNames.push("bias"),s&&this.variableNames.push("preluActivationWeights"),r&&this.variableNames.push("leakyreluAlpha"),this.userCode=`
|
|
${A}
|
|
|
|
const ivec2 strides = ivec2(${i}, ${l});
|
|
const ivec2 pads = ivec2(${a}, ${o});
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int d2 = coords[${x}];
|
|
|
|
ivec2 xRCCorner =
|
|
ivec2(coords[${g}], coords[${y}]) * strides - pads;
|
|
int xRCorner = xRCCorner.x;
|
|
int xCCorner = xRCCorner.y;
|
|
|
|
// Convolve x(?, ?, d1) with w(:, :, d1, d2) to get y(yR, yC, d2).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
float dotProd = 0.0;
|
|
for (int wR = 0; wR < ${p}; wR++) {
|
|
int xR = xRCorner + wR * ${u};
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wC = 0; wC < ${d}; wC++) {
|
|
int xC = xCCorner + wC * ${c};
|
|
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
continue;
|
|
}
|
|
|
|
for (int d1 = 0; d1 < ${h}; d1 += 4) {
|
|
vec4 wValues = vec4(
|
|
getW(wR, wC, d1, d2),
|
|
getW(wR, wC, d1 + 1, d2),
|
|
getW(wR, wC, d1 + 2, d2),
|
|
getW(wR, wC, d1 + 3, d2)
|
|
);
|
|
|
|
if (${m}) {
|
|
vec4 xValues = vec4(
|
|
getX(batch, xR, xC, d1),
|
|
getX(batch, xR, xC, d1 + 1),
|
|
getX(batch, xR, xC, d1 + 2),
|
|
getX(batch, xR, xC, d1 + 3)
|
|
);
|
|
dotProd += dot(xValues, wValues);
|
|
} else {
|
|
vec4 xValues = vec4(
|
|
getX(batch, d1, xR, xC),
|
|
getX(batch, d1 + 1, xR, xC),
|
|
getX(batch, d1 + 2, xR, xC),
|
|
getX(batch, d1 + 3, xR, xC)
|
|
);
|
|
dotProd += dot(xValues, wValues);
|
|
}
|
|
}
|
|
|
|
if (${f===1}) {
|
|
|
|
if (${m}) {
|
|
dotProd +=
|
|
getX(batch, xR, xC, ${h}) *
|
|
getW(wR, wC, ${h}, d2);
|
|
} else {
|
|
dotProd +=
|
|
getX(batch, ${h}, xR, xC) *
|
|
getW(wR, wC, ${h}, d2);
|
|
}
|
|
|
|
} else if (${f===2}) {
|
|
vec2 wValues = vec2(
|
|
getW(wR, wC, ${h}, d2),
|
|
getW(wR, wC, ${h} + 1, d2)
|
|
);
|
|
|
|
if (${m}) {
|
|
vec2 xValues = vec2(
|
|
getX(batch, xR, xC, ${h}),
|
|
getX(batch, xR, xC, ${h} + 1)
|
|
);
|
|
dotProd += dot(xValues, wValues);
|
|
} else {
|
|
vec2 xValues = vec2(
|
|
getX(batch, ${h}, xR, xC),
|
|
getX(batch, ${h} + 1, xR, xC)
|
|
);
|
|
dotProd += dot(xValues, wValues);
|
|
}
|
|
|
|
} else if (${f===3}) {
|
|
vec3 wValues = vec3(
|
|
getW(wR, wC, ${h}, d2),
|
|
getW(wR, wC, ${h} + 1, d2),
|
|
getW(wR, wC, ${h} + 2, d2)
|
|
);
|
|
|
|
if (${m}) {
|
|
vec3 xValues = vec3(
|
|
getX(batch, xR, xC, ${h}),
|
|
getX(batch, xR, xC, ${h} + 1),
|
|
getX(batch, xR, xC, ${h} + 2)
|
|
);
|
|
dotProd += dot(xValues, wValues);
|
|
} else {
|
|
vec3 xValues = vec3(
|
|
getX(batch, ${h}, xR, xC),
|
|
getX(batch, ${h} + 1, xR, xC),
|
|
getX(batch, ${h} + 2, xR, xC)
|
|
);
|
|
dotProd += dot(xValues, wValues);
|
|
}
|
|
|
|
}
|
|
}
|
|
}
|
|
|
|
float result = dotProd;
|
|
${w}
|
|
${b}
|
|
setOutput(result);
|
|
}
|
|
`}},Sre=class{constructor(e){this.variableNames=["x","W"],this.outputShape=e.outShape;let t=e.padInfo.front,n=e.padInfo.top,s=e.padInfo.left,r=e.strideDepth,a=e.strideHeight,o=e.strideWidth,i=e.dilationDepth,l=e.dilationHeight,u=e.dilationWidth,c=e.filterDepth,p=e.filterHeight,d=e.filterWidth,h=Math.floor(e.inChannels/4)*4,f=e.inChannels%4;this.userCode=`
|
|
const ivec3 strides = ivec3(${r}, ${a}, ${o});
|
|
const ivec3 pads = ivec3(${t}, ${n}, ${s});
|
|
|
|
void main() {
|
|
ivec5 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
int d2 = coords.u;
|
|
|
|
ivec3 xFRCCorner = ivec3(coords.y, coords.z, coords.w) * strides - pads;
|
|
int xFCorner = xFRCCorner.x;
|
|
int xRCorner = xFRCCorner.y;
|
|
int xCCorner = xFRCCorner.z;
|
|
|
|
// Convolve x(?, ?, ?, d1) with w(:, :, :, d1, d2) to get
|
|
// y(yF, yR, yC, d2). ? = to be determined. : = across all
|
|
// values in that axis.
|
|
float dotProd = 0.0;
|
|
for (int wF = 0; wF < ${c}; wF++) {
|
|
int xF = xFCorner + wF * ${i};
|
|
|
|
if (xF < 0 || xF >= ${e.inDepth}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wR = 0; wR < ${p}; wR++) {
|
|
int xR = xRCorner + wR * ${l};
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int wC = 0; wC < ${d}; wC++) {
|
|
int xC = xCCorner + wC * ${u};
|
|
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
continue;
|
|
}
|
|
|
|
for (int d1 = 0; d1 < ${h}; d1 += 4) {
|
|
vec4 xValues = vec4(
|
|
getX(batch, xF, xR, xC, d1),
|
|
getX(batch, xF, xR, xC, d1 + 1),
|
|
getX(batch, xF, xR, xC, d1 + 2),
|
|
getX(batch, xF, xR, xC, d1 + 3)
|
|
);
|
|
vec4 wValues = vec4(
|
|
getW(wF, wR, wC, d1, d2),
|
|
getW(wF, wR, wC, d1 + 1, d2),
|
|
getW(wF, wR, wC, d1 + 2, d2),
|
|
getW(wF, wR, wC, d1 + 3, d2)
|
|
);
|
|
|
|
dotProd += dot(xValues, wValues);
|
|
}
|
|
|
|
if (${f===1}) {
|
|
dotProd +=
|
|
getX(batch, xF, xR, xC, ${h}) *
|
|
getW(wF, wR, wC, ${h}, d2);
|
|
} else if (${f===2}) {
|
|
vec2 xValues = vec2(
|
|
getX(batch, xF, xR, xC, ${h}),
|
|
getX(batch, xF, xR, xC, ${h} + 1)
|
|
);
|
|
vec2 wValues = vec2(
|
|
getW(wF, wR, wC, ${h}, d2),
|
|
getW(wF, wR, wC, ${h} + 1, d2)
|
|
);
|
|
dotProd += dot(xValues, wValues);
|
|
} else if (${f===3}) {
|
|
vec3 xValues = vec3(
|
|
getX(batch, xF, xR, xC, ${h}),
|
|
getX(batch, xF, xR, xC, ${h} + 1),
|
|
getX(batch, xF, xR, xC, ${h} + 2)
|
|
);
|
|
vec3 wValues = vec3(
|
|
getW(wF, wR, wC, ${h}, d2),
|
|
getW(wF, wR, wC, ${h} + 1, d2),
|
|
getW(wF, wR, wC, ${h} + 2, d2)
|
|
);
|
|
dotProd += dot(xValues, wValues);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}},Cre=class{constructor(e,t){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.customUniforms=[{name:"inputShape",type:"ivec4"},{name:"pad",type:"ivec2"},{name:"stride",type:"ivec2"},{name:"dilation",type:"ivec2"},{name:"inChannels",type:"int"},{name:"itemsPerBlockRow",type:"int"},{name:"outWidth",type:"int"}],this.outputShape=e,this.enableShapeUniforms=xs(this.outputShape.length);let{dataFormat:n}=t,s=as(),r=n==="channelsLast",a=r?1:2,o=r?2:3,i=this.enableShapeUniforms?"if(blockIndex < outShape[2] && pos < outShape[1]) {":`if(blockIndex < ${e[2]} && pos < ${e[1]}) {`,l="";for(let u=0;u<=1;u++)for(let c=0;c<=1;c++)l+=`
|
|
blockIndex = rc.z + ${c};
|
|
pos = rc.y + ${u};
|
|
|
|
${i}
|
|
offsetY = int(blockIndex / outWidth) * stride[0] - pad[0];
|
|
d0 = offsetY + dilation[0] * (pos / itemsPerBlockRow);
|
|
|
|
if(d0 < inputShape[${a}] && d0 >= 0) {
|
|
// Use custom imod instead mod. On Intel GPU, mod may generate
|
|
// unexpected value.
|
|
// https://github.com/tensorflow/tfjs/issues/5447
|
|
offsetX = imod(blockIndex, outWidth) * stride[1] - pad[1];
|
|
d1 = offsetX + dilation[1] * (imod(pos, itemsPerBlockRow) /
|
|
inChannels);
|
|
|
|
if(d1 < inputShape[${o}] && d1 >= 0) {
|
|
|
|
ch = imod(pos, inChannels);
|
|
|
|
if (${r}) {
|
|
innerDims = vec2(d1, ch);
|
|
result[${u*2+c}] = getChannel(
|
|
getA(rc.x, d0, int(innerDims.x),
|
|
int(innerDims.y)), innerDims);
|
|
} else {
|
|
innerDims = vec2(d0, d1);
|
|
result[${u*2+c}] = getChannel(
|
|
getA(rc.x, ch, int(innerDims.x),
|
|
int(innerDims.y)), innerDims);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
`;this.userCode=`
|
|
void main() {
|
|
ivec3 rc = getOutputCoords();
|
|
|
|
vec4 result = vec4(0);
|
|
|
|
int blockIndex, pos, offsetY, d0, offsetX, d1, ch;
|
|
vec2 innerDims;
|
|
|
|
${l}
|
|
|
|
${s.output} = result;
|
|
}
|
|
`}};function Om(e,t){let n=e.length;return n>=3?t?[...e.slice(0,-3),e[n-3]*e[n-2],e[n-1]]:[...e.slice(0,-3),e[n-3],e[n-2]*e[n-1]]:!t&&n===1&&e[0]>1?[e[0],1]:null}function I9({x:e,filter:t,convInfo:n,backend:s,bias:r=null,preluActivationWeights:a=null,leakyreluAlpha:o=0,activation:i=null}){let l=e.shape,u=s.texData.get(e.dataId),c=n.inChannels,p=l[0]*l[1]*l[2],d=n.outChannels,h=n.dataFormat==="channelsLast",f=!1,m=!1,g,y=[];if(a!=null){let b=Om(a.shape,h);b!=null&&(a=be({inputs:{x:a},backend:s,attrs:{shape:b}}),y.push(a))}if(r!=null){let b=Om(r.shape,h);b!=null&&(r=be({inputs:{x:r},backend:s,attrs:{shape:b}}),y.push(r))}if(!((p===1||d===1)&&c>y9)&&u.isPacked&&h&&u.texture!=null&&l[2]%2!==0&&v.arraysEqual(u.shape.slice(-3),l.slice(-3))){let b=l[0]*l[1]*(l[2]+1),w={dataId:e.dataId,shape:[1,b,n.inChannels],dtype:e.dtype},k=u.shape;u.shape=u.shape.slice(),u.shape[u.shape.length-2]++,v.assert(Dp(u.shape,w.shape),()=>`packed reshape ${u.shape} to ${w.shape} isn't free`);let S=be({inputs:{x:t},backend:s,attrs:{shape:[1,n.inChannels,n.outChannels]}});y.push(S);let E=Fm({a:w,b:S,backend:s,transposeA:f,transposeB:m,bias:r,activation:i,preluActivationWeights:a,leakyreluAlpha:o}),R=s.texData.get(E.dataId);v.assert(R.isPacked,()=>"batchMatMul result is expected to be packed"),u.shape=k,R.shape=n.outShape,g=Ps({inputs:{x:E},backend:s}),g.shape=n.outShape,y.push(E)}else{let b=n.outHeight*n.outWidth,w=be({inputs:{x:e},backend:s,attrs:{shape:h?[n.batchSize,b,n.inChannels]:[n.batchSize,n.inChannels,b]}}),k=be({inputs:{x:t},backend:s,attrs:{shape:[1,n.inChannels,n.outChannels]}}),S=Fm({a:h?w:k,b:h?k:w,transposeA:!h,transposeB:m,backend:s,bias:r,activation:i,preluActivationWeights:a,leakyreluAlpha:o});g=be({inputs:{x:S},backend:s,attrs:{shape:n.outShape}}),y.push(w),y.push(k),y.push(S)}for(let b of y)s.disposeIntermediateTensorInfo(b);return g}function S9({x:e,filter:t,convInfo:n,backend:s,bias:r=null,preluActivationWeights:a=null,leakyreluAlpha:o=0,activation:i=null}){let{filterWidth:l,filterHeight:u,inChannels:c,outWidth:p,outHeight:d,dataFormat:h}=n,f=h==="channelsLast",m=l*u*c,g=d*p,y=[n.batchSize,m,g],x=!0,A=!1,b=[];if(a!=null){let Z=Om(a.shape,f);Z!=null&&(a=be({inputs:{x:a},backend:s,attrs:{shape:Z}}),b.push(a))}if(r!=null){let Z=Om(r.shape,f);Z!=null&&(r=be({inputs:{x:r},backend:s,attrs:{shape:Z}}),b.push(r))}let w=be({inputs:{x:t},backend:s,attrs:{shape:[1,m,v.sizeFromShape(t.shape)/m]}});b.push(w);let k=new Cre(y,n),S=[e.shape,[n.padInfo.top,n.padInfo.left],[n.strideHeight,n.strideWidth],[n.dilationHeight,n.dilationWidth],[n.inChannels],[n.filterWidth*n.inChannels],[n.outWidth]],E=s.runWebGLProgram(k,[e],"float32",S),R=be({inputs:{x:E},backend:s,attrs:{shape:y}});b.push(E),b.push(R);let $=r!=null,_=a!=null,P=i==="leakyrelu",C=i?w2(i,!0):null,F=new g9(f?R.shape:w.shape,f?w.shape:R.shape,f?[n.batchSize,g,n.outChannels]:[n.batchSize,n.outChannels,g],x,A,$,C,_,P),G=f?[R,w]:[w,R];if(r&&G.push(r),_&&G.push(a),P){let Z=s.makeTensorInfo([],"float32",v.createScalarValue(o,"float32"));G.push(Z),b.push(Z)}let K=s.runWebGLProgram(F,G,"float32"),z=be({inputs:{x:K},backend:s,attrs:{shape:n.outShape}});b.push(K);for(let Z of b)s.disposeIntermediateTensorInfo(Z);return z}function Tre(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dataFormat:l,dilations:u,dimRoundingMode:c}=s,p=T.convertConv2DDataFormat(l),d=T.computeConv2DInfo(r.shape,a.shape,o,u,i,c,!1,p),h;if(d.filterHeight===1&&d.filterWidth===1&&d.dilationHeight===1&&d.dilationWidth===1&&d.strideHeight===1&&d.strideWidth===1&&(d.padInfo.type==="SAME"||d.padInfo.type==="VALID"))h=I9({x:r,filter:a,convInfo:d,backend:n});else if(q().getBool("WEBGL_CONV_IM2COL"))h=S9({x:r,filter:a,convInfo:d,backend:n});else{let m=new k9(d);h=n.runWebGLProgram(m,[r,a],"float32")}let f=be({inputs:{x:h},backend:n,attrs:{shape:d.outShape}});return n.disposeIntermediateTensorInfo(h),f}var Nre={kernelName:Ao,backendName:"webgl",kernelFunc:Tre},Ere=class{constructor(e){this.variableNames=["x","dy"],this.outputShape=e.filterShape;let t=e.strideHeight,n=e.strideWidth,s=e.padInfo.top,r=e.padInfo.left,a=e.dataFormat==="channelsLast";this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int wR = coords.x;
|
|
int wC = coords.y;
|
|
int d1 = coords.z;
|
|
int d2 = coords.w;
|
|
|
|
// Convolve x(?, ?, d1) with dy(:, :, d2) to get dw(wR, wC, d1, d2).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
float dotProd = 0.0;
|
|
|
|
for (int b = 0; b < ${e.batchSize}; b++) {
|
|
for (int yR = 0; yR < ${e.outHeight}; yR++) {
|
|
int xR = wR + yR * ${t} - ${s};
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int yC = 0; yC < ${e.outWidth}; yC++) {
|
|
int xC = wC + yC * ${n} - ${r};
|
|
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
continue;
|
|
}
|
|
|
|
if (${a}) {
|
|
float dyValue = getDy(b, yR, yC, d2);
|
|
float xValue = getX(b, xR, xC, d1);
|
|
dotProd += (xValue * dyValue);
|
|
} else {
|
|
float dyValue = getDy(b, d2, yR, yC);
|
|
float xValue = getX(b, d1, xR, xC);
|
|
dotProd += (xValue * dyValue);
|
|
}
|
|
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}},Rre=class{constructor(e){this.variableNames=["dy","W"],this.outputShape=e.inShape;let t=e.filterHeight,n=e.filterWidth,s=e.strideHeight,r=e.strideWidth,a=e.dataFormat==="channelsLast",o=t-1-e.padInfo.top,i=n-1-e.padInfo.left,l=a?1:2,u=a?2:3,c=a?3:1;this.userCode=`
|
|
const ivec2 pads = ivec2(${o}, ${i});
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int d1 = coords[${c}];
|
|
|
|
ivec2 dyCorner = ivec2(coords[${l}], coords[${u}]) - pads;
|
|
int dyRCorner = dyCorner.x;
|
|
int dyCCorner = dyCorner.y;
|
|
|
|
// Convolve dy(?, ?, d2) with w(:, :, d1, d2) to compute dx(xR, xC, d1).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
float dotProd = 0.0;
|
|
for (int wR = 0; wR < ${t}; wR++) {
|
|
float dyR = float(dyRCorner + wR) / ${s}.0;
|
|
|
|
if (dyR < 0.0 || dyR >= ${e.outHeight}.0 || fract(dyR) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyR = int(dyR);
|
|
|
|
int wRPerm = ${t} - 1 - wR;
|
|
|
|
for (int wC = 0; wC < ${n}; wC++) {
|
|
float dyC = float(dyCCorner + wC) / ${r}.0;
|
|
|
|
if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
|
|
fract(dyC) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyC = int(dyC);
|
|
|
|
int wCPerm = ${n} - 1 - wC;
|
|
|
|
for (int d2 = 0; d2 < ${e.outChannels}; d2++) {
|
|
|
|
if (${a}) {
|
|
float xValue = getDy(batch, idyR, idyC, d2);
|
|
float wValue = getW(wRPerm, wCPerm, d1, d2);
|
|
dotProd += xValue * wValue;
|
|
} else {
|
|
float xValue = getDy(batch, d2, idyR, idyC);
|
|
float wValue = getW(wRPerm, wCPerm, d1, d2);
|
|
dotProd += xValue * wValue;
|
|
}
|
|
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}},_re=class{constructor(e){this.variableNames=["x","dy"],this.outputShape=e.filterShape;let t=e.strideDepth,n=e.strideHeight,s=e.strideWidth,r=e.padInfo.front,a=e.padInfo.top,o=e.padInfo.left;this.userCode=`
|
|
void main() {
|
|
ivec5 coords = getOutputCoords();
|
|
int wF = coords.x;
|
|
int wR = coords.y;
|
|
int wC = coords.z;
|
|
int d1 = coords.w;
|
|
int d2 = coords.u;
|
|
|
|
float dotProd = 0.0;
|
|
|
|
for (int b = 0; b < ${e.batchSize}; b++) {
|
|
for (int yF = 0; yF < ${e.outDepth}; yF++) {
|
|
int xF = wF + yF * ${t} - ${r};
|
|
|
|
if (xF < 0 || xF >= ${e.inDepth}) {
|
|
continue;
|
|
}
|
|
|
|
for (int yR = 0; yR < ${e.outHeight}; yR++) {
|
|
int xR = wR + yR * ${n} - ${a};
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int yC = 0; yC < ${e.outWidth}; yC++) {
|
|
int xC = wC + yC * ${s} - ${o};
|
|
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
continue;
|
|
}
|
|
|
|
float dyValue = getDy(b, yF, yR, yC, d2);
|
|
float xValue = getX(b, xF, xR, xC, d1);
|
|
dotProd += (xValue * dyValue);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}},Dre=class{constructor(e){this.variableNames=["dy","W"],this.outputShape=e.inShape;let t=e.filterDepth,n=e.filterHeight,s=e.filterWidth,r=e.strideDepth,a=e.strideHeight,o=e.strideWidth,i=t-1-e.padInfo.front,l=n-1-e.padInfo.top,u=s-1-e.padInfo.left;this.userCode=`
|
|
const ivec3 pads = ivec3(${i}, ${l}, ${u});
|
|
|
|
void main() {
|
|
ivec5 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
int d1 = coords.u;
|
|
|
|
|
|
ivec3 dyCorner = ivec3(coords.y, coords.z, coords.w) - pads;
|
|
int dyFCorner = dyCorner.x;
|
|
int dyRCorner = dyCorner.y;
|
|
int dyCCorner = dyCorner.z;
|
|
|
|
float dotProd = 0.0;
|
|
for (int wF = 0; wF < ${t}; wF++) {
|
|
float dyF = float(dyFCorner + wF) / ${r}.0;
|
|
|
|
if (dyF < 0.0 || dyF >= ${e.outDepth}.0 || fract(dyF) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyF = int(dyF);
|
|
|
|
int wFPerm = ${t} - 1 - wF;
|
|
|
|
for (int wR = 0; wR < ${n}; wR++) {
|
|
float dyR = float(dyRCorner + wR) / ${a}.0;
|
|
|
|
if (dyR < 0.0 || dyR >= ${e.outHeight}.0 ||
|
|
fract(dyR) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyR = int(dyR);
|
|
|
|
int wRPerm = ${n} - 1 - wR;
|
|
|
|
for (int wC = 0; wC < ${s}; wC++) {
|
|
float dyC = float(dyCCorner + wC) / ${o}.0;
|
|
|
|
if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
|
|
fract(dyC) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyC = int(dyC);
|
|
|
|
int wCPerm = ${s} - 1 - wC;
|
|
|
|
for (int d2 = 0; d2 < ${e.outChannels}; d2++) {
|
|
float xValue = getDy(batch, idyF, idyR, idyC, d2);
|
|
float wValue = getW(wFPerm, wRPerm, wCPerm, d1, d2);
|
|
dotProd += xValue * wValue;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}};function $re(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,dy:a}=t,{strides:o,pad:i,dataFormat:l,dimRoundingMode:u,filterShape:c}=s,p=T.convertConv2DDataFormat(l),d=T.computeConv2DInfo(r.shape,c,o,1,i,u,!1,p),h=new Ere(d);return n.runWebGLProgram(h,[r,a],"float32")}var Pre={kernelName:Xm,backendName:"webgl",kernelFunc:$re};function Fre(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,filter:a}=t,{inputShape:o,strides:i,pad:l,dataFormat:u,dimRoundingMode:c}=s,p=T.convertConv2DDataFormat(u),d=T.computeConv2DInfo(o,a.shape,i,1,l,c,!1,p),h=new Rre(d);return n.runWebGLProgram(h,[r,a],"float32")}var Ore={kernelName:xo,backendName:"webgl",kernelFunc:Fre};function Mre(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dilations:l}=s,u=T.computeConv3DInfo(r.shape,a.shape,o,l,i),c=new Sre(u);return n.runWebGLProgram(c,[r,a],"float32")}var zre={kernelName:Wp,backendName:"webgl",kernelFunc:Mre};function Lre(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,dy:a}=t,{strides:o,pad:i,filterShape:l}=s,u=T.computeConv3DInfo(r.shape,l,o,1,i),c=new _re(u);return n.runWebGLProgram(c,[r,a],"float32")}var Bre={kernelName:Km,backendName:"webgl",kernelFunc:Lre};function Wre(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,filter:a}=t,{pad:o,strides:i,inputShape:l}=s,u=T.computeConv3DInfo(l,a.shape,i,1,o),c=new Dre(u);return n.runWebGLProgram(c,[r,a],"float32")}var Vre={kernelName:Zm,backendName:"webgl",kernelFunc:Wre},Ure=ld+`
|
|
return cos(x);
|
|
`,Gre=dt({opSnippet:Ure}),Hre={kernelName:bo,backendName:"webgl",kernelFunc:Gre},jre=`
|
|
float e2x = exp(-x);
|
|
return (e2x + 1.0 / e2x) / 2.0;
|
|
`,qre=dt({opSnippet:jre}),Xre={kernelName:vo,backendName:"webgl",kernelFunc:qre},Kre=class{constructor(e,t,n,s,r){this.variableNames=["Image","Boxes","BoxInd"],this.outputShape=[];let[a,o,i,l]=e,[u]=t,[c,p]=n;this.outputShape=[u,c,p,l];let d=s==="bilinear"?1:0,[h,f]=[`${o-1}.0`,`${i-1}.0`],[m,g,y]=c>1?[`${(o-1)/(c-1)}`,"(y2-y1) * height_ratio",`y1*${h} + float(y)*(height_scale)`]:["0.0","0.0",`0.5 * (y1+y2) * ${h}`],[x,A,b]=p>1?[`${(i-1)/(p-1)}`,"(x2-x1) * width_ratio",`x1*${f} + float(x)*(width_scale)`]:["0.0","0.0",`0.5 * (x1+x2) * ${f}`];this.userCode=`
|
|
const float height_ratio = float(${m});
|
|
const float width_ratio = float(${x});
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int y = coords[1];
|
|
int x = coords[2];
|
|
int d = coords[3];
|
|
|
|
// get box vals
|
|
float y1 = getBoxes(b,0);
|
|
float x1 = getBoxes(b,1);
|
|
float y2 = getBoxes(b,2);
|
|
float x2 = getBoxes(b,3);
|
|
|
|
// get image in batch index
|
|
int bInd = round(getBoxInd(b));
|
|
if(bInd < 0 || bInd >= ${a}) {
|
|
return;
|
|
}
|
|
|
|
float height_scale = ${g};
|
|
float width_scale = ${A};
|
|
|
|
float in_y = ${y};
|
|
if( in_y < 0.0 || in_y > ${h} ) {
|
|
setOutput(float(${r}));
|
|
return;
|
|
}
|
|
float in_x = ${b};
|
|
if( in_x < 0.0 || in_x > ${f} ) {
|
|
setOutput(float(${r}));
|
|
return;
|
|
}
|
|
|
|
vec2 sourceFracIndexCR = vec2(in_x,in_y);
|
|
if(${d} == 1) {
|
|
// Compute the four integer indices.
|
|
ivec2 sourceFloorCR = ivec2(sourceFracIndexCR);
|
|
ivec2 sourceCeilCR = ivec2(ceil(sourceFracIndexCR));
|
|
|
|
float topLeft = getImage(b, sourceFloorCR.y, sourceFloorCR.x, d);
|
|
float bottomLeft = getImage(b, sourceCeilCR.y, sourceFloorCR.x, d);
|
|
float topRight = getImage(b, sourceFloorCR.y, sourceCeilCR.x, d);
|
|
float bottomRight = getImage(b, sourceCeilCR.y, sourceCeilCR.x, d);
|
|
|
|
vec2 fracCR = sourceFracIndexCR - vec2(sourceFloorCR);
|
|
|
|
float top = topLeft + (topRight - topLeft) * fracCR.x;
|
|
float bottom = bottomLeft + (bottomRight - bottomLeft) * fracCR.x;
|
|
float newValue = top + (bottom - top) * fracCR.y;
|
|
setOutput(newValue);
|
|
} else {
|
|
// Compute the coordinators of nearest neighbor point.
|
|
ivec2 sourceNearestCR = ivec2(floor(
|
|
sourceFracIndexCR + vec2(0.5,0.5)));
|
|
float newValue = getImage(b, sourceNearestCR.y, sourceNearestCR.x, d);
|
|
setOutput(newValue);
|
|
}
|
|
}
|
|
`}},Zre=e=>{let{inputs:t,backend:n,attrs:s}=e,{image:r,boxes:a,boxInd:o}=t,{cropSize:i,method:l,extrapolationValue:u}=s,c=new Kre(r.shape,a.shape,i,l,u);return n.runWebGLProgram(c,[r,a,o],"float32")},Yre={kernelName:dl,backendName:"webgl",kernelFunc:Zre},Pp;(function(e){e.Prod="*",e.Sum="+"})(Pp||(Pp={}));var T7=class{constructor(e,t,n,s){this.op=e,this.outputShape=t,this.variableNames=["x"],this.customUniforms=[{name:"index",type:"float"}];let r=this.outputShape.length,a=this.op===Pp.Prod?"1.0":"0.0",o=n?a:`getX(${N7(r,"coords",this.op)})`,i=this.outputShape[this.outputShape.length-1],l="",u="";n?(l=s?`end != ${i-1}`:"end != 0",u=s?"end + 1":"end - 1"):(l=s?`end + pow2 < ${i}`:"end >= pow2",u=s?"end + pow2":"end - pow2"),this.userCode=`
|
|
void main() {
|
|
${wt(r)} coords = getOutputCoords();
|
|
int end = ${E7(r,"coords",this.op)};
|
|
float val = ${o};
|
|
int pow2 = int(pow(2.0, index));
|
|
if (${l}) {
|
|
int idx = ${u};
|
|
${E7(r,"coords",this.op)} = idx;
|
|
val ${this.op}= getX(${N7(r,"coords",this.op)});
|
|
}
|
|
setOutput(val);
|
|
}
|
|
`}};function N7(e,t,n){if(e===1)return`${t}`;if(e===2)return`${t}.x, ${t}.y`;if(e===3)return`${t}.x, ${t}.y, ${t}.z`;if(e===4)return`${t}.x, ${t}.y, ${t}.z, ${t}.w`;throw new Error(`Cumulative ${n} for rank ${e} is not yet supported`)}function E7(e,t,n){if(e===1)return`${t}`;if(e===2)return`${t}.y`;if(e===3)return`${t}.z`;if(e===4)return`${t}.w`;throw new Error(`Cumulative ${n} for rank ${e} is not yet supported`)}function C9(e,t,n,s,r,a){let o=t.shape.length,i=T.getAxesPermutation([s],o),l=t;i!=null&&(l=ts({inputs:{x:t},backend:n,attrs:{perm:i}}));let u=T.getInnerMostAxes(1,o)[0];if(u!==o-1)throw new Error(`WebGL cumprod shader expects an inner-most axis=${t.shape.length-1} but got axis=${s}`);let c=l.shape[u],p=Ps({inputs:{x:l},backend:n});for(let d=0;d<=Math.ceil(Math.log2(c))-1;d++){let h=new T7(e,l.shape,!1,a),f=[[d]],m=p;p=n.runWebGLProgram(h,[p],p.dtype,f),n.disposeIntermediateTensorInfo(m)}if(r){let d=new T7(e,l.shape,r,a),h=p;p=n.runWebGLProgram(d,[p],p.dtype),n.disposeIntermediateTensorInfo(h)}if(i!=null){let d=T.getUndoAxesPermutation(i),h=ts({inputs:{x:p},backend:n,attrs:{perm:d}});return n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(l),h}return p}function Jre(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,exclusive:o,reverse:i}=s;return C9(Pp.Prod,r,n,a,o,i)}var Qre={kernelName:cl,backendName:"webgl",kernelFunc:Jre};function eae(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,exclusive:o,reverse:i}=s;return C9(Pp.Sum,r,n,a,o,i)}var tae={kernelName:wo,backendName:"webgl",kernelFunc:eae};function nae(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,weights:a}=t,{size:o,binaryOutput:i}=s;if(r.shape.length===1){let l=n.readSync(r.dataId),u=n.readSync(a.dataId),c=a9(l,u,a.dtype,a.shape,o);return n.makeTensorInfo([o],a.dtype,c)}else if(r.shape.length===2){let l=n.bufferSync(r),u=n.bufferSync(a),c=fte(l,u,o,i);return n.makeTensorInfo(c.shape,a.dtype,c.values)}throw new Error(`Error in denseBincount: input must be at most rank 2, but got rank${r.shape.length}.`)}var sae={kernelName:Ym,backendName:"webgl",kernelFunc:nae},rae=class{constructor(e,t,n){this.variableNames=["x"],this.outputShape=[],this.outputShape=e,this.blockSize=t,this.dataFormat=n,this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int h = ${this.getHeightCoordString()};
|
|
int w = ${this.getWidthCoordString()};
|
|
int d = ${this.getDepthCoordString()};
|
|
|
|
int in_h = h / ${t};
|
|
int offset_h = imod(h, ${t});
|
|
int in_w = w / ${t};
|
|
int offset_w = imod(w, ${t});
|
|
int offset_d = (offset_h * ${t} + offset_w) *
|
|
${this.getOutputDepthSize()};
|
|
int in_d = d + offset_d;
|
|
|
|
float result = ${this.getInputSamplingString()};
|
|
setOutput(result);
|
|
}
|
|
`}getHeightCoordString(){return this.dataFormat==="NHWC"?"coords[1]":"coords[2]"}getWidthCoordString(){return this.dataFormat==="NHWC"?"coords[2]":"coords[3]"}getDepthCoordString(){return this.dataFormat==="NHWC"?"coords[3]":"coords[1]"}getOutputDepthSize(){return this.dataFormat==="NHWC"?this.outputShape[3]:this.outputShape[1]}getInputSamplingString(){return this.dataFormat==="NHWC"?"getX(b, in_h, in_w, in_d)":"getX(b, in_d, in_h, in_w)"}};function aae(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockSize:a,dataFormat:o}=s,i=r.shape[0],l=o==="NHWC"?r.shape[1]:r.shape[2],u=o==="NHWC"?r.shape[2]:r.shape[3],c=o==="NHWC"?r.shape[3]:r.shape[1],p=l*a,d=u*a,h=c/(a*a),f=o==="NHWC"?[i,p,d,h]:[i,h,p,d],m=new rae(f,a,o);return n.runWebGLProgram(m,[r],r.dtype)}var oae={kernelName:pl,backendName:"webgl",kernelFunc:aae},T9=class{constructor(e,t=!1,n=null,s=!1,r=!1){this.variableNames=["x","W"],this.customUniforms=[{name:"pads",type:"ivec2"},{name:"strides",type:"ivec2"},{name:"dilations",type:"ivec2"},{name:"inDims",type:"ivec2"}],this.outputShape=e.outShape,this.enableShapeUniforms=xs(this.outputShape.length);let a=e.filterHeight,o=e.filterWidth,i=e.outChannels/e.inChannels,l="",u="";n&&(s?l=`float activation(float a) {
|
|
float b = getPreluActivationWeightsAtOutCoords();
|
|
${n}
|
|
}`:r?l=`float activation(float a) {
|
|
float b = getLeakyreluAlphaAtOutCoords();
|
|
${n}
|
|
}`:l=`
|
|
float activation(float x) {
|
|
${n}
|
|
}
|
|
`,u="result = activation(result);");let c=t?"result += getBiasAtOutCoords();":"";t&&this.variableNames.push("bias"),s&&this.variableNames.push("preluActivationWeights"),r&&this.variableNames.push("leakyreluAlpha"),this.userCode=`
|
|
${l}
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
ivec2 xRCCorner = coords.yz * strides - pads;
|
|
int d2 = coords.w;
|
|
int d1 = d2 / ${i};
|
|
int q = d2 - d1 * ${i};
|
|
|
|
int xRCorner = xRCCorner.x;
|
|
int xCCorner = xRCCorner.y;
|
|
|
|
// Convolve x(?, ?, d1) with w(:, :, d1, q) to get y(yR, yC, d2).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
float dotProd = 0.0;
|
|
// TO DO(dsmilkov): Flatten the two for loops and vec4 the operations.
|
|
for (int wR = 0; wR < ${a}; wR++) {
|
|
int xR = xRCorner + wR * dilations[0];
|
|
|
|
if (xR < 0 || xR >= inDims[0]) {
|
|
continue;
|
|
}
|
|
|
|
for (int wC = 0; wC < ${o}; wC++) {
|
|
int xC = xCCorner + wC * dilations[1];
|
|
|
|
if (xC < 0 || xC >= inDims[1]) {
|
|
continue;
|
|
}
|
|
|
|
float xVal = getX(batch, xR, xC, d1);
|
|
float wVal = getW(wR, wC, d1, q);
|
|
dotProd += xVal * wVal;
|
|
}
|
|
}
|
|
|
|
float result = dotProd;
|
|
${c}
|
|
${u}
|
|
setOutput(result);
|
|
}
|
|
`}},N9=class{constructor(e,t=!1,n=null,s=!1,r=!1){this.variableNames=["x","W"],this.packedInputs=!0,this.packedOutput=!0,this.customUniforms=[{name:"pads",type:"ivec2"},{name:"strides",type:"ivec2"},{name:"dilations",type:"ivec2"},{name:"inDims",type:"ivec2"}],this.outputShape=e.outShape,this.enableShapeUniforms=xs(this.outputShape.length);let a=e.outChannels/e.inChannels,o=e.padInfo.left,i=e.strideWidth,l=e.dilationWidth,u=e.filterHeight,c=e.filterWidth,p=c,d=`
|
|
int xR; int xC; int xCOffset;
|
|
vec4 wTexel; vec4 previous; vec4 final;`;for(let g=0;g<c;g++)d+=`
|
|
vec4 xTexelC${g*2};
|
|
int xTexelC${g*2}Ready;
|
|
vec4 xTexelC${g*2+1};
|
|
int xTexelC${g*2+1}Ready;
|
|
vec4 xC${g};`;d+=`
|
|
for (int r = 0; r < ${u}; r++) {
|
|
`;for(let g=0;g<c;g++)d+=`
|
|
xTexelC${g*2} = vec4(0.0);
|
|
xTexelC${g*2}Ready = 0;
|
|
xTexelC${g*2+1} = vec4(0.0);
|
|
xTexelC${g*2+1}Ready = 0;
|
|
xC${g} = vec4(0.0);`;d+=`
|
|
xR = xRCorner + r * dilations[0];
|
|
if (xR >=0 && xR < inDims[0]) {
|
|
`;for(let g=0;g<(p+1)/2;g++){let y=g*2;if(d+=`
|
|
xC = xCCorner + ${y*l};
|
|
`,i===1){if(y<c&&(o%2===1?(d+=`
|
|
xCOffset = xC + 1;
|
|
if (xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${y}Ready == 0) {
|
|
xTexelC${y} = getX(batch, xR, xCOffset, d1);
|
|
|
|
// Need to manually clear unused channels in case
|
|
// we're reading from recycled texture.
|
|
if (xCOffset + 1 >= inDims[1]) {
|
|
xTexelC${y}.zw = vec2(0.0);
|
|
}
|
|
xTexelC${y}Ready = 1;
|
|
}
|
|
`,l===1&&y>0?d+=`
|
|
xC${y} = vec4(xTexelC${y-2}.zw, xTexelC${y}.xy);
|
|
`:d+=`
|
|
xCOffset = xC + 1 - 2;
|
|
|
|
if (xCOffset >= 0 && xCOffset < inDims[1]) {
|
|
previous = getX(batch, xR, xCOffset, d1);
|
|
|
|
// Need to manually clear unused channels in case
|
|
// we're reading from recycled texture.
|
|
if (xCOffset + 1 >= inDims[1]) {
|
|
previous.zw = vec2(0.0);
|
|
}
|
|
|
|
xC${y} = vec4(previous.zw, xTexelC${y}.xy);
|
|
} else {
|
|
xC${y} = vec4(0.0, 0.0, xTexelC${y}.xy);
|
|
}
|
|
`):d+=`
|
|
if (xC >= 0 && xC < inDims[1] && xTexelC${y}Ready == 0) {
|
|
xTexelC${y} = getX(batch, xR, xC, d1);
|
|
if (xC + 1 >= inDims[1]) {
|
|
xTexelC${y}.zw = vec2(0.0);
|
|
}
|
|
xTexelC${y}Ready = 1;
|
|
}
|
|
|
|
xC${y} = xTexelC${y};
|
|
`,y+1<c)){let x=o%2===0?v.nearestLargerEven(l):l;l%2===0&&o%2===1||l%2!==0&&o%2!==1?(d+=`
|
|
xCOffset = xC + imod(pads[1], 2) + ${x};
|
|
|
|
if (xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${y+1}Ready == 0) {
|
|
xTexelC${y+1} = getX(batch, xR, xCOffset, d1);
|
|
|
|
// Need to manually clear unused channels in case
|
|
// we're reading from recycled texture.
|
|
if (xCOffset + 1 >= inDims[1]) {
|
|
xTexelC${y+1}.zw = vec2(0.0);
|
|
}
|
|
xTexelC${y+1}Ready = 1;
|
|
}
|
|
`,l>1&&(d+=`
|
|
xCOffset -= 2;
|
|
if (xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${y}Ready == 0) {
|
|
xTexelC${y} = getX(batch, xR, xCOffset, d1);
|
|
xTexelC${y}Ready = 1;
|
|
}
|
|
`),d+=`
|
|
xC${y+1} = vec4(xTexelC${y}.zw, xTexelC${y+1}.xy);
|
|
`):x===1?d+=`
|
|
xC${y+1} = xTexelC${y};
|
|
`:d+=`
|
|
xCOffset = xC + ${x};
|
|
|
|
if (xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${y+1}Ready == 0) {
|
|
xTexelC${y+1} = getX(batch, xR, xCOffset, d1);
|
|
if (xCOffset + 1 >= inDims[1]) {
|
|
xTexelC${y+1}.zw = vec2(0.0);
|
|
}
|
|
xTexelC${y+1}Ready = 1;
|
|
}
|
|
|
|
xC${y+1} = xTexelC${y+1};
|
|
`}}else y<c&&(o%2===1?(d+=`
|
|
xCOffset = xC + 1 - strides[1];
|
|
if(xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${y}Ready == 0) {
|
|
xTexelC${y} = getX(batch, xR, xCOffset, d1);
|
|
// Need to manually clear unused channels in case
|
|
// we're reading from recycled texture.
|
|
if (xCOffset + 1 >= inDims[1]) {
|
|
xTexelC${y}.zw = vec2(0.0);
|
|
}
|
|
xTexelC${y}Ready = 1;
|
|
}
|
|
|
|
if(xC + 1 >= 0 && xC + 1 < inDims[1] && xTexelC${y+1}Ready == 0) {
|
|
xTexelC${y+1} = getX(batch, xR, xC + 1, d1);
|
|
// Need to manually clear unused channels in case
|
|
// we're reading from recycled texture.
|
|
if (xC + 2 >= inDims[1]) {
|
|
xTexelC${y+1}.zw = vec2(0.0);
|
|
}
|
|
xTexelC${y+1}Ready = 1;
|
|
}
|
|
|
|
xC${y} = vec4(xTexelC${y}.zw, xTexelC${y+1}.zw);
|
|
`,y+1<c&&(d+=`
|
|
final = vec4(0.0);
|
|
xCOffset = xC + 1 + strides[1];
|
|
if(xCOffset >= 0 && xCOffset < inDims[1]) {
|
|
final = getX(batch, xR, xCOffset, d1);
|
|
}
|
|
xC${y+1} = vec4(xTexelC${y+1}.xy, final.xy);
|
|
`)):(d+=`
|
|
if(xC >= 0 && xC < inDims[1] && xTexelC${y}Ready == 0) {
|
|
xTexelC${y} = getX(batch, xR, xC, d1);
|
|
if (xC + 1 >= inDims[1]) {
|
|
xTexelC${y}.zw = vec2(0.0);
|
|
}
|
|
xTexelC${y}Ready = 1;
|
|
}
|
|
|
|
xCOffset = xC + strides[1];
|
|
if(xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${y+1}Ready == 0) {
|
|
xTexelC${y+1} = getX(batch, xR, xCOffset, d1);
|
|
if (xCOffset + 1 >= inDims[1]) {
|
|
xTexelC${y+1}.zw = vec2(0.);
|
|
}
|
|
xTexelC${y+1}Ready = 1;
|
|
}
|
|
|
|
xC${y} = vec4(
|
|
xTexelC${y}.xy, xTexelC${y+1}.xy);
|
|
`,y+1<c&&(d+=`
|
|
xC${y+1} = vec4(xTexelC${y}.zw, xTexelC${y+1}.zw);
|
|
`)));y<c&&(d+=`
|
|
wTexel = getW(r, ${y}, d1, q);
|
|
dotProd += xC${y} * vec4(wTexel.xz, wTexel.xz);
|
|
`,y+1<c&&(d+=`
|
|
wTexel = getW(r, ${y+1}, d1, q);
|
|
dotProd += xC${y+1} * vec4(wTexel.xz, wTexel.xz);
|
|
`))}d+=`
|
|
}
|
|
`,d+=`
|
|
}
|
|
`;let h="",f="";n&&(s?h=`vec4 activation(vec4 a) {
|
|
vec4 b = getPreluActivationWeightsAtOutCoords();
|
|
${n}
|
|
}`:r?h=`vec4 activation(vec4 a) {
|
|
vec4 b = getLeakyreluAlphaAtOutCoords();
|
|
${n}
|
|
}`:h=`vec4 activation(vec4 x) {
|
|
${n}
|
|
}`,f="result = activation(result);");let m=t?"result += getBiasAtOutCoords();":"";t&&this.variableNames.push("bias"),s&&this.variableNames.push("preluActivationWeights"),r&&this.variableNames.push("leakyreluAlpha"),this.userCode=`
|
|
${h}
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
ivec2 xRCCorner = coords.yz * strides - pads;
|
|
int d2 = coords.w;
|
|
int d1 = d2 / ${a};
|
|
int q = d2 - d1 * ${a};
|
|
int xRCorner = xRCCorner.x;
|
|
int xCCorner = xRCCorner.y;
|
|
|
|
//intialize dotProd with a small epsilon seems to reduce GPU accuracy loss.
|
|
vec4 dotProd = vec4(0.000000000000001);
|
|
|
|
${d}
|
|
|
|
vec4 result = dotProd - vec4(0.000000000000001);
|
|
${m}
|
|
${f}
|
|
setOutput(result);
|
|
}
|
|
`}};function iae(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dilations:l,dimRoundingMode:u}=s,c=l;c==null&&(c=[1,1]),v.assert(T.eitherStridesOrDilationsAreOne(o,c),()=>`Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${o} and dilations '${c}'`);let p=T.computeConv2DInfo(r.shape,a.shape,o,c,i,u,!0),d;q().getBool("WEBGL_PACK_DEPTHWISECONV")&&p.strideWidth<=2&&p.outChannels/p.inChannels===1?d=new N9(p):d=new T9(p);let h=[[p.padInfo.top,p.padInfo.left],[p.strideHeight,p.strideWidth],[p.dilationHeight,p.dilationWidth],[p.inHeight,p.inWidth]];return n.runWebGLProgram(d,[r,a],"float32",h)}var lae={kernelName:ko,backendName:"webgl",kernelFunc:iae},uae=class{constructor(e){this.variableNames=["x","dy"],this.outputShape=e.filterShape;let t=e.strideHeight,n=e.strideWidth,s=e.padInfo.top,r=e.padInfo.left,a=e.outChannels/e.inChannels;this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int wR = coords.x;
|
|
int wC = coords.y;
|
|
int d1 = coords.z;
|
|
int dm = coords.w;
|
|
int d2 = d1 * ${a} + dm;
|
|
|
|
float dotProd = 0.0;
|
|
|
|
// TO DO: Vec4 over the batch size
|
|
for (int b = 0; b < ${e.batchSize}; b++) {
|
|
for (int yR = 0; yR < ${e.outHeight}; yR++) {
|
|
int xR = wR + yR * ${t} - ${s};
|
|
|
|
if (xR < 0 || xR >= ${e.inHeight}) {
|
|
continue;
|
|
}
|
|
|
|
for (int yC = 0; yC < ${e.outWidth}; yC++) {
|
|
int xC = wC + yC * ${n} - ${r};
|
|
|
|
if (xC < 0 || xC >= ${e.inWidth}) {
|
|
continue;
|
|
}
|
|
|
|
float dyValue = getDy(b, yR, yC, d2);
|
|
float xValue = getX(b, xR, xC, d1);
|
|
dotProd += (xValue * dyValue);
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}},cae=class{constructor(e){this.variableNames=["dy","W"],this.outputShape=e.inShape;let t=e.filterHeight,n=e.filterWidth,s=e.strideHeight,r=e.strideWidth,a=t-1-e.padInfo.top,o=n-1-e.padInfo.left,i=e.outChannels/e.inChannels;this.userCode=`
|
|
const ivec2 pads = ivec2(${a}, ${o});
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int d1 = coords[3];
|
|
ivec2 dyCorner = coords.yz - pads;
|
|
int dyRCorner = dyCorner.x;
|
|
int dyCCorner = dyCorner.y;
|
|
|
|
float dotProd = 0.0;
|
|
|
|
for (int wR = 0; wR < ${t}; wR++) {
|
|
float dyR = float(dyRCorner + wR) / ${s}.0;
|
|
|
|
if (dyR < 0.0 || dyR >= ${e.outHeight}.0 || fract(dyR) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyR = int(dyR);
|
|
|
|
int wRPerm = ${t} - 1 - wR;
|
|
|
|
for (int wC = 0; wC < ${n}; wC++) {
|
|
float dyC = float(dyCCorner + wC) / ${r}.0;
|
|
|
|
if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
|
|
fract(dyC) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyC = int(dyC);
|
|
|
|
int wCPerm = ${n} - 1 - wC;
|
|
|
|
// TO DO: Vec4 over the channelMul
|
|
for (int dm = 0; dm < ${i}; dm++) {
|
|
int d2 = d1 * ${i} + dm;
|
|
float xValue = getDy(batch, idyR, idyC, d2);
|
|
float wValue = getW(wRPerm, wCPerm, d1, dm);
|
|
dotProd += xValue * wValue;
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}};function dae(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,dy:a}=t,{strides:o,dilations:i,pad:l,dimRoundingMode:u,filterShape:c}=s,p=T.computeConv2DInfo(r.shape,c,o,i,l,u,!0),d=new uae(p);return n.runWebGLProgram(d,[r,a],"float32")}var pae={kernelName:Jm,backendName:"webgl",kernelFunc:dae};function hae(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,filter:a}=t,{strides:o,dilations:i,pad:l,dimRoundingMode:u,inputShape:c}=s,p=T.computeConv2DInfo(c,a.shape,o,i,l,u,!0),d=new cae(p);return n.runWebGLProgram(d,[r,a],"float32")}var fae={kernelName:Qm,backendName:"webgl",kernelFunc:hae},mae=class{constructor(e){this.variableNames=["X"],this.outputShape=[e,e],this.userCode=`
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
float val = coords[0] == coords[1] ? getX(coords[0]) : 0.0;
|
|
setOutput(val);
|
|
}
|
|
`}};function gae(e){let{inputs:t,backend:n}=e,{x:s}=t,r=[...s.shape,...s.shape],a=v.sizeFromShape(s.shape),o=be({inputs:{x:s},backend:n,attrs:{shape:[a]}}),i=new mae(a),l=n.runWebGLProgram(i,[o],o.dtype),u=be({inputs:{x:l},backend:n,attrs:{shape:r}});return n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(l),u}var yae={kernelName:e0,backendName:"webgl",kernelFunc:gae},Aae=class{constructor(e){this.variableNames=["x","W"],this.outputShape=e.outShape;let{inHeight:t,inWidth:n,padInfo:s,strideHeight:r,strideWidth:a,filterHeight:o,filterWidth:i,dilationHeight:l,dilationWidth:u}=e,{top:c,left:p}=s;this.userCode=`
|
|
const ivec2 strides = ivec2(${r}, ${a});
|
|
const ivec2 pads = ivec2(${c}, ${p});
|
|
const float neg_infinity = -3.4e38;
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
int d1 = coords.w;
|
|
ivec2 outTopLeftCorner =
|
|
coords.yz * strides - pads;
|
|
int hBeg = outTopLeftCorner.x;
|
|
int wBeg = outTopLeftCorner.y;
|
|
|
|
float curVal = neg_infinity;
|
|
for (int h = 0; h < ${o}; h++) {
|
|
int hIn = hBeg + h * ${l};
|
|
|
|
if (hIn >= 0 && hIn < ${t}) {
|
|
for (int w = 0; w < ${i}; w++) {
|
|
int wIn = wBeg + w * ${u};
|
|
|
|
if (wIn >= 0 && wIn < ${n}) {
|
|
float xVal = getX(batch, hIn, wIn, d1);
|
|
float wVal = getW(h, w, d1);
|
|
|
|
float val = xVal + wVal;
|
|
if (val > curVal) {
|
|
curVal = val;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
float result = curVal;
|
|
setOutput(result);
|
|
}
|
|
`}};function xae(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dilations:l}=s,u=T.computeDilation2DInfo(r.shape,a.shape,o,i,"NHWC",l),c,p=new Aae(u);c=n.runWebGLProgram(p,[r,a],"float32");let d=be({inputs:{x:c},backend:n,attrs:{shape:u.outShape}});return n.disposeIntermediateTensorInfo(c),d}var bae={kernelName:Vp,backendName:"webgl",kernelFunc:xae};function vae(e){let{inputs:t,backend:n,attrs:s}=e,{equation:r}=s,a=t,{allDims:o,summedDims:i,idDims:l}=T.decodeEinsumEquation(r,a.length);T.checkEinsumDimSizes(o.length,l,a);let{path:u,steps:c}=T.getEinsumComputePath(i,l),p=c.length,d=null,h=o.length,f=[];for(let m=0;m<p;++m){for(let g of c[m]){let{permutationIndices:y,expandDims:x}=T.getEinsumPermutation(h,l[g]),A;T.isIdentityPermutation(y)?A=a[g]:(A=ts({inputs:{x:a[g]},backend:n,attrs:{perm:y}}),f.push(A));let b=A.shape.slice();for(let w=0;w<x.length;++w)b.splice(x[w],0,1);v.arraysEqual(A.shape,b)||(A=be({inputs:{x:A},backend:n,attrs:{shape:b}}),f.push(A)),d===null?d=A:(d=Jx({inputs:{a:A,b:d},backend:n}),f.push(d))}m<p-1&&(u[m]>=0&&(d=I2({inputs:{x:d},backend:n,attrs:{axis:u[m]-(o.length-h),keepDims:!1}}),f.push(d)),h--)}for(let m of f)m!==d&&n.disposeIntermediateTensorInfo(m);return d}var wae={kernelName:Up,backendName:"webgl",kernelFunc:vae},kae="return (x >= 0.0) ? x : (exp(x) - 1.0);",Iae=`
|
|
vec4 result;
|
|
|
|
result.r = (x.r >= 0.0) ? x.r : (exp(x.r) - 1.0);
|
|
result.g = (x.g >= 0.0) ? x.g : (exp(x.g) - 1.0);
|
|
result.b = (x.b >= 0.0) ? x.b : (exp(x.b) - 1.0);
|
|
result.a = (x.a >= 0.0) ? x.a : (exp(x.a) - 1.0);
|
|
|
|
return result;
|
|
`,Sae=dt({opSnippet:kae,packedOpSnippet:Iae}),Cae={kernelName:So,backendName:"webgl",kernelFunc:Sae},Tae="return (b >= 1.0) ? a : a * (b + 1.0);",Nae=`
|
|
vec4 bGTEZero = vec4(greaterThanEqual(b, vec4(0.)));
|
|
return (bGTEZero * a) + ((vec4(1.0) - bGTEZero) * (a * (b + vec4(1.0))));
|
|
`,Eae=e=>{let{inputs:t,backend:n}=e,{dy:s,y:r}=t,a=q().getBool("WEBGL_PACK_BINARY_OPERATIONS")?new Mh(Nae,s.shape,r.shape):new uc(Tae,s.shape,r.shape);return n.runWebGLProgram(a,[s,r],s.dtype)},Rae={kernelName:t0,backendName:"webgl",kernelFunc:Eae},_ae=`
|
|
return vec4(equal(a, b));
|
|
`,Dae="return float(a == b);",$ae=_n({opSnippet:Dae,packedOpSnippet:_ae,dtype:"bool",cpuKernelImpl:yte}),Pae={kernelName:hl,backendName:"webgl",kernelFunc:$ae},Fae=`
|
|
// Error function is calculated approximately with elementary function.
|
|
// See "Handbook of Mathematical Functions with Formulas,
|
|
// Graphs, and Mathematical Tables", Abramowitz and Stegun.
|
|
float p = ${T.ERF_P};
|
|
float a1 = ${T.ERF_A1};
|
|
float a2 = ${T.ERF_A2};
|
|
float a3 = ${T.ERF_A3};
|
|
float a4 = ${T.ERF_A4};
|
|
float a5 = ${T.ERF_A5};
|
|
|
|
float sign = sign(x);
|
|
x = abs(x);
|
|
float t = 1.0 / (1.0 + p * x);
|
|
return sign * (1.0 - (((((a5*t + a4)*t) + a3)*t + a2)*t + a1)*t*exp(-x*x));
|
|
`,Oae=dt({opSnippet:Fae}),Mae={kernelName:wc,backendName:"webgl",kernelFunc:Oae},zae=ld+`
|
|
return exp(x);
|
|
`,Lae=`
|
|
vec4 result = exp(x);
|
|
bvec4 isNaN = isnan(x);
|
|
result.r = isNaN.r ? x.r : result.r;
|
|
result.g = isNaN.g ? x.g : result.g;
|
|
result.b = isNaN.b ? x.b : result.b;
|
|
result.a = isNaN.a ? x.a : result.a;
|
|
|
|
return result;
|
|
`,E9=dt({opSnippet:zae,packedOpSnippet:Lae,cpuKernelImpl:Ate,dtype:"float32"}),Bae={kernelName:Co,backendName:"webgl",kernelFunc:E9};function ly(e){let{inputs:t,attrs:n,backend:s}=e,{dim:r}=n,{input:a}=t,o=a.shape.length,i=a.shape.slice(),l=r;return r<0&&(v.assert(-(o+1)<=r,()=>`Axis must be in the interval [${-(o+1)}, ${o}]`),l=o+r+1),i.splice(l,0,1),be({inputs:{x:a},backend:s,attrs:{shape:i}})}var Wae={kernelName:fl,backendName:"webgl",kernelFunc:ly},R7="return exp(x) - 1.0;",Vae=dt({opSnippet:R7,packedOpSnippet:R7,cpuKernelImpl:xte}),Uae={kernelName:ml,backendName:"webgl",kernelFunc:Vae},_7=class{constructor(e,t,n){this.variableNames=["real","imag"];let s=t[1];this.outputShape=t;let r=n?`2.0 * ${Math.PI}`:`-2.0 * ${Math.PI}`,a=n?`${s}.0`:"1.0",o;if(e==="real")o="return real * expR - imag * expI;";else if(e==="imag")o="return real * expI + imag * expR;";else throw new Error(`FFT component must be either "real" or "imag", got ${e}.`);this.userCode=`
|
|
const float exponentMultiplier = ${r};
|
|
|
|
float unaryOpComplex(float real, float expR, float imag, float expI) {
|
|
${o}
|
|
}
|
|
|
|
float mulMatDFT(int batch, int index) {
|
|
float indexRatio = float(index) / float(${s});
|
|
float exponentMultiplierTimesIndexRatio =
|
|
exponentMultiplier * indexRatio;
|
|
|
|
float result = 0.0;
|
|
|
|
for (int i = 0; i < ${s}; i++) {
|
|
// x = (-2|2 * PI / N) * index * i;
|
|
float x = exponentMultiplierTimesIndexRatio * float(i);
|
|
float expR = cos(x);
|
|
float expI = sin(x);
|
|
float real = getReal(batch, i);
|
|
float imag = getImag(batch, i);
|
|
|
|
result +=
|
|
unaryOpComplex(real, expR, imag, expI) / ${a};
|
|
}
|
|
|
|
return result;
|
|
}
|
|
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
setOutput(mulMatDFT(coords[0], coords[1]));
|
|
}
|
|
`}};function R9(e,t,n){let s=n.texData.get(e.dataId),r=v.sizeFromShape(e.shape),a=e.shape[e.shape.length-1],o=r/a,i=be({inputs:{x:e},backend:n,attrs:{shape:[o,a]}}),l=i.shape,u=new _7("real",l,t),c=new _7("imag",l,t),p=[{dataId:s.complexTensorInfos.real.dataId,dtype:s.complexTensorInfos.real.dtype,shape:l},{dataId:s.complexTensorInfos.imag.dataId,dtype:s.complexTensorInfos.imag.dtype,shape:l}],d=n.runWebGLProgram(u,p,"float32"),h=n.runWebGLProgram(c,p,"float32"),f=pi({inputs:{real:d,imag:h},backend:n});n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h);let m=be({inputs:{x:f},backend:n,attrs:{shape:e.shape}});return n.disposeIntermediateTensorInfo(i),n.disposeIntermediateTensorInfo(f),m}function Gae(e){let{inputs:t,backend:n}=e,{input:s}=t;return R9(s,!1,n)}var Hae={kernelName:n0,backendName:"webgl",kernelFunc:Gae},jae=class{constructor(e,t){this.outputShape=[],this.customUniforms=[{name:"value",type:"float"}],this.variableNames=["x"],this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
// Input can be obtained from uniform value.
|
|
setOutput(value);
|
|
}
|
|
`}};function Lh(e){let{backend:t,attrs:n}=e,{shape:s,value:r}=n,{dtype:a}=n;if(a=a||v.inferDtype(r),a==="string"){let o=v.getArrayFromDType(a,v.sizeFromShape(s));return o.fill(r),t.makeTensorInfo(s,a,o)}else{let o=new jae(s,r),i=[[r]];return t.runWebGLProgram(o,[],a,i)}}var qae={kernelName:kc,backendName:"webgl",kernelFunc:Lh},Xae=class{constructor(e){this.variableNames=["Image"],this.outputShape=[];let t=e[2];this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int x = coords[2];
|
|
|
|
int coordX = ${t} - x - 1;
|
|
float outputValue;
|
|
if(coordX >= 0 && coordX < ${t}) {
|
|
outputValue = getImage(coords[0], coords[1], coordX, coords[3]);
|
|
} else {
|
|
outputValue = getImage(coords[0], coords[1], coords[2], coords[3]);
|
|
}
|
|
setOutput(outputValue);
|
|
}
|
|
`}},Kae={kernelName:gl,backendName:"webgl",kernelFunc:({inputs:e,backend:t})=>{let{image:n}=e,s=t,r=new Xae(n.shape);return s.runWebGLProgram(r,[n],n.dtype)}},D7="return floor(x);",Zae=dt({opSnippet:D7,packedOpSnippet:D7,cpuKernelImpl:bte}),Yae={kernelName:To,backendName:"webgl",kernelFunc:Zae},Jae=`
|
|
float s = sign(a) * sign(b);
|
|
int ia = round(a);
|
|
int ib = round(b);
|
|
if (ib != 0) {
|
|
// Windows (D3D) wants guaranteed non-zero int division at compile-time.
|
|
return float(idiv(ia, ib, s));
|
|
} else {
|
|
return NAN;
|
|
}
|
|
`,Qae=`
|
|
ivec4 ia = round(a);
|
|
ivec4 ib = round(b);
|
|
bvec4 cond = notEqual(ib, ivec4(0));
|
|
ivec4 result = ivec4(0);
|
|
vec4 s = sign(a) * sign(b);
|
|
|
|
// Windows (D3D) wants guaranteed non-zero int division at compile-time.
|
|
if (cond[0]) {
|
|
result[0] = idiv(ia[0], ib[0], s[0]);
|
|
}
|
|
if (cond[1]) {
|
|
result[1] = idiv(ia[1], ib[1], s[1]);
|
|
}
|
|
if (cond[2]) {
|
|
result[2] = idiv(ia[2], ib[2], s[2]);
|
|
}
|
|
if (cond[3]) {
|
|
result[3] = idiv(ia[3], ib[3], s[3]);
|
|
}
|
|
return vec4(result);
|
|
`,eoe=_n({opSnippet:Jae,packedOpSnippet:Qae,dtype:"int32"}),toe={kernelName:No,backendName:"webgl",kernelFunc:eoe},noe=class{constructor(e){this.variableNames=["A"];let t=as(),[n,s]=e;this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
ivec3 coords = getOutputCoords();
|
|
int texR = coords[0];
|
|
int texC = coords[1];
|
|
int depth = coords[2];
|
|
vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${s}.0, ${n}.0);
|
|
|
|
vec4 values = ${t.texture2D}(A, uv);
|
|
float value;
|
|
if (depth == 0) {
|
|
value = values.r;
|
|
} else if (depth == 1) {
|
|
value = values.g;
|
|
} else if (depth == 2) {
|
|
value = values.b;
|
|
} else if (depth == 3) {
|
|
value = values.a;
|
|
}
|
|
|
|
setOutput(floor(value * 255.0 + 0.5));
|
|
}
|
|
`}},soe=class{constructor(e){this.variableNames=["A"],this.packedInputs=!1,this.packedOutput=!0;let t=as(),[n,s]=e;this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
ivec3 coords = getOutputCoords();
|
|
int texR = coords[0];
|
|
int texC = coords[1];
|
|
int depth = coords[2];
|
|
|
|
vec4 result = vec4(0.);
|
|
|
|
for(int row=0; row<=1; row++) {
|
|
for(int col=0; col<=1; col++) {
|
|
texC = coords[1] + row;
|
|
depth = coords[2] + col;
|
|
|
|
vec2 uv = (vec2(texC, texR) + halfCR) /
|
|
vec2(${s}.0, ${n}.0);
|
|
vec4 values = ${t.texture2D}(A, uv);
|
|
float value;
|
|
if (depth == 0) {
|
|
value = values.r;
|
|
} else if (depth == 1) {
|
|
value = values.g;
|
|
} else if (depth == 2) {
|
|
value = values.b;
|
|
} else if (depth == 3) {
|
|
value = values.a;
|
|
}
|
|
|
|
result[row * 2 + col] = floor(value * 255.0 + 0.5);
|
|
}
|
|
}
|
|
|
|
${t.output} = result;
|
|
}
|
|
`}},roe={kernelName:bp,backendName:"webgl",kernelFunc:aoe},Fu;function aoe(e){let{inputs:t,backend:n,attrs:s}=e,{pixels:r}=t,{numChannels:a}=s,o=typeof HTMLVideoElement!="undefined"&&r instanceof HTMLVideoElement,i=typeof HTMLImageElement!="undefined"&&r instanceof HTMLImageElement,[l,u]=o?[r.videoWidth,r.videoHeight]:[r.width,r.height],c=[u,l],p=[u,l,a];(i||o)&&(Fu==null&&(Fu=document.createElement("canvas").getContext("2d")),Fu.canvas.width=l,Fu.canvas.height=u,Fu.drawImage(r,0,0,l,u),r=Fu.canvas);let d=n.makeTensorInfo(c,"int32");n.texData.get(d.dataId).usage=Xs.PIXELS,n.gpgpu.uploadPixelDataToTexture(n.getTexture(d.dataId),r);let h=q().getBool("WEBGL_PACK")?new soe(p):new noe(p),f=n.runWebGLProgram(h,[d],"int32");return n.disposeData(d.dataId),f}function ooe(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dataFormat:c,dilations:p,dimRoundingMode:d,activation:h,leakyreluAlpha:f}=s,m=T.convertConv2DDataFormat(c),g=T.computeConv2DInfo(r.shape,a.shape,l,p,u,d,!1,m),y,x=[];if(g.filterHeight===1&&g.filterWidth===1&&g.dilationHeight===1&&g.dilationWidth===1&&g.strideHeight===1&&g.strideWidth===1&&(g.padInfo.type==="SAME"||g.padInfo.type==="VALID"))y=I9({x:r,filter:a,convInfo:g,backend:n,bias:o,activation:h,preluActivationWeights:i,leakyreluAlpha:f});else if(q().getBool("WEBGL_CONV_IM2COL"))y=S9({x:r,filter:a,convInfo:g,backend:n,bias:o,activation:h,preluActivationWeights:i,leakyreluAlpha:f});else{let b=o!=null,w=i!=null,k=h==="leakyrelu",S=h?w2(h,!1):null,E=new k9(g,b,S,w,k),R=[r,a],$=(_,P)=>{if(P==="NCHW"&&_.shape.length===1&&_.shape[0]!==1){let C=be({inputs:{x:_},backend:n,attrs:{shape:[_.shape[0],1,1]}});return x.push(C),C}return _};if(b&&R.push($(o,c)),w&&R.push($(i,c)),k){let _=n.makeTensorInfo([],"float32",v.createScalarValue(f,"float32"));R.push(_),x.push(_)}y=n.runWebGLProgram(E,R,"float32")}let A=be({inputs:{x:y},backend:n,attrs:{shape:g.outShape}});return x.push(y),x.forEach(b=>n.disposeIntermediateTensorInfo(b)),A}var ioe={kernelName:Qa,backendName:"webgl",kernelFunc:ooe};function loe(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dilations:c,dimRoundingMode:p,activation:d,leakyreluAlpha:h}=s,f=[],m=c;m==null&&(m=[1,1]),v.assert(T.eitherStridesOrDilationsAreOne(l,m),()=>`Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${l} and dilations '${m}'`);let g=T.computeConv2DInfo(r.shape,a.shape,l,m,u,p,!0),y=q().getBool("WEBGL_PACK_DEPTHWISECONV")&&g.strideWidth<=2&&g.outChannels/g.inChannels===1,x=d?w2(d,y):null,A=[r,a],b=o!=null,w=i!=null,k=d==="leakyrelu";if(b&&A.push(o),w&&A.push(i),k){let $=n.makeTensorInfo([],"float32",v.createScalarValue(h,"float32"));A.push($),f.push($)}let S;y?S=new N9(g,b,x,w,k):S=new T9(g,b,x,w,k);let E=[[g.padInfo.top,g.padInfo.left],[g.strideHeight,g.strideWidth],[g.dilationHeight,g.dilationWidth],[g.inHeight,g.inWidth]],R=n.runWebGLProgram(S,A,"float32",E);return f.forEach($=>n.disposeIntermediateTensorInfo($)),R}var uoe={kernelName:eo,backendName:"webgl",kernelFunc:loe},coe=class{constructor(e,t,n,s){this.sliceDim=e,this.strides=t,this.paramsShape=s,this.variableNames=["x","indices"],this.outputShape=n;let r=wt(t.length),a=wt(n.length),o=this.sliceDim>1?"strides[j]":"strides",i=wt(s.length),l=s.length>1?"paramsShape[j]":"paramsShape";this.userCode=`
|
|
${r} strides = ${r}(${this.strides});
|
|
${i} paramsShape = ${i}(${this.paramsShape});
|
|
void main() {
|
|
${a} coords = getOutputCoords();
|
|
int flattenIndex = 0;
|
|
bool out_of_bounds = false;
|
|
for (int j = 0; j < ${this.sliceDim}; j++) {
|
|
int index = round(getIndices(coords[0], j));
|
|
out_of_bounds = out_of_bounds || index < 0;
|
|
out_of_bounds = out_of_bounds || index >= ${l};
|
|
flattenIndex += index * ${o};
|
|
}
|
|
setOutput(out_of_bounds ? 0.0 : getX(flattenIndex, coords[1]));
|
|
}
|
|
`}};function doe(e){let{inputs:t,backend:n}=e,{params:s,indices:r}=t,a=r.shape,o=a[a.length-1],i=v.sizeFromShape(s.shape),[l,u,c,p]=T.prepareAndValidate(s,r),d=be({inputs:{x:r},backend:n,attrs:{shape:[u,o]}}),h=be({inputs:{x:s},backend:n,attrs:{shape:[v.sizeFromShape(s.shape)/c,c]}});if(n.shouldExecuteOnCPU([s,r])||s.dtype==="string"){let y=n.readSync(r.dataId),x=n.bufferSync(s),A=vte(y,x,s.dtype,u,o,c,p,s.shape,i);return n.makeTensorInfo(l,s.dtype,A.values)}let f=new coe(o,p,[u,c],s.shape),m=n.runWebGLProgram(f,[h,d],h.dtype),g=be({inputs:{x:m},backend:n,attrs:{shape:l}});return n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(m),g}var poe={kernelName:Al,backendName:"webgl",kernelFunc:doe},hoe=class{constructor(e,t){this.variableNames=["A","indices"],this.outputShape=t,this.rank=t.length;let n=wt(this.rank),s=foe(e,2);this.userCode=`
|
|
void main() {
|
|
${n} resRC = getOutputCoords();
|
|
int index = int(getIndices(resRC.x, resRC.z));
|
|
float inBounds = (index >= 0) && (index < ${e[2]}) ? 1.0 : 0.0;
|
|
setOutput(inBounds * getA(${s}));
|
|
}
|
|
`}};function foe(e,t){let n=["resRC.x","resRC.y","resRC.z","resRC.w"],s=[];for(let r=0;r<e.length;r++)r===2?s.push("index"):s.push(`${n[r]}`);return s.join()}function _9(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,indices:a}=t,{axis:o,batchDims:i}=s,l=v.parseAxisParam(o,r.shape)[0];if(q().get("DEBUG")){let x=n.readSync(a.dataId),A=r.shape[l];for(let b=0;b<x.length;++b){let w=x[b];v.assert(w<=A-1&&w>=0,()=>`GatherV2: the index value ${w} is not in [0, ${A-1}]`)}}let u=T.segment_util.collectGatherOpShapeInfo(r,a,l,i),c=v.sizeFromShape(a.shape),p=[],d=be({inputs:{x:r},backend:n,attrs:{shape:[u.batchSize,u.outerSize,u.dimSize,u.sliceSize]}}),h=be({inputs:{x:a},backend:n,attrs:{shape:[u.batchSize,c/u.batchSize]}});p.push(d),p.push(h);let f=[u.batchSize,u.outerSize,c/u.batchSize,u.sliceSize];if(n.shouldExecuteOnCPU([r,a])||r.dtype==="string"){let x=n.bufferSync(h),A=n.bufferSync(d),b=wte(A,x,f);return p.forEach(w=>n.disposeIntermediateTensorInfo(w)),n.makeTensorInfo(u.outputShape,b.dtype,b.values)}let m=new hoe(d.shape,f),g=n.runWebGLProgram(m,[d,h],d.dtype);p.push(g);let y=be({inputs:{x:g},backend:n,attrs:{shape:u.outputShape}});return p.forEach(x=>n.disposeIntermediateTensorInfo(x)),y}var moe={kernelName:yl,backendName:"webgl",kernelFunc:_9},goe="return float(a > b);",yoe=`
|
|
return vec4(greaterThan(a, b));
|
|
`,Aoe=_n({opSnippet:goe,packedOpSnippet:yoe,cpuKernelImpl:kte,dtype:"bool"}),xoe={kernelName:xl,backendName:"webgl",kernelFunc:Aoe},boe="return float(a >= b);",voe=`
|
|
return vec4(greaterThanEqual(a, b));
|
|
`,woe=_n({opSnippet:boe,packedOpSnippet:voe,dtype:"bool",cpuKernelImpl:Ite}),koe={kernelName:Ro,backendName:"webgl",kernelFunc:woe};function Ioe(e){let{inputs:t,backend:n}=e,{input:s}=t;return R9(s,!0,n)}var Soe={kernelName:s0,backendName:"webgl",kernelFunc:Ioe},Coe="return float(!isnan(x) && !isinf(x));",Toe=dt({opSnippet:Coe,dtype:"bool"}),Noe={kernelName:Ic,backendName:"webgl",kernelFunc:Toe},Eoe="return float(isinf(x));",Roe=dt({opSnippet:Eoe,dtype:"bool"}),_oe={kernelName:Sc,backendName:"webgl",kernelFunc:Roe},Doe="return float(isnan(x));",$oe=dt({opSnippet:Doe,dtype:"bool"}),Poe={kernelName:Cc,backendName:"webgl",kernelFunc:$oe},Foe="return float(a < b);",Ooe=`
|
|
return vec4(lessThan(a, b));
|
|
`,Moe=_n({opSnippet:Foe,packedOpSnippet:Ooe,cpuKernelImpl:Ste,dtype:"bool"}),zoe={kernelName:bl,backendName:"webgl",kernelFunc:Moe},Loe="return float(a <= b);",Boe=`
|
|
return vec4(lessThanEqual(a, b));
|
|
`,Woe=_n({opSnippet:Loe,packedOpSnippet:Boe,cpuKernelImpl:Cte,dtype:"bool"}),Voe={kernelName:vl,backendName:"webgl",kernelFunc:Woe};function Uoe(e){let{backend:t,attrs:n}=e,{start:s,stop:r,num:a}=n,o=Tte(s,r,a);return t.makeTensorInfo([o.length],"float32",o)}var Goe={kernelName:r0,backendName:"webgl",kernelFunc:Uoe},Hoe=ld+`
|
|
return x < 0.0 ? 0./0. : log(x);
|
|
`,joe=`
|
|
vec4 result = log(x);
|
|
bvec4 isNaN = isnan(x);
|
|
result.r = isNaN.r ? x.r : (x.r < 0.0 ? 0./0. : result.r);
|
|
result.g = isNaN.g ? x.g : (x.g < 0.0 ? 0./0. : result.g);
|
|
result.b = isNaN.b ? x.b : (x.b < 0.0 ? 0./0. : result.b);
|
|
result.a = isNaN.a ? x.a : (x.a < 0.0 ? 0./0. : result.a);
|
|
return result;
|
|
`,qoe=dt({opSnippet:Hoe,packedOpSnippet:joe,cpuKernelImpl:Nte}),Xoe={kernelName:$o,backendName:"webgl",kernelFunc:qoe},Koe=ld+`
|
|
return log(1.0 + x);
|
|
`,Zoe=dt({opSnippet:Koe}),Yoe={kernelName:Tc,backendName:"webgl",kernelFunc:Zoe},Joe="return float(a >= 1.0 && b >= 1.0);",Qoe=`
|
|
return vec4(
|
|
vec4(greaterThanEqual(a, vec4(1.0))) *
|
|
vec4(greaterThanEqual(b, vec4(1.0))));
|
|
`,eie=_n({opSnippet:Joe,packedOpSnippet:Qoe,dtype:"bool"}),tie={kernelName:wl,backendName:"webgl",kernelFunc:eie},nie="return float(!(x >= 1.0));",sie=dt({opSnippet:nie}),rie={kernelName:kl,backendName:"webgl",kernelFunc:sie},aie="return float(a >= 1.0 || b >= 1.0);",oie=`
|
|
return min(
|
|
vec4(greaterThanEqual(a, vec4(1.0))) +
|
|
vec4(greaterThanEqual(b, vec4(1.0))),
|
|
vec4(1.0));
|
|
`,iie=_n({opSnippet:aie,packedOpSnippet:oie,dtype:"bool"}),lie={kernelName:Nc,backendName:"webgl",kernelFunc:iie},uie=class{constructor(e,t,n,s,r){this.variableNames=["x"],this.outputShape=[];let a=t,o=e[3]-1;this.outputShape=e;let i,l=`float(${n}) + float(${s}) * sum`;r===.5?i=`inversesqrt(${l})`:r===1?i=`1.0/(${l})`:i=`exp(log(${l}) * float(-${r}));`,this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int r = coords[1];
|
|
int c = coords[2];
|
|
int d = coords[3];
|
|
float x = getX(b, r, c, d);
|
|
float sum = 0.0;
|
|
for (int j = -${a}; j <= ${a}; j++) {
|
|
int idx = d + j;
|
|
if (idx >= 0 && idx <= ${o}) {
|
|
float z = getX(b, r, c, idx);
|
|
sum += z * z;
|
|
}
|
|
}
|
|
float val = x * ${i};
|
|
setOutput(val);
|
|
}
|
|
`}},cie=class{constructor(e,t,n,s,r){this.variableNames=["x"],this.outputShape=[],this.packedInputs=!0,this.packedOutput=!0;let a=t,o=e[3]-1;this.outputShape=e;let i,l=`float(${n}) + float(${s}) * sum`;r===.5?i=`inversesqrt(${l})`:r===1?i=`1.0/(${l})`:i=`exp(log(${l}) * float(-${r}));`,this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords.x;
|
|
int r = coords.y;
|
|
int c = coords.z;
|
|
int d = coords.w;
|
|
|
|
bool hasNextCol = d < ${this.outputShape[3]};
|
|
bool hasNextRow = c < ${this.outputShape[2]};
|
|
|
|
vec4 sum = vec4(0.);
|
|
vec4 xFragAtOutputCoords = getX(b, r, c, d);
|
|
|
|
vec4 xAtOutputCoords = vec4(
|
|
getChannel(xFragAtOutputCoords, vec2(c, d)),
|
|
hasNextCol ?
|
|
getChannel(xFragAtOutputCoords, vec2(c, d + 1)) : 0.0,
|
|
hasNextRow ?
|
|
getChannel(xFragAtOutputCoords , vec2(c + 1, d)) : 0.0,
|
|
(hasNextRow && hasNextCol) ?
|
|
getChannel(xFragAtOutputCoords, vec2(c + 1, d + 1)) : 0.0
|
|
);
|
|
|
|
int firstChannel = d - ${a};
|
|
vec2 cache = vec2(0.);
|
|
if(firstChannel >= 0){
|
|
vec4 firstChannelFrag = getX(b, r, c, firstChannel);
|
|
cache.x = getChannel(firstChannelFrag, vec2(c, firstChannel));
|
|
if(hasNextRow){
|
|
cache.y = getChannel(firstChannelFrag, vec2(c + 1, firstChannel));
|
|
}
|
|
}
|
|
|
|
ivec2 depth = ivec2(d, d + 1);
|
|
for (int j = - ${a}; j <= ${a}; j++) {
|
|
ivec2 idx = depth + j;
|
|
bvec2 aboveLowerBound = greaterThanEqual(idx, ivec2(0));
|
|
bvec2 belowUpperBound = lessThanEqual(idx, ivec2(${o}));
|
|
|
|
bool depthInRange = aboveLowerBound.x && belowUpperBound.x;
|
|
bool depthPlusOneInRange = aboveLowerBound.y && belowUpperBound.y;
|
|
|
|
if(depthInRange || depthPlusOneInRange){
|
|
vec4 z = vec4(0.);
|
|
vec4 xFragAtCurrentDepth;
|
|
z.xz = cache.xy;
|
|
if(depthPlusOneInRange && hasNextCol){
|
|
xFragAtCurrentDepth = idx.y != d ?
|
|
getX(b, r, c, idx.y) : xFragAtOutputCoords;
|
|
z.y = getChannel(xFragAtCurrentDepth, vec2(c, idx.y));
|
|
if(hasNextRow){
|
|
z.w = getChannel(xFragAtCurrentDepth, vec2(c + 1, idx.y));
|
|
}
|
|
}
|
|
cache.xy = z.yw;
|
|
sum += z * z;
|
|
}
|
|
}
|
|
vec4 result = xAtOutputCoords * ${i};
|
|
setOutput(result);
|
|
}
|
|
`}},die=e=>{let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{depthRadius:a,bias:o,alpha:i,beta:l}=s,u=q().getBool("WEBGL_PACK_NORMALIZATION")?new cie(r.shape,a,o,i,l):new uie(r.shape,a,o,i,l);return n.runWebGLProgram(u,[r],r.dtype)},pie={kernelName:Hp,backendName:"webgl",kernelFunc:die},hie=class{constructor(e,t,n,s,r){this.variableNames=["inputImage","outputImage","dy"],this.outputShape=[],this.outputShape=e,this.depth=e[3],this.depthRadius=t,this.bias=n,this.alpha=s,this.beta=r,this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int r = coords[1];
|
|
int c = coords[2];
|
|
|
|
float result = 0.0;
|
|
for (int d = 0; d < ${this.depth}; ++d) {
|
|
int depthBegin = int(max(0.0, float(d - ${t})));
|
|
int depthEnd = int(min(float(${this.depth}),
|
|
float(d + ${t} + 1)));
|
|
|
|
const int MIN_DEPTH_BEGIN = 0;
|
|
const int MAX_DEPTH_END = ${this.depth};
|
|
|
|
float norm = 0.0;
|
|
for (int k = MIN_DEPTH_BEGIN; k < MAX_DEPTH_END; ++k) {
|
|
if (k < depthBegin){
|
|
continue;
|
|
}
|
|
else if (k >= depthBegin && k < depthEnd) {
|
|
norm += getInputImage(b, r, c, k) * getInputImage(b, r, c, k);
|
|
}
|
|
else {
|
|
break;
|
|
}
|
|
}
|
|
|
|
norm = float(${s}) * norm + float(${n});
|
|
|
|
for(int k = MIN_DEPTH_BEGIN; k < MAX_DEPTH_END; ++k){
|
|
if (k < depthBegin){
|
|
continue;
|
|
}
|
|
else if (k >= depthBegin && k < depthEnd){
|
|
float dyi = -2.0 * float(${s})
|
|
* float(${r})
|
|
* getInputImage(b ,r ,c, k) * getOutputImage(b, r, c, d)
|
|
/ norm;
|
|
if (k == d) {
|
|
dyi += pow(norm, -1.0 * ${r});
|
|
}
|
|
if (k == coords[3]) {
|
|
dyi *= getDy(b, r, c, d);
|
|
result += dyi;
|
|
}
|
|
}
|
|
else {
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
setOutput(result);
|
|
}
|
|
`}},fie=e=>{let{inputs:t,backend:n,attrs:s}=e,{x:r,y:a,dy:o}=t,{depthRadius:i,bias:l,alpha:u,beta:c}=s,p=new hie(r.shape,i,l,u,c);return n.runWebGLProgram(p,[r,a,o],r.dtype)},mie={kernelName:a0,backendName:"webgl",kernelFunc:fie};function gie(e,t,n,s){let r=v.sizeFromShape(t),o=v.sizeFromShape(e.shape)/r,i=be({inputs:{x:e},attrs:{shape:[o,r]},backend:s}),l=au(i,e.dtype,"max",s),u=be({inputs:{x:l},attrs:{shape:n},backend:s});return s.disposeIntermediateTensorInfo(i),s.disposeIntermediateTensorInfo(l),u}function D9(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{reductionIndices:a,keepDims:o}=s,i=r.shape.length,l=v.parseAxisParam(a,r.shape),u=l,c=T.getAxesPermutation(u,i),p=c!=null,d=n.shouldExecuteOnCPU([r]),h=r;if(p){if(d){let A=n.texData.get(h.dataId).values,b=new Array(i);for(let S=0;S<b.length;S++)b[S]=r.shape[c[S]];let w=Yx(A,r.shape,r.dtype,c,b);h=n.makeTensorInfo(b,r.dtype);let k=n.texData.get(h.dataId);k.values=w}else h=k2(r,c,n);u=T.getInnerMostAxes(u.length,i)}T.assertAxesAreInnerMostDims("max",u,i);let[f,m]=T.computeOutAndReduceShapes(h.shape,u),g=f;o&&(g=T.expandShapeToKeepDim(f,l));let y;if(d){let A=n.texData.get(h.dataId).values,b=Ete(A,v.sizeFromShape(m),g,r.dtype);y=n.makeTensorInfo(g,r.dtype);let w=n.texData.get(y.dataId);w.values=b}else y=gie(h,m,g,n);return p&&n.disposeIntermediateTensorInfo(h),y}var yie={kernelName:Po,backendName:"webgl",kernelFunc:D9},Aie=d9+`
|
|
return max(a, b);
|
|
`,xie=`
|
|
vec4 result = vec4(max(a, b));
|
|
vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
|
|
`+v2+`
|
|
return result;
|
|
`,bie=_n({opSnippet:Aie,packedOpSnippet:xie,cpuKernelImpl:Rte}),vie={kernelName:Fo,backendName:"webgl",kernelFunc:bie};function wie(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t;nd(r,"maxPool");let{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=s,u=1;v.assert(T.eitherStridesOrDilationsAreOne(o,u),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${o} and dilations '${u}'`);let c=T.computePool2DInfo(r.shape,a,o,u,i,l);if(c.filterWidth===1&&c.filterHeight===1&&v.arraysEqual(c.inShape,c.outShape))return Ps({inputs:{x:r},backend:n});let p=new $p(c,"max",!1);return n.runWebGLProgram(p,[r],r.dtype)}var kie={kernelName:Oo,backendName:"webgl",kernelFunc:wie};function Iie(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{filterSize:a,strides:o,pad:i,dataFormat:l,dimRoundingMode:u}=s,c=[1,1,1],p=T.computePool3DInfo(r.shape,a,o,c,i,u,l),d=new Qx(p,"max",!1);return n.runWebGLProgram(d,[r],r.dtype)}var Sie={kernelName:jp,backendName:"webgl",kernelFunc:Iie},Cie=class{constructor(e){this.variableNames=["dy","maxPos"],this.outputShape=e.inShape;let t=e.strideHeight,n=e.strideWidth,s=e.dilationHeight,r=e.effectiveFilterHeight,a=e.effectiveFilterWidth,o=r-1-e.padInfo.top,i=a-1-e.padInfo.left,l=r*a-1;this.userCode=`
|
|
const ivec2 pads = ivec2(${o}, ${i});
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int d = coords[3];
|
|
|
|
ivec2 dyRCCorner = coords.yz - pads;
|
|
int dyRCorner = dyRCCorner.x;
|
|
int dyCCorner = dyRCCorner.y;
|
|
|
|
// Convolve dy(?, ?, d) with pos mask(:, :, d) to get dx(xR, xC, d).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
float dotProd = 0.0;
|
|
for (int wR = 0; wR < ${r};
|
|
wR += ${s}) {
|
|
float dyR = float(dyRCorner + wR) / ${t}.0;
|
|
|
|
if (dyR < 0.0 || dyR >= ${e.outHeight}.0 || fract(dyR) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyR = int(dyR);
|
|
|
|
for (int wC = 0; wC < ${a}; wC++) {
|
|
float dyC = float(dyCCorner + wC) / ${n}.0;
|
|
|
|
if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
|
|
fract(dyC) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyC = int(dyC);
|
|
|
|
float dyValue = getDy(b, idyR, idyC, d);
|
|
int maxPosValue = ${l} - int(getMaxPos(b, idyR, idyC, d));
|
|
|
|
// Get the current value, check it against the value from the
|
|
// position matrix.
|
|
int curPosValue = wR * ${a} + wC;
|
|
float mask = float(maxPosValue == curPosValue ? 1.0 : 0.0);
|
|
|
|
dotProd += dyValue * mask;
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}},Tie=class{constructor(e){this.variableNames=["dy","maxPos"],this.outputShape=e.inShape;let t=e.strideDepth,n=e.strideHeight,s=e.strideWidth,r=e.dilationDepth,a=e.dilationHeight,o=e.dilationWidth,i=e.effectiveFilterDepth,l=e.effectiveFilterHeight,u=e.effectiveFilterWidth,c=i-1-e.padInfo.front,p=l-1-e.padInfo.top,d=u-1-e.padInfo.left,h=i*l*u-1;this.userCode=`
|
|
const ivec3 pads = ivec3(${c}, ${p}, ${d});
|
|
|
|
void main() {
|
|
ivec5 coords = getOutputCoords();
|
|
int batch = coords.x;
|
|
int ch = coords.u;
|
|
|
|
ivec3 dyCorner = ivec3(coords.y, coords.z, coords.w) - pads;
|
|
int dyDCorner = dyCorner.x;
|
|
int dyRCorner = dyCorner.y;
|
|
int dyCCorner = dyCorner.z;
|
|
|
|
// Convolve dy(?, ?, ?, ch) with pos mask(:, :, :, d) to get
|
|
// dx(xD, xR, xC, ch).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
float dotProd = 0.0;
|
|
|
|
for (int wD = 0; wD < ${i};
|
|
wD += ${r}) {
|
|
float dyD = float(dyDCorner + wD) / ${t}.0;
|
|
|
|
if (dyD < 0.0 || dyD >= ${e.outDepth}.0 || fract(dyD) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyD = int(dyD);
|
|
|
|
for (int wR = 0; wR < ${l};
|
|
wR += ${a}) {
|
|
float dyR = float(dyRCorner + wR) / ${n}.0;
|
|
|
|
if (dyR < 0.0 || dyR >= ${e.outHeight}.0 ||
|
|
fract(dyR) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyR = int(dyR);
|
|
|
|
for (int wC = 0; wC < ${u};
|
|
wC += ${o}) {
|
|
float dyC = float(dyCCorner + wC) / ${s}.0;
|
|
|
|
if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
|
|
fract(dyC) > 0.0) {
|
|
continue;
|
|
}
|
|
int idyC = int(dyC);
|
|
|
|
float dyValue = getDy(batch, idyD, idyR, idyC, ch);
|
|
int maxPosValue = ${h} -
|
|
int(getMaxPos(batch, idyD, idyR, idyC, ch));
|
|
|
|
// Get the current value, check it against the value from the
|
|
// position matrix.
|
|
int curPosValue =
|
|
wD * ${l} * ${u} +
|
|
wR * ${u} + wC;
|
|
float mask = float(maxPosValue == curPosValue ? 1.0 : 0.0);
|
|
|
|
dotProd += dyValue * mask;
|
|
}
|
|
}
|
|
}
|
|
setOutput(dotProd);
|
|
}
|
|
`}};function Nie(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,input:a}=t,o=a,{filterSize:i,strides:l,pad:u,dimRoundingMode:c}=s,p=[1,1,1],d=T.computePool3DInfo(o.shape,i,l,p,u,c),h=new Qx(d,"max",!0),f=n.runWebGLProgram(h,[o],o.dtype),m=new Tie(d),g=n.runWebGLProgram(m,[r,f],o.dtype);return n.disposeIntermediateTensorInfo(f),g}var Eie={kernelName:i0,backendName:"webgl",kernelFunc:Nie};function Rie(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,input:a,output:o}=t,i=a;nd([a,o],"maxPoolGrad");let{filterSize:l,strides:u,pad:c,dimRoundingMode:p}=s,d=T.computePool2DInfo(i.shape,l,u,1,c,p),h=!0,f=new $p(d,"max",h),m=n.runWebGLProgram(f,[i],i.dtype),g=new Cie(d),y=n.runWebGLProgram(g,[r,m],i.dtype);return n.disposeIntermediateTensorInfo(m),y}var _ie={kernelName:o0,backendName:"webgl",kernelFunc:Rie};function Die(e,t,n,s){let r=new $p(n,"max",!1),a=s.runWebGLProgram(r,[e],"float32");r=new $p(n,"max",!0,!0,t);let o=s.runWebGLProgram(r,[e],"float32");return[a,o]}var $ie={kernelName:l0,backendName:"webgl",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{x:s}=e,{filterSize:r,strides:a,pad:o,includeBatchInIndex:i}=t,l=n;v.assert(s.shape.length===4,()=>`Error in maxPool: input must be rank 4 but got rank ${s.shape.length}.`);let u=[1,1];v.assert(T.eitherStridesOrDilationsAreOne(a,u),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${a} and dilations '${u}'`);let c=T.computePool2DInfo(s.shape,r,a,u,o),[p,d]=Die(s,i,c,l);return[p,d]}};function Pie(e,t,n,s){let r=v.sizeFromShape(t),o=v.sizeFromShape(e.shape)/r,i=be({inputs:{x:e},attrs:{shape:[o,r]},backend:s}),l=au(i,"float32","mean",s),u=be({inputs:{x:l},attrs:{shape:n},backend:s});return s.disposeIntermediateTensorInfo(i),s.disposeIntermediateTensorInfo(l),u}var Fie={kernelName:Mo,backendName:"webgl",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{x:s}=e,{keepDims:r,axis:a}=t,o=n,i=s.shape.length,l=v.parseAxisParam(a,s.shape),u=l,c=T.getAxesPermutation(u,i),p=c!=null,d=o.shouldExecuteOnCPU([s]),h=[],f=s;if(p){if(d){let b=o.texData.get(f.dataId).values,w=new Array(i);for(let E=0;E<w.length;E++)w[E]=s.shape[c[E]];let k=Yx(b,s.shape,s.dtype,c,w);f=o.makeTensorInfo(w,s.dtype);let S=o.texData.get(f.dataId);S.values=k}else f=k2(s,c,o);h.push(f),u=T.getInnerMostAxes(u.length,i)}T.assertAxesAreInnerMostDims("sum",u,i);let[m,g]=T.computeOutAndReduceShapes(f.shape,u),y=m;r&&(y=T.expandShapeToKeepDim(m,l));let x=Pie(f,g,y,o);for(let A of h)o.disposeIntermediateTensorInfo(A);return x}};function Oie(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s,i=r.shape.length,l=v.parseAxisParam(a,r.shape),u=l,c=T.getAxesPermutation(u,i),p=r;c!=null&&(p=ts({inputs:{x:r},backend:n,attrs:{perm:c}}),u=T.getInnerMostAxes(u.length,r.shape.length)),T.assertAxesAreInnerMostDims("min",u,i);let[d,h]=T.computeOutAndReduceShapes(p.shape,u),f=v.sizeFromShape(h),m=be({inputs:{x:p},backend:n,attrs:{shape:[-1,f]}}),g=au(m,m.dtype,"min",n),y;if(o){let x=T.expandShapeToKeepDim(d,l);y=be({inputs:{x:g},backend:n,attrs:{shape:x}})}else y=be({inputs:{x:g},backend:n,attrs:{shape:d}});return n.disposeIntermediateTensorInfo(m),n.disposeIntermediateTensorInfo(g),c!=null&&n.disposeIntermediateTensorInfo(p),y}var Mie={kernelName:zo,backendName:"webgl",kernelFunc:Oie},zie=d9+`
|
|
return min(a, b);
|
|
`,Lie=`
|
|
vec4 result = vec4(min(a, b));
|
|
vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
|
|
`+v2+`
|
|
return result;
|
|
`,Bie=_n({opSnippet:zie,packedOpSnippet:Lie,cpuKernelImpl:_te}),Wie={kernelName:Lo,backendName:"webgl",kernelFunc:Bie},Vie=class{constructor(e,t,n){this.variableNames=["x"],this.outputShape=t.map((u,c)=>u[0]+e[c]+u[1]);let s=e.length,r=wt(s),a=t.map(u=>u[0]).join(","),o=t.map((u,c)=>u[0]+e[c]).join(","),i=["coords[0]","coords[1]","coords[2]","coords[3]"].slice(0,s),l=n==="reflect"?0:1;if(s===1){this.userCode=`
|
|
int start = ${a};
|
|
int end = ${o};
|
|
|
|
void main() {
|
|
int outC = getOutputCoords();
|
|
if (outC < start) {
|
|
outC = start * 2 - outC - ${l};
|
|
} else if(outC >= end) {
|
|
outC = (end - 1) * 2 - outC + ${l};
|
|
}
|
|
setOutput(getX(outC - start));
|
|
}
|
|
`;return}this.userCode=`
|
|
${r} start = ${r}(${a});
|
|
${r} end = ${r}(${o});
|
|
|
|
void main() {
|
|
${r} outC = getOutputCoords();
|
|
for (int i = 0; i < ${s}; i++) {
|
|
if (outC[i] < start[i]) {
|
|
outC[i] = start[i] * 2 - outC[i] - ${l};
|
|
} else if(outC[i] >= end[i]) {
|
|
outC[i] = (end[i] - 1) * 2 - outC[i] + ${l};
|
|
}
|
|
}
|
|
${r} coords = outC - start;
|
|
setOutput(getX(${i}));
|
|
}
|
|
`}},Uie=class{constructor(e,t,n){this.variableNames=["x"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=t.map((h,f)=>h[0]+e[f]+h[1]);let s=e.length,r=wt(s),a=t.map(h=>h[0]).join(","),o=t.map((h,f)=>h[0]+e[f]).join(","),i=es("rc",s),l=es("source",s),u=`${i[s-1]} < ${this.outputShape[s-1]}`,c=s===1?"source":`vec2(${l.slice(-2).join()})`,p=n==="reflect"?0:1,d="";if(s===1){let h=`
|
|
${r} source = rc;
|
|
if (source < start) {
|
|
source = start * 2 - source - ${p};
|
|
} else if (source >= end) {
|
|
source = (end - 1) * 2 - source + ${p};
|
|
}
|
|
source -= start;
|
|
`;d=`
|
|
${r} rc = outputLoc;
|
|
${h}
|
|
result[0] = getChannel(getX(${l.join()}), ${c});
|
|
${i[s-1]} += 1;
|
|
if(${u}) {
|
|
${h}
|
|
result[1] = getChannel(getX(${l.join()}), ${c});
|
|
}
|
|
`}else{let h=`
|
|
${r} source = rc;
|
|
${r} lt = ${r}(lessThan(source, start));
|
|
${r} gte = ${r}(greaterThanEqual(source, end));
|
|
${r} orig = 1 - (lt + gte);
|
|
source = orig * source +
|
|
lt * (start * 2 - source - ${p}) +
|
|
gte * ((end - 1) * 2 - source + ${p});
|
|
source -= start;
|
|
`;d=`
|
|
${r} rc = outputLoc;
|
|
${h}
|
|
result[0] = getChannel(getX(${l.join()}), ${c});
|
|
${i[s-1]} += 1;
|
|
if(${u}) {
|
|
${h}
|
|
result[1] = getChannel(getX(${l.join()}), ${c});
|
|
}
|
|
rc = outputLoc;
|
|
${i[s-2]} += 1;
|
|
if(${i[s-2]} < ${this.outputShape[s-2]}) {
|
|
${h}
|
|
result[2] = getChannel(getX(${l.join()}), ${c});
|
|
${i[s-1]} += 1;
|
|
if(${u}) {
|
|
${h}
|
|
result[3] = getChannel(getX(${l.join()}), ${c});
|
|
}
|
|
}
|
|
`}this.userCode=`
|
|
const ${r} start = ${r}(${a});
|
|
const ${r} end = ${r}(${o});
|
|
|
|
void main() {
|
|
${r} outputLoc = getOutputCoords();
|
|
vec4 result = vec4(0.);
|
|
${d}
|
|
setOutput(result);
|
|
}
|
|
`}},Gie=({inputs:e,backend:t,attrs:n})=>{let{x:s}=e,{paddings:r,mode:a}=n,o=q().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new Uie(s.shape,r,a):new Vie(s.shape,r,a);return t.runWebGLProgram(o,[s],s.dtype)},Hie={kernelName:Bo,backendName:"webgl",kernelFunc:Gie},jie=`if (b == 0.0) return NAN;
|
|
return mod(a, b);`,qie=`
|
|
vec4 result = mod(a, b);
|
|
vec4 isNaN = vec4(equal(b, vec4(0.0)));
|
|
`+v2+`
|
|
return result;
|
|
`,Xie=_n({opSnippet:jie,packedOpSnippet:qie}),Kie={kernelName:Ec,backendName:"webgl",kernelFunc:Xie},Zie=class{constructor(e,t,n){this.variableNames=["probs"],this.customUniforms=[{name:"seed",type:"float"}],this.outputShape=[e,n],this.userCode=`
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
|
|
float r = random(seed);
|
|
float cdf = 0.0;
|
|
|
|
for (int i = 0; i < ${t-1}; i++) {
|
|
cdf += getProbs(batch, i);
|
|
|
|
if (r < cdf) {
|
|
setOutput(float(i));
|
|
return;
|
|
}
|
|
}
|
|
|
|
// If no other event happened, last event happened.
|
|
setOutput(float(${t-1}));
|
|
}
|
|
`}},Yie=`
|
|
if (a == b) {
|
|
return 1.0;
|
|
};
|
|
return a / b;`,Jie=`
|
|
// vec4 one = vec4(equal(a, b));
|
|
// return one + (vec4(1.0) - one) * a / b;
|
|
vec4 result = a / b;
|
|
if(a.x == b.x) {
|
|
result.x = 1.;
|
|
}
|
|
if(a.y == b.y) {
|
|
result.y = 1.;
|
|
}
|
|
if(a.z == b.z) {
|
|
result.z = 1.;
|
|
}
|
|
if(a.w == b.w) {
|
|
result.w = 1.;
|
|
}
|
|
|
|
return result;
|
|
`,$9=_n({opSnippet:Yie,packedOpSnippet:Jie,checkOutOfBounds:!0}),Qie={kernelName:Io,backendName:"webgl",kernelFunc:$9},$7="return a - b;",P9=_n({opSnippet:$7,packedOpSnippet:$7,supportsComplex:!0,cpuKernelImpl:Xte}),ele={kernelName:si,backendName:"webgl",kernelFunc:P9};function F9(e){let{inputs:t,backend:n,attrs:s}=e,{logits:r}=t,{dim:a}=s,o=v.parseAxisParam([a],r.shape),i=D9({inputs:{x:r},backend:n,attrs:{reductionIndices:o,keepDims:!1}}),l=T.expandShapeToKeepDim(i.shape,o),u=be({inputs:{x:i},backend:n,attrs:{shape:l}}),c=P9({inputs:{a:r,b:u},backend:n}),p=E9({inputs:{x:c},backend:n}),d=I2({inputs:{x:p},backend:n,attrs:{axis:o,keepDims:!1}}),h=be({inputs:{x:d},backend:n,attrs:{shape:l}}),f=$9({inputs:{a:p,b:h},backend:n});return n.disposeIntermediateTensorInfo(i),n.disposeIntermediateTensorInfo(u),n.disposeIntermediateTensorInfo(c),n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),f}var tle={kernelName:ti,backendName:"webgl",kernelFunc:F9};function nle(e){let{inputs:t,backend:n,attrs:s}=e,{logits:r}=t,{numSamples:a,seed:o,normalized:i}=s,l=i?r:F9({inputs:{logits:r},backend:n,attrs:{dim:r.shape.length-1}}),u=l.shape[0],c=l.shape[1],p=new Zie(u,c,a),d=[[o]],h=n.runWebGLProgram(p,[l],"int32",d);return i||n.disposeIntermediateTensorInfo(l),h}var sle={kernelName:u0,backendName:"webgl",kernelFunc:nle},rle=fr+`
|
|
return -x;
|
|
`,ale=`
|
|
vec4 result = -x;
|
|
bvec4 isNaN = isnan(x);
|
|
|
|
result.r = isNaN.r ? x.r : result.r;
|
|
result.g = isNaN.g ? x.g : result.g;
|
|
result.b = isNaN.b ? x.b : result.b;
|
|
result.a = isNaN.a ? x.a : result.a;
|
|
|
|
return result;
|
|
`;function ole(e){let{inputs:t,backend:n}=e,{x:s}=t;if(n.shouldExecuteOnCPU([s])){let a=n.texData.get(s.dataId),[o,i]=$te(a.values,s.shape,s.dtype);return n.makeTensorInfo(i,s.dtype,o)}let r;return q().getBool("WEBGL_PACK_UNARY_OPERATIONS")?r=new Ui(s.shape,ale):r=new pa(s.shape,rle),n.runWebGLProgram(r,[s],s.dtype)}var ile={kernelName:Il,backendName:"webgl",kernelFunc:ole},lle=pr.nonMaxSuppressionV3Impl;function ule(e){T.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");let{inputs:t,backend:n,attrs:s}=e,{boxes:r,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l}=s,u=n.readSync(r.dataId),c=n.readSync(a.dataId),{selectedIndices:p}=lle(u,c,o,i,l);return n.makeTensorInfo([p.length],"int32",new Int32Array(p))}var cle={kernelName:Cl,backendName:"webgl",kernelFunc:ule},dle=pr.nonMaxSuppressionV4Impl;function ple(e){T.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");let{inputs:t,backend:n,attrs:s}=e,{boxes:r,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,padToMaxOutputSize:u}=s,c=n.readSync(r.dataId),p=n.readSync(a.dataId),{selectedIndices:d,validOutputs:h}=dle(c,p,o,i,l,u);return[n.makeTensorInfo([d.length],"int32",new Int32Array(d)),n.makeTensorInfo([],"int32",new Int32Array([h]))]}var hle={kernelName:Rc,backendName:"webgl",kernelFunc:ple},fle=pr.nonMaxSuppressionV5Impl;function mle(e){T.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");let{inputs:t,backend:n,attrs:s}=e,{boxes:r,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,softNmsSigma:u}=s,c=n.readSync(r.dataId),p=n.readSync(a.dataId),d=o,h=i,f=l,m=u,{selectedIndices:g,selectedScores:y}=fle(c,p,d,h,f,m);return[n.makeTensorInfo([g.length],"int32",new Int32Array(g)),n.makeTensorInfo([y.length],"float32",new Float32Array(y))]}var gle={kernelName:Tl,backendName:"webgl",kernelFunc:mle},yle=class{constructor(e,t,n,s){this.variableNames=["indices"],this.outputShape=[e,t],this.userCode=`
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int index = round(getIndices(coords.x));
|
|
setOutput(mix(float(${s}), float(${n}),
|
|
float(index == coords.y)));
|
|
}
|
|
`}},Ale=e=>{let{inputs:t,backend:n,attrs:s}=e,{indices:r}=t,{depth:a,onValue:o,offValue:i}=s,l=v.sizeFromShape(r.shape),u=new yle(l,a,o,i),c=be({inputs:{x:r},backend:n,attrs:{shape:[l]}}),p=n.runWebGLProgram(u,[c],r.dtype);n.disposeIntermediateTensorInfo(c);let d=[...r.shape,a],h=be({inputs:{x:p},backend:n,attrs:{shape:d}});return n.disposeIntermediateTensorInfo(p),h},xle={kernelName:El,backendName:"webgl",kernelFunc:Ale};function Mm(e){let{inputs:t,backend:n}=e,{x:s}=t;if(s.dtype==="complex64"){let r=zh({inputs:{input:s},backend:n}),a=Mm({inputs:{x:r},backend:n}),o=S2({inputs:{input:s},backend:n}),i=Mm({inputs:{x:o},backend:n}),l=pi({inputs:{real:a,imag:i},backend:n});return n.disposeIntermediateTensorInfo(r),n.disposeIntermediateTensorInfo(a),n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}else return Lh({attrs:{shape:s.shape,dtype:s.dtype,value:s.dtype==="string"?"":0},backend:n})}var ble={kernelName:Hl,backendName:"webgl",kernelFunc:Mm};function O9(e){let{inputs:t,backend:n}=e,{x:s}=t;if(s.dtype==="string")throw new Error("onesLike is not supported under string dtype");if(s.dtype==="complex64"){let r=zh({inputs:{input:s},backend:n}),a=O9({inputs:{x:r},backend:n}),o=S2({inputs:{input:s},backend:n}),i=Mm({inputs:{x:o},backend:n}),l=pi({inputs:{real:a,imag:i},backend:n});return n.disposeIntermediateTensorInfo(r),n.disposeIntermediateTensorInfo(a),n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}else return Lh({attrs:{shape:s.shape,dtype:s.dtype,value:1},backend:n})}var vle={kernelName:Nl,backendName:"webgl",kernelFunc:O9};function wle(e){let{inputs:t,backend:n,attrs:s}=e,{axis:r}=s;if(t.length===1)return ly({inputs:{input:t[0]},backend:n,attrs:{dim:r}});let a=t[0].shape,o=t[0].dtype;t.forEach(c=>{v.assertShapesMatch(a,c.shape,"All tensors passed to stack must have matching shapes"),v.assert(o===c.dtype,()=>"All tensors passed to stack must have matching dtypes")});let i=[],l=t.map(c=>{let p=ly({inputs:{input:c},backend:n,attrs:{dim:r}});return i.push(p),p}),u=w9({inputs:l,backend:n,attrs:{axis:r}});return i.forEach(c=>n.disposeIntermediateTensorInfo(c)),u}var kle={kernelName:Rl,backendName:"webgl",kernelFunc:wle},Ile=class{constructor(e,t,n){this.variableNames=["x"],this.customUniforms=[{name:"value",type:"float"}],this.outputShape=t.map((l,u)=>l[0]+e[u]+l[1]);let s=e.length,r=wt(s),a=t.map(l=>l[0]).join(","),o=t.map((l,u)=>l[0]+e[u]).join(","),i=["coords[0]","coords[1]","coords[2]","coords[3]"].slice(0,s);if(s===1){this.userCode=`
|
|
int start = ${a};
|
|
int end = ${o};
|
|
|
|
void main() {
|
|
int outC = getOutputCoords();
|
|
if (outC < start || outC >= end) {
|
|
setOutput(value);
|
|
} else {
|
|
setOutput(getX(outC - start));
|
|
}
|
|
}
|
|
`;return}this.userCode=`
|
|
${r} start = ${r}(${a});
|
|
${r} end = ${r}(${o});
|
|
|
|
void main() {
|
|
${r} outC = getOutputCoords();
|
|
if (any(lessThan(outC, start)) || any(greaterThanEqual(outC, end))) {
|
|
setOutput(value);
|
|
} else {
|
|
${r} coords = outC - start;
|
|
setOutput(getX(${i}));
|
|
}
|
|
}
|
|
`}},Sle=class{constructor(e,t,n){this.variableNames=["x"],this.packedInputs=!0,this.packedOutput=!0,this.customUniforms=[{name:"value",type:"float"}],this.outputShape=t.map((f,m)=>f[0]+e[m]+f[1]);let s=e.length,r=wt(s),a=t.map(f=>f[0]).join(","),o=t.map((f,m)=>f[0]+e[m]).join(","),i=es("rc",s),l=es("source",s),u=`${i[s-1]} < ${this.outputShape[s-1]}`,c=s===1?"source":`vec2(${l.slice(-2).join()})`,p=[`${r} rc = outputLoc;`,`${i[s-1]} += 1;
|
|
if(${u}) {
|
|
`,s===1?"":`}
|
|
rc = outputLoc;
|
|
${i[s-2]} += 1;
|
|
if(${i[s-2]} < ${this.outputShape[s-2]}) {`,s===1?"":` ${i[s-1]} += 1;
|
|
if(${u}) {`],d=s===1?"rc < start || rc >= end":"any(lessThan(rc, start)) || any(greaterThanEqual(rc, end))",h="";for(let f=0,m=s===1?2:4;f<m;f++)h+=`
|
|
${p[f]}
|
|
if (${d}) {
|
|
result[${f}] = float(value);
|
|
} else {
|
|
${r} source = rc - start;
|
|
result[${f}] = getChannel(getX(${l.join()}), ${c});
|
|
}
|
|
`;h+=s===1?"} ":"}}",this.userCode=`
|
|
const ${r} start = ${r}(${a});
|
|
const ${r} end = ${r}(${o});
|
|
|
|
void main() {
|
|
${r} outputLoc = getOutputCoords();
|
|
vec4 result = vec4(0.);
|
|
${h}
|
|
setOutput(result);
|
|
}
|
|
`}},M9=e=>{let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{paddings:a,constantValue:o}=s;if(v.sizeFromShape(r.shape)===0){let u=a.map((c,p)=>c[0]+r.shape[p]+c[1]);return Lh({backend:n,attrs:{shape:u,value:o,dtype:r.dtype}})}let i=q().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new Sle(r.shape,a,o):new Ile(r.shape,a,o),l=[[o]];return n.runWebGLProgram(i,[r],r.dtype,l)},Cle={kernelName:Vo,backendName:"webgl",kernelFunc:M9},Tle=`
|
|
if(a < 0.0 && floor(b) < b){
|
|
return NAN;
|
|
}
|
|
if (b == 0.0) {
|
|
return 1.0;
|
|
}
|
|
return (round(mod(b, 2.0)) != 1) ?
|
|
pow(abs(a), b) : sign(a) * pow(abs(a), b);
|
|
`,Nle=`
|
|
// isModRound1 has 1 for components with round(mod(b, 2.0)) == 1, 0 otherwise.
|
|
vec4 isModRound1 = vec4(equal(round(mod(b, 2.0)), ivec4(1)));
|
|
vec4 multiplier = sign(a) * isModRound1 + (vec4(1.0) - isModRound1);
|
|
vec4 result = multiplier * pow(abs(a), b);
|
|
|
|
// Ensure that a^0 = 1, including 0^0 = 1 as this correspond to TF and JS
|
|
bvec4 isExpZero = equal(b, vec4(0.0));
|
|
result.r = isExpZero.r ? 1.0 : result.r;
|
|
result.g = isExpZero.g ? 1.0 : result.g;
|
|
result.b = isExpZero.b ? 1.0 : result.b;
|
|
result.a = isExpZero.a ? 1.0 : result.a;
|
|
|
|
vec4 isNaN = vec4(lessThan(a, vec4(0.0))) * vec4(lessThan(floor(b), b));
|
|
`+v2+`
|
|
return result;
|
|
`,Ele=_n({opSnippet:Tle,packedOpSnippet:Nle}),Rle={kernelName:Uo,backendName:"webgl",kernelFunc:Ele};function _le(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s,i=r.shape.length,l=[],u=v.parseAxisParam(a,r.shape),c=u,p=T.getAxesPermutation(c,i),d=r;p!=null&&(d=ts({inputs:{x:r},backend:n,attrs:{perm:p}}),c=T.getInnerMostAxes(c.length,i),l.push(d)),T.assertAxesAreInnerMostDims("prod",c,i);let h;if(n.shouldExecuteOnCPU([d])){let f=n.texData.get(d.dataId).values,{outVals:m,outShape:g,outDtype:y}=Fte(d.shape,d.dtype,f,c);h=n.makeTensorInfo(g,y,m)}else{let[f,m]=T.computeOutAndReduceShapes(d.shape,c),g=v.sizeFromShape(m),y=be({inputs:{x:d},backend:n,attrs:{shape:[-1,g]}}),x=sh(r.dtype),A=au(y,x,"prod",n);h=be({inputs:{x:A},backend:n,attrs:{shape:f}}),l.push(y),l.push(A)}if(o){l.push(h);let f=T.expandShapeToKeepDim(h.shape,u);h=be({inputs:{x:h},backend:n,attrs:{shape:f}})}return l.forEach(f=>n.disposeIntermediateTensorInfo(f)),h}var Dle={kernelName:Ho,backendName:"webgl",kernelFunc:_le},z9=e=>{let{backend:t,attrs:n}=e,{start:s,stop:r,step:a,dtype:o}=n,i=Ote(s,r,a,o);return t.makeTensorInfo([i.length],o,i)},$le={kernelName:_c,backendName:"webgl",kernelFunc:z9},Ple="return 1.0 / x;",Fle=dt({opSnippet:Ple}),Ole={kernelName:Dc,backendName:"webgl",kernelFunc:Fle},Mle=fr+`
|
|
return (x < 0.0) ? 0.0 : x;
|
|
`,zle=`
|
|
vec4 result = x * vec4(greaterThanEqual(x, vec4(0.0)));
|
|
bvec4 isNaN = isnan(x);
|
|
|
|
result.r = isNaN.r ? x.r : result.r;
|
|
result.g = isNaN.g ? x.g : result.g;
|
|
result.b = isNaN.b ? x.b : result.b;
|
|
result.a = isNaN.a ? x.a : result.a;
|
|
|
|
return result;
|
|
`,Lle=dt({opSnippet:Mle,packedOpSnippet:zle}),Ble={kernelName:jo,backendName:"webgl",kernelFunc:Lle},Wle=fr+`
|
|
return (x < 0.0) ? 0.0 : min(6.0, x);
|
|
`,Vle=`
|
|
vec4 result = min(x, vec4(6.)) * vec4(greaterThanEqual(x, vec4(0.0)));
|
|
bvec4 isNaN = isnan(x);
|
|
|
|
result.r = isNaN.r ? x.r : result.r;
|
|
result.g = isNaN.g ? x.g : result.g;
|
|
result.b = isNaN.b ? x.b : result.b;
|
|
result.a = isNaN.a ? x.a : result.a;
|
|
|
|
return result;
|
|
`,Ule=dt({opSnippet:Wle,packedOpSnippet:Vle}),Gle={kernelName:Ko,backendName:"webgl",kernelFunc:Ule},Hle=class{constructor(e,t,n,s,r){this.variableNames=["A"],this.outputShape=[];let[a,o,i,l]=e;this.outputShape=[a,t,n,l];let u=[s&&t>1?o-1:o,s&&n>1?i-1:i],c=[s&&t>1?t-1:t,s&&n>1?n-1:n],p;r?p="(vec2(yRC) + vec2(0.5)) * effectiveInputOverOutputRatioRC - vec2(0.5)":p="vec2(yRC) * effectiveInputOverOutputRatioRC",this.userCode=`
|
|
const vec2 effectiveInputOverOutputRatioRC = vec2(
|
|
${u[0]/c[0]},
|
|
${u[1]/c[1]});
|
|
const vec2 inputShapeRC = vec2(${o}.0, ${i}.0);
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int d = coords[3];
|
|
ivec2 yRC = coords.yz;
|
|
|
|
// Fractional source index.
|
|
vec2 sourceFracIndexRC = ${p};
|
|
|
|
// Compute the four integer indices.
|
|
ivec2 sourceFloorRC = ivec2(max(sourceFracIndexRC, vec2(0.0)));
|
|
ivec2 sourceCeilRC = ivec2(
|
|
min(inputShapeRC - 1.0, ceil(sourceFracIndexRC)));
|
|
|
|
float topLeft = getA(b, sourceFloorRC.x, sourceFloorRC.y, d);
|
|
float bottomLeft = getA(b, sourceCeilRC.x, sourceFloorRC.y, d);
|
|
float topRight = getA(b, sourceFloorRC.x, sourceCeilRC.y, d);
|
|
float bottomRight = getA(b, sourceCeilRC.x, sourceCeilRC.y, d);
|
|
|
|
vec2 fracRC = sourceFracIndexRC - vec2(sourceFloorRC);
|
|
|
|
float top = topLeft + (topRight - topLeft) * fracRC.y;
|
|
float bottom = bottomLeft + (bottomRight - bottomLeft) * fracRC.y;
|
|
float newValue = top + (bottom - top) * fracRC.x;
|
|
|
|
setOutput(newValue);
|
|
}
|
|
`}},jle=class{constructor(e,t,n,s,r){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=[];let[a,o,i,l]=e;this.outputShape=[a,t,n,l];let u=[s&&t>1?o-1:o,s&&n>1?i-1:i],c=[s&&t>1?t-1:t,s&&n>1?n-1:n],p;r?p="(vec3(yRC) + vec3(0.5)) * effectiveInputOverOutputRatioRC - vec3(0.5)":p="vec3(yRC) * effectiveInputOverOutputRatioRC",this.userCode=`
|
|
const vec3 effectiveInputOverOutputRatioRC = vec3(
|
|
${u[0]/c[0]},
|
|
${u[1]/c[1]},
|
|
${u[1]/c[1]});
|
|
const vec3 inputShapeRC = vec3(${o}.0, ${i}.0,
|
|
${i}.0);
|
|
|
|
float getAValue(int b, int r, int c, int d) {
|
|
return getChannel(getA(b, r, c, d), vec2(c, d));
|
|
}
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int d = coords[3];
|
|
// Calculate values for next column in yRC.z.
|
|
ivec3 yRC = coords.yzz + ivec3(0, 0, 1);
|
|
|
|
// Fractional source index.
|
|
vec3 sourceFracIndexRC = ${p};
|
|
|
|
// Compute the four integer indices.
|
|
ivec3 sourceFloorRC = ivec3(max(sourceFracIndexRC, vec3(0.0)));
|
|
ivec3 sourceCeilRC = ivec3(
|
|
min(inputShapeRC - 1.0, ceil(sourceFracIndexRC)));
|
|
|
|
// Should we calculate next column and row elements in 2x2 packed cell.
|
|
bool hasNextCol = d < ${l-1};
|
|
bool hasNextRow = coords.z < ${n-1};
|
|
|
|
// In parallel, construct four corners for all four components in
|
|
// packed 2x2 cell.
|
|
vec4 topLeft = vec4(
|
|
getAValue(b, sourceFloorRC.x, sourceFloorRC.y, d),
|
|
hasNextCol ? getAValue(b, sourceFloorRC.x, sourceFloorRC.y, d + 1)
|
|
: 0.0,
|
|
hasNextRow ? getAValue(b, sourceFloorRC.x, sourceFloorRC.z, d)
|
|
: 0.0,
|
|
(hasNextRow && hasNextCol) ?
|
|
getAValue(b, sourceFloorRC.x, sourceFloorRC.z, d + 1) : 0.0);
|
|
|
|
vec4 bottomLeft = vec4(
|
|
getAValue(b, sourceCeilRC.x, sourceFloorRC.y, d),
|
|
hasNextCol ? getAValue(b, sourceCeilRC.x, sourceFloorRC.y, d + 1)
|
|
: 0.0,
|
|
hasNextRow ? getAValue(b, sourceCeilRC.x, sourceFloorRC.z, d)
|
|
: 0.0,
|
|
(hasNextRow && hasNextCol) ?
|
|
getAValue(b, sourceCeilRC.x, sourceFloorRC.z, d + 1) : 0.0);
|
|
|
|
vec4 topRight = vec4(
|
|
getAValue(b, sourceFloorRC.x, sourceCeilRC.y, d),
|
|
hasNextCol ? getAValue(b, sourceFloorRC.x, sourceCeilRC.y, d + 1)
|
|
: 0.0,
|
|
hasNextRow ? getAValue(b, sourceFloorRC.x, sourceCeilRC.z, d)
|
|
: 0.0,
|
|
(hasNextRow && hasNextCol) ?
|
|
getAValue(b, sourceFloorRC.x, sourceCeilRC.z, d + 1) : 0.0);
|
|
|
|
vec4 bottomRight = vec4(
|
|
getAValue(b, sourceCeilRC.x, sourceCeilRC.y, d),
|
|
hasNextCol ? getAValue(b, sourceCeilRC.x, sourceCeilRC.y, d + 1)
|
|
: 0.0,
|
|
hasNextRow ? getAValue(b, sourceCeilRC.x, sourceCeilRC.z, d)
|
|
: 0.0,
|
|
(hasNextRow && hasNextCol) ?
|
|
getAValue(b, sourceCeilRC.x, sourceCeilRC.z, d + 1) : 0.0);
|
|
|
|
vec3 fracRC = sourceFracIndexRC - vec3(sourceFloorRC);
|
|
|
|
vec4 top = mix(topLeft, topRight, fracRC.yyzz);
|
|
vec4 bottom = mix(bottomLeft, bottomRight, fracRC.yyzz);
|
|
vec4 newValue = mix(top, bottom, fracRC.x);
|
|
|
|
setOutput(newValue);
|
|
}
|
|
`}};function qle(e){let{inputs:t,backend:n,attrs:s}=e,{images:r}=t,{alignCorners:a,halfPixelCenters:o,size:i}=s,[l,u]=i,c=q().getBool("WEBGL_PACK_IMAGE_OPERATIONS")?new jle(r.shape,l,u,a,o):new Hle(r.shape,l,u,a,o);return n.runWebGLProgram(c,[r],"float32")}var Xle={kernelName:Xo,backendName:"webgl",kernelFunc:qle},Kle=class{constructor(e,t,n){this.variableNames=["dy"],this.outputShape=[],this.outputShape=t;let[,s,r]=t,[,a,o]=e,i=[n&&a>1?s-1:s,n&&o>1?r-1:r],l=[n&&a>1?a-1:a,n&&o>1?o-1:o],u=i[0]/l[0],c=i[1]/l[1],p=1/u,d=1/c,h=Math.ceil(p)*2+2,f=Math.ceil(d)*2+2;this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int d = coords[3];
|
|
int r = coords[1];
|
|
int c = coords[2];
|
|
|
|
float accumulator = 0.0;
|
|
|
|
const float heightScale = float(${u});
|
|
const float widthScale = float(${c});
|
|
|
|
const float invHeightScale = float(${p});
|
|
const float invWidthScale = float(${d});
|
|
|
|
const int winHeight = int(${h});
|
|
const int winWidth = int(${f});
|
|
|
|
// Compute bounds for where in dy we will look
|
|
float startRLerp = floor(float(r) * invHeightScale);
|
|
int startDyR = int(startRLerp - float(winHeight / 2));
|
|
|
|
float startCLerp = floor(float(c) * invWidthScale);
|
|
int startDyC = int(startCLerp - float(winWidth / 2));
|
|
|
|
// Loop over dy
|
|
for (int dyROffset = 0; dyROffset < winHeight; dyROffset++) {
|
|
int dyR = dyROffset + startDyR;
|
|
|
|
// Guard against the window exceeding the bounds of dy
|
|
if (dyR < 0 || dyR >= ${a}) {
|
|
continue;
|
|
}
|
|
|
|
for (int dyCOffset = 0; dyCOffset < winWidth; dyCOffset++) {
|
|
int dyC = dyCOffset + startDyC;
|
|
|
|
// Guard against the window exceeding the bounds of dy
|
|
if (dyC < 0 || dyC >= ${o}) {
|
|
continue;
|
|
}
|
|
|
|
float dxR = float(dyR) * heightScale;
|
|
int topDxRIndex = int(floor(dxR));
|
|
int bottomDxRIndex = int(min(ceil(dxR), ${s-1}.0));
|
|
float dxRLerp = dxR - float(topDxRIndex);
|
|
float inverseDxRLerp = 1.0 - dxRLerp;
|
|
|
|
float dxC = float(dyC) * widthScale;
|
|
int leftDxCIndex = int(floor(dxC));
|
|
int rightDxCIndex = int(min(ceil(dxC), ${r-1}.0));
|
|
float dxCLerp = dxC - float(leftDxCIndex);
|
|
float inverseDxCLerp = 1.0 - dxCLerp;
|
|
|
|
if (r == topDxRIndex && c == leftDxCIndex) {
|
|
// topLeft
|
|
accumulator +=
|
|
getDy(b, dyR, dyC, d) * inverseDxRLerp * inverseDxCLerp;
|
|
}
|
|
|
|
if (r == topDxRIndex && c == rightDxCIndex) {
|
|
// topRight
|
|
accumulator += getDy(b, dyR, dyC, d) * inverseDxRLerp * dxCLerp;
|
|
}
|
|
|
|
if (r == bottomDxRIndex && c == leftDxCIndex) {
|
|
// bottomLeft
|
|
accumulator += getDy(b, dyR, dyC, d) * dxRLerp * inverseDxCLerp;
|
|
}
|
|
|
|
if (r == bottomDxRIndex && c == rightDxCIndex) {
|
|
// bottomRight
|
|
accumulator += getDy(b, dyR, dyC, d) * dxRLerp * dxCLerp;
|
|
}
|
|
}
|
|
}
|
|
// End loop over dy
|
|
|
|
setOutput(accumulator);
|
|
}
|
|
`}};function Zle(e){let{inputs:t,backend:n,attrs:s}=e,{images:r,dy:a}=t,{alignCorners:o}=s,i=new Kle(a.shape,r.shape,o);return n.runWebGLProgram(i,[a],a.dtype)}var Yle={kernelName:d0,backendName:"webgl",kernelFunc:Zle},Jle=class{constructor(e,t,n,s,r){this.variableNames=["A"],this.outputShape=[];let[a,o,i,l]=e;this.outputShape=[a,t,n,l];let u=[s&&t>1?o-1:o,s&&n>1?i-1:i],c=[s&&t>1?t-1:t,s&&n>1?n-1:n],p=s?"0.5":"0.0",d;r?d="max((vec2(yRC) + vec2(0.5)) * effectiveInputOverOutputRatioRC, vec2(0.0))":d="vec2(yRC) * effectiveInputOverOutputRatioRC",this.userCode=`
|
|
const vec2 effectiveInputOverOutputRatioRC = vec2(
|
|
${u[0]/c[0]},
|
|
${u[1]/c[1]});
|
|
const vec2 inputShapeRC = vec2(${o}.0, ${i}.0);
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int d = coords[3];
|
|
ivec2 yRC = coords.yz;
|
|
|
|
// Fractional source index.
|
|
vec2 sourceFracIndexRC = ${d};
|
|
|
|
// Compute the coordinators of nearest neighbor point.
|
|
ivec2 sourceNearestRC = ivec2(
|
|
min(inputShapeRC - 1.0, floor(sourceFracIndexRC + ${p})));
|
|
float newValue = getA(b, sourceNearestRC.x, sourceNearestRC.y, d);
|
|
|
|
setOutput(newValue);
|
|
}
|
|
`}},Qle=class{constructor(e,t,n,s,r){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=[];let[a,o,i,l]=e;this.outputShape=[a,t,n,l];let u=[s&&t>1?o-1:o,s&&n>1?i-1:i],c=[s&&t>1?t-1:t,s&&n>1?n-1:n],p=s?"0.5":"0.0",d;r?d="max((vec3(yRC) + vec3(0.5)) * effectiveInputOverOutputRatioRC, vec3(0.0))":d="vec3(yRC) * effectiveInputOverOutputRatioRC",this.userCode=`
|
|
const vec3 effectiveInputOverOutputRatioRC = vec3(
|
|
${u[0]/c[0]},
|
|
${u[1]/c[1]},
|
|
${u[1]/c[1]});
|
|
const vec3 inputShapeRC = vec3(${o}.0, ${i}.0,
|
|
${i}.0);
|
|
|
|
float getAValue(int b, int r, int c, int d) {
|
|
return getChannel(getA(b, r, c, d), vec2(c, d));
|
|
}
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int d = coords[3];
|
|
// Calculate values for next column in yRC.z.
|
|
ivec3 yRC = coords.yzz + ivec3(0, 0, 1);
|
|
|
|
// Fractional source index.
|
|
vec3 sourceFracIndexRC = ${d};
|
|
|
|
// Compute the coordinators of nearest neighbor point.
|
|
ivec3 sourceNearestRC = ivec3(
|
|
min(inputShapeRC - 1.0, floor(sourceFracIndexRC + ${p})));
|
|
|
|
// Should we calculate next column and row elements in 2x2 packed cell.
|
|
bool hasNextCol = d < ${l-1};
|
|
bool hasNextRow = coords.z < ${n-1};
|
|
|
|
vec4 newValue = vec4(
|
|
getAValue(b, sourceNearestRC.x, sourceNearestRC.y, d),
|
|
hasNextCol ? getAValue(b, sourceNearestRC.x, sourceNearestRC.y, d + 1)
|
|
: 0.0,
|
|
hasNextRow ? getAValue(b, sourceNearestRC.x, sourceNearestRC.z, d)
|
|
: 0.0,
|
|
(hasNextRow && hasNextCol) ?
|
|
getAValue(b, sourceNearestRC.x, sourceNearestRC.z, d + 1) : 0.0);
|
|
|
|
setOutput(newValue);
|
|
}
|
|
`}};function eue(e){let{inputs:t,backend:n,attrs:s}=e,{images:r}=t,{alignCorners:a,halfPixelCenters:o,size:i}=s,[l,u]=i,c=q().getBool("WEBGL_PACK_IMAGE_OPERATIONS")?new Qle(r.shape,l,u,a,o):new Jle(r.shape,l,u,a,o);return n.runWebGLProgram(c,[r],r.dtype)}var tue={kernelName:qo,backendName:"webgl",kernelFunc:eue},nue=class{constructor(e,t,n){this.variableNames=["dy"],this.outputShape=[],this.outputShape=t;let[,s,r]=t,[,a,o]=e,i=[n&&a>1?s-1:s,n&&o>1?r-1:r],l=[n&&a>1?a-1:a,n&&o>1?o-1:o],u=i[0]/l[0],c=i[1]/l[1],p=1/u,d=1/c,h=Math.ceil(p)*2+2,f=Math.ceil(d)*2+2;this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int b = coords[0];
|
|
int d = coords[3];
|
|
int r = coords[1];
|
|
int c = coords[2];
|
|
|
|
float accumulator = 0.0;
|
|
|
|
const float heightScale = float(${u});
|
|
const float widthScale = float(${c});
|
|
|
|
const float invHeightScale = float(${p});
|
|
const float invWidthScale = float(${d});
|
|
|
|
const int winHeight = int(${h});
|
|
const int winWidth = int(${f});
|
|
|
|
// Compute bounds for where in dy we will look
|
|
float startRLerp = floor(float(r) * invHeightScale);
|
|
int startDyR = int(floor(startRLerp - float(winHeight / 2)));
|
|
|
|
float startCLerp = floor(float(c) * invWidthScale);
|
|
int startDyC = int(floor(startCLerp - float(winWidth / 2)));
|
|
|
|
// Loop over dy
|
|
for (int dyROffset = 0; dyROffset < winHeight; dyROffset++) {
|
|
int dyR = dyROffset + startDyR;
|
|
|
|
// Guard against the window exceeding the bounds of dy
|
|
if (dyR < 0 || dyR >= ${a}) {
|
|
continue;
|
|
}
|
|
|
|
for (int dyCOffset = 0; dyCOffset < winWidth; dyCOffset++) {
|
|
int dyC = dyCOffset + startDyC;
|
|
|
|
// Guard against the window exceeding the bounds of dy
|
|
if (dyC < 0 || dyC >= ${o}) {
|
|
continue;
|
|
}
|
|
|
|
float sourceFracRow =
|
|
float(${i[0]}) *
|
|
(float(dyR) / float(${l[0]}));
|
|
|
|
float sourceFracCol =
|
|
float(${i[1]}) *
|
|
(float(dyC) / float(${l[1]}));
|
|
|
|
int sourceNearestRow = int(min(
|
|
float(int(${s}) - 1),
|
|
${n} ? float(round(sourceFracRow)) :
|
|
float(floor(sourceFracRow))));
|
|
|
|
int sourceNearestCol = int(min(
|
|
float(int(${r}) - 1),
|
|
${n} ? float(round(sourceFracCol)) :
|
|
float(floor(sourceFracCol))));
|
|
|
|
if (r == sourceNearestRow && c == sourceNearestCol) {
|
|
accumulator += getDy(b, dyR, dyC, d);
|
|
}
|
|
}
|
|
}
|
|
// End loop over dy
|
|
|
|
setOutput(accumulator);
|
|
}
|
|
`}};function sue(e){let{inputs:t,backend:n,attrs:s}=e,{images:r,dy:a}=t,{alignCorners:o}=s,i=new nue(a.shape,r.shape,o);return n.runWebGLProgram(i,[a],a.dtype)}var rue={kernelName:c0,backendName:"webgl",kernelFunc:sue},aue=class{constructor(e,t){this.variableNames=["x"];let n=e.length;if(n>4)throw new Error(`WebGL backend: Reverse of rank-${n} tensor is not yet supported`);if(this.outputShape=e,n===1){this.userCode=`
|
|
void main() {
|
|
int coord = getOutputCoords();
|
|
setOutput(getX(${e[0]} - coord - 1));
|
|
}
|
|
`;return}let s=o=>t.indexOf(o)!==-1&&e[o]!==1?`${e[o]} - coords[${o}] - 1`:`coords[${o}]`,r=e.map((o,i)=>s(i)).join(","),a=wt(n);this.userCode=`
|
|
void main() {
|
|
${a} coords = getOutputCoords();
|
|
setOutput(getX(${r}));
|
|
}
|
|
`}},oue=class{constructor(e,t){this.variableNames=["x"],this.packedInputs=!0,this.packedOutput=!0;let n=e.length;if(n>4)throw new Error(`WebGL backend: Reverse of rank-${n} tensor is not yet supported`);this.outputShape=e;let s=es("rc",n),r=`${s[n-1]} + 1 < ${this.outputShape[n-1]}`,a=`${s[n-2]} + 1 < ${this.outputShape[n-2]}`,o=wt(n);n===1?this.userCode=`
|
|
void main(){
|
|
int rc = getOutputCoords();
|
|
vec4 result = vec4(0.);
|
|
result.r = getChannel(getX(${e[0]} - rc - 1),
|
|
${e[0]} - rc - 1);
|
|
if(${r}){
|
|
result.g = getChannel(getX(${e[0]} - (rc + 1) - 1),
|
|
${e[0]} - (rc + 1) - 1);
|
|
}
|
|
setOutput(result);
|
|
}
|
|
`:this.userCode=`
|
|
void main() {
|
|
${o} rc = getOutputCoords();
|
|
vec4 result = vec4(0.);
|
|
result.r = ${i(s.slice())};
|
|
if(${r}){
|
|
result.g = ${l(s.slice())};
|
|
}
|
|
if(${a}) {
|
|
result.b = ${u(s.slice())};
|
|
if(${r}) {
|
|
result.a = ${c(s.slice())};
|
|
}
|
|
}
|
|
setOutput(result);
|
|
}
|
|
`;function i(h){return p(h)}function l(h){return h[n-1]="("+h[n-1]+" + 1)",p(h)}function u(h){return h[n-2]="("+h[n-2]+" + 1)",p(h)}function c(h){return h[n-1]="("+h[n-1]+" + 1)",h[n-2]="("+h[n-2]+" + 1)",p(h)}function p(h){let f=e.map((y,x)=>d(x,h)),m=f.join(","),g=f.slice(-2).join(",");return`getChannel(getX(${m}), vec2(${g}))`}function d(h,f){return t.indexOf(h)!==-1&&e[h]!==1?`${e[h]} - ${f[h]} - 1`:`${f[h]}`}}};function iue(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{dims:a}=s,o=r.shape.length,i=v.parseAxisParam(a,r.shape);if(o===0)return Ps({inputs:{x:r},backend:n});let l=q().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new oue(r.shape,i):new aue(r.shape,i);return n.runWebGLProgram(l,[r],r.dtype)}var lue={kernelName:Dl,backendName:"webgl",kernelFunc:iue},uue=class{constructor(e,t){this.variableNames=["Image"],this.outputShape=[],this.customUniforms=[{name:"params",type:"vec4"}];let n=e[1],s=e[2];this.outputShape=e;let r="";typeof t=="number"?r=`float outputValue = ${t.toFixed(2)};`:r=`
|
|
vec3 fill = vec3(${t.join(",")});
|
|
float outputValue = fill[coords[3]];`,this.userCode=`
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
int x = coords[2];
|
|
int y = coords[1];
|
|
float coordXFloat = (float(x) - params[0]) * params[3] -
|
|
(float(y) - params[1]) * params[2];
|
|
float coordYFloat = (float(x) - params[0]) * params[2] +
|
|
(float(y) - params[1]) * params[3];
|
|
int coordX = int(round(coordXFloat + params[0]));
|
|
int coordY = int(round(coordYFloat + params[1]));
|
|
${r}
|
|
if(coordX >= 0 && coordX < ${s} && coordY >= 0 && coordY < ${n}) {
|
|
outputValue = getImage(coords[0], coordY, coordX, coords[3]);
|
|
}
|
|
setOutput(outputValue);
|
|
}
|
|
`}},cue={kernelName:jl,backendName:"webgl",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{image:s}=e,{radians:r,fillValue:a,center:o}=t,i=n,l=new uue(s.shape,a),[u,c]=T.getImageCenter(o,s.shape[1],s.shape[2]),p=[[u,c,Math.sin(r),Math.cos(r)]];return i.runWebGLProgram(l,[s],s.dtype,p)}},due=`
|
|
// OpenGL ES does not support round function.
|
|
// The algorithm is based on banker's rounding.
|
|
float base = floor(x);
|
|
if ((x - base) < 0.5) {
|
|
return floor(x);
|
|
} else if ((x - base) > 0.5) {
|
|
return ceil(x);
|
|
} else {
|
|
if (mod(base, 2.0) == 0.0) {
|
|
return base;
|
|
} else {
|
|
return base + 1.0;
|
|
}
|
|
}
|
|
`,pue=dt({opSnippet:due}),hue={kernelName:$l,backendName:"webgl",kernelFunc:pue},fue="return inversesqrt(x);",mue=dt({opSnippet:fue,cpuKernelImpl:Mte}),gue={kernelName:Zo,backendName:"webgl",kernelFunc:mue},L9=class{constructor(e,t,n,s,r,a,o=!0){this.variableNames=["updates","indices","defaultValue"],this.outputShape=a;let i=wt(r.length),l=wt(a.length),u="";n===1?u="i":n===2&&(u="i, j");let c=`getIndices(${u})`,p="";s===1?p="i":s===2&&(p="i, coords[1]");let d=`getUpdates(${p})`,h=t>1?"strides[j]":"strides";this.userCode=`
|
|
${i} strides = ${i}(${r});
|
|
|
|
void main() {
|
|
${l} coords = getOutputCoords();
|
|
float sum = 0.0;
|
|
bool found = false;
|
|
for (int i = 0; i < ${e}; i++) {
|
|
int flattenedIndex = 0;
|
|
for (int j = 0; j < ${t}; j++) {
|
|
int index = round(${c});
|
|
flattenedIndex += index * ${h};
|
|
}
|
|
if (flattenedIndex == coords[0]) {
|
|
sum += ${d};
|
|
found = true;
|
|
}
|
|
}
|
|
setOutput(mix(getDefaultValue(), sum, float(found)));
|
|
}
|
|
`}};function yue(e){let{inputs:t,backend:n,attrs:s}=e,{indices:r,updates:a}=t,{shape:o}=s,{sliceRank:i,numUpdates:l,sliceSize:u,strides:c,outputSize:p}=T.calculateShapes(a,r,o),d=[p/u,u];if(p===0)return n.makeTensorInfo(o,r.dtype);let h=be({inputs:{x:r},backend:n,attrs:{shape:[l,i]}}),f=be({inputs:{x:a},backend:n,attrs:{shape:[l,u]}}),m=n.makeTensorInfo([],"float32",new Float32Array([0])),g=new L9(l,i,h.shape.length,f.shape.length,c,d),y=n.runWebGLProgram(g,[f,h,m],f.dtype),x=be({inputs:{x:y},backend:n,attrs:{shape:o}});return n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(f),n.disposeIntermediateTensorInfo(y),n.disposeIntermediateTensorInfo(m),x}var Aue={kernelName:Pl,backendName:"webgl",kernelFunc:yue},xue=class{constructor(e,t,n,s){this.variableNames=["sortedSequence","values"],this.customUniforms=[{name:"numInputs",type:"int"}],this.outputShape=[e,n];let r="while (left < right) {",a=`for (int i = 0; i < ${Math.ceil(Math.log2(t+1))}; ++i) { if (left >= right) break;`,o=q().getNumber("WEBGL_VERSION")===2?r:a,i=s==="left"?"<":"<=";this.userCode=`
|
|
int findBound(int batch, float value) {
|
|
int left = 0;
|
|
int right = numInputs;
|
|
int mid;
|
|
${o}
|
|
mid = (left + right) / 2;
|
|
if (getSortedSequence(batch, mid) ${i} value) {
|
|
left = mid + 1;
|
|
} else {
|
|
right = mid;
|
|
}
|
|
}
|
|
return right;
|
|
}
|
|
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int valueIndex = coords[1];
|
|
|
|
float value = getValues(batch, valueIndex);
|
|
|
|
setOutput(float(findBound(batch, value)));
|
|
}
|
|
`}};function bue(e){let{inputs:t,backend:n,attrs:s}=e,{sortedSequence:r,values:a}=t,{side:o}=s,i=new xue(r.shape[0],r.shape[1],a.shape[1],o),l=[[r.shape[1]]];return n.runWebGLProgram(i,[r,a],"int32",l)}var vue={kernelName:p0,backendName:"webgl",kernelFunc:bue},wue=class{constructor(e,t,n){this.variableNames=["c","a","b"],this.outputShape=t;let s,r;if(n>4)throw Error(`Where for rank ${n} is not yet supported`);if(n===1)r="resRC",s="resRC";else{let o=["resRC.x","resRC.y","resRC.z","resRC.w"],i=[],l=[];for(let u=0;u<t.length;u++)l.push(`${o[u]}`),u<e&&i.push(`${o[u]}`);s=i.join(),r=l.join()}let a=wt(n);this.userCode=`
|
|
void main() {
|
|
${a} resRC = getOutputCoords();
|
|
float cVal = getC(${s});
|
|
if (cVal >= 1.0) {
|
|
setOutput(getA(${r}));
|
|
} else {
|
|
setOutput(getB(${r}));
|
|
}
|
|
}
|
|
`}};function kue(e){let{inputs:t,backend:n}=e,{condition:s,t:r,e:a}=t,o=new wue(s.shape.length,r.shape,r.shape.length);return n.runWebGLProgram(o,[s,r,a],Mn(r.dtype,a.dtype))}var Iue={kernelName:Fl,backendName:"webgl",kernelFunc:kue},Sue=`
|
|
// Stable and Attracting Fixed Point (0, 1) for Normalized Weights.
|
|
// see: https://arxiv.org/abs/1706.02515
|
|
float scaleAlpha = ${T.SELU_SCALEALPHA};
|
|
float scale = ${T.SELU_SCALE};
|
|
return (x >= 0.0) ? scale * x : scaleAlpha * (exp(x) - 1.0);
|
|
`,Cue=dt({opSnippet:Sue}),Tue={kernelName:$c,backendName:"webgl",kernelFunc:Cue},Nue=ld+`
|
|
return 1.0 / (1.0 + exp(-1.0 * x));
|
|
`,Eue=`
|
|
vec4 result = 1.0 / (1.0 + exp(-1.0 * x));
|
|
bvec4 isNaN = isnan(x);
|
|
|
|
result.r = isNaN.r ? x.r : result.r;
|
|
result.g = isNaN.g ? x.g : result.g;
|
|
result.b = isNaN.b ? x.b : result.b;
|
|
result.a = isNaN.a ? x.a : result.a;
|
|
|
|
return result;
|
|
`,Rue=dt({opSnippet:Nue,packedOpSnippet:Eue,cpuKernelImpl:Lte}),_ue={kernelName:Jo,backendName:"webgl",kernelFunc:Rue},Due=`
|
|
if (isnan(x)) { return 0.0; }
|
|
return sign(x);
|
|
`,$ue=dt({opSnippet:Due}),Pue={kernelName:Pc,backendName:"webgl",kernelFunc:$ue},Fue=ld+`
|
|
return sin(x);
|
|
`,Oue=dt({opSnippet:Fue}),Mue={kernelName:Yo,backendName:"webgl",kernelFunc:Oue},zue=`
|
|
float e2x = exp(x);
|
|
return (e2x - 1.0 / e2x) / 2.0;
|
|
`,Lue=dt({opSnippet:zue}),Bue={kernelName:Ml,backendName:"webgl",kernelFunc:Lue},Wue=`
|
|
float epsilon = 1.1920928955078125e-7;
|
|
float threshold = log(epsilon) + 2.0;
|
|
|
|
bool too_large = x > -threshold;
|
|
bool too_small = x < threshold;
|
|
|
|
float result;
|
|
float exp_x = exp(x);
|
|
|
|
if (too_large){
|
|
result = x;
|
|
}
|
|
else if (too_small){
|
|
result = exp_x;
|
|
}
|
|
else{
|
|
result = log(exp_x + 1.0);
|
|
}
|
|
return result;
|
|
`,Vue=dt({opSnippet:Wue}),Uue={kernelName:Fc,backendName:"webgl",kernelFunc:Vue},Gue=e=>{let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockShape:a,paddings:o}=s;v.assert(r.shape.length<=4,()=>"spaceToBatchND for rank > 4 with a WebGL backend not implemented yet");let i=a.reduce((y,x)=>y*x),l=[[0,0]];l.push(...o);for(let y=1+a.length;y<r.shape.length;++y)l.push([0,0]);let u=[],c=M9({inputs:{x:r},backend:n,attrs:{paddings:l,constantValue:0}}),p=T.getReshaped(c.shape,a,i,!1),d=T.getPermuted(p.length,a.length,!1),h=T.getReshapedPermuted(c.shape,a,i,!1),f=be({inputs:{x:c},backend:n,attrs:{shape:p}}),m=ts({inputs:{x:f},backend:n,attrs:{perm:d}}),g=be({inputs:{x:m},backend:n,attrs:{shape:h}});return u.push(c),u.push(f),u.push(m),u.forEach(y=>n.disposeIntermediateTensorInfo(y)),g},Hue={kernelName:zl,backendName:"webgl",kernelFunc:Gue};function jue(e){let{inputs:t,backend:n}=e,{indices:s,values:r,denseShape:a,defaultValue:o}=t;if(a.shape.length!==1)throw new Error(`Dense shape must be a vector, saw:
|
|
${a.shape}`);if(s.shape.length!==2)throw new Error(`Indices must be a matrix, saw:
|
|
${s.shape}`);if(r.shape.length!==1)throw new Error(`Values must be a vector, saw:
|
|
${r.shape}`);if(o.shape.length!==0)throw new Error(`Default value must be a scalar, saw:
|
|
${o.shape}`);let i=n.readSync(s.dataId),l=n.readSync(r.dataId),u=n.readSync(a.dataId),c=n.readSync(o.dataId)[0],[p,d,h,f,m]=Wte(i,s.shape,s.dtype,l,r.dtype,u,c);return[n.makeTensorInfo(d,s.dtype,p),n.makeTensorInfo([d[0]],r.dtype,h),n.makeTensorInfo([f.length],"bool",new Uint8Array(f.map(g=>Number(g)))),n.makeTensorInfo([m.length],s.dtype,new Int32Array(m))]}var que={kernelName:Xp,backendName:"webgl",kernelFunc:jue};function Xue(e){let{inputs:t,backend:n}=e,{inputIndices:s,inputShape:r,newShape:a}=t;if(s.shape.length!==2)throw new Error(`Input indices should be a matrix but received shape ${s.shape}`);if(r.shape.length!==1)throw new Error(`Input shape should be a vector but received shape ${r.shape}`);if(a.shape.length!==1)throw new Error(`Target shape should be a vector but received shape ${a.shape}`);let o=Array.from(n.readSync(r.dataId)),i=n.readSync(s.dataId),l=Array.from(n.readSync(a.dataId)),[u,c,p]=Vte(i,s.shape,s.dtype,o,l);return[n.makeTensorInfo(c,s.dtype,u),n.makeTensorInfo([p.length],a.dtype,new Int32Array(p))]}var Kue={kernelName:Oc,backendName:"webgl",kernelFunc:Xue};function Zue(e){let{inputs:t,backend:n}=e,{data:s,indices:r,segmentIds:a}=t;if(s.shape.length<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(r.shape.length!==1)throw new Error(`Indices should be a vector but received shape
|
|
${r.shape}`);if(a.shape.length!==1)throw new Error(`Segment ids should be a vector but received shape
|
|
${a.shape}`);let o=n.readSync(s.dataId),i=n.readSync(r.dataId),l=n.readSync(a.dataId),[u,c]=i9(o,s.shape,s.dtype,i,l,!0);return n.makeTensorInfo(c,s.dtype,u)}var Yue={kernelName:Kp,backendName:"webgl",kernelFunc:Zue};function Jue(e){let{inputs:t,backend:n}=e,{data:s,indices:r,segmentIds:a}=t;if(s.shape.length<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(r.shape.length!==1)throw new Error(`Indices should be a vector but received shape
|
|
${r.shape}`);if(a.shape.length!==1)throw new Error(`Segment ids should be a vector but received shape
|
|
${a.shape}`);let o=n.readSync(s.dataId),i=n.readSync(r.dataId),l=n.readSync(a.dataId),[u,c]=i9(o,s.shape,s.dtype,i,l);return n.makeTensorInfo(c,s.dtype,u)}var Que={kernelName:Zp,backendName:"webgl",kernelFunc:Jue};function ece(e){let{inputs:t,backend:n,attrs:s}=e,{sparseIndices:r,sparseValues:a,defaultValue:o}=t,{outputShape:i}=s,{sliceRank:l,numUpdates:u,sliceSize:c,strides:p,outputSize:d}=T.calculateShapes(a,r,i),h=!1;if(a.dtype==="string"){let y=n.bufferSync(r),x=n.bufferSync(a),A=v.decodeString(n.readSync(o.dataId)[0]),b=zte(y,x,i,d,c,u,l,p,A,h);return n.makeTensorInfo(i,b.dtype,b.values)}let f=new L9(u,l,r.shape.length,a.shape.length,p,[d,1],h),m=n.runWebGLProgram(f,[a,r,o],a.dtype),g=be({inputs:{x:m},backend:n,attrs:{shape:i}});return n.disposeIntermediateTensorInfo(m),g}var tce={kernelName:Yp,backendName:"webgl",kernelFunc:ece};function nce(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{numOrSizeSplits:a,axis:o}=s,i=v.parseAxisParam(o,r.shape)[0],l=T.prepareSplitSize(r,a,i),u=r.shape.length,c=new Array(u).fill(0),p=r.shape.slice();return l.map(d=>{let h=[...p];h[i]=d;let f=ud({inputs:{x:r},backend:n,attrs:{begin:c,size:h}});return c[i]+=d,f})}var sce={kernelName:Ll,backendName:"webgl",kernelFunc:nce},P7="return sqrt(x);",rce=dt({opSnippet:P7,packedOpSnippet:P7,cpuKernelImpl:Ute}),ace={kernelName:Qo,backendName:"webgl",kernelFunc:rce},oce="return x * x;",ice=dt({opSnippet:oce}),lce={kernelName:Mc,backendName:"webgl",kernelFunc:ice},F7="return (a - b) * (a - b);",uce=_n({opSnippet:F7,packedOpSnippet:F7}),cce={kernelName:ni,backendName:"webgl",kernelFunc:uce};function dce({inputs:e,attrs:t,backend:n}){let{x:s}=e,r=fr+`
|
|
return x > 0.0 ? 1.0 : float(${t.alpha});
|
|
`,a=new pa(s.shape,r);return n.runWebGLProgram(a,[s],s.dtype)}var pce={kernelName:ai,backendName:"webgl",kernelFunc:dce},hce=class{constructor(e,t,n){this.variableNames=["x"],this.outputShape=n;let s=n.length,r=wt(n.length),a=wt(n.length),o="";if(s===1)o="coords * strides + begin";else{let i=0;o=n.map((l,u)=>(i++,n.length===1?`coords * strides[${u}] + begin[${u}]`:`coords[${i-1}] * strides[${u}] + begin[${u}]`)).join(",")}this.userCode=`
|
|
${r} begin = ${r}(${e});
|
|
${r} strides = ${r}(${t});
|
|
|
|
void main() {
|
|
${a} coords = getOutputCoords();
|
|
setOutput(getX(${o}));
|
|
}
|
|
`}};function fce(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{begin:a,end:o,strides:i,beginMask:l,endMask:u,ellipsisMask:c,newAxisMask:p,shrinkAxisMask:d}=s,{finalShapeSparse:h,finalShape:f,isIdentity:m,sliceDim0:g,isSimpleSlice:y,begin:x,end:A,strides:b}=Ut.sliceInfo(r.shape,a,o,i,l,u,c,p,d),w;if(m)w=be({inputs:{x:r},backend:n,attrs:{shape:f}});else if(g||y){v.assert(r.shape.length>=1,()=>`Input must have rank at least 1, got: ${r.shape.length}`);let S=Ut.computeOutShape(x,A,b),E=ud({inputs:{x:r},backend:n,attrs:{begin:x,size:S}});w=be({inputs:{x:E},backend:n,attrs:{shape:f}}),n.disposeIntermediateTensorInfo(E)}else if(n.shouldExecuteOnCPU([r])){let E=n.readSync(r.dataId),R=Le(r.shape,r.dtype,E),$=Gte(h,R,b,x);w=n.makeTensorInfo(f,r.dtype,$.values)}else{let E=new hce(x,b,h);w=n.runWebGLProgram(E,[r],r.dtype)}let k=be({inputs:{x:w},backend:n,attrs:{shape:f}});return n.disposeIntermediateTensorInfo(w),k}var mce={kernelName:Bl,backendName:"webgl",kernelFunc:fce};function gce(e){let{inputs:t,backend:n,attrs:s}=e,{separator:r,nGramWidths:a,leftPad:o,rightPad:i,padWidth:l,preserveShortSequences:u}=s,{data:c,dataSplits:p}=t,d=n.readSync(c.dataId),h=n.readSync(p.dataId),[f,m]=Hte(d,h,r,a,o,i,l,u);return[n.makeTensorInfo([f.length],"string",f),n.makeTensorInfo(p.shape,"int32",m)]}var yce={kernelName:zc,backendName:"webgl",kernelFunc:gce};function Ace(e){let{inputs:t,backend:n,attrs:s}=e,{skipEmpty:r}=s,{input:a,delimiter:o}=t;if(a.dtype!=="string")throw new Error("Input must be of datatype string");if(a.shape.length!==1)throw new Error(`Input must be a vector, got shape: ${a.shape}`);if(o.shape.length!==0)throw new Error(`Delimiter must be a scalar, got shape: ${o.shape}`);let i=n.readSync(a.dataId),l=n.readSync(o.dataId)[0],[u,c,p]=jte(i,l,r),d=c.length;return[n.makeTensorInfo([d,2],"int32",u),n.makeTensorInfo([d],"string",c),n.makeTensorInfo([2],"int32",new Int32Array(p))]}var xce={kernelName:Jp,backendName:"webgl",kernelFunc:Ace};function bce(e){let{inputs:t,backend:n,attrs:s}=e,{numBuckets:r}=s,{input:a}=t;if(a.dtype!=="string")throw new Error("Input must be of datatype string");if(r<=0)throw new Error("Number of buckets must be at least 1");let o=n.readSync(a.dataId),i=qte(o,r);return n.makeTensorInfo(a.shape,"int32",i)}var vce={kernelName:Qp,backendName:"webgl",kernelFunc:bce},wce="return tan(x);",kce=dt({opSnippet:wce}),Ice={kernelName:Wl,backendName:"webgl",kernelFunc:kce},Sce=`
|
|
float e2x = exp(-2.0 * abs(x));
|
|
return sign(x) * (1.0 - e2x) / (1.0 + e2x);
|
|
`,Cce=dt({opSnippet:Sce}),Tce={kernelName:ri,backendName:"webgl",kernelFunc:Cce},Nce=class{constructor(e,t){this.variableNames=["A"];let n=new Array(e.length);for(let a=0;a<n.length;a++)n[a]=e[a]*t[a];this.outputShape=n,this.rank=n.length;let s=wt(this.rank),r=Ece(e);this.userCode=`
|
|
void main() {
|
|
${s} resRC = getOutputCoords();
|
|
setOutput(getA(${r}));
|
|
}
|
|
`}};function Ece(e){let t=e.length;if(t>5)throw Error(`Tile for rank ${t} is not yet supported`);if(t===1)return`imod(resRC, ${e[0]})`;let n=["resRC.x","resRC.y","resRC.z","resRC.w","resRC.u"],s=[];for(let r=0;r<e.length;r++)s.push(`imod(${n[r]}, ${e[r]})`);return s.join()}function B9(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{reps:a}=s;if(r.dtype==="string"||r.shape.length>5){let l=n.readSync(r.dataId),u=r.dtype==="string"?l.map(d=>v.decodeString(d)):l,c=Le(r.shape,r.dtype,u),p=Kte(c,a);return n.makeTensorInfo(p.shape,p.dtype,p.values)}let o=new Nce(r.shape,a);return n.runWebGLProgram(o,[r],r.dtype)}var Rce={kernelName:va,backendName:"webgl",kernelFunc:B9},_ce=class{constructor(e){this.variableNames=["x","indices"],this.customUniforms=[{name:"n",type:"int"},{name:"firstPass",type:"int"},{name:"negativeInf",type:"float"},{name:"dir",type:"int"},{name:"inc",type:"int"}],this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int elemIdx = coords[1];
|
|
|
|
// We compare elements pair-wise within a group of size 2 * inc.
|
|
// The comparing rule for each group alternates between ascending
|
|
// and descending. Within each group, we compare each pair at
|
|
// positions i and i+inc. To decide whether an element at position i
|
|
// is x0 or x1, we mod it by 2 * inc, if the result is smaller than
|
|
// inc, it is in the first half of the group, we denote it as x0,
|
|
// otherwise we denote it as x1.
|
|
// For example, as shown in the Bitonic top K paper referenced above,
|
|
// Figure5(a) shows that element[1] is in the
|
|
// second half of the group when group size is 2, but it is in the
|
|
// first half of the group when group size is 4.
|
|
|
|
bool isFirstInPair = imod(elemIdx, 2 * inc) < inc;
|
|
int i = isFirstInPair ? elemIdx : elemIdx - inc;
|
|
|
|
int i0 = firstPass == 1 ? i : int(getIndices(batch, i));
|
|
int i1 = firstPass == 1 ? i + inc : int(getIndices(batch, i + inc));
|
|
float x0 = i0 < n ? getX(batch, i0) : negativeInf;
|
|
float x1 = i1 < n ? getX(batch, i1) : negativeInf;
|
|
|
|
// Denotes which direction indices are in (ascending or descending).
|
|
bool reverse = imod(elemIdx, 2 * dir) >= dir;
|
|
bool isGreater = x0 > x1 || (x0 == x1 && i1 > i0);
|
|
if (reverse == isGreater) { // Elements in opposite order of direction
|
|
int iTemp = i0;
|
|
i0 = i1;
|
|
i1 = iTemp;
|
|
}
|
|
if (isFirstInPair) {
|
|
setOutput(float(i0));
|
|
} else {
|
|
setOutput(float(i1));
|
|
}
|
|
}
|
|
`}},Dce=class{constructor(e){this.variableNames=["x","indices"],this.customUniforms=[{name:"n",type:"int"},{name:"firstPass",type:"int"},{name:"k",type:"int"}],this.outputShape=e,this.userCode=`
|
|
void main() {
|
|
// Takes max of indices (0, k), (1, k + 1), (2, k + 2) ...
|
|
ivec2 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int elemIdx = coords[1];
|
|
|
|
// The output size is half of the previous size.
|
|
// If the previous sequence is | | | | _ _ _ _ | | | | _ _ _ _ (k=4),
|
|
// we only need to output the indices at positions |, the indices at
|
|
// positions _ can be thrown away, see Figure5(b) After Phase 2
|
|
// (Merge phase) in the Bitonic Top K paper referenced above.
|
|
// For example, the paper shows we only need to output the orange bars.
|
|
// The output sequence should look like this | | | | | | | |.
|
|
// Because the sequence is halved, to map the output index back
|
|
// to the previous sequence to find the corresponding value,
|
|
// we need to double the index. When we double the index,
|
|
// we basically interpolate a position, so 2i looks like
|
|
// | _ | _ | _ | _ | _ | _ | _. We move the | to the first k position
|
|
// of each 2k positions by - elemIdx % k. E.g. for output at
|
|
// index 4,5,6,7, we want to get the corresponding element at
|
|
// original index 8,9,10,11, for output at index 8,9,10,11,
|
|
// we want to get the corresponding element at original index
|
|
// 16,17,18,19, so on and so forth.
|
|
|
|
int i = elemIdx < k ? elemIdx : (elemIdx * 2 - imod(elemIdx, k));
|
|
int i0 = firstPass == 1 ? i : int(getIndices(batch, i));
|
|
int i1 = firstPass == 1 ? i + k : int(getIndices(batch, i + k));
|
|
|
|
float x0 = getX(batch, i0);
|
|
float x1 = i1 < n ? getX(batch, i1) : x0;
|
|
|
|
setOutput(x0 >= x1 ? float(i0) : float(i1));
|
|
}
|
|
`}};function $i(e,t){t!==null&&e.disposeIntermediateTensorInfo(t)}function O7(e){let t=1;for(;t<e;)t*=2;return t}function $ce(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{k:a,sorted:o}=s,i=q().getNumber("TOPK_LAST_DIM_CPU_HANDOFF_SIZE_THRESHOLD"),l=q().getNumber("TOPK_K_CPU_HANDOFF_THRESHOLD"),u=r.shape,c=u[u.length-1];if(n.shouldExecuteOnCPU([r])||c<i||a>l){let $=n.readSync(r.dataId),[_,P]=Zte($,u,r.dtype,a,o);return[n.makeTensorInfo(_.shape,_.dtype,_.values),n.makeTensorInfo(P.shape,P.dtype,P.values)]}if(a===0)return u[u.length-1]=0,[n.makeTensorInfo(u,r.dtype,[]),n.makeTensorInfo(u,"int32",[])];if(c===1)return[r,Lh({attrs:{shape:u,dtype:"int32",value:0},backend:n})];let p=n.texData.get(r.dataId),d=p!==null&&p.isPacked,h=d?n.unpackTensor(r):r,m=v.sizeFromShape(u)/c,g=be({inputs:{x:h},attrs:{shape:[m,c]},backend:n});d&&$i(n,h);let y=O7(a),x=O7(c),A=null,b=()=>A===null?[g,g]:[g,A],w=($,_,P)=>{let C=b(),F=new _ce(P),K=[[c],[A===null?1:0],[Number.NEGATIVE_INFINITY],[$],[_]],z=A;A=n.runWebGLProgram(F,C,"int32",K),$i(n,z)};for(let $=1;$<y;$*=2){let _=$*2;for(let P=$;P>=1;P/=2)w(_,P,[m,x])}for(let $=x;$>y;$/=2){let _=b(),P=new Dce([m,$/2]),F=[[c],[A===null?1:0],[y]],G=A;A=n.runWebGLProgram(P,_,"int32",F),$i(n,G);let K=y/2,z=K*2;for(let Z=K;Z>=1;Z/=2)w(z,Z,A.shape)}let k=A;A=ud({inputs:{x:A},backend:n,attrs:{begin:0,size:[m,a]}}),$i(n,k);let S=_9({inputs:{x:g,indices:A},backend:n,attrs:{axis:1,batchDims:1}});$i(n,g);let E=u.slice(0,-1);E.push(a),k=A,A=be({inputs:{x:A},attrs:{shape:E},backend:n}),$i(n,k);let R=S;return S=be({inputs:{x:S},attrs:{shape:E},backend:n}),$i(n,R),[S,A]}var Pce={kernelName:Vl,backendName:"webgl",kernelFunc:$ce},Fce=class{constructor(e,t,n,s,r,a){this.variableNames=["Image","Transforms"],this.outputShape=a;let o=n==="nearest"?1:2,i;switch(s){case"constant":i=1;break;case"reflect":i=2;break;case"wrap":i=3;break;case"nearest":i=4;break;default:i=1;break}this.userCode=`
|
|
float mapCoord(float outCoord, float len) {
|
|
float inCoord = outCoord;
|
|
if(${i} == 2) {
|
|
if (inCoord < 0.0) {
|
|
if (len <= 1.0) {
|
|
inCoord = 0.0;
|
|
} else {
|
|
float sz2 = 2.0 * len;
|
|
if (inCoord < sz2) {
|
|
inCoord = sz2 * float(int(float(-inCoord / sz2))) +
|
|
inCoord;
|
|
}
|
|
inCoord = inCoord < -len ? inCoord + sz2 : -inCoord - 1.0;
|
|
}
|
|
} else if (inCoord > len - 1.0) {
|
|
if (len <= 1.0) {
|
|
inCoord = 0.0;
|
|
} else {
|
|
float sz2 = 2.0 * len;
|
|
inCoord -= sz2 * float(int(float(inCoord / sz2)));
|
|
if (inCoord >= len) {
|
|
inCoord = sz2 - inCoord - 1.0;
|
|
}
|
|
}
|
|
}
|
|
return clamp(inCoord, 0.0, len - 1.0);
|
|
} else if (${i} == 3) {
|
|
if (inCoord < 0.0) {
|
|
if (len <= 1.0) {
|
|
inCoord = 0.0;
|
|
} else {
|
|
float sz = len - 1.0;
|
|
inCoord += len * (float(int(float(-inCoord / sz))) + 1.0);
|
|
}
|
|
} else if (inCoord > len - 1.0) {
|
|
if (len <= 1.0) {
|
|
inCoord = 0.0;
|
|
} else {
|
|
float sz = len - 1.0;
|
|
inCoord -= len * float(int(float(inCoord / sz)));
|
|
}
|
|
}
|
|
return clamp(inCoord, 0.0, len - 1.0);
|
|
} else if (${i} == 4) {
|
|
return clamp(outCoord, 0.0, len - 1.0);
|
|
} else {
|
|
return outCoord;
|
|
}
|
|
}
|
|
|
|
float readWithFillValue(int batch, int coordY, int coordX,
|
|
int channel) {
|
|
float outputValue;
|
|
if (0 <= coordY && coordY < ${e} && 0 <= coordX && coordX < ${t}) {
|
|
outputValue = getImage(batch, coordY, coordX, channel);
|
|
} else {
|
|
outputValue = float(${r});
|
|
}
|
|
return outputValue;
|
|
}
|
|
|
|
void main() {
|
|
ivec4 coords = getOutputCoords();
|
|
float outputValue;
|
|
int batch = coords[0];
|
|
int x = coords[2];
|
|
int y = coords[1];
|
|
int channel = coords[3];
|
|
float xf = float(x);
|
|
float yf = float(y);
|
|
float a1 = getTransforms(batch, 0);
|
|
float a2 = getTransforms(batch, 1);
|
|
float a3 = getTransforms(batch, 2);
|
|
float b1 = getTransforms(batch, 3);
|
|
float b2 = getTransforms(batch, 4);
|
|
float b3 = getTransforms(batch, 5);
|
|
float c1 = getTransforms(batch, 6);
|
|
float c2 = getTransforms(batch, 7);
|
|
float projection = c1 * xf + c2 * yf + 1.0;
|
|
if (projection == 0.0) {
|
|
outputValue = float(${r});
|
|
} else {
|
|
float inX = (a1 * xf + a2 * yf + a3) / projection;
|
|
float inY = (b1 * xf + b2 * yf + b3) / projection;
|
|
float mapX = mapCoord(inX, float(${t}));
|
|
float mapY = mapCoord(inY, float(${e}));
|
|
|
|
if (${o} == 1) {
|
|
int coordY = int(round(mapY));
|
|
int coordX = int(round(mapX));
|
|
outputValue = readWithFillValue(batch, coordY, coordX,
|
|
channel);
|
|
} else {
|
|
float yFloor = floor(mapY);
|
|
float xFloor = floor(mapX);
|
|
float yCeil = yFloor + 1.0;
|
|
float xCeil = xFloor + 1.0;
|
|
float valueYFloor = (xCeil - mapX) *
|
|
readWithFillValue(batch, int(yFloor), int(xFloor), channel) +
|
|
(mapX - xFloor) *
|
|
readWithFillValue(batch, int(yFloor), int(xCeil), channel);
|
|
float valueYCeil = (xCeil - mapX) *
|
|
readWithFillValue(batch, int(yCeil), int(xFloor), channel) +
|
|
(mapX - xFloor) *
|
|
readWithFillValue(batch, int(yCeil), int(xCeil), channel);
|
|
outputValue = (yCeil - mapY) * valueYFloor +
|
|
(mapY - yFloor) * valueYCeil;
|
|
}
|
|
}
|
|
setOutput(outputValue);
|
|
}
|
|
`}};function Oce(e){let{inputs:t,backend:n,attrs:s}=e,{image:r,transforms:a}=t,{interpolation:o,fillMode:i,fillValue:l,outputShape:u}=s,[c,p,d,h]=r.shape,[f,m]=u!=null?u:[p,d],g=[c,f,m,h],y=new Fce(p,d,o,i,l,g);return n.runWebGLProgram(y,[r,a],"float32")}var Mce={kernelName:Ul,backendName:"webgl",kernelFunc:Oce};function zce(e){let{inputs:t,attrs:n,backend:s}=e,{axis:r}=n,{x:a}=t;nd(a,"unique"),console.warn("WARNING: ","UI might be locked temporarily as data is being downloaded");let o=s.readSync(a.dataId),{outputValues:i,outputShape:l,indices:u}=Yte(o,r,a.shape,a.dtype);return[s.makeTensorInfo(l,a.dtype,i),s.makeTensorInfo([u.length],"int32",u)]}var Lce={kernelName:h0,backendName:"webgl",kernelFunc:zce};function Bce(e){let{inputs:t,backend:n,attrs:s}=e,{value:r}=t,{axis:a}=s;a<0&&(a+=r.shape.length);let o=r,i=o.shape.length,l=r.shape[a],u=new Array(i-1),c=0;for(let m=0;m<i;m++)m!==a&&(u[c++]=o.shape[m]);let p=[],d=new Array(i).fill(0),h=o.shape.slice();h[a]=1;let f=new Array(l);for(let m=0;m<f.length;m++){d[a]=m;let g=ud({inputs:{x:o},backend:n,attrs:{begin:d,size:h}}),y=be({inputs:{x:g},backend:n,attrs:{shape:u}});f[m]=y,p.push(g)}return p.forEach(m=>n.disposeIntermediateTensorInfo(m)),f}var Wce={kernelName:Gl,backendName:"webgl",kernelFunc:Bce},Vce=class{constructor(e,t){this.variableNames=["x","segmentIds"];let n=e.windowSize,s=e.batchSize,r=e.inSize,a=e.numSegments,o=a*Math.ceil(r/n);this.outputShape=[s,o];let i="0.0",l="sumValue",u=Math.floor(n/4)*4,c=n%4,p=`
|
|
sumValue += dot(values, segFilter);
|
|
`,d="";r%n>0&&(d=`
|
|
if (inIdx < 0 || inIdx >= ${r}) {
|
|
return initializationValue;
|
|
}
|
|
`);let h="";r%n>0&&(h=`
|
|
if (inIdx < 0 || inIdx >= ${r}) {
|
|
return -1.0;
|
|
}
|
|
`),this.userCode=`
|
|
const float initializationValue = ${i};
|
|
|
|
float getValue(int batch, int inIdx) {
|
|
${d}
|
|
return getX(batch, inIdx);
|
|
}
|
|
|
|
float getSegmentIdAtIndex(int inIdx) {
|
|
${h}
|
|
return getSegmentIds(inIdx);
|
|
}
|
|
|
|
void main() {
|
|
ivec2 coords = getOutputCoords();
|
|
int batch = coords[0];
|
|
int outIdx = coords[1];
|
|
int inOffset = int(floor(float(outIdx) / float(
|
|
${a})) * float(${n}));
|
|
int currentSeg = int(mod(float(outIdx), float(${a})));
|
|
|
|
float sumValue = 0.0;
|
|
|
|
for (int i = 0; i < ${u}; i += 4) {
|
|
int inIdx = inOffset + i;
|
|
vec4 values = vec4(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1),
|
|
getValue(batch, inIdx + 2),
|
|
getValue(batch, inIdx + 3)
|
|
);
|
|
|
|
vec4 segFilter = vec4(
|
|
int(getSegmentIdAtIndex(inIdx)) == currentSeg ? 1 : 0,
|
|
int(getSegmentIdAtIndex(inIdx + 1)) == currentSeg ? 1 : 0,
|
|
int(getSegmentIdAtIndex(inIdx + 2)) == currentSeg ? 1 : 0,
|
|
int(getSegmentIdAtIndex(inIdx + 3)) == currentSeg ? 1 : 0
|
|
);
|
|
|
|
${p}
|
|
}
|
|
|
|
int inIdx = inOffset + ${u};
|
|
if (${c===1}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, inIdx),
|
|
initializationValue,
|
|
initializationValue,
|
|
initializationValue
|
|
);
|
|
|
|
int inIdxSeg = int(getSegmentIdAtIndex(inIdx));
|
|
|
|
vec4 segFilter = vec4(
|
|
int(getSegmentIdAtIndex(inIdx)) == currentSeg ? 1 : 0,
|
|
0,
|
|
0,
|
|
0
|
|
);
|
|
|
|
${p}
|
|
} else if (${c===2}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1),
|
|
initializationValue,
|
|
initializationValue
|
|
);
|
|
|
|
vec4 segFilter = vec4(
|
|
int(getSegmentIdAtIndex(inIdx)) == currentSeg ? 1 : 0,
|
|
int(getSegmentIdAtIndex(inIdx + 1)) == currentSeg ? 1 : 0,
|
|
0,
|
|
0
|
|
);
|
|
|
|
${p}
|
|
} else if (${c===3}) {
|
|
vec4 values = vec4(
|
|
getValue(batch, inIdx),
|
|
getValue(batch, inIdx + 1),
|
|
getValue(batch, inIdx + 2),
|
|
initializationValue
|
|
);
|
|
|
|
vec4 segFilter = vec4(
|
|
int(getSegmentIdAtIndex(inIdx)) == currentSeg ? 1 : 0,
|
|
int(getSegmentIdAtIndex(inIdx + 1)) == currentSeg ? 1 : 0,
|
|
int(getSegmentIdAtIndex(inIdx + 2)) == currentSeg ? 1 : 0,
|
|
0
|
|
);
|
|
|
|
${p}
|
|
}
|
|
setOutput(${l});
|
|
}
|
|
`}};function Uce(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,segmentIds:a}=t,{numSegments:o}=s,i=r.shape.length,l=[],u=0,c=T.getAxesPermutation([u],i),p=r;c!=null&&(p=ts({inputs:{x:r},backend:n,attrs:{perm:c}}),l.push(p),u=T.getInnerMostAxes(1,i)[0]);let d=T.segment_util.computeOutShape(p.shape,u,o),h=v.sizeFromShape([p.shape[u]]),f=be({inputs:{x:p},backend:n,attrs:{shape:[-1,h]}});l.push(f);let m=sh(r.dtype),g=(b,w,k,S,E)=>{let R=b.shape[0],$=b.shape[1],_=T.segment_util.segOpComputeOptimalWindowSize($,E),P={windowSize:_,inSize:$,batchSize:R,numSegments:E},C=new Vce(P,w),F=n.compileAndRun(C,[b,k],S);if(l.push(F),F.shape[1]===E)return F;let G=z9({backend:n,attrs:{start:0,stop:E,step:1,dtype:"float32"}}),K=B9({inputs:{x:G},backend:n,attrs:{reps:[$/_]}});return l.push(G),l.push(K),g(F,w,K,S,E)},y=g(f,"unsortedSegmentSum",a,m,o),x=be({inputs:{x:y},backend:n,attrs:{shape:d}}),A=x;if(c!=null){l.push(x);let b=T.getUndoAxesPermutation(c);A=ts({inputs:{x:A},backend:n,attrs:{perm:b}})}return l.forEach(b=>n.disposeIntermediateTensorInfo(b)),A}var Gce={kernelName:eh,backendName:"webgl",kernelFunc:Uce},Hce=[jne,Xne,Yne,ese,nse,ase,ise,use,hse,mse,Ase,vse,Ise,Nse,_se,$se,Fse,Lse,Wse,Use,qse,ere,nre,rre,cre,pre,gre,Tne,xre,Ire,Nre,Pre,Ore,zre,Bre,Vre,Hre,Xre,Yre,Qre,tae,sae,oae,lae,pae,fae,yae,bae,wae,Cae,Rae,Pae,Mae,Bae,Wae,Uae,Hae,qae,Kae,Yae,toe,roe,ioe,uoe,poe,moe,xoe,koe,Cne,Soe,wre,Noe,_oe,Poe,Ene,zoe,Voe,Goe,Xoe,Yoe,tie,rie,lie,pie,mie,yie,vie,kie,Sie,Eie,_ie,$ie,Fie,Mie,Wie,Hie,Kie,sle,Pne,ile,cle,hle,gle,ore,xle,vle,kle,Cle,Rle,_ne,Dle,$le,ire,Qie,Ole,Ble,Gle,One,Xle,Yle,tue,rue,lue,cue,hue,gue,Aue,vue,Iue,Tue,_ue,Pue,Mue,Bue,Jse,tle,Uue,Hue,que,Kue,Yue,Que,tce,sce,ace,lce,cce,pce,mce,yce,xce,vce,ele,Une,Ice,Tce,Rce,Pce,Mce,Gne,Lce,Wce,Gce,ble];for(let e of Hce)dr(e);var Ht;(function(e){e[e.float32=0]="float32",e[e.int32=1]="int32",e[e.bool=2]="bool",e[e.string=3]="string",e[e.complex64=4]="complex64"})(Ht||(Ht={}));var Fp;(function(e){e[e.linear=0]="linear",e[e.relu=1]="relu",e[e.relu6=2]="relu6",e[e.prelu=3]="prelu",e[e.leakyrelu=4]="leakyrelu",e[e.sigmoid=5]="sigmoid",e[e.elu=6]="elu"})(Fp||(Fp={}));var W9;function jce(e){W9=e.wasm.cwrap(Ja,null,["number","array","number","number","array","number","number","number","number","number","number","number","number"])}function qce(e){let{inputs:t,backend:n,attrs:s}=e,{a:r,b:a,bias:o,preluActivationWeights:i}=t;if(r.dtype!=="float32"||a.dtype!=="float32")throw new Error("_FusedMatMul for non non-float32 tensors not yet supported.");let{transposeA:l,transposeB:u,activation:c,leakyreluAlpha:p}=s,d=n.dataIdMap.get(r.dataId).id,h=n.dataIdMap.get(a.dataId).id,f=0;if(o!=null){let E=n.dataIdMap.get(o.dataId);if(E.shape.length!==1)throw new Error(`_FusedMatMul only supports rank-1 bias but got rank ${E.shape.length}.`);f=E.id}let m=i==null?0:n.dataIdMap.get(i.dataId).id,g=Fp[c];if(g==null)throw new Error(`${c} activation not yet supported for FusedConv2D in the wasm backend.`);let y=l?r.shape[2]:r.shape[1],x=u?a.shape[1]:a.shape[2],A=Xl.assertAndGetBroadcastShape(r.shape.slice(0,-2),a.shape.slice(0,-2)),b=n.makeOutput([...A,y,x],r.dtype),w=n.dataIdMap.get(b.dataId).id,k=new Uint8Array(new Int32Array(r.shape).buffer),S=new Uint8Array(new Int32Array(a.shape).buffer);return W9(d,k,r.shape.length,h,S,a.shape.length,l,u,g,f,m,p||0,w),b}var Xce={kernelName:Ja,backendName:"wasm",setupFunc:jce,kernelFunc:qce};function In(e,t){let n;function s(a){n=a.wasm.cwrap(e,null,["number","number","number"])}function r(a){let{backend:o,inputs:{x:i}}=a,l=o.dataIdMap.get(i.dataId).id,u=o.makeOutput(i.shape,t||i.dtype),c=o.dataIdMap.get(u.dataId).id;return v.sizeFromShape(u.shape)===0||n(l,Ht[i.dtype],c),u}return{kernelName:e,backendName:"wasm",setupFunc:s,kernelFunc:r}}var Kce=In(il);function Dn(e,t,n){let s;function r(o){s=o.wasm.cwrap(e,null,["number","array","number","number","array","number","number","number"])}function a(o){let{backend:i,inputs:l}=o,{a:u,b:c}=l,p=i.dataIdMap.get(u.dataId).id,d=i.dataIdMap.get(c.dataId).id,h=n!=null?n:u.dtype,f=T.assertAndGetBroadcastShape(u.shape,c.shape),m=i.makeOutput(f,h);if(v.sizeFromShape(f)===0)return m;let g=new Uint8Array(new Int32Array(u.shape).buffer),y=new Uint8Array(new Int32Array(c.shape).buffer),x=i.dataIdMap.get(m.dataId).id;return(()=>s(p,g,u.shape.length,d,y,c.shape.length,Ht[u.dtype],x))(),m}return{kernelName:e,backendName:"wasm",setupFunc:r,kernelFunc:a}}var Zce=!0,Yce=Dn(xa,Zce),V9;function Jce(e){V9=e.wasm.cwrap(po,null,["array","number","number","number"])}function Qce(e){let{inputs:t,backend:n}=e,s=n.makeOutput(t[0].shape,t[0].dtype);if(v.sizeFromShape(s.shape)===0)return s;let r=t.map(i=>n.dataIdMap.get(i.dataId).id),a=new Uint8Array(new Int32Array(r).buffer),o=n.dataIdMap.get(s.dataId).id;return V9(a,r.length,Ht[s.dtype],o),s}var ede={kernelName:po,backendName:"wasm",setupFunc:Jce,kernelFunc:Qce};function C2(e){let{inputs:{x:t},backend:n}=e,s=n.makeOutput(t.shape,t.dtype),r=n.typedArrayFromHeap(t);return n.typedArrayFromHeap(s).set(r),s}var tde={kernelName:_o,backendName:"wasm",kernelFunc:C2},U9;function nde(e){U9=e.wasm.cwrap(Hr,null,["number","array","number","number","number","array","number"])}function lo(e){let{inputs:t,backend:n,attrs:s}=e,[r,a]=rde(t.x.shape,s.perm),o=!0;for(let f=0;f<a.length;f++)a[f]!==f&&(o=!1);let i=sde(t.x.shape,s.perm),l={dataId:t.x.dataId,shape:r,dtype:t.x.dtype};if(o){let f=C2({inputs:t,backend:n});return f.shape=i,f}let u=n.makeOutput(i,l.dtype),c=n.dataIdMap.get(l.dataId).id,p=n.dataIdMap.get(u.dataId).id,d=new Uint8Array(new Int32Array(a).buffer),h=new Uint8Array(new Int32Array(l.shape).buffer);return U9(c,h,l.shape.length,Ht[l.dtype],p,d,a.length),u}function sde(e,t){let n=new Array(e.length);for(let s=0;s<n.length;s++)n[s]=e[t[s]];return n}function rde(e,t){let n=[],s=[];for(let r=0;r<e.length;++r)e[r]!==1&&n.push(e[r]),e[t[r]]!==1&&s.push(t[r]);for(let r=0;r<s.length;++r){let a=-1;for(let o=0;o<s.length;++o)s[o]>=r&&(a===-1||s[a]>s[o])&&(a=o);s[a]=r}return[n,s]}var ade={kernelName:Hr,backendName:"wasm",kernelFunc:lo,setupFunc:nde};function hi(e,t,n){let s=e.shape,r=e.shape.length,a=v.parseAxisParam(t,s),o=a,i=T.getAxesPermutation(o,r),l=null,u=!1;if(i!=null){let c=new Array(r);for(let h=0;h<c.length;h++)c[h]=s[i[h]];o=T.getInnerMostAxes(o.length,r),l=lo({inputs:{x:e},attrs:{perm:i},backend:n});let p=n.dataIdMap.get(e.dataId).id;n.dataIdMap.get(l.dataId).id!==p&&(u=!0)}return{transposed:l,originalAxes:a,axes:o,inputWasTransposed:u}}var G9;function ode(e){G9=e.wasm.cwrap(fc,null,["number, number, number"])}function ide(e){let{backend:t,inputs:n,attrs:s}=e,{axis:r,keepDims:a}=s,{x:o}=n,l=t.dataIdMap.get(o.dataId).id,u=o,{transposed:c,axes:p,originalAxes:d,inputWasTransposed:h}=hi(o,r,t);if(h){let A=t.dataIdMap.get(c.dataId).id;u=c,l=A}let f=u.shape.length;T.assertAxesAreInnerMostDims("all",p,f);let[m,g]=T.computeOutAndReduceShapes(u.shape,p),y=v.sizeFromShape(g),x=t.makeOutput(m,o.dtype);if(v.sizeFromShape(u.shape)!==0){let A=t.dataIdMap.get(x.dataId).id;G9(l,y,A)}if(h&&t.disposeData(c.dataId),a){let A=T.expandShapeToKeepDim(x.shape,d);x.shape=A}return x}var lde={kernelName:fc,backendName:"wasm",setupFunc:ode,kernelFunc:ide},H9;function ude(e){H9=e.wasm.cwrap(mc,null,["number, number, number"])}function cde(e){let{backend:t,inputs:n,attrs:s}=e,{axis:r,keepDims:a}=s,{x:o}=n,l=t.dataIdMap.get(o.dataId).id,u=o,{transposed:c,axes:p,originalAxes:d,inputWasTransposed:h}=hi(o,r,t);if(h){let A=t.dataIdMap.get(c.dataId).id;u=c,l=A}let f=u.shape.length;T.assertAxesAreInnerMostDims("any",p,f);let[m,g]=T.computeOutAndReduceShapes(u.shape,p),y=v.sizeFromShape(g),x=t.makeOutput(m,o.dtype);if(v.sizeFromShape(u.shape)!==0){let A=t.dataIdMap.get(x.dataId).id;H9(l,y,A)}if(h&&t.disposeData(c.dataId),a){let A=T.expandShapeToKeepDim(x.shape,d);x.shape=A}return x}var dde={kernelName:mc,backendName:"wasm",setupFunc:ude,kernelFunc:cde},j9;function pde(e){j9=e.wasm.cwrap(ho,null,["number","number","number","number","number"])}function hde(e){let{backend:t,inputs:n,attrs:s}=e,{axis:r}=s,{x:a}=n,o=t.dataIdMap.get(a.dataId).id,i=o,l=a,{transposed:u,axes:c,inputWasTransposed:p}=hi(a,r,t);if(p){let y=t.dataIdMap.get(u.dataId).id;y!==o&&(l=u,i=y)}let d=l.shape.slice(0,-1),h=t.makeOutput(d,"int32"),f=t.dataIdMap.get(h.dataId).id,m=v.sizeFromShape(h.shape),g=l.shape[c[0]];return j9(i,Ht[l.dtype],m,g,f),p&&t.disposeData(u.dataId),h}var fde={kernelName:ho,backendName:"wasm",kernelFunc:hde,setupFunc:pde},q9;function mde(e){q9=e.wasm.cwrap(fo,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function gde(e){let{inputs:t,attrs:n,backend:s}=e,r=t.x,a=s.dataIdMap.get(r.dataId).id,{filterSize:o,strides:i,pad:l,dimRoundingMode:u}=n,c=T.computePool2DInfo(r.shape,o,i,1,l,u),p=c.filterHeight,d=c.filterWidth,h=c.padInfo.top,f=c.padInfo.right,m=c.padInfo.bottom,g=c.padInfo.left,y=c.strideHeight,x=c.strideWidth,A=c.inChannels;if(c.dataFormat!=="channelsLast")throw new Error(`wasm backend does not support dataFormat:'${c.dataFormat}'. Please use 'channelsLast'.`);if(c.dilationWidth!==1||c.dilationHeight!==1)throw new Error(`was backend only supports average pooling with dilation = [1, 1], got [${c.dilationHeight}, ${c.dilationWidth}].`);let b=s.makeOutput(c.outShape,"float32"),w=s.dataIdMap.get(b.dataId).id;return q9(a,r.shape[0],r.shape[1],r.shape[2],p,d,h,f,m,g,y,x,A,w),b}var yde={kernelName:fo,backendName:"wasm",setupFunc:mde,kernelFunc:gde};function hs(e){let{inputs:t,attrs:n}=e,{x:s}=t,{shape:r}=n,a=v.sizeFromShape(s.shape),o=v.inferFromImplicitShape(r,a);return v.assert(a===v.sizeFromShape(o),()=>`new shape: ${o}, old shape: ${s.shape}. New shape and old shape must have the same number of elements.`),e.backend.incRef(s.dataId),{dataId:s.dataId,shape:o,dtype:s.dtype}}var Ade={kernelName:_l,backendName:"wasm",kernelFunc:hs},X9;function xde(e){X9=e.wasm.cwrap(mo,null,["number","array","number","number","array","number","number","number","number"])}function bde(e){let{inputs:t,backend:n,attrs:s}=e,{a:r,b:a}=t,{transposeA:o,transposeB:i}=s;if(r.dtype!=="float32"||a.dtype!=="float32")throw new Error("BatchMatMul for non non-float32 tensors not yet supported.");let l=r.shape.length,u=a.shape.length,c=o?r.shape[l-2]:r.shape[l-1],p=i?a.shape[u-1]:a.shape[u-2],d=o?r.shape[l-1]:r.shape[l-2],h=i?a.shape[u-2]:a.shape[u-1],f=r.shape.slice(0,-2),m=a.shape.slice(0,-2),g=v.sizeFromShape(f),y=v.sizeFromShape(m),A=Xl.assertAndGetBroadcastShape(r.shape.slice(0,-2),a.shape.slice(0,-2)).concat([d,h]);v.assert(c===p,()=>`Error in matMul: inner shapes (${c}) and (${p}) of Tensors with shapes ${r.shape} and ${a.shape} and transposeA=${o} and transposeB=${i} must match.`);let b=o?[g,c,d]:[g,d,c],w=i?[y,h,p]:[y,p,h],k=hs({inputs:{x:r},backend:n,attrs:{shape:b}}),S=hs({inputs:{x:a},backend:n,attrs:{shape:w}}),E=n.dataIdMap.get(k.dataId).id,R=n.dataIdMap.get(S.dataId).id,$=o?k.shape[2]:k.shape[1],_=i?S.shape[1]:S.shape[2],P=Math.max(g,y),C=n.makeOutput([P,$,_],k.dtype),F=n.dataIdMap.get(C.dataId).id,G=new Uint8Array(new Int32Array(k.shape).buffer),K=new Uint8Array(new Int32Array(S.shape).buffer);return X9(E,G,k.shape.length,R,K,S.shape.length,o,i,F),n.disposeData(k.dataId),n.disposeData(S.dataId),C.shape=A,C}var vde={kernelName:mo,backendName:"wasm",setupFunc:xde,kernelFunc:bde};function al(e){let{inputs:{x:t},attrs:{begin:n,size:s},backend:r}=e,[a,o]=Ut.parseSliceParams(t,n,s),i=Ut.isSliceContinous(t.shape,a,o),l=r.readSync(t.dataId),u=r.makeOutput(o,t.dtype),c=v.computeStrides(t.shape),p=r.dataIdMap.get(u.dataId);if(i){let f=Ut.computeFlatOffset(a,c);return t.dtype==="string"?p.stringBytes=l.slice(f,f+v.sizeFromShape(o)):r.typedArrayFromHeap(u).set(l.subarray(f,f+v.sizeFromShape(o))),u}if(t.dtype==="string"){let f=_m(l,a,o,t.shape,t.dtype);return p.stringBytes=f,u}let d=r.typedArrayFromHeap(u),h=t.shape.length;if(h===2)wde(l,c[0],d,a,o);else if(h===3)kde(l,c[0],c[1],d,a,o);else if(h===4)Ide(l,c[0],c[1],c[2],d,a,o);else{let f=_m(l,a,o,t.shape,t.dtype);d.set(f)}return u}function wde(e,t,n,s,r){let a=0,o=s[0],i=s[1],l=o+r[0];for(let u=o;u<l;u++){let c=u*t+i;n.set(e.subarray(c,c+r[1]),a),a+=r[1]}}function kde(e,t,n,s,r,a){let o=0,i=r[0],l=r[1],u=r[2],c=i+a[0],p=l+a[1];for(let d=i;d<c;d++)for(let h=l;h<p;h++){let f=d*t+h*n+u;s.set(e.subarray(f,f+a[2]),o),o+=a[2]}}function Ide(e,t,n,s,r,a,o){let i=0,l=a[0],u=a[1],c=a[2],p=l+o[0],d=u+o[1],h=c+o[2],f=a[3];for(let m=l;m<p;m++)for(let g=u;g<d;g++)for(let y=c;y<h;y++){let x=m*t+g*n+y*s+f;r.set(e.subarray(x,x+o[3]),i),i+=o[3]}}var Sde={kernelName:Ol,backendName:"wasm",kernelFunc:al};function Cde(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockShape:a,crops:o}=s,i=a.reduce((y,x)=>y*x),l=T.getReshaped(r.shape,a,i),u=T.getPermuted(l.length,a.length),c=T.getReshapedPermuted(r.shape,a,i),p=T.getSliceBeginCoords(o,a.length),d=T.getSliceSize(c,o,a.length),h=hs({inputs:{x:r},backend:n,attrs:{shape:l}}),f=lo({inputs:{x:h},backend:n,attrs:{perm:u}}),m=hs({inputs:{x:f},backend:n,attrs:{shape:c}}),g=al({inputs:{x:m},backend:n,attrs:{begin:p,size:d}});return n.disposeData(h.dataId),n.disposeData(f.dataId),n.disposeData(h.dataId),g}var Tde={kernelName:ll,backendName:"wasm",kernelFunc:Cde};function cd(e){let{inputs:{x:t},attrs:{dtype:n},backend:s}=e,r=s.makeOutput(t.shape,n),a=s.typedArrayFromHeap(t);return s.typedArrayFromHeap(r).set(a),r}var Nde={kernelName:go,backendName:"wasm",kernelFunc:cd},Ede=In(yo),K9;function Rde(e){K9=e.wasm.cwrap(ba,null,["number","number","number","number"])}function _de(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{clipValueMin:a,clipValueMax:o}=s,i=n.dataIdMap.get(r.dataId).id,l=n.makeOutput(r.shape,r.dtype),u=n.dataIdMap.get(l.dataId).id;return K9(i,a,o,u),l}var Dde={kernelName:ba,backendName:"wasm",setupFunc:Rde,kernelFunc:_de};function Z9(e){let{inputs:t,backend:n}=e,s=v.parseAxisParam(e.attrs.axis,t[0].shape)[0],r=T.computeOutShape(t.map(h=>h.shape),s),a=t.filter(h=>v.sizeFromShape(h.shape)>0);if(a.length===1)return C2({inputs:{x:a[0]},backend:n});let o=n.makeOutput(r,t[0].dtype);if(v.sizeFromShape(r)===0)return o;let i=a.map(h=>h.shape);if(T.assertParamsConsistent(i,s),a[0].dtype==="string"){let h=a.map(A=>{let b=v.sizeFromShape(A.shape.slice(s));return hs({inputs:{x:A},backend:n,attrs:{shape:[-1,b]}})}),f=h.map(A=>({vals:n.readSync(A.dataId),shape:A.shape}));r=T.computeOutShape(h.map(A=>A.shape),1);let m=h[0].shape[0]===1,g=Ex(f,r,t[0].dtype,m),y=T.computeOutShape(a.map(A=>A.shape),s);o.shape=y;let x=n.dataIdMap.get(o.dataId);return x.stringBytes=T.fromStringArrayToUint8(g),h.forEach(A=>n.disposeData(A.dataId)),o}let l=v.sizeFromShape(a[0].shape.slice(0,s)),u=0,c=a.map(h=>{let f=v.sizeFromShape(h.shape.slice(s));return u+=f,f}),p=a.map(h=>n.typedArrayFromHeap(h)),d=n.typedArrayFromHeap(o);for(let h=0;h<l;h++){let f=h*u;for(let m=0;m<p.length;m++){let g=c[m],y=h*g,x=p[m].subarray(y,y+g);d.set(x,f),f+=g}}return o}var $de={kernelName:ul,backendName:"wasm",kernelFunc:Z9},Y9;function Pde(e){Y9=e.wasm.cwrap(Ao,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function Fde(e){let{inputs:t,attrs:n,backend:s}=e,{x:r,filter:a}=t,o=s.dataIdMap.get(r.dataId).id,i=s.dataIdMap.get(a.dataId).id,{strides:l,dilations:u,pad:c,dimRoundingMode:p,dataFormat:d}=n,h=T.convertConv2DDataFormat(d),f=T.computeConv2DInfo(r.shape,a.shape,l,u,c,p,!1,h),m=f.filterHeight,g=f.filterWidth,y=f.padInfo.top,x=f.padInfo.right,A=f.padInfo.bottom,b=f.padInfo.left,w=f.dilationHeight,k=f.dilationWidth,S=f.strideHeight,E=f.strideWidth,R=f.inChannels,$=f.outChannels,_=f.padInfo.type==="SAME"?1:0;if(f.dataFormat!=="channelsLast")throw new Error(`wasm backend Conv2D does not support dataFormat:'${f.dataFormat}'. Please use 'channelsLast'.`);let P=s.makeOutput(f.outShape,"float32"),C=s.dataIdMap.get(P.dataId).id;return Y9(o,r.shape[0],r.shape[1],r.shape[2],i,m,g,y,x,A,b,_,w,k,S,E,R,$,C),P}var Ode={kernelName:Ao,backendName:"wasm",setupFunc:Pde,kernelFunc:Fde},J9;function Mde(e){J9=e.wasm.cwrap(xo,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function zde(e){let{backend:t,inputs:n,attrs:s}=e,{dy:r,filter:a}=n,{strides:o,pad:i,dataFormat:l,dimRoundingMode:u,inputShape:c}=s,p=1,d=T.convertConv2DDataFormat(l),h=T.computeConv2DInfo(c,a.shape,o,p,i,u,!1,d),{batchSize:f,filterHeight:m,filterWidth:g,inChannels:y,inHeight:x,inWidth:A,outChannels:b,outHeight:w,outWidth:k,strideHeight:S,strideWidth:E}=h,R=m-1-h.padInfo.top,$=g-1-h.padInfo.left,_=h.dataFormat==="channelsLast",P=v.computeStrides(h.inShape),C=v.computeStrides(r.shape),[F,G,K]=v.computeStrides(a.shape),z=P[0],Z=_?P[1]:P[2],J=_?P[2]:1,te=_?1:P[1],B=C[0],oe=_?C[1]:C[2],Q=_?C[2]:1,ae=_?1:C[1],ie=t.makeOutput(h.inShape,"float32"),me=t.dataIdMap.get(ie.dataId).id,we=t.dataIdMap.get(r.dataId).id,Re=t.dataIdMap.get(a.dataId).id;return J9(we,Re,f,m,g,x,A,y,w,k,b,S,E,R,$,F,G,K,z,Z,J,te,B,oe,Q,ae,me),ie}var Lde={kernelName:xo,backendName:"wasm",setupFunc:Mde,kernelFunc:zde},Bde=In(bo),Wde=In(vo),uy;(function(e){e[e.bilinear=0]="bilinear",e[e.nearest=1]="nearest"})(uy||(uy={}));var Q9;function Vde(e){Q9=e.wasm.cwrap(dl,null,["number","number","number","number","array","number","number","number","number","number"])}function Ude(e){let{backend:t,inputs:n,attrs:s}=e,{method:r,extrapolationValue:a,cropSize:o}=s,{image:i,boxes:l,boxInd:u}=n,c=l.shape[0],[p,d]=o,h=[c,p,d,i.shape[3]],f=t.dataIdMap.get(i.dataId),m;i.dtype!=="float32"&&(m=cd({backend:t,inputs:{x:i},attrs:{dtype:"float32"}}),f=t.dataIdMap.get(m.dataId));let g=f.id,y=t.dataIdMap.get(l.dataId).id,x=t.dataIdMap.get(u.dataId).id,A=t.makeOutput(h,"float32"),b=t.dataIdMap.get(A.dataId).id,w=new Uint8Array(new Int32Array(i.shape).buffer);return Q9(g,y,x,c,w,p,d,uy[r],a,b),m!=null&&t.disposeData(m.dataId),A}var Gde={kernelName:dl,backendName:"wasm",setupFunc:Vde,kernelFunc:Ude},eC;function Hde(e){eC=e.wasm.cwrap(cl,null,["number","number","number","number","number","number"])}function jde(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,exclusive:o,reverse:i}=s,l=r.shape.length;v.assert(r.dtype==="float32"||r.dtype==="int32",()=>`cumprod does not support ${r.dtype} tensors in the WASM backend`);let u=T.getAxesPermutation([a],l),c=r;u!==null&&(c=lo({inputs:{x:r},attrs:{perm:u},backend:n}));let p=T.getInnerMostAxes(1,l)[0];T.assertAxesAreInnerMostDims("cumprod",[p],l);let d=n.makeOutput(c.shape,c.dtype),h=c.shape[p],f=n.dataIdMap.get(c.dataId).id,m=n.dataIdMap.get(d.dataId).id;eC(f,o?1:0,i?1:0,h,m,Ht[r.dtype]);let g=d;if(u!==null){let y=T.getUndoAxesPermutation(u);g=lo({inputs:{x:d},attrs:{perm:y},backend:n}),n.disposeData(c.dataId),n.disposeData(d.dataId)}return g}var qde={kernelName:cl,backendName:"wasm",setupFunc:Hde,kernelFunc:jde},tC;function Xde(e){tC=e.wasm.cwrap(wo,null,["number","number","number","number","number","number"])}function Kde(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,exclusive:o,reverse:i}=s,l=r.shape.length;v.assert(r.dtype==="float32"||r.dtype==="int32",()=>`cumsum does not support ${r.dtype} tensors in the WASM backend`);let u=T.getAxesPermutation([a],l),c=r;u!==null&&(c=lo({inputs:{x:r},attrs:{perm:u},backend:n}));let p=T.getInnerMostAxes(1,l)[0];T.assertAxesAreInnerMostDims("cumsum",[p],l);let d=n.makeOutput(c.shape,c.dtype),h=c.shape[p],f=n.dataIdMap.get(c.dataId).id,m=n.dataIdMap.get(d.dataId).id;tC(f,o?1:0,i?1:0,h,m,Ht[r.dtype]);let g=d;if(u!==null){let y=T.getUndoAxesPermutation(u);g=lo({inputs:{x:d},attrs:{perm:y},backend:n}),n.disposeData(c.dataId),n.disposeData(d.dataId)}return g}var Zde={kernelName:wo,backendName:"wasm",setupFunc:Xde,kernelFunc:Kde},nC;function Yde(e){nC=e.wasm.cwrap(pl,null,["number","number","number","array","number","array","array","number","number"])}function Jde(e){let{backend:t,inputs:n,attrs:s}=e,{x:r}=n,{blockSize:a,dataFormat:o}=s,i=r.shape[0],l=o==="NHWC"?r.shape[1]:r.shape[2],u=o==="NHWC"?r.shape[2]:r.shape[3],c=o==="NHWC"?r.shape[3]:r.shape[1],p=l*a,d=u*a,h=c/(a*a),f=o==="NHWC"?[i,p,d,h]:[i,h,p,d],m=t.makeOutput(f,"float32"),y=t.dataIdMap.get(r.dataId).id,x=new Uint8Array(new Int32Array(v.computeStrides(r.shape)).buffer),A=new Uint8Array(new Int32Array(f).buffer),b=new Uint8Array(new Int32Array(v.computeStrides(f)).buffer),w=t.dataIdMap.get(m.dataId).id;return nC(y,a,o==="NHWC"?1:0,x,r.shape.length-1,A,b,f.length,w),m}var Qde={kernelName:pl,backendName:"wasm",setupFunc:Yde,kernelFunc:Jde},sC;function epe(e){sC=e.wasm.cwrap(ko,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function tpe(e){let{inputs:t,attrs:n,backend:s}=e,{x:r,filter:a}=t,o=s.dataIdMap.get(r.dataId).id,i=s.dataIdMap.get(a.dataId).id,{strides:l,dilations:u,pad:c,dimRoundingMode:p}=n,d=u==null?[1,1]:u,h=T.computeConv2DInfo(r.shape,a.shape,l,d,c,p,!0),f=h.filterHeight,m=h.filterWidth,g=h.padInfo.top,y=h.padInfo.right,x=h.padInfo.bottom,A=h.padInfo.left,b=h.dilationHeight,w=h.dilationWidth,k=h.strideHeight,S=h.strideWidth,E=h.inChannels,R=h.outChannels,$=h.padInfo.type==="SAME"?1:0;if(h.dataFormat!=="channelsLast")throw new Error(`wasm backend DepthwiseConv2dNative does not support dataFormat:'${h.dataFormat}'. Please use 'channelsLast'.`);let _=s.makeOutput(h.outShape,"float32"),P=s.dataIdMap.get(_.dataId).id;return sC(o,r.shape[0],r.shape[1],r.shape[2],i,f,m,g,y,x,A,$,b,w,k,S,E,R,P),_}var npe={kernelName:ko,backendName:"wasm",setupFunc:epe,kernelFunc:tpe},spe=In(So),rpe=!1,ape=Dn(hl,rpe,"bool"),ope=In(Co,"float32");function cy(e){let{inputs:t,attrs:n,backend:s}=e,{input:r}=t,{dim:a}=n,o=r.shape.length,i=r.shape.slice(),l=a;return a<0&&(v.assert(-(o+1)<=a,()=>`Axis must be in the interval [${-(o+1)}, ${o}]`),l=o+a+1),i.splice(l,0,1),hs({inputs:{x:r},backend:s,attrs:{shape:i}})}var ipe={kernelName:fl,backendName:"wasm",kernelFunc:cy};function rC(e){let{attrs:{shape:t,value:n,dtype:s},backend:r}=e,a=r.makeOutput(t,s);return r.typedArrayFromHeap(a).fill(n),a}var lpe={kernelName:kc,backendName:"wasm",kernelFunc:rC},aC;function upe(e){aC=e.wasm.cwrap(gl,null,["number","number","number","number","number","number"])}function cpe(e){let{inputs:t,backend:n}=e,{image:s}=t,r=n.makeOutput(s.shape,s.dtype),a=n.dataIdMap.get(s.dataId).id,o=n.dataIdMap.get(r.dataId).id,[i,l,u,c]=s.shape;return aC(a,i,l,u,c,o),r}var dpe={kernelName:gl,backendName:"wasm",kernelFunc:cpe,setupFunc:upe},ppe=In(To),hpe=!1,fpe=Dn(No,hpe),oC;function mpe(e){oC=e.wasm.cwrap(Eo,null,["number","number","number","number","number","number","number"])}function gpe(e){let{backend:t,inputs:n,attrs:s}=e,{varianceEpsilon:r}=s,{x:a,mean:o,variance:i,offset:l,scale:u}=n,c=t.dataIdMap.get(a.dataId).id,p=t.dataIdMap.get(o.dataId).id,d=t.dataIdMap.get(i.dataId).id,h=l!=null?t.dataIdMap.get(l.dataId).id:0,f=u!=null?t.dataIdMap.get(u.dataId).id:0,m=t.makeOutput(a.shape,a.dtype);if(v.sizeFromShape(a.shape)===0)return m;let g=t.dataIdMap.get(m.dataId).id;return oC(c,p,d,h,f,r,g),m}var ype={kernelName:Eo,backendName:"wasm",setupFunc:mpe,kernelFunc:gpe},iC;function Ape(e){iC=e.wasm.cwrap(Qa,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function xpe(e){let{inputs:t,attrs:n,backend:s}=e,{x:r,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dilations:c,dataFormat:p,dimRoundingMode:d,activation:h,leakyreluAlpha:f}=n,m=T.computeConv2DInfo(r.shape,a.shape,l,c,u,d),g=Fp[h];if(g==null)throw new Error(`${h} activation not yet supported for FusedConv2D in the wasm backend.`);let y=s.dataIdMap.get(r.dataId).id,x=s.dataIdMap.get(a.dataId).id,A=m.outChannels,b=0;if(o!=null){let Q=s.dataIdMap.get(o.dataId);if(Q.shape.length!==1)throw new Error(`FusedConv2D only supports rank-1 bias but got rank ${Q.shape.length}.`);if(Q.shape[0]!==A)throw new Error(`FusedConv2D bias shape (${Q.shape}) does not match the number of output channels (${A})`);b=Q.id}let w=m.filterHeight,k=m.filterWidth,S=m.padInfo.top,E=m.padInfo.right,R=m.padInfo.bottom,$=m.padInfo.left,_=m.dilationHeight,P=m.dilationWidth,C=m.strideHeight,F=m.strideWidth,G=m.inChannels,K=m.padInfo.type==="SAME"?1:0,z=m.batchSize,Z=m.inHeight,J=m.inWidth;if(p!=="NHWC")throw new Error(`wasm backend FusedConv2D does not support dataFormat:'${p}'. Please use 'NHWC'.`);let te=s.makeOutput(m.outShape,"float32"),B=s.dataIdMap.get(te.dataId).id,oe=i==null?0:s.dataIdMap.get(i.dataId).id;return iC(y,z,Z,J,x,w,k,b,S,E,R,$,K,_,P,C,F,G,A,g,oe,f||0,B),te}var bpe={kernelName:Qa,backendName:"wasm",setupFunc:Ape,kernelFunc:xpe},lC;function vpe(e){lC=e.wasm.cwrap(eo,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function wpe(e){let{inputs:t,attrs:n,backend:s}=e,{x:r,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dilations:c,dataFormat:p,dimRoundingMode:d,activation:h,leakyreluAlpha:f}=n,m=T.computeConv2DInfo(r.shape,a.shape,l,c,u,d,!0),g=Fp[h];if(g==null)throw new Error(`${h} activation not yet supported for FusedDepthwiseConv2D in the wasm backend.`);let y=s.dataIdMap.get(r.dataId).id,x=s.dataIdMap.get(a.dataId).id,A=m.outChannels,b=0;if(o!=null){let Q=s.dataIdMap.get(o.dataId);if(Q.shape.length!==1)throw new Error(`FusedDepthwiseConv2D only supports rank-1 bias but got rank ${Q.shape.length}.`);if(Q.shape[0]!==A)throw new Error(`FusedDepthwiseConv2D bias shape (${Q.shape}) does not match the number of output channels (${A})`);b=Q.id}let w=m.filterHeight,k=m.filterWidth,S=m.padInfo.top,E=m.padInfo.right,R=m.padInfo.bottom,$=m.padInfo.left,_=m.dilationHeight,P=m.dilationWidth,C=m.strideHeight,F=m.strideWidth,G=m.inChannels,K=m.padInfo.type==="SAME"?1:0,z=m.batchSize,Z=m.inHeight,J=m.inWidth;if(p!=="NHWC")throw new Error(`wasm backend FusedDepthwiseConv2D does not support dataFormat:'${p}'. Please use 'NHWC'.`);let te=s.makeOutput(m.outShape,"float32"),B=s.dataIdMap.get(te.dataId).id,oe=i==null?0:s.dataIdMap.get(i.dataId).id;return lC(y,z,Z,J,x,w,k,b,S,E,R,$,K,_,P,C,F,G,A,g,oe,f||0,B),te}var kpe={kernelName:eo,backendName:"wasm",setupFunc:vpe,kernelFunc:wpe},uC;function Ipe(e){uC=e.wasm.cwrap(Al,null,["number","number","number","number","number","number","array","number"])}function Spe(e){let{backend:t,inputs:n}=e,{params:s,indices:r}=n,[a,o,i,l]=Ly.prepareAndValidate(s,r),u=t.makeOutput(a,s.dtype);if(o===0)return u;let c=r.shape,p=c[c.length-1],h=t.dataIdMap.get(s.dataId).id,m=t.dataIdMap.get(r.dataId).id,g=new Uint8Array(new Int32Array(l).buffer),y=t.dataIdMap.get(u.dataId).id;return uC(h,Ht[s.dtype],m,o,p,i,g,y),u}var Cpe={kernelName:Al,backendName:"wasm",setupFunc:Ipe,kernelFunc:Spe},cC;function Tpe(e){cC=e.wasm.cwrap("Gather",null,["number","number","array","number","number","number","array","number"])}function Npe(e){let{backend:t,inputs:n,attrs:s}=e,{x:r,indices:a}=n,{axis:o,batchDims:i}=s,l=v.parseAxisParam(o,r.shape)[0],u=t.readSync(a.dataId),c=r.shape[l];for(let R=0;R<u.length;++R){let $=u[R];v.assert($<=c-1&&$>=0,()=>`GatherV2: the index value ${$} is not in [0, ${c-1}]`)}let p=T.segment_util.collectGatherOpShapeInfo(r,a,l,i),d=hs({inputs:{x:r},attrs:{shape:[p.batchSize,p.outerSize,p.dimSize,p.sliceSize]},backend:t}),h=v.sizeFromShape(a.shape),f=hs({inputs:{x:a},attrs:{shape:[p.batchSize,h/p.batchSize]},backend:t}),m=[p.batchSize,p.outerSize,h/p.batchSize,p.sliceSize],g=t.makeOutput(m,r.dtype);if(v.sizeFromShape(r.shape)===0)return g;let y=d.shape.length-1,A=t.dataIdMap.get(d.dataId).id,w=t.dataIdMap.get(f.dataId).id,k=t.dataIdMap.get(g.dataId).id,S=new Uint8Array(new Int32Array(v.computeStrides(d.shape)).buffer),E=new Uint8Array(new Int32Array(v.computeStrides(m)).buffer);return cC(A,Ht[r.dtype],S,y,w,p.batchSize,E,k),t.disposeData(d.dataId),t.disposeData(f.dataId),g.shape=p.outputShape,g}var Epe={kernelName:yl,backendName:"wasm",setupFunc:Tpe,kernelFunc:Npe},Rpe=!1,_pe=Dn(xl,Rpe,"bool"),Dpe=!1,$pe=Dn(Ro,Dpe,"bool"),dC;function Ppe(e){dC=e.wasm.cwrap(Do,null,["number","number","number","number"])}function Fpe(e){let{inputs:{x:t},attrs:{alpha:n},backend:s}=e,r=s.dataIdMap.get(t.dataId).id,a=s.makeOutput(t.shape,"float32");if(v.sizeFromShape(t.shape)!==0){let o=s.dataIdMap.get(a.dataId).id;dC(r,Ht[t.dtype],n,o)}return a}var Ope={kernelName:Do,backendName:"wasm",setupFunc:Ppe,kernelFunc:Fpe},Mpe=!1,zpe=Dn(bl,Mpe,"bool"),Lpe=!1,Bpe=Dn(vl,Lpe,"bool"),Wpe=In($o),Vpe=!1,Upe=Dn(wl,Vpe,"bool"),Gpe=In(kl),Hpe=!1,jpe=Dn(Nc,Hpe,"bool"),qpe=!1,Xpe=Dn(f6,qpe,"bool"),pC;function Kpe(e){pC=e.wasm.cwrap(Po,null,["number","number","number","number"])}function Zpe(e){let{backend:t,inputs:n,attrs:s}=e,{reductionIndices:r,keepDims:a}=s,{x:o}=n,l=t.dataIdMap.get(o.dataId).id,u=o,{transposed:c,axes:p,originalAxes:d,inputWasTransposed:h}=hi(o,r,t);if(h){let A=t.dataIdMap.get(c.dataId).id;u=c,l=A}let f=u.shape.length;T.assertAxesAreInnerMostDims("max",p,f);let[m,g]=T.computeOutAndReduceShapes(u.shape,p),y=v.sizeFromShape(g),x=t.makeOutput(m,o.dtype);if(v.sizeFromShape(u.shape)!==0){let A=t.dataIdMap.get(x.dataId).id;pC(l,Ht[o.dtype],y,A)}if(h&&t.disposeData(c.dataId),a){let A=T.expandShapeToKeepDim(x.shape,d);x.shape=A}return x}var Ype={kernelName:Po,backendName:"wasm",setupFunc:Kpe,kernelFunc:Zpe},Jpe=!1,Qpe=Dn(Fo,Jpe),hC;function ehe(e){hC=e.wasm.cwrap(Oo,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function the(e){let{inputs:t,attrs:n,backend:s}=e,r=t.x,a=s.dataIdMap.get(r.dataId).id;v.assert(r.dtype==="float32",()=>`Error in MaxPool: only float32 input is supported. Got ${r.dtype}.`);let{filterSize:o,strides:i,pad:l,dimRoundingMode:u}=n,c=T.computePool2DInfo(r.shape,o,i,1,l,u),p=c.filterHeight,d=c.filterWidth,h=c.padInfo.top,f=c.padInfo.right,m=c.padInfo.bottom,g=c.padInfo.left,y=c.dilationHeight,x=c.dilationWidth,A=c.strideHeight,b=c.strideWidth,w=c.inChannels,k=c.outChannels;if(c.dataFormat!=="channelsLast")throw new Error(`wasm backend does not support dataFormat:'${c.dataFormat}'. Please use 'channelsLast'.`);let S=s.makeOutput(c.outShape,"float32"),E=s.dataIdMap.get(S.dataId).id;return hC(a,r.shape[0],r.shape[1],r.shape[2],p,d,h,f,m,g,y,x,A,b,w,k,E),S}var nhe={kernelName:Oo,backendName:"wasm",setupFunc:ehe,kernelFunc:the},fC;function she(e){fC=e.wasm.cwrap(Mo,null,["number, number, number"])}function rhe(e){let{backend:t,inputs:n,attrs:s}=e,{axis:r,keepDims:a}=s,{x:o}=n,i=t.dataIdMap.get(o.dataId).id,l=i,u=o,{transposed:c,axes:p,originalAxes:d,inputWasTransposed:h}=hi(o,r,t),f=p;if(h){let b=t.dataIdMap.get(c.dataId).id;b!==i&&(u=c,l=b,f=T.getInnerMostAxes(f.length,u.shape.length))}T.assertAxesAreInnerMostDims("mean",f,u.shape.length);let[m,g]=T.computeOutAndReduceShapes(u.shape,f),y=v.sizeFromShape(g),x=u;u.dtype!=="float32"&&(x=cd({backend:t,inputs:{x:u},attrs:{dtype:"float32"}}),l=t.dataIdMap.get(x.dataId).id);let A=t.makeOutput(m,"float32");if(v.sizeFromShape(u.shape)!==0){let b=t.dataIdMap.get(A.dataId).id;fC(l,y,b)}if(h&&t.disposeData(c.dataId),a){let b=T.expandShapeToKeepDim(A.shape,d);A.shape=b}return u.dtype!=="float32"&&t.disposeData(x.dataId),A}var ahe={kernelName:Mo,backendName:"wasm",setupFunc:she,kernelFunc:rhe},mC;function ohe(e){mC=e.wasm.cwrap(zo,null,["number","number","number","number"])}function ihe(e){let{backend:t,inputs:n,attrs:s}=e,{axis:r,keepDims:a}=s,{x:o}=n,i=t.dataIdMap.get(o.dataId).id,l=i,u=o,{transposed:c,axes:p,originalAxes:d,inputWasTransposed:h}=hi(o,r,t);if(h){let A=t.dataIdMap.get(c.dataId).id;A!==i&&(u=c,l=A)}let f=u.shape.length;T.assertAxesAreInnerMostDims("min",p,f);let[m,g]=T.computeOutAndReduceShapes(u.shape,p),y=v.sizeFromShape(g),x=t.makeOutput(m,u.dtype);if(v.sizeFromShape(u.shape)!==0){let A=t.dataIdMap.get(x.dataId).id;mC(l,Ht[o.dtype],y,A)}if(h&&t.disposeData(c.dataId),a){let A=T.expandShapeToKeepDim(x.shape,d);x.shape=A}return x}var lhe={kernelName:zo,backendName:"wasm",setupFunc:ohe,kernelFunc:ihe},uhe=!1,che=Dn(Lo,uhe),dy;(function(e){e[e.reflect=0]="reflect",e[e.symmetric=1]="symmetric"})(dy||(dy={}));var gC;function dhe(e){gC=e.wasm.cwrap(Bo,null,["number","array","number","number","array","array","number","number"])}function phe(e){let{inputs:{x:t},backend:n,attrs:{paddings:s,mode:r}}=e,a=s.map((f,m)=>f[0]+t.shape[m]+f[1]),o=n.dataIdMap.get(t.dataId).id,i=n.makeOutput(a,t.dtype),l=n.dataIdMap.get(i.dataId).id,u=new Uint8Array(new Int32Array(t.shape).buffer),c=s.map(f=>f[0]),p=s.map(f=>f[1]),d=new Uint8Array(new Int32Array(c).buffer),h=new Uint8Array(new Int32Array(p).buffer);return gC(o,u,t.shape.length,Ht[t.dtype],d,h,dy[r],l),i}var hhe={kernelName:Bo,backendName:"wasm",kernelFunc:phe,setupFunc:dhe},fhe=!0,mhe=Dn(Wo,fhe),ghe=In(Il);function eb(e,t){let n=new Int32Array(e.wasm.HEAPU8.buffer,t,4),s=n[0],r=n[1],a=n[2],o=n[3];return e.wasm._free(t),{pSelectedIndices:s,selectedSize:r,pSelectedScores:a,pValidOutputs:o}}var yC;function yhe(e){yC=e.wasm.cwrap(Cl,"number",["number","number","number","number","number"])}function Ahe(e){let{backend:t,inputs:n,attrs:s}=e,{iouThreshold:r,maxOutputSize:a,scoreThreshold:o}=s,{boxes:i,scores:l}=n,u=t.dataIdMap.get(i.dataId).id,c=t.dataIdMap.get(l.dataId).id,p=yC(u,c,a,r,o),{pSelectedIndices:d,selectedSize:h,pSelectedScores:f,pValidOutputs:m}=eb(t,p);return t.wasm._free(f),t.wasm._free(m),t.makeOutput([h],"int32",d)}var xhe={kernelName:Cl,backendName:"wasm",setupFunc:yhe,kernelFunc:Ahe},AC;function bhe(e){AC=e.wasm.cwrap(Rc,"number",["number","number","number","number","number","bool"])}function vhe(e){let{backend:t,inputs:n,attrs:s}=e,{iouThreshold:r,maxOutputSize:a,scoreThreshold:o,padToMaxOutputSize:i}=s,{boxes:l,scores:u}=n,c=t.dataIdMap.get(l.dataId).id,p=t.dataIdMap.get(u.dataId).id,d=AC(c,p,a,r,o,i),{pSelectedIndices:h,selectedSize:f,pSelectedScores:m,pValidOutputs:g}=eb(t,d);t.wasm._free(m);let y=t.makeOutput([f],"int32",h),x=t.makeOutput([],"int32",g);return[y,x]}var whe={kernelName:Rc,backendName:"wasm",setupFunc:bhe,kernelFunc:vhe},xC;function khe(e){xC=e.wasm.cwrap(Tl,"number",["number","number","number","number","number","number"])}function Ihe(e){let{backend:t,inputs:n,attrs:s}=e,{iouThreshold:r,maxOutputSize:a,scoreThreshold:o,softNmsSigma:i}=s,{boxes:l,scores:u}=n,c=t.dataIdMap.get(l.dataId).id,p=t.dataIdMap.get(u.dataId).id,d=xC(c,p,a,r,o,i),{pSelectedIndices:h,selectedSize:f,pSelectedScores:m,pValidOutputs:g}=eb(t,d);t.wasm._free(g);let y=t.makeOutput([f],"int32",h),x=t.makeOutput([f],"float32",m);return[y,x]}var She={kernelName:Tl,backendName:"wasm",setupFunc:khe,kernelFunc:Ihe},Che=!1,The=Dn(Sl,Che,"bool"),bC;function Nhe(e){bC=e.wasm.cwrap(El,null,["number","number","number","number","number"])}function Ehe(e){let{inputs:t,backend:n,attrs:s}=e,{indices:r}=t,{depth:a,onValue:o,offValue:i}=s,l=n.makeOutput([...r.shape,a],"int32"),u=n.dataIdMap.get(l.dataId).id,p=n.dataIdMap.get(r.dataId).id;return bC(p,a,o,i,u),l}var Rhe={kernelName:El,backendName:"wasm",setupFunc:Nhe,kernelFunc:Ehe};function _he(e){let{inputs:{x:t},backend:n}=e,s=n.makeOutput(t.shape,t.dtype);return n.typedArrayFromHeap(s).fill(1),s}var Dhe={kernelName:Nl,backendName:"wasm",kernelFunc:_he};function $he(e){let{inputs:t,backend:n,attrs:s}=e,{axis:r}=s;if(t.length===1)return cy({inputs:{input:t[0]},backend:n,attrs:{dim:r}});let a=t[0].shape,o=t[0].dtype;t.forEach(c=>{v.assertShapesMatch(a,c.shape,"All tensors passed to stack must have matching shapes"),v.assert(o===c.dtype,()=>"All tensors passed to stack must have matching dtypes")});let i=[],l=t.map(c=>{let p=cy({inputs:{input:c},backend:n,attrs:{dim:r}});return i.push(p),p}),u=Z9({inputs:l,backend:n,attrs:{axis:r}});return i.forEach(c=>n.disposeData(c.dataId)),u}var Phe={kernelName:Rl,backendName:"wasm",kernelFunc:$he},vC;function Fhe(e){vC=e.wasm.cwrap(Vo,null,["number","array","number","number","array","array","number","number"])}function Ohe(e){let{inputs:{x:t},backend:n,attrs:{paddings:s,constantValue:r}}=e,a=s.map((m,g)=>m[0]+t.shape[g]+m[1]);if(v.sizeFromShape(t.shape)===0)return rC({backend:n,attrs:{shape:a,value:r,dtype:t.dtype}});let o=n.dataIdMap.get(t.dataId).id,i=n.makeOutput(a,t.dtype),u=n.dataIdMap.get(i.dataId).id,c=new Uint8Array(new Int32Array(t.shape).buffer),p=s.map(m=>m[0]),d=s.map(m=>m[1]),h=new Uint8Array(new Int32Array(p).buffer),f=new Uint8Array(new Int32Array(d).buffer);return vC(o,c,t.shape.length,Ht[t.dtype],h,f,r,u),i}var wC={kernelName:Vo,backendName:"wasm",kernelFunc:Ohe,setupFunc:Fhe},Mhe=!1,zhe=Dn(Uo,Mhe),kC;function Lhe(e){kC=e.wasm.cwrap(Go,null,["number","number","number"])}function Bhe(e){let{inputs:t,backend:n}=e,{x:s,alpha:r}=t,a=n.dataIdMap.get(s.dataId).id,o=n.dataIdMap.get(r.dataId).id,i=a,l=s,u=l;l.dtype!=="float32"&&(u=cd({backend:n,inputs:{x:s},attrs:{dtype:"float32"}}),i=n.dataIdMap.get(u.dataId).id);let c=n.makeOutput(s.shape,"float32"),p=n.dataIdMap.get(c.dataId).id;return kC(i,o,p),l.dtype!=="float32"&&n.disposeData(u.dataId),c}var Whe={kernelName:Go,backendName:"wasm",setupFunc:Lhe,kernelFunc:Bhe},IC;function Vhe(e){IC=e.wasm.cwrap(Ho,null,["number","number","number","number"])}function Uhe(e){let{backend:t,inputs:n,attrs:s}=e,{axis:r,keepDims:a}=s,{x:o}=n,i=t.dataIdMap.get(o.dataId).id,l=i,u=o,{transposed:c,axes:p,originalAxes:d,inputWasTransposed:h}=hi(o,r,t),f=p;if(h){let A=t.dataIdMap.get(c.dataId).id;A!==i&&(u=c,l=A,f=T.getInnerMostAxes(f.length,u.shape.length))}T.assertAxesAreInnerMostDims("prod",f,u.shape.length);let[m,g]=T.computeOutAndReduceShapes(u.shape,f),y=v.sizeFromShape(g),x=t.makeOutput(m,u.dtype);if(v.sizeFromShape(u.shape)!==0){let A=t.dataIdMap.get(x.dataId).id;IC(l,y,Ht[x.dtype],A)}if(h&&t.disposeData(c.dataId),a){let A=T.expandShapeToKeepDim(x.shape,d);x.shape=A}return x}var Ghe={kernelName:Ho,backendName:"wasm",setupFunc:Vhe,kernelFunc:Uhe},Hhe=e=>{let{backend:t,attrs:n}=e,{start:s,stop:r,step:a,dtype:o}=n,i=Dx(s,r,a,o),l=t.makeOutput([i.length],o);return t.typedArrayFromHeap(l).set(i),l},jhe={kernelName:_c,backendName:"wasm",kernelFunc:Hhe},qhe=!0,Xhe=Dn(Io,qhe),Khe=In(jo),Zhe=In(Ko),SC;function Yhe(e){SC=e.wasm.cwrap(Xo,null,["number","number","number","number","number","number","number","number","number","number"])}function Jhe(e){let{backend:t,inputs:n,attrs:s}=e,{images:r}=n,{alignCorners:a,halfPixelCenters:o,size:i}=s,[l,u]=i,[c,p,d,h]=r.shape,f=[c,l,u,h],m=t.dataIdMap.get(r.dataId),g;m.dtype!=="float32"&&(g=cd({backend:t,inputs:{x:r},attrs:{dtype:"float32"}}),m=t.dataIdMap.get(g.dataId));let y=m.id,x=t.makeOutput(f,"float32");if(v.sizeFromShape(r.shape)===0)return x;let A=t.dataIdMap.get(x.dataId).id;return SC(y,c,p,d,h,l,u,a?1:0,o?1:0,A),g!=null&&t.disposeData(g.dataId),x}var Qhe={kernelName:Xo,backendName:"wasm",setupFunc:Yhe,kernelFunc:Jhe},CC;function efe(e){CC=e.wasm.cwrap(qo,null,["number","number","number","number","number","number","number","number","number","number"])}function tfe(e){let{backend:t,inputs:n,attrs:s}=e,{images:r}=n,{alignCorners:a,halfPixelCenters:o,size:i}=s,[l,u]=i,[c,p,d,h]=r.shape,f=[c,l,u,h],m=t.makeOutput(f,"float32");if(v.sizeFromShape(r.shape)===0)return m;let g=t.dataIdMap.get(r.dataId),y;g.dtype!=="float32"&&(y=cd({backend:t,inputs:{x:r},attrs:{dtype:"float32"}}),g=t.dataIdMap.get(y.dataId));let x=g.id,A=t.dataIdMap.get(m.dataId).id;return CC(x,c,p,d,h,l,u,a?1:0,o?1:0,A),y!=null&&t.disposeData(y.dataId),m}var nfe={kernelName:qo,backendName:"wasm",setupFunc:efe,kernelFunc:tfe},TC;function sfe(e){TC=e.wasm.cwrap(Dl,null,["number","array","number","array","number","number"])}function rfe(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{dims:a}=s,o=v.parseAxisParam(a,r.shape);if(r.shape.length===0)return C2({inputs:{x:r},backend:n});let i=n.makeOutput(r.shape,r.dtype),l=n.dataIdMap.get(r.dataId).id,u=n.dataIdMap.get(i.dataId).id,c=new Uint8Array(new Int32Array(o).buffer),p=new Uint8Array(new Int32Array(r.shape).buffer);TC(l,c,o.length,p,r.shape.length,u);let d=hs({inputs:{x:i},attrs:{shape:r.shape},backend:n});return n.disposeData(i.dataId),d}var afe={kernelName:Dl,backendName:"wasm",kernelFunc:rfe,setupFunc:sfe},NC;function ofe(e){NC=e.wasm.cwrap(jl,null,["number","number","number","number","number","number","number","number","array","number","number"])}function ife(e){let{inputs:t,backend:n,attrs:s}=e,{image:r}=t,{radians:a,fillValue:o,center:i}=s,l=n.makeOutput(r.shape,r.dtype),u=n.dataIdMap.get(r.dataId).id,c=n.dataIdMap.get(l.dataId).id,[p,d,h,f]=r.shape,[m,g]=T.getImageCenter(i,d,h),y=o===0,x=255,A=typeof o=="number"?[o,o,o,y?0:x]:[...o,x],b=new Uint8Array(new Int32Array(A).buffer);return NC(u,p,d,h,f,a,m,g,b,A.length,c),l}var lfe={kernelName:jl,backendName:"wasm",kernelFunc:ife,setupFunc:ofe},ufe=In($l),cfe=In(Zo),EC;function dfe(e){EC=e.wasm.cwrap(Pl,null,["number","number","number","number","number","number","array","number","number"])}function pfe(e){let{backend:t,inputs:n,attrs:s}=e,{indices:r,updates:a}=n,{shape:o}=s,i=t.makeOutput(o,a.dtype);if(v.sizeFromShape(o)===0)return i;let{sliceRank:l,numUpdates:u,sliceSize:c,strides:p,outputSize:d}=By.calculateShapes(a,r,o),f=t.dataIdMap.get(r.dataId).id,g=t.dataIdMap.get(a.dataId).id,y=new Uint8Array(new Int32Array(p).buffer),x=t.dataIdMap.get(i.dataId).id;return EC(f,g,Ht[a.dtype],l,u,c,y,d,x),i}var hfe={kernelName:Pl,backendName:"wasm",setupFunc:dfe,kernelFunc:pfe},RC;function ffe(e){RC=e.wasm.cwrap("SelectV2",null,["number","number","number","number","number"])}function mfe(e){let{inputs:t,backend:n}=e,{condition:s,t:r,e:a}=t,o=n.dataIdMap.get(s.dataId).id,i=n.dataIdMap.get(r.dataId).id,l=n.dataIdMap.get(a.dataId).id,u=n.makeOutput(r.shape,r.dtype),c=n.dataIdMap.get(u.dataId).id,p=s.shape.length,d=r.shape.length,h=p===0||p>1||d===1?1:v.sizeFromShape(r.shape.slice(1));return RC(o,i,l,h,c),u}var gfe={kernelName:Fl,backendName:"wasm",kernelFunc:mfe,setupFunc:ffe},_C;function yfe(e){_C=e.wasm.cwrap(Jo,null,["number","number"])}function Afe(e){let{backend:t,inputs:{x:n}}=e,s=t.dataIdMap.get(n.dataId).id,r=t.makeOutput(n.shape,n.dtype),a=t.dataIdMap.get(r.dataId).id;return v.sizeFromShape(r.shape)===0||_C(s,a),r}var xfe={kernelName:"Sigmoid",backendName:"wasm",setupFunc:yfe,kernelFunc:Afe},bfe=In(Yo),DC;function vfe(e){DC=e.wasm.cwrap(ti,null,["number","number","number","number"])}function wfe(e){let{backend:t,inputs:{logits:n},attrs:{dim:s}}=e,r=t.dataIdMap.get(n.dataId).id,a=t.makeOutput(n.shape,n.dtype),o=t.dataIdMap.get(a.dataId).id,i=n.shape[s],l=v.sizeFromShape(n.shape)/i;return v.sizeFromShape(a.shape)===0||DC(r,o,i,l),a}var kfe={kernelName:ti,backendName:"wasm",setupFunc:vfe,kernelFunc:wfe};function Ife(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockShape:a,paddings:o}=s,i=v.sizeFromShape(a),l=[[0,0]];l.push(...o);for(let k=1+a.length;k<r.shape.length;++k)l.push([0,0]);let u=wC.kernelFunc({inputs:{x:r},backend:n,attrs:{paddings:l,constantValue:0}}),c=T.getReshaped(u.shape,a,i,!1),p=T.getPermuted(c.length,a.length,!1),d=T.getReshapedPermuted(u.shape,a,i,!1),m=hs({inputs:{x:u},backend:n,attrs:{shape:c}}),x=lo({inputs:{x:m},backend:n,attrs:{perm:p}}),w=hs({inputs:{x},backend:n,attrs:{shape:d}});return n.disposeData(u.dataId),n.disposeData(m.dataId),n.disposeData(x.dataId),w}var Sfe={kernelName:zl,backendName:"wasm",kernelFunc:Ife},$C;function Cfe(e){$C=e.wasm.cwrap("SparseFillEmptyRows","number",["number","number","number","number","number","number","number","number","number","number","number","number"])}function Tfe(e){let{backend:t,inputs:n}=e,{indices:s,values:r,denseShape:a,defaultValue:o}=n,i=s.shape[0],l=s.shape[1],u=t.readSync(a.dataId)[0],c=[i+u,l],p=t.dataIdMap.get(s.dataId).id,d=t.dataIdMap.get(r.dataId).id,h=t.dataIdMap.get(o.dataId).id,f=t.makeOutput(c,s.dtype),m=t.dataIdMap.get(f.dataId).id,g=t.makeOutput(c.slice(0,1),r.dtype),y=t.dataIdMap.get(g.dataId).id,x=t.makeOutput([u],"bool"),A=t.dataIdMap.get(x.dataId).id,b=t.makeOutput([i],s.dtype),w=t.dataIdMap.get(b.dataId).id,k=t.makeOutput([4],"int32"),S=t.dataIdMap.get(k.dataId).id,E=$C(p,d,Ht[r.dtype],i,u,l,h,m,y,A,w,S),R=t.readSync(k.dataId),$;switch(R[0]){case 1:{$=T.getSparseFillEmptyRowsIndicesDenseShapeMismatch(R[1]);break}case 2:{$=T.getSparseFillEmptyRowsNegativeIndexErrorMessage(R[1],R[2]);break}case 3:$=T.getSparseFillEmptyRowsOutOfRangeIndexErrorMessage(R[1],R[2],R[3]);break;default:$=""}if(t.disposeData(k.dataId),$)throw t.disposeData(f.dataId),t.disposeData(g.dataId),t.disposeData(x.dataId),t.disposeData(b.dataId),new Error($);let _=f,P=g;return E!==c[0]&&(_=al({inputs:{x:f},attrs:{begin:0,size:[E,l]},backend:t}),P=al({inputs:{x:g},attrs:{begin:0,size:E},backend:t}),t.disposeData(f.dataId),t.disposeData(g.dataId)),[_,P,x,b]}var Nfe={kernelName:Xp,backendName:"wasm",setupFunc:Cfe,kernelFunc:Tfe},PC;function Efe(e){PC=e.wasm.cwrap(Oc,null,["number","number","number","number","number","number","number"])}function Rfe(e){let{backend:t,inputs:n}=e,{inputIndices:s,inputShape:r,newShape:a}=n;if(s.shape.length!==2)throw new Error(`Input indices should be a matrix but received shape
|
|
${s.shape}`);if(r.shape.length!==1)throw new Error(`Input shape should be a vector but received shape
|
|
${r.shape}`);if(a.shape.length!==1)throw new Error(`Target shape should be a vector but received shape ${a.shape}`);let o=t.dataIdMap.get(s.dataId).id,i=t.dataIdMap.get(r.dataId).id,l=t.dataIdMap.get(a.dataId).id,u=s.shape[0],c=v.sizeFromShape(a.shape),p=t.makeOutput([u,c],s.dtype),d=t.dataIdMap.get(p.dataId).id,h=t.makeOutput([c],a.dtype),f=t.dataIdMap.get(h.dataId).id,m=t.makeOutput([3],"int32"),g=t.dataIdMap.get(m.dataId).id;PC(o,i,l,u,d,f,g);let y=t.readSync(m.dataId),x;switch(y[0]){case 0:{x=T.getSparseReshapeMultipleNegativeOneOutputDimErrorMessage(y[1],y[2]);break}case 1:{x=T.getSparseReshapeNegativeOutputDimErrorMessage(y[1],y[2]);break}case 2:x=T.getSparseReshapeEmptyTensorZeroOutputDimErrorMessage();break;case 3:{let A=Array.from(t.readSync(r.dataId)),b=Array.from(t.readSync(h.dataId));x=T.getSparseReshapeInputOutputMultipleErrorMessage(A,b);break}case 4:{let A=Array.from(t.readSync(r.dataId)),b=Array.from(t.readSync(h.dataId));x=T.getSparseReshapeInputOutputMismatchErrorMessage(A,b);break}default:x=""}if(t.disposeData(m.dataId),x)throw t.disposeData(p.dataId),t.disposeData(h.dataId),new Error(x);return[p,h]}var _fe={kernelName:Oc,backendName:"wasm",setupFunc:Efe,kernelFunc:Rfe},FC;function OC(e){FC=e.wasm.cwrap("SparseSegmentReduction",null,["number","number","number","number","number","number","number","number","number"])}function MC(e,t){let{backend:n,inputs:s}=e,{data:r,indices:a,segmentIds:o}=s,i=a.shape[0],l=n.readSync(o.dataId,i-1,i)[0],c=i>0?l+1:0;if(c<0)throw new Error(T.getSparseSegmentReductionNegativeSegmentIdsErrorMessage());let p=r.shape.slice();p[0]=c;let d=n.dataIdMap.get(r.dataId).id,h=n.dataIdMap.get(a.dataId).id,f=n.dataIdMap.get(o.dataId).id,m=n.makeOutput(p,r.dtype),g=n.dataIdMap.get(m.dataId).id,y=n.makeOutput([4],"int32"),x=n.dataIdMap.get(y.dataId).id;FC(d,Ht[r.dtype],r.shape[0],h,f,g,x,t,0);let A=n.readSync(y.dataId),b;switch(A[0]){case 0:{b=T.getSparseSegmentReductionNegativeSegmentIdsErrorMessage();break}case 1:{b=T.getSparseSegmentReductionNonIncreasingSegmentIdsErrorMessage();break}case 2:b=T.getSparseSegmentReductionSegmentIdOutOfRangeErrorMessage(A[1],A[2]);break;case 3:b=T.getSparseSegmentReductionIndicesOutOfRangeErrorMessage(A[1],A[2],A[3]);break;default:b=""}if(n.disposeData(y.dataId),b)throw n.disposeData(m.dataId),new Error(b);return m}function Dfe(e){return MC(e,!0)}var $fe={kernelName:Kp,backendName:"wasm",setupFunc:OC,kernelFunc:Dfe};function Pfe(e){return MC(e,!1)}var Ffe={kernelName:Zp,backendName:"wasm",setupFunc:OC,kernelFunc:Pfe};function Ofe(e){let{inputs:t,attrs:n,backend:s}=e,{x:r}=t,{numOrSizeSplits:a,axis:o}=n,i=v.parseAxisParam(o,r.shape)[0],l=T.prepareSplitSize(r,a,i),u=new Array(r.shape.length).fill(0),c=r.shape.slice();return l.map(p=>{let d=[...c];d[i]=p;let h=al({inputs:{x:r},attrs:{begin:u,size:d},backend:s});return u[i]+=p,h})}var Mfe={kernelName:Ll,backendName:"wasm",kernelFunc:Ofe},zfe=In(Qo),Lfe=In(Mc),Bfe=!0,Wfe=Dn(ni,Bfe),zC;function Vfe(e){zC=e.wasm.cwrap(ai,null,["number","number","number","number"])}function Ufe(e){let{backend:t,inputs:n,attrs:s}=e,{alpha:r}=s,{x:a}=n,o=t.dataIdMap.get(a.dataId).id,i=t.makeOutput(a.shape,a.dtype),l=t.dataIdMap.get(i.dataId).id;return zC(o,r,Ht[a.dtype],l),i}var Gfe={kernelName:ai,backendName:"wasm",setupFunc:Vfe,kernelFunc:Ufe},LC;function Hfe(e){LC=e.wasm.cwrap(Bl,null,["number","array","number","array","array","array","array","array","number","number"])}function jfe(e){let{backend:t,inputs:n,attrs:s}=e,{x:r}=n,{begin:a,end:o,strides:i,beginMask:l,endMask:u,ellipsisMask:c,newAxisMask:p,shrinkAxisMask:d}=s,{finalShapeSparse:h,finalShape:f,isIdentity:m,sliceDim0:g,isSimpleSlice:y,begin:x,end:A,strides:b}=Ut.sliceInfo(r.shape,a,o,i,l,u,c,p,d),w;if(m)w=hs({inputs:{x:r},backend:t,attrs:{shape:f}});else if(g||y){v.assert(r.shape.length>=1,()=>`Input must have rank at least 1, got: ${r.shape.length}`);let k=Ut.computeOutShape(x,A,b),S=al({inputs:{x:r},backend:t,attrs:{begin:x,size:k}});w=hs({inputs:{x:S},backend:t,attrs:{shape:f}}),t.disposeData(S.dataId)}else{let k=t.makeOutput(h,"float32"),S=t.dataIdMap.get(r.dataId).id,E=new Uint8Array(new Int32Array(v.computeStrides(r.shape)).buffer),R=new Uint8Array(new Int32Array(x).buffer),$=new Uint8Array(new Int32Array(A).buffer),_=new Uint8Array(new Int32Array(b).buffer),P=new Uint8Array(new Int32Array(h).buffer),C=new Uint8Array(new Int32Array(v.computeStrides(h)).buffer),F=t.dataIdMap.get(k.dataId).id;LC(S,E,r.shape.length,R,$,_,P,C,h.length,F),w=hs({inputs:{x:k},backend:t,attrs:{shape:f}}),t.disposeData(k.dataId)}return w}var qfe={kernelName:Bl,backendName:"wasm",setupFunc:Hfe,kernelFunc:jfe};function Xfe(e){let{backend:t,inputs:n,attrs:s}=e,{data:r,dataSplits:a}=n,{separator:o,nGramWidths:i,leftPad:l,rightPad:u,padWidth:c,preserveShortSequences:p}=s,d=t.readSync(r.dataId),h=t.readSync(a.dataId),[f,m]=Px(d,h,o,i,l,u,c,p),g=t.makeOutput([f.length],"string"),y=t.dataIdMap.get(g.dataId);y.stringBytes=f;let x=t.makeOutput(a.shape,"int32");return t.typedArrayFromHeap(x).set(m),[g,x]}var Kfe={kernelName:zc,backendName:"wasm",kernelFunc:Xfe};function Zfe(e){let{backend:t,inputs:n,attrs:s}=e,{input:r,delimiter:a}=n,{skipEmpty:o}=s,i=t.readSync(r.dataId),l=t.readSync(a.dataId),[u,c,p]=Fx(i,l[0],o),d=c.length,h=t.makeOutput([d,2],"int32");t.typedArrayFromHeap(h).set(u);let m=t.makeOutput([d],"string"),g=t.dataIdMap.get(m.dataId);g.stringBytes=c;let y=t.makeOutput([2],"int32");return t.typedArrayFromHeap(y).set(p),[h,m,y]}var Yfe={kernelName:Jp,backendName:"wasm",kernelFunc:Zfe};function Jfe(e){let{backend:t,inputs:n,attrs:s}=e,{input:r}=n,{numBuckets:a}=s,o=t.readSync(r.dataId),i=Ox(o,a),l=t.makeOutput(r.shape,"int32");return t.typedArrayFromHeap(l).set(i),l}var Qfe={kernelName:Qp,backendName:"wasm",kernelFunc:Jfe},eme=!0,tme=Dn(si,eme),BC;function nme(e){BC=e.wasm.cwrap(ei,null,["number","number","number","number"])}function sme(e){let{backend:t,inputs:n,attrs:s}=e,{axis:r,keepDims:a}=s,{x:o}=n,i=t.dataIdMap.get(o.dataId).id,l=i,u=o,{transposed:c,axes:p,originalAxes:d,inputWasTransposed:h}=hi(o,r,t),f=p;if(h){let A=t.dataIdMap.get(c.dataId).id;A!==i&&(u=c,l=A,f=T.getInnerMostAxes(f.length,u.shape.length))}T.assertAxesAreInnerMostDims("sum",f,u.shape.length);let[m,g]=T.computeOutAndReduceShapes(u.shape,f),y=v.sizeFromShape(g),x=t.makeOutput(m,u.dtype);if(v.sizeFromShape(u.shape)!==0){let A=t.dataIdMap.get(x.dataId).id;BC(l,y,Ht[x.dtype],A)}if(h&&t.disposeData(c.dataId),a){let A=T.expandShapeToKeepDim(x.shape,d);x.shape=A}return x}var rme={kernelName:ei,backendName:"wasm",setupFunc:nme,kernelFunc:sme},ame=In(Wl),ome=In(ri),WC;function ime(e){WC=e.wasm.cwrap(va,null,["number","array","number","array","number","number"])}function lme(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,a=n.dataIdMap.get(r.dataId).id,{reps:o}=s,i=new Array(r.shape.length);for(let d=0;d<i.length;d++)i[d]=r.shape[d]*o[d];let l=new Uint8Array(new Int32Array(r.shape).buffer),u=new Uint8Array(new Int32Array(i).buffer),c=n.makeOutput(i,r.dtype),p=n.dataIdMap.get(c.dataId).id;return WC(a,l,r.shape.length,u,i.length,Ht[c.dtype],p),c}var ume={kernelName:va,backendName:"wasm",setupFunc:ime,kernelFunc:lme},VC;function cme(e){VC=e.wasm.cwrap(Vl,null,["number","array","number","number","number","bool","number","number"])}var dme=({inputs:e,backend:t,attrs:n})=>{let{x:s}=e,{k:r,sorted:a}=n,o=t.dataIdMap.get(s.dataId).id,i=new Uint8Array(new Int32Array(s.shape).buffer),l=s.shape.slice();l[l.length-1]=r;let u=t.makeOutput(l,s.dtype),c=t.dataIdMap.get(u.dataId).id,p=t.makeOutput(l,"int32"),d=t.dataIdMap.get(p.dataId).id;return VC(o,i,s.shape.length,Ht[s.dtype],r,a,c,d),[u,p]},pme={kernelName:Vl,backendName:"wasm",setupFunc:cme,kernelFunc:dme},UC;function hme(e){UC=e.wasm.cwrap(Ul,null,["number","number","bool","number","number","number","number","number","number","array","number","number","number","number","number"])}function fme(e){let{backend:t,inputs:n,attrs:s}=e,{image:r,transforms:a}=n,{interpolation:o,fillMode:i,fillValue:l,outputShape:u}=s,[c,p,d,h]=r.shape,[f,m]=u!=null?u:[p,d],g=[c,f,m,h],y=new Uint8Array(new Int32Array(v.computeStrides(r.shape)).buffer),x=t.makeOutput(g,r.dtype),A=t.dataIdMap.get(x.dataId).id,w=t.dataIdMap.get(r.dataId).id,S=t.dataIdMap.get(a.dataId).id,E=o==="nearest"?1:2,R;switch(i){case"constant":R=1;break;case"reflect":R=2;break;case"wrap":R=3;break;case"nearest":R=4;break;default:R=1;break}return UC(w,S,a.shape[0]>1,c,f,m,h,d,p,y,r.shape.length-1,E,R,l,A),x}var mme={kernelName:Ul,backendName:"wasm",setupFunc:hme,kernelFunc:fme};function gme(e){let{inputs:t,backend:n,attrs:s}=e,{value:r}=t,{axis:a}=s;a<0&&(a+=r.shape.length);let o=r.shape[a],i=r.shape.length,l=new Array(i-1),u=0;for(let h=0;h<i;h++)h!==a&&(l[u++]=r.shape[h]);let c=new Array(o),p=new Array(i).fill(0),d=r.shape.slice();d[a]=1;for(let h=0;h<c.length;h++)p[a]=h,c[h]=al({inputs:{x:r},attrs:{begin:p,size:d},backend:n});return c.map(({dataId:h,dtype:f})=>({dataId:h,dtype:f,shape:l}))}var yme={kernelName:Gl,backendName:"wasm",kernelFunc:gme};function Ame(e){let{inputs:{x:t},backend:n}=e,s=n.makeOutput(t.shape,t.dtype);return n.typedArrayFromHeap(s).fill(0),s}var xme={kernelName:Hl,backendName:"wasm",kernelFunc:Ame},bme=[Xce,Kce,Yce,ede,lde,dde,fde,yde,vde,Tde,Nde,Ede,Dde,$de,Ode,Lde,Bde,Wde,Gde,qde,Zde,Qde,npe,spe,ape,ope,ipe,lpe,dpe,ppe,fpe,ype,bpe,kpe,Cpe,Epe,_pe,$pe,tde,Ope,zpe,Bpe,Wpe,Upe,Gpe,jpe,Xpe,Ype,Qpe,nhe,ahe,lhe,che,hhe,mhe,ghe,xhe,whe,She,The,Rhe,Dhe,Phe,wC,zhe,Whe,Ghe,jhe,Xhe,Khe,Zhe,Ade,Qhe,nfe,afe,lfe,ufe,cfe,hfe,gfe,xfe,bfe,Sde,kfe,Sfe,Nfe,_fe,$fe,Ffe,Mfe,zfe,Lfe,Wfe,Gfe,qfe,Kfe,Yfe,Qfe,tme,rme,ame,ome,ume,pme,mme,ade,yme,xme];for(let e of bme)dr(e);var py=q();py.registerFlag("WASM_HAS_SIMD_SUPPORT",async()=>WebAssembly.validate(new Uint8Array([0,97,115,109,1,0,0,0,1,4,1,96,0,0,3,2,1,0,10,9,1,7,0,65,0,253,15,26,11])));py.registerFlag("WASM_HAS_MULTITHREAD_SUPPORT",async()=>{if(py.get("IS_NODE"))return!1;try{return new MessageChannel().port1.postMessage(new SharedArrayBuffer(1)),WebAssembly.validate(new Uint8Array([0,97,115,109,1,0,0,0,1,4,1,96,0,0,3,2,1,0,5,4,1,3,1,1,10,11,1,9,0,65,0,254,16,2,0,26,11]))}catch(e){return!1}});var M7=uo(k_()),vme=uo(I_()),z7=uo(S_()),L7=M7.default||M7,wme=z7.default||z7,GC=class extends cc{constructor(e){super(),this.wasm=e,this.dataIdNextNumber=1,this.wasm.tfjs.initWithThreadsCount(HC),hy=this.wasm.tfjs.getThreadsCount(),this.dataIdMap=new Mp(this,sn())}write(e,t,n){let s={id:this.dataIdNextNumber++};return this.move(s,e,t,n,1),s}numDataIds(){return this.dataIdMap.numDataIds()}async time(e){let t=v.now();return e(),{kernelMs:v.now()-t}}move(e,t,n,s,r){let a=this.dataIdNextNumber++;if(s==="string"){let u=t;this.dataIdMap.set(e,{id:a,stringBytes:u,shape:n,dtype:s,memoryOffset:null,refCount:r});return}let o=v.sizeFromShape(n),i=o*v.bytesPerElement(s),l=this.wasm._malloc(i);this.dataIdMap.set(e,{id:a,memoryOffset:l,shape:n,dtype:s,refCount:r}),this.wasm.tfjs.registerTensor(a,o,l),t!=null&&this.wasm.HEAPU8.set(new Uint8Array(t.buffer,t.byteOffset,i),l)}async read(e){return this.readSync(e)}readSync(e,t,n){let{memoryOffset:s,dtype:r,shape:a,stringBytes:o}=this.dataIdMap.get(e);if(r==="string")return(t==null||t===0)&&(n==null||n>=o.length)?o:o.slice(t,n);t=t||0,n=n||v.sizeFromShape(a);let i=v.bytesPerElement(r),l=this.wasm.HEAPU8.slice(s+t*i,s+n*i);return Sme(l.buffer,r)}disposeData(e,t=!1){if(this.dataIdMap.has(e)){let n=this.dataIdMap.get(e);if(n.refCount--,!t&&n.refCount>0)return!1;this.wasm._free(n.memoryOffset),this.wasm.tfjs.disposeData(n.id),this.dataIdMap.delete(e)}return!0}refCount(e){return this.dataIdMap.has(e)?this.dataIdMap.get(e).refCount:0}incRef(e){let t=this.dataIdMap.get(e);t!=null&&t.refCount++}floatPrecision(){return 32}getMemoryOffset(e){return this.dataIdMap.get(e).memoryOffset}dispose(){this.wasm.tfjs.dispose(),"PThread"in this.wasm&&this.wasm.PThread.terminateAllThreads(),this.wasm=null}memory(){return{unreliable:!1}}makeOutput(e,t,n){let s;if(n==null)s=this.write(null,e,t);else{let r=this.dataIdNextNumber++;s={id:r},this.dataIdMap.set(s,{id:r,memoryOffset:n,shape:e,dtype:t,refCount:1});let a=v.sizeFromShape(e);this.wasm.tfjs.registerTensor(r,a,n)}return{dataId:s,shape:e,dtype:t}}typedArrayFromHeap({shape:e,dtype:t,dataId:n}){let s=this.wasm.HEAPU8.buffer,{memoryOffset:r}=this.dataIdMap.get(n),a=v.sizeFromShape(e);switch(t){case"float32":return new Float32Array(s,r,a);case"int32":return new Int32Array(s,r,a);case"bool":return new Uint8Array(s,r,a);default:throw new Error(`Unknown dtype ${t}`)}}};function kme(e){return(t,n)=>(v.fetch(e,{credentials:"same-origin"}).then(s=>{s.ok||t.env.a(`failed to load wasm binary file at '${e}'`),s.arrayBuffer().then(r=>{WebAssembly.instantiate(r,t).then(a=>{n(a.instance,a.module)})})}),{})}function B7(e,t,n){if(zm!=null)return zm;let s="tfjs-backend-wasm.wasm";return e&&t?s="tfjs-backend-wasm-threaded-simd.wasm":e&&(s="tfjs-backend-wasm-simd.wasm"),gp!=null&&gp[s]!=null?gp[s]:n+s}async function Ime(){let[e,t]=await Promise.all([q().getAsync("WASM_HAS_SIMD_SUPPORT"),q().getAsync("WASM_HAS_MULTITHREAD_SUPPORT")]);return new Promise((n,s)=>{let r={};r.locateFile=(i,l)=>{if(i.endsWith(".worker.js")){let u=vme.wasmWorkerContents.replace(/\n/g,"\\n"),c=new Blob([u],{type:"application/javascript"});return URL.createObjectURL(c)}return i.endsWith(".wasm")?B7(e,t,pp!=null?pp:l):l+i},tb&&(r.instantiateWasm=kme(B7(e,t,pp!=null?pp:"")));let a=!1;r.onAbort=()=>{if(a||yp)return;yp=!0,s({message:"Make sure the server can serve the `.wasm` file relative to the bundled js file. For more details see https://github.com/tensorflow/tfjs/blob/master/tfjs-backend-wasm/README.md#using-bundlers"})};let o;t&&e&&zm==null?(r.mainScriptUrlOrBlob=new Blob(["var WasmBackendModuleThreadedSimd = "+L7.toString()],{type:"text/javascript"}),o=L7(r)):o=wme(r),o.then(i=>{a=!0,yp=!1;let l=null;i.tfjs={init:i.cwrap("init",null,[]),initWithThreadsCount:i.cwrap("init_with_threads_count",null,["number"]),getThreadsCount:i.cwrap("get_threads_count","number",[]),registerTensor:i.cwrap("register_tensor",null,["number","number","number"]),disposeData:i.cwrap("dispose_data",l,["number"]),dispose:i.cwrap("dispose",l,[])},n({wasm:i})}).catch(s)})}function Sme(e,t){switch(t){case"float32":return new Float32Array(e);case"int32":return new Int32Array(e);case"bool":return new Uint8Array(e);default:throw new Error(`Unknown dtype ${t}`)}}var Cme=["tfjs-backend-wasm.wasm","tfjs-backend-wasm-simd.wasm","tfjs-backend-wasm-threaded-simd.wasm"],zm=null,pp=null,gp={},yp=!1,tb=!1;function Tme(e,t=!1){if(Fy("setWasmPath has been deprecated in favor of setWasmPaths and will be removed in a future release."),yp)throw new Error("The WASM backend was already initialized. Make sure you call `setWasmPath()` before you call `tf.setBackend()` or `tf.ready()`");zm=e,tb=t}function nb(e,t=!1){if(yp)throw new Error("The WASM backend was already initialized. Make sure you call `setWasmPaths()` before you call `tf.setBackend()` or `tf.ready()`");if(typeof e=="string")pp=e;else{gp=e;let n=Cme.filter(s=>gp[s]==null);if(n.length>0)throw new Error(`There were no entries found for the following binaries: ${n.join(",")}. Please either call setWasmPaths with a map providing a path for each binary, or with a string indicating the directory where all the binaries can be found.`)}tb=t}var HC=-1,hy=-1;function Nme(e){HC=e}function Eme(){if(hy===-1)throw new Error("WASM backend not initialized.");return hy}var Rme="3.19.0",_me=2;ql("wasm",async()=>{let{wasm:e}=await Ime();return new GC(e)},_me);var Sa=q();Sa.registerFlag("WEBGPU_DEFERRED_SUBMIT_BATCH_SIZE",()=>15);Sa.registerFlag("WEBGPU_CPU_FORWARD",()=>!0);Sa.registerFlag("WEBGPU_MATMUL_WORK_PER_THREAD",()=>4);Sa.registerFlag("WEBGPU_MATMUL_PROGRAM_TYPE",()=>-1);Sa.registerFlag("WEBGPU_USE_NAIVE_CONV2D_TRANSPOSE",()=>!1);Sa.registerFlag("WEBGPU_USE_LOW_POWER_GPU",()=>!1);Sa.registerFlag("WEBGPU_CPU_HANDOFF_SIZE_THRESHOLD",()=>1e3);Sa.registerFlag("WEBGPU_USE_PROFILE_TOOL",()=>!1);Sa.registerFlag("WEBGPU_IMPORT_EXTERNAL_TEXTURE",()=>!1);var Ye;(function(e){e[e.MUL=0]="MUL",e[e.ADD=1]="ADD",e[e.SUB=2]="SUB",e[e.DIV=3]="DIV",e[e.EQUAL=4]="EQUAL",e[e.GREATER=5]="GREATER",e[e.GREATER_EQUAL=6]="GREATER_EQUAL",e[e.LESS=7]="LESS",e[e.LESS_EQUAL=8]="LESS_EQUAL",e[e.LOGICAL_AND=9]="LOGICAL_AND",e[e.NOT_EQUAL=10]="NOT_EQUAL",e[e.SQUARED_DIFFERENCE=11]="SQUARED_DIFFERENCE",e[e.INT_DIV=12]="INT_DIV",e[e.POW=13]="POW",e[e.PRELU=14]="PRELU",e[e.MAX=15]="MAX",e[e.MIN=16]="MIN",e[e.COMPLEX_MULTIPLY_REAL=17]="COMPLEX_MULTIPLY_REAL",e[e.COMPLEX_MULTIPLY_IMAG=18]="COMPLEX_MULTIPLY_IMAG"})(Ye||(Ye={}));var Dme="return a + b;",$me="return areal * breal - aimag * bimag;",Pme="return areal * bimag + aimag * breal;",Fme="return a / b;",Ome="return a * b;",Mme="return (a - b) * (a - b);",zme="return a - b;",Lme="return f32(a == b);",Bme="return vec4<f32>(a == b);",Wme="return f32(a > b);",Vme="return vec4<f32>(a > b);",Ume="return f32(a >= b);",Gme="return vec4<f32>(a >= b);",Hme="return f32(a < b);",jme="return vec4<f32>(a < b);",qme="return f32(a <= b);",Xme="return vec4<f32>(a <= b);",Kme="return f32(f32(a) >= 1.0 && f32(b) >= 1.0);",Zme=`return (vec4<f32>(a >= vec4<f32>(1.0)) *
|
|
vec4<f32>(b >= vec4<f32>(1.0)));`,Yme=`
|
|
if (isnan(a)) { return a; }
|
|
if (isnan(b)) { return b; }
|
|
`,jC=`
|
|
if (isNaN.r) {
|
|
resultTemp.r = uniforms.NAN;
|
|
}
|
|
if (isNaN.g) {
|
|
resultTemp.g = uniforms.NAN;
|
|
}
|
|
if (isNaN.b) {
|
|
resultTemp.b = uniforms.NAN;
|
|
}
|
|
if (isNaN.a) {
|
|
resultTemp.a = uniforms.NAN;
|
|
}
|
|
`,Jme=`
|
|
let s = sign(a) * sign(b);
|
|
let ia = i32(round(a));
|
|
let ib = i32(round(b));
|
|
return f32(idiv(ia, ib, s));
|
|
`,Qme=`
|
|
let ia = vec4<i32>(round(a));
|
|
let ib = vec4<i32>(round(b));
|
|
let cond = ib != vec4<i32>(0);
|
|
var resultTemp = vec4<i32>(0);
|
|
let s = sign(a) * sign(b);
|
|
|
|
// Windows (D3D) wants guaranteed non-zero int division at compile-time.
|
|
if (cond[0]) {
|
|
resultTemp[0] = idiv(ia[0], ib[0], s[0]);
|
|
}
|
|
if (cond[1]) {
|
|
resultTemp[1] = idiv(ia[1], ib[1], s[1]);
|
|
}
|
|
if (cond[2]) {
|
|
resultTemp[2] = idiv(ia[2], ib[2], s[2]);
|
|
}
|
|
if (cond[3]) {
|
|
resultTemp[3] = idiv(ia[3], ib[3], s[3]);
|
|
}
|
|
return vec4<f32>(resultTemp);
|
|
`,e0e="return f32(a != b);",t0e="return vec4<f32>(a != b);",n0e=`
|
|
if(a < 0.0 && floor(b) < b) {
|
|
return uniforms.NAN;
|
|
}
|
|
if (b == 0.0) {
|
|
return 1.0;
|
|
}
|
|
if (round(abs(b) % 2.0) != 1.0) {
|
|
return pow(abs(a), b);
|
|
}
|
|
return sign(a) * pow(abs(a), b);
|
|
`,s0e=`
|
|
let isModRound1Bool = vec4<i32>(round(abs(b) % vec4<f32>(2.0))) == vec4<i32>(1);
|
|
let isModRound1 = vec4<f32>(isModRound1Bool);
|
|
let multiplier = sign(a) * isModRound1 + (vec4<f32>(1.0) - isModRound1);
|
|
var resultTemp = multiplier * pow(abs(a), b);
|
|
|
|
// Ensure that a^0 = 1, including 0^0 = 1 as this correspond to TF and JS
|
|
let isExpZero = b == vec4<f32>(0.0);
|
|
if (isExpZero.r) {
|
|
resultTemp.r = 1.0;
|
|
}
|
|
if (isExpZero.g) {
|
|
resultTemp.g = 1.0;
|
|
}
|
|
if (isExpZero.b) {
|
|
resultTemp.b = 1.0;
|
|
}
|
|
if (isExpZero.a) {
|
|
resultTemp.a = 1.0;
|
|
}
|
|
let isNaN = a < vec4<f32>(0.0) & floor(b) < b;
|
|
${jC}
|
|
return resultTemp;
|
|
`,r0e="if (a < 0.0) { return b * a; } return a;",a0e=`
|
|
let aLessThanZero = vec4<f32>(a < vec4<f32>(0.0));
|
|
return (aLessThanZero * (b * a)) + ((vec4<f32>(1.0) - aLessThanZero) * a);
|
|
`;function W7(e,t){let n=t?jC:Yme;return t?`
|
|
var resultTemp = vec4<f32>(${e}(a, b));
|
|
let isNaN = isnanVec4(a) | isnanVec4(b);
|
|
`+n+`
|
|
return resultTemp;
|
|
`:n+`
|
|
return ${e}(a, b);
|
|
`}function Lm(e,t){switch(e){case Ye.MUL:return Ome;case Ye.ADD:return Dme;case Ye.SUB:return zme;case Ye.DIV:return Fme;case Ye.EQUAL:return t?Bme:Lme;case Ye.GREATER:return t?Vme:Wme;case Ye.GREATER_EQUAL:return t?Gme:Ume;case Ye.LESS:return t?jme:Hme;case Ye.LESS_EQUAL:return t?Xme:qme;case Ye.LOGICAL_AND:return t?Zme:Kme;case Ye.NOT_EQUAL:return t?t0e:e0e;case Ye.SQUARED_DIFFERENCE:return Mme;case Ye.INT_DIV:return t?Qme:Jme;case Ye.PRELU:return t?a0e:r0e;case Ye.MAX:return W7("max",t);case Ye.MIN:return W7("min",t);case Ye.POW:return t?s0e:n0e;case Ye.COMPLEX_MULTIPLY_REAL:return $me;case Ye.COMPLEX_MULTIPLY_IMAG:return Pme;default:throw new Error(`BinaryType ${e} is not implemented!`)}}var Fe;(function(e){e[e.ABS=0]="ABS",e[e.CEIL=1]="CEIL",e[e.COS=2]="COS",e[e.COSH=3]="COSH",e[e.ELU=4]="ELU",e[e.EXP=5]="EXP",e[e.EXPM1=6]="EXPM1",e[e.FLOOR=7]="FLOOR",e[e.LINEAR=8]="LINEAR",e[e.LOG=9]="LOG",e[e.LOGICAL_NOT=10]="LOGICAL_NOT",e[e.NEG=11]="NEG",e[e.RELU=12]="RELU",e[e.RELU6=13]="RELU6",e[e.LEAKYRELU=14]="LEAKYRELU",e[e.RSQRT=15]="RSQRT",e[e.SIN=16]="SIN",e[e.SINH=17]="SINH",e[e.SIGMOID=18]="SIGMOID",e[e.SQRT=19]="SQRT",e[e.SQUARE=20]="SQUARE",e[e.TANH=21]="TANH",e[e.TO_INT=22]="TO_INT"})(Fe||(Fe={}));var o0e="return abs(a);",i0e="return ceil(a);",l0e="return cos(a);",u0e=`
|
|
let e2x = exp(-a);
|
|
return (e2x + 1.0 / e2x) / 2.0;
|
|
`,c0e="return exp(a) - 1.0;",d0e="if (a >= 0.0) { return a; } return (exp(a) - 1.0);",p0e=`
|
|
var resFloat = exp(a) - vec4<f32>(1.0);
|
|
if (a.r >= 0.0) {
|
|
resFloat.r = a.r;
|
|
}
|
|
if (a.g >= 0.0) {
|
|
resFloat.g = a.g;
|
|
}
|
|
if (a.b >= 0.0) {
|
|
resFloat.b = a.b;
|
|
}
|
|
if (a.a >= 0.0) {
|
|
resFloat.a = a.a;
|
|
}
|
|
return resFloat;
|
|
`,h0e="return exp(a);",f0e="return floor(a);",m0e="return a;",g0e=`if (a < 0.0) { return 1.0/0.0; }
|
|
return log(a);`,y0e="return f32(!(a >= 1.0));",A0e="return -a;",x0e="if (a < 0.0) { return uniforms.alpha * a; } return a;",b0e=`
|
|
let aLessThanZero = vec4<f32>(a < vec4<f32>(0.0));
|
|
return (aLessThanZero * (uniforms.alpha * a)) + ((vec4<f32>(1.0) - aLessThanZero) * a);
|
|
`,v0e="return select(a, 0.0, a < 0.0);",w0e="return clamp(a, 0.0, 6.0);",k0e="return clamp(a, vec4<f32>(0.0, 0.0, 0.0, 0.0), vec4<f32>(6.0, 6.0, 6.0, 6.0));",I0e=`
|
|
return select(a, vec4<f32>(0.0), a < vec4<f32>(0.0));
|
|
`,S0e="return 1.0/sqrt(a);",C0e="return 1.0 / (1.0 + exp(-1.0 * a));",T0e="return sin(a);",N0e=`
|
|
let e2x = exp(a);
|
|
return (e2x - 1.0 / e2x) / 2.0;
|
|
`,E0e="return sqrt(a);",R0e="return a * a;",_0e=`
|
|
let e2x = exp(-2.0 * abs(a));
|
|
return sign(a) * (1.0 - e2x) / (1.0 + e2x);
|
|
`,D0e="return f32(i32((a)));";function Oi(e,t){switch(e){case Fe.ABS:return o0e;case Fe.COS:return l0e;case Fe.COSH:return u0e;case Fe.CEIL:return i0e;case Fe.ELU:return t?p0e:d0e;case Fe.EXP:return h0e;case Fe.EXPM1:return c0e;case Fe.FLOOR:return f0e;case Fe.LINEAR:return m0e;case Fe.LOG:return g0e;case Fe.LOGICAL_NOT:return y0e;case Fe.NEG:return A0e;case Fe.LEAKYRELU:return t?b0e:x0e;case Fe.RELU:return t?I0e:v0e;case Fe.RELU6:return t?k0e:w0e;case Fe.RSQRT:return S0e;case Fe.SIGMOID:return C0e;case Fe.SIN:return T0e;case Fe.SINH:return N0e;case Fe.SQRT:return E0e;case Fe.SQUARE:return R0e;case Fe.TANH:return _0e;case Fe.TO_INT:return D0e;default:throw new Error(`BinaryType ${e} is not implemented!`)}}var nn=e=>{switch(e){case 1:return"f32";case 2:return"vec2<f32>";case 3:return"vec3<f32>";case 4:return"vec4<f32>";default:throw new Error(`${e}-component is not supported.`)}};function Ca(e,t=!1,n=!1,s=3){if(e===null)return"";let r="";if(e==="linear")r=Oi(Fe.LINEAR);else if(e==="relu")r=Oi(Fe.RELU,n);else if(e==="elu")r=Oi(Fe.ELU,n);else if(e==="relu6")r=Oi(Fe.RELU6,n);else if(e==="prelu")r=Lm(Ye.PRELU,n);else if(e==="sigmoid")r=Oi(Fe.SIGMOID,n);else if(e==="leakyrelu")r=Oi(Fe.LEAKYRELU,n);else throw new Error(`Activation ${e} has not been implemented for the WebGPU backend.`);let o=nn(n?4:1),i="";return t?i=`
|
|
fn activation(a : ${o}, coords : vec${s}<i32>) -> ${o} {
|
|
let b = getPreluActivationWeightsByOutputCoords(coords);
|
|
${r}
|
|
}`:i=`
|
|
fn activation(a : ${o}, coords : vec${s}<i32>) -> ${o} {
|
|
${r}
|
|
}`,i}function dd(e,t){return`
|
|
${e?"value = value + getBiasByOutputCoords(coords);":""}
|
|
${t?"value = activation(value, coords);":""}
|
|
`}function $0e(e,t){if(Math.max(...e)>3)throw new Error("Cannot symbolically compute strides for rank > 4 tensor.");let n=e.length,s=e.map(a=>`${t}[${a}]`),r=new Array(n-1);r[n-2]=s[n-1];for(let a=n-3;a>=0;--a)r[a]=`(${r[a+1]} * ${s[a+1]})`;return r}var P0e=(e,t,n,s)=>{let r={dtype:s.dtype,shape:s.shape},a=F0e(n,r,t),o=e.createShaderModule({code:a,label:t.constructor.name});return e.createComputePipeline({compute:{module:o,entryPoint:"main"},label:t.constructor.name,layout:"auto"})};function Tn(e){if(e<=1)return"i32";if(e===2)return"vec2<i32>";if(e===3)return"vec3<i32>";if(e===4)return"vec4<i32>";if(e===5)return"vec5";if(e===6)return"vec6";throw Error(`GPU for rank ${e} is not yet supported`)}function Za(e){if(e===0)return"x";if(e===1)return"y";if(e===2)return"z";if(e===3)return"w";if(e===4)return"u";if(e===5)return"v";throw Error(`Index ${e} is not yet supported`)}function lt(){return`
|
|
${pd()}
|
|
let index = getGlobalIndex();
|
|
`}function pd(){return`
|
|
${T2()}
|
|
fn main(@builtin(local_invocation_id) LocalId : vec3<u32>,
|
|
@builtin(global_invocation_id) GlobalId : vec3<u32>,
|
|
@builtin(num_workgroups) NumWorkgroups: vec3<u32>) {
|
|
localId = LocalId;
|
|
globalId = GlobalId;
|
|
numWorkgroups = NumWorkgroups;
|
|
`}function T2(){return`
|
|
@compute @workgroup_size(workGroupSizeX, workGroupSizeY, workGroupSizeZ)
|
|
`}function F0e(e,t,n){let s=[];if(s.push(`
|
|
const workGroupSizeX = ${n.workGroupSize[0]}u;
|
|
const workGroupSizeY = ${n.workGroupSize[1]}u;
|
|
const workGroupSizeZ = ${n.workGroupSize[2]}u;
|
|
|
|
var<private> localId: vec3<u32>;
|
|
var<private> globalId: vec3<u32>;
|
|
var<private> numWorkgroups: vec3<u32>;
|
|
|
|
// Only used when the y/z dimension of workgroup size is 1.
|
|
fn getGlobalIndex() -> i32 {
|
|
${qC(n)?" return i32(globalId.x);":` let localInvocationIndex = localId.z * workGroupSizeX * workGroupSizeY +
|
|
localId.y * workGroupSizeX + localId.x;
|
|
let workGroupID = (globalId - localId)/vec3<u32>(
|
|
workGroupSizeX, workGroupSizeY, workGroupSizeZ);
|
|
|
|
return i32((workGroupID.z * numWorkgroups.x * numWorkgroups.y +
|
|
workGroupID.y * numWorkgroups.x + workGroupID.x) *
|
|
(workGroupSizeX * workGroupSizeY * workGroupSizeZ) +
|
|
localInvocationIndex);
|
|
`}
|
|
}
|
|
`),n.isFromPixels)return s.push(`
|
|
struct Uniform {
|
|
size : i32,
|
|
numChannels : i32,
|
|
outShapeStrides : vec2<i32>,
|
|
};
|
|
|
|
@group(0) @binding(0) var<storage, read_write> result: array<${Ap(t.dtype,n.isVec4)}>;
|
|
@group(0) @binding(2) var<uniform> uniforms: Uniform;
|
|
`),[V7,s.join(`
|
|
`),U7(t.shape),n.getUserCode()].join(`
|
|
`);let r=!1,a=!1,o="struct Uniforms { NAN : f32, ";n.variableNames.forEach((f,m)=>{let g=Tn(e[m].shape.length);(g==="vec5"||g==="vec6")&&(a=!0),(r||a)&&(o+="@align(16) "),r=a,o+=`${f.charAt(0).toLowerCase()+f.slice(1)}Shape : ${g}, `});let i=Tn(t.shape.length);a=i==="vec5"||i==="vec6",(r||a)&&(o+="@align(16) "),r=a,o+=`outShape : ${i}, `;let l=t.shape.length-1,u=Tn(l);a=u==="vec5"||u==="vec6",(r||a)&&(o+="@align(16) "),r=a,o+=`
|
|
outShapeStrides: ${u}, `,n.size&&(r&&(o+="@align(16) "),r=!1,o+="size : i32, "),n.uniforms&&(r&&(o+="@align(16) "),o+=n.uniforms),o+="};",s.push(o),n.atomic?s.push(`
|
|
@group(0) @binding(0) var<storage, read_write> result: array<atomic<i32>>;
|
|
`):s.push(`
|
|
@group(0) @binding(0) var<storage, read_write> result: array<${Ap(t.dtype,n.isVec4)}>;
|
|
`),n.variableNames.forEach((f,m)=>{s.push(`
|
|
@group(0) @binding(${1+m}) var<storage, read> ${f}: array<${n.variableTypes?n.variableTypes[m]:Ap(e[m].dtype,n.isVec4)}>;
|
|
`)}),o!==""&&s.push(`
|
|
@group(0) @binding(${1+n.variableNames.length}) var<uniform> uniforms: Uniforms;
|
|
`);let c=B0e(t.shape,n.dispatchLayout),p=[V7,s.join(`
|
|
`),U7(t.shape),c,W0e(t.shape.length)];n.atomic||p.push(V0e(t.shape,t.dtype,n.isVec4));let d=e.map((f,m)=>L0e(f,t.shape,n.variableTypes?n.variableTypes[m]==="vec4<f32>":n.isVec4,n.dispatchLayout.x.length===t.shape.length)).join(`
|
|
`);return p.push(d),p.push(n.getUserCode()),p.join(`
|
|
`)}function O0e(e,t,n,s){let r=e.shaderKey;if(e.isFromPixels)return r;let a=n.map(c=>c.dtype).concat(s.dtype),o=n.map(c=>T.getBroadcastDims(c.shape,s.shape)),i=n.map(c=>v.arraysEqual(c.shape,s.shape)).join("_"),l=o.map(c=>c.join("_")).join(";"),u=qC(e)?"flatDispatch":"";return r+="_"+(e.workGroupSize?e.workGroupSize.join(","):"")+t.map(c=>c.length).join(",")+a.join(",")+e.variableNames.join(",")+l+i+u,r}var V7=`
|
|
struct vec5 {x: i32, y: i32, z: i32, w: i32, u: i32};
|
|
struct vec6 {x: i32, y: i32, z: i32, w: i32, u: i32, v: i32};
|
|
|
|
// Checks whether coordinates lie within the bounds of the shape.
|
|
fn coordsInBounds2D(coord : vec2<i32>, shape : vec2<i32>) -> bool {
|
|
return all(coord >= vec2<i32>(0)) && all(coord < shape);
|
|
}
|
|
fn coordsInBounds3D(coord : vec3<i32>, shape : vec3<i32>) -> bool {
|
|
return all(coord >= vec3<i32>(0)) && all(coord < shape);
|
|
}
|
|
fn coordsInBounds4D(coord : vec4<i32>, shape : vec4<i32>) -> bool {
|
|
return all(coord >= vec4<i32>(0)) && all(coord < shape);
|
|
}
|
|
|
|
fn getIndexFromCoords1D(coord : i32, shape : i32) -> i32 {
|
|
return coord;
|
|
}
|
|
fn getIndexFromCoords2D(coords : vec2<i32>, shape : vec2<i32>) -> i32 {
|
|
return dot(coords, vec2<i32>(shape.y, 1));
|
|
}
|
|
fn getIndexFromCoords3D(coords : vec3<i32>, shape : vec3<i32>) -> i32 {
|
|
return dot(coords, vec3<i32>(shape.y * shape.z, shape.z, 1));
|
|
}
|
|
fn getIndexFromCoords4D(coords : vec4<i32>, shape : vec4<i32>) -> i32 {
|
|
return dot(coords, vec4<i32>(
|
|
shape.y * shape.z * shape.w, shape.z * shape.w, shape.w, 1));
|
|
}
|
|
fn getIndexFromCoords5D(coords : vec5, shape : vec5) -> i32 {
|
|
let shapeStrides: vec5 = vec5(shape.y * shape.z * shape.w * shape.u, shape.z * shape.w * shape.u, shape.w * shape.u, shape.u, 1);
|
|
return coords.x*shapeStrides.x + coords.y*shapeStrides.y + coords.z*shapeStrides.z + coords.w*shapeStrides.w + coords.u*shapeStrides.u;
|
|
}
|
|
fn getIndexFromCoords6D(coords : vec6, shape : vec6) -> i32 {
|
|
let shapeStrides: vec6 = vec6(shape.y * shape.z * shape.w * shape.u * shape.v, shape.z * shape.w * shape.u * shape.v, shape.w * shape.u * shape.v, shape.u * shape.v, shape.v, 1);
|
|
return coords.x*shapeStrides.x + coords.y*shapeStrides.y + coords.z*shapeStrides.z + coords.w*shapeStrides.w + coords.u*shapeStrides.u + coords.v*shapeStrides.v;
|
|
}
|
|
|
|
fn idiv(a: i32, b: i32, sign: f32) -> i32 {
|
|
var res: i32 = a / b;
|
|
let mod: i32 = a % b;
|
|
if (sign < 0. && mod != 0) {
|
|
res = res - 1;
|
|
}
|
|
return res;
|
|
}
|
|
|
|
// NaN defination in IEEE 754-1985 is :
|
|
// - sign = either 0 or 1.
|
|
// - biased exponent = all 1 bits.
|
|
// - fraction = anything except all 0 bits (since all 0 bits represents infinity).
|
|
// https://en.wikipedia.org/wiki/IEEE_754-1985#Representation_of_non-numbers
|
|
fn isnan(val: f32) -> bool {
|
|
let floatToUint: u32 = bitcast<u32>(val);
|
|
return (floatToUint & 0x7fffffffu) > 0x7f800000u;
|
|
}
|
|
fn isnanVec4(val : vec4<f32>) -> vec4<bool> {
|
|
return vec4<bool>(isnan(val[0]), isnan(val[1]), isnan(val[2]), isnan(val[3]));
|
|
}
|
|
`;function U7(e){let t=e.length;if(t<=1)return"fn getCoordsFromIndex(index : i32) -> i32 { return index; }";let n=v.computeStrides(e),s=Tn(t),r=[];for(let o=0;o<t;o++)r.push(`d${o}`);if(n.length===1)return` fn getCoordsFromIndex(index : i32) -> vec2<i32> {
|
|
let d0 = index / uniforms.outShapeStrides; let d1 = index - d0 * uniforms.outShapeStrides;
|
|
return vec2<i32>(d0, d1);
|
|
}`;let a;return a="var index2 = index;"+n.map((o,i)=>{let l=`let ${r[i]} = index2 / uniforms.outShapeStrides.${Za(i)}`,u=i===n.length-1?`let ${r[i+1]} = index2 - ${r[i]} * uniforms.outShapeStrides.${Za(i)}`:`index2 = index2 - ${r[i]} * uniforms.outShapeStrides.${Za(i)}`;return`${l}; ${u};`}).join(""),`
|
|
fn getCoordsFromIndex(index : i32) -> ${s} {
|
|
${a}
|
|
return ${s}(${r.join(",")});
|
|
}
|
|
`}function M0e(e,t){let n=e.name,s=e.shape.length,r=Tn(s),a="get"+n.charAt(0).toUpperCase()+n.slice(1),o=["d0","d1","d2","d3","d4","d5"].slice(0,s),i=o.map(c=>`${c} : i32`).join(", ");if(s<1)return t?`
|
|
fn ${a}() -> vec4<f32> {
|
|
return vec4<f32>(${n}[0]);
|
|
}
|
|
`:`
|
|
fn ${a}() ->f32 {
|
|
return f32(${n}[0]);
|
|
}
|
|
`;let l=`uniforms.${n.charAt(0).toLowerCase()+n.slice(1)}Shape`,u=`${s}D`;return s===0&&(u="1D"),t?`
|
|
fn ${a}(${i}) -> vec4<f32> {
|
|
return vec4<f32>(${n}[getIndexFromCoords${u}(${r}(${o.join(",")}),
|
|
${l}) / 4]);
|
|
}
|
|
`:`
|
|
fn ${a}(${i}) -> f32 {
|
|
return f32(${n}[getIndexFromCoords${u}(${r}(${o.join(",")}),
|
|
${l})]);
|
|
}
|
|
`}function z0e(e,t,n,s){let r=e.name,a=r.charAt(0).toUpperCase()+r.slice(1),o="get"+a+"ByOutput",i=e.shape.length,l=t.length,u=Tn(l);if(v.arraysEqual(e.shape,t)&&s)return n?`
|
|
fn ${o}Index(globalIndex : i32) -> vec4<f32> {
|
|
return vec4<f32>(${r}[globalIndex]);
|
|
}
|
|
|
|
fn ${o}Coords(coords : ${u}) -> vec4<f32> {
|
|
return vec4<f32>(${r}[${l>1?"getOutputIndexFromCoords(coords)":"coords"} / 4]);
|
|
}
|
|
`:`
|
|
fn ${o}Index(globalIndex : i32) -> f32 {
|
|
return f32(${r}[globalIndex]);
|
|
}
|
|
|
|
fn ${o}Coords(coords : ${u}) -> f32 {
|
|
return f32(${r}[${l>1?"getOutputIndexFromCoords(coords)":"coords"}]);
|
|
}
|
|
`;let c=T.getBroadcastDims(e.shape,t),p=l-i,d="";if(i===0)return n?`
|
|
fn ${o}Index(globalIndex : i32) -> vec4<f32> {
|
|
return get${a}();
|
|
}
|
|
|
|
fn ${o}Coords(coords : ${u}) -> vec4<f32> {
|
|
return get${a}();
|
|
}
|
|
`:`
|
|
fn ${o}Index(globalIndex : i32) -> f32{
|
|
return get${a}();
|
|
}
|
|
|
|
fn ${o}Coords(coords : ${u}) -> f32{
|
|
return get${a}();
|
|
}
|
|
`;l<2&&c.length>=1?d="coords = 0;":d=c.map(g=>`coords.${Za(g+p)} = 0;`).join(`
|
|
`);let h="";if(l<2&&i>0)h="coords";else if(l>1){let g=Tn(i),y=e.shape.map((x,A)=>`coords.${Za(A+p)}`).join(", ");h=`${g}(${y})`}else h="coords";let f=`uniforms.${r.charAt(0).toLowerCase()+r.slice(1)}Shape`,m=`${i}D`;return n?`
|
|
fn ${o}Index(globalIndex : i32) -> vec4<f32> {
|
|
var coords = getCoordsFromIndex(globalIndex);
|
|
${d}
|
|
return ${r}[getIndexFromCoords${m}(${h}, ${f}) / 4];
|
|
}
|
|
|
|
fn ${o}Coords(coordsIn : ${u}) -> vec4<f32> {
|
|
var coords = coordsIn;
|
|
${d}
|
|
return ${r}[getIndexFromCoords${m}(${h}, ${f}) / 4];
|
|
}
|
|
`:`
|
|
fn ${o}Index(globalIndex : i32) -> f32 {
|
|
var coords = getCoordsFromIndex(globalIndex);
|
|
${d}
|
|
return f32(${r}[getIndexFromCoords${m}(${h}, ${f})]);
|
|
}
|
|
|
|
fn ${o}Coords(coordsIn : ${u}) -> f32 {
|
|
var coords = coordsIn;
|
|
${d}
|
|
return f32(${r}[getIndexFromCoords${m}(${h}, ${f})]);
|
|
}
|
|
`}function L0e(e,t,n,s){let r=M0e(e,n);return e.shape.length<=t.length&&(r+=z0e(e,t,n,s)),r}function B0e(e,t){let{x:n,y:s=[],z:r=[]}=t,a=e.length;if(n.length===a)return`fn getOutputCoords() -> ${Tn(a)}{
|
|
let globalIndex = getGlobalIndex();
|
|
return getCoordsFromIndex(globalIndex);
|
|
}
|
|
`;let o="",i=[n,s,r],l=0;for(let d=0;d<i.length;d++){let h=i[d];if(h.length!==0)if(l+=h.length,h.length===1)o+=`let d${h[0]} = i32(globalId[${d}]);`;else{let f=$0e(h,"uniforms.outShape");o+=`var index${d} = i32(globalId[${d}]);`;for(let m=0;m<f.length;m++)o+=`let d${h[m]} = index${d} / ${f[m]};`,m===f.length-1?o+=`let d${h[m+1]} = index${d} - d${h[m]} * ${f[m]};`:o+=`index${d} = index${d} - d${h[m]} * ${f[m]};`}}let u=[];for(let d=0;d<l;d++)u.push(`d${d}`);let c=Tn(l),p=`fn getOutputCoords() -> ${c} {
|
|
${o}
|
|
`;return u.length===0?p+=`return ${c}(0); }`:p+=`return ${c}(${u.join(",")}); }`,p}function W0e(e){let t="";switch(e){case 0:case 1:t+=`
|
|
fn getOutputIndexFromCoords(coords : i32) -> i32 {
|
|
return coords;
|
|
}
|
|
`;break;case 2:t+=`
|
|
fn getOutputIndexFromCoords(coords : vec2<i32>) -> i32 {
|
|
return dot(coords, vec2<i32>(uniforms.outShapeStrides, 1));
|
|
}
|
|
`;break;case 3:t+=`
|
|
fn getOutputIndexFromCoords(coords : vec3<i32>) -> i32 {
|
|
return dot(coords, vec3<i32>(uniforms.outShapeStrides.x, uniforms.outShapeStrides.y, 1));
|
|
}
|
|
`;break;case 4:t+=`
|
|
fn getOutputIndexFromCoords(coords : vec4<i32>) -> i32 {
|
|
return dot(coords, vec4<i32>(
|
|
uniforms.outShapeStrides.x, uniforms.outShapeStrides.y, uniforms.outShapeStrides.z, 1));
|
|
}
|
|
`;break;case 5:t+=`
|
|
fn getOutputIndexFromCoords(coords : vec5) -> i32 {
|
|
return coords.x * uniforms.outShapeStrides.x +
|
|
coords.y * uniforms.outShapeStrides.y +
|
|
coords.z * uniforms.outShapeStrides.z +
|
|
coords.w * uniforms.outShapeStrides.w +
|
|
coords.u;
|
|
}
|
|
`;break;case 6:t+=`
|
|
fn getOutputIndexFromCoords(coords : vec6) -> i32 {
|
|
return coords.x * uniforms.outShapeStrides.x +
|
|
coords.y * uniforms.outShapeStrides.y +
|
|
coords.z * uniforms.outShapeStrides.z +
|
|
coords.w * uniforms.outShapeStrides.w +
|
|
coords.u * uniforms.outShapeStrides.u +
|
|
coords.v;
|
|
}
|
|
`;break;default:v.assert(!1,()=>`Unsupported ${e}D shape`);break}return t}function qC(e){return e.dispatch[1]===1&&e.dispatch[2]===1}function Ap(e,t){return e==="float32"?t?"vec4<f32>":"f32":e==="int32"||e==="bool"?t?"vec4<i32>":"i32":e}function V0e(e,t,n){let s=e.length,r=Ap(t,n),a;if(n?a=`fn setOutputAtIndex(flatIndex : i32, value : vec4<f32>) {
|
|
result[flatIndex] = ${r}(value);
|
|
}
|
|
fn setOutputAtIndexI32(flatIndex : i32, value : vec4<i32>) {
|
|
result[flatIndex] = ${r}(value);
|
|
}`:a=`fn setOutputAtIndex(flatIndex : i32, value : f32) {
|
|
result[flatIndex] = ${r}(value);
|
|
}
|
|
fn setOutputAtIndexI32(flatIndex : i32, value : i32) {
|
|
result[flatIndex] = ${r}(value);
|
|
}`,s>=2){let o=["d0","d1","d2","d3","d4","d5"].slice(0,s),i=Tn(s);n?a+=`
|
|
fn setOutputAtCoords(${o.map(l=>`${l} : i32`).join(", ")}, value : vec4<f32>) {
|
|
let flatIndex = getOutputIndexFromCoords(${i}(${o.join(", ")}));
|
|
setOutputAtIndex(flatIndex / 4, value);
|
|
}
|
|
fn setOutputAtCoordsI32(${o.map(l=>`${l} : i32`).join(", ")}, value : vec4<i32>) {
|
|
let flatIndex = getOutputIndexFromCoords(${i}(${o.join(", ")}));
|
|
setOutputAtIndexI32(flatIndex / 4, value);
|
|
}
|
|
`:a+=`
|
|
fn setOutputAtCoords(${o.map(l=>`${l} : i32`).join(", ")}, value : f32) {
|
|
let flatIndex = getOutputIndexFromCoords(${i}(${o.join(", ")}));
|
|
setOutputAtIndex(flatIndex, value);
|
|
}
|
|
fn setOutputAtCoordsI32(${o.map(l=>`${l} : i32`).join(", ")}, value : i32) {
|
|
let flatIndex = getOutputIndexFromCoords(${i}(${o.join(", ")}));
|
|
setOutputAtIndexI32(flatIndex, value);
|
|
}
|
|
`}return a}var XC={};Ve(XC,{ArrayBufferToTypedArray:()=>YC,GPUBytesPerElement:()=>ZC,MatMulProgramType:()=>js,computeDispatch:()=>We,computeWorkGroupSizeForConv2d:()=>sb,computeWorkGroupSizeForMatMul:()=>KC,computeWorkPerThreadForConv2d:()=>rb,flatDispatchLayout:()=>at,isWebGPUSupported:()=>ab,tilesFitEvenlyIntoShape:()=>U0e});var qi=e=>{let t=1;for(let n=0;n<e.length;n++)t*=e[n];return t};function U0e(e,t){if(e.length!==t.length)throw new Error(`Cannot compute whether rank ${e.length} tiles fit evenly into rank ${t.length} shape - ranks must match.`);return t.every((n,s)=>n%e[s]===0)}function We(e,t,n=[1,1,1],s=[1,1,1]){let[r,a,o]=[Math.ceil(qi(e.x.map(i=>t[i]))/(n[0]*s[0])),e.y?Math.ceil(qi(e.y.map(i=>t[i]))/(n[1]*s[1])):1,e.z?Math.ceil(qi(e.z.map(i=>t[i]))/(n[2]*s[2])):1];return[r,a,o]}function sb(e,t,n=!1){if(n)return[8,8,1];let s=qi(e.x.map(a=>t[a])),r=qi(e.y.map(a=>t[a]));return s<=4?[4,16,1]:r<=4?[16,4,1]:[16,16,1]}function KC(e,t,n){return e===1?[32,1,1]:n===1?[1,32,1]:[8,8,1]}function rb(e,t,n=!1){if(n)return[4,4,1];let s=qi(e.x.map(a=>t[a])),r=qi(e.y.map(a=>t[a]));return s<=4?[1,2,1]:r<=4?[2,1,1]:[2,2,1]}function at(e){return{x:e.map((t,n)=>n)}}function ZC(e){if(e==="float32"||e==="int32"||e==="bool"||e==="string")return 4;if(e==="complex64")return 8;throw new Error(`Unknown dtype ${e}`)}function YC(e,t){if(t==="float32")return new Float32Array(e);if(t==="int32")return new Int32Array(e);if(t==="bool"||t==="string")return Uint8Array.from(new Int32Array(e));throw new Error(`Unknown dtype ${t}`)}function ab(){return(typeof window!="undefined"||typeof WorkerGlobalScope!="undefined")&&!!navigator.gpu}var js;(function(e){e[e.MatMulPackedVec4Program=0]="MatMulPackedVec4Program",e[e.MatMulReduceProgram=1]="MatMulReduceProgram",e[e.MatMulSplitKProgram=2]="MatMulSplitKProgram",e[e.MatMulSmallOutputSizeProgram=3]="MatMulSmallOutputSizeProgram",e[e.MatMulPackedProgram=4]="MatMulPackedProgram",e[e.MatMulMax=5]="MatMulMax"})(js||(js={}));function JC(e,t,n,s,r=!1,a=!1,o=!1,i=1){v.assert(n&&i===1||!n,()=>`transposeA ${n} is not compatible with component size ${i}`);let l=`
|
|
let batch = ${e?"0":"batchIn"};
|
|
let batchASize = uniforms.aShape[1] * uniforms.aShape[2];
|
|
${n?`value = A[(batch * batchASize + col * uniforms.aShape[2] + row) / ${i}];`:`value = A[(batch * batchASize + row * uniforms.aShape[2] + col) / ${i}];`}
|
|
|
|
`,u;return s===!1?u=`value = B[(batch * batchBSize + row * uniforms.bShape[2] + col) / ${i}];`:u=`value = B[(batch * batchBSize + col * uniforms.bShape[2] + row) / ${i}];`,`
|
|
fn mm_readA(batchIn: i32, row: i32, colIn: i32) -> ${nn(i)} {
|
|
var value = ${nn(i)}(0.0);
|
|
let col = colIn * ${i};
|
|
${r&&o?l:`
|
|
${n?"if(row < uniforms.dimAOuter && col < uniforms.dimInner)":"if(row < uniforms.aShape[1] && col < uniforms.aShape[2])"}
|
|
{
|
|
${l}
|
|
}
|
|
`}
|
|
return value;
|
|
}
|
|
|
|
fn mm_readB(batchIn: i32, row: i32, colIn: i32) -> ${nn(i)} {
|
|
let col = colIn * ${i};
|
|
let batch = ${t?"0":"batchIn"};
|
|
let batchBSize = uniforms.bShape[1] * uniforms.bShape[2];
|
|
var value = ${nn(i)}(0.0);
|
|
${u}
|
|
return value;
|
|
}
|
|
`}function N2(e,t,n,s,r,a,o=!1,i=!1,l=!1,u=1){return`
|
|
${JC(n,s,r,a,o,i,l,u)}
|
|
fn mm_write(batch: i32, row: i32, colIn: i32, valueIn: ${nn(u)}) {
|
|
let col = colIn * ${u};
|
|
${o&&i?"":"if (row < uniforms.dimAOuter && col < uniforms.dimBOuter)"}
|
|
{
|
|
var value = valueIn;
|
|
let coords = vec3<i32>(batch, row, col);
|
|
${dd(e,t)}
|
|
setOutputAtCoords(coords[0], coords[1], coords[2], value);
|
|
}
|
|
}
|
|
`}var G0e=e=>e?`
|
|
mm_Asub[inputRow][inputCol] = mm_readA(batch,
|
|
t * TileInner + inputRow,
|
|
globalRowStart + inputCol);
|
|
`:`
|
|
mm_Asub[inputRow][inputCol] = mm_readA(batch,
|
|
globalRowStart + inputRow,
|
|
t * TileInner + inputCol);
|
|
`,H0e=e=>e?"let ACached = mm_Asub[k][tileRow + innerRow];":"let ACached = mm_Asub[tileRow + innerRow][k];";function ob(e,t,n=!1,s=32){let r=e[1]*t[1],a=e[0]*t[0],o=n?r:s,i=n?s:r;v.assert(i%t[1]===0&&o%t[0]===0&&s%t[1]===0,()=>`tileAHight ${i} must be divisible by workGroupSize[1]${t[1]}, tileAWidth ${o} must be divisible by workGroupSize[0]${t[0]}, tileInner ${s} must be divisible by workGroupSize[1]${t[1]}`);let l=i/t[1],u=o/t[0],c=s/t[1];return`
|
|
var<workgroup> mm_Asub : array<array<f32, ${o}>, ${i}>;
|
|
var<workgroup> mm_Bsub : array<array<f32, ${a}>, ${s}>;
|
|
const RowPerThread = ${e[1]};
|
|
const ColPerThread = ${e[0]};
|
|
const TileInner = ${s};
|
|
|
|
@compute @workgroup_size(workGroupSizeX, workGroupSizeY, workGroupSizeZ)
|
|
fn main(@builtin(local_invocation_id) LocalId : vec3<u32>,
|
|
@builtin(global_invocation_id) GlobalId : vec3<u32>,
|
|
@builtin(num_workgroups) NumWorkgroups: vec3<u32>,
|
|
@builtin(workgroup_id) workgroupId: vec3<u32>) {
|
|
localId = LocalId;
|
|
globalId = GlobalId;
|
|
numWorkgroups = NumWorkgroups;
|
|
|
|
let tileRow = i32(localId.y) * RowPerThread;
|
|
let tileCol = i32(localId.x) * ColPerThread;
|
|
|
|
let globalRow = i32(globalId.y) * RowPerThread;
|
|
let globalCol = i32(globalId.x) * ColPerThread;
|
|
let batch = i32(globalId.z);
|
|
let globalRowStart = i32(workgroupId.y) * ${r};
|
|
|
|
let numTiles = (uniforms.dimInner - 1) / TileInner + 1;
|
|
|
|
var acc : array<array<f32, ColPerThread>, RowPerThread>;
|
|
|
|
// Without this initialization strange values show up in acc.
|
|
for (var innerRow = 0; innerRow < RowPerThread; innerRow = innerRow + 1) {
|
|
for (var innerCol = 0; innerCol < ColPerThread; innerCol = innerCol + 1) {
|
|
acc[innerRow][innerCol] = 0.0;
|
|
}
|
|
}
|
|
|
|
let tileRowA = i32(localId.y) * ${l};
|
|
let tileColA = i32(localId.x) * ${u};
|
|
let tileRowB = i32(localId.y) * ${c};
|
|
// Loop over shared dimension.
|
|
for (var t = 0; t < numTiles; t = t + 1) {
|
|
// Load one tile of A into local memory.
|
|
for (var innerRow = 0; innerRow < ${l}; innerRow = innerRow + 1) {
|
|
for (var innerCol = 0; innerCol < ${u}; innerCol = innerCol + 1) {
|
|
let inputRow = tileRowA + innerRow;
|
|
let inputCol = tileColA + innerCol;
|
|
${G0e(n)}
|
|
}
|
|
}
|
|
|
|
// Load one tile of B into local memory.
|
|
for (var innerRow = 0; innerRow < ${c}; innerRow = innerRow + 1) {
|
|
for (var innerCol = 0; innerCol < ColPerThread; innerCol = innerCol + 1) {
|
|
let inputRow = tileRowB + innerRow;
|
|
let inputCol = tileCol + innerCol;
|
|
mm_Bsub[inputRow][inputCol] = mm_readB(batch,
|
|
t * TileInner + inputRow,
|
|
globalCol + innerCol);
|
|
}
|
|
}
|
|
|
|
workgroupBarrier();
|
|
|
|
// Compute acc values for a single thread.
|
|
var BCached : array<f32, ColPerThread>;
|
|
for (var k = 0; k < TileInner; k = k + 1) {
|
|
for (var inner = 0; inner < ColPerThread; inner = inner + 1) {
|
|
BCached[inner] = mm_Bsub[k][tileCol + inner];
|
|
}
|
|
|
|
for (var innerRow = 0; innerRow < RowPerThread; innerRow = innerRow + 1) {
|
|
${H0e(n)}
|
|
for (var innerCol = 0; innerCol < ColPerThread; innerCol = innerCol + 1) {
|
|
acc[innerRow][innerCol] = acc[innerRow][innerCol] + ACached * BCached[innerCol];
|
|
}
|
|
}
|
|
}
|
|
|
|
workgroupBarrier();
|
|
}
|
|
|
|
for (var innerRow = 0; innerRow < RowPerThread; innerRow = innerRow + 1) {
|
|
for (var innerCol = 0; innerCol < ColPerThread; innerCol = innerCol + 1) {
|
|
mm_write(batch, globalRow + innerRow, globalCol + innerCol,
|
|
acc[innerRow][innerCol]);
|
|
}
|
|
}
|
|
}
|
|
`}var j0e=e=>e?`
|
|
mm_readA(batch, colA, globalRow),
|
|
mm_readA(batch, colA + 1, globalRow),
|
|
mm_readA(batch, colA + 2, globalRow),
|
|
mm_readA(batch, colA + 3, globalRow)
|
|
`:`
|
|
mm_readA(batch, globalRow, colA),
|
|
mm_readA(batch, globalRow, colA + 1),
|
|
mm_readA(batch, globalRow, colA + 2),
|
|
mm_readA(batch, globalRow, colA + 3)
|
|
`;function q0e(e,t=!1){return v.assert(e[1]===1&&e[2]===1,()=>`A linear work group size is required. But got ${e}.`),`
|
|
const TileSize = ${e[0]*4};
|
|
var<workgroup> mm_Asub : array<vec4<f32>, ${e[0]}>;
|
|
|
|
${pd()}
|
|
let tileCol = i32(localId.x);
|
|
let globalCol = i32(globalId.x);
|
|
let globalRow = i32(globalId.y);
|
|
|
|
let numTiles = (uniforms.dimInner - 1) / TileSize + 1;
|
|
let batch = i32(globalId.z);
|
|
// Without this initialization strange values show up in acc.
|
|
var acc = 0.0;
|
|
|
|
// Loop over shared dimension.
|
|
for (var t = 0; t < numTiles; t = t + 1) {
|
|
// Load one tile of A into local memory.
|
|
let colA = t * TileSize + tileCol * 4;
|
|
mm_Asub[tileCol] = vec4<f32>(${j0e(t)});
|
|
workgroupBarrier();
|
|
|
|
// Compute acc values for a single thread.
|
|
for (var k = 0; k < TileSize / 4; k = k + 1) {
|
|
let rowB = t * TileSize + k * 4;
|
|
let BCached = vec4<f32>(mm_readB(batch, rowB, globalCol),
|
|
mm_readB(batch, rowB + 1, globalCol),
|
|
mm_readB(batch, rowB + 2, globalCol),
|
|
mm_readB(batch, rowB + 3, globalCol));
|
|
|
|
let ACached = mm_Asub[k];
|
|
acc = acc + dot(ACached, BCached);
|
|
}
|
|
|
|
workgroupBarrier();
|
|
}
|
|
|
|
mm_write(batch, globalRow, globalCol, acc);
|
|
}
|
|
`}var X0e=class{constructor(e,t,n,s,r,a=!1,o=!1,i=null,l=null,u=null){this.variableNames=["A","B"],this.uniforms="dimAOuter : i32, dimBOuter : i32, dimInner : i32,",this.workGroupSize=[16,16,1],this.outputShape=t,this.dispatchLayout={x:[2],y:[1],z:[0]};let c=a?e[1]:e[2];this.workGroupSize=KC(t[1],c,t[2]),(t[1]===1||t[2]===1)&&(n=1),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[n,n,1]),v.arraysEqual(this.dispatch,[1,1,1])&&(n=1,this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[n,n,1]));let p=i!=null,d=u!=null;p&&this.variableNames.push("bias"),d&&this.variableNames.push("preluActivationWeights"),this.workPerThread=n,this.transposeA=a,this.transposeB=o,this.addBias=p,this.activation=l,this.hasPreluActivationWeights=d,this.batchAEqualOne=s,this.batchBEqualOne=r,[this.fitAOuter,this.fitBOuter,this.fitInner]=this.getShapeFit(t[1],t[2],c),this.shaderKey=`matMulPacked_${this.workPerThread}_${a}_${o}_${this.activation}_${this.fitAOuter}_${this.fitBOuter}_${this.fitInner}_${this.outputShape[1]>1}_${this.batchAEqualOne}_${this.batchBEqualOne}`}getShapeFit(e,t,n){let s=this.workGroupSize[1]*this.workPerThread,r=this.workGroupSize[0]*this.workPerThread;this.tileInner=32,this.outputShape[1]===1&&(this.tileInner=this.workGroupSize[0]*4);let a=e%s===0,o=t%r===0,i=n%this.tileInner===0;return[a,o,i]}getUserCode(){return`
|
|
${Ca(this.activation,this.hasPreluActivationWeights)}
|
|
${N2(this.addBias,this.activation,this.batchAEqualOne,this.batchBEqualOne,!1,this.transposeB,this.fitAOuter,this.fitBOuter,this.fitInner)}
|
|
${this.outputShape[1]>1?ob([this.workPerThread,this.workPerThread,1],this.workGroupSize,this.transposeA,this.tileInner):q0e(this.workGroupSize,this.transposeA)}
|
|
`}},K0e=(e,t)=>e?`
|
|
mm_Asub[inputRow][inputCol] = mm_readA(batch,
|
|
t * TileInner + inputRow,
|
|
globalRowStart / ${t} + inputCol);
|
|
`:`
|
|
mm_Asub[inputRow][inputCol] = mm_readA(batch,
|
|
globalRow + innerRow,
|
|
t * TileInner / ${t} + inputCol);
|
|
`,Z0e=(e,t)=>e?`
|
|
let ACached0 = mm_Asub[k * InnerElementSize][localRow];
|
|
let ACached1 = mm_Asub[k * InnerElementSize + 1][localRow];
|
|
let ACached2 = mm_Asub[k * InnerElementSize + 2][localRow];
|
|
${t===3?"":"let ACached3 = mm_Asub[k * InnerElementSize + 3][localRow];"}
|
|
for (var i = 0; i < RowPerThread; i = i + 1) {
|
|
acc[i] = BCached[0] * ACached0[i] + acc[i];
|
|
acc[i] = BCached[1] * ACached1[i] + acc[i];
|
|
acc[i] = BCached[2] * ACached2[i] + acc[i];
|
|
${t===3?"":"acc[i] = BCached[3] * ACached3[i] + acc[i];"}
|
|
}`:`
|
|
for (var i = 0; i < RowPerThread; i = i + 1) {
|
|
let ACached = mm_Asub[tileRow + i][k];
|
|
acc[i] = BCached[0] * ACached.x + acc[i];
|
|
acc[i] = BCached[1] * ACached.y + acc[i];
|
|
acc[i] = BCached[2] * ACached.z + acc[i];
|
|
${t===3?"":"acc[i] = BCached[3] * ACached.w + acc[i];"}
|
|
}`;function ib(e,t,n,s,r=4,a=!1){let o=a?t:s,i=a?s:t,l=a?e[1]:r;return v.assert((a&&t===n||s%4===0||s%3===0)&&e[0]===4&&(r===3||r===4),()=>`tileInner ${s} must be divisible by 4|3. ColPerThread ${e[0]} must be 4.
|
|
innerElementSize ${r} must be 3|4.`),`
|
|
var<workgroup> mm_Asub : array<array<vec${l}<f32>, ${o/l}>, ${i}>;
|
|
var<workgroup> mm_Bsub : array<array<vec4<f32>, ${n/e[0]}>, ${s}>;
|
|
|
|
const RowPerThread = ${e[1]};
|
|
const ColPerThread = ${e[0]};
|
|
const InnerElementSize = ${r};
|
|
const TileInner = ${s};
|
|
|
|
@compute @workgroup_size(workGroupSizeX, workGroupSizeY, workGroupSizeZ)
|
|
fn main(@builtin(local_invocation_id) LocalId : vec3<u32>,
|
|
@builtin(global_invocation_id) GlobalId : vec3<u32>,
|
|
@builtin(num_workgroups) NumWorkgroups: vec3<u32>,
|
|
@builtin(workgroup_id) workgroupId: vec3<u32>) {
|
|
localId = LocalId;
|
|
globalId = GlobalId;
|
|
numWorkgroups = NumWorkgroups;
|
|
|
|
let localRow = i32(localId.y);
|
|
let tileRow = ${t===1?"0":"localRow * RowPerThread"};
|
|
let tileCol = i32(localId.x);
|
|
|
|
let globalRow = ${t===1?"0":"i32(globalId.y) * RowPerThread"};
|
|
let globalCol = i32(globalId.x);
|
|
let batch = i32(globalId.z);
|
|
let globalRowStart = i32(workgroupId.y) * ${t};
|
|
|
|
let numTiles = (uniforms.dimInner - 1) / TileInner + 1;
|
|
|
|
var acc: array<vec4<f32>, RowPerThread>;
|
|
var BCached : array<vec4<f32>, 4>;
|
|
|
|
// Loop over shared dimension.
|
|
let RowPerThreadB = TileInner / i32(workGroupSizeY);
|
|
let tileRowB = localRow * RowPerThreadB;
|
|
for (var t = 0; t < numTiles; t = t + 1) {
|
|
// Load one tile of A into local memory.
|
|
for (var innerRow = 0; innerRow < RowPerThread; innerRow = innerRow + 1) {
|
|
let inputRow = tileRow + innerRow;
|
|
let inputCol = tileCol;
|
|
${K0e(a,l)}
|
|
}
|
|
|
|
// Load one tile of B into local memory.
|
|
for (var innerRow = 0; innerRow < RowPerThreadB; innerRow = innerRow + 1) {
|
|
let inputRow = tileRowB + innerRow;
|
|
let inputCol = tileCol;
|
|
mm_Bsub[inputRow][inputCol] = mm_readB(batch, t * TileInner + inputRow, globalCol);
|
|
}
|
|
|
|
workgroupBarrier();
|
|
|
|
// Compute acc values for a single thread.
|
|
for (var k = 0; k < TileInner / InnerElementSize; k = k + 1) {
|
|
BCached[0] = mm_Bsub[k * InnerElementSize][tileCol];
|
|
BCached[1] = mm_Bsub[k * InnerElementSize + 1][tileCol];
|
|
BCached[2] = mm_Bsub[k * InnerElementSize + 2][tileCol];
|
|
${r===3?"":"BCached[3] = mm_Bsub[k * InnerElementSize + 3][tileCol];"}
|
|
|
|
${Z0e(a,r)}
|
|
}
|
|
|
|
workgroupBarrier();
|
|
}
|
|
|
|
for (var innerRow = 0; innerRow < RowPerThread; innerRow = innerRow + 1) {
|
|
mm_write(batch, globalRow + innerRow, globalCol, acc[innerRow]);
|
|
}
|
|
}`}var Y0e=class{constructor(e,t,n,s,r=!1,a=null,o=null,i=null){this.variableNames=["A","B"],this.uniforms="dimAOuter : i32, dimBOuter : i32, dimInner : i32,",this.workGroupSize=[8,8,1],this.isVec4=!0,this.outputShape=t,this.dispatchLayout={x:[2],y:[1],z:[0]},t[1]===1&&!r?this.elementsPerThread=[4,1,1]:this.elementsPerThread=[4,4,1],this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,this.elementsPerThread);let l=a!=null,u=i!=null;l&&this.variableNames.push("bias"),u&&this.variableNames.push("preluActivationWeights"),this.tileAOuter=t[1]===1&&!r?1:this.workGroupSize[1]*this.elementsPerThread[1],this.tileBOuter=this.workGroupSize[0]*this.elementsPerThread[0],this.tileInner=this.tileBOuter,this.aShape=e,this.addBias=l,this.activation=o,this.hasPreluActivationWeights=u,this.batchAEqualOne=n,this.batchBEqualOne=s,this.transposeA=r;let c=r?e[1]:e[2];this.fitAOuter=t[1]%this.tileAOuter===0,this.fitBOuter=t[2]%this.tileBOuter===0,this.fitInner=c%this.tileInner===0,this.shaderKey=`matMulPackedVec4_${this.activation}_${this.fitAOuter}_${this.fitBOuter}_${this.fitInner}_${this.elementsPerThread}_${this.batchAEqualOne}_${this.batchBEqualOne}_${this.transposeA}`}getUserCode(){return`
|
|
${Ca(this.activation,this.hasPreluActivationWeights,!0)}
|
|
${N2(this.addBias,this.activation,this.batchAEqualOne,this.batchBEqualOne,!1,!1,this.fitAOuter,this.fitBOuter,this.fitInner,4)}
|
|
${ib(this.elementsPerThread,this.tileAOuter,this.tileBOuter,this.tileInner,4,this.transposeA)}
|
|
`}};function J0e(){return`
|
|
var<workgroup> sumValues : array<f32, workGroupSizeX>;
|
|
${pd()}
|
|
let coords = getOutputCoords();
|
|
let batch = coords[0];
|
|
let row = coords[1];
|
|
let col = coords[2];
|
|
var sum = 0.0;
|
|
let Length = uniforms.dimInner;
|
|
for (var k = i32(localId.x); k < Length; k = k + i32(workGroupSizeX)) {
|
|
let dataA = mm_readA(batch, row, k);
|
|
let dataB = mm_readB(batch, k, col);
|
|
sum = sum + dataA * dataB;
|
|
}
|
|
sumValues[localId.x] = sum;
|
|
workgroupBarrier();
|
|
|
|
for(var currentSize = workGroupSizeX / 2u; currentSize > 1u;
|
|
currentSize = currentSize / 2u) {
|
|
if (localId.x < currentSize)
|
|
{
|
|
sumValues[localId.x] = sumValues[localId.x] + sumValues[localId.x + currentSize];
|
|
}
|
|
workgroupBarrier();
|
|
}
|
|
|
|
if (localId.x == 0u) {
|
|
sum = sumValues[0] + sumValues[1];
|
|
mm_write(batch, row, col, sum);
|
|
}
|
|
}
|
|
`}var Q0e=class{constructor(e,t,n,s=!1,r=!1,a=null,o=null,i=null){this.variableNames=["A","B"],this.uniforms="dimAOuter : i32, dimBOuter : i32, dimInner : i32,",this.workGroupSize=[256,1,1],this.outputShape=e,this.dispatchLayout={x:[],y:[1,2],z:[0]},this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize);let l=a!=null,u=i!=null;l&&this.variableNames.push("bias"),u&&this.variableNames.push("preluActivationWeights"),this.transposeA=s,this.transposeB=r,this.addBias=l,this.activation=o,this.hasPreluActivationWeights=u,this.batchAEqualOne=t,this.batchBEqualOne=n,this.shaderKey=`matMulReduce_${this.activation}_${s}_${r}_${this.batchAEqualOne}_${this.batchBEqualOne}`}getUserCode(){return`
|
|
${Ca(this.activation,this.hasPreluActivationWeights)}
|
|
${N2(this.addBias,this.activation,this.batchAEqualOne,this.batchBEqualOne,this.transposeA,this.transposeB)}
|
|
${J0e()}
|
|
`}};function e2e(e){let t=e[1],n=e[0],s=t>n?t:n;return`
|
|
var<workgroup> mm_Asub : array<array<f32, ${s}>, ${t}>;
|
|
var<workgroup> mm_Bsub : array<array<f32, ${n}>, ${s}>;
|
|
|
|
// If the output size is small for matrix multiplication, avoid to use vec4
|
|
// and handle some elements per thread to optimally utilize the ALU.
|
|
// Read data from global memory to registers firstly, then store them into
|
|
// shared memory, so it is instruction-Level parallelism for arithmetic
|
|
// operations and others handle IO operations between barrier api, makes ALU
|
|
// and load/store units work simultaneously, could improves the performance.
|
|
${pd()}
|
|
let tileRow = i32(localId.y);
|
|
let tileCol = i32(localId.x);
|
|
let globalRow = i32(globalId.y);
|
|
let globalCol = i32(globalId.x);
|
|
let batch = i32(globalId.z);
|
|
|
|
// uniforms.dimInner should be greater than 0.
|
|
let numTiles = (uniforms.dimInner - 1) / ${s} + 1;
|
|
var acc = 0.0;
|
|
|
|
var globalColA = tileCol;
|
|
var globalRowB = 0;
|
|
var regA = mm_readA(batch, globalRow, globalColA);
|
|
var regB0 = mm_readB(batch, globalRowB + 2 * tileRow, globalCol);
|
|
var regB1 = mm_readB(batch, globalRowB + 2 * tileRow + 1, globalCol);
|
|
globalColA = globalColA + ${s};
|
|
globalRowB = globalRowB + ${s};
|
|
|
|
for (var t = 0; t < numTiles; t = t + 1) {
|
|
mm_Asub[tileRow][tileCol] = regA;
|
|
mm_Bsub[2 * tileRow][tileCol] = regB0;
|
|
mm_Bsub[2 * tileRow + 1][tileCol] = regB1;
|
|
|
|
workgroupBarrier();
|
|
|
|
regA = mm_readA(batch, globalRow, globalColA);
|
|
regB0 = mm_readB(batch, globalRowB + 2 * tileRow, globalCol);
|
|
regB1 = mm_readB(batch, globalRowB + 2 * tileRow + 1, globalCol);
|
|
globalColA = globalColA + ${s};
|
|
globalRowB = globalRowB + ${s};
|
|
|
|
for (var k = 0; k < ${s}; k = k + 1) {
|
|
acc = acc + mm_Asub[tileRow][k] * mm_Bsub[k][tileCol];
|
|
}
|
|
workgroupBarrier();
|
|
}
|
|
|
|
mm_write(batch, globalRow, globalCol, acc);
|
|
}
|
|
`}var t2e=class{constructor(e,t,n,s=!1,r=!1,a=null,o=null,i=null){this.variableNames=["A","B"],this.uniforms="dimAOuter : i32, dimBOuter : i32, dimInner : i32,",this.workGroupSize=[16,8,1],this.outputShape=n,this.dispatchLayout={x:[2],y:[1],z:[0]},this.dispatch=[Math.ceil(n[2]/this.workGroupSize[0]),Math.ceil(n[1]/this.workGroupSize[1]),n[0]];let l=a!=null;l&&this.variableNames.push("bias");let u=i!=null;u&&this.variableNames.push("preluActivationWeights"),this.transposeA=s,this.transposeB=r,this.addBias=l,this.activation=o,this.hasPreluActivationWeights=u,this.batchAEqualOne=e[0]===1,this.batchBEqualOne=t[0]===1,this.shaderKey=`matMulSmallOutputSize_${this.activation}_${s}_${r}_${this.batchAEqualOne}_${this.batchBEqualOne}`}getUserCode(){return`
|
|
${Ca(this.activation,this.hasPreluActivationWeights)}
|
|
${N2(this.addBias,this.activation,this.batchAEqualOne,this.batchBEqualOne,this.transposeA,this.transposeB)}
|
|
${e2e(this.workGroupSize)}
|
|
`}},n2e=class{constructor(e,t,n,s,r=!1,a=!1){this.variableNames=["A","B"],this.uniforms="dimAOuter : i32, dimBOuter : i32, dimInner : i32,",this.workGroupSize=[8,8,1],this.atomic=!0,this.tileInner=32,v.assert(e[0]===1,()=>"MatMulSplitKProgram only supports batch = 1."),this.outputShape=e,this.dispatchLayout={x:[2],y:[1],z:[0,3]},this.elementsPerThread=[4,4,this.tileInner],this.outputShape[1]<16&&(this.elementsPerThread[1]=1),this.outputShape[2]<16&&(this.elementsPerThread[0]=1),this.dispatch=We(this.dispatchLayout,[this.outputShape[0],this.outputShape[1],this.outputShape[2],t],this.workGroupSize,this.elementsPerThread),this.transposeA=r,this.transposeB=a,this.batchAEqualOne=n,this.batchBEqualOne=s,this.shaderKey=`matMulSplitK_${r}_${a}_${n}_${s}_${this.elementsPerThread}`}getUserCode(){let e=`
|
|
var oldValue = atomicLoad(&(result[flatIndex]));
|
|
var exchanged = false;
|
|
for (; !exchanged;) {
|
|
let newValueF32 = bitcast<f32>(oldValue) + value;
|
|
let newValue = bitcast<i32>(newValueF32);
|
|
let res = atomicCompareExchangeWeak(&(result[flatIndex]), oldValue, newValue);
|
|
oldValue = res.old_value;
|
|
exchanged = res.exchanged;
|
|
}
|
|
`;return`
|
|
${JC(this.batchAEqualOne,this.batchBEqualOne,this.transposeA,this.transposeB)}
|
|
fn mm_write(batch: i32, row : i32, col : i32, valueIn : f32) {
|
|
if (row < uniforms.dimAOuter && col < uniforms.dimBOuter) {
|
|
let coords = vec3<i32>(batch, row, col);
|
|
let flatIndex = getOutputIndexFromCoords(coords);
|
|
var value = valueIn;
|
|
// The problem is that we should initialize output to zero before using.
|
|
// Otherwise, the original value will be added to the result.
|
|
${e}
|
|
}
|
|
}
|
|
|
|
${this.makeMatMulSplitKSource()}
|
|
`}makeMatMulSplitKSource(){let e=this.workGroupSize[1]*this.elementsPerThread[1],t=this.workGroupSize[0]*this.elementsPerThread[0],n=this.elementsPerThread[1],s=this.elementsPerThread[0],r=this.tileInner/this.workGroupSize[0],a=this.tileInner/this.workGroupSize[1];return v.assert(this.tileInner%this.workGroupSize[0]===0&&this.tileInner%this.workGroupSize[1]===0,()=>`tileInner ${this.tileInner} must be divisible by workGroupSize[0]${this.workGroupSize[0]} and workGroupSize[1]${this.workGroupSize[1]}`),`
|
|
var<workgroup> mm_Asub : array<array<f32, ${this.tileInner}>, ${e}>;
|
|
var<workgroup> mm_Bsub : array<array<f32, ${t}>, ${this.tileInner}>;
|
|
${pd()}
|
|
let tileRow = i32(localId.y) * ${n};
|
|
let tileCol = i32(localId.x) * ${s};
|
|
|
|
let globalRow = i32(globalId.y) * ${n};
|
|
let globalCol = i32(globalId.x) * ${s};
|
|
let batch = 0;
|
|
let kStart = i32(globalId.z) * ${this.tileInner};
|
|
|
|
// Load one tile of A into local memory.
|
|
let tileColA = i32(localId.x) * ${r};
|
|
for (var innerRow = 0; innerRow < ${n}; innerRow = innerRow + 1) {
|
|
for (var innerCol = 0; innerCol < ${r}; innerCol = innerCol + 1) {
|
|
let inputRow = tileRow + innerRow;
|
|
let inputCol = tileColA + innerCol;
|
|
mm_Asub[inputRow][inputCol] = mm_readA(${this.batchAEqualOne?0:"batch"},
|
|
globalRow + innerRow,
|
|
kStart + inputCol);
|
|
}
|
|
}
|
|
// Load one tile of B into local memory.
|
|
let tileRowB = i32(localId.y) * ${a};
|
|
for (var innerRow = 0; innerRow < ${a}; innerRow = innerRow + 1) {
|
|
for (var innerCol = 0; innerCol < ${s}; innerCol = innerCol + 1) {
|
|
let inputRow = tileRowB + innerRow;
|
|
let inputCol = tileCol + innerCol;
|
|
mm_Bsub[inputRow][inputCol] = mm_readB(${this.batchBEqualOne?0:"batch"},
|
|
kStart + inputRow,
|
|
globalCol + innerCol);
|
|
}
|
|
}
|
|
|
|
workgroupBarrier();
|
|
|
|
var acc : array<array<f32, ${s}>, ${n}>;
|
|
// Loop over shared dimension. Compute acc values for a single thread.
|
|
for (var k = 0; k < ${this.tileInner}; k = k + 1) {
|
|
var BCached : array<f32, ${s}>;
|
|
for (var inner = 0; inner < ${s}; inner = inner + 1) {
|
|
BCached[inner] = mm_Bsub[k][tileCol + inner];
|
|
}
|
|
|
|
for (var innerRow = 0; innerRow < ${n}; innerRow = innerRow + 1) {
|
|
let ACached = mm_Asub[tileRow + innerRow][k];
|
|
for (var innerCol = 0; innerCol < ${s}; innerCol = innerCol + 1) {
|
|
acc[innerRow][innerCol] = acc[innerRow][innerCol] + ACached * BCached[innerCol];
|
|
}
|
|
}
|
|
}
|
|
|
|
for (var innerRow = 0; innerRow < ${n}; innerRow = innerRow + 1) {
|
|
for (var innerCol = 0; innerCol < ${s}; innerCol = innerCol + 1) {
|
|
mm_write(batch, globalRow + innerRow, globalCol + innerCol, acc[innerRow][innerCol]);
|
|
}
|
|
}
|
|
}
|
|
`}},s2e=class{constructor(e,t=null,n=null,s=null){this.uniforms="",this.variableNames=["x"],this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.addBias=t!=null,this.hasPreluActivationWeights=s!=null,this.activation=n,this.addBias&&this.variableNames.push("bias"),this.hasPreluActivationWeights&&this.variableNames.push("preluActivationWeights"),this.shaderKey=`biasActivation_${n}`}getUserCode(){return`
|
|
${Ca(this.activation,this.hasPreluActivationWeights)}
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
var value = getXByOutputIndex(index);
|
|
${dd(this.addBias,this.activation)}
|
|
setOutputAtIndex(index, value);
|
|
}
|
|
}
|
|
`}},r2e=class{constructor(e){this.variableNames=[],this.outputShape=[],this.uniforms="value : f32,",this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey="fill"}getUserCode(){return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
setOutputAtIndex(index, uniforms.value);
|
|
}
|
|
}
|
|
`}};function ou(e){let{backend:t,attrs:n}=e,{shape:s,value:r}=n,{dtype:a}=n;if(a=a||v.inferDtype(r),a==="string"){let o=v.getArrayFromDType(a,v.sizeFromShape(s));return o.fill(r),t.makeTensorInfo(s,a,o)}else{let o=new r2e(s),i=[{type:"float32",data:[r]}];return t.runWebGPUProgram(o,[],a,i)}}var a2e={kernelName:kc,backendName:"webgpu",kernelFunc:ou};function Ue(e){let{inputs:t,attrs:n}=e,{x:s}=t,{shape:r}=n,a=v.sizeFromShape(s.shape),o=v.inferFromImplicitShape(r,a),i=v.sizeFromShape(o);return v.assert(a===i,()=>`The new shape (${o}) has ${i} elements and the old shape (${s.shape}) has ${a} elements. The new shape and old shape must have the same number of elements.`),e.backend.incRef(s.dataId),{dataId:s.dataId,shape:o,dtype:s.dtype}}var o2e={kernelName:_l,backendName:"webgpu",kernelFunc:Ue};function lb({a:e,b:t,transposeA:n,transposeB:s,backend:r,bias:a=null,preluActivationWeights:o=null,leakyreluAlpha:i=0,activation:l=null}){let u=e.shape.length,c=t.shape.length,p=n?e.shape[u-2]:e.shape[u-1],d=s?t.shape[c-1]:t.shape[c-2],h=n?e.shape[u-1]:e.shape[u-2],f=s?t.shape[c-2]:t.shape[c-1],m=e.shape.slice(0,-2),g=t.shape.slice(0,-2),y=v.sizeFromShape(m),x=v.sizeFromShape(g),b=Xl.assertAndGetBroadcastShape(e.shape.slice(0,-2),t.shape.slice(0,-2)).concat([h,f]);v.assert(p===d,()=>`Error in matMul: inner shapes (${p}) and (${d}) of Tensors with shapes ${e.shape} and ${t.shape} and transposeA=${n} and transposeB=${s} must match.`);let w=n?[y,p,h]:[y,h,p],k=s?[x,f,d]:[x,d,f],S=Ue({inputs:{x:e},backend:r,attrs:{shape:w}}),E=Ue({inputs:{x:t},backend:r,attrs:{shape:k}}),R=[S,E],$=Math.max(y,x),_=y===1,P=x===1,C=(p%4===0&&!n||h%4===0&&n)&&f%4===0&&!s,F=[S,E],G=[{type:"int32",data:[h]},{type:"int32",data:[f]},{type:"int32",data:[p]}],K,z,Z=[$,h,f],J=q().get("WEBGPU_MATMUL_PROGRAM_TYPE");switch(J<0&&(h*f<=128?J=js.MatMulReduceProgram:$===1&&h<=128&&f<=48&&d>=2e3?J=js.MatMulSplitKProgram:h<=16&&(f<=512||d>=2*f)||f<=16&&(h<=512||p>=2*h)?J=js.MatMulSmallOutputSizeProgram:C?J=js.MatMulPackedVec4Program:J=js.MatMulPackedProgram),J){case js.MatMulPackedVec4Program:K=new Y0e(w,Z,_,P,n,a,l,o);break;case js.MatMulReduceProgram:K=new Q0e(Z,_,P,n,s,a,l,o);break;case js.MatMulSplitKProgram:{if(z=ou({backend:r,attrs:{shape:Z,value:0,dtype:e.dtype}}),K=new n2e(Z,d,_,P,n,s),a||l){z=r.runWebGPUProgram(K,F,e.dtype,G,z);let B=new s2e(z.shape,a,l,o),oe=null,Q=[z];a&&Q.push(a),o&&Q.push(o),l==="leakyrelu"&&(oe=[{type:"float32",data:[i]}],B.uniforms+=" alpha : f32,");let ae=r.runWebGPUProgram(B,Q,z.dtype,oe);R.push(z);let ie=Ue({inputs:{x:ae},backend:r,attrs:{shape:b}});R.push(ae);for(let me of R)r.disposeData(me.dataId);return ie}break}case js.MatMulSmallOutputSizeProgram:K=new t2e(w,k,Z,n,s,a,l,o);break;case js.MatMulPackedProgram:K=new X0e(w,Z,q().get("WEBGPU_MATMUL_WORK_PER_THREAD"),_,P,n,s,a,l,o);break;default:throw new Error(`Unsupported MatMulProgramType ${J}.`)}a&&F.push(a),o&&F.push(o),l==="leakyrelu"&&(G.push({type:"float32",data:[i]}),K.uniforms+=" alpha : f32,"),z=r.runWebGPUProgram(K,F,e.dtype,G,z);let te=Ue({inputs:{x:z},backend:r,attrs:{shape:b}});R.push(z);for(let B of R)r.disposeData(B.dataId);return te}function i2e(e){let{inputs:t,backend:n,attrs:s}=e,{a:r,b:a,bias:o,preluActivationWeights:i}=t,{transposeA:l,transposeB:u,activation:c,leakyreluAlpha:p}=s;return lb({a:r,b:a,transposeA:l,transposeB:u,backend:n,bias:o,preluActivationWeights:i,leakyreluAlpha:p,activation:c})}var l2e={kernelName:Ja,backendName:"webgpu",kernelFunc:i2e},G7=class{constructor(e,t,n){this.variableNames=["AReal","AImag","BReal","BImag"],this.workGroupSize=[128,1,1],this.size=!0,this.outputShape=T.assertAndGetBroadcastShape(t,n),this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey=`binaryOpComplex_${e}`,this.op=e}getUserCode(){return`
|
|
fn binaryOpComplex(
|
|
areal : f32, aimag : f32, breal : f32, bimag : f32) -> f32 {
|
|
${Lm(this.op,!1)}
|
|
}
|
|
|
|
${lt()}
|
|
if(index < uniforms.size) {
|
|
let areal = getARealByOutputIndex(index);
|
|
let aimag = getAImagByOutputIndex(index);
|
|
let breal = getBRealByOutputIndex(index);
|
|
let bimag = getBImagByOutputIndex(index);
|
|
setOutputAtIndex(index, binaryOpComplex(areal, aimag, breal, bimag));
|
|
}
|
|
}
|
|
`}},fy=class{constructor(e,t,n){this.size=!0,this.variableNames=["A","B"],this.outputShape=T.assertAndGetBroadcastShape(t,n),this.dispatchLayout=at(this.outputShape),this.op=e,this.useSharedMemoryWithA=t.length===1&&n.length>1&&t[0]<1024,this.useSharedMemoryWithB=n.length===1&&t.length>1&&n[0]<1024,this.useSharedMemoryWithA||this.useSharedMemoryWithB?(this.isVec4=!1,this.lastDimensionSize=this.useSharedMemoryWithB?n[0]:t[0],this.shaderKey=`binary_${this.type}_${e}_${this.lastDimensionSize}_${this.useSharedMemoryWithB}`,this.type="shared",this.workGroupSize=[256,1,1],this.lastDimensionSize<256?this.workPerThread=1:this.lastDimensionSize<512?this.workPerThread=2:this.workPerThread=4):(v.arraysEqual(t,n)&&v.sizeFromShape(t)%4===0?(this.isVec4=!0,this.type="vec4",this.workPerThread=4):(this.isVec4=!1,this.type="plain",this.workPerThread=1),this.shaderKey=`binary_${this.type}_${e}`,this.workGroupSize=[128,1,1]),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[this.workPerThread,1,1])}getUserCode(){let e;if(this.type==="shared"){let t=this.lastDimensionSize>1?`coords[${this.outputShape.length-1}]`:"0",n=this.useSharedMemoryWithB?`let a = getAByOutputCoords(coords);
|
|
let b = sharedBuf[${t}];`:`let a = sharedBuf[${t}];
|
|
let b = getBByOutputCoords(coords);`;e=`
|
|
fn binaryOperation(a : f32, b : f32) -> f32 {
|
|
${Lm(this.op,this.isVec4)}
|
|
}
|
|
var<workgroup> sharedBuf : array<f32, ${this.lastDimensionSize}>;
|
|
${lt()}
|
|
|
|
// Fill in the shared memory buffer. Here we need a loop to make sure
|
|
// that all data in A|B are uploaded when |sharedMemorySize| is larger
|
|
// than work group size.
|
|
for(var localIndex = i32(localId.x); localIndex < ${this.lastDimensionSize}; localIndex = localIndex + ${this.workGroupSize[0]}) {
|
|
sharedBuf[localIndex] = f32(${this.useSharedMemoryWithB?"B":"A"}[localIndex]);
|
|
}
|
|
workgroupBarrier();
|
|
|
|
for(var i = 0; i < ${this.workPerThread}; i = i + 1) {
|
|
let flatIndex = index * ${this.workPerThread} + i;
|
|
if(flatIndex < uniforms.size) {
|
|
let coords = getCoordsFromIndex(flatIndex);
|
|
|
|
${n}
|
|
setOutputAtIndex(flatIndex, binaryOperation(a, b));
|
|
}
|
|
}
|
|
}
|
|
`}else{let t=this.type==="vec4"?"vec4<f32>":"f32",n=Lm(this.op,this.isVec4);e=`
|
|
fn binaryOperation(a : ${t}, b : ${t}) -> ${t} {
|
|
${n}
|
|
}
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let a = getAByOutputIndex(index);
|
|
let b = getBByOutputIndex(index);
|
|
setOutputAtIndex(index, binaryOperation(a, b));
|
|
}
|
|
}
|
|
`}return e}};function Fs(e){let{inputs:t}=e,{x:n}=t;return e.backend.incRef(n.dataId),{dataId:n.dataId,shape:n.shape,dtype:n.dtype}}var u2e={kernelName:_o,backendName:"webgpu",kernelFunc:Fs};function hd(e){let{inputs:t,backend:n}=e,{real:s,imag:r}=t,a=n.makeTensorInfo(s.shape,"complex64"),o=n.tensorMap.get(a.dataId),i=Fs({inputs:{x:s},backend:n}),l=Fs({inputs:{x:r},backend:n});return o.complexTensorInfos={real:i,imag:l},a}var c2e={kernelName:Lp,backendName:"webgpu",kernelFunc:hd},Bh=class{constructor(e,t){this.variableNames=["A"],this.size=!0;let n=128;this.workGroupSize=[n,1,1],this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.op=t,this.shaderKey=`unary_${t}`}getUserCode(){return`
|
|
fn unaryOperation(a : f32) -> f32 {
|
|
${Oi(this.op,!1)}
|
|
}
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let a = getAByOutputIndex(index);
|
|
setOutputAtIndex(index, unaryOperation(a));
|
|
}
|
|
}
|
|
`}};function $n({opType:e,cpuKernelImpl:t,dtype:n}){return({inputs:s,backend:r})=>{let{x:a}=s,o=r,i=n||a.dtype;if(o.shouldExecuteOnCPU([a])&&t!=null){let u=o.tensorMap.get(a.dataId),c=t(u.values,i);return o.makeTensorInfo(a.shape,i,c)}let l=new Bh(a.shape,e);return o.runWebGPUProgram(l,[a],i)}}function os({opType:e,cpuKernelImpl:t,supportsComplex:n=!1,dtype:s}){return({inputs:r,backend:a})=>{let{a:o,b:i}=r,l=a;if(n&&o.dtype==="complex64"){let p=l.tensorMap.get(o.dataId),d=l.tensorMap.get(i.dataId),h,f;if(e!==Ye.MUL)[h,f]=[[p.complexTensorInfos.real,d.complexTensorInfos.real],[p.complexTensorInfos.imag,d.complexTensorInfos.imag]].map(g=>{let[y,x]=g,A={dataId:y.dataId,dtype:y.dtype,shape:o.shape},b={dataId:x.dataId,dtype:x.dtype,shape:i.shape},w=new fy(e,o.shape,i.shape);return l.runWebGPUProgram(w,[A,b],Mn(y.dtype,x.dtype))});else{let g=new G7(Ye.COMPLEX_MULTIPLY_REAL,o.shape,i.shape),y=new G7(Ye.COMPLEX_MULTIPLY_IMAG,o.shape,i.shape),x=[{dataId:p.complexTensorInfos.real.dataId,dtype:p.complexTensorInfos.real.dtype,shape:o.shape},{dataId:p.complexTensorInfos.imag.dataId,dtype:p.complexTensorInfos.imag.dtype,shape:o.shape},{dataId:d.complexTensorInfos.real.dataId,dtype:d.complexTensorInfos.real.dtype,shape:i.shape},{dataId:d.complexTensorInfos.imag.dataId,dtype:d.complexTensorInfos.imag.dtype,shape:i.shape}];h=l.runWebGPUProgram(g,x,"float32"),f=l.runWebGPUProgram(y,x,"float32")}let m=hd({inputs:{real:h,imag:f},backend:l});return l.disposeData(h.dataId),l.disposeData(f.dataId),m}let u=s||Mn(o.dtype,i.dtype);if((o.dtype==="string"||i.dtype==="string"||l.shouldExecuteOnCPU([o,i]))&&t!=null){let p=l.tensorMap.get(o.dataId).values,d=l.tensorMap.get(i.dataId).values,h=o.dtype==="string"?T.fromUint8ToStringArray(p):p,f=o.dtype==="string"?T.fromUint8ToStringArray(d):d,[m,g]=t(o.shape,i.shape,h,f,u);return l.makeTensorInfo(g,u,m)}let c=new fy(e,o.shape,i.shape);return l.runWebGPUProgram(c,[o,i],u)}}var{addImpl:d2e,ceilImpl:p2e,concatImpl:h2e,equalImpl:f2e,expImpl:m2e,expm1Impl:g2e,floorImpl:y2e,gatherNdImpl:A2e,gatherV2Impl:x2e,greaterEqualImpl:b2e,greaterImpl:v2e,lessEqualImpl:w2e,lessImpl:k2e,logImpl:I2e,maxImpl:S2e,maximumImpl:C2e,minimumImpl:T2e,multiplyImpl:N2e,negImpl:E2e,notEqualImpl:R2e,prodImpl:_2e,rangeImpl:D2e,rsqrtImpl:$2e,scatterImpl:P2e,simpleAbsImpl:F2e,sliceImpl:O2e,stridedSliceImpl:M2e,stringNGramsImpl:z2e,subImpl:L2e,tileImpl:B2e,topKImpl:W2e,transposeImpl:V2e,uniqueImpl:Uxe}=Cx,U2e=$n({opType:Fe.ABS,cpuKernelImpl:F2e}),G2e={kernelName:il,backendName:"webgpu",kernelFunc:U2e},H2e=os({opType:Ye.ADD,cpuKernelImpl:d2e,supportsComplex:!0}),j2e={kernelName:xa,backendName:"webgpu",kernelFunc:H2e},q2e=class{constructor(e){this.workPerThread=4,this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e[0],this.variableNames=e.map((t,n)=>`T${n}`),this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[this.workPerThread,1,1]),this.shaderKey="addN"}getUserCode(){let e=[];this.variableNames.forEach(s=>{e.push(`let v${s} = get${s}ByOutputCoords(coords);`)});let t=this.variableNames.map(s=>`v${s}`).join(" + ");return`
|
|
${lt()}
|
|
for (var i = 0; i < ${this.workPerThread}; i = i + 1) {
|
|
let flatIndex = index * ${this.workPerThread} + i;
|
|
if (flatIndex < uniforms.size) {
|
|
let coords = getCoordsFromIndex(flatIndex);
|
|
${e.join(`
|
|
`)}
|
|
setOutputAtIndex(flatIndex, ${t});
|
|
}
|
|
}
|
|
}
|
|
`}};function X2e(e){let{inputs:t,backend:n}=e,s=t;if(s.length===1)return Fs({inputs:{x:s[0]},backend:n});let r=s.map(i=>i.dtype).reduce((i,l)=>Mn(i,l)),a=s.map(i=>i.shape),o=new q2e(a);return n.runWebGPUProgram(o,s,r)}var K2e={kernelName:po,backendName:"webgpu",kernelFunc:X2e},QC=class{constructor(e,t,n){this.workGroupSize=[64,1,1],this.variableNames=["x"],this.uniforms="infinityValue : f32,",this.size=!0;let s=[t];T.assertAxesAreInnerMostDims("arg"+n.charAt(0).toUpperCase()+n.slice(1),s,e.length),this.op=n==="min"?"<":">";let[r]=T.computeOutAndReduceShapes(e,s);this.outputShape=r.length===0?[1]:r,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,[1,1,1]),this.inputShape=e,this.shaderKey=`argMinMax${this.op}`}getUserCode(){let e=`
|
|
var<workgroup> xBestIndices : array<i32, ${this.workGroupSize[0]}>;
|
|
var<workgroup> xBestValues : array<f32, ${this.workGroupSize[0]}>;
|
|
`,t=()=>this.inputShape.length===1?"uniforms.xShape":`uniforms.xShape.${Za(this.inputShape.length-1)}`,n=()=>{let r="";if(this.outputShape.length===1)this.inputShape.length!==1&&(r+="outputCoords,");else for(let a=0;a<this.outputShape.length;a++)r+=`outputCoords.${Za(a)},`;return r};return`
|
|
fn DIV_CEIL(a : u32, b : u32) -> u32 {
|
|
return ((a - 1u) / b + 1u);
|
|
}
|
|
|
|
${e}
|
|
|
|
${lt()}
|
|
let outputIndex = index / i32(workGroupSizeX);
|
|
let reduceLength = ${t()};
|
|
|
|
var bestIndex = i32(localId.x);
|
|
var bestValue = uniforms.infinityValue;
|
|
let outputCoords = getCoordsFromIndex(outputIndex);
|
|
for (var k = i32(localId.x); k < reduceLength && outputIndex < uniforms.size;
|
|
k = k + i32(workGroupSizeX)) {
|
|
let candidate = getX(${n()} k);
|
|
if (!isnan(candidate) && candidate ${this.op} bestValue) {
|
|
bestValue = candidate;
|
|
bestIndex = k;
|
|
}
|
|
}
|
|
xBestValues[localId.x] = bestValue;
|
|
xBestIndices[localId.x] = bestIndex;
|
|
workgroupBarrier();
|
|
|
|
var reduceSize = min(u32(reduceLength), workGroupSizeX);
|
|
for (var currentSize = reduceSize / 2u; reduceSize > 1u;
|
|
currentSize = reduceSize / 2u) {
|
|
let interval = DIV_CEIL(reduceSize, 2u);
|
|
if (localId.x < currentSize) {
|
|
let candidate = xBestValues[localId.x + interval];
|
|
if (candidate ${this.op} bestValue) {
|
|
bestValue = candidate;
|
|
xBestValues[localId.x] = bestValue;
|
|
xBestIndices[localId.x] = xBestIndices[localId.x + interval];
|
|
}
|
|
}
|
|
reduceSize = interval;
|
|
workgroupBarrier();
|
|
}
|
|
|
|
if (localId.x == 0u && outputIndex < uniforms.size) {
|
|
setOutputAtIndexI32(outputIndex, xBestIndices[localId.x]);
|
|
}
|
|
}
|
|
`}},Z2e=class{constructor(e,t){this.variableNames=["A"],this.workGroupSize=[16,16,1];let n=new Array(e.length);for(let s=0;s<n.length;s++)n[s]=e[t[s]];this.outputShape=n,this.dispatchLayout={x:[0],y:[1]},this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[1,1,1]),this.shaderKey="transposeShared"}getUserCode(){return`
|
|
const TILE_DIM = ${this.workGroupSize[0]};
|
|
var<workgroup> tile : array<array<f32, ${this.workGroupSize[0]+1}>, ${this.workGroupSize[0]}>;
|
|
${T2()}
|
|
fn main(@builtin(local_invocation_id) localId : vec3<u32>,
|
|
@builtin(workgroup_id) workgroupId : vec3<u32>) {
|
|
var x = i32(workgroupId.x) * TILE_DIM + i32(localId.x);
|
|
var y = i32(workgroupId.y) * TILE_DIM + i32(localId.y);
|
|
let width = uniforms.outShape[0];
|
|
let height = uniforms.outShape[1];
|
|
if (x < width && y < height) {
|
|
tile[localId.y][localId.x] = A[y * width + x];
|
|
}
|
|
workgroupBarrier();
|
|
|
|
x = i32(workgroupId.y) * TILE_DIM + i32(localId.x);
|
|
y = i32(workgroupId.x) * TILE_DIM + i32(localId.y);
|
|
if (x < height && y < width) {
|
|
setOutputAtIndex((y * height + x), tile[localId.x]
|
|
[localId.y]);
|
|
}
|
|
}
|
|
`}},Y2e=class{constructor(e,t){this.variableNames=["A"],this.workPerThread=4,this.workGroupSize=[64,1,1],this.size=!0;let n=new Array(e.length);for(let s=0;s<n.length;s++)n[s]=e[t[s]];this.outputShape=n,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[this.workPerThread,1,1]),this.newDim=t,this.shaderKey=`transpose_${t}`}getUserCode(){let e=Tn(this.outputShape.length),t=J2e(this.newDim);return`
|
|
${lt()}
|
|
|
|
for(var i = 0; i < ${this.workPerThread}; i = i + 1) {
|
|
let flatIndex = index * ${this.workPerThread} + i;
|
|
if(flatIndex < uniforms.size) {
|
|
let resRC = getCoordsFromIndex(flatIndex);
|
|
setOutputAtIndex(flatIndex, A[getIndexFromCoords${this.outputShape.length}D(
|
|
${e}(${t}), uniforms.aShape)]);
|
|
}
|
|
}
|
|
}
|
|
`}};function J2e(e){let t=e.length;if(t>6)throw Error(`Transpose for rank ${t} is not yet supported`);let n=new Array(t);for(let s=0;s<e.length;s++)n[e[s]]=`resRC.${Za(s)}`;return n.join()}function Aa(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{perm:a}=s,o=n,i=r.shape.length,l=new Array(i);for(let c=0;c<l.length;c++)l[c]=r.shape[a[c]];if(n.shouldExecuteOnCPU([r])){let p=o.tensorMap.get(r.dataId).values,d=V2e(p,r.shape,r.dtype,a,l);return n.makeTensorInfo(l,r.dtype,d)}if(r.shape.length===2&&v.arraysEqual(a,[1,0])){let c=new Z2e(r.shape,a);return o.runWebGPUProgram(c,[r],r.dtype)}let u=new Y2e(r.shape,a);return o.runWebGPUProgram(u,[r],r.dtype)}var Q2e={kernelName:Hr,backendName:"webgpu",kernelFunc:Aa};function e1e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a}=s,o=v.parseAxisParam(a,r.shape),i=T.getAxesPermutation(o,r.shape.length),l=r,u=[];i!=null&&(l=Aa({inputs:{x:r},backend:n,attrs:{perm:i}}),u.push(l),o=T.getInnerMostAxes(o.length,l.shape.length)),T.assertAxesAreInnerMostDims("argMax",[o[0]],l.shape.length);let c=new QC(l.shape,o[0],"max"),p=[{type:"float32",data:[Number.NEGATIVE_INFINITY]}],d=n.runWebGPUProgram(c,[l],"int32",p);return u.forEach(h=>n.disposeData(h.dataId)),d}var t1e={kernelName:ho,backendName:"webgpu",kernelFunc:e1e};function n1e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a}=s,o=v.parseAxisParam(a,r.shape),i=T.getAxesPermutation(o,r.shape.length),l=r,u=[];i!=null&&(l=Aa({inputs:{x:r},backend:n,attrs:{perm:i}}),u.push(l),o=T.getInnerMostAxes(o.length,l.shape.length)),T.assertAxesAreInnerMostDims("argMin",[o[0]],l.shape.length);let c=new QC(l.shape,o[0],"min"),p=[{type:"float32",data:[Number.POSITIVE_INFINITY]}],d=n.runWebGPUProgram(c,[l],"int32",p);return u.forEach(h=>n.disposeData(h.dataId)),d}var s1e={kernelName:gc,backendName:"webgpu",kernelFunc:n1e},eT=class{constructor(e,t){this.variableNames=["x"],this.uniforms="stride : vec2<i32>, pad : vec2<i32>, dilation : vec2<i32>, convDims : vec2<i32>, filterDims : vec2<i32>,",this.workGroupSize=[128,1,1],this.size=!0,this.outputShape=e.outShape,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey=`pool2D_${t}`,this.poolType=t}getUserCode(){let e="resultValue = max(value, resultValue);";this.poolType==="avg"&&(e="resultValue = resultValue + value; count = count + 1.0;");let t="resultValue";return this.poolType==="avg"&&(t="resultValue / count"),`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let batch = coords[0];
|
|
let xRCCorner = vec2<i32>(coords.yz) * uniforms.stride - uniforms.pad;
|
|
let xRCorner = xRCCorner.x;
|
|
let xCCorner = xRCCorner.y;
|
|
|
|
var resultValue = ${this.poolType==="avg"?"0.0":"-1.0 / pow(10.0, -20.0)"};
|
|
var count = 0.0;
|
|
|
|
for (var wR = 0; wR < uniforms.filterDims.x; wR = wR + uniforms.dilation.x) {
|
|
let xR = xRCorner + wR;
|
|
|
|
if (xR < 0 || xR >= uniforms.convDims.x) {
|
|
continue;
|
|
}
|
|
|
|
for (var wC = 0; wC < uniforms.filterDims.y; wC = wC + uniforms.dilation.y) {
|
|
let xC = xCCorner + wC;
|
|
if (xC < 0 || xC >= uniforms.convDims.y) {
|
|
continue;
|
|
}
|
|
|
|
let value = getX(batch, xR, xC, coords[3]);
|
|
${e}
|
|
}
|
|
}
|
|
|
|
setOutputAtIndex(index, ${t});
|
|
}
|
|
}
|
|
`}},tT=class{constructor(e){this.variableNames=["x"],this.uniforms="stride : vec2<i32>,",this.workGroupSize=[256,1,1],this.size=!0,this.outputShape=e.outShape,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey="poolWithFilterSizeEqualsOne"}getUserCode(){return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let batch = coords[0];
|
|
let d = coords[3];
|
|
|
|
let xRCCorner = coords.yz * uniforms.stride;
|
|
let xRCorner = xRCCorner.x;
|
|
let xCCorner = xRCCorner.y;
|
|
|
|
let value = getX(batch, xRCorner, xCCorner, d);
|
|
setOutputAtIndex(index, value);
|
|
}
|
|
}
|
|
`}};function r1e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=s,u=1,c=T.computePool2DInfo(r.shape,a,o,u,i,l);if(c.filterWidth===1&&c.filterHeight===1&&v.arraysEqual(c.inShape,c.outShape))return Fs({inputs:{x:r},backend:n});let p,d=[{type:"int32",data:[c.strideHeight,c.strideWidth]}];return c.filterHeight===1&&c.filterWidth===1?p=new tT(c):(p=new eT(c,"avg"),d.push({type:"int32",data:[c.padInfo.top,c.padInfo.left]},{type:"int32",data:[c.dilationHeight,c.dilationWidth]},{type:"int32",data:[c.inHeight,c.inWidth]},{type:"int32",data:[c.effectiveFilterHeight,c.effectiveFilterWidth]})),n.runWebGPUProgram(p,[r],r.dtype,d)}var a1e={kernelName:fo,backendName:"webgpu",kernelFunc:r1e};function o1e(e){let{inputs:t,backend:n,attrs:s}=e,{a:r,b:a}=t,{transposeA:o,transposeB:i}=s;return lb({a:r,b:a,transposeA:o,transposeB:i,backend:n})}var i1e={kernelName:mo,backendName:"webgpu",kernelFunc:o1e},l1e=class{constructor(e,t){this.variableNames=["source"],this.workPerThread=1,this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=t,this.rank=t.length,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[this.workPerThread,1,1]),this.start=e,this.uniforms=`start : ${Tn(e.length)}, `,this.shaderKey="slice"}getUserCode(){let e=Tn(this.rank),t=u1e(this.rank),n;return this.start.length===1?n=this.outputShape.map((r,a)=>"sourceLoc = uniforms.start + coords;"):n=this.outputShape.map((r,a)=>`sourceLoc.${my[a]} = uniforms.start[${a}] + coords.${my[a]};`),`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
var sourceLoc : ${e};
|
|
let coords = getCoordsFromIndex(index);
|
|
${n.join(`
|
|
`)}
|
|
setOutputAtIndex(index, getSource(${t}));
|
|
}
|
|
}
|
|
`}},my=["x","y","z","w","u","v"];function u1e(e){if(e===1)return"sourceLoc";if(e<=6)return my.slice(0,e).map(t=>`sourceLoc.${t}`).join(",");throw Error(`Slicing for rank ${e} is not yet supported`)}function fd(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{begin:a,size:o}=s,[i,l]=Ut.parseSliceParams(r,a,o);if(Ut.assertParamsValid(r,i,l),n.shouldExecuteOnCPU([r])||r.dtype==="string"){let p=n.tensorMap.get(r.dataId),d=O2e(p.values,i,l,r.shape,r.dtype);return n.makeTensorInfo(l,r.dtype,d)}if(v.sizeFromShape(l)===0)return n.makeTensorInfo(l,r.dtype,[]);let u=new l1e(i,l),c=[{type:"int32",data:i}];return n.runWebGPUProgram(u,[r],r.dtype,c)}var c1e={kernelName:Ol,backendName:"webgpu",kernelFunc:fd},d1e=e=>{let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockShape:a,crops:o}=s;v.assert(r.shape.length<=4,()=>"batchToSpaceND for rank > 4 with a WebGPU backend not implemented yet");let i=a.reduce((x,A)=>x*A),l=T.getReshaped(r.shape,a,i),u=T.getPermuted(l.length,a.length),c=T.getReshapedPermuted(r.shape,a,i),p=T.getSliceBeginCoords(o,a.length),d=T.getSliceSize(c,o,a.length),h=[],f=Ue({inputs:{x:r},backend:n,attrs:{shape:l}}),m=Aa({inputs:{x:f},backend:n,attrs:{perm:u}}),g=Ue({inputs:{x:m},backend:n,attrs:{shape:c}}),y=fd({inputs:{x:g},backend:n,attrs:{begin:p,size:d}});return h.push(f),h.push(m),h.push(g),h.forEach(x=>n.disposeData(x.dataId)),y},p1e={kernelName:ll,backendName:"webgpu",kernelFunc:d1e},nT=os({opType:Ye.NOT_EQUAL,dtype:"bool",cpuKernelImpl:R2e}),h1e={kernelName:Sl,backendName:"webgpu",kernelFunc:nT};function Wh(e){let{inputs:t,backend:n}=e,{input:s}=t,r=n.tensorMap.get(s.dataId);return Fs({inputs:{x:r.complexTensorInfos.real},backend:n})}var f1e={kernelName:qp,backendName:"webgpu",kernelFunc:Wh};function m1e(e,t){let n=new Bh(e.shape,Fe.TO_INT),s=t.runWebGPUProgram(n,[e],"int32");return{dataId:s.dataId,shape:s.shape,dtype:s.dtype}}function gy(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{dtype:a}=s;if(a==="complex64"){if(r.dtype==="complex64")return Fs({inputs:{x:r},backend:n});let o=Wt(r.shape),i=gy({inputs:{x:r},backend:n,attrs:{dtype:"float32"}}),l=hd({inputs:{real:i,imag:o},backend:n});return o.dispose(),n.disposeData(i.dataId),l}if(r.dtype==="complex64"){let o=Wh({inputs:{input:r},backend:n}),i=gy({inputs:{x:o},backend:n,attrs:{dtype:a}});return n.disposeData(o.dataId),i}if(!v.hasEncodingLoss(r.dtype,a)){let o=Fs({inputs:{x:r},backend:n});return{dataId:o.dataId,shape:o.shape,dtype:a}}if(a==="int32")return m1e(r,n);if(a==="bool"){let o=n.makeTensorInfo([],"bool",v.getTypedArrayFromDType("bool",1)),l=nT({inputs:{a:r,b:o},backend:n});return n.disposeData(o.dataId),l}throw new Error(`Error in Cast: failed to cast ${r.dtype} to ${a}`)}var g1e={kernelName:go,backendName:"webgpu",kernelFunc:gy},y1e=$n({opType:Fe.CEIL,cpuKernelImpl:p2e}),A1e={kernelName:yo,backendName:"webgpu",kernelFunc:y1e},x1e=class{constructor(e){this.variableNames=["A"],this.uniforms="minVal : f32, maxVal : f32,",this.workPerThread=4,this.workGroupSize=[64,1,1],this.isVec4=!0,this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[this.workPerThread,1,1]),this.shaderKey="clipVec4"}getUserCode(){return`
|
|
${lt()}
|
|
if(index < uniforms.size) {
|
|
let value = getAByOutputIndex(index);
|
|
var clampedValue : vec4<f32>;
|
|
for (var i = 0; i < 4; i = i + 1) {
|
|
if (isnan(value[i])) {
|
|
clampedValue[i] = value[i];
|
|
} else {
|
|
clampedValue[i] = clamp(value[i], uniforms.minVal, uniforms.maxVal);
|
|
}
|
|
}
|
|
|
|
setOutputAtIndex(index, clampedValue);
|
|
}
|
|
}
|
|
`}},b1e=class{constructor(e){this.variableNames=["A"],this.uniforms="minVal : f32, maxVal : f32,",this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey="clip"}getUserCode(){return`
|
|
${lt()}
|
|
if(index < uniforms.size) {
|
|
let value = getAByOutputIndex(index);
|
|
if (isnan(value)) {
|
|
setOutputAtIndex(index, value);
|
|
return;
|
|
}
|
|
setOutputAtIndex(index, clamp(value, uniforms.minVal, uniforms.maxVal));
|
|
}
|
|
}
|
|
`}};function v1e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{clipValueMin:a,clipValueMax:o}=s,i,l=[{type:"float32",data:[a]},{type:"float32",data:[o]}];return v.sizeFromShape(r.shape)%4===0?i=new x1e(r.shape):i=new b1e(r.shape),n.runWebGPUProgram(i,[r],r.dtype,l)}var w1e={kernelName:ba,backendName:"webgpu",kernelFunc:v1e},k1e=class{constructor(e){this.uniforms="",this.workPerThread=4,this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=T.computeOutShape(e,1),this.variableNames=e.map((t,n)=>`T${n}`),this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[this.workPerThread,1,1]),this.offsetLength=e.length-1;for(let t=0;t<this.offsetLength;t++)this.uniforms+=`offset${t} : i32,`;this.shaderKey="concat"}getUserCode(){let e=[];if(this.offsetLength>0){e.push("if (yC < uniforms.offset0){ setOutputAtCoords(coords.x, coords.y, getT0(yR, yC)); }");for(let r=1;r<this.offsetLength;r++)e.push(`else if (yC < uniforms.offset${[r]}){ setOutputAtCoords(coords.x, coords.y, getT${r}(yR, yC - uniforms.offset${r-1})); }`);let n=this.offsetLength,s=this.offsetLength-1;e.push(`else { setOutputAtCoords(coords.x, coords.y, getT${n}(yR, yC - uniforms.offset${s})); }`)}else e.push("setOutputAtCoords(coords.x, coords.y, getT0(yR, yC));");return`
|
|
${lt()}
|
|
for(var i = 0; i < ${this.workPerThread}; i = i + 1) {
|
|
let flatIndex = index * ${this.workPerThread} + i;
|
|
if(flatIndex < uniforms.size) {
|
|
let coords = getCoordsFromIndex(flatIndex);
|
|
let yR = coords.x;
|
|
let yC = coords.y;
|
|
|
|
${e.join(`
|
|
`)}
|
|
}
|
|
}
|
|
}
|
|
`}};function E2(e){let{inputs:t,backend:n}=e,{input:s}=t,r=n.tensorMap.get(s.dataId);return Fs({inputs:{x:r.complexTensorInfos.imag},backend:n})}var I1e={kernelName:Gp,backendName:"webgpu",kernelFunc:E2};function hp(e,t,n){let s=e[0].dtype;if(s==="complex64"){let f=e.map(A=>Wh({inputs:{input:A},backend:n})),m=e.map(A=>E2({inputs:{input:A},backend:n})),g=hp(f,t,n),y=hp(m,t,n),x=hd({inputs:{real:g,imag:y},backend:n});return f.forEach(A=>n.disposeData(A.dataId)),m.forEach(A=>n.disposeData(A.dataId)),n.disposeData(g.dataId),n.disposeData(y.dataId),x}let r=n.shouldExecuteOnCPU(e);if(s==="string"&&(r=!0),r){let f=e.map(w=>{let k=v.sizeFromShape(w.shape.slice(t));return Ue({inputs:{x:w},backend:n,attrs:{shape:[-1,k]}})}),m=f.map(w=>({vals:n.readSync(w.dataId),shape:w.shape})),g=T.computeOutShape(f.map(w=>w.shape),1),y=f[0].shape[0]===1,x=h2e(m,g,s,y),A=T.computeOutShape(e.map(w=>w.shape),t),b=n.makeTensorInfo(A,s,x);return f.forEach(w=>n.disposeData(w.dataId)),b}let a=n.device.limits.maxStorageBuffersPerShaderStage-1;if(e.length>a){let f=[];for(let g=0;g<e.length;g+=a){let y=e.slice(g,g+a);f.push(hp(y,t,n))}let m=hp(f,t,n);for(let g of f)n.disposeData(g.dataId);return m}let{tensors2D:o,outShape:i}=S1e(e,t,n),l=o.map(f=>f.shape),u=new k1e(l),c=[],p=new Array(l.length-1);if(p.length>0){p[0]=l[0][1],c.push({type:"int32",data:[p[0]]});for(let f=1;f<p.length;f++)p[f]=p[f-1]+l[f][1],c.push({type:"int32",data:[p[f]]})}let d=n.runWebGPUProgram(u,o,o[0].dtype,c);o.forEach(f=>n.disposeData(f.dataId));let h=Ue({inputs:{x:d},backend:n,attrs:{shape:i}});return n.disposeData(d.dataId),h}function S1e(e,t,n){let s=T.computeOutShape(e.map(a=>a.shape),t);return{tensors2D:e.map(a=>Ue({inputs:{x:a},backend:n,attrs:{shape:[v.sizeFromShape(a.shape.slice(0,t)),v.sizeFromShape(a.shape.slice(t))]}})),outShape:s}}function sT(e){let{inputs:t,backend:n,attrs:s}=e,{axis:r}=s,a=v.parseAxisParam(r,t[0].shape)[0],o=T.computeOutShape(t.map(u=>u.shape),a);if(v.sizeFromShape(o)===0)return n.makeTensorInfo(o,t[0].dtype,[]);let i=t.filter(u=>v.sizeFromShape(u.shape)>0);if(i.length===1)return Fs({inputs:{x:i[0]},backend:n});let l=i.map(u=>u.shape);return T.assertParamsConsistent(l,a),hp(i,a,n)}var C1e={kernelName:ul,backendName:"webgpu",kernelFunc:sT};function T1e(e,t,n,s,r=!1,a=null,o=!1,i=4,l=4,u=4){let c=R=>{switch(R){case 1:return"resData = x[xIndex];";case 3:return"resData = vec3<f32>(x[xIndex], x[xIndex + 1], x[xIndex + 2]);";case 4:return"resData = x[xIndex / 4];";default:throw new Error(`innerElementSize ${R} is not supported.`)}},p=R=>{switch(R){case 1:return"return W[row * uniforms.wShape[3] + colIn];";case 4:return"return W[row * uniforms.wShape[3] / 4 + colIn];";default:throw new Error(`innerElementSize ${R} is not supported.`)}},d=e?`
|
|
let coord = vec4<i32>(batch, xRow, xCol, xCh);
|
|
`:`
|
|
let coord = vec4<i32>(batch, xCh, xRow, xCol);
|
|
`,h=e?`
|
|
let coords = vec4<i32>(
|
|
batch,
|
|
row / outWidth,
|
|
row % outWidth,
|
|
col);
|
|
`:`
|
|
let coords = vec4<i32>(
|
|
batch,
|
|
row,
|
|
col / outWidth,
|
|
col % outWidth);
|
|
`,f=e?"uniforms.xShape[1]":"uniforms.xShape[2]",m=e?"uniforms.xShape[2]":"uniforms.xShape[3]",g=e?"row":"col",y=e?"col":"row",x=`
|
|
let inChannels = uniforms.wShape[2];
|
|
let outWidth = ${e?"uniforms.outShape[2]":"uniforms.outShape[3]"};
|
|
let outRow = ${g} / outWidth;
|
|
let outCol = ${g} % outWidth;
|
|
|
|
let WRow = ${y} / (uniforms.filterDims[1] * inChannels);
|
|
let WCol = ${y} / inChannels % uniforms.filterDims[1];
|
|
let xRow = outRow * uniforms.stride[0] + uniforms.dilation[0] * WRow - uniforms.pad[0];
|
|
let xCol = outCol * uniforms.stride[1] + uniforms.dilation[1] * WCol - uniforms.pad[1];
|
|
let xCh = ${y} % inChannels;
|
|
var resData = ${nn(i)}(0.0);
|
|
// The bounds checking is always needed since we use it to pad zero for
|
|
// the 'same' padding type.
|
|
if (xRow >= 0 && xRow < ${f} && xCol >= 0 && xCol < ${m}) {
|
|
${d}
|
|
let xIndex = getIndexFromCoords4D(coord, uniforms.xShape);
|
|
${c(i)}
|
|
}
|
|
return resData;`,A=e?t&&s?`
|
|
let col = colIn * ${i};
|
|
${x}`:`
|
|
let col = colIn * ${i};
|
|
if (row < uniforms.dimAOuter && col < uniforms.dimInner) {
|
|
${x}
|
|
}
|
|
return ${nn(i)}(0.0);`:s&&n?`
|
|
let col = colIn * ${i};
|
|
${x}`:`
|
|
let col = colIn * ${i};
|
|
if (row < uniforms.dimInner && col < uniforms.dimBOuter) {
|
|
${x}
|
|
}
|
|
return ${nn(i)}(0.0);`,b=`${p(l)}`,w=nn(u),k=nn(e?i:l),S=nn(e?l:i);return`
|
|
${Ca(a,o,u===4,4)}
|
|
fn mm_readA(batch: i32, row : i32, colIn : i32) -> ${k} {
|
|
${e?A:b}
|
|
}
|
|
|
|
fn mm_readB(batch: i32, row : i32, colIn : i32) -> ${S} {
|
|
${e?b:A}
|
|
}
|
|
|
|
fn mm_write(batch: i32, row : i32, colIn : i32, valueIn : ${w}) {
|
|
let col = colIn * ${u};
|
|
if (row < uniforms.dimAOuter && col < uniforms.dimBOuter)
|
|
{
|
|
var value = valueIn;
|
|
let outWidth = ${e?"uniforms.outShape[2]":"uniforms.outShape[3]"};
|
|
${h}
|
|
${dd(r,a)}
|
|
setOutputAtCoords(coords[0], coords[1], coords[2], coords[3], value);
|
|
}
|
|
}`}var N1e=class{constructor(e,t,n,s,r=!1,a=null,o=!1){this.variableNames=["x","W"],this.uniforms="filterDims : vec2<i32>, pad : vec2<i32>, stride : vec2<i32>, dilation : vec2<i32>, dimAOuter : i32, dimBOuter : i32, dimInner : i32,",this.outputShape=e.outShape,this.isChannelsLast=e.dataFormat==="channelsLast",this.isVec4=((e.inChannels%4===0||e.inChannels%3===0)&&this.isChannelsLast||e.outWidth%4===0&&!this.isChannelsLast)&&e.outChannels%4===0,this.dispatchLayout=this.isChannelsLast?{x:[3],y:[1,2],z:[0]}:{x:[2,3],y:[1],z:[0]},this.workGroupSize=sb(this.dispatchLayout,this.outputShape,this.isVec4),this.elementsPerThread=rb(this.dispatchLayout,this.outputShape,this.isVec4),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,this.elementsPerThread),this.isVec4?(this.isChannelsLast&&e.inChannels%4!==0?(this.innerElementSize=3,this.variableTypes=["f32","vec4<f32>"]):(this.innerElementSize=4,this.variableTypes=["vec4<f32>","vec4<f32>"]),r&&(this.variableNames.push("bias"),this.variableTypes.push("vec4<f32>")),o&&(this.variableNames.push("preluActivationWeights"),this.variableTypes.push("vec4<f32>"))):(this.innerElementSize=this.elementsPerThread[0],r&&this.variableNames.push("bias"),o&&this.variableNames.push("preluActivationWeights")),this.addBias=r,this.activation=a,this.hasPreluActivationWeights=o,this.tileAOuter=this.workGroupSize[1]*this.elementsPerThread[1],this.tileBOuter=this.workGroupSize[0]*this.elementsPerThread[0],this.tileInner=Math.max(this.workGroupSize[0]*this.innerElementSize,this.workGroupSize[1]),this.fitAOuter=t%this.tileAOuter===0,this.fitBOuter=n%this.tileBOuter===0,this.fitInner=s%this.tileInner===0,this.shaderKey=`conv2DMM_${this.elementsPerThread}_${this.activation}}_${this.fitAOuter}_${this.fitBOuter}_${this.fitInner}_${this.isVec4}_${this.innerElementSize}_${this.isChannelsLast}`}getUserCode(){let e=this.isVec4?ib(this.elementsPerThread,this.tileAOuter,this.tileBOuter,this.tileInner,this.innerElementSize,!this.isChannelsLast):ob(this.elementsPerThread,this.workGroupSize,!this.isChannelsLast,this.tileInner),t=this.isVec4?[this.isChannelsLast?this.innerElementSize:4,4,4]:[1,1,1];return`
|
|
${T1e(this.isChannelsLast,this.fitAOuter,this.fitBOuter,this.fitInner,this.addBias,this.activation,this.hasPreluActivationWeights,t[0],t[1],t[2])}
|
|
${e}
|
|
`}};function H7(e,t){let n=e.length;return n>=3?t?[...e.slice(0,-3),e[n-3]*e[n-2],e[n-1]]:[...e.slice(0,-3),e[n-3],e[n-2]*e[n-1]]:!t&&n===1&&e[0]>1?[e[0],1]:null}function E1e({x:e,filter:t,convInfo:n,backend:s,bias:r=null,preluActivationWeights:a=null,leakyreluAlpha:o=0,activation:i=null}){let l=n.dataFormat==="channelsLast",u=!l,c=!1,p=l&&n.filterHeight===n.inHeight&&n.filterWidth===n.inWidth&&n.padInfo.type==="VALID",d=[],h,f;if(p){let y=n.inHeight*n.inWidth*n.inChannels;h=Ue({inputs:{x:e},backend:s,attrs:{shape:[1,n.batchSize,y]}}),f=Ue({inputs:{x:t},backend:s,attrs:{shape:[1,y,n.outChannels]}})}else h=Ue({inputs:{x:e},backend:s,attrs:{shape:l?[n.batchSize,n.inHeight*n.inWidth,n.inChannels]:[n.batchSize,n.inChannels,n.inHeight*n.inWidth]}}),f=Ue({inputs:{x:t},backend:s,attrs:{shape:[1,n.inChannels,n.outChannels]}});if(d.push(h),d.push(f),a!=null){let y=H7(a.shape,l);y!=null&&(a=Ue({inputs:{x:a},backend:s,attrs:{shape:y}}),d.push(a))}if(r!=null){let y=H7(r.shape,l);y!=null&&(r=Ue({inputs:{x:r},backend:s,attrs:{shape:y}}),d.push(r))}let m=lb({a:l?h:f,b:l?f:h,transposeA:u,transposeB:c,backend:s,bias:r,activation:i,preluActivationWeights:a,leakyreluAlpha:o}),g=Ue({inputs:{x:m},backend:s,attrs:{shape:n.outShape}});d.push(m);for(let y of d)s.disposeData(y.dataId);return g}function rT({x:e,filter:t,convInfo:n,backend:s,bias:r=null,preluActivationWeights:a=null,leakyreluAlpha:o=0,activation:i=null}){let l=r!=null,u=a!=null,c=n.dataFormat==="channelsLast";if(c&&n.filterHeight===n.inHeight&&n.filterWidth===n.inWidth&&n.padInfo.type==="VALID"||n.filterHeight===1&&n.filterWidth===1&&n.dilationHeight===1&&n.dilationWidth===1&&n.strideHeight===1&&n.strideWidth===1&&(n.padInfo.type==="SAME"||n.padInfo.type==="VALID"))return E1e({x:e,filter:t,convInfo:n,backend:s,bias:r,activation:i,preluActivationWeights:a,leakyreluAlpha:o});let d=c?n.outHeight*n.outWidth:n.outChannels,h=c?n.outChannels:n.outHeight*n.outWidth,f=n.filterHeight*n.filterWidth*n.inChannels,m=[n.padInfo.top,n.padInfo.left],g=[{type:"int32",data:[n.filterHeight,n.filterWidth]},{type:"int32",data:[...m]},{type:"int32",data:[n.strideHeight,n.strideWidth]},{type:"int32",data:[n.dilationHeight,n.dilationWidth]},{type:"int32",data:[d]},{type:"int32",data:[h]},{type:"int32",data:[f]}],y=new N1e(n,d,h,f,l,i,u),x=[],A=[e,t];l&&(!c&&r.shape.length===1&&(r=Ue({inputs:{x:r},backend:s,attrs:{shape:[r.shape[0],1,1]}}),x.push(r)),A.push(r)),u&&(!c&&a.shape.length===1&&(a=Ue({inputs:{x:a},backend:s,attrs:{shape:[a.shape[0],1,1]}}),x.push(a)),A.push(a)),i==="leakyrelu"&&(g.push({type:"float32",data:[o]}),y.uniforms+=" alpha : f32,");let b=s.runWebGPUProgram(y,A,e.dtype,g);for(let w of x)s.disposeData(w.dataId);return b}function R1e(e){let{inputs:t,attrs:n,backend:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dataFormat:l,dilations:u,dimRoundingMode:c}=n,p=T.convertConv2DDataFormat(l),d=T.computeConv2DInfo(r.shape,a.shape,o,u,i,c,!1,p);return rT({x:r,filter:a,convInfo:d,backend:s})}var _1e={kernelName:Ao,backendName:"webgpu",kernelFunc:R1e};function D1e(e=4){let t=a=>{switch(a){case 1:return"return W[getIndexFromCoords4D(coord, uniforms.wShape)];";case 4:return`
|
|
let coord1 = vec4<i32>(coordX, coordY, col + 1, rowInner);
|
|
let coord2 = vec4<i32>(coordX, coordY, col + 2, rowInner);
|
|
let coord3 = vec4<i32>(coordX, coordY, col + 3, rowInner);
|
|
let v0 = W[getIndexFromCoords4D(coord, uniforms.wShape)];
|
|
let v1 = W[getIndexFromCoords4D(coord1, uniforms.wShape)];
|
|
let v2 = W[getIndexFromCoords4D(coord2, uniforms.wShape)];
|
|
let v3 = W[getIndexFromCoords4D(coord3, uniforms.wShape)];
|
|
return vec4<f32>(v0, v1, v2, v3);
|
|
`;default:throw new Error(`innerElementSize ${a} is not supported.`)}},s=`if (row < uniforms.dimAOuter && col < uniforms.dimInner) {
|
|
${`
|
|
let outRow = row / uniforms.outShape[2];
|
|
let outCol = row % uniforms.outShape[2];
|
|
|
|
let WRow = col / (uniforms.filterDims[1] * uniforms.outBackprop[3]);
|
|
let WCol = col / uniforms.outBackprop[3] % uniforms.filterDims[1];
|
|
let xR = f32(outRow - uniforms.pads[0] + WRow) / f32(uniforms.stride[0]);
|
|
let xC = f32(outCol - uniforms.pads[1] + WCol) / f32(uniforms.stride[1]);
|
|
if (xR < 0.0 || xR >= f32(uniforms.outBackprop[1]) || fract(xR) > 0.0) {
|
|
return ${nn(e)}(0.0);
|
|
}
|
|
if (xC < 0.0 || xC >= f32(uniforms.outBackprop[2]) || fract(xC) > 0.0) {
|
|
return ${nn(e)}(0.0);
|
|
}
|
|
let coord = vec4<i32>(
|
|
batch,
|
|
i32(xR),
|
|
i32(xC),
|
|
col % uniforms.outBackprop[3]);
|
|
return x[getIndexFromCoords4D(coord, uniforms.xShape)/${e}];`}
|
|
}
|
|
return ${nn(e)}(0.0);`;return`
|
|
fn mm_readA(batch: i32, row : i32, colIn : i32) -> ${nn(e)} {
|
|
let col = colIn * ${e};
|
|
${s}
|
|
}
|
|
|
|
fn mm_readB(batch: i32, row : i32, colIn : i32) -> ${nn(e)} {
|
|
let col = colIn * ${e};
|
|
let coordX = uniforms.filterDims.x - 1 -
|
|
row / (uniforms.filterDims[1] * uniforms.outBackprop[3]);
|
|
let coordY = uniforms.filterDims.y - 1 -
|
|
(row / uniforms.outBackprop[3]) % uniforms.filterDims[1];
|
|
if (row < uniforms.dimInner && col < uniforms.dimBOuter &&
|
|
coordX >= 0 && coordY >= 0) {
|
|
let rowInner = row % uniforms.outBackprop[3];
|
|
let coord = vec4<i32>(coordX, coordY, col, rowInner);
|
|
${t(e)}
|
|
}
|
|
return ${nn(e)}(0.0);
|
|
}
|
|
|
|
fn mm_write(batch: i32, row : i32, colIn : i32, valueInput : ${nn(e)}) {
|
|
let col = colIn * ${e};
|
|
if (row < uniforms.dimAOuter && (col + ${e-1}) < uniforms.dimBOuter) {
|
|
var value = valueInput;
|
|
let outCoord = vec4<i32>(
|
|
batch,
|
|
row / uniforms.outShape[2],
|
|
row % uniforms.outShape[2],
|
|
col);
|
|
result[getIndexFromCoords4D(outCoord, uniforms.outShape)/${e}] = value;
|
|
}
|
|
}`}var $1e=class{constructor(e){this.variableNames=["x","W"],this.uniforms="filterDims : vec2<i32>, pads : vec2<i32>, stride : vec2<i32>, outBackprop : vec4<i32>, dimAOuter : i32, dimBOuter : i32, dimInner : i32,",this.outputShape=e.inShape,v.assert(e.dataFormat==="channelsLast",()=>"TODO: NCHW is unimplemented"),this.isVec4=e.inChannels%4===0&&e.outChannels%4===0,this.dispatchLayout={x:[3],y:[1,2],z:[0]},this.workGroupSize=sb(this.dispatchLayout,this.outputShape,this.isVec4),this.elementsPerThread=rb(this.dispatchLayout,this.outputShape,this.isVec4),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,this.elementsPerThread),this.isVec4?(this.innerElementSize=4,this.variableTypes=["vec4<f32>","f32"]):this.innerElementSize=this.elementsPerThread[0],this.tileAOuter=this.workGroupSize[1]*this.elementsPerThread[1],this.tileBOuter=this.workGroupSize[0]*this.elementsPerThread[0],this.tileInner=Math.max(this.workGroupSize[0]*this.innerElementSize,this.workGroupSize[1]),this.shaderKey=`conv2DDerInputMM_${this.isVec4}_${this.elementsPerThread}_${this.innerElementSize}`}getUserCode(){let e=this.isVec4?ib(this.elementsPerThread,this.tileAOuter,this.tileBOuter,this.tileInner,this.innerElementSize):ob(this.elementsPerThread,this.workGroupSize);return`
|
|
${D1e(this.isVec4?4:1)}
|
|
${e}
|
|
`}},P1e=class{constructor(e){this.variableNames=["dy","W"],this.uniforms="filterDims : vec2<i32>, pads : vec2<i32>, stride : vec2<i32>, outBackprop : vec4<i32>,",this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e.inShape,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.isChannelsLast=e.dataFormat==="channelsLast",this.shaderKey=`conv2DDerInput_${this.isChannelsLast}`}getUserCode(){let e=this.isChannelsLast?1:2,t=this.isChannelsLast?2:3,n=this.isChannelsLast?3:1;return`
|
|
${lt()} {
|
|
if(index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let batch = coords[0];
|
|
let d1 = coords[${n}];
|
|
|
|
let dyCorner = vec2<i32>(coords[${e}]), coords[${t}]) - uniforms.pads;
|
|
let dyRCorner = dyCorner.x;
|
|
let dyCCorner = dyCorner.y;
|
|
|
|
// Convolve dy(?, ?, d2) with w(:, :, d1, d2) to compute dx(xR, xC, d1).
|
|
// ? = to be determined. : = across all values in that axis.
|
|
var dotProd = 0.0;
|
|
for (var wR = 0; wR < uniforms.filterDims.x; wR = wR + 1) {
|
|
let dyR = (f32(dyRCorner) + f32(wR)) / f32(uniforms.stride.x);
|
|
let wRPerm = uniforms.filterDims.x - 1 - wR;
|
|
if (dyR < 0.0 || dyR >= f32(uniforms.outBackprop[1]) || fract(dyR) > 0.0 ||
|
|
wRPerm < 0) {
|
|
continue;
|
|
}
|
|
let idyR = dyR;
|
|
|
|
for (var wC = 0; wC < uniforms.filterDims.y; wC = wC + 1) {
|
|
let dyC = (f32(dyCCorner) + f32(wC)) / f32(uniforms.stride.y);
|
|
let wCPerm = uniforms.filterDims.y - 1 - wC;
|
|
if (dyC < 0.0 || dyC >= f32(uniforms.outBackprop[2]) ||
|
|
fract(dyC) > 0.0 || wCPerm < 0) {
|
|
continue;
|
|
}
|
|
let idyC = dyC;
|
|
|
|
for (var d2 = 0; d2 < uniforms.outBackprop[3]; d2 = d2 + 1) {
|
|
if (${this.isChannelsLast}) {
|
|
let xValue = getDy(batch, idyR, idyC, d2);
|
|
let wValue = getW(wRPerm, wCPerm, d1, d2);
|
|
dotProd = dotProd + xValue * wValue;
|
|
} else {
|
|
let xValue = getDy(batch, d2, idyR, idyC);
|
|
let wValue = getW(wRPerm, wCPerm, d1, d2);
|
|
dotProd = dotProd + xValue * wValue;
|
|
}
|
|
|
|
}
|
|
}
|
|
}
|
|
setOutputAtIndex(index, dotProd);
|
|
}
|
|
}
|
|
`}};function F1e(e){let{inputs:t,backend:n,attrs:s}=e,{dy:r,filter:a}=t,{inputShape:o,strides:i,pad:l,dataFormat:u,dimRoundingMode:c}=s,p=T.convertConv2DDataFormat(u),d=T.computeConv2DInfo(o,a.shape,i,1,l,c,!1,p),h=[{type:"int32",data:[d.filterHeight,d.filterWidth]},{type:"int32",data:[d.filterHeight-1-d.padInfo.top,d.filterWidth-1-d.padInfo.left]},{type:"int32",data:[d.strideHeight,d.strideWidth]},{type:"int32",data:[d.batchSize,d.outHeight,d.outWidth,d.outChannels]}],f;if(q().getBool("WEBGPU_USE_NAIVE_CONV2D_TRANSPOSE"))f=new P1e(d);else{f=new $1e(d);let m=d.inShape[1]*d.inShape[2],g=d.inShape[3],y=d.filterHeight*d.filterWidth*d.outChannels;h.push({type:"uint32",data:[m]},{type:"uint32",data:[g]},{type:"uint32",data:[y]})}return n.runWebGPUProgram(f,[r,a],"float32",h)}var O1e={kernelName:xo,backendName:"webgpu",kernelFunc:F1e},M1e=$n({opType:Fe.COS}),z1e={kernelName:bo,backendName:"webgpu",kernelFunc:M1e},L1e=$n({opType:Fe.COSH}),B1e={kernelName:vo,backendName:"webgpu",kernelFunc:L1e},W1e=class{constructor(e,t,n,s){this.variableNames=["Image","Boxes","BoxInd"],this.uniforms="extrapolationValue : f32,",this.workGroupSize=[64,1,1],this.size=!0;let[r]=t;this.outputShape=[r,n[0],n[1],e],this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.methodId=s==="bilinear"?1:0,this.cropHeightBiggerThan1=this.outputShape[1]>1,this.cropWidthBiggerThan1=this.outputShape[2]>1,this.shaderKey=`cropAndResize_${this.methodId}_${this.cropHeightBiggerThan1}_${this.cropWidthBiggerThan1}`}getUserCode(){let[e,t]=["f32(uniforms.imageShape[1] - 1)","f32(uniforms.imageShape[2] - 1)"],[n,s,r]=this.cropHeightBiggerThan1?[`(${e} / f32(uniforms.outShape[1] - 1))`,"(y2-y1) * height_ratio",`y1*${e} + f32(y)*(height_scale)`]:["0.0","0.0",`0.5 * (y1+y2) * ${e}`],[a,o,i]=this.cropWidthBiggerThan1?[`(${t} / f32(uniforms.outShape[2] - 1))`,"(x2-x1) * width_ratio",`x1*${t} + f32(x)*(width_scale)`]:["0.0","0.0",`0.5 * (x1+x2) * ${t}`];return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let height_ratio = f32(${n});
|
|
let width_ratio = f32(${a});
|
|
let b = coords[0];
|
|
let y = coords[1];
|
|
let x = coords[2];
|
|
let d = coords[3];
|
|
// get box vals
|
|
let y1 = getBoxes(b, 0);
|
|
let x1 = getBoxes(b, 1);
|
|
let y2 = getBoxes(b, 2);
|
|
let x2 = getBoxes(b, 3);
|
|
// get image in batch index
|
|
let bInd = i32(round(getBoxInd(b)));
|
|
if(bInd < 0 || bInd >= uniforms.outShape[0]) {
|
|
return;
|
|
}
|
|
let height_scale = ${s};
|
|
let width_scale = ${o};
|
|
let in_y = ${r};
|
|
if( in_y < 0.0 || in_y > ${e} ) {
|
|
setOutputAtIndex(index, uniforms.extrapolationValue);
|
|
return;
|
|
}
|
|
let in_x = ${i};
|
|
if( in_x < 0.0 || in_x > ${t} ) {
|
|
setOutputAtIndex(index, uniforms.extrapolationValue);
|
|
return;
|
|
}
|
|
let sourceFracIndexCR = vec2<f32>(in_x,in_y);
|
|
if(${this.methodId} == 1) {
|
|
// Compute the four integer indices.
|
|
let sourceFloorCR = vec2<i32>(sourceFracIndexCR);
|
|
let sourceCeilCR = vec2<i32>(ceil(sourceFracIndexCR));
|
|
let topLeft = getImage(bInd, sourceFloorCR.y, sourceFloorCR.x, d);
|
|
let bottomLeft = getImage(bInd, sourceCeilCR.y, sourceFloorCR.x, d);
|
|
let topRight = getImage(bInd, sourceFloorCR.y, sourceCeilCR.x, d);
|
|
let bottomRight = getImage(bInd, sourceCeilCR.y, sourceCeilCR.x, d);
|
|
let fracCR = sourceFracIndexCR - vec2<f32>(sourceFloorCR);
|
|
let top = topLeft + (topRight - topLeft) * fracCR.x;
|
|
let bottom = bottomLeft + (bottomRight - bottomLeft) * fracCR.x;
|
|
let newValue = top + (bottom - top) * fracCR.y;
|
|
setOutputAtIndex(index, newValue);
|
|
} else {
|
|
// Compute the coordinators of nearest neighbor point.
|
|
let sourceNearestCR = vec2<i32>(floor(
|
|
sourceFracIndexCR + vec2<f32>(0.5,0.5)));
|
|
let newValue = getImage(
|
|
bInd, sourceNearestCR.y, sourceNearestCR.x, d);
|
|
setOutputAtIndex(index, newValue);
|
|
}
|
|
}
|
|
}
|
|
`}},V1e=e=>{let{inputs:t,backend:n,attrs:s}=e,{image:r,boxes:a,boxInd:o}=t,{cropSize:i,method:l,extrapolationValue:u}=s,c=new W1e(r.shape[3],a.shape,i,l),p=[{type:"float32",data:[u]}];return n.runWebGPUProgram(c,[r,a,o],"float32",p)},U1e={kernelName:dl,backendName:"webgpu",kernelFunc:V1e},Op;(function(e){e.Prod="*",e.Sum="+"})(Op||(Op={}));var j7=class{constructor(e,t,n,s){this.variableNames=["x"],this.uniforms="index : f32,",this.size=!0;let r=128;this.workGroupSize=[r,1,1],this.outputShape=t,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.exclusive=n,this.reverse=s,this.op=e,this.shaderKey=`cum_${this.op}_${this.exclusive}_${this.reverse}`}getUserCode(){let e=this.outputShape.length,t=this.op===Op.Prod?"1.0":"0.0",n=this.exclusive?t:`getX(${q7(e,"coords",this.op)})`,s=this.outputShape[this.outputShape.length-1],r="",a="";return this.exclusive?(r=this.reverse?`end != ${s-1}`:"end != 0",a=this.reverse?"end + 1":"end - 1"):(r=this.reverse?`end + pow2 < ${s}`:"end >= pow2",a=this.reverse?"end + pow2":"end - pow2"),`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
var coords = getCoordsFromIndex(index);
|
|
|
|
let end = ${X7(e,"coords",this.op)};
|
|
var val = ${n};
|
|
let pow2 = i32(pow(2.0, uniforms.index));
|
|
if (${r}) {
|
|
let idx = ${a};
|
|
${X7(e,"coords",this.op)} = idx;
|
|
val ${this.op}= getX(${q7(e,"coords",this.op)});
|
|
}
|
|
setOutputAtIndex(index, val);
|
|
}
|
|
}
|
|
`}};function q7(e,t,n){if(e===1)return`${t}`;if(e===2)return`${t}.x, ${t}.y`;if(e===3)return`${t}.x, ${t}.y, ${t}.z`;if(e===4)return`${t}.x, ${t}.y, ${t}.z, ${t}.w`;throw Error(`Cumulative ${n} for rank ${e} is not yet supported`)}function X7(e,t,n){if(e===1)return`${t}`;if(e===2)return`${t}.y`;if(e===3)return`${t}.z`;if(e===4)return`${t}.w`;throw Error(`Cumulative ${n} for rank ${e} is not yet supported`)}function aT(e,t,n,s,r,a){let o=t.shape.length,i=T.getAxesPermutation([s],o),l=t;i!=null&&(l=Aa({inputs:{x:t},backend:n,attrs:{perm:i}}));let u=T.getInnerMostAxes(1,o)[0];if(u!==o-1)throw new Error(`WebGPU cumprod shader expects an inner-most axis=${t.shape.length-1} but got axis=${s}`);let c=l.shape[u],p=Fs({inputs:{x:l},backend:n});for(let d=0;d<=Math.ceil(Math.log2(c))-1;d++){let h=new j7(e,l.shape,!1,a),f=p,m=[{type:"float32",data:[d]}];p=n.runWebGPUProgram(h,[p],p.dtype,m),n.disposeData(f.dataId)}if(r){let d=new j7(e,l.shape,r,a),h=p,f=[{type:"float32",data:[0]}];p=n.runWebGPUProgram(d,[p],p.dtype,f),n.disposeData(h.dataId)}if(i!=null){let d=T.getUndoAxesPermutation(i),h=Aa({inputs:{x:p},backend:n,attrs:{perm:d}});return n.disposeData(p.dataId),n.disposeData(l.dataId),h}return p}function G1e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,exclusive:o,reverse:i}=s;return aT(Op.Prod,r,n,a,o,i)}var H1e={kernelName:cl,backendName:"webgpu",kernelFunc:G1e};function j1e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,exclusive:o,reverse:i}=s;return aT(Op.Sum,r,n,a,o,i)}var q1e={kernelName:wo,backendName:"webgpu",kernelFunc:j1e},X1e=class{constructor(e,t){this.variableNames=["x"],this.workGroupSize=[64,1,1],this.size=!0,this.uniforms="blockSize : i32,",this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey=`depthToSpace_${t}`,this.dataFormat=t}getUserCode(){return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let b = coords[0];
|
|
let h = ${this.getHeightCoordString()};
|
|
let w = ${this.getWidthCoordString()};
|
|
let d = ${this.getDepthCoordString()};
|
|
|
|
let in_h = h / uniforms.blockSize;
|
|
let offset_h = h % uniforms.blockSize;
|
|
let in_w = w / uniforms.blockSize;
|
|
let offset_w = w % uniforms.blockSize;
|
|
let offset_d = (offset_h * uniforms.blockSize + offset_w) *
|
|
${this.getOutputDepthSize()};
|
|
let in_d = d + offset_d;
|
|
|
|
let rlt = ${this.getInputSamplingString()};
|
|
setOutputAtIndex(index, rlt);
|
|
}
|
|
}`}getHeightCoordString(){return this.dataFormat==="NHWC"?"coords[1]":"coords[2]"}getWidthCoordString(){return this.dataFormat==="NHWC"?"coords[2]":"coords[3]"}getDepthCoordString(){return this.dataFormat==="NHWC"?"coords[3]":"coords[1]"}getOutputDepthSize(){return this.dataFormat==="NHWC"?"uniforms.outShape[3]":"uniforms.outShape[1]"}getInputSamplingString(){return this.dataFormat==="NHWC"?"getX(b, in_h, in_w, in_d)":"getX(b, in_d, in_h, in_w)"}};function K1e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockSize:a,dataFormat:o}=s,i=r.shape[0],l=o==="NHWC"?r.shape[1]:r.shape[2],u=o==="NHWC"?r.shape[2]:r.shape[3],c=o==="NHWC"?r.shape[3]:r.shape[1],p=l*a,d=u*a,h=c/(a*a),f=o==="NHWC"?[i,p,d,h]:[i,h,p,d],m=[{type:"int32",data:[a]}],g=new X1e(f,o);return n.runWebGPUProgram(g,[r],r.dtype,m)}var Z1e={kernelName:pl,backendName:"webgpu",kernelFunc:K1e},Y1e=class{constructor(e,t,n,s=!1,r=null,a=!1){this.variableNames=["x","W"],this.uniforms="pad : vec2<i32>, inDims : vec2<i32>,",this.workGroupSize=[16,16,1],this.outputShape=e,this.dispatchLayout={x:[3],y:[2],z:[0,1]},this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),s&&this.variableNames.push("bias"),a&&this.variableNames.push("preluActivationWeights"),this.addBias=s,this.activation=r,this.hasPreluActivation=a,this.filterHeight=t,this.filterWidth=n,this.shaderKey=`depthwiseNCHW_${this.activation}_${this.filterHeight}_${this.filterWidth}`}getUserCode(){let e=this.filterWidth*this.filterHeight,t=this.workGroupSize[0]*this.workGroupSize[1]*this.workGroupSize[2],n=this.workGroupSize[1]+this.filterHeight-1,s=this.workGroupSize[0]+this.filterWidth-1;return`
|
|
${Ca(this.activation,this.hasPreluActivation,!1,4)}
|
|
|
|
var<workgroup> mm_Asub : array<array<f32, ${s}>, ${n}>;
|
|
var<workgroup> mm_Bsub : array<array<f32, ${this.filterWidth}>, ${this.filterHeight}>;
|
|
fn readX(batch : i32, channel : i32, row : i32, col : i32) -> f32 {
|
|
var value = 0.0;
|
|
if (row >=0 && row < uniforms.inDims[0] && col >=0 && col < uniforms.inDims[1])
|
|
{
|
|
value = getX(batch, channel, row, col);
|
|
}
|
|
return value;
|
|
}
|
|
|
|
${T2()}
|
|
fn main(@builtin(local_invocation_id) LocalId : vec3<u32>,
|
|
@builtin(global_invocation_id) GlobalId : vec3<u32>,
|
|
@builtin(local_invocation_index) LocalIndex: u32,
|
|
@builtin(num_workgroups) NumWorkgroups: vec3<u32>) {
|
|
localId = LocalId;
|
|
globalId = GlobalId;
|
|
let localIndex = i32(LocalIndex);
|
|
numWorkgroups = NumWorkgroups;
|
|
let coords = getOutputCoords();
|
|
let batch = coords[0];
|
|
let xRCCorner = vec2<i32>(coords.zw) - uniforms.pad;
|
|
let channelMul = uniforms.wShape[3];
|
|
let d1 = coords[1] / channelMul;
|
|
let q = coords[1] % channelMul;
|
|
|
|
let inputRowStart = xRCCorner.x;
|
|
let inputColStart = xRCCorner.y;
|
|
|
|
let localRow = i32(localId.y);
|
|
let localCol = i32(localId.x);
|
|
|
|
// Load one tile of X into local memory.
|
|
for (var inputRow = localRow; inputRow < ${n}; inputRow = inputRow + ${this.workGroupSize[1]}) {
|
|
for (var inputCol = localCol; inputCol < ${s}; inputCol = inputCol + ${this.workGroupSize[0]}) {
|
|
let rowOffset = inputRow - localRow;
|
|
let colOffset = inputCol - localCol;
|
|
mm_Asub[inputRow][inputCol] = readX(batch, d1, inputRowStart + rowOffset, inputColStart + colOffset);
|
|
}
|
|
}
|
|
|
|
// Load one tile of W into local memory.
|
|
var wIndex = localIndex;
|
|
${e<t?`if (wIndex < ${e})`:`for(; wIndex < ${e}; wIndex = wIndex + ${t})`}
|
|
|
|
{
|
|
let wRow = wIndex / ${this.filterWidth};
|
|
let wCol = wIndex % ${this.filterWidth};
|
|
mm_Bsub[wRow][wCol] = getW(wRow, wCol, d1, q);
|
|
}
|
|
|
|
workgroupBarrier();
|
|
|
|
var value = 0.0;
|
|
for (var wR = 0; wR < ${this.filterHeight}; wR = wR + 1) {
|
|
for (var wC = 0; wC < ${this.filterWidth}; wC = wC + 1) {
|
|
let xVal = mm_Asub[localRow + wR][localCol + wC];
|
|
let wVal = mm_Bsub[wR][wC];
|
|
value = fma(xVal, wVal, value);
|
|
}
|
|
}
|
|
${dd(this.addBias,this.activation)}
|
|
if (coordsInBounds4D(coords, uniforms.outShape)) {
|
|
setOutputAtCoords(coords[0], coords[1], coords[2], coords[3], value);
|
|
}
|
|
}
|
|
`}},oT=class{constructor(e,t=!1,n=null,s=!1){this.variableNames=["x","W"],this.uniforms="pad : vec2<i32>, inDims : vec2<i32>,",this.workGroupSize=[4,4,4],this.isVec4=!0,this.outputShape=e.outShape,this.dispatchLayout={x:[3],y:[2],z:[0,1]},this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[4,4,1]),v.assert(e.dataFormat==="channelsLast",()=>"TODO: NCHW is unimplemented"),t&&this.variableNames.push("bias"),s&&this.variableNames.push("preluActivationWeights"),this.convInfo=e,this.addBias=t,this.activation=n,this.hasPreluActivation=s,this.shaderKey=`depthwiseVec4_${n}_${this.convInfo.filterHeight}_${this.convInfo.filterWidth}`}getUserCode(){let e=4+this.convInfo.filterWidth-1;return`
|
|
${Ca(this.activation,this.hasPreluActivation,!0,4)}
|
|
fn readX(batch : i32, row : i32, col : i32, channel : i32) -> vec4<f32> {
|
|
var value = vec4<f32>(0.0);
|
|
if (row >=0 && row < uniforms.inDims[0] && col >=0 && col < uniforms.inDims[1])
|
|
{
|
|
value = getX(batch, row, col, channel);
|
|
}
|
|
return value;
|
|
}
|
|
${T2()}
|
|
fn main(@builtin(global_invocation_id) globalId: vec3<u32>) {
|
|
let batch = i32(globalId.z) / uniforms.outShape[1];
|
|
let r = i32(globalId.z) % uniforms.outShape[1];
|
|
let c = i32(globalId.y) * 4;
|
|
let d1 = i32(globalId.x) * 4;
|
|
let xRCCorner = vec2<i32>(r, c) - uniforms.pad;
|
|
|
|
let xRCorner = xRCCorner.x;
|
|
let xCCorner = xRCCorner.y;
|
|
var xVals : array<vec4<f32>, ${e}>;
|
|
var dotProd : array<vec4<f32>, 4>;
|
|
dotProd[0] = vec4<f32>(0.0);
|
|
dotProd[1] = vec4<f32>(0.0);
|
|
dotProd[2] = vec4<f32>(0.0);
|
|
dotProd[3] = vec4<f32>(0.0);
|
|
|
|
// Use constant instead of uniform can give better performance.
|
|
for (var wR = 0; wR < ${this.convInfo.filterHeight}; wR = wR + 1) {
|
|
let xR = xRCorner + wR;
|
|
for (var i = 0; i < ${e}; i++)
|
|
{
|
|
xVals[i] = readX(batch, xR, xCCorner + i, d1);
|
|
}
|
|
for (var wC = 0; wC < ${this.convInfo.filterWidth}; wC = wC + 1) {
|
|
let wValue = getW(wR, wC, d1, 0);
|
|
dotProd[0] = dotProd[0] + xVals[0 + wC] * wValue;
|
|
dotProd[1] = dotProd[1] + xVals[1 + wC] * wValue;
|
|
dotProd[2] = dotProd[2] + xVals[2 + wC] * wValue;
|
|
dotProd[3] = dotProd[3] + xVals[3 + wC] * wValue;
|
|
}
|
|
}
|
|
|
|
for (var i = 0; i < 4; i = i + 1) {
|
|
let coords = vec4<i32>(batch, r, c + i, d1);
|
|
if (coordsInBounds4D(coords, uniforms.outShape)) {
|
|
var value = dotProd[i];
|
|
${dd(this.addBias,this.activation)}
|
|
setOutputAtCoords(coords[0], coords[1], coords[2], coords[3], value);
|
|
}
|
|
}
|
|
}
|
|
`}},iT=class{constructor(e,t=!1,n=null,s=!1){this.variableNames=["x","W"],this.uniforms=`pad : vec2<i32>, inDims : vec2<i32>, filterHeight : i32,
|
|
filterWidth : i32, stride : vec2<i32>, dilation : vec2<i32>,`,this.workGroupSize=[256,1,1],this.outputShape=e.outShape,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.isChannelsLast=e.dataFormat==="channelsLast",t&&this.variableNames.push("bias"),s&&this.variableNames.push("preluActivationWeights"),this.convInfo=e,this.addBias=t,this.activation=n,this.hasPreluActivation=s,this.shaderKey=`depthwise_${this.activation}_${this.isChannelsLast}`}getUserCode(){let e=this.isChannelsLast?"getX(batch, xR, xC, d1);":"getX(batch, d1, xR, xC);";return`
|
|
${Ca(this.activation,this.hasPreluActivation,!1,4)}
|
|
|
|
${pd()}
|
|
let coords = getOutputCoords();
|
|
let batch = coords[0];
|
|
let xRCCorner = vec2<i32>(coords.${this.isChannelsLast?"yz":"zw"}) * uniforms.stride - uniforms.pad;
|
|
let d2 = coords[${this.isChannelsLast?3:1}];
|
|
let channelMul = uniforms.wShape[3];
|
|
let d1 = d2 / channelMul;
|
|
let q = d2 % channelMul;
|
|
|
|
let inputRowStart = xRCCorner.x;
|
|
let inputColStart = xRCCorner.y;
|
|
let inputRowEnd = inputRowStart + uniforms.filterHeight *
|
|
uniforms.dilation[0];
|
|
let inputColEnd = inputColStart + uniforms.filterWidth *
|
|
uniforms.dilation[1];
|
|
|
|
// Convolve x(?, ?, d1)|x(d1, ?, ?) with w(:, :, d1, q) to get
|
|
// y(yR, yC, d2)|y(d2, yR, yC). ? = to be determined. : = across all
|
|
// values in that axis. x(?, ?, d1) and y(yR, yC, d2) is for NHWC.
|
|
// x(d1, ?, ?) and y(d2, yR, yC) is for NCHW.
|
|
var value = 0.0;
|
|
|
|
// Extract if checking out of for loop for performance.
|
|
if (inputRowStart >= 0 && inputColStart >= 0 &&
|
|
inputRowEnd < uniforms.inDims[0] &&
|
|
inputColEnd < uniforms.inDims[1]) {
|
|
for (var wR = 0; wR < uniforms.filterHeight; wR = wR + 1) {
|
|
let xR = inputRowStart + wR * uniforms.dilation[0];
|
|
|
|
for (var wC = 0; wC < uniforms.filterWidth; wC = wC + 1) {
|
|
let xC = inputColStart + wC * uniforms.dilation[1];
|
|
|
|
let xVal = ${e};
|
|
let wVal = getW(wR, wC, d1, q);
|
|
value = value + xVal * wVal;
|
|
}
|
|
}
|
|
} else {
|
|
for (var wR = 0; wR < uniforms.filterHeight; wR = wR + 1) {
|
|
let xR = inputRowStart + wR * uniforms.dilation[0];
|
|
|
|
if (xR < 0 || xR >= uniforms.inDims[0]) {
|
|
continue;
|
|
}
|
|
|
|
for (var wC = 0; wC < uniforms.filterWidth; wC = wC + 1) {
|
|
let xC = inputColStart + wC * uniforms.dilation[1];
|
|
|
|
if (xC < 0 || xC >= uniforms.inDims[1]) {
|
|
continue;
|
|
}
|
|
|
|
let xVal = ${e};
|
|
let wVal = getW(wR, wC, d1, q);
|
|
value = value + xVal * wVal;
|
|
}
|
|
}
|
|
}
|
|
${dd(this.addBias,this.activation)}
|
|
if (coordsInBounds4D(coords, uniforms.outShape)) {
|
|
setOutputAtCoords(coords[0], coords[1], coords[2], coords[3], value);
|
|
}
|
|
}
|
|
`}};function J1e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a}=t,{strides:o,pad:i,dataFormat:l,dilations:u,dimRoundingMode:c}=s,p=T.convertConv2DDataFormat(l),d=u;d==null&&(d=[1,1]);let h=T.computeConv2DInfo(r.shape,a.shape,o,d,i,c,!0,p),f=[{type:"int32",data:[h.padInfo.top,h.padInfo.left]},{type:"int32",data:[h.inHeight,h.inWidth]}],m=h.dataFormat==="channelsLast",g;return!m&&h.inHeight>16&&h.inWidth>16&&h.strideHeight===1&&h.strideWidth===1&&h.dilationWidth===1&&h.dilationHeight===1&&h.inChannels===h.outChannels?g=new Y1e(h.outShape,h.filterHeight,h.filterWidth):m&&h.inHeight>4&&h.inWidth>4&&h.strideHeight===1&&h.strideWidth===1&&h.inChannels===h.outChannels&&h.dilationHeight===1&&h.dilationWidth===1&&h.inChannels%4===0?g=new oT(h):(g=new iT(h),f.push({type:"int32",data:[h.filterHeight]},{type:"int32",data:[h.filterWidth]},{type:"int32",data:[h.strideHeight,h.strideWidth]},{type:"int32",data:[h.dilationHeight,h.dilationWidth]})),n.runWebGPUProgram(g,[r,a],r.dtype,f)}var Q1e={kernelName:ko,backendName:"webgpu",kernelFunc:J1e},lT=os({opType:Ye.MUL,cpuKernelImpl:N2e,supportsComplex:!0}),ege={kernelName:Wo,backendName:"webgpu",kernelFunc:lT},tge=class{constructor(e,t){this.workGroupSize=[64,1,1],this.variableNames=["x"],this.uniforms="reduceSize : i32,",this.size=!0,this.inputShape=[e.batchSize,e.inSize];let[n]=T.computeOutAndReduceShapes(this.inputShape,[1]);this.outputShape=n.length===0?[1]:n,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,[1,1,1]),this.reduceType=t,this.shaderKey=`reduce_${t}`}getUserCode(){let e="",t="0.0";this.reduceType==="min"||this.reduceType==="max"?(e=`
|
|
if (isnan(candidate)) {
|
|
bestValue = uniforms.NAN;
|
|
} else if (!isnan(bestValue) && candidate ${this.reduceType==="min"?"<":">"} bestValue)
|
|
{ bestValue = candidate; }`,t="f32(x[offset])"):this.reduceType==="sum"||this.reduceType==="mean"?e=" bestValue = bestValue + candidate; ":this.reduceType==="prod"&&(e=" bestValue = bestValue * candidate; ",t="1.0");let n=this.reduceType==="mean"?"setOutputAtIndex(outputIndex, bestValue / f32(uniforms.reduceSize));":"setOutputAtIndex(outputIndex, bestValue);";return`
|
|
fn DIV_CEIL(a : u32, b : u32) -> u32 {
|
|
return ((a - 1u) / b + 1u);
|
|
}
|
|
|
|
${`
|
|
var<workgroup> xBestValues : array<f32, ${this.workGroupSize[0]}>;
|
|
`}
|
|
fn getOffset(outputIndex : i32) -> i32 {
|
|
let outputCoords = getCoordsFromIndex(outputIndex);
|
|
let offset = ${this.outputShape.length===1?"outputCoords":"outputCoords[0]"} * uniforms.reduceSize;
|
|
return offset;
|
|
}
|
|
${lt()}
|
|
let outputIndex = index / i32(workGroupSizeX);
|
|
let offset = getOffset(outputIndex);
|
|
var bestValue = ${t};
|
|
let Length = uniforms.reduceSize;
|
|
let WorkPerThread = DIV_CEIL(u32(Length), workGroupSizeX);
|
|
for (var k = i32(localId.x); k < Length && outputIndex < uniforms.size;
|
|
k = k + i32(workGroupSizeX)) {
|
|
let candidate = f32(x[offset + k]);
|
|
${e}
|
|
}
|
|
xBestValues[localId.x] = bestValue;
|
|
workgroupBarrier();
|
|
|
|
var reduceSize = min(u32(Length), workGroupSizeX);
|
|
for (var currentSize = reduceSize / 2u; reduceSize > 1u;
|
|
currentSize = reduceSize / 2u) {
|
|
let interval = DIV_CEIL(reduceSize, 2u);
|
|
if (localId.x < currentSize) {
|
|
let candidate = xBestValues[localId.x + interval];
|
|
${e}
|
|
xBestValues[localId.x] = bestValue;
|
|
}
|
|
reduceSize = interval;
|
|
workgroupBarrier();
|
|
}
|
|
|
|
if (localId.x == 0u && outputIndex < uniforms.size) {
|
|
${n}
|
|
}
|
|
}
|
|
`}};function Vh(e,t,n,s,r){let a=e.shape.length,o=[],i=v.parseAxisParam(t,e.shape),l=i,u=T.getAxesPermutation(l,a),c=e;u!=null&&(c=Aa({inputs:{x:e},attrs:{perm:u},backend:r}),l=T.getInnerMostAxes(l.length,a),o.push(c)),T.assertAxesAreInnerMostDims(s,l,a);let[p,d]=T.computeOutAndReduceShapes(c.shape,l),h=p;n&&(h=T.expandShapeToKeepDim(p,i));let f;if((s==="max"||s==="prod")&&r.shouldExecuteOnCPU([c])){let m=r.tensorMap.get(c.dataId).values;switch(s){case"max":let g=S2e(m,v.sizeFromShape(d),h,e.dtype);f=r.makeTensorInfo(h,e.dtype,g);break;case"prod":let{outVals:y,outShape:x,outDtype:A}=_2e(c.shape,c.dtype,m,l);f=r.makeTensorInfo(x,A,y);break;default:throw new Error(`${s} CPU implementation is not yet supported.`)}}else{let m=v.sizeFromShape(d),y=v.sizeFromShape(c.shape)/m,x={windowSize:m,inSize:m,batchSize:y,outSize:1},A=s==="mean"?"float32":sh(e.dtype),b=[{type:"int32",data:[m]}],w=new tge(x,s),k=r.runWebGPUProgram(w,[c],A,b);o.push(k),f=Ue({inputs:{x:k},attrs:{shape:h},backend:r})}return o.forEach(m=>r.disposeData(m.dataId)),f}function ub(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;return Vh(r,a,o,"sum",n)}var nge={kernelName:ei,backendName:"webgpu",kernelFunc:ub};function sge(e){let{inputs:t,backend:n,attrs:s}=e,{equation:r}=s,a=t,{allDims:o,summedDims:i,idDims:l}=T.decodeEinsumEquation(r,a.length);T.checkEinsumDimSizes(o.length,l,a);let{path:u,steps:c}=T.getEinsumComputePath(i,l),p=c.length,d=null,h=o.length,f=[];for(let m=0;m<p;++m){for(let g of c[m]){let{permutationIndices:y,expandDims:x}=T.getEinsumPermutation(h,l[g]),A;T.isIdentityPermutation(y)?A=a[g]:(A=Aa({inputs:{x:a[g]},backend:n,attrs:{perm:y}}),f.push(A));let b=A.shape.slice();for(let w=0;w<x.length;++w)b.splice(x[w],0,1);v.arraysEqual(A.shape,b)||(A=Ue({inputs:{x:A},backend:n,attrs:{shape:b}}),f.push(A)),d===null?d=A:(d=lT({inputs:{a:A,b:d},backend:n}),f.push(d))}m<p-1&&(u[m]>=0&&(d=ub({inputs:{x:d},backend:n,attrs:{axis:u[m]-(o.length-h),keepDims:!1}}),f.push(d)),h--)}for(let m of f)m!==d&&n.disposeData(m.dataId);return d}var rge={kernelName:Up,backendName:"webgpu",kernelFunc:sge},age=$n({opType:Fe.ELU}),oge={kernelName:So,backendName:"webgpu",kernelFunc:age},ige=os({opType:Ye.EQUAL,dtype:"bool",cpuKernelImpl:f2e}),lge={kernelName:hl,backendName:"webgpu",kernelFunc:ige},uT=$n({opType:Fe.EXP,cpuKernelImpl:m2e,dtype:"float32"}),uge={kernelName:Co,backendName:"webgpu",kernelFunc:uT};function yy(e){let{inputs:t,attrs:n,backend:s}=e,{dim:r}=n,{input:a}=t,o=a.shape.length,i=a.shape.slice(),l=r;return r<0&&(v.assert(-(o+1)<=r,()=>`Axis must be in the interval [${-(o+1)}, ${o}]`),l=o+r+1),i.splice(l,0,1),Ue({inputs:{x:a},backend:s,attrs:{shape:i}})}var cge={kernelName:fl,backendName:"webgpu",kernelFunc:yy},dge=$n({opType:Fe.EXPM1,cpuKernelImpl:g2e}),pge={kernelName:ml,backendName:"webgpu",kernelFunc:dge},hge=class{constructor(e){this.outputShape=[],this.variableNames=["x"],this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey="flipLeftRight"}getUserCode(){return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let coordX = uniforms.xShape[2] - coords[2] - 1;
|
|
let outputValue = getX(coords[0], coords[1], coordX, coords[3]);
|
|
setOutputAtIndex(index, outputValue);
|
|
}
|
|
}
|
|
`}},fge={kernelName:gl,backendName:"webgpu",kernelFunc:({inputs:e,backend:t})=>{let{image:n}=e,s=t,r=new hge(n.shape);return s.runWebGPUProgram(r,[n],n.dtype)}},mge=$n({opType:Fe.FLOOR,cpuKernelImpl:y2e}),gge={kernelName:To,backendName:"webgpu",kernelFunc:mge},yge=os({opType:Ye.INT_DIV,dtype:"int32"}),Age={kernelName:No,backendName:"webgpu",kernelFunc:yge},xge=class{constructor(e,t,n=!1){this.isFromPixels=!0,this.outputShape=[0],this.variableNames=[],this.workGroupSize=[256,1,1],this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[t,1,1]),this.importVideo=n,this.shaderKey=`fromPixels_${this.importVideo}`}getUserCode(){let e=this.importVideo?"textureLoad(src, vec2<i32>(coords.yx));":"textureLoad(src, vec2<i32>(coords.yx), 0)";return`
|
|
@binding(1) @group(0) var src: ${this.importVideo?"texture_external":"texture_2d<f32>"};
|
|
${lt()}
|
|
let flatIndex = index * uniforms.numChannels;
|
|
if (flatIndex < uniforms.size) {
|
|
let coords = getCoordsFromIndex(flatIndex);
|
|
let values = ${e};
|
|
for (var i = 0; i < uniforms.numChannels; i = i + 1) {
|
|
result[flatIndex + i] = i32(floor(255.0 * values[i]));
|
|
}
|
|
}
|
|
}
|
|
`}},bge={kernelName:bp,backendName:"webgpu",kernelFunc:vge},Ou,Jf=new Map;function vge(e){let{inputs:t,backend:n,attrs:s}=e,{pixels:r}=t,{numChannels:a}=s;if(r==null)throw new Error("pixels passed to tf.browser.fromPixels() can not be null");let o=typeof HTMLVideoElement!="undefined"&&r instanceof HTMLVideoElement,i=typeof HTMLImageElement!="undefined"&&r instanceof HTMLImageElement,l=typeof HTMLCanvasElement!="undefined"&&r instanceof HTMLCanvasElement||typeof OffscreenCanvas!="undefined"&&r instanceof OffscreenCanvas,u=typeof ImageBitmap!="undefined"&&r instanceof ImageBitmap,[c,p]=o?[r.videoWidth,r.videoHeight]:[r.width,r.height],d=[p,c,a],h=q().getBool("WEBGPU_IMPORT_EXTERNAL_TEXTURE")&&o,f=o||i;if(u||l||f){let x;if(h){let $=r;if(!Jf.has($)||Jf.get($).expired){let _={source:$};Jf.set($,n.device.importExternalTexture(_))}x={width:c,height:p,format:null,usage:null,texture:Jf.get($)}}else{f&&(Ou==null&&(Ou=document.createElement("canvas").getContext("2d")),Ou.canvas.width=c,Ou.canvas.height=p,Ou.drawImage(r,0,0,c,p),r=Ou.canvas);let $=GPUTextureUsage.COPY_DST|GPUTextureUsage.RENDER_ATTACHMENT|GPUTextureUsage.TEXTURE_BINDING,_="rgba8unorm",P=n.textureManager.acquireTexture(d[1],d[0],_,$);n.queue.copyExternalImageToTexture({source:r},{texture:P},[d[1],d[0]]),x={width:c,height:p,format:_,usage:$,texture:P}}let A=v.sizeFromShape(d),b=v.computeStrides(d),w=new xge(d,a,h),k=[{type:"uint32",data:[A]},{type:"uint32",data:[a]},{type:"uint32",data:[...b]}],S=n.makeTensorInfo([p,c],"int32"),E=n.tensorMap.get(S.dataId);E.resourceInfo=x;let R=n.runWebGPUProgram(w,[S],"int32",k);return n.disposeData(S.dataId),R}let m=r.data,g=m;if(a!=null&&a!==4){g=new Uint8Array(r.width*r.height*a);let x=m.length,A=0;for(let b=0;b<x;b++)b%4<a&&(g[A++]=m[b])}let y=n.makeTensorInfo(d,"int32",new Int32Array(g));return n.uploadToGPU(y.dataId),y}var wge=class{constructor(e,t,n,s,r){this.uniforms="varianceEpsilon : f32,",this.workGroupSize=[128,1,1],this.size=!0,this.variableNames=["x","mean","variance"],T.assertAndGetBroadcastShape(e,t),T.assertAndGetBroadcastShape(e,n),this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),s!=null&&(T.assertAndGetBroadcastShape(e,s),this.variableNames.push("offset")),r!=null&&(T.assertAndGetBroadcastShape(e,r),this.variableNames.push("scale")),this.offsetShape=s,this.scaleShape=r,this.shaderKey="batchNorm"}getUserCode(){let e="0.0";this.offsetShape!=null&&(e="getOffsetByOutputIndex(index)");let t="1.0";return this.scaleShape!=null&&(t="getScaleByOutputIndex(index)"),`
|
|
${lt()}
|
|
if (index < uniforms.size)
|
|
{
|
|
let xValue = getXByOutputIndex(index);
|
|
let meanValue = getMeanByOutputIndex(index);
|
|
let varianValue = getVarianceByOutputIndex(index);
|
|
let offsetValue = ${e};
|
|
let scaleValue = ${t};
|
|
let inv = scaleValue * inverseSqrt(varianValue + f32(uniforms.varianceEpsilon));
|
|
setOutputAtIndex(index,dot(vec3<f32>(xValue, -meanValue, offsetValue), vec3<f32>(inv, inv, 1.0)));
|
|
}
|
|
}
|
|
`}},kge={kernelName:Eo,backendName:"webgpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{x:s,scale:r,offset:a,mean:o,variance:i}=e,{varianceEpsilon:l}=t,u=n,c=[s,o,i],p=null;a!=null&&(p=a.shape,c.push(a));let d=null;r!=null&&(d=r.shape,c.push(r));let h=new wge(s.shape,o.shape,i.shape,p,d),f=[{type:"float32",data:[l]}];return u.runWebGPUProgram(h,c,s.dtype,f)}};function Ige(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dataFormat:c,dilations:p,dimRoundingMode:d,activation:h,leakyreluAlpha:f}=s,m=T.convertConv2DDataFormat(c),g=T.computeConv2DInfo(r.shape,a.shape,l,p,u,d,!1,m);return rT({x:r,filter:a,convInfo:g,backend:n,bias:o,preluActivationWeights:i,leakyreluAlpha:f,activation:h})}var Sge={kernelName:Qa,backendName:"webgpu",kernelFunc:Ige};function Cge(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dilations:c,dimRoundingMode:p,activation:d,leakyreluAlpha:h}=s,f=c;f==null&&(f=[1,1]),v.assert(T.eitherStridesOrDilationsAreOne(l,f),()=>`Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${l} and dilations '${f}'`);let m=T.computeConv2DInfo(r.shape,a.shape,l,f,u,p,!0),g=[r,a],y=o!=null,x=i!=null;y&&g.push(o),x&&g.push(i);let A=[{type:"int32",data:[m.padInfo.top,m.padInfo.left]},{type:"int32",data:[m.inHeight,m.inWidth]}],b;return m.inHeight>4&&m.inWidth>4&&m.strideHeight===1&&m.strideWidth===1&&m.inChannels===m.outChannels&&m.dilationHeight===1&&m.dilationWidth===1&&m.inChannels%4===0?b=new oT(m,y,d,x):(b=new iT(m,y,d,x),A.push({type:"int32",data:[m.filterHeight]},{type:"int32",data:[m.filterWidth]},{type:"int32",data:[m.strideHeight,m.strideWidth]},{type:"int32",data:[m.dilationHeight,m.dilationWidth]})),d==="leakyrelu"&&(A.push({type:"float32",data:[h]}),b.uniforms+=" alpha : f32,"),n.runWebGPUProgram(b,g,"float32",A)}var Tge={kernelName:eo,backendName:"webgpu",kernelFunc:Cge},Nge=class{constructor(e,t){this.variableNames=["A","indices"],this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=t,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey=`gathernd_${e}`,this.sliceDim=e,this.uniforms=`sliceDim : i32, strides : ${Tn(e)},`}getUserCode(){let e;return this.sliceDim>1?e="uniforms.strides[j]":e="uniforms.strides",`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
var flattenIndex = 0;
|
|
for (var j = 0; j < uniforms.sliceDim; j = j + 1) {
|
|
let indexTemp = i32(round(getIndices(coords[0], j)));
|
|
let strideNum = ${e};
|
|
flattenIndex = flattenIndex + indexTemp * strideNum;
|
|
}
|
|
|
|
setOutputAtIndex(index, getA(flattenIndex, coords[1]));
|
|
}
|
|
}
|
|
`}};function Ege(e){let{inputs:t,backend:n}=e,{params:s,indices:r}=t,a=r.shape,o=a[a.length-1],i=v.sizeFromShape(s.shape),[l,u,c,p]=T.prepareAndValidate(s,r),d=Ue({inputs:{x:r},backend:n,attrs:{shape:[u,o]}}),h=Ue({inputs:{x:s},backend:n,attrs:{shape:[v.sizeFromShape(s.shape)/c,c]}});if(n.shouldExecuteOnCPU([s,r])||s.dtype==="string"){let x=n.readSync(r.dataId),A=n.bufferSync(s),b=A2e(x,A,s.dtype,u,o,c,p,s.shape,i);return n.makeTensorInfo(l,s.dtype,b.values)}let f=new Nge(o,[u,c]),m=[{type:"int32",data:[o]},{type:"int32",data:p}],g=n.runWebGPUProgram(f,[h,d],h.dtype,m),y=Ue({inputs:{x:g},backend:n,attrs:{shape:l}});return n.disposeData(d.dataId),n.disposeData(h.dataId),n.disposeData(g.dataId),y}var Rge={kernelName:Al,backendName:"webgpu",kernelFunc:Ege},_ge=class{constructor(e,t){this.variableNames=["A","indices"],this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e.slice(),this.aShape=e,this.outputShape=t,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey="gather"}getUserCode(){let e=Dge(this.aShape);return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let resRC = getCoordsFromIndex(index);
|
|
let indexZ = i32(getIndices(resRC.x, resRC.z));
|
|
let inBounds = select(0.0, 1.0, indexZ >= 0 && indexZ < uniforms.aShape[2]);
|
|
setOutputAtIndex(index, inBounds * getA(${e}));
|
|
}
|
|
}
|
|
`}};function Dge(e){let t=["resRC.x","resRC.y","resRC.z","resRC.w"],n=[];for(let s=0;s<e.length;s++)s===2?n.push("indexZ"):n.push(`${t[s]}`);return n.join()}function cT(e){let{inputs:t,backend:n,attrs:s}=e,{x:r,indices:a}=t,{axis:o,batchDims:i}=s,l=v.parseAxisParam(o,r.shape)[0],u=T.segment_util.collectGatherOpShapeInfo(r,a,l,i),c=v.sizeFromShape(a.shape),p=[],d=Ue({inputs:{x:r},backend:n,attrs:{shape:[u.batchSize,u.outerSize,u.dimSize,u.sliceSize]}}),h=Ue({inputs:{x:a},backend:n,attrs:{shape:[u.batchSize,c/u.batchSize]}});p.push(d),p.push(h);let f=[u.batchSize,u.outerSize,c/u.batchSize,u.sliceSize];if(n.shouldExecuteOnCPU([r,a])){let A=n.tensorMap.get(h.dataId).values,b=Le(h.shape,h.dtype,A),k=n.tensorMap.get(d.dataId).values,S=Le(d.shape,d.dtype,k),E=x2e(S,b,f);return p.forEach(R=>n.disposeData(R.dataId)),n.makeTensorInfo(u.outputShape,E.dtype,E.values)}let m=new _ge(d.shape,f),g=n.runWebGPUProgram(m,[d,h],d.dtype);p.push(g);let y=Ue({inputs:{x:g},backend:n,attrs:{shape:u.outputShape}});return p.forEach(x=>n.disposeData(x.dataId)),y}var $ge={kernelName:yl,backendName:"webgpu",kernelFunc:cT},Pge=os({opType:Ye.GREATER,cpuKernelImpl:v2e,dtype:"bool"}),Fge={kernelName:xl,backendName:"webgpu",kernelFunc:Pge},Oge=os({opType:Ye.GREATER_EQUAL,dtype:"bool",cpuKernelImpl:b2e}),Mge={kernelName:Ro,backendName:"webgpu",kernelFunc:Oge};function zge(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{alpha:a}=s,o=[{type:"float32",data:[a]}],i=new Bh(r.shape,Fe.LEAKYRELU);return i.uniforms="alpha : f32,",n.runWebGPUProgram(i,[r],"float32",o)}var Lge={kernelName:Do,backendName:"webgpu",kernelFunc:zge},Bge=os({opType:Ye.LESS,dtype:"bool",cpuKernelImpl:k2e}),Wge={kernelName:bl,backendName:"webgpu",kernelFunc:Bge},Vge=os({opType:Ye.LESS_EQUAL,dtype:"bool",cpuKernelImpl:w2e}),Uge={kernelName:vl,backendName:"webgpu",kernelFunc:Vge},Gge=$n({opType:Fe.LOG,cpuKernelImpl:I2e}),Hge={kernelName:$o,backendName:"webgpu",kernelFunc:Gge},jge=os({opType:Ye.LOGICAL_AND,dtype:"bool"}),qge={kernelName:wl,backendName:"webgpu",kernelFunc:jge},Xge=$n({opType:Fe.LOGICAL_NOT}),Kge={kernelName:kl,backendName:"webgpu",kernelFunc:Xge};function dT(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{reductionIndices:a,keepDims:o}=s;return Vh(r,a,o,"max",n)}var Zge={kernelName:Po,backendName:"webgpu",kernelFunc:dT},Yge=os({opType:Ye.MAX,cpuKernelImpl:C2e}),Jge={kernelName:Fo,backendName:"webgpu",kernelFunc:Yge};function Qge(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=s,u=1,c=T.computePool2DInfo(r.shape,a,o,u,i,l),p,d=[];if(c.filterHeight===1&&c.filterWidth===1){if(v.arraysEqual(c.inShape,c.outShape))return Fs({inputs:{x:r},backend:n});p=new tT(c),d.push({type:"int32",data:[c.strideHeight,c.strideWidth]})}else p=new eT(c,"max"),d.push({type:"int32",data:[c.strideHeight,c.strideWidth]},{type:"int32",data:[c.padInfo.top,c.padInfo.left]},{type:"int32",data:[c.dilationHeight,c.dilationWidth]},{type:"int32",data:[c.inHeight,c.inWidth]},{type:"int32",data:[c.effectiveFilterHeight,c.effectiveFilterWidth]});return n.runWebGPUProgram(p,[r],r.dtype,d)}var e3e={kernelName:Oo,backendName:"webgpu",kernelFunc:Qge};function t3e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{keepDims:a,axis:o}=s;return Vh(r,o,a,"mean",n)}var n3e={kernelName:Mo,backendName:"webgpu",kernelFunc:t3e};function s3e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;return Vh(r,a,o,"min",n)}var r3e={kernelName:zo,backendName:"webgpu",kernelFunc:s3e},a3e=os({opType:Ye.MIN,cpuKernelImpl:T2e}),o3e={kernelName:Lo,backendName:"webgpu",kernelFunc:a3e},i3e=class{constructor(e,t,n){this.uniforms="",this.variableNames=["x"],this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=t.map((s,r)=>s[0]+e[r]+s[1]),this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.xShape=e,t.map((s,r)=>{this.uniforms+=` pad${r} : vec2<i32>,`}),this.offset=n==="reflect"?0:1,this.shaderKey=`mirrorPad_${n}`}getUserCode(){let e=this.xShape.length,t=this.xShape.map((l,u)=>`uniforms.pad${u}[0]`).join(","),n=this.xShape.map((l,u)=>`uniforms.pad${u}[0] + uniforms.xShape${e>1?`[${u}]`:""}`).join(","),s=e===1?"start":"start[i]",r=e===1?"end":"end[i]",a=e===1?"outC":"outC[i]",o=Tn(e),i=e>1?["coords[0]","coords[1]","coords[2]","coords[3]"].slice(0,e):"coords";return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let start = ${o}(${t});
|
|
let end = ${o}(${n});
|
|
var outC = getCoordsFromIndex(index);
|
|
for (var i = 0; i < ${e}; i = i + 1) {
|
|
if (${a} < ${s}) {
|
|
${a} = ${s} * 2 - ${a} - ${this.offset};
|
|
} else if(${a} >= ${r}) {
|
|
${a} = (${r} - 1) * 2 - ${a} + ${this.offset};
|
|
}
|
|
}
|
|
let coords = outC - start;
|
|
setOutputAtIndex(index, getX(${i}));
|
|
}
|
|
}
|
|
`}},l3e={kernelName:Bo,backendName:"webgpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{x:s}=e,{paddings:r,mode:a}=t,o=n,i=r.map(c=>({type:"int32",data:[c[0],c[1]]})),l=new i3e(s.shape,r,a);return o.runWebGPUProgram(l,[s],s.dtype,i)}};function u3e(e){let{inputs:t,backend:n}=e,{x:s}=t;if(n.shouldExecuteOnCPU([s])){let a=n.tensorMap.get(s.dataId),[o,i]=E2e(a.values,s.shape,s.dtype);return n.makeTensorInfo(i,s.dtype,o)}let r=new Bh(s.shape,Fe.NEG);return n.runWebGPUProgram(r,[s],s.dtype)}var c3e={kernelName:Il,backendName:"webgpu",kernelFunc:u3e};function d3e(e){console.warn("tf.nonMaxSuppression() in webgpu locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");let{inputs:t,backend:n,attrs:s}=e,{boxes:r,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l}=s,u=n.readSync(r.dataId),c=n.readSync(a.dataId),{selectedIndices:p}=pr.nonMaxSuppressionV3Impl(u,c,o,i,l);return n.makeTensorInfo([p.length],"int32",new Int32Array(p))}var p3e={kernelName:Cl,backendName:"webgpu",kernelFunc:d3e};function h3e(e){console.warn("tf.nonMaxSuppression() in webgpu locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");let{inputs:t,backend:n,attrs:s}=e,{boxes:r,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,softNmsSigma:u}=s,c=n.readSync(r.dataId),p=n.readSync(a.dataId),d=o,h=i,f=l,m=u,{selectedIndices:g,selectedScores:y}=pr.nonMaxSuppressionV5Impl(c,p,d,h,f,m);return[n.makeTensorInfo([g.length],"int32",new Int32Array(g)),n.makeTensorInfo([y.length],"float32",new Float32Array(y))]}var f3e={kernelName:Tl,backendName:"webgpu",kernelFunc:h3e};function Bm(e){let{inputs:t,backend:n}=e,{x:s}=t;if(s.dtype==="complex64"){let r=Wh({inputs:{input:s},backend:n}),a=Bm({inputs:{x:r},backend:n}),o=E2({inputs:{input:s},backend:n}),i=Bm({inputs:{x:o},backend:n}),l=hd({inputs:{real:a,imag:i},backend:n});return n.disposeData(r.dataId),n.disposeData(a.dataId),n.disposeData(o.dataId),n.disposeData(i.dataId),l}else return ou({attrs:{shape:s.shape,dtype:s.dtype,value:s.dtype==="string"?"":0},backend:n})}var m3e={kernelName:Hl,backendName:"webgpu",kernelFunc:Bm};function pT(e){let{inputs:t,backend:n}=e,{x:s}=t;if(s.dtype==="string")throw new Error("onesLike is not supported under string dtype");if(s.dtype==="complex64"){let r=Wh({inputs:{input:s},backend:n}),a=pT({inputs:{x:r},backend:n}),o=E2({inputs:{input:s},backend:n}),i=Bm({inputs:{x:o},backend:n}),l=hd({inputs:{real:a,imag:i},backend:n});return n.disposeData(r.dataId),n.disposeData(a.dataId),n.disposeData(o.dataId),n.disposeData(i.dataId),l}else return ou({attrs:{shape:s.shape,dtype:s.dtype,value:1},backend:n})}var g3e={kernelName:Nl,backendName:"webgpu",kernelFunc:pT};function y3e(e){let{inputs:t,backend:n,attrs:s}=e,{axis:r}=s;if(t.length===1)return yy({inputs:{input:t[0]},backend:n,attrs:{dim:r}});let a=t[0].shape,o=t[0].dtype;t.forEach(c=>{v.assertShapesMatch(a,c.shape,"All tensors passed to stack must have matching shapes"),v.assert(o===c.dtype,()=>"All tensors passed to stack must have matching dtypes")});let i=[],l=t.map(c=>{let p=yy({inputs:{input:c},backend:n,attrs:{dim:r}});return i.push(p),p}),u=sT({inputs:l,backend:n,attrs:{axis:r}});return i.forEach(c=>n.disposeData(c.dataId)),u}var A3e={kernelName:Rl,backendName:"webgpu",kernelFunc:y3e},x3e=class{constructor(e,t){this.variableNames=["x"],this.uniforms="constantValue : f32,",this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=t.map((n,s)=>n[0]+e[s]+n[1]),this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),t.map((n,s)=>{this.uniforms+=` pad${s} : vec2<i32>,`}),this.xShape=e,this.shaderKey="pad"}getUserCode(){let e=this.xShape.length,t=Tn(e),n=this.xShape.map((c,p)=>`uniforms.pad${p}[0]`).join(","),s=this.xShape.map((c,p)=>`uniforms.pad${p}[0] + uniforms.xShape${e>1?`[${p}]`:""}`).join(","),r=e>1?`${t}(${n})`:`${n}`,a=e>1?`${t}(${s})`:`${s}`,o=e>1?"any(outC < start)":"outC < start",i=e>1?"any(outC >= end)":"outC >= end",l=e>1?["coords[0]","coords[1]","coords[2]","coords[3]"].slice(0,e):"coords";return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let start = ${r};
|
|
let end = ${a};
|
|
let outC = getCoordsFromIndex(index);
|
|
|
|
if (${o} || ${i}) {
|
|
setOutputAtIndex(index, uniforms.constantValue);
|
|
} else {
|
|
let coords = outC - start;
|
|
setOutputAtIndex(index, getX(${l}));
|
|
}
|
|
}
|
|
}
|
|
`}},hT=e=>{let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{paddings:a,constantValue:o}=s;if(a.every(u=>v.arraysEqual(u,[0,0])))return Fs({inputs:{x:r},backend:n});if(v.sizeFromShape(r.shape)===0){let u=a.map((c,p)=>c[0]+r.shape[p]+c[1]);return ou({backend:n,attrs:{shape:u,value:o,dtype:r.dtype}})}let i=[{type:"float32",data:[o]}];a.map(u=>i.push({type:"int32",data:[u[0],u[1]]}));let l=new x3e(r.shape,a);return n.runWebGPUProgram(l,[r],r.dtype,i)},b3e={kernelName:Vo,backendName:"webgpu",kernelFunc:hT},v3e=os({opType:Ye.POW}),w3e={kernelName:Uo,backendName:"webgpu",kernelFunc:v3e};function k3e(e){let{inputs:t,backend:n}=e,{x:s,alpha:r}=t,a=new fy(Ye.PRELU,s.shape,r.shape);return n.runWebGPUProgram(a,[s,r],"float32")}var I3e={kernelName:Go,backendName:"webgpu",kernelFunc:k3e};function S3e(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{axis:a,keepDims:o}=s;return Vh(r,a,o,"prod",n)}var C3e={kernelName:Ho,backendName:"webgpu",kernelFunc:S3e},T3e=e=>{let{backend:t,attrs:n}=e,{start:s,stop:r,step:a,dtype:o}=n,i=D2e(s,r,a,o);return t.makeTensorInfo([i.length],o,i)},N3e={kernelName:_c,backendName:"webgpu",kernelFunc:T3e},fT=os({opType:Ye.DIV}),E3e={kernelName:Io,backendName:"webgpu",kernelFunc:fT},R3e=$n({opType:Fe.RELU}),_3e={kernelName:jo,backendName:"webgpu",kernelFunc:R3e},D3e=$n({opType:Fe.RELU6}),$3e={kernelName:Ko,backendName:"webgpu",kernelFunc:D3e},P3e=class{constructor(e,t,n){this.variableNames=["x"],this.uniforms="adjustHeightWidth : vec2<f32>, halfPixelCenters : f32,",this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=[e[0],t,n,e[3]],this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey="resizeBilinear"}getUserCode(){return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let b = coords[0];
|
|
let d = coords[3];
|
|
let rc = coords.yz;
|
|
|
|
let effectiveInSize = vec2<f32>(
|
|
f32(uniforms.xShape.y) - uniforms.adjustHeightWidth[0],
|
|
f32(uniforms.xShape.z) - uniforms.adjustHeightWidth[1]);
|
|
|
|
let effectiveOutSize = vec2<f32>(
|
|
f32(uniforms.outShape.y) - uniforms.adjustHeightWidth[0],
|
|
f32(uniforms.outShape.z) - uniforms.adjustHeightWidth[1]);
|
|
|
|
let effectiveInputOverOutputRatioRC =
|
|
effectiveInSize / effectiveOutSize;
|
|
|
|
// Fractional source index
|
|
let sourceFracIndexRC =
|
|
(vec2<f32>(rc) + vec2<f32>(uniforms.halfPixelCenters)) *
|
|
effectiveInputOverOutputRatioRC - vec2<f32>(uniforms.halfPixelCenters);
|
|
|
|
// Compute the four integer indices.
|
|
let sourceFloorRC = vec2<i32>(sourceFracIndexRC);
|
|
let sourceCeilRC = vec2<i32>(
|
|
min(vec2<f32>(uniforms.xShape.yz) - vec2<f32>(1.0), ceil(sourceFracIndexRC)));
|
|
|
|
let topLeft = getX(b, sourceFloorRC.x, sourceFloorRC.y, d);
|
|
let bottomLeft = getX(b, sourceCeilRC.x, sourceFloorRC.y, d);
|
|
let topRight = getX(b, sourceFloorRC.x, sourceCeilRC.y, d);
|
|
let bottomRight = getX(b, sourceCeilRC.x, sourceCeilRC.y, d);
|
|
|
|
let fracRC = sourceFracIndexRC - vec2<f32>(sourceFloorRC);
|
|
|
|
let top = topLeft + (topRight - topLeft) * fracRC.y;
|
|
let bottom = bottomLeft + (bottomRight - bottomLeft) * fracRC.y;
|
|
let newValue = top + (bottom - top) * fracRC.x;
|
|
|
|
setOutputAtIndex(index, newValue);
|
|
}
|
|
}
|
|
`}};function F3e(e){let{inputs:t,backend:n,attrs:s}=e,{images:r}=t,{alignCorners:a,size:o,halfPixelCenters:i}=s,[l,u]=o,c=a&&l>1?1:0,p=a&&u>1?1:0,h=[{type:"float32",data:[c,p]},{type:"float32",data:[i?.5:0]}],f=new P3e(r.shape,l,u);return n.runWebGPUProgram(f,[r],"float32",h)}var O3e={kernelName:Xo,backendName:"webgpu",kernelFunc:F3e},M3e=class{constructor(e,t,n,s){this.variableNames=["x"],this.uniforms="adjustHeightWidth : vec2<f32>, roundBase : f32,",this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=[e[0],t,n,e[3]],this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.halfPixelCenters=s,this.shaderKey=`resizeNearest_${s}`}getUserCode(){let e;return this.halfPixelCenters?e="max((vec2<f32>(rc) + vec2<f32>(0.5)) * effectiveInputOverOutputRatioRC, vec2<f32>(0.0))":e="vec2<f32>(rc) * effectiveInputOverOutputRatioRC",`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let b = coords[0];
|
|
let d = coords[3];
|
|
let rc = coords.yz;
|
|
|
|
let effectiveInSize = vec2<f32>(
|
|
f32(uniforms.xShape.y) - uniforms.adjustHeightWidth[0],
|
|
f32(uniforms.xShape.z) - uniforms.adjustHeightWidth[1]);
|
|
|
|
let effectiveOutSize = vec2<f32>(
|
|
f32(uniforms.outShape.y) - uniforms.adjustHeightWidth[0],
|
|
f32(uniforms.outShape.z) - uniforms.adjustHeightWidth[1]);
|
|
|
|
let effectiveInputOverOutputRatioRC =
|
|
effectiveInSize / effectiveOutSize;
|
|
|
|
// Fractional source index
|
|
let sourceFracIndexRC = ${e};
|
|
|
|
// Compute the coordinators of nearest neighbor point.
|
|
let inputShapeRC = vec2<f32>(f32(uniforms.xShape.y), f32(uniforms.xShape.z));
|
|
let sourceNearestRC = vec2<i32>(
|
|
min(inputShapeRC - 1.0, floor(sourceFracIndexRC + uniforms.roundBase)));
|
|
let newValue = getX(b, sourceNearestRC.x, sourceNearestRC.y, d);
|
|
|
|
setOutputAtIndex(index, newValue);
|
|
}
|
|
}
|
|
`}};function z3e(e){let{inputs:t,backend:n,attrs:s}=e,{images:r}=t,{alignCorners:a,halfPixelCenters:o,size:i}=s,[l,u]=i,c=a&&l>1?1:0,p=a&&u>1?1:0,h=[{type:"float32",data:[c,p]},{type:"float32",data:[a?.5:0]}],f=new M3e(r.shape,l,u,o);return n.runWebGPUProgram(f,[r],r.dtype,h)}var L3e={kernelName:qo,backendName:"webgpu",kernelFunc:z3e},B3e=class{constructor(e,t){this.outputShape=[],this.variableNames=["x"],this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.uniforms=`centerX : f32, centerY : f32, sinRadians : f32,
|
|
cosRadians : f32,`,this.shaderKey="rotate",this.outputShape=e,typeof t=="number"?(this.uniforms+=" fillValue : f32,",this.fillSnippet="var outputValue = uniforms.fillValue;",this.shaderKey+="_float"):(this.uniforms+=" fillValue : vec3<f32>,",this.fillSnippet="var outputValue = uniforms.fillValue[coords[3]];",this.shaderKey+="_vec3")}getUserCode(){return`
|
|
${lt()}
|
|
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
let coordXFloat = (f32(coords[2]) - uniforms.centerX) *
|
|
uniforms.cosRadians - (f32(coords[1]) - uniforms.centerY) *
|
|
uniforms.sinRadians;
|
|
let coordYFloat = (f32(coords[2]) - uniforms.centerX) *
|
|
uniforms.sinRadians + (f32(coords[1]) - uniforms.centerY) *
|
|
uniforms.cosRadians;
|
|
let coordX = i32(round(coordXFloat + uniforms.centerX));
|
|
let coordY = i32(round(coordYFloat + uniforms.centerY));
|
|
${this.fillSnippet}
|
|
if(coordX >= 0 && coordX < uniforms.xShape[2] && coordY >= 0 &&
|
|
coordY < uniforms.xShape[1]) {
|
|
outputValue = getX(coords[0], coordY, coordX, coords[3]);
|
|
}
|
|
setOutputAtIndex(index, outputValue);
|
|
}
|
|
}
|
|
`}},W3e={kernelName:jl,backendName:"webgpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{image:s}=e,{radians:r,fillValue:a,center:o}=t,i=n,l=new B3e(s.shape,a),[u,c]=T.getImageCenter(o,s.shape[1],s.shape[2]),p=[{type:"float32",data:[u]},{type:"float32",data:[c]},{type:"float32",data:[Math.sin(r)]},{type:"float32",data:[Math.cos(r)]}];return typeof a=="number"?p.push({type:"float32",data:[Number.parseFloat(a.toFixed(2))]}):p.push({type:"float32",data:a}),i.runWebGPUProgram(l,[s],s.dtype,p)}},V3e=$n({opType:Fe.RSQRT,cpuKernelImpl:$2e}),U3e={kernelName:Zo,backendName:"webgpu",kernelFunc:V3e},lm=class{constructor(e,t,n,s,r,a,o,i=!0){this.variableNames=["updates","indices"],this.workGroupSize=[64,1,1],this.atomic=!0,this.outputShape=a,this.type=o,this.sumDupeIndices=i,this.dispatchLayout=at(e),this.dispatch=We(this.dispatchLayout,e,this.workGroupSize),this.sliceDimGreaterThanOne=t>1,this.shaderKey=`scatter_${n}_${s}_${this.sliceDimGreaterThanOne}_${o}_${i}`;let l=Tn(r.length);this.uniforms=`sliceDim : i32, strides: ${l}, size: i32,`,this.updatesRank=s,this.indicesRank=n}getUserCode(){let e="";this.indicesRank===1?e="coords[0]":this.indicesRank===2&&(e="coords[0], j");let t=`getIndices(${e})`,n=this.sliceDimGreaterThanOne?"uniforms.strides[j]":"uniforms.strides",s="",r="";this.dispatchLayout.x.length===1?(s="flattenedIndex",r=`
|
|
fn getUpdatesCoordsFromFlatIndex(index : i32) -> i32 {
|
|
return index;
|
|
}
|
|
`):this.dispatchLayout.x.length===2&&(s="vec2<i32>(flattenedIndex, coords[1])",r=`
|
|
fn getUpdatesCoordsFromFlatIndex(index : i32) -> vec2<i32> {
|
|
// N.B. |updates| could be a scalar tensor, conceptually representing a
|
|
// 2D tensor with all values equal to that. By design, its size must be
|
|
// the same as |outShape[1]| in one dimension, and |indicesShape[0]|
|
|
// gives the other.
|
|
let sliceSize = uniforms.outShape[1];
|
|
let d0 = index / sliceSize;
|
|
let d1 = index - d0 * sliceSize;
|
|
return vec2<i32>(d0, d1);
|
|
}
|
|
`);let o=`getUpdates(${Array.from({length:this.updatesRank},(u,c)=>`coords[${c}]`).join(", ")})`,i=(u,c)=>{let p=`atomicAdd(${u}, bitcast<i32>(${c}))`;this.type==="float32"&&(p=`
|
|
{
|
|
var oldBits = 0;
|
|
var newBits = bitcast<i32>(${c});
|
|
loop {
|
|
let info = atomicCompareExchangeWeak(${u}, oldBits, newBits);
|
|
if (info.exchanged) {
|
|
break;
|
|
}
|
|
oldBits = info.old_value;
|
|
let oldValue = bitcast<f32>(oldBits);
|
|
let newValue = oldValue + (${c});
|
|
newBits = bitcast<i32>(newValue);
|
|
}
|
|
}
|
|
`);let d=`atomicStore(${u}, bitcast<i32>(${c}));`;return this.sumDupeIndices?p:d};return`
|
|
${r}
|
|
|
|
${lt()}
|
|
|
|
if (index < uniforms.size) {
|
|
let coords = getUpdatesCoordsFromFlatIndex(index);
|
|
var flattenedIndex = 0;
|
|
for (var j = 0; j < uniforms.sliceDim; j = j + 1) {
|
|
let indexInside = i32(round(${t}));
|
|
flattenedIndex = flattenedIndex + indexInside * ${n};
|
|
}
|
|
let updateValue =
|
|
${Ap(this.type,!1)}(${o});
|
|
let flatIndex = getOutputIndexFromCoords(${s});
|
|
|
|
${i("&result[flatIndex]","updateValue")};
|
|
}
|
|
}`}};function G3e(e){let{inputs:t,backend:n,attrs:s}=e,{indices:r,updates:a}=t,{shape:o}=s,{sliceRank:i,numUpdates:l,sliceSize:u,strides:c,outputSize:p}=T.calculateShapes(a,r,o),d=[p/u,u];if(p===0)return n.makeTensorInfo(o,r.dtype);let h=Ue({inputs:{x:r},backend:n,attrs:{shape:[l,i]}}),f=Ue({inputs:{x:a},backend:n,attrs:{shape:[l,u]}}),m=f.dtype,g=ou({backend:n,attrs:{shape:d,value:0,dtype:m}}),y=v.sizeFromShape(f.shape),x=[{type:"int32",data:[i]},{type:"int32",data:c},{type:"int32",data:[y]}],A=new lm(f.shape,i,h.shape.length,f.shape.length,c,d,m),b=n.runWebGPUProgram(A,[f,h],m,x,g),w=Ue({inputs:{x:b},backend:n,attrs:{shape:o}});return n.disposeData(h.dataId),n.disposeData(f.dataId),n.disposeData(b.dataId),w}var H3e={kernelName:Pl,backendName:"webgpu",kernelFunc:G3e},j3e=class{constructor(e,t,n){this.variableNames=["c","a","b"],this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=t,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.cRank=e,this.rank=n,this.shaderKey="select"}getUserCode(){let e,t;if(this.rank>4)throw Error(`Where for rank ${this.rank} is not yet supported`);if(this.rank===1)t="resRC",e="resRC";else{let s=["resRC.x","resRC.y","resRC.z","resRC.w"],r=[],a=[];for(let o=0;o<this.outputShape.length;o++)a.push(`${s[o]}`),o<this.cRank&&r.push(`${s[o]}`);e=r.join(),t=a.join()}return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let resRC = getCoordsFromIndex(index);
|
|
let cVal = getC(${e});
|
|
if (cVal >= 1.0) {
|
|
setOutputAtIndex(index, getA(${t}));
|
|
} else {
|
|
setOutputAtIndex(index, getB(${t}));
|
|
}
|
|
}
|
|
}
|
|
`}};function q3e(e){let{inputs:t,backend:n}=e,{condition:s,t:r,e:a}=t,o=new j3e(s.shape.length,r.shape,r.shape.length);return n.runWebGPUProgram(o,[s,r,a],Mn(r.dtype,a.dtype))}var X3e={kernelName:Fl,backendName:"webgpu",kernelFunc:q3e},K3e=$n({opType:Fe.SIGMOID}),Z3e={kernelName:Jo,backendName:"webgpu",kernelFunc:K3e},Y3e=$n({opType:Fe.SIN}),J3e={kernelName:Yo,backendName:"webgpu",kernelFunc:Y3e},Q3e=$n({opType:Fe.SINH}),eye={kernelName:Ml,backendName:"webgpu",kernelFunc:Q3e},mT=os({opType:Ye.SUB,cpuKernelImpl:L2e,supportsComplex:!0}),tye={kernelName:si,backendName:"webgpu",kernelFunc:mT};function nye(e){let{inputs:t,backend:n,attrs:s}=e,{logits:r}=t,{dim:a}=s,o=v.parseAxisParam([a],r.shape),i=dT({inputs:{x:r},backend:n,attrs:{reductionIndices:o,keepDims:!1}}),l=T.expandShapeToKeepDim(i.shape,o),u=Ue({inputs:{x:i},backend:n,attrs:{shape:l}}),c=mT({inputs:{a:r,b:u},backend:n}),p=uT({inputs:{x:c},backend:n}),d=ub({inputs:{x:p},backend:n,attrs:{axis:o,keepDims:!1}}),h=Ue({inputs:{x:d},backend:n,attrs:{shape:l}}),f=fT({inputs:{a:p,b:h},backend:n});return n.disposeData(i.dataId),n.disposeData(u.dataId),n.disposeData(c.dataId),n.disposeData(p.dataId),n.disposeData(d.dataId),n.disposeData(h.dataId),f}var sye={kernelName:ti,backendName:"webgpu",kernelFunc:nye},rye=e=>{let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{blockShape:a,paddings:o}=s;v.assert(r.shape.length<=4,()=>"spaceToBatchND for rank > 4 with a WebGPU backend not implemented yet");let i=a.reduce((y,x)=>y*x),l=[[0,0]];l.push(...o);for(let y=1+a.length;y<r.shape.length;++y)l.push([0,0]);let u=[],c=hT({inputs:{x:r},backend:n,attrs:{paddings:l,constantValue:0}}),p=T.getReshaped(c.shape,a,i,!1),d=T.getPermuted(p.length,a.length,!1),h=T.getReshapedPermuted(c.shape,a,i,!1),f=Ue({inputs:{x:c},backend:n,attrs:{shape:p}}),m=Aa({inputs:{x:f},backend:n,attrs:{perm:d}}),g=Ue({inputs:{x:m},backend:n,attrs:{shape:h}});return u.push(c),u.push(f),u.push(m),u.forEach(y=>n.disposeData(y.dataId)),g},aye={kernelName:zl,backendName:"webgpu",kernelFunc:rye},oye=class{constructor(e,t){this.variableNames=["A"],this.workGroupSize=[64,1,1],this.size=!0;let n=new Array(e.length);for(let s=0;s<n.length;s++)n[s]=e[s]*t[s];this.outputShape=n,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.rank=this.outputShape.length,this.shaderKey="tile"}getUserCode(){let e=iye(this.rank,"uniforms.");return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let resRC = getCoordsFromIndex(index);
|
|
setOutputAtIndex(index, getA(${e}));
|
|
}
|
|
}
|
|
`}};function iye(e,t=""){if(e>=5)throw Error(`Tile for rank ${e} is not yet supported`);if(e===1)return`(resRC % ${t}aShape)`;let n=["resRC.x","resRC.y","resRC.z","resRC.w"],s=[];for(let r=0;r<e;r++)s.push(`(${n[r]} % ${t}aShape[${r}])`);return s.join()}function gT(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{reps:a}=s;if(n.shouldExecuteOnCPU([r])||r.dtype==="string"||r.shape.length>=5){let l=n.readSync(r.dataId),u=r.dtype==="string"?l.map(d=>v.decodeString(d)):l,c=Le(r.shape,r.dtype,u),p=B2e(c,a);return n.makeTensorInfo(p.shape,p.dtype,p.values)}let o=new oye(r.shape,a);return n.runWebGPUProgram(o,[r],r.dtype)}var lye={kernelName:va,backendName:"webgpu",kernelFunc:gT};function uye(e){let{inputs:t,backend:n,attrs:s}=e,{sparseIndices:r,sparseValues:a,defaultValue:o}=t,{outputShape:i}=s,{sliceRank:l,numUpdates:u,sliceSize:c,strides:p,outputSize:d}=T.calculateShapes(a,r,i),h=!1;if(a.dtype==="string"){let E=n.bufferSync(r),R=n.bufferSync(a),$=v.decodeString(n.readSync(o.dataId)[0]),_=P2e(E,R,i,d,c,u,l,p,$,h);return n.makeTensorInfo(i,_.dtype,_.values)}let f=[d/c,c],m=Ue({inputs:{x:r},backend:n,attrs:{shape:[u,l]}}),g=a.shape.length?Ue({inputs:{x:a},backend:n,attrs:{shape:[u,c]}}):Fs({inputs:{x:a},backend:n}),y=g.dtype,x=n.makeTensorInfo([],y,v.makeZerosTypedArray(1,y)),A=Ue({inputs:{x:o},backend:n,attrs:{shape:Array(f.length).fill(1)}}),b=gT({inputs:{x:A},backend:n,attrs:{reps:f}}),w=v.sizeFromShape([u,c]),k=[{type:"int32",data:[l]},{type:"int32",data:p},{type:"int32",data:[w]}];switch(u){case 0:break;case 1:{let E=new lm([u,c],l,m.shape.length,g.shape.length,p,f,y,h);n.runWebGPUProgram(E,[g,m],y,k,b)}break;default:{let E=new lm([u,c],l,m.shape.length,x.shape.length,p,f,y,h);n.runWebGPUProgram(E,[x,m],y,k,b)}{let E=new lm([u,c],l,m.shape.length,g.shape.length,p,f,y);n.runWebGPUProgram(E,[g,m],y,k,b)}}let S=Ue({inputs:{x:b},backend:n,attrs:{shape:i}});return n.disposeData(m.dataId),n.disposeData(g.dataId),n.disposeData(A.dataId),n.disposeData(x.dataId),n.disposeData(b.dataId),S}var cye={kernelName:Yp,backendName:"webgpu",kernelFunc:uye};function dye(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{numOrSizeSplits:a,axis:o}=s,i=v.parseAxisParam(o,r.shape)[0],l=T.prepareSplitSize(r,a,i),u=r.shape.length,c=new Array(u).fill(0),p=r.shape.slice();return l.map(d=>{let h=[...p];h[i]=d;let f=fd({inputs:{x:r},backend:n,attrs:{begin:c,size:h}});return c[i]+=d,f})}var pye={kernelName:Ll,backendName:"webgpu",kernelFunc:dye},hye=$n({opType:Fe.SQRT}),fye={kernelName:Qo,backendName:"webgpu",kernelFunc:hye},mye={kernelName:Mc,backendName:"webgpu",kernelFunc:({inputs:e,backend:t})=>{let{x:n}=e,s=t,r=new Bh(n.shape,Fe.SQUARE);return s.runWebGPUProgram(r,[n],n.dtype)}},gye=os({opType:Ye.SQUARED_DIFFERENCE}),yye={kernelName:ni,backendName:"webgpu",kernelFunc:gye},Aye=class{constructor(e){this.variableNames=["x"],this.workPerThread=1,this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize,[this.workPerThread,1,1]);let t=Tn(this.outputShape.length);this.uniforms=`begin : ${t}, strides : ${t}, `,this.shaderKey="stridedSlice"}getUserCode(){let e=this.outputShape.length,t="";if(e===1)t="coords * uniforms.strides + uniforms.begin";else{let s=0;t=this.outputShape.map((r,a)=>(s++,this.outputShape.length===1?`coords * uniforms.strides[${a}] + uniforms.begin[${a}]`:`coords[${s-1}] * uniforms.strides[${a}] + uniforms.begin[${a}]`)).join(",")}return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
setOutputAtIndex(index, getX(${t}));
|
|
}
|
|
}
|
|
`}};function xye(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{begin:a,end:o,strides:i,beginMask:l,endMask:u,ellipsisMask:c,newAxisMask:p,shrinkAxisMask:d}=s,{finalShapeSparse:h,finalShape:f,isIdentity:m,sliceDim0:g,isSimpleSlice:y,begin:x,end:A,strides:b}=Ut.sliceInfo(r.shape,a,o,i,l,u,c,p,d),w;if(m)w=Ue({inputs:{x:r},backend:n,attrs:{shape:f}});else if(g||y){v.assert(r.shape.length>=1,()=>`Input must have rank at least 1, got: ${r.shape.length}`);let k=Ut.computeOutShape(x,A,b),S=fd({inputs:{x:r},backend:n,attrs:{begin:x,size:k}});w=Ue({inputs:{x:S},backend:n,attrs:{shape:f}}),n.disposeData(S.dataId)}else if(n.shouldExecuteOnCPU([r])){let S=n.readSync(r.dataId),E=Le(r.shape,r.dtype,S),R=M2e(h,E,b,x);w=n.makeTensorInfo(f,r.dtype,R.values)}else{let S=new Aye(h),E=[{type:"int32",data:x},{type:"int32",data:b}],R=n.runWebGPUProgram(S,[r],r.dtype,E);w=Ue({inputs:{x:R},backend:n,attrs:{shape:f}}),n.disposeData(R.dataId)}return w}var bye={kernelName:Bl,backendName:"webgpu",kernelFunc:xye};function vye(e){let{inputs:t,backend:n,attrs:s}=e,{separator:r,nGramWidths:a,leftPad:o,rightPad:i,padWidth:l,preserveShortSequences:u}=s,{data:c,dataSplits:p}=t,d=n.readSync(c.dataId),h=n.readSync(p.dataId),[f,m]=z2e(d,h,r,a,o,i,l,u);return[n.makeTensorInfo([f.length],"string",f),n.makeTensorInfo(p.shape,"int32",m)]}var wye={kernelName:zc,backendName:"webgpu",kernelFunc:vye},kye=$n({opType:Fe.TANH}),Iye={kernelName:ri,backendName:"webgpu",kernelFunc:kye},Sye=class{constructor(e){this.variableNames=["x","indices"],this.workGroupSize=[256,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.uniforms=`inputSize : i32, firstPass : i32, negativeInf : f32,
|
|
dir : i32, inc : i32,`,this.shaderKey="swap"}getUserCode(){return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let outC = getCoordsFromIndex(index);
|
|
let batch = outC[0];
|
|
let elemIdx = outC[1];
|
|
// We compare elements pair-wise within a group of size 2 * inc.
|
|
// The comparing rule for each group alternates between ascending
|
|
// and descending. Within each group, we compare each pair at
|
|
// positions i and i+inc. To decide whether an element at position i
|
|
// is x0 or x1, we mod it by 2 * inc, if the result is smaller than
|
|
// inc, it is in the first half of the group, we denote it as x0,
|
|
// otherwise we denote it as x1.
|
|
// For example, as shown in the Bitonic top K paper referenced
|
|
// above, Figure5(a) shows that element[1] is in the second half of
|
|
// the group when group size is 2, but it is in the first half of
|
|
// the group when group size is 4.
|
|
let isFirstInPair = elemIdx % (2 * uniforms.inc) < uniforms.inc;
|
|
var i = 0;
|
|
if (isFirstInPair) {
|
|
i = elemIdx;
|
|
} else {
|
|
i = elemIdx - uniforms.inc;
|
|
}
|
|
|
|
var i0 = 0;
|
|
if (uniforms.firstPass == 1) {
|
|
i0 = i;
|
|
} else {
|
|
i0 = i32(getIndices(batch, i));
|
|
}
|
|
|
|
var i1 = 0;
|
|
if (uniforms.firstPass == 1) {
|
|
i1 = i + uniforms.inc;
|
|
} else {
|
|
i1 = i32(getIndices(batch, i + uniforms.inc));
|
|
}
|
|
|
|
var x0 = f32(0.0);
|
|
var x1 = f32(0.0);
|
|
if (i0 < uniforms.inputSize) {
|
|
x0 = getX(batch, i0);
|
|
} else {
|
|
x0 = uniforms.negativeInf;
|
|
}
|
|
if (i1 < uniforms.inputSize) {
|
|
x1 = getX(batch, i1);
|
|
} else {
|
|
x1 = uniforms.negativeInf;
|
|
}
|
|
|
|
let reverse = elemIdx % (2 * uniforms.dir) >= uniforms.dir;
|
|
let isGreater = x0 > x1 || (x0 == x1 && i1 > i0);
|
|
if (reverse == isGreater) {
|
|
// Elements in opposite order of direction
|
|
let iTemp = i0;
|
|
i0 = i1;
|
|
i1 = iTemp;
|
|
}
|
|
if (isFirstInPair) {
|
|
setOutputAtIndex(index, f32(i0));
|
|
} else {
|
|
setOutputAtIndex(index, f32(i1));
|
|
}
|
|
}
|
|
}
|
|
`}},Cye=class{constructor(e){this.variableNames=["x","indices"],this.workGroupSize=[256,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.uniforms="inputSize : i32, firstPass : i32, k : i32,",this.shaderKey="merge"}getUserCode(){return`
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let outC = getCoordsFromIndex(index);
|
|
let batch = outC[0];
|
|
let elemIdx = outC[1];
|
|
// The output size is half of the previous size.
|
|
// If the previous sequence is | | | | _ _ _ _ | | | | _ _ _ _
|
|
// (k=4), we only need to output the indices at positions |, the
|
|
// indices at positions _ can be thrown away, see Figure5(b) After
|
|
// Phase 2 (Merge phase) in the Bitonic Top K paper referenced
|
|
// above.
|
|
// For example, the paper shows we only need to output the orange
|
|
// bars. The output sequence should look like this | | | | | | | |.
|
|
// Because the sequence is halved, to map the output index back to
|
|
// the previous sequence to find the corresponding value, we need
|
|
// to double the index. When we double the index, we basically
|
|
// interpolate a position, so 2i looks like
|
|
// | _ | _ | _ | _ | _ | _ | _. We move the | to the first k
|
|
// position of each 2k positions by - elemIdx % k. E.g. for output
|
|
// at index 4,5,6,7, we want to get the corresponding element at
|
|
// original index 8,9,10,11, for output at index 8,9,10,11,
|
|
// we want to get the corresponding element at original index
|
|
// 16,17,18,19, so on and so forth.
|
|
|
|
var i = 0;
|
|
if (elemIdx < uniforms.k) {
|
|
i = elemIdx;
|
|
} else {
|
|
i = elemIdx * 2 - elemIdx % uniforms.k;
|
|
}
|
|
var i0 = 0;
|
|
if (uniforms.firstPass == 1) {
|
|
i0 = i;
|
|
} else {
|
|
i0 = i32(getIndices(batch, i));
|
|
}
|
|
var i1 = 0;
|
|
if (uniforms.firstPass == 1) {
|
|
i1 = i + uniforms.k;
|
|
} else {
|
|
i1 = i32(getIndices(batch, i + uniforms.k));
|
|
}
|
|
|
|
let x0 = getX(batch, i0);
|
|
var x1 = f32(0.0);
|
|
if (i1 < uniforms.inputSize) {
|
|
x1 = getX(batch, i1);
|
|
} else {
|
|
x1 = x0;
|
|
}
|
|
|
|
if (x0 >= x1) {
|
|
setOutputAtIndex(index, f32(i0));
|
|
} else {
|
|
setOutputAtIndex(index, f32(i1));
|
|
}
|
|
}
|
|
}
|
|
`}};function Mu(e,t){t!==null&&e.disposeData(t.dataId)}function K7(e){let t=1;for(;t<e;)t*=2;return t}function Tye(e){let{inputs:t,backend:n,attrs:s}=e,{x:r}=t,{k:a,sorted:o}=s,i=r.shape,l=i[i.length-1];if(n.shouldExecuteOnCPU([r])){let w=n.readSync(r.dataId),[k,S]=W2e(w,i,r.dtype,a,o);return[n.makeTensorInfo(k.shape,k.dtype,k.values),n.makeTensorInfo(S.shape,S.dtype,S.values)]}if(a===0)return i[i.length-1]=0,[n.makeTensorInfo(i,r.dtype,[]),n.makeTensorInfo(i,"int32",[])];if(l===1)return[r,ou({attrs:{shape:i,dtype:"int32",value:0},backend:n})];let c=v.sizeFromShape(i)/l,p=Ue({inputs:{x:r},attrs:{shape:[c,l]},backend:n}),d=K7(a),h=K7(l),f=null,m=()=>f===null?[p,p]:[p,f],g=(w,k,S)=>{let E=m(),R=new Sye(S),_=[{type:"int32",data:[l]},{type:"int32",data:[f===null?1:0]},{type:"float32",data:[Number.NEGATIVE_INFINITY]},{type:"int32",data:[w]},{type:"int32",data:[k]}],P=f;f=n.runWebGPUProgram(R,E,"int32",_),Mu(n,P)};for(let w=1;w<d;w*=2){let k=w*2;for(let S=w;S>=1;S/=2)g(k,S,[c,h])}for(let w=h;w>d;w/=2){let k=m(),S=new Cye([c,w/2]),R=[{type:"int32",data:[l]},{type:"int32",data:[f===null?1:0]},{type:"int32",data:[d]}],$=f;f=n.runWebGPUProgram(S,k,"int32",R),Mu(n,$);let _=d/2,P=_*2;for(let C=_;C>=1;C/=2)g(P,C,f.shape)}let y=f;f=fd({inputs:{x:f},backend:n,attrs:{begin:0,size:[c,a]}}),Mu(n,y);let x=cT({inputs:{x:p,indices:f},backend:n,attrs:{axis:1,batchDims:1}});Mu(n,p);let A=i.slice(0,-1);A.push(a),y=f,f=Ue({inputs:{x:f},attrs:{shape:A},backend:n}),Mu(n,y);let b=x;return x=Ue({inputs:{x},attrs:{shape:A},backend:n}),Mu(n,b),[x,f]}var Nye={kernelName:Vl,backendName:"webgpu",kernelFunc:Tye},Eye=class{constructor(e){this.variableNames=["Image","Transforms"],this.uniforms="interpolationModeId : i32, fillModeId : i32, fillValue : f32,",this.workGroupSize=[64,1,1],this.size=!0,this.outputShape=e,this.dispatchLayout=at(this.outputShape),this.dispatch=We(this.dispatchLayout,this.outputShape,this.workGroupSize),this.shaderKey="transform"}getUserCode(){return`
|
|
fn mapCoord(outCoord : f32, len : f32) -> f32{
|
|
var inCoord = outCoord;
|
|
if(uniforms.fillModeId == 2) {
|
|
if (inCoord < 0.0) {
|
|
if (len <= 1.0) {
|
|
inCoord = 0.0;
|
|
} else {
|
|
let sz2 = 2.0 * len;
|
|
if (inCoord < sz2) {
|
|
inCoord = sz2 * f32(i32(f32(-inCoord / sz2))) +
|
|
inCoord;
|
|
}
|
|
if (inCoord < -len) {
|
|
inCoord = inCoord + sz2;
|
|
} else {
|
|
inCoord = -inCoord - 1.0;
|
|
}
|
|
}
|
|
} else if (inCoord > len - 1.0) {
|
|
if (len <= 1.0) {
|
|
inCoord = 0.0;
|
|
} else {
|
|
let sz2 = 2.0 * len;
|
|
inCoord = inCoord - sz2 * f32(i32(f32(inCoord / sz2)));
|
|
if (inCoord >= len) {
|
|
inCoord = sz2 - inCoord - 1.0;
|
|
}
|
|
}
|
|
}
|
|
return clamp(inCoord, 0.0, len - 1.0);
|
|
} else if (uniforms.fillModeId == 3) {
|
|
if (inCoord < 0.0) {
|
|
if (len <= 1.0) {
|
|
inCoord = 0.0;
|
|
} else {
|
|
let sz = len - 1.0;
|
|
inCoord = inCoord + len * (f32(i32(f32(-inCoord / sz))) + 1.0);
|
|
}
|
|
} else if (inCoord > len - 1.0) {
|
|
if (len <= 1.0) {
|
|
inCoord = 0.0;
|
|
} else {
|
|
let sz = len - 1.0;
|
|
inCoord = inCoord - len * f32(i32(f32(inCoord / sz)));
|
|
}
|
|
}
|
|
return clamp(inCoord, 0.0, len - 1.0);
|
|
} else if (uniforms.fillModeId == 4) {
|
|
return clamp(outCoord, 0.0, len - 1.0);
|
|
}
|
|
return outCoord;
|
|
}
|
|
fn readWithFillValue(batch : i32, coordY : i32, coordX : i32,
|
|
channel : i32) -> f32 {
|
|
var outputValue : f32;
|
|
if (0 <= coordY && coordY < uniforms.imageShape[1] && 0 <= coordX && coordX < uniforms.imageShape[2]) {
|
|
outputValue = getImage(batch, coordY, coordX, channel);
|
|
} else {
|
|
outputValue = uniforms.fillValue;
|
|
}
|
|
return outputValue;
|
|
}
|
|
|
|
${lt()}
|
|
if (index < uniforms.size) {
|
|
let coords = getCoordsFromIndex(index);
|
|
var outputValue : f32;
|
|
let batch = coords[0];
|
|
let x = coords[2];
|
|
let y = coords[1];
|
|
let channel = coords[3];
|
|
let xf = f32(x);
|
|
let yf = f32(y);
|
|
let a1 = getTransforms(batch, 0);
|
|
let a2 = getTransforms(batch, 1);
|
|
let a3 = getTransforms(batch, 2);
|
|
let b1 = getTransforms(batch, 3);
|
|
let b2 = getTransforms(batch, 4);
|
|
let b3 = getTransforms(batch, 5);
|
|
let c1 = getTransforms(batch, 6);
|
|
let c2 = getTransforms(batch, 7);
|
|
let projection = c1 * xf + c2 * yf + 1.0;
|
|
if (projection == 0.0) {
|
|
outputValue = uniforms.fillValue;
|
|
} else {
|
|
let inX = (a1 * xf + a2 * yf + a3) / projection;
|
|
let inY = (b1 * xf + b2 * yf + b3) / projection;
|
|
let mapX = mapCoord(inX, f32(uniforms.imageShape[2]));
|
|
let mapY = mapCoord(inY, f32(uniforms.imageShape[1]));
|
|
|
|
if (uniforms.interpolationModeId == 1) {
|
|
let coordY = i32(round(mapY));
|
|
let coordX = i32(round(mapX));
|
|
outputValue = readWithFillValue(batch, coordY, coordX,
|
|
channel);
|
|
} else {
|
|
let yFloor = floor(mapY);
|
|
let xFloor = floor(mapX);
|
|
let yCeil = yFloor + 1.0;
|
|
let xCeil = xFloor + 1.0;
|
|
let valueYFloor = (xCeil - mapX) *
|
|
readWithFillValue(batch, i32(yFloor), i32(xFloor), channel) +
|
|
(mapX - xFloor) *
|
|
readWithFillValue(batch, i32(yFloor), i32(xCeil), channel);
|
|
let valueYCeil = (xCeil - mapX) *
|
|
readWithFillValue(batch, i32(yCeil), i32(xFloor), channel) +
|
|
(mapX - xFloor) *
|
|
readWithFillValue(batch, i32(yCeil), i32(xCeil), channel);
|
|
outputValue = (yCeil - mapY) * valueYFloor +
|
|
(mapY - yFloor) * valueYCeil;
|
|
}
|
|
}
|
|
setOutputAtIndex(index, outputValue);
|
|
}
|
|
}
|
|
`}};function Rye(e){let{inputs:t,backend:n,attrs:s}=e,{image:r,transforms:a}=t,{interpolation:o,fillMode:i,fillValue:l,outputShape:u}=s,[c,p,d,h]=r.shape,[f,m]=u!=null?u:[p,d],g=[c,f,m,h],y=new Eye(g),x=o==="nearest"?1:2,A;switch(i){case"constant":A=1;break;case"reflect":A=2;break;case"wrap":A=3;break;case"nearest":A=4;break;default:A=1;break}let b=[{type:"int32",data:[x]},{type:"int32",data:[A]},{type:"float32",data:[l]}];return n.runWebGPUProgram(y,[r,a],"float32",b)}var _ye={kernelName:Ul,backendName:"webgpu",kernelFunc:Rye};function Dye(e){let{inputs:t,backend:n,attrs:s}=e,{value:r}=t,{axis:a}=s;a<0&&(a+=r.shape.length);let o=r,i=o.shape.length,l=r.shape[a],u=new Array(i-1),c=0;for(let m=0;m<i;m++)m!==a&&(u[c++]=o.shape[m]);let p=[],d=new Array(i).fill(0),h=o.shape.slice();h[a]=1;let f=new Array(l);for(let m=0;m<f.length;m++){d[a]=m;let g=fd({inputs:{x:o},backend:n,attrs:{begin:d,size:h}}),y=Ue({inputs:{x:g},backend:n,attrs:{shape:u}});f[m]=y,p.push(g)}return p.forEach(m=>n.disposeData(m.dataId)),f}var $ye={kernelName:Gl,backendName:"webgpu",kernelFunc:Dye},Pye=[l2e,G2e,j2e,K2e,t1e,s1e,a1e,i1e,p1e,g1e,A1e,w1e,c2e,C1e,_1e,O1e,z1e,B1e,U1e,H1e,q1e,Z1e,Q1e,rge,oge,lge,uge,cge,pge,a2e,fge,bge,gge,Age,kge,Sge,Tge,Rge,$ge,Fge,Mge,u2e,I1e,Lge,Wge,Uge,Hge,qge,Kge,Zge,Jge,e3e,n3e,r3e,o3e,l3e,ege,c3e,p3e,f3e,h1e,g3e,A3e,b3e,w3e,I3e,C3e,N3e,f1e,E3e,_3e,$3e,o2e,O3e,L3e,W3e,U3e,H3e,X3e,Z3e,J3e,eye,c1e,bye,wye,sye,aye,cye,pye,fye,mye,yye,tye,nge,Iye,lye,Nye,_ye,Q2e,$ye,m3e];for(let e of Pye)dr(e);var Fye=class{constructor(e){this.device=e,this.numUsedBuffers=0,this.numFreeBuffers=0,this.freeBuffers=new Map,this.usedBuffers=new Map,this.numBytesUsed=0,this.numBytesAllocated=0}acquireUploadBuffer(e,t){return this.acquireBuffer(e,t,!0)}acquireBuffer(e,t,n=!1){let s=Z7(e,t);if(this.freeBuffers.has(s)||this.freeBuffers.set(s,[]),this.usedBuffers.has(s)||this.usedBuffers.set(s,[]),this.numBytesUsed+=e,this.numUsedBuffers++,this.freeBuffers.get(s).length>0){this.numFreeBuffers--;let a=this.freeBuffers.get(s).shift();return this.usedBuffers.get(s).push(a),a}this.numBytesAllocated+=e;let r=this.device.createBuffer({size:e,usage:t,mappedAtCreation:n});return this.usedBuffers.get(s).push(r),r}releaseBuffer(e,t,n){if(this.freeBuffers.size===0)return;let s=Z7(t,n);this.freeBuffers.has(s)||this.freeBuffers.set(s,[]),this.freeBuffers.get(s).push(e),this.numFreeBuffers++,this.numUsedBuffers--;let r=this.usedBuffers.get(s),a=r.indexOf(e);if(a<0)throw new Error("Cannot release a buffer that was never provided by this buffer manager");r.splice(a,1),this.numBytesUsed-=t}releaseUploadBuffer(e,t,n){e.mapAsync(GPUMapMode.WRITE).then(()=>{this.releaseBuffer(e,t,n)},s=>{})}getNumUsedBuffers(){return this.numUsedBuffers}getNumFreeBuffers(){return this.numFreeBuffers}dispose(){this.freeBuffers.forEach((e,t)=>{e.forEach(n=>{n.destroy()})}),this.usedBuffers.forEach((e,t)=>{e.forEach(n=>{n.destroy()})}),this.freeBuffers=new Map,this.usedBuffers=new Map,this.numUsedBuffers=0,this.numFreeBuffers=0,this.numBytesUsed=0,this.numBytesAllocated=0}};function Z7(e,t){return`${e}_${t}`}var Oye=class{constructor(e){this.device=e,this.numUsedTextures=0,this.numFreeTextures=0,this.freeTextures=new Map,this.usedTextures=new Map,this.numBytesUsed=0,this.numBytesAllocated=0}acquireTexture(e,t,n,s){let r=J7(n),a=e*t*r,o=Y7(e,t,n,s);if(this.freeTextures.has(o)||this.freeTextures.set(o,[]),this.usedTextures.has(o)||this.usedTextures.set(o,[]),this.numBytesUsed+=a,this.numUsedTextures++,this.freeTextures.get(o).length>0){this.numFreeTextures--;let l=this.freeTextures.get(o).shift();return this.usedTextures.get(o).push(l),l}this.numBytesAllocated+=a;let i=this.device.createTexture({size:[e,t],format:n,usage:s});return this.usedTextures.get(o).push(i),i}releaseTexture(e,t,n,s,r){if(this.freeTextures.size===0)return;let a=Y7(t,n,s,r);this.freeTextures.has(a)||this.freeTextures.set(a,[]),this.freeTextures.get(a).push(e),this.numFreeTextures++,this.numUsedTextures--;let o=this.usedTextures.get(a),i=o.indexOf(e);if(i<0)throw new Error("Cannot release a texture that was never provided by this texture manager");o.splice(i,1);let l=J7(s),u=t*n*l;this.numBytesUsed-=u}getNumUsedTextures(){return this.numUsedTextures}getNumFreeTextures(){return this.numFreeTextures}dispose(){this.freeTextures.forEach((e,t)=>{e.forEach(n=>{n.destroy()})}),this.usedTextures.forEach((e,t)=>{e.forEach(n=>{n.destroy()})}),this.freeTextures=new Map,this.usedTextures=new Map,this.numUsedTextures=0,this.numFreeTextures=0,this.numBytesUsed=0,this.numBytesAllocated=0}};function Y7(e,t,n,s){return`${e}_${t}_${n}_${s}`}function J7(e){if(e==="rgba8unorm")return 16;throw new Error(`${e} is not supported!`)}var Mye=q().getNumber("WEBGPU_CPU_HANDOFF_SIZE_THRESHOLD"),zye=(e,t)=>{let n=e.limits.maxComputeWorkgroupsPerDimension,s=t.dispatchLayout,r=t.dispatch;if(r.every(o=>o<=n))return r;v.assert(r[0]>n&&s.y===void 0&&s.z===void 0,()=>"Dispatch size exceeds WebGPU limits in Y or Z dimension.");let a=Math.ceil(Math.sqrt(r[0]));return a>n?(a=Math.ceil(Math.cbrt(r[0])),v.assert(a<=n,()=>"Total dispatch size exceeds WebGPU maximum."),[a,a,a]):[a,a,1]},R2=class extends cc{constructor(e,t=!1){if(super(),this.commandQueueOwnedIds=new WeakSet,this.dispatchNumberInEncoder=0,this.disposed=!1,this.downloadWaitMs=0,this.tensorDataPendingDisposal=[],this.stagingPendingDisposal=[],this.uniformPendingDisposal=[],this.uploadWaitMs=0,!ab())throw new Error("WebGPU is not supported on this device");this.pipelineCache={},this.device=e,this.queue=e.queue,this.currentCommandEncoder=null,this.currentComputePass=null,this.supportTimeQuery=t,this.bufferManager=new Fye(this.device),this.textureManager=new Oye(this.device),this.tensorMap=new Mp(this,sn()),this.supportTimeQuery&&(this.querySet=this.device.createQuerySet({type:"timestamp",count:2})),q().getBool("WEBGPU_USE_PROFILE_TOOL")&&(this.dummyCanvas=document.createElement("canvas"),this.dummyCanvas.width=1,this.dummyCanvas.height=1,this.dummyContext=this.dummyCanvas.getContext("webgpu"),this.dummyContext.configure({device:e,format:"bgra8unorm"}),document.body.appendChild(this.dummyCanvas))}nextDataId(){return R2.nextDataId++}floatPrecision(){return 32}defaultGpuBufferUsage(){return GPUBufferUsage.STORAGE|GPUBufferUsage.COPY_SRC|GPUBufferUsage.COPY_DST}disposeData(e,t=!1){if(this.tensorDataPendingDisposal.indexOf(e)>=0)return!1;if(!this.tensorMap.has(e))return!0;let n=this.tensorMap.get(e);if(this.decRef(e),!t&&n.refCount>0)return!1;if(this.commandQueueOwnedIds.has(e))return this.tensorDataPendingDisposal.push(e),!1;let{complexTensorInfos:s}=this.tensorMap.get(e);return s!=null&&(this.disposeData(s.real.dataId,t),this.disposeData(s.imag.dataId,t)),this.releaseResource(e),this.tensorMap.delete(e),!0}memory(){return{numBytesInGPU:this.bufferManager.numBytesUsed,numBytesAllocatedInGPU:this.bufferManager.numBytesAllocated,unreliable:!1}}releaseResource(e){let t=this.tensorMap.get(e);if(!(!t||!t.resourceInfo)){if("texture"in t.resourceInfo){let n=t.resourceInfo;n.texture instanceof GPUTexture&&this.textureManager.releaseTexture(n.texture,n.width,n.height,n.format,n.usage),n.texture=null}else{let n=t.resourceInfo;this.bufferManager.releaseBuffer(n.buffer,n.size,n.usage),n.buffer=null}t.resourceInfo=null}}refCount(e){return this.tensorMap.has(e)?this.tensorMap.get(e).refCount:0}incRef(e){let t=this.tensorMap.get(e);t.refCount++}decRef(e){if(this.tensorMap.has(e)){let t=this.tensorMap.get(e);t.refCount--}}write(e,t,n){if(n==="complex64"&&e!=null)throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");let s={id:this.nextDataId()};return this.tensorMap.set(s,{dtype:n,shape:t,values:e,refCount:1}),s}move(e,t,n,s,r){if(s==="complex64")throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");this.tensorMap.set(e,{dtype:s,shape:n,values:t,refCount:r})}submitQueue(){this.ensureComputePassEnded(),this.queue.submit([this.currentCommandEncoder.finish()]),this.currentCommandEncoder=null,this.dispatchNumberInEncoder=0,this.commandQueueOwnedIds=new WeakSet,this.tensorDataPendingDisposal.forEach(e=>{this.releaseResource(e),this.tensorMap.delete(e)}),this.uniformPendingDisposal.forEach(e=>this.bufferManager.releaseBuffer(e.buffer,e.size,e.usage)),this.stagingPendingDisposal.forEach(e=>this.bufferManager.releaseUploadBuffer(e.buffer,e.size,e.usage)),this.tensorDataPendingDisposal=[],this.uniformPendingDisposal=[],this.stagingPendingDisposal=[]}ensureCommandEncoderReady(){this.currentCommandEncoder||(this.currentCommandEncoder=this.device.createCommandEncoder())}ensureComputePassEnded(){this.currentComputePass&&(this.currentComputePass.end(),this.currentComputePass=null)}getComputePass(){return this.currentComputePass||(this.currentComputePass=this.currentCommandEncoder.beginComputePass()),this.currentComputePass}async getBufferData(e,t){let n=this.bufferManager.acquireBuffer(t,GPUBufferUsage.COPY_DST|GPUBufferUsage.MAP_READ);this.ensureCommandEncoderReady(),this.ensureComputePassEnded(),this.currentCommandEncoder.copyBufferToBuffer(e,0,n,0,t),this.submitQueue(),await n.mapAsync(GPUMapMode.READ);let s=n.getMappedRange().slice(0);return n.unmap(),n!=null&&this.bufferManager.releaseBuffer(n,t,GPUBufferUsage.COPY_DST|GPUBufferUsage.MAP_READ),q().getBool("WEBGPU_USE_PROFILE_TOOL")&&(v.assert(this.dummyContext!==void 0,()=>"Fail to get context for profiling tool"),this.dummyContext.getCurrentTexture()),s}convertAndCacheOnCPU(e,t){let n=this.tensorMap.get(e);return this.releaseResource(e),n.values=t,n.values}readSync(e){let t=this.tensorMap.get(e),{values:n}=t;if(n==null)throw new Error("WebGPU readSync is only available for CPU-resident tensors.");return n}async read(e){if(!this.tensorMap.has(e))throw new Error(`Tensor ${e} was not registered!`);let t=this.tensorMap.get(e),{values:n}=t;if(n!=null)return this.convertAndCacheOnCPU(e,n);let s;if(t.dtype==="complex64"){let r=await Promise.all([this.read(t.complexTensorInfos.real.dataId),this.read(t.complexTensorInfos.imag.dataId)]),a=r[0],o=r[1];s=T.mergeRealAndImagArrays(a,o)}else{let r=t.resourceInfo,a=await this.getBufferData(r.buffer,r.size);s=YC(a,t.dtype)}return this.convertAndCacheOnCPU(e,s),s}readToGPU(e){let t=this.tensorMap.get(e),{values:n,dtype:s,shape:r,resourceInfo:a}=t;if(s==="complex64")throw new Error("Does not support reading buffer for complex64 dtype.");if(a==null)throw n!=null?new Error("Data is not on GPU but on CPU."):new Error("There is no data on GPU or CPU.");let o=a.size,i=this.bufferManager.acquireBuffer(o,a.usage);this.ensureCommandEncoderReady(),this.ensureComputePassEnded(),this.currentCommandEncoder.copyBufferToBuffer(a.buffer,0,i,0,o),this.submitQueue();let l=this.makeTensorInfo(r,s),u=sn().makeTensorFromTensorInfo(l),c=this.tensorMap.get(l.dataId);return c.resourceInfo={size:o,usage:this.defaultGpuBufferUsage(),buffer:i},{tensorRef:u,buffer:i,bufSize:o}}bufferSync(e){let t=this.readSync(e.dataId);if(e.dtype==="string")try{let n=t.map(s=>v.decodeString(s));return Le(e.shape,e.dtype,n)}catch(n){throw new Error("Failed to decode encoded string bytes into utf-8")}return Le(e.shape,e.dtype,t)}async time(e){let t=this.activeTimers,n=[],s=!1;this.programTimersStack==null?(this.programTimersStack=n,s=!0):this.activeTimers.push(n),this.activeTimers=n,e();let r=v.flatten(this.activeTimers.map(l=>l.query)).filter(l=>l!=null),a=v.flatten(this.activeTimers.map(l=>l.name)).filter(l=>l!=null);this.activeTimers=t,s&&(this.programTimersStack=null);let o={uploadWaitMs:this.uploadWaitMs,downloadWaitMs:this.downloadWaitMs,kernelMs:null,wallMs:null},i=await Promise.all(r);return o.kernelMs=v.sum(i),o.getExtraProfileInfo=()=>i.map((l,u)=>({name:a[u],ms:l})).map(l=>`${l.name}: ${l.ms}`).join(", "),this.uploadWaitMs=0,this.downloadWaitMs=0,o}makeTensorInfo(e,t,n){return t==="string"&&n!=null&&n.length>0&&v.isString(n[0])&&(n=n.map(r=>v.encodeString(r))),{dataId:this.write(n,e,t),shape:e,dtype:t}}tensorToBinding(e){if(!e)return null;let t=this.tensorMap.get(e.dataId);if("texture"in t.resourceInfo){let s=t.resourceInfo;return s.texture instanceof GPUExternalTexture?s.texture:s.texture.createView()}let n=t.resourceInfo;return{offset:0,size:n.size,buffer:n.buffer}}async getQueryTime(e){return this.supportTimeQuery?this.getTimeFromQuerySet(e):0}uploadToGPU(e){let t=this.tensorMap.get(e);if(t.resourceInfo)return;let n=ZC(t.dtype)*v.sizeFromShape(t.shape),s=this.bufferManager.acquireBuffer(n,this.defaultGpuBufferUsage());if(t.resourceInfo={size:n,usage:this.defaultGpuBufferUsage(),buffer:s},t.values){let r=this.bufferManager.acquireUploadBuffer(n,GPUBufferUsage.MAP_WRITE|GPUBufferUsage.COPY_SRC),a=r.getMappedRange();t.dtype==="int32"||t.dtype==="bool"?new Int32Array(a).set(t.values):new Float32Array(a).set(t.values),r.unmap(),this.ensureCommandEncoderReady(),this.ensureComputePassEnded(),this.currentCommandEncoder.copyBufferToBuffer(r,0,s,0,n);let o={size:n,usage:GPUBufferUsage.MAP_WRITE|GPUBufferUsage.COPY_SRC,buffer:r};this.stagingPendingDisposal.push(o)}}makeUniforms(e){let t=0,n=0,s=[];e.forEach(i=>{i.data.length===0&&(i.data=[1]);let l;switch(i.data.length){case 1:l=4;break;case 2:l=8;break;case 3:l=16;break;case 4:l=16;break;case 5:l=16;break;case 6:l=16;break;default:v.assert(!1,()=>`Unsupported ${i.data.length}D shape`)}(n===5||n===6)&&(l=16),t=Math.ceil(t/l)*l,n=i.data.length,s.push(t),t+=i.data.length*4});let r=new ArrayBuffer(t);e.forEach((i,l)=>{let u=s[l];i.type==="int32"?new Int32Array(r,u,i.data.length).set(i.data):i.type==="uint32"?new Uint32Array(r,u,i.data.length).set(i.data):new Float32Array(r,u,i.data.length).set(i.data)});let a=this.bufferManager.acquireBuffer(t,GPUBufferUsage.COPY_DST|GPUBufferUsage.UNIFORM);this.queue.writeBuffer(a,0,r,0,t);let o={size:t,usage:GPUBufferUsage.COPY_DST|GPUBufferUsage.UNIFORM,buffer:a};return this.uniformPendingDisposal.push(o),{offset:0,size:t,buffer:a}}runWebGPUProgram(e,t,n,s,r){if(r||(r=this.makeTensorInfo(e.outputShape,n)),v.sizeFromShape(r.shape)===0)return this.tensorMap.get(r.dataId).values=v.getTypedArrayFromDType(r.dtype,0),r;this.uploadToGPU(r.dataId),e.dispatch=zye(this.device,e);let a=[],o=[];if(!e.isFromPixels){a.push({type:"float32",data:[NaN]}),o=t.concat(r).map(g=>g.shape);let f="int32";o.map(g=>{a.push({type:f,data:g})});let m=v.computeStrides(r.shape);if(a.push({type:f,data:m}),e.size){let g=v.sizeFromShape(e.outputShape);a.push({type:f,data:[e.isVec4?g/4:g]})}}let i=t.map((f,m)=>{if(f.dtype==="complex64")throw new Error("GPGPUProgram does not support complex64 input. For complex64 dtypes, please separate the program into real and imaginary parts.");return this.uploadToGPU(f.dataId),{dtype:this.tensorMap.get(f.dataId).dtype,shape:f.shape,name:e.variableNames[m]}}),l=O0e(e,o,i,r),u;l in this.pipelineCache?u=this.pipelineCache[l]:(u=P0e(this.device,e,i,r),this.pipelineCache[l]=u),s&&(a=[...a,...s]);let c=[this.tensorToBinding(r),...t.map(f=>this.tensorToBinding(f)),this.makeUniforms(a)],p=this.device.createBindGroup({layout:u.getBindGroupLayout(0),entries:c.map((f,m)=>({binding:m,resource:f}))});this.ensureCommandEncoderReady();let d=this.getComputePass(),h=this.activeTimers!=null;return h&&this.supportTimeQuery&&d.writeTimestamp(this.querySet,0),d.setPipeline(u),d.setBindGroup(0,p),d.dispatchWorkgroups(e.dispatch[0],e.dispatch[1],e.dispatch[2]),h&&this.supportTimeQuery&&d.writeTimestamp(this.querySet,1),this.dispatchNumberInEncoder++,t.forEach(f=>{this.commandQueueOwnedIds.add(f.dataId)}),this.commandQueueOwnedIds.add(r.dataId),q().get("WEBGPU_DEFERRED_SUBMIT_BATCH_SIZE")<=this.dispatchNumberInEncoder&&this.submitQueue(),h&&this.activeTimers.push({name:e.constructor.name,query:this.getQueryTime(this.querySet)}),r}async getTimeFromQuerySet(e){let t=this.bufferManager.acquireBuffer(16,GPUBufferUsage.COPY_SRC|GPUBufferUsage.QUERY_RESOLVE),n=this.bufferManager.acquireBuffer(16,GPUBufferUsage.MAP_READ|GPUBufferUsage.COPY_DST);this.ensureCommandEncoderReady(),this.ensureComputePassEnded(),this.currentCommandEncoder.resolveQuerySet(e,0,2,t,0),this.currentCommandEncoder.copyBufferToBuffer(t,0,n,0,16),this.submitQueue(),await n.mapAsync(GPUMapMode.READ);let s=new BigUint64Array(n.getMappedRange()),r=Number(s[1]-s[0]);return n.unmap(),this.bufferManager.releaseBuffer(n,16,GPUBufferUsage.MAP_READ|GPUBufferUsage.COPY_DST),this.bufferManager.releaseBuffer(t,16,GPUBufferUsage.COPY_SRC|GPUBufferUsage.QUERY_RESOLVE),r/1e6}shouldExecuteOnCPU(e,t=Mye){return q().getBool("WEBGPU_CPU_FORWARD")&&e.every(n=>this.tensorMap.get(n.dataId).resourceInfo==null&&v.sizeFromShape(n.shape)<t)}numDataIds(){return this.tensorMap.numDataIds()-this.tensorDataPendingDisposal.length}dispose(){this.disposed||(this.bufferManager.dispose(),this.textureManager.dispose(),this.disposed=!0)}};R2.nextDataId=0;var yT={};Ve(yT,{WebGPUBackend:()=>R2,webgpu_util:()=>XC});ab()&&ql("webgpu",async()=>{q().set("CHECK_COMPUTATION_FOR_ERRORS",!1);let e={powerPreference:q().get("WEBGPU_USE_LOW_POWER_GPU")?"low-power":"high-performance"},t=await navigator.gpu.requestAdapter(e),n=t.limits,s={},r=t.features.has("timestamp-query");s.requiredLimits={maxComputeWorkgroupStorageSize:n.maxComputeWorkgroupStorageSize,maxComputeWorkgroupsPerDimension:n.maxComputeWorkgroupsPerDimension,maxStorageBufferBindingSize:n.maxStorageBufferBindingSize},r?s.requiredFeatures=["timestamp-query"]:console.warn("This device doesn't support timestamp-query extension. Start Chrome browser with flag --disable-dawn-features=disallow_unsafe_apis then try again. Or zero will shown for the kernel time when profiling mode isenabled. Using performance.now is not workable for webgpu sinceit doesn't support synchronously to read data from GPU.");let a=await t.requestDevice(s);return new R2(a,r)},3);var Lye="3.19.0",Bye="3.19.0",Wye="3.19.0",Vye="3.19.0",Uye="3.19.0",Gye="3.19.0",Hye="3.19.0",Uh={tfjs:Lye,"tfjs-core":Bye,"tfjs-data":Wye,"tfjs-layers":Vye,"tfjs-converter":Uye,"tfjs-backend-webgl":Gye,"tfjs-backend-wasm":Hye};var AT=`
|
|
precision highp float;
|
|
attribute vec2 pos;
|
|
attribute vec2 uv;
|
|
varying vec2 vUv;
|
|
uniform float flipY;
|
|
void main(void) {
|
|
vUv = uv;
|
|
gl_Position = vec4(pos.x, pos.y*flipY, 0.0, 1.);
|
|
}
|
|
`;var xT=`
|
|
precision highp float;
|
|
varying vec2 vUv;
|
|
uniform sampler2D texture;
|
|
uniform float m[20];
|
|
void main(void) {
|
|
vec4 c = texture2D(texture, vUv);
|
|
gl_FragColor.r = m[0] * c.r + m[1] * c.g + m[2] * c.b + m[3] * c.a + m[4];
|
|
gl_FragColor.g = m[5] * c.r + m[6] * c.g + m[7] * c.b + m[8] * c.a + m[9];
|
|
gl_FragColor.b = m[10] * c.r + m[11] * c.g + m[12] * c.b + m[13] * c.a + m[14];
|
|
gl_FragColor.a = m[15] * c.r + m[16] * c.g + m[17] * c.b + m[18] * c.a + m[19];
|
|
}
|
|
`,bT=`
|
|
precision highp float;
|
|
varying vec2 vUv;
|
|
uniform sampler2D texture;
|
|
uniform float m[20];
|
|
void main(void) {
|
|
vec4 c = texture2D(texture, vUv);
|
|
gl_FragColor.r = m[0] * c.r + m[1] * c.g + m[2] * c.b + m[4];
|
|
gl_FragColor.g = m[5] * c.r + m[6] * c.g + m[7] * c.b + m[9];
|
|
gl_FragColor.b = m[10] * c.r + m[11] * c.g + m[12] * c.b + m[14];
|
|
gl_FragColor.a = c.a;
|
|
}
|
|
`,vT=`
|
|
precision highp float;
|
|
varying vec2 vUv;
|
|
uniform vec2 size;
|
|
uniform sampler2D texture;
|
|
vec2 pixelate(vec2 coord, vec2 size) {
|
|
return floor( coord / size ) * size;
|
|
}
|
|
void main(void) {
|
|
gl_FragColor = vec4(0.0);
|
|
vec2 coord = pixelate(vUv, size);
|
|
gl_FragColor += texture2D(texture, coord);
|
|
}
|
|
`,wT=`
|
|
precision highp float;
|
|
varying vec2 vUv;
|
|
uniform sampler2D texture;
|
|
uniform vec2 px;
|
|
void main(void) {
|
|
gl_FragColor = vec4(0.0);
|
|
gl_FragColor += texture2D(texture, vUv + vec2(-7.0*px.x, -7.0*px.y))*0.0044299121055113265;
|
|
gl_FragColor += texture2D(texture, vUv + vec2(-6.0*px.x, -6.0*px.y))*0.00895781211794;
|
|
gl_FragColor += texture2D(texture, vUv + vec2(-5.0*px.x, -5.0*px.y))*0.0215963866053;
|
|
gl_FragColor += texture2D(texture, vUv + vec2(-4.0*px.x, -4.0*px.y))*0.0443683338718;
|
|
gl_FragColor += texture2D(texture, vUv + vec2(-3.0*px.x, -3.0*px.y))*0.0776744219933;
|
|
gl_FragColor += texture2D(texture, vUv + vec2(-2.0*px.x, -2.0*px.y))*0.115876621105;
|
|
gl_FragColor += texture2D(texture, vUv + vec2(-1.0*px.x, -1.0*px.y))*0.147308056121;
|
|
gl_FragColor += texture2D(texture, vUv )*0.159576912161;
|
|
gl_FragColor += texture2D(texture, vUv + vec2( 1.0*px.x, 1.0*px.y))*0.147308056121;
|
|
gl_FragColor += texture2D(texture, vUv + vec2( 2.0*px.x, 2.0*px.y))*0.115876621105;
|
|
gl_FragColor += texture2D(texture, vUv + vec2( 3.0*px.x, 3.0*px.y))*0.0776744219933;
|
|
gl_FragColor += texture2D(texture, vUv + vec2( 4.0*px.x, 4.0*px.y))*0.0443683338718;
|
|
gl_FragColor += texture2D(texture, vUv + vec2( 5.0*px.x, 5.0*px.y))*0.0215963866053;
|
|
gl_FragColor += texture2D(texture, vUv + vec2( 6.0*px.x, 6.0*px.y))*0.00895781211794;
|
|
gl_FragColor += texture2D(texture, vUv + vec2( 7.0*px.x, 7.0*px.y))*0.0044299121055113265;
|
|
}
|
|
`,kT=`
|
|
precision highp float;
|
|
varying vec2 vUv;
|
|
uniform sampler2D texture;
|
|
uniform vec2 px;
|
|
uniform float m[9];
|
|
void main(void) {
|
|
vec4 c11 = texture2D(texture, vUv - px); // top left
|
|
vec4 c12 = texture2D(texture, vec2(vUv.x, vUv.y - px.y)); // top center
|
|
vec4 c13 = texture2D(texture, vec2(vUv.x + px.x, vUv.y - px.y)); // top right
|
|
vec4 c21 = texture2D(texture, vec2(vUv.x - px.x, vUv.y) ); // mid left
|
|
vec4 c22 = texture2D(texture, vUv); // mid center
|
|
vec4 c23 = texture2D(texture, vec2(vUv.x + px.x, vUv.y) ); // mid right
|
|
vec4 c31 = texture2D(texture, vec2(vUv.x - px.x, vUv.y + px.y) ); // bottom left
|
|
vec4 c32 = texture2D(texture, vec2(vUv.x, vUv.y + px.y) ); // bottom center
|
|
vec4 c33 = texture2D(texture, vUv + px ); // bottom right
|
|
gl_FragColor =
|
|
c11 * m[0] + c12 * m[1] + c22 * m[2] +
|
|
c21 * m[3] + c22 * m[4] + c23 * m[5] +
|
|
c31 * m[6] + c32 * m[7] + c33 * m[8];
|
|
gl_FragColor.a = c22.a;
|
|
}
|
|
`;var cb=(e,t,n)=>{let s=new RegExp("\\b"+t+" \\w+ (\\w+)","ig");e.replace(s,(r,a)=>(n[a]=0,r))},db=class{constructor(t,n,s){ge(this,"uniform",{});ge(this,"attribute",{});ge(this,"gl");ge(this,"id");ge(this,"compile",(t,n)=>{let s=this.gl.createShader(n);return s?(this.gl.shaderSource(s,t),this.gl.compileShader(s),this.gl.getShaderParameter(s,this.gl.COMPILE_STATUS)?s:(le(`filter: gl compile failed: ${this.gl.getShaderInfoLog(s)}`),null)):(le("filter: could not create shader"),null)});this.gl=t;let r=this.compile(n,this.gl.VERTEX_SHADER),a=this.compile(s,this.gl.FRAGMENT_SHADER);if(this.id=this.gl.createProgram(),!(!r||!a)){if(!this.id){le("filter: could not create webgl program");return}if(this.gl.attachShader(this.id,r),this.gl.attachShader(this.id,a),this.gl.linkProgram(this.id),!this.gl.getProgramParameter(this.id,this.gl.LINK_STATUS)){le(`filter: gl link failed: ${this.gl.getProgramInfoLog(this.id)}`);return}this.gl.useProgram(this.id),cb(n,"attribute",this.attribute);for(let o in this.attribute)this.attribute[o]=this.gl.getAttribLocation(this.id,o);cb(n,"uniform",this.uniform),cb(s,"uniform",this.uniform);for(let o in this.uniform)this.uniform[o]=this.gl.getUniformLocation(this.id,o)}}};function IT(){let e=0,t=null,n=!1,s=-1,r=[null,null],a=[],o=null,i=null,l=is(100,100),u={},c={INTERMEDIATE:1},p=l.getContext("webgl");if(!p){le("filter: cannot get webgl context");return}this.gl=p;function d(x,A){if(!(x===l.width&&A===l.height)){if(l.width=x,l.height=A,!o){let b=new Float32Array([-1,-1,0,1,1,-1,1,1,-1,1,0,0,-1,1,0,0,1,-1,1,1,1,1,1,0]);o=p.createBuffer(),p.bindBuffer(p.ARRAY_BUFFER,o),p.bufferData(p.ARRAY_BUFFER,b,p.STATIC_DRAW),p.pixelStorei(p.UNPACK_PREMULTIPLY_ALPHA_WEBGL,!0)}p.viewport(0,0,l.width,l.height),r=[null,null]}}function h(x,A){let b=p.createFramebuffer();p.bindFramebuffer(p.FRAMEBUFFER,b);let w=p.createRenderbuffer();p.bindRenderbuffer(p.RENDERBUFFER,w);let k=p.createTexture();return p.bindTexture(p.TEXTURE_2D,k),p.texImage2D(p.TEXTURE_2D,0,p.RGBA,x,A,0,p.RGBA,p.UNSIGNED_BYTE,null),p.texParameteri(p.TEXTURE_2D,p.TEXTURE_MAG_FILTER,p.LINEAR),p.texParameteri(p.TEXTURE_2D,p.TEXTURE_MIN_FILTER,p.LINEAR),p.texParameteri(p.TEXTURE_2D,p.TEXTURE_WRAP_S,p.CLAMP_TO_EDGE),p.texParameteri(p.TEXTURE_2D,p.TEXTURE_WRAP_T,p.CLAMP_TO_EDGE),p.framebufferTexture2D(p.FRAMEBUFFER,p.COLOR_ATTACHMENT0,p.TEXTURE_2D,k,0),p.bindTexture(p.TEXTURE_2D,null),p.bindFramebuffer(p.FRAMEBUFFER,null),{fbo:b,texture:k}}function f(x){return r[x]=r[x]||h(l.width,l.height),r[x]}function m(x=0){if(!i)return;let A=null,b=null,w=!1;e===0?A=t:A=f(s).texture||null,e++,n&&!(x&c.INTERMEDIATE)?(b=null,w=e%2===0):(s=(s+1)%2,b=f(s).fbo||null),p.bindTexture(p.TEXTURE_2D,A),p.bindFramebuffer(p.FRAMEBUFFER,b),p.uniform1f(i.uniform.flipY,w?-1:1),p.drawArrays(p.TRIANGLES,0,6)}function g(x){if(u[x])return i=u[x],p.useProgram((i?i.id:null)||null),i;if(i=new db(p,AT,x),!i)return le("filter: could not get webgl program"),null;let A=Float32Array.BYTES_PER_ELEMENT,b=4*A;return p.enableVertexAttribArray(i.attribute.pos),p.vertexAttribPointer(i.attribute.pos,2,p.FLOAT,!1,b,0*A),p.enableVertexAttribArray(i.attribute.uv),p.vertexAttribPointer(i.attribute.uv,2,p.FLOAT,!1,b,2*A),u[x]=i,i}let y={colorMatrix:x=>{let A=new Float32Array(x);A[4]/=255,A[9]/=255,A[14]/=255,A[19]/=255;let b=A[18]===1&&A[3]===0&&A[8]===0&&A[13]===0&&A[15]===0&&A[16]===0&&A[17]===0&&A[19]===0?bT:xT,w=g(b);!w||(p.uniform1fv(w.uniform.m,A),m())},brightness:x=>{let A=(x||0)+1;y.colorMatrix([A,0,0,0,0,0,A,0,0,0,0,0,A,0,0,0,0,0,1,0])},saturation:x=>{let A=(x||0)*2/3+1,b=(A-1)*-.5;y.colorMatrix([A,b,b,0,0,b,A,b,0,0,b,b,A,0,0,0,0,0,1,0])},desaturate:()=>{y.saturation(-1)},contrast:x=>{let A=(x||0)+1,b=-128*(A-1);y.colorMatrix([A,0,0,0,b,0,A,0,0,b,0,0,A,0,b,0,0,0,1,0])},negative:()=>{y.contrast(-2)},hue:x=>{x=(x||0)/180*Math.PI;let A=Math.cos(x),b=Math.sin(x),w=.213,k=.715,S=.072;y.colorMatrix([w+A*(1-w)+b*-w,k+A*-k+b*-k,S+A*-S+b*(1-S),0,0,w+A*-w+b*.143,k+A*(1-k)+b*.14,S+A*-S+b*-.283,0,0,w+A*-w+b*-(1-w),k+A*-k+b*k,S+A*(1-S)+b*S,0,0,0,0,0,1,0])},desaturateLuminance:()=>{y.colorMatrix([.2764723,.929708,.0938197,0,-37.1,.2764723,.929708,.0938197,0,-37.1,.2764723,.929708,.0938197,0,-37.1,0,0,0,1,0])},sepia:()=>{y.colorMatrix([.393,.7689999,.18899999,0,0,.349,.6859999,.16799999,0,0,.272,.5339999,.13099999,0,0,0,0,0,1,0])},brownie:()=>{y.colorMatrix([.5997023498159715,.34553243048391263,-.2708298674538042,0,47.43192855600873,-.037703249837783157,.8609577587992641,.15059552388459913,0,-36.96841498319127,.24113635128153335,-.07441037908422492,.44972182064877153,0,-7.562075277591283,0,0,0,1,0])},vintagePinhole:()=>{y.colorMatrix([.6279345635605994,.3202183420819367,-.03965408211312453,0,9.651285835294123,.02578397704808868,.6441188644374771,.03259127616149294,0,7.462829176470591,.0466055556782719,-.0851232987247891,.5241648018700465,0,5.159190588235296,0,0,0,1,0])},kodachrome:()=>{y.colorMatrix([1.1285582396593525,-.3967382283601348,-.03992559172921793,0,63.72958762196502,-.16404339962244616,1.0835251566291304,-.05498805115633132,0,24.732407896706203,-.16786010706155763,-.5603416277695248,1.6014850761964943,0,35.62982807460946,0,0,0,1,0])},technicolor:()=>{y.colorMatrix([1.9125277891456083,-.8545344976951645,-.09155508482755585,0,11.793603434377337,-.3087833385928097,1.7658908555458428,-.10601743074722245,0,-70.35205161461398,-.231103377548616,-.7501899197440212,1.847597816108189,0,30.950940869491138,0,0,0,1,0])},polaroid:()=>{y.colorMatrix([1.438,-.062,-.062,0,0,-.122,1.378,-.122,0,0,-.016,-.016,1.483,0,0,0,0,0,1,0])},shiftToBGR:()=>{y.colorMatrix([0,0,1,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,1,0])},convolution:x=>{let A=new Float32Array(x),b=1/l.width,w=1/l.height,k=g(kT);!k||(p.uniform1fv(k.uniform.m,A),p.uniform2f(k.uniform.px,b,w),m())},detectEdges:()=>{y.convolution.call(this,[0,1,0,1,-4,1,0,1,0])},sobelX:()=>{y.convolution.call(this,[-1,0,1,-2,0,2,-1,0,1])},sobelY:()=>{y.convolution.call(this,[-1,-2,-1,0,0,0,1,2,1])},sharpen:x=>{let A=x||1;y.convolution.call(this,[0,-1*A,0,-1*A,1+4*A,-1*A,0,-1*A,0])},emboss:x=>{let A=x||1;y.convolution.call(this,[-2*A,-1*A,0,-1*A,1,1*A,0,1*A,2*A])},blur:x=>{let A=x/7/l.width,b=x/7/l.height,w=g(wT);!w||(p.uniform2f(w.uniform.px,0,b),m(c.INTERMEDIATE),p.uniform2f(w.uniform.px,A,0),m())},pixelate:x=>{let A=x/l.width,b=x/l.height,w=g(vT);!w||(p.uniform2f(w.uniform.size,A,b),m())}};this.add=function(x){let A=Array.prototype.slice.call(arguments,1),b=y[x];a.push({func:b,args:A})},this.reset=function(){a=[]},this.get=function(){return a},this.apply=function(x){d(x.width,x.height),e=0,t||(t=p.createTexture()),p.bindTexture(p.TEXTURE_2D,t),p.texParameteri(p.TEXTURE_2D,p.TEXTURE_WRAP_S,p.CLAMP_TO_EDGE),p.texParameteri(p.TEXTURE_2D,p.TEXTURE_WRAP_T,p.CLAMP_TO_EDGE),p.texParameteri(p.TEXTURE_2D,p.TEXTURE_MIN_FILTER,p.NEAREST),p.texParameteri(p.TEXTURE_2D,p.TEXTURE_MAG_FILTER,p.NEAREST),p.texImage2D(p.TEXTURE_2D,0,p.RGBA,p.RGBA,p.UNSIGNED_BYTE,x);for(let A=0;A<a.length;A++){n=A===a.length-1;let b=a[A];b.func.apply(this,b.args||[])}return l},this.draw=function(x){return this.add("brightness",0),this.apply(x)}}async function _2(e){let t=e.shape.length===4?st(e):e,n=Zt(t,3,2),s=[ga(n[0]),ga(n[1]),ga(n[2])],r=[mn(n[0]),mn(n[1]),mn(n[2])],a=await Promise.all(r.map(h=>h.data())),o=.99*Math.max(a[0][0],a[1][0],a[2][0]),i=[fe(n[0],s[0]),fe(n[1],s[1]),fe(n[2],s[2])],l=[fe(r[0],s[0]),fe(r[1],s[1]),fe(r[2],s[2])],u=[he(o,l[0]),he(o,l[1]),he(o,l[2])],c=[L(i[0],u[0]),L(i[1],u[1]),L(i[2],u[2])],p=on([c[0],c[1],c[2]],2),d=U(p,[1,t.shape[0],t.shape[1],3]);return ee([...n,...s,...r,...i,...l,...u,...c,p,t]),d}var D2=3840,mt=null,cn=null,md=null,$t,Ta={inputSum:0,cacheDiff:1,sumMethod:0,inputTensor:void 0};function is(e,t){let n;if(pe.browser)if(pe.worker){if(typeof OffscreenCanvas=="undefined")throw new Error("canvas error: attempted to run in web worker but OffscreenCanvas is not supported");n=new OffscreenCanvas(e,t)}else{if(typeof document=="undefined")throw new Error("canvas error: attempted to run in browser but DOM is not defined");n=document.createElement("canvas"),n.width=e,n.height=t}else typeof pe.Canvas!="undefined"?n=new pe.Canvas(e,t):typeof globalThis.Canvas!="undefined"&&(n=new globalThis.Canvas(e,t));return n}function $2(e,t){let n=t||is(e.width,e.height);return n.getContext("2d").drawImage(e,0,0),n}async function gd(e,t,n=!0){if(!e)return t.debug&&le("input error: input is missing"),{tensor:null,canvas:null};if(!(e instanceof nt)&&!(typeof Image!="undefined"&&e instanceof Image)&&!(typeof pe.Canvas!="undefined"&&e instanceof pe.Canvas)&&!(typeof globalThis.Canvas!="undefined"&&e instanceof globalThis.Canvas)&&!(typeof ImageData!="undefined"&&e instanceof ImageData)&&!(typeof ImageBitmap!="undefined"&&e instanceof ImageBitmap)&&!(typeof HTMLImageElement!="undefined"&&e instanceof HTMLImageElement)&&!(typeof HTMLMediaElement!="undefined"&&e instanceof HTMLMediaElement)&&!(typeof HTMLVideoElement!="undefined"&&e instanceof HTMLVideoElement)&&!(typeof HTMLCanvasElement!="undefined"&&e instanceof HTMLCanvasElement)&&!(typeof OffscreenCanvas!="undefined"&&e instanceof OffscreenCanvas))throw new Error("input error: type is not recognized");if(e instanceof nt){let s=null;if(e.isDisposedInternal)throw new Error("input error: attempted to use tensor but it is disposed");if(!e.shape)throw new Error("input error: attempted to use tensor without a shape");if(e.shape.length===3){if(e.shape[2]===3)s=Kt(e,0);else if(e.shape[2]===4){let r=ci(e,[0,0,0],[-1,-1,3]);s=Kt(r,0),ee(r)}}else e.shape.length===4&&(e.shape[3]===3?s=On(e):e.shape[3]===4&&(s=no(e,[0,0,0,0],[-1,-1,-1,3])));if(s==null||s.shape.length!==4||s.shape[0]!==1||s.shape[3]!==3)throw new Error(`input error: attempted to use tensor with unrecognized shape: ${e.shape}`);if(s.dtype==="int32"){let r=ye(s,"float32");ee(s),s=r}return{tensor:s,canvas:t.filter.return?cn:null}}else{if(typeof e.readyState!="undefined"&&e.readyState<=2)return t.debug&&le("input stream is not ready"),{tensor:null,canvas:mt};let s=e.naturalWidth||e.videoWidth||e.width||e.shape&&e.shape[1]>0,r=e.naturalHeight||e.videoHeight||e.height||e.shape&&e.shape[2]>0;if(!s||!r)return t.debug&&le("cannot determine input dimensions"),{tensor:null,canvas:mt};let a=s,o=r;if(a>D2&&(a=D2,o=Math.trunc(a*r/s)),o>D2&&(o=D2,a=Math.trunc(o*s/r)),(t.filter.width||0)>0?a=t.filter.width:(t.filter.height||0)>0&&(a=s*((t.filter.height||0)/r)),(t.filter.height||0)>0?o=t.filter.height:(t.filter.width||0)>0&&(o=r*((t.filter.width||0)/s)),!a||!o)throw new Error("input error: cannot determine dimension");(!mt||(mt==null?void 0:mt.width)!==a||(mt==null?void 0:mt.height)!==o)&&(mt=is(a,o));let i=mt.getContext("2d");if(typeof ImageData!="undefined"&&e instanceof ImageData?i.putImageData(e,0,0):t.filter.flip&&typeof i.translate!="undefined"?(i.translate(s,0),i.scale(-1,1),i.drawImage(e,0,0,s,r,0,0,mt==null?void 0:mt.width,mt==null?void 0:mt.height),i.setTransform(1,0,0,1,0,0)):i.drawImage(e,0,0,s,r,0,0,mt==null?void 0:mt.width,mt==null?void 0:mt.height),(!cn||mt.width!==cn.width||(mt==null?void 0:mt.height)!==(cn==null?void 0:cn.height))&&(cn=is(mt.width,mt.height)),t.filter.enabled&&pe.webgl.supported?($t||($t=pe.browser?new IT:null),pe.filter=!!$t,!$t||!$t.add?(t.debug&&le("input process error: cannot initialize filters"),pe.webgl.supported=!1,t.filter.enabled=!1,$2(mt,cn)):($t.reset(),t.filter.brightness!==0&&$t.add("brightness",t.filter.brightness),t.filter.contrast!==0&&$t.add("contrast",t.filter.contrast),t.filter.sharpness!==0&&$t.add("sharpen",t.filter.sharpness),t.filter.blur!==0&&$t.add("blur",t.filter.blur),t.filter.saturation!==0&&$t.add("saturation",t.filter.saturation),t.filter.hue!==0&&$t.add("hue",t.filter.hue),t.filter.negative&&$t.add("negative"),t.filter.sepia&&$t.add("sepia"),t.filter.vintage&&$t.add("brownie"),t.filter.sepia&&$t.add("sepia"),t.filter.kodachrome&&$t.add("kodachrome"),t.filter.technicolor&&$t.add("technicolor"),t.filter.polaroid&&$t.add("polaroid"),t.filter.pixelate!==0&&$t.add("pixelate",t.filter.pixelate),$t.get()>0?cn=$t.apply(mt):cn=$t.draw(mt))):($2(mt,cn),$t&&($t=null),pe.filter=!!$t),!n)return{tensor:null,canvas:cn};if(!cn)throw new Error("canvas error: cannot create output");let l,u=3;if(typeof ImageData!="undefined"&&e instanceof ImageData||e.data&&e.width&&e.height)if(pe.browser&&Js)l=Js?Js.fromPixels(e):null;else{u=e.data.length/e.height/e.width;let d=new Uint8Array(e.data.buffer);l=ct(d,[e.height,e.width,u],"int32")}else if((!md||cn.width!==md.width||cn.height!==md.height)&&(md=is(cn.width,cn.height)),Js&&pe.browser)t.backend==="webgl"||t.backend==="humangl"||t.backend==="webgpu"?l=Js.fromPixels(cn):(md=$2(cn),l=Js.fromPixels(md));else{let f=$2(cn).getContext("2d").getImageData(0,0,a,o);u=f.data.length/a/o;let m=new Uint8Array(f.data.buffer);l=ct(m,[a,o,u])}if(u===4){let d=ci(l,[0,0,0],[-1,-1,3]);ee(l),l=d}if(!l)throw new Error("input error: cannot create tensor");let c=ye(l,"float32"),p=t.filter.equalization?await _2(c):Kt(c,0);return ee([l,c]),{tensor:p,canvas:t.filter.return?cn:null}}}async function ST(e,t){let n=!1;if(e.cacheSensitivity===0||!t.shape||t.shape.length!==4||t.shape[1]>2048||t.shape[2]>2048)return n;if(!Ta.inputTensor)Ta.inputTensor=On(t);else if(Ta.inputTensor.shape[1]!==t.shape[1]||Ta.inputTensor.shape[2]!==t.shape[2])ee(Ta.inputTensor),Ta.inputTensor=On(t);else{let s={};s.diff=fe(t,Ta.inputTensor),s.squared=L(s.diff,s.diff),s.sum=ke(s.squared);let a=(await s.sum.data())[0]/(t.shape[1]||1)/(t.shape[2]||1)/255/3;ee([Ta.inputTensor,s.diff,s.squared,s.sum]),Ta.inputTensor=On(t),n=a<=(e.cacheSensitivity||0)}return n}async function CT(e,t,n){let s={};if(!t||!n||t.shape.length!==4||t.shape.length!==n.shape.length)return e.debug||le("invalid input tensor or tensor shapes do not match:",t.shape,n.shape),0;if(t.shape[0]!==1||n.shape[0]!==1||t.shape[3]!==3||n.shape[3]!==3)return e.debug||le("input tensors must be of shape [1, height, width, 3]:",t.shape,n.shape),0;s.input1=On(t),s.input2=t.shape[1]!==n.shape[1]||t.shape[2]!==n.shape[2]?Se.resizeBilinear(n,[t.shape[1],t.shape[2]]):On(n),s.diff=fe(s.input1,s.input2),s.squared=L(s.diff,s.diff),s.sum=ke(s.squared);let a=(await s.sum.data())[0]/(t.shape[1]||1)/(t.shape[2]||1)/255/3;return ee([s.input1,s.input2,s.diff,s.squared,s.sum]),a}var pb=class{constructor(){ge(this,"browser");ge(this,"node");ge(this,"worker");ge(this,"platform","");ge(this,"agent","");ge(this,"backends",[]);ge(this,"initial");ge(this,"filter");ge(this,"tfjs");ge(this,"offscreen");ge(this,"perfadd",!1);ge(this,"wasm",{supported:void 0,backend:void 0,simd:void 0,multithread:void 0});ge(this,"webgl",{supported:void 0,backend:void 0,version:void 0,renderer:void 0});ge(this,"webgpu",{supported:void 0,backend:void 0,adapter:void 0});ge(this,"cpu",{model:void 0,flags:[]});ge(this,"kernels",[]);ge(this,"Canvas");ge(this,"Image");ge(this,"ImageData");if(this.browser=typeof navigator!="undefined",this.node=typeof process!="undefined"&&typeof process.versions!="undefined"&&typeof process.versions.node!="undefined",this.tfjs={version:Uh["tfjs-core"]},this.offscreen=typeof OffscreenCanvas!="undefined",this.initial=!0,this.worker=this.browser&&this.offscreen?typeof WorkerGlobalScope!="undefined":void 0,typeof navigator!="undefined"){let t=navigator.userAgent.match(/\(([^()]+)\)/g);if(t&&t[0]){let n=t[0].match(/\(([^()]+)\)/g);this.platform=n&&n[0]?n[0].replace(/\(|\)/g,""):"",this.agent=navigator.userAgent.replace(t[0],""),this.platform[1]&&(this.agent=this.agent.replace(t[1],"")),this.agent=this.agent.replace(/ /g," ")}}else typeof process!="undefined"&&(this.platform=`${process.platform} ${process.arch}`,this.agent=`NodeJS ${process.version}`)}async updateBackend(){this.backends=Object.keys(sn().registryFactory),this.wasm.supported=typeof WebAssembly!="undefined",this.wasm.backend=this.backends.includes("wasm"),this.wasm.supported&&this.wasm.backend&&ss()==="wasm"&&(this.wasm.simd=await q().getAsync("WASM_HAS_SIMD_SUPPORT"),this.wasm.multithread=await q().getAsync("WASM_HAS_MULTITHREAD_SUPPORT"));let t=is(100,100),n=t?t.getContext("webgl2"):void 0;if(this.webgl.supported=typeof n!="undefined",this.webgl.backend=this.backends.includes("webgl"),this.webgl.supported&&this.webgl.backend&&(ss()==="webgl"||ss()==="humangl")){let s=Ys().gpgpu!=="undefined"?await Ys().getGPGPUContext().gl:null;s&&(this.webgl.version=s.getParameter(s.VERSION),this.webgl.renderer=s.getParameter(s.RENDERER))}this.webgpu.supported=this.browser&&typeof navigator.gpu!="undefined",this.webgpu.backend=this.backends.includes("webgpu");try{this.webgpu.supported&&(this.webgpu.adapter=(await navigator.gpu.requestAdapter()).name)}catch(s){this.webgpu.supported=!1}try{this.kernels=qr(ss()).map(s=>s.kernelName.toLowerCase())}catch(s){}}async updateCPU(){let t={model:"",flags:[]};this.node&&this.platform.startsWith("linux"),this.cpu?this.cpu=t:Object.defineProperty(this,"cpu",{value:t})}},pe=new pb;var hb={};ia(hb,{age:()=>Kye,antispoof:()=>Zye,blazeface:()=>Qye,"blazeface-back":()=>Yye,"blazeface-front":()=>Jye,"blazepose-detector2d":()=>eAe,"blazepose-detector3d":()=>tAe,"blazepose-full":()=>nAe,"blazepose-heavy":()=>sAe,"blazepose-lite":()=>rAe,default:()=>MAe,efficientpose:()=>aAe,emotion:()=>oAe,faceboxes:()=>iAe,facemesh:()=>hAe,"facemesh-attention":()=>uAe,"facemesh-attention-alt":()=>lAe,"facemesh-detection-full":()=>cAe,"facemesh-detection-short":()=>dAe,"facemesh-orig":()=>pAe,faceres:()=>mAe,"faceres-deep":()=>fAe,gear:()=>gAe,gender:()=>AAe,"gender-ssrnet-imdb":()=>yAe,handdetect:()=>xAe,"handlandmark-full":()=>bAe,"handlandmark-lite":()=>vAe,"handlandmark-sparse":()=>wAe,handskeleton:()=>kAe,handtrack:()=>IAe,iris:()=>SAe,liveness:()=>CAe,"mb3-centernet":()=>TAe,meet:()=>NAe,mobileface:()=>EAe,mobilefacenet:()=>RAe,"movenet-lightning":()=>_Ae,"movenet-multipose":()=>DAe,"movenet-thunder":()=>$Ae,nanodet:()=>PAe,posenet:()=>FAe,selfie:()=>OAe});var Kye=161240,Zye=853098,Yye=538928,Jye=402048,Qye=538928,eAe=7499400,tAe=5928856,nAe=6338290,sAe=27501554,rAe=2725490,aAe=5651240,oAe=820516,iAe=2013002,lAe=2387598,uAe=2382414,cAe=1026192,dAe=201268,pAe=2955780,hAe=1477958,fAe=13957620,mAe=6978814,gAe=1498916,yAe=161236,AAe=201808,xAe=3515612,bAe=5431368,vAe=2023432,wAe=5286322,kAe=5502280,IAe=2964837,SAe=2599092,CAe=592976,TAe=4030290,NAe=372228,EAe=2183192,RAe=5171976,_Ae=4650216,DAe=9448838,$Ae=12477112,PAe=7574558,FAe=5032780,OAe=212886,MAe={age:Kye,antispoof:Zye,"blazeface-back":Yye,"blazeface-front":Jye,blazeface:Qye,"blazepose-detector2d":eAe,"blazepose-detector3d":tAe,"blazepose-full":nAe,"blazepose-heavy":sAe,"blazepose-lite":rAe,efficientpose:aAe,emotion:oAe,faceboxes:iAe,"facemesh-attention-alt":lAe,"facemesh-attention":uAe,"facemesh-detection-full":cAe,"facemesh-detection-short":dAe,"facemesh-orig":pAe,facemesh:hAe,"faceres-deep":fAe,faceres:mAe,gear:gAe,"gender-ssrnet-imdb":yAe,gender:AAe,handdetect:xAe,"handlandmark-full":bAe,"handlandmark-lite":vAe,"handlandmark-sparse":wAe,handskeleton:kAe,handtrack:IAe,iris:SAe,liveness:CAe,"mb3-centernet":TAe,meet:NAe,mobileface:EAe,mobilefacenet:RAe,"movenet-lightning":_Ae,"movenet-multipose":DAe,"movenet-thunder":$Ae,nanodet:PAe,posenet:FAe,selfie:OAe};var ls={cacheModels:!0,cacheSupported:!0,verbose:!0,debug:!1,modelBasePath:""},Or={};async function zAe(e,t){return ls.debug&&le("load model fetch:",e,t),fetch(e,t)}function TT(e){ls.cacheModels=e.cacheModels,ls.verbose=e.debug,ls.modelBasePath=e.modelBasePath}async function Ge(e){var u,c,p;let t=pv(ls.modelBasePath,e||"");t.toLowerCase().endsWith(".json")||(t+=".json");let n=t.includes("/")?t.split("/"):t.split("\\"),s=n[n.length-1].replace(".json",""),r="indexeddb://"+s;Or[s]={name:s,sizeFromManifest:0,sizeLoadedWeights:0,sizeDesired:hb[s],inCache:!1},ls.cacheSupported=typeof window!="undefined"&&typeof window.localStorage!="undefined"&&typeof window.indexedDB!="undefined";let a={};try{a=ls.cacheSupported&&ls.cacheModels?await Ts.listModels():{}}catch(d){ls.cacheSupported=!1}Or[s].inCache=ls.cacheSupported&&ls.cacheModels&&Object.keys(a).includes(r);let o=typeof fetch=="undefined"?{}:{fetchFunc:(d,h)=>zAe(d,h)},i=new $h(Or[s].inCache?r:t,o),l=!1;try{i.findIOHandler(),ls.debug&&le("model load handler:",i.handler);let d=await i.handler.load();Or[s].sizeFromManifest=((u=d==null?void 0:d.weightData)==null?void 0:u.byteLength)||0,i.loadSync(d),Or[s].sizeLoadedWeights=((p=(c=i==null?void 0:i.artifacts)==null?void 0:c.weightData)==null?void 0:p.byteLength)||0,ls.verbose&&le("load model:",i.modelUrl,{bytes:Or[s].sizeLoadedWeights},ls),l=!0}catch(d){le("error loading model:",t,d)}if(l&&ls.cacheModels&&ls.cacheSupported&&!Or[s].inCache)try{let d=await i.save(r);le("model saved:",r,d)}catch(d){le("error saving model:",t,d)}return i}var fb="2.9.2";var m1={};ia(m1,{Models:()=>Qh,getModelStats:()=>T4,load:()=>N4,reset:()=>f1,validate:()=>E4});var mr,mb=[],WAe=["white","black","asian","indian","other"],VAe=[15,23,28,35.5,45.5,55.5,65],NT=0,ET=0,gb=Number.MAX_SAFE_INTEGER;async function RT(e){var t;return pe.initial&&(mr=null),mr?e.debug&&le("cached model:",mr.modelUrl):mr=await Ge((t=e.face.gear)==null?void 0:t.modelPath),mr}async function yb(e,t,n,s){var o,i;if(!mr)return{age:0,gender:"unknown",genderScore:0,race:[]};let r=gb<(((o=t.face.gear)==null?void 0:o.skipFrames)||0),a=(((i=t.face.gear)==null?void 0:i.skipTime)||0)>ue()-ET;return t.skipAllowed&&a&&r&&NT===s&&mb[n]?(gb++,mb[n]):(gb=0,new Promise(async l=>{var y,x;if(!(mr!=null&&mr.inputs[0].shape))return;let u={},c=[[0,.1,.9,.9]];u.resize=Se.cropAndResize(e,c,[0],[mr.inputs[0].shape[2],mr.inputs[0].shape[1]]);let p={age:0,gender:"unknown",genderScore:0,race:[]};(y=t.face.gear)!=null&&y.enabled&&([u.age,u.gender,u.race]=mr.execute(u.resize,["age_output","gender_output","race_output"]));let d=await u.gender.data();p.gender=d[0]>d[1]?"male":"female",p.genderScore=Math.round(100*(d[0]>d[1]?d[0]:d[1]))/100;let h=await u.race.data();for(let A=0;A<h.length;A++)h[A]>(((x=t.face.gear)==null?void 0:x.minConfidence)||.2)&&p.race.push({score:Math.round(100*h[A])/100,race:WAe[A]});p.race.sort((A,b)=>b.score-A.score);let m=Array.from(await u.age.data()).map((A,b)=>[VAe[b],A]).sort((A,b)=>b[1]-A[1]),g=m[0][0];for(let A=1;A<m.length;A++)g+=m[A][1]*(m[A][0]-g);p.age=Math.round(10*g)/10,Object.keys(u).forEach(A=>ee(u[A])),mb[n]=p,NT=s,ET=ue(),l(p)}))}var rt={tf255:255,tf1:1,tf2:2,tf05:.5,tf127:127.5,rgb:[.2989,.587,.114]};function DT(){rt.tf255=Ce(255,"float32"),rt.tf1=Ce(1,"float32"),rt.tf2=Ce(2,"float32"),rt.tf05=Ce(.5,"float32"),rt.tf127=Ce(127.5,"float32"),rt.rgb=Ft([.2989,.587,.114],"float32")}var Os,P2=[],$T=0,PT=0,Ab=Number.MAX_SAFE_INTEGER;async function FT(e){return pe.initial&&(Os=null),Os?e.debug&&le("cached model:",Os.modelUrl):Os=await Ge(e.face.ssrnet.modelPathAge),Os}async function xb(e,t,n,s){var o,i,l,u;if(!Os)return{age:0};let r=Ab<(((o=t.face.ssrnet)==null?void 0:o.skipFrames)||0),a=(((i=t.face.ssrnet)==null?void 0:i.skipTime)||0)>ue()-PT;return t.skipAllowed&&r&&a&&$T===s&&((l=P2[n])==null?void 0:l.age)&&((u=P2[n])==null?void 0:u.age)>0?(Ab++,P2[n]):(Ab=0,new Promise(async c=>{if(!(Os!=null&&Os.inputs)||!Os.inputs[0]||!Os.inputs[0].shape)return;let p={};p.resize=Se.resizeBilinear(e,[Os.inputs[0].shape[2],Os.inputs[0].shape[1]],!1),p.enhance=L(p.resize,rt.tf255);let d={age:0};if(t.face.ssrnet.enabled&&(p.age=Os.execute(p.enhance)),p.age){let h=await p.age.data();d.age=Math.trunc(10*h[0])/10}Object.keys(p).forEach(h=>ee(p[h])),P2[n]=d,$T=s,PT=ue(),c(d)}))}var gr,F2=[],MT=0,zT=0,bb=Number.MAX_SAFE_INTEGER,vb=[.2989,.587,.114];async function LT(e){return pe.initial&&(gr=null),gr?e.debug&&le("cached model:",gr.modelUrl):gr=await Ge(e.face.ssrnet.modelPathGender),gr}async function wb(e,t,n,s){var o,i,l,u;if(!gr)return{gender:"unknown",genderScore:0};let r=bb<(((o=t.face.ssrnet)==null?void 0:o.skipFrames)||0),a=(((i=t.face.ssrnet)==null?void 0:i.skipTime)||0)>ue()-zT;return t.skipAllowed&&r&&a&&MT===s&&((l=F2[n])==null?void 0:l.gender)&&((u=F2[n])==null?void 0:u.genderScore)>0?(bb++,F2[n]):(bb=0,new Promise(async c=>{if(!(gr!=null&&gr.inputs[0].shape))return;let p={};p.resize=Se.resizeBilinear(e,[gr.inputs[0].shape[2],gr.inputs[0].shape[1]],!1),p.enhance=Y(()=>{let[f,m,g]=Zt(p.resize,3,3),y=L(f,vb[0]),x=L(m,vb[1]),A=L(g,vb[2]),b=m0([y,x,A]);return L(fe(b,rt.tf05),2)});let d={gender:"unknown",genderScore:0};t.face.ssrnet.enabled&&(p.gender=gr.execute(p.enhance));let h=await p.gender.data();d.gender=h[0]>h[1]?"female":"male",d.genderScore=h[0]>h[1]?Math.trunc(100*h[0])/100:Math.trunc(100*h[1])/100,Object.keys(p).forEach(f=>ee(p[f])),F2[n]=d,MT=s,zT=ue(),c(d)}))}var Ln,O2=[],kb=Number.MAX_SAFE_INTEGER,WT=0,VT=0;async function UT(e){var t;return pe.initial&&(Ln=null),Ln?e.debug&&le("cached model:",Ln.modelUrl):Ln=await Ge((t=e.face.antispoof)==null?void 0:t.modelPath),Ln}async function Ib(e,t,n,s){var o,i;if(!Ln)return 0;let r=(((o=t.face.antispoof)==null?void 0:o.skipTime)||0)>ue()-VT,a=kb<(((i=t.face.antispoof)==null?void 0:i.skipFrames)||0);return t.skipAllowed&&r&&a&&WT===s&&O2[n]?(kb++,O2[n]):(kb=0,new Promise(async l=>{let u=Se.resizeBilinear(e,[Ln!=null&&Ln.inputs[0].shape?Ln.inputs[0].shape[2]:0,Ln!=null&&Ln.inputs[0].shape?Ln.inputs[0].shape[1]:0],!1),c=Ln==null?void 0:Ln.execute(u),p=(await c.data())[0];O2[n]=Math.round(100*p)/100,WT=s,VT=ue(),ee([u,c]),l(O2[n])}))}var yr={silhouette:[10,338,297,332,284,251,389,356,454,323,361,288,397,365,379,378,400,377,152,148,176,149,150,136,172,58,132,93,234,127,162,21,54,103,67,109],lipsUpperOuter:[185,40,39,37,0,267,269,270,409],lipsLowerOuter:[61,146,91,181,84,17,314,405,321,375,291],lipsUpperInner:[191,80,81,82,13,312,311,310,415],lipsLowerInner:[78,95,88,178,87,14,317,402,318,324,308],lipsLowerSemiOuter:[76,77,90,180,85,16,315,404,320,307,306],lipsUpperSemiOuter:[184,74,73,72,11,302,303,304,408],lipsLowerSemiInner:[62,96,89,179,86,15,316,403,319,325,292],lipsUpperSemiInner:[183,42,41,38,12,268,271,272,407],rightEyeUpper0:[246,161,160,159,158,157,173],rightEyeLower0:[33,7,163,144,145,153,154,155,133],rightEyeUpper1:[247,30,29,27,28,56,190],rightEyeLower1:[130,25,110,24,23,22,26,112,243],rightEyeUpper2:[113,225,224,223,222,221,189],rightEyeLower2:[226,31,228,229,230,231,232,233,244],rightEyeLower3:[143,111,117,118,119,120,121,128,245],rightEyebrowUpper:[156,70,63,105,66,107,55,193],rightEyebrowLower:[35,124,46,53,52,65],rightEyeIris:[473,474,475,476,477],leftEyeUpper0:[466,388,387,386,385,384,398],leftEyeLower0:[263,249,390,373,374,380,381,382,362],leftEyeUpper1:[467,260,259,257,258,286,414],leftEyeLower1:[359,255,339,254,253,252,256,341,463],leftEyeUpper2:[342,445,444,443,442,441,413],leftEyeLower2:[446,261,448,449,450,451,452,453,464],leftEyeLower3:[372,340,346,347,348,349,350,357,465],leftEyebrowUpper:[383,300,293,334,296,336,285,417],leftEyebrowLower:[265,353,276,283,282,295],leftEyeIris:[468,469,470,471,472],midwayBetweenEyes:[168],noseTip:[1],noseBottom:[2],noseRightCorner:[98],noseLeftCorner:[327],rightCheek:[205],leftCheek:[425]},Sb={count:468,mouth:13,symmetryLine:[13,yr.midwayBetweenEyes[0]]},iu={leftEye:0,rightEye:1,nose:2,mouth:3,leftEar:4,rightEar:5,symmetryLine:[3,2]},Cb=[{key:"EyeUpper0",indices:[9,10,11,12,13,14,15]},{key:"EyeUpper1",indices:[25,26,27,28,29,30,31]},{key:"EyeUpper2",indices:[41,42,43,44,45,46,47]},{key:"EyeLower0",indices:[0,1,2,3,4,5,6,7,8]},{key:"EyeLower1",indices:[16,17,18,19,20,21,22,23,24]},{key:"EyeLower2",indices:[32,33,34,35,36,37,38,39,40]},{key:"EyeLower3",indices:[54,55,56,57,58,59,60,61,62]},{key:"EyebrowUpper",indices:[63,64,65,66,67,68,69,70]},{key:"EyebrowLower",indices:[48,49,50,51,52,53]}],Hh=[[.499976992607117,.652534008026123],[.500025987625122,.547487020492554],[.499974012374878,.602371990680695],[.482113003730774,.471979022026062],[.500150978565216,.527155995368958],[.499909996986389,.498252987861633],[.499523013830185,.40106201171875],[.289712011814117,.380764007568359],[.499954998493195,.312398016452789],[.499987006187439,.269918978214264],[.500023007392883,.107050001621246],[.500023007392883,.666234016418457],[.5000159740448,.679224014282227],[.500023007392883,.692348003387451],[.499976992607117,.695277988910675],[.499976992607117,.70593398809433],[.499976992607117,.719385027885437],[.499976992607117,.737019002437592],[.499967992305756,.781370997428894],[.499816000461578,.562981009483337],[.473773002624512,.573909997940063],[.104906998574734,.254140973091125],[.365929991006851,.409575998783112],[.338757991790771,.41302502155304],[.311120003461838,.409460008144379],[.274657994508743,.389131009578705],[.393361985683441,.403706014156342],[.345234006643295,.344011008739471],[.370094001293182,.346076011657715],[.319321990013123,.347265005111694],[.297903001308441,.353591024875641],[.24779200553894,.410809993743896],[.396889001131058,.842755019664764],[.280097991228104,.375599980354309],[.106310002505779,.399955987930298],[.2099249958992,.391353011131287],[.355807989835739,.534406006336212],[.471751004457474,.65040397644043],[.474155008792877,.680191993713379],[.439785003662109,.657229006290436],[.414617002010345,.66654098033905],[.450374007225037,.680860996246338],[.428770989179611,.682690978050232],[.374971002340317,.727805018424988],[.486716985702515,.547628998756409],[.485300987958908,.527395009994507],[.257764995098114,.314490020275116],[.401223003864288,.455172002315521],[.429818987846375,.548614978790283],[.421351999044418,.533740997314453],[.276895999908447,.532056987285614],[.483370006084442,.499586999416351],[.33721199631691,.282882988452911],[.296391993761063,.293242990970612],[.169294998049736,.193813979625702],[.447580009698868,.302609980106354],[.392390012741089,.353887975215912],[.354490011930466,.696784019470215],[.067304998636246,.730105042457581],[.442739009857178,.572826027870178],[.457098007202148,.584792017936707],[.381974011659622,.694710969924927],[.392388999462128,.694203019142151],[.277076005935669,.271932005882263],[.422551989555359,.563233017921448],[.385919004678726,.281364023685455],[.383103013038635,.255840003490448],[.331431001424789,.119714021682739],[.229923993349075,.232002973556519],[.364500999450684,.189113974571228],[.229622006416321,.299540996551514],[.173287004232407,.278747975826263],[.472878992557526,.666198015213013],[.446828007698059,.668527007102966],[.422762006521225,.673889994621277],[.445307999849319,.580065965652466],[.388103008270264,.693961024284363],[.403039008378983,.706539988517761],[.403629004955292,.693953037261963],[.460041999816895,.557139039039612],[.431158006191254,.692366003990173],[.452181994915009,.692366003990173],[.475387006998062,.692366003990173],[.465828001499176,.779190003871918],[.472328990697861,.736225962638855],[.473087012767792,.717857003211975],[.473122000694275,.704625964164734],[.473033010959625,.695277988910675],[.427942007780075,.695277988910675],[.426479011774063,.703539967536926],[.423162013292313,.711845993995667],[.4183090031147,.720062971115112],[.390094995498657,.639572978019714],[.013953999616206,.560034036636353],[.499913990497589,.58014702796936],[.413199990987778,.69539999961853],[.409626007080078,.701822996139526],[.468080013990402,.601534962654114],[.422728985548019,.585985004901886],[.463079988956451,.593783974647522],[.37211999297142,.47341400384903],[.334562003612518,.496073007583618],[.411671012639999,.546965003013611],[.242175996303558,.14767599105835],[.290776997804642,.201445996761322],[.327338010072708,.256527006626129],[.399509996175766,.748921036720276],[.441727995872498,.261676013469696],[.429764986038208,.187834024429321],[.412198007106781,.108901023864746],[.288955003023148,.398952007293701],[.218936994671822,.435410976409912],[.41278201341629,.398970007896423],[.257135003805161,.355440020561218],[.427684992551804,.437960982322693],[.448339998722076,.536936044692993],[.178560003638268,.45755398273468],[.247308000922203,.457193970680237],[.286267012357712,.467674970626831],[.332827985286713,.460712015628815],[.368755996227264,.447206974029541],[.398963987827301,.432654976844788],[.476410001516342,.405806005001068],[.189241006970406,.523923993110657],[.228962004184723,.348950982093811],[.490725994110107,.562400996685028],[.404670000076294,.485132992267609],[.019469000399113,.401564002037048],[.426243007183075,.420431017875671],[.396993011236191,.548797011375427],[.266469985246658,.376977026462555],[.439121007919312,.51895797252655],[.032313998788595,.644356966018677],[.419054001569748,.387154996395111],[.462783008813858,.505746960639954],[.238978996872902,.779744982719421],[.198220998048782,.831938028335571],[.107550002634525,.540755033493042],[.183610007166862,.740257024765015],[.134409993886948,.333683013916016],[.385764002799988,.883153975009918],[.490967005491257,.579378008842468],[.382384985685349,.508572995662689],[.174399003386497,.397670984268188],[.318785011768341,.39623498916626],[.343364000320435,.400596976280212],[.396100014448166,.710216999053955],[.187885001301765,.588537991046906],[.430987000465393,.944064974784851],[.318993002176285,.898285031318665],[.266247987747192,.869701027870178],[.500023007392883,.190576016902924],[.499976992607117,.954452991485596],[.366169989109039,.398822009563446],[.393207013607025,.39553701877594],[.410373002290726,.391080021858215],[.194993004202843,.342101991176605],[.388664990663528,.362284004688263],[.365961998701096,.355970978736877],[.343364000320435,.355356991291046],[.318785011768341,.35834002494812],[.301414996385574,.363156020641327],[.058132998645306,.319076001644135],[.301414996385574,.387449026107788],[.499987989664078,.618434011936188],[.415838003158569,.624195992946625],[.445681989192963,.566076993942261],[.465844005346298,.620640993118286],[.49992299079895,.351523995399475],[.288718998432159,.819945991039276],[.335278987884521,.852819979190826],[.440512001514435,.902418971061707],[.128294005990028,.791940987110138],[.408771991729736,.373893976211548],[.455606997013092,.451801002025604],[.499877005815506,.908990025520325],[.375436991453171,.924192011356354],[.11421000212431,.615022003650665],[.448662012815475,.695277988910675],[.4480200111866,.704632043838501],[.447111994028091,.715808033943176],[.444831997156143,.730794012546539],[.430011987686157,.766808986663818],[.406787008047104,.685672998428345],[.400738000869751,.681069016456604],[.392399996519089,.677703022956848],[.367855995893478,.663918972015381],[.247923001646996,.601333022117615],[.452769994735718,.420849978923798],[.43639200925827,.359887003898621],[.416164010763168,.368713974952698],[.413385987281799,.692366003990173],[.228018000721931,.683571994304657],[.468268007040024,.352671027183533],[.411361992359161,.804327011108398],[.499989002943039,.469825029373169],[.479153990745544,.442654013633728],[.499974012374878,.439637005329132],[.432112008333206,.493588984012604],[.499886006116867,.866917014122009],[.49991300702095,.821729004383087],[.456548988819122,.819200992584229],[.344549000263214,.745438992977142],[.37890899181366,.574010014533997],[.374292999505997,.780184984207153],[.319687992334366,.570737957954407],[.357154995203018,.604269981384277],[.295284003019333,.621580958366394],[.447750002145767,.862477004528046],[.410986006259918,.508723020553589],[.31395098567009,.775308012962341],[.354128003120422,.812552988529205],[.324548006057739,.703992962837219],[.189096003770828,.646299958229065],[.279776990413666,.71465802192688],[.1338230073452,.682700991630554],[.336768001317978,.644733011722565],[.429883986711502,.466521978378296],[.455527991056442,.548622965812683],[.437114000320435,.558896005153656],[.467287987470627,.529924988746643],[.414712011814117,.335219979286194],[.37704598903656,.322777986526489],[.344107985496521,.320150971412659],[.312875986099243,.32233202457428],[.283526003360748,.333190023899078],[.241245999932289,.382785975933075],[.102986000478268,.468762993812561],[.267612010240555,.424560010433197],[.297879010438919,.433175981044769],[.333433985710144,.433878004550934],[.366427004337311,.426115989685059],[.396012008190155,.416696012020111],[.420121014118195,.41022801399231],[.007561000064015,.480777025222778],[.432949006557465,.569517970085144],[.458638995885849,.479089021682739],[.473466008901596,.545744001865387],[.476087987422943,.563830018043518],[.468472003936768,.555056989192963],[.433990985155106,.582361996173859],[.483518004417419,.562983989715576],[.482482999563217,.57784903049469],[.42645001411438,.389798998832703],[.438998997211456,.39649498462677],[.450067013502121,.400434017181396],[.289712011814117,.368252992630005],[.276670008897781,.363372981548309],[.517862021923065,.471948027610779],[.710287988185883,.380764007568359],[.526226997375488,.573909997940063],[.895093023777008,.254140973091125],[.634069979190826,.409575998783112],[.661242008209229,.41302502155304],[.688880026340485,.409460008144379],[.725341975688934,.389131009578705],[.606630027294159,.40370500087738],[.654766023159027,.344011008739471],[.629905998706818,.346076011657715],[.680678009986877,.347265005111694],[.702096998691559,.353591024875641],[.75221198797226,.410804986953735],[.602918028831482,.842862963676453],[.719901978969574,.375599980354309],[.893692970275879,.399959981441498],[.790081977844238,.391354024410248],[.643998026847839,.534487962722778],[.528249025344849,.65040397644043],[.525849997997284,.680191040039062],[.560214996337891,.657229006290436],[.585384011268616,.66654098033905],[.549625992774963,.680860996246338],[.57122802734375,.682691991329193],[.624852001667023,.72809898853302],[.513050019741058,.547281980514526],[.51509702205658,.527251958847046],[.742246985435486,.314507007598877],[.598631024360657,.454979002475739],[.570338010787964,.548575043678284],[.578631997108459,.533622980117798],[.723087012767792,.532054007053375],[.516445994377136,.499638974666595],[.662801027297974,.282917976379395],[.70362401008606,.293271005153656],[.830704987049103,.193813979625702],[.552385985851288,.302568018436432],[.607609987258911,.353887975215912],[.645429015159607,.696707010269165],[.932694971561432,.730105042457581],[.557260990142822,.572826027870178],[.542901992797852,.584792017936707],[.6180260181427,.694710969924927],[.607590973377228,.694203019142151],[.722943007946014,.271963000297546],[.577413976192474,.563166975975037],[.614082992076874,.281386971473694],[.616907000541687,.255886018276215],[.668509006500244,.119913995265961],[.770092010498047,.232020974159241],[.635536015033722,.189248979091644],[.77039098739624,.299556016921997],[.826722025871277,.278755009174347],[.527121007442474,.666198015213013],[.553171992301941,.668527007102966],[.577238023281097,.673889994621277],[.554691970348358,.580065965652466],[.611896991729736,.693961024284363],[.59696102142334,.706539988517761],[.596370995044708,.693953037261963],[.539958000183105,.557139039039612],[.568841993808746,.692366003990173],[.547818005084991,.692366003990173],[.52461302280426,.692366003990173],[.534089982509613,.779141008853912],[.527670979499817,.736225962638855],[.526912987232208,.717857003211975],[.526877999305725,.704625964164734],[.526966989040375,.695277988910675],[.572058022022247,.695277988910675],[.573521018028259,.703539967536926],[.57683801651001,.711845993995667],[.581691026687622,.720062971115112],[.609944999217987,.639909982681274],[.986046016216278,.560034036636353],[.5867999792099,.69539999961853],[.590372025966644,.701822996139526],[.531915009021759,.601536989212036],[.577268004417419,.585934996604919],[.536915004253387,.593786001205444],[.627542972564697,.473352015018463],[.665585994720459,.495950996875763],[.588353991508484,.546862006187439],[.757824003696442,.14767599105835],[.709249973297119,.201507985591888],[.672684013843536,.256581008434296],[.600408971309662,.74900496006012],[.55826598405838,.261672019958496],[.570303976535797,.187870979309082],[.588165998458862,.109044015407562],[.711045026779175,.398952007293701],[.781069993972778,.435405015945435],[.587247014045715,.398931980133057],[.742869973182678,.355445981025696],[.572156012058258,.437651991844177],[.55186802148819,.536570012569427],[.821442008018494,.457556009292603],[.752701997756958,.457181990146637],[.71375697851181,.467626988887787],[.66711300611496,.460672974586487],[.631101012229919,.447153985500336],[.6008620262146,.432473003864288],[.523481011390686,.405627012252808],[.810747981071472,.523926019668579],[.771045982837677,.348959028720856],[.509127020835876,.562718033790588],[.595292985439301,.485023975372314],[.980530977249146,.401564002037048],[.573499977588654,.420000016689301],[.602994978427887,.548687994480133],[.733529984951019,.376977026462555],[.560611009597778,.519016981124878],[.967685997486115,.644356966018677],[.580985009670258,.387160003185272],[.537728011608124,.505385041236877],[.760966002941132,.779752969741821],[.801778972148895,.831938028335571],[.892440974712372,.54076099395752],[.816350996494293,.740260004997253],[.865594983100891,.333687007427216],[.614073991775513,.883246004581451],[.508952975273132,.579437971115112],[.617941975593567,.508316040039062],[.825608015060425,.397674977779388],[.681214988231659,.39623498916626],[.656635999679565,.400596976280212],[.603900015354156,.710216999053955],[.81208598613739,.588539004325867],[.56801301240921,.944564998149872],[.681007981300354,.898285031318665],[.733752012252808,.869701027870178],[.633830010890961,.398822009563446],[.606792986392975,.39553701877594],[.589659988880157,.391062021255493],[.805015981197357,.342108011245728],[.611334979534149,.362284004688263],[.634037971496582,.355970978736877],[.656635999679565,.355356991291046],[.681214988231659,.35834002494812],[.698584973812103,.363156020641327],[.941866993904114,.319076001644135],[.698584973812103,.387449026107788],[.584177017211914,.624107003211975],[.554318010807037,.566076993942261],[.534153997898102,.62064003944397],[.711217999458313,.819975018501282],[.664629995822906,.852871000766754],[.559099972248077,.902631998062134],[.871706008911133,.791940987110138],[.591234028339386,.373893976211548],[.544341027736664,.451583981513977],[.624562978744507,.924192011356354],[.88577002286911,.615028977394104],[.551338016986847,.695277988910675],[.551980018615723,.704632043838501],[.552887976169586,.715808033943176],[.555167973041534,.730794012546539],[.569944024085999,.767035007476807],[.593203008174896,.685675978660583],[.599261999130249,.681069016456604],[.607599973678589,.677703022956848],[.631937980651855,.663500010967255],[.752032995223999,.601315021514893],[.547226011753082,.420395016670227],[.563543975353241,.359827995300293],[.583841025829315,.368713974952698],[.586614012718201,.692366003990173],[.771915018558502,.683578014373779],[.531597018241882,.352482974529266],[.588370978832245,.804440975189209],[.52079701423645,.442565023899078],[.567984998226166,.493479013442993],[.543282985687256,.819254994392395],[.655317008495331,.745514988899231],[.621008992195129,.574018001556396],[.625559985637665,.78031200170517],[.680198013782501,.570719003677368],[.64276397228241,.604337990283966],[.704662978649139,.621529996395111],[.552012026309967,.862591981887817],[.589071989059448,.508637011051178],[.685944974422455,.775357007980347],[.645735025405884,.812640011310577],[.675342977046967,.703978002071381],[.810858011245728,.646304965019226],[.72012197971344,.714666962623596],[.866151988506317,.682704985141754],[.663187026977539,.644596993923187],[.570082008838654,.466325998306274],[.544561982154846,.548375964164734],[.562758982181549,.558784961700439],[.531987011432648,.530140042304993],[.585271000862122,.335177004337311],[.622952997684479,.32277899980545],[.655896008014679,.320163011550903],[.687132000923157,.322345972061157],[.716481983661652,.333200991153717],[.758756995201111,.382786989212036],[.897013008594513,.468769013881683],[.732392013072968,.424547016620636],[.70211398601532,.433162987232208],[.66652500629425,.433866024017334],[.633504986763,.426087975502014],[.603875994682312,.416586995124817],[.579657971858978,.409945011138916],[.992439985275269,.480777025222778],[.567192018032074,.569419980049133],[.54136598110199,.478899002075195],[.526564002037048,.546118021011353],[.523913025856018,.563830018043518],[.531529009342194,.555056989192963],[.566035985946655,.582329034805298],[.51631098985672,.563053965568542],[.5174720287323,.577877044677734],[.573594987392426,.389806985855103],[.560697972774506,.395331978797913],[.549755990505219,.399751007556915],[.710287988185883,.368252992630005],[.723330020904541,.363372981548309]],lu=[127,34,139,11,0,37,232,231,120,72,37,39,128,121,47,232,121,128,104,69,67,175,171,148,157,154,155,118,50,101,73,39,40,9,151,108,48,115,131,194,204,211,74,40,185,80,42,183,40,92,186,230,229,118,202,212,214,83,18,17,76,61,146,160,29,30,56,157,173,106,204,194,135,214,192,203,165,98,21,71,68,51,45,4,144,24,23,77,146,91,205,50,187,201,200,18,91,106,182,90,91,181,85,84,17,206,203,36,148,171,140,92,40,39,193,189,244,159,158,28,247,246,161,236,3,196,54,68,104,193,168,8,117,228,31,189,193,55,98,97,99,126,47,100,166,79,218,155,154,26,209,49,131,135,136,150,47,126,217,223,52,53,45,51,134,211,170,140,67,69,108,43,106,91,230,119,120,226,130,247,63,53,52,238,20,242,46,70,156,78,62,96,46,53,63,143,34,227,173,155,133,123,117,111,44,125,19,236,134,51,216,206,205,154,153,22,39,37,167,200,201,208,36,142,100,57,212,202,20,60,99,28,158,157,35,226,113,160,159,27,204,202,210,113,225,46,43,202,204,62,76,77,137,123,116,41,38,72,203,129,142,64,98,240,49,102,64,41,73,74,212,216,207,42,74,184,169,170,211,170,149,176,105,66,69,122,6,168,123,147,187,96,77,90,65,55,107,89,90,180,101,100,120,63,105,104,93,137,227,15,86,85,129,102,49,14,87,86,55,8,9,100,47,121,145,23,22,88,89,179,6,122,196,88,95,96,138,172,136,215,58,172,115,48,219,42,80,81,195,3,51,43,146,61,171,175,199,81,82,38,53,46,225,144,163,110,246,33,7,52,65,66,229,228,117,34,127,234,107,108,69,109,108,151,48,64,235,62,78,191,129,209,126,111,35,143,163,161,246,117,123,50,222,65,52,19,125,141,221,55,65,3,195,197,25,7,33,220,237,44,70,71,139,122,193,245,247,130,33,71,21,162,153,158,159,170,169,150,188,174,196,216,186,92,144,160,161,2,97,167,141,125,241,164,167,37,72,38,12,145,159,160,38,82,13,63,68,71,226,35,111,158,153,154,101,50,205,206,92,165,209,198,217,165,167,97,220,115,218,133,112,243,239,238,241,214,135,169,190,173,133,171,208,32,125,44,237,86,87,178,85,86,179,84,85,180,83,84,181,201,83,182,137,93,132,76,62,183,61,76,184,57,61,185,212,57,186,214,207,187,34,143,156,79,239,237,123,137,177,44,1,4,201,194,32,64,102,129,213,215,138,59,166,219,242,99,97,2,94,141,75,59,235,24,110,228,25,130,226,23,24,229,22,23,230,26,22,231,112,26,232,189,190,243,221,56,190,28,56,221,27,28,222,29,27,223,30,29,224,247,30,225,238,79,20,166,59,75,60,75,240,147,177,215,20,79,166,187,147,213,112,233,244,233,128,245,128,114,188,114,217,174,131,115,220,217,198,236,198,131,134,177,132,58,143,35,124,110,163,7,228,110,25,356,389,368,11,302,267,452,350,349,302,303,269,357,343,277,452,453,357,333,332,297,175,152,377,384,398,382,347,348,330,303,304,270,9,336,337,278,279,360,418,262,431,304,408,409,310,415,407,270,409,410,450,348,347,422,430,434,313,314,17,306,307,375,387,388,260,286,414,398,335,406,418,364,367,416,423,358,327,251,284,298,281,5,4,373,374,253,307,320,321,425,427,411,421,313,18,321,405,406,320,404,405,315,16,17,426,425,266,377,400,369,322,391,269,417,465,464,386,257,258,466,260,388,456,399,419,284,332,333,417,285,8,346,340,261,413,441,285,327,460,328,355,371,329,392,439,438,382,341,256,429,420,360,364,394,379,277,343,437,443,444,283,275,440,363,431,262,369,297,338,337,273,375,321,450,451,349,446,342,467,293,334,282,458,461,462,276,353,383,308,324,325,276,300,293,372,345,447,382,398,362,352,345,340,274,1,19,456,248,281,436,427,425,381,256,252,269,391,393,200,199,428,266,330,329,287,273,422,250,462,328,258,286,384,265,353,342,387,259,257,424,431,430,342,353,276,273,335,424,292,325,307,366,447,345,271,303,302,423,266,371,294,455,460,279,278,294,271,272,304,432,434,427,272,407,408,394,430,431,395,369,400,334,333,299,351,417,168,352,280,411,325,319,320,295,296,336,319,403,404,330,348,349,293,298,333,323,454,447,15,16,315,358,429,279,14,15,316,285,336,9,329,349,350,374,380,252,318,402,403,6,197,419,318,319,325,367,364,365,435,367,397,344,438,439,272,271,311,195,5,281,273,287,291,396,428,199,311,271,268,283,444,445,373,254,339,263,466,249,282,334,296,449,347,346,264,447,454,336,296,299,338,10,151,278,439,455,292,407,415,358,371,355,340,345,372,390,249,466,346,347,280,442,443,282,19,94,370,441,442,295,248,419,197,263,255,359,440,275,274,300,383,368,351,412,465,263,467,466,301,368,389,380,374,386,395,378,379,412,351,419,436,426,322,373,390,388,2,164,393,370,462,461,164,0,267,302,11,12,374,373,387,268,12,13,293,300,301,446,261,340,385,384,381,330,266,425,426,423,391,429,355,437,391,327,326,440,457,438,341,382,362,459,457,461,434,430,394,414,463,362,396,369,262,354,461,457,316,403,402,315,404,403,314,405,404,313,406,405,421,418,406,366,401,361,306,408,407,291,409,408,287,410,409,432,436,410,434,416,411,264,368,383,309,438,457,352,376,401,274,275,4,421,428,262,294,327,358,433,416,367,289,455,439,462,370,326,2,326,370,305,460,455,254,449,448,255,261,446,253,450,449,252,451,450,256,452,451,341,453,452,413,464,463,441,413,414,258,442,441,257,443,442,259,444,443,260,445,444,467,342,445,459,458,250,289,392,290,290,328,460,376,433,435,250,290,392,411,416,433,341,463,464,453,464,465,357,465,412,343,412,399,360,363,440,437,399,456,420,456,363,401,435,288,372,383,353,339,255,249,448,261,255,133,243,190,133,155,112,33,246,247,33,130,25,398,384,286,362,398,414,362,463,341,263,359,467,263,249,255,466,467,260,75,60,166,238,239,79,162,127,139,72,11,37,121,232,120,73,72,39,114,128,47,233,232,128,103,104,67,152,175,148,173,157,155,119,118,101,74,73,40,107,9,108,49,48,131,32,194,211,184,74,185,191,80,183,185,40,186,119,230,118,210,202,214,84,83,17,77,76,146,161,160,30,190,56,173,182,106,194,138,135,192,129,203,98,54,21,68,5,51,4,145,144,23,90,77,91,207,205,187,83,201,18,181,91,182,180,90,181,16,85,17,205,206,36,176,148,140,165,92,39,245,193,244,27,159,28,30,247,161,174,236,196,103,54,104,55,193,8,111,117,31,221,189,55,240,98,99,142,126,100,219,166,218,112,155,26,198,209,131,169,135,150,114,47,217,224,223,53,220,45,134,32,211,140,109,67,108,146,43,91,231,230,120,113,226,247,105,63,52,241,238,242,124,46,156,95,78,96,70,46,63,116,143,227,116,123,111,1,44,19,3,236,51,207,216,205,26,154,22,165,39,167,199,200,208,101,36,100,43,57,202,242,20,99,56,28,157,124,35,113,29,160,27,211,204,210,124,113,46,106,43,204,96,62,77,227,137,116,73,41,72,36,203,142,235,64,240,48,49,64,42,41,74,214,212,207,183,42,184,210,169,211,140,170,176,104,105,69,193,122,168,50,123,187,89,96,90,66,65,107,179,89,180,119,101,120,68,63,104,234,93,227,16,15,85,209,129,49,15,14,86,107,55,9,120,100,121,153,145,22,178,88,179,197,6,196,89,88,96,135,138,136,138,215,172,218,115,219,41,42,81,5,195,51,57,43,61,208,171,199,41,81,38,224,53,225,24,144,110,105,52,66,118,229,117,227,34,234,66,107,69,10,109,151,219,48,235,183,62,191,142,129,126,116,111,143,7,163,246,118,117,50,223,222,52,94,19,141,222,221,65,196,3,197,45,220,44,156,70,139,188,122,245,139,71,162,145,153,159,149,170,150,122,188,196,206,216,92,163,144,161,164,2,167,242,141,241,0,164,37,11,72,12,144,145,160,12,38,13,70,63,71,31,226,111,157,158,154,36,101,205,203,206,165,126,209,217,98,165,97,237,220,218,237,239,241,210,214,169,140,171,32,241,125,237,179,86,178,180,85,179,181,84,180,182,83,181,194,201,182,177,137,132,184,76,183,185,61,184,186,57,185,216,212,186,192,214,187,139,34,156,218,79,237,147,123,177,45,44,4,208,201,32,98,64,129,192,213,138,235,59,219,141,242,97,97,2,141,240,75,235,229,24,228,31,25,226,230,23,229,231,22,230,232,26,231,233,112,232,244,189,243,189,221,190,222,28,221,223,27,222,224,29,223,225,30,224,113,247,225,99,60,240,213,147,215,60,20,166,192,187,213,243,112,244,244,233,245,245,128,188,188,114,174,134,131,220,174,217,236,236,198,134,215,177,58,156,143,124,25,110,7,31,228,25,264,356,368,0,11,267,451,452,349,267,302,269,350,357,277,350,452,357,299,333,297,396,175,377,381,384,382,280,347,330,269,303,270,151,9,337,344,278,360,424,418,431,270,304,409,272,310,407,322,270,410,449,450,347,432,422,434,18,313,17,291,306,375,259,387,260,424,335,418,434,364,416,391,423,327,301,251,298,275,281,4,254,373,253,375,307,321,280,425,411,200,421,18,335,321,406,321,320,405,314,315,17,423,426,266,396,377,369,270,322,269,413,417,464,385,386,258,248,456,419,298,284,333,168,417,8,448,346,261,417,413,285,326,327,328,277,355,329,309,392,438,381,382,256,279,429,360,365,364,379,355,277,437,282,443,283,281,275,363,395,431,369,299,297,337,335,273,321,348,450,349,359,446,467,283,293,282,250,458,462,300,276,383,292,308,325,283,276,293,264,372,447,346,352,340,354,274,19,363,456,281,426,436,425,380,381,252,267,269,393,421,200,428,371,266,329,432,287,422,290,250,328,385,258,384,446,265,342,386,387,257,422,424,430,445,342,276,422,273,424,306,292,307,352,366,345,268,271,302,358,423,371,327,294,460,331,279,294,303,271,304,436,432,427,304,272,408,395,394,431,378,395,400,296,334,299,6,351,168,376,352,411,307,325,320,285,295,336,320,319,404,329,330,349,334,293,333,366,323,447,316,15,315,331,358,279,317,14,316,8,285,9,277,329,350,253,374,252,319,318,403,351,6,419,324,318,325,397,367,365,288,435,397,278,344,439,310,272,311,248,195,281,375,273,291,175,396,199,312,311,268,276,283,445,390,373,339,295,282,296,448,449,346,356,264,454,337,336,299,337,338,151,294,278,455,308,292,415,429,358,355,265,340,372,388,390,466,352,346,280,295,442,282,354,19,370,285,441,295,195,248,197,457,440,274,301,300,368,417,351,465,251,301,389,385,380,386,394,395,379,399,412,419,410,436,322,387,373,388,326,2,393,354,370,461,393,164,267,268,302,12,386,374,387,312,268,13,298,293,301,265,446,340,380,385,381,280,330,425,322,426,391,420,429,437,393,391,326,344,440,438,458,459,461,364,434,394,428,396,262,274,354,457,317,316,402,316,315,403,315,314,404,314,313,405,313,421,406,323,366,361,292,306,407,306,291,408,291,287,409,287,432,410,427,434,411,372,264,383,459,309,457,366,352,401,1,274,4,418,421,262,331,294,358,435,433,367,392,289,439,328,462,326,94,2,370,289,305,455,339,254,448,359,255,446,254,253,449,253,252,450,252,256,451,256,341,452,414,413,463,286,441,414,286,258,441,258,257,442,257,259,443,259,260,444,260,467,445,309,459,250,305,289,290,305,290,460,401,376,435,309,250,392,376,411,433,453,341,464,357,453,465,343,357,412,437,343,399,344,360,440,420,437,456,360,420,363,361,401,288,265,372,353,390,339,249,339,448,255];var GAe=[127,234,132,58,172,150,149,148,152,377,378,379,397,288,361,454,356,70,63,105,66,107,336,296,334,293,300,168,6,195,4,98,97,2,326,327,33,160,158,133,153,144,362,385,387,263,373,380,57,40,37,0,267,270,287,321,314,17,84,91,78,81,13,311,308,402,14,178],HAe=[33,133,362,263,1,62,308,159,145,386,374,6,102,331,2,13,14,70,105,107,336,334,300,54,10,284,50,280,234,454,58,288,152],jAe=[33,133,362,263,1,78,308],X8e=GAe.map(e=>Hh[e]),K8e=HAe.map(e=>Hh[e]),Z8e=jAe.map(e=>Hh[e]);function fi(e){let t=e.map(n=>n[0]);return t.push(e[e.length-1][1]),t}var qAe=[[61,146],[146,91],[91,181],[181,84],[84,17],[17,314],[314,405],[405,321],[321,375],[375,291],[61,185],[185,40],[40,39],[39,37],[37,0],[0,267],[267,269],[269,270],[270,409],[409,291],[78,95],[95,88],[88,178],[178,87],[87,14],[14,317],[317,402],[402,318],[318,324],[324,308],[78,191],[191,80],[80,81],[81,82],[82,13],[13,312],[312,311],[311,310],[310,415],[415,308]],XAe=[[263,249],[249,390],[390,373],[373,374],[374,380],[380,381],[381,382],[382,362],[263,466],[466,388],[388,387],[387,386],[386,385],[385,384],[384,398],[398,362]],KAe=[[276,283],[283,282],[282,295],[295,285],[300,293],[293,334],[334,296],[296,336]],ZAe=[[474,475],[475,476],[476,477],[477,474]],YAe=[[33,7],[7,163],[163,144],[144,145],[145,153],[153,154],[154,155],[155,133],[33,246],[246,161],[161,160],[160,159],[159,158],[158,157],[157,173],[173,133]],JAe=[[46,53],[53,52],[52,65],[65,55],[70,63],[63,105],[105,66],[66,107]],QAe=[[469,470],[470,471],[471,472],[472,469]],e5e=[[10,338],[338,297],[297,332],[332,284],[284,251],[251,389],[389,356],[356,454],[454,323],[323,361],[361,288],[288,397],[397,365],[365,379],[379,378],[378,400],[400,377],[377,152],[152,148],[148,176],[176,149],[149,150],[150,136],[136,172],[172,58],[58,132],[132,93],[93,234],[234,127],[127,162],[162,21],[21,54],[54,103],[103,67],[67,109],[109,10]],Y8e={lips:fi(qAe),leftEye:fi(XAe),leftEyebrow:fi(KAe),leftIris:fi(ZAe),rightEye:fi(YAe),rightEyebrow:fi(JAe),rightIris:fi(QAe),faceOval:fi(e5e)};var yd=e=>[Math.abs(e.endPoint[0]-e.startPoint[0]),Math.abs(e.endPoint[1]-e.startPoint[1])],M2=e=>[e.startPoint[0]+(e.endPoint[0]-e.startPoint[0])/2,e.startPoint[1]+(e.endPoint[1]-e.startPoint[1])/2,1],z2=(e,t)=>e?[Math.trunc(Math.max(0,e.startPoint[0])),Math.trunc(Math.max(0,e.startPoint[1])),Math.trunc(Math.min(t.shape[2]||0,e.endPoint[0])-Math.max(0,e.startPoint[0])),Math.trunc(Math.min(t.shape[1]||0,e.endPoint[1])-Math.max(0,e.startPoint[1]))]:[0,0,0,0],L2=(e,t)=>e?[e.startPoint[0]/(t.shape[2]||0),e.startPoint[1]/(t.shape[1]||0),(e.endPoint[0]-e.startPoint[0])/(t.shape[2]||0),(e.endPoint[1]-e.startPoint[1])/(t.shape[1]||0)]:[0,0,0,0],qT=(e,t)=>{let n=[e.startPoint[0]*t[0],e.startPoint[1]*t[1]],s=[e.endPoint[0]*t[0],e.endPoint[1]*t[1]];return{startPoint:n,endPoint:s,landmarks:e.landmarks,confidence:e.confidence}},Nb=(e,t,n)=>{let s=t.shape[1],r=t.shape[2],a=[e.startPoint[1]/s,e.startPoint[0]/r,e.endPoint[1]/s,e.endPoint[0]/r],o=Se.cropAndResize(t,[a],[0],n),i=he(o,rt.tf255);return ee(o),i},B2=(e,t)=>{let n=M2(e),s=yd(e),r=[t*s[0]/2,t*s[1]/2];return{startPoint:[n[0]-r[0],n[1]-r[1]],endPoint:[n[0]+r[0],n[1]+r[1]],landmarks:e.landmarks,confidence:e.confidence}},W2=e=>{let t=M2(e),n=yd(e),s=Math.max(...n)/2;return{startPoint:[Math.round(t[0]-s),Math.round(t[1]-s)],endPoint:[Math.round(t[0]+s),Math.round(t[1]+s)],landmarks:e.landmarks,confidence:e.confidence}},XT=e=>{let t=e.map(s=>s[0]),n=e.map(s=>s[1]);return{startPoint:[Math.min(...t),Math.min(...n)],endPoint:[Math.max(...t),Math.max(...n)],landmarks:e}},Eb=[[1,0,0],[0,1,0],[0,0,1]],t5e=e=>e-2*Math.PI*Math.floor((e+Math.PI)/(2*Math.PI)),n5e=(e,t)=>t5e(Math.PI/2-Math.atan2(-(t[1]-e[1]),t[0]-e[0]));var HT=(e,t)=>[[1,0,e],[0,1,t],[0,0,1]],uu=(e,t)=>{let n=0;for(let s=0;s<e.length;s++)n+=e[s]*t[s];return n},s5e=(e,t)=>{let n=[];for(let s=0;s<e.length;s++)n.push(e[s][t]);return n},jT=(e,t)=>{let n=[],s=e.length;for(let r=0;r<s;r++){n.push([]);for(let a=0;a<s;a++)n[r].push(uu(e[r],s5e(t,a)))}return n},KT=(e,t)=>{let n=Math.cos(e),s=Math.sin(e),r=[[n,-s,0],[s,n,0],[0,0,1]],a=HT(t[0],t[1]),o=jT(a,r),i=HT(-t[0],-t[1]);return jT(o,i)},r5e=e=>{let t=[[e[0][0],e[1][0]],[e[0][1],e[1][1]]],n=[e[0][2],e[1][2]],s=[-uu(t[0],n),-uu(t[1],n)];return[t[0].concat(s[0]),t[1].concat(s[1]),[0,0,1]]},a5e=(e,t)=>[uu(e,t[0]),uu(e,t[1])];function ZT(e){let t=e===192?{strides:[4],anchors:[1]}:{strides:[e/16,e/8],anchors:[2,6]},n=[];for(let s=0;s<t.strides.length;s++){let r=t.strides[s],a=Math.floor((e+r-1)/r),o=Math.floor((e+r-1)/r),i=t.anchors[s];for(let l=0;l<a;l++){let u=r*(l+.5);for(let c=0;c<o;c++){let p=r*(c+.5);for(let d=0;d<i;d++)n.push([p,u])}}}return n}function YT(e,t,n,s,r){let a=yd(t),o=e.map(h=>[a[0]/r*(h[0]-r/2),a[1]/r*(h[1]-r/2),h[2]||0]),i=n&&n!==0&&Math.abs(n)>.2,l=i?KT(n,[0,0]):Eb,u=i?o.map(h=>[...a5e(h,l),h[2]]):o,c=i?r5e(s):Eb,p=M2(t),d=[uu(p,c[0]),uu(p,c[1])];return u.map(h=>[Math.trunc(h[0]+d[0]),Math.trunc(h[1]+d[1]),Math.trunc(h[2]||0)])}function JT(e,t,n,s){let r=t.landmarks.length>=Sb.count?Sb.symmetryLine:iu.symmetryLine,a=0,o=Eb,i;if(e&&pe.kernels.includes("rotatewithoffset"))if(a=n5e(t.landmarks[r[0]],t.landmarks[r[1]]),a&&a!==0&&Math.abs(a)>.2){let u=M2(t),c=[u[0]/n.shape[2],u[1]/n.shape[1]],p=Se.rotateWithOffset(n,a,0,c);o=KT(-a,u),i=Nb(t,p,[s,s]),ee(p)}else i=Nb(t,n,[s,s]);else i=Nb(t,n,[s,s]);return[a,o,i]}var o5e=e=>{let t=e.map(s=>s[0]),n=e.map(s=>s[1]);return[Math.min(...t)+(Math.max(...t)-Math.min(...t))/2,Math.min(...n)+(Math.max(...n)-Math.min(...n))/2]},QT=(e,t)=>{let n=o5e(e),s=yd(t);return{startPoint:[n[0]-s[0]/2,n[1]-s[1]/2],endPoint:[n[0]+s[0]/2,n[1]+s[1]/2]}};var eN=6,i5e=1.4,ea,tN=null,mi=0,jh=null,Ad=()=>mi;async function nN(e){var t;return pe.initial&&(ea=null),ea?e.debug&&le("cached model:",ea.modelUrl):ea=await Ge((t=e.face.detector)==null?void 0:t.modelPath),mi=ea.inputs[0].shape?ea.inputs[0].shape[2]:0,jh=Ce(mi,"int32"),tN=ir(ZT(mi)),ea}function l5e(e){let t={};t.boxStarts=Oe(e,[0,1],[-1,2]),t.centers=ce(t.boxStarts,tN),t.boxSizes=Oe(e,[0,3],[-1,2]),t.boxSizesNormalized=he(t.boxSizes,jh),t.centersNormalized=he(t.centers,jh),t.halfBoxSize=he(t.boxSizesNormalized,rt.tf2),t.starts=fe(t.centersNormalized,t.halfBoxSize),t.ends=ce(t.centersNormalized,t.halfBoxSize),t.startNormalized=L(t.starts,jh),t.endNormalized=L(t.ends,jh);let n=Kl([t.startNormalized,t.endNormalized],1);return Object.keys(t).forEach(s=>ee(t[s])),n}async function sN(e,t){var i,l,u,c;if(!e||e.isDisposedInternal||e.shape.length!==4||e.shape[1]<1||e.shape[2]<1)return[];let n={};n.resized=Se.resizeBilinear(e,[mi,mi]),n.div=he(n.resized,rt.tf127),n.normalized=fe(n.div,rt.tf05);let s=ea==null?void 0:ea.execute(n.normalized);if(Array.isArray(s)&&s.length>2){let p=s.sort((d,h)=>d.size-h.size);n.concat384=Ct([p[0],p[2]],2),n.concat512=Ct([p[1],p[3]],2),n.concat=Ct([n.concat512,n.concat384],1),n.batch=st(n.concat,0)}else Array.isArray(s)?n.batch=st(s[0]):n.batch=st(s);ee(s),n.boxes=l5e(n.batch),n.logits=Oe(n.batch,[0,0],[-1,1]),n.sigmoid=Cn(n.logits),n.scores=st(n.sigmoid),n.nms=await Se.nonMaxSuppressionAsync(n.boxes,n.scores,((i=t.face.detector)==null?void 0:i.maxDetected)||0,((l=t.face.detector)==null?void 0:l.iouThreshold)||0,((u=t.face.detector)==null?void 0:u.minConfidence)||0);let r=await n.nms.array(),a=[],o=await n.scores.data();for(let p=0;p<r.length;p++){let d=o[r[p]];if(d>(((c=t.face.detector)==null?void 0:c.minConfidence)||0)){let h={};h.bbox=Oe(n.boxes,[r[p],0],[1,-1]),h.slice=Oe(n.batch,[r[p],eN-1],[1,-1]),h.squeeze=st(h.slice),h.landmarks=U(h.squeeze,[eN,-1]);let f=await h.bbox.data(),m={startPoint:[f[0],f[1]],endPoint:[f[2],f[3]],landmarks:await h.landmarks.array(),confidence:d},g=qT(m,[(e.shape[2]||0)/mi,(e.shape[1]||0)/mi]),y=B2(g,t.face.scale||i5e),x=W2(y);a.push(x),Object.keys(h).forEach(A=>ee(h[A]))}}return Object.keys(n).forEach(p=>ee(n[p])),a}var V2={};ia(V2,{connected:()=>Db,kpt:()=>_b});var _b=["nose","leftEyeInside","leftEye","leftEyeOutside","rightEyeInside","rightEye","rightEyeOutside","leftEar","rightEar","leftMouth","rightMouth","leftShoulder","rightShoulder","leftElbow","rightElbow","leftWrist","rightWrist","leftPinky","rightPinky","leftIndex","rightIndex","leftThumb","rightThumb","leftHip","rightHip","leftKnee","rightKnee","leftAnkle","rightAnkle","leftHeel","rightHeel","leftFoot","rightFoot","bodyCenter","bodyTop","leftPalm","leftHand","rightPalm","rightHand"],Db={shoulders:["leftShoulder","rightShoulder"],hips:["rightHip","leftHip"],mouth:["leftMouth","rightMouth"],leftLegUpper:["leftHip","leftKnee"],leftLegLower:["leftKnee","leftAnkle"],leftFoot:["leftAnkle","leftHeel","leftFoot"],leftTorso:["leftShoulder","leftHip"],leftArmUpper:["leftShoulder","leftElbow"],leftArmLower:["leftElbow","leftWrist"],leftHand:["leftWrist","leftPalm"],leftHandPinky:["leftPalm","leftPinky"],leftHandIndex:["leftPalm","leftIndex"],leftHandThumb:["leftPalm","leftThumb"],leftEyeOutline:["leftEyeInside","leftEyeOutside"],rightLegUpper:["rightHip","rightKnee"],rightLegLower:["rightKnee","rightAnkle"],rightFoot:["rightAnkle","rightHeel","rightFoot"],rightTorso:["rightShoulder","rightHip"],rightArmUpper:["rightShoulder","rightElbow"],rightArmLower:["rightElbow","rightWrist"],rightHand:["rightWrist","rightPalm"],rightHandPinky:["rightPalm","rightPinky"],rightHandIndex:["rightPalm","rightIndex"],rightHandThumb:["rightPalm","rightThumb"],rightEyeOutline:["rightEyeInside","rightEyeOutside"]};var aN=224,u5e,c5e=5,U2=[8,16,32,32,32];async function oN(){let e=[],t=0;for(;t<c5e;){let n=0,s=t;for(;s<U2.length&&U2[s]===U2[t];)n+=2,s++;let r=U2[t],a=Math.ceil(aN/r),o=Math.ceil(aN/r);for(let i=0;i<a;++i)for(let l=0;l<o;++l)for(let u=0;u<n;++u)e.push({x:(l+.5)/o,y:(i+.5)/a});t=s}u5e={x:Ft(e.map(n=>n.x)),y:Ft(e.map(n=>n.y))}}function Na(e,t=[1,1]){let n=[e.map(i=>i[0]),e.map(i=>i[1])],s=[Math.min(...n[0]),Math.min(...n[1])],r=[Math.max(...n[0]),Math.max(...n[1])],a=[s[0],s[1],r[0]-s[0],r[1]-s[1]],o=[a[0]/t[0],a[1]/t[1],a[2]/t[0],a[3]/t[1]];return{box:a,boxRaw:o}}function iN(e,t=[1,1]){let n=[e.map(u=>u[0]),e.map(u=>u[1])],s=[Math.min(...n[0]),Math.min(...n[1])],r=[Math.max(...n[0]),Math.max(...n[1])],a=[(s[0]+r[0])/2,(s[1]+r[1])/2],o=Math.max(a[0]-s[0],a[1]-s[1],-a[0]+r[0],-a[1]+r[1]),i=[Math.trunc(a[0]-o),Math.trunc(a[1]-o),Math.trunc(2*o),Math.trunc(2*o)],l=[i[0]/t[0],i[1]/t[1],i[2]/t[0],i[3]/t[1]];return{box:i,boxRaw:l}}function G2(e,t){let n=[e[2]*t,e[3]*t];return[e[0]-(n[0]-e[2])/2,e[1]-(n[1]-e[3])/2,n[0],n[1]]}var cN={initial:!0},Ms={detector:null,landmarks:null},xd={detector:[224,224],landmarks:[256,256]},$b=Number.MAX_SAFE_INTEGER,p5e={landmarks:["ld_3d","activation_segmentation","activation_heatmap","world_3d","output_poseflag"],detector:[]},j2=null,qh,gi=[[0,0],[0,0],[0,0],[0,0]],lN=0,uN=e=>1-1/(1+Math.exp(e));async function dN(e){if(cN.initial&&(Ms.detector=null),!Ms.detector&&e.body.detector&&e.body.detector.modelPath){Ms.detector=await Ge(e.body.detector.modelPath);let t=Object.values(Ms.detector.modelSignature.inputs);xd.detector[0]=Array.isArray(t)?parseInt(t[0].tensorShape.dim[1].size):0,xd.detector[1]=Array.isArray(t)?parseInt(t[0].tensorShape.dim[2].size):0}else e.debug&&Ms.detector&&le("cached model:",Ms.detector.modelUrl);return await oN(),Ms.detector}async function pN(e){if(cN.initial&&(Ms.landmarks=null),Ms.landmarks)e.debug&&le("cached model:",Ms.landmarks.modelUrl);else{Ms.landmarks=await Ge(e.body.modelPath);let t=Object.values(Ms.landmarks.modelSignature.inputs);xd.landmarks[0]=Array.isArray(t)?parseInt(t[0].tensorShape.dim[1].size):0,xd.landmarks[1]=Array.isArray(t)?parseInt(t[0].tensorShape.dim[2].size):0}return Ms.landmarks}async function h5e(e,t){let n={};if(!e.shape||!e.shape[1]||!e.shape[2])return e;let s;if(qh&&(n.cropped=Se.cropAndResize(e,[qh],[0],[e.shape[1],e.shape[2]])),e.shape[1]!==e.shape[2]){let r=[e.shape[2]>e.shape[1]?Math.trunc((e.shape[2]-e.shape[1])/2):0,e.shape[2]>e.shape[1]?Math.trunc((e.shape[2]-e.shape[1])/2):0],a=[e.shape[1]>e.shape[2]?Math.trunc((e.shape[1]-e.shape[2])/2):0,e.shape[1]>e.shape[2]?Math.trunc((e.shape[1]-e.shape[2])/2):0];gi=[[0,0],r,a,[0,0]],n.pad=Qs(n.cropped||e,gi),n.resize=Se.resizeBilinear(n.pad,[t,t]),s=he(n.resize,rt.tf255)}else e.shape[1]!==t?(n.resize=Se.resizeBilinear(n.cropped||e,[t,t]),s=he(n.resize,rt.tf255)):s=he(n.cropped||e,rt.tf255);return Object.keys(n).forEach(r=>ee(n[r])),s}function f5e(e,t){for(let n of e)n.position=[Math.trunc(n.position[0]*(t[0]+gi[2][0]+gi[2][1])/t[0]-gi[2][0]),Math.trunc(n.position[1]*(t[1]+gi[1][0]+gi[1][1])/t[1]-gi[1][0]),n.position[2]],n.positionRaw=[n.position[0]/t[0],n.position[1]/t[1],2*n.position[2]/(t[0]+t[1])];if(qh)for(let n of e)n.positionRaw=[n.positionRaw[0]+qh[1],n.positionRaw[1]+qh[0],n.positionRaw[2]],n.position=[Math.trunc(n.positionRaw[0]*t[0]),Math.trunc(n.positionRaw[1]*t[1]),n.positionRaw[2]];return e}async function m5e(e){let t=e.find(i=>i.part==="leftPalm"),n=e.find(i=>i.part==="leftWrist"),s=e.find(i=>i.part==="leftIndex");t.position[2]=((n.position[2]||0)+(s.position[2]||0))/2;let r=e.find(i=>i.part==="rightPalm"),a=e.find(i=>i.part==="rightWrist"),o=e.find(i=>i.part==="rightIndex");r.position[2]=((a.position[2]||0)+(o.position[2]||0))/2}async function g5e(e,t,n){var f;let s={};[s.ld,s.segmentation,s.heatmap,s.world,s.poseflag]=(f=Ms.landmarks)==null?void 0:f.execute(e,p5e.landmarks);let r=(await s.poseflag.data())[0],a=await s.ld.data(),o=await s.world.data();Object.keys(s).forEach(m=>ee(s[m]));let i=[],l=5;for(let m=0;m<a.length/l;m++){let g=uN(a[l*m+3]),y=uN(a[l*m+4]),x=Math.trunc(100*g*y*r)/100,A=[a[l*m+0]/xd.landmarks[0],a[l*m+1]/xd.landmarks[1],a[l*m+2]+0],b=[Math.trunc(n[0]*A[0]),Math.trunc(n[1]*A[1]),A[2]],w=[o[l*m+0],o[l*m+1],o[l*m+2]+0];i.push({part:_b[m],positionRaw:A,position:b,distance:w,score:x})}if(r<(t.body.minConfidence||0))return null;m5e(i);let u=f5e(i,n),c=u.map(m=>m.position),p=Na(c,[n[0],n[1]]),d={};for(let[m,g]of Object.entries(Db)){let y=[];for(let x=0;x<g.length-1;x++){let A=u.find(w=>w.part===g[x]),b=u.find(w=>w.part===g[x+1]);A&&b&&y.push([A.position,b.position])}d[m]=y}return{id:0,score:Math.trunc(100*r)/100,box:p.box,boxRaw:p.boxRaw,keypoints:u,annotations:d}}async function Pb(e,t){let n=[e.shape[2]||0,e.shape[1]||0],s=(t.body.skipTime||0)>ue()-lN,r=$b<(t.body.skipFrames||0);if(t.skipAllowed&&s&&r&&j2!==null)$b++;else{let a={};a.landmarks=await h5e(e,256),j2=await g5e(a.landmarks,t,n),Object.keys(a).forEach(o=>ee(a[o])),lN=ue(),$b=0}return j2?[j2]:[]}var bd=[{class:1,label:"person"},{class:2,label:"bicycle"},{class:3,label:"car"},{class:4,label:"motorcycle"},{class:5,label:"airplane"},{class:6,label:"bus"},{class:7,label:"train"},{class:8,label:"truck"},{class:9,label:"boat"},{class:10,label:"traffic light"},{class:11,label:"fire hydrant"},{class:12,label:"stop sign"},{class:13,label:"parking meter"},{class:14,label:"bench"},{class:15,label:"bird"},{class:16,label:"cat"},{class:17,label:"dog"},{class:18,label:"horse"},{class:19,label:"sheep"},{class:20,label:"cow"},{class:21,label:"elephant"},{class:22,label:"bear"},{class:23,label:"zebra"},{class:24,label:"giraffe"},{class:25,label:"backpack"},{class:26,label:"umbrella"},{class:27,label:"handbag"},{class:28,label:"tie"},{class:29,label:"suitcase"},{class:30,label:"frisbee"},{class:31,label:"skis"},{class:32,label:"snowboard"},{class:33,label:"sports ball"},{class:34,label:"kite"},{class:35,label:"baseball bat"},{class:36,label:"baseball glove"},{class:37,label:"skateboard"},{class:38,label:"surfboard"},{class:39,label:"tennis racket"},{class:40,label:"bottle"},{class:41,label:"wine glass"},{class:42,label:"cup"},{class:43,label:"fork"},{class:44,label:"knife"},{class:45,label:"spoon"},{class:46,label:"bowl"},{class:47,label:"banana"},{class:48,label:"apple"},{class:49,label:"sandwich"},{class:50,label:"orange"},{class:51,label:"broccoli"},{class:52,label:"carrot"},{class:53,label:"hot dog"},{class:54,label:"pizza"},{class:55,label:"donut"},{class:56,label:"cake"},{class:57,label:"chair"},{class:58,label:"couch"},{class:59,label:"potted plant"},{class:60,label:"bed"},{class:61,label:"dining table"},{class:62,label:"toilet"},{class:63,label:"tv"},{class:64,label:"laptop"},{class:65,label:"mouse"},{class:66,label:"remote"},{class:67,label:"keyboard"},{class:68,label:"cell phone"},{class:69,label:"microwave"},{class:70,label:"oven"},{class:71,label:"toaster"},{class:72,label:"sink"},{class:73,label:"refrigerator"},{class:74,label:"book"},{class:75,label:"clock"},{class:76,label:"vase"},{class:77,label:"scissors"},{class:78,label:"teddy bear"},{class:79,label:"hair drier"},{class:80,label:"toothbrush"}];var Ea,cu=0,Fb=[],fN=0,Ob=Number.MAX_SAFE_INTEGER;async function mN(e){if(pe.initial&&(Ea=null),Ea)e.debug&&le("cached model:",Ea.modelUrl);else{Ea=await Ge(e.object.modelPath);let t=Object.values(Ea.modelSignature.inputs);cu=Array.isArray(t)?parseInt(t[0].tensorShape.dim[2].size):0}return Ea}async function y5e(e,t,n){if(!e)return[];let s={},r=[],a=await e.array();s.squeeze=st(e);let o=Zt(s.squeeze,6,1);s.stack=on([o[1],o[0],o[3],o[2]],1),s.boxes=st(s.stack),s.scores=st(o[4]),s.classes=st(o[5]),ee([e,...o]),s.nms=await Se.nonMaxSuppressionAsync(s.boxes,s.scores,n.object.maxDetected,n.object.iouThreshold,n.object.minConfidence||0);let i=await s.nms.data(),l=0;for(let u of Array.from(i)){let c=Math.trunc(100*a[0][u][4])/100,p=a[0][u][5],d=bd[p].label,[h,f]=[a[0][u][0]/cu,a[0][u][1]/cu],m=[h,f,a[0][u][2]/cu-h,a[0][u][3]/cu-f],g=[Math.trunc(m[0]*t[0]),Math.trunc(m[1]*t[1]),Math.trunc(m[2]*t[0]),Math.trunc(m[3]*t[1])];r.push({id:l++,score:c,class:p,label:d,box:g,boxRaw:m})}return Object.keys(s).forEach(u=>ee(s[u])),r}async function Mb(e,t){let n=(t.object.skipTime||0)>ue()-fN,s=Ob<(t.object.skipFrames||0);return t.skipAllowed&&n&&s&&Fb.length>0?(Ob++,Fb):(Ob=0,new Promise(async r=>{let a=[e.shape[2]||0,e.shape[1]||0],o=Se.resizeBilinear(e,[cu,cu]),i=t.object.enabled?Ea==null?void 0:Ea.execute(o,["tower_0/detections"]):null;fN=ue(),ee(o);let l=await y5e(i,a,t);Fb=l,r(l)}))}var q2={};ia(q2,{connected:()=>Lb,kpt:()=>zb});var zb=["head","neck","rightShoulder","rightElbow","rightWrist","chest","leftShoulder","leftElbow","leftWrist","bodyCenter","rightHip","rightKnee","rightAnkle","leftHip","leftKnee","leftAnkle"],Lb={leftLeg:["leftHip","leftKnee","leftAnkle"],rightLeg:["rightHip","rightKnee","rightAnkle"],torso:["leftShoulder","rightShoulder","rightHip","leftHip","leftShoulder"],leftArm:["leftShoulder","leftElbow","leftWrist"],rightArm:["rightShoulder","rightElbow","rightWrist"],head:[]};var Bn,yN=0,us={id:0,keypoints:[],box:[0,0,0,0],boxRaw:[0,0,0,0],score:0,annotations:{}},Bb=Number.MAX_SAFE_INTEGER;async function AN(e){return pe.initial&&(Bn=null),Bn?e.debug&&le("cached model:",Bn.modelUrl):Bn=await Ge(e.body.modelPath),Bn}async function A5e(e,t){let[n,s]=e.shape,r=U(e,[s*n]),a=mn(r,0),o=(await a.data())[0];if(o>t){let i=Es(r,0),l=Yl(i,n),u=(await l.data())[0],c=he(i,n),p=(await c.data())[0];return ee([r,a,i,l,c]),[u,p,o]}else return ee([r,a]),[0,0,o]}async function Wb(e,t){let n=(t.body.skipTime||0)>ue()-yN,s=Bb<(t.body.skipFrames||0);return t.skipAllowed&&n&&s&&Object.keys(us.keypoints).length>0?(Bb++,[us]):(Bb=0,new Promise(async r=>{var p;let a=Y(()=>{if(!(Bn!=null&&Bn.inputs[0].shape))return null;let d=Se.resizeBilinear(e,[Bn.inputs[0].shape[2],Bn.inputs[0].shape[1]],!1),h=L(d,rt.tf2);return fe(h,rt.tf1)}),o;if(t.body.enabled&&(o=Bn==null?void 0:Bn.execute(a)),yN=ue(),ee(a),o){us.keypoints.length=0;let d=st(o);ee(o);let h=En(d,2);ee(d);for(let f=0;f<h.length;f++){let[m,g,y]=await A5e(h[f],t.body.minConfidence);y>(((p=t.body)==null?void 0:p.minConfidence)||0)&&us.keypoints.push({score:Math.round(100*y)/100,part:zb[f],positionRaw:[m/Bn.inputs[0].shape[2],g/Bn.inputs[0].shape[1]],position:[Math.round(e.shape[2]*m/Bn.inputs[0].shape[2]),Math.round(e.shape[1]*g/Bn.inputs[0].shape[1])]})}h.forEach(f=>ee(f))}us.score=us.keypoints.reduce((d,h)=>h.score>d?h.score:d,0);let i=us.keypoints.map(d=>d.position[0]),l=us.keypoints.map(d=>d.position[1]);us.box=[Math.min(...i),Math.min(...l),Math.max(...i)-Math.min(...i),Math.max(...l)-Math.min(...l)];let u=us.keypoints.map(d=>d.positionRaw[0]),c=us.keypoints.map(d=>d.positionRaw[1]);us.boxRaw=[Math.min(...u),Math.min(...c),Math.max(...u)-Math.min(...u),Math.max(...c)-Math.min(...c)];for(let[d,h]of Object.entries(Lb)){let f=[];for(let m=0;m<h.length-1;m++){let g=us.keypoints.find(x=>x.part===h[m]),y=us.keypoints.find(x=>x.part===h[m+1]);g&&y&&g.score>(t.body.minConfidence||0)&&y.score>(t.body.minConfidence||0)&&f.push([g.position,y.position])}us.annotations[d]=f}r([us])}))}var x5e=["angry","disgust","fear","happy","sad","surprise","neutral"],tr,X2=[],bN=0,vN=0,Vb=Number.MAX_SAFE_INTEGER;async function wN(e){var t;return pe.initial&&(tr=null),tr?e.debug&&le("cached model:",tr.modelUrl):tr=await Ge((t=e.face.emotion)==null?void 0:t.modelPath),tr}async function Ub(e,t,n,s){var o,i;if(!tr)return[];let r=Vb<(((o=t.face.emotion)==null?void 0:o.skipFrames)||0),a=(((i=t.face.emotion)==null?void 0:i.skipTime)||0)>ue()-vN;return t.skipAllowed&&a&&r&&bN===s&&X2[n]&&X2[n].length>0?(Vb++,X2[n]):(Vb=0,new Promise(async l=>{var c,p;let u=[];if((c=t.face.emotion)!=null&&c.enabled){let d={},h=tr!=null&&tr.inputs[0].shape?tr.inputs[0].shape[2]:0;d.resize=Se.resizeBilinear(e,[h,h],!1),d.channels=L(d.resize,rt.rgb),d.grayscale=ke(d.channels,3,!0),d.grayscaleSub=fe(d.grayscale,rt.tf05),d.grayscaleMul=L(d.grayscaleSub,rt.tf2),d.emotion=tr==null?void 0:tr.execute(d.grayscaleMul),vN=ue();let f=await d.emotion.data();for(let m=0;m<f.length;m++)f[m]>(((p=t.face.emotion)==null?void 0:p.minConfidence)||0)&&u.push({score:Math.min(.99,Math.trunc(100*f[m])/100),emotion:x5e[m]});u.sort((m,g)=>g.score-m.score),Object.keys(d).forEach(m=>ee(d[m]))}X2[n]=u,bN=s,l(u)}))}var zs,Gb=[],IN=0,SN=0,CN=Number.MAX_SAFE_INTEGER;async function TN(e){return pe.initial&&(zs=null),zs?e.debug&&le("cached model:",zs.modelUrl):zs=await Ge(e.face.mobilefacenet.modelPath),zs}async function Hb(e,t,n,s){var o,i;if(!zs)return[];let r=CN<(((o=t.face.mobilefacenet)==null?void 0:o.skipFrames)||0),a=(((i=t.face.mobilefacenet)==null?void 0:i.skipTime)||0)>ue()-SN;return t.skipAllowed&&a&&r&&IN===s&&Gb[n]?(CN++,Gb[n]):new Promise(async l=>{var c;let u=[];if(((c=t.face.mobilefacenet)==null?void 0:c.enabled)&&(zs==null?void 0:zs.inputs[0].shape)){let p={};p.crop=Se.resizeBilinear(e,[zs.inputs[0].shape[2],zs.inputs[0].shape[1]],!1),p.data=zs==null?void 0:zs.execute(p.crop);let d=await p.data.data();u=Array.from(d),Object.keys(p).forEach(h=>ee(p[h]))}Gb[n]=u,IN=s,SN=ue(),l(u)})}var Ls,jb=[],EN=0,RN=0,_N=Number.MAX_SAFE_INTEGER;async function DN(e){return pe.initial&&(Ls=null),Ls?e.debug&&le("cached model:",Ls.modelUrl):Ls=await Ge(e.face.insightface.modelPath),Ls}async function qb(e,t,n,s){var o,i;if(!Ls)return[];let r=_N<(((o=t.face.insightface)==null?void 0:o.skipFrames)||0),a=(((i=t.face.insightface)==null?void 0:i.skipTime)||0)>ue()-RN;return t.skipAllowed&&a&&r&&EN===s&&jb[n]?(_N++,jb[n]):new Promise(async l=>{var c;let u=[];if(((c=t.face.insightface)==null?void 0:c.enabled)&&(Ls==null?void 0:Ls.inputs[0].shape)){let p={};p.crop=Se.resizeBilinear(e,[Ls.inputs[0].shape[2],Ls.inputs[0].shape[1]],!1),p.data=Ls==null?void 0:Ls.execute(p.crop);let d=await p.data.data();u=Array.from(d),Object.keys(p).forEach(h=>ee(p[h]))}jb[n]=u,EN=s,RN=ue(),l(u)})}var Ra,yi=0,b5e=2.3,Xb=yr.leftEyeLower0,Kb=yr.rightEyeLower0,vd={leftBounds:[Xb[0],Xb[Xb.length-1]],rightBounds:[Kb[0],Kb[Kb.length-1]]},wd={upperCenter:3,lowerCenter:4,index:71,numCoordinates:76};async function MN(e){var t;return pe.initial&&(Ra=null),Ra?e.debug&&le("cached model:",Ra.modelUrl):Ra=await Ge((t=e.face.iris)==null?void 0:t.modelPath),yi=Ra.inputs[0].shape?Ra.inputs[0].shape[2]:0,yi===-1&&(yi=64),Ra}function K2(e,t,n,s){for(let r=0;r<Cb.length;r++){let{key:a,indices:o}=Cb[r],i=yr[`${n}${a}`];if(!s||s.includes(a))for(let l=0;l<o.length;l++){let u=o[l];e[i[l]]=[t[u][0],t[u][1],(t[u][2]+e[i[l]][2])/2]}}}var v5e=e=>{let t=e[vd.leftBounds[0]][2],n=e[vd.rightBounds[0]][2];return t-n},PN=(e,t,n,s,r,a=!1)=>{let o=W2(B2(XT([e[n],e[s]]),b5e)),i=yd(o),l=Se.cropAndResize(t,[[o.startPoint[1]/r,o.startPoint[0]/r,o.endPoint[1]/r,o.endPoint[0]/r]],[0],[yi,yi]);if(a&&pe.kernels.includes("flipleftright")){let u=Se.flipLeftRight(l);ee(l),l=u}return{box:o,boxSize:i,crop:l}},FN=(e,t,n,s=!1)=>{let r=[];for(let a=0;a<wd.numCoordinates;a++){let o=e[a*3],i=e[a*3+1],l=e[a*3+2];r.push([(s?1-o/yi:o/yi)*n[0]+t.startPoint[0],i/yi*n[1]+t.startPoint[1],l])}return{rawCoords:r,iris:r.slice(wd.index)}},ON=(e,t,n)=>{let s=e[yr[`${n}EyeUpper0`][wd.upperCenter]][2],r=e[yr[`${n}EyeLower0`][wd.lowerCenter]][2],a=(s+r)/2;return t.map((o,i)=>{let l=a;return i===2?l=s:i===4&&(l=r),[o[0],o[1],l]})};async function zN(e,t,n,s){if(!Ra)return n.debug&&le("face mesh iris detection requested, but model is not loaded"),e;let{box:r,boxSize:a,crop:o}=PN(e,t,vd.leftBounds[0],vd.leftBounds[1],s,!0),{box:i,boxSize:l,crop:u}=PN(e,t,vd.rightBounds[0],vd.rightBounds[1],s,!0),c=Ct([o,u]);ee(o),ee(u);let p=Ra.execute(c);ee(c);let d=await p.data();ee(p);let h=d.slice(0,wd.numCoordinates*3),{rawCoords:f,iris:m}=FN(h,r,a,!0),g=d.slice(wd.numCoordinates*3),{rawCoords:y,iris:x}=FN(g,i,l,!1),A=v5e(e);Math.abs(A)<30?(K2(e,f,"left",null),K2(e,y,"right",null)):A<1?K2(e,f,"left",["EyeUpper0","EyeLower0"]):K2(e,y,"right",["EyeUpper0","EyeLower0"]);let b=ON(e,m,"left"),w=ON(e,x,"right");return e.concat(b).concat(w)}var w5e=[[61,146],[146,91],[91,181],[181,84],[84,17],[17,314],[314,405],[405,321],[321,375],[375,291],[61,185],[185,40],[40,39],[39,37],[37,0],[0,267],[267,269],[269,270],[270,409],[409,291],[78,95],[95,88],[88,178],[178,87],[87,14],[14,317],[317,402],[402,318],[318,324],[324,308],[78,191],[191,80],[80,81],[81,82],[82,13],[13,312],[312,311],[311,310],[310,415],[415,308]],k5e=[[263,249],[249,390],[390,373],[373,374],[374,380],[380,381],[381,382],[382,362],[263,466],[466,388],[388,387],[387,386],[386,385],[385,384],[384,398],[398,362]],I5e=[[276,283],[283,282],[282,295],[295,285],[300,293],[293,334],[334,296],[296,336]],S5e=[[474,475],[475,476],[476,477],[477,474]],C5e=[[33,7],[7,163],[163,144],[144,145],[145,153],[153,154],[154,155],[155,133],[33,246],[246,161],[161,160],[160,159],[159,158],[158,157],[157,173],[173,133]],T5e=[[46,53],[53,52],[52,65],[65,55],[70,63],[63,105],[105,66],[66,107]],N5e=[[469,470],[470,471],[471,472],[472,469]],E5e=[[10,338],[338,297],[297,332],[332,284],[284,251],[251,389],[389,356],[356,454],[454,323],[323,361],[361,288],[288,397],[397,365],[365,379],[379,378],[378,400],[400,377],[377,152],[152,148],[148,176],[176,149],[149,150],[150,136],[136,172],[172,58],[58,132],[132,93],[93,234],[234,127],[127,162],[162,21],[21,54],[54,103],[103,67],[67,109],[109,10]];function Ai(e){let t=e.map(n=>n[0]);return t.push(e[e.length-1][1]),t}var R5e={lips:Ai(w5e),leftEye:Ai(k5e),leftEyebrow:Ai(I5e),leftIris:Ai(S5e),rightEye:Ai(C5e),rightEyebrow:Ai(T5e),rightIris:Ai(N5e),faceOval:Ai(E5e)},_5e=Object.entries(R5e).map(([e,t])=>t.map(n=>[n,e])).flat(),Eke=new Map(_5e),Xh=[61,146,91,181,84,17,314,405,321,375,291,185,40,39,37,0,267,269,270,409,78,95,88,178,87,14,317,402,318,324,308,191,80,81,82,13,312,311,310,415,76,77,90,180,85,16,315,404,320,307,306,184,74,73,72,11,302,303,304,408,62,96,89,179,86,15,316,403,319,325,292,183,42,41,38,12,268,271,272,407],du=[33,7,163,144,145,153,154,155,133,246,161,160,159,158,157,173,130,25,110,24,23,22,26,112,243,247,30,29,27,28,56,190,226,31,228,229,230,231,232,233,244,113,225,224,223,222,221,189,35,124,46,53,52,65,143,111,117,118,119,120,121,128,245,156,70,63,105,66,107,55,193],pu=[263,249,390,373,374,380,381,382,362,466,388,387,386,385,384,398,359,255,339,254,253,252,256,341,463,467,260,259,257,258,286,414,446,261,448,449,450,451,452,453,464,342,445,444,443,442,441,413,265,353,276,283,282,295,372,340,346,347,348,349,350,357,465,383,300,293,334,296,336,285,417];async function WN(e,t){let n={lips:await t.filter(a=>a.size===160)[0].data(),irisL:await t.filter(a=>a.size===10)[0].data(),eyeL:await t.filter(a=>a.size===142)[0].data(),irisR:await t.filter(a=>a.size===10)[1].data(),eyeR:await t.filter(a=>a.size===142)[1].data()},s=du.reduce((a,o)=>a+=e[o][2],0)/du.length;for(let a=0;a<n.irisL.length/2;a++)e.push([n.irisL[2*a+0],n.irisL[2*a+1],s]);let r=pu.reduce((a,o)=>a+=e[o][2],0)/pu.length;for(let a=0;a<n.irisR.length/2;a++)e.push([n.irisR[2*a+0],n.irisR[2*a+1],r]);for(let a=0;a<n.eyeL.length/2;a++)e[du[a]]=[n.eyeL[2*a+0],n.eyeL[2*a+1],e[du[a]][2]];for(let a=0;a<n.eyeR.length/2;a++)e[pu[a]]=[n.eyeR[2*a+0],n.eyeR[2*a+1],e[pu[a]][2]];for(let a=0;a<n.lips.length/2;a++)e[Xh[a]]=[n.lips[2*a+0],n.lips[2*a+1],e[Xh[a]][2]];return e}var ta={boxes:[],skipped:Number.MAX_SAFE_INTEGER,timestamp:0},Wn=null,hu=0;async function VN(e,t){var i,l,u,c,p,d,h,f,m,g,y;let n=(((i=t.face.detector)==null?void 0:i.skipTime)||0)>ue()-ta.timestamp,s=ta.skipped<(((l=t.face.detector)==null?void 0:l.skipFrames)||0);!t.skipAllowed||!n||!s||ta.boxes.length===0?(ta.boxes=await sN(e,t),ta.timestamp=ue(),ta.skipped=0):ta.skipped++;let r=[],a=[],o=0;for(let x=0;x<ta.boxes.length;x++){let A=ta.boxes[x],b=0,w,k={id:o++,mesh:[],meshRaw:[],box:[0,0,0,0],boxRaw:[0,0,0,0],score:0,boxScore:0,faceScore:0,annotations:{}};if([b,w,k.tensor]=JT((u=t.face.detector)==null?void 0:u.rotation,A,e,(c=t.face.mesh)!=null&&c.enabled?hu:Ad()),(p=t==null?void 0:t.filter)!=null&&p.equalization){let S=await _2(k.tensor);ee(k.tensor),k.tensor=S}if(k.boxScore=Math.round(100*A.confidence)/100,(d=t.face.mesh)!=null&&d.enabled)if(!Wn)t.debug&&le("face mesh detection requested, but model is not loaded");else{let S=Wn.execute(k.tensor),R=await S.find($=>$.shape[$.shape.length-1]===1).data();if(k.faceScore=Math.round(100*R[0])/100,k.faceScore<(((h=t.face.detector)==null?void 0:h.minConfidence)||1)){if(A.confidence=k.faceScore,(f=t.face.mesh)!=null&&f.keepInvalid){k.box=z2(A,e),k.boxRaw=L2(A,e),k.score=k.boxScore,k.mesh=A.landmarks.map($=>[(A.startPoint[0]+A.endPoint[0])/2+(A.endPoint[0]+A.startPoint[0])*$[0]/Ad(),(A.startPoint[1]+A.endPoint[1])/2+(A.endPoint[1]+A.startPoint[1])*$[1]/Ad()]),k.meshRaw=k.mesh.map($=>[$[0]/(e.shape[2]||1),$[1]/(e.shape[1]||1),($[2]||0)/hu]);for(let $ of Object.keys(iu))k.annotations[$]=[k.mesh[iu[$]]]}}else{let $=S.find(F=>F.shape[F.shape.length-1]===1404),_=U($,[-1,3]),P=await _.array();ee(_),(m=t.face.attention)!=null&&m.enabled?P=await WN(P,S):(g=t.face.iris)!=null&&g.enabled&&(P=await zN(P,k.tensor,t,hu)),k.mesh=YT(P,A,b,w,hu),k.meshRaw=k.mesh.map(F=>[F[0]/(e.shape[2]||0),F[1]/(e.shape[1]||0),(F[2]||0)/hu]);for(let F of Object.keys(yr))k.annotations[F]=yr[F].map(G=>k.mesh[G]);k.score=k.faceScore;let C={...QT(k.mesh,A),confidence:A.confidence,landmarks:A.landmarks};k.box=z2(C,e),k.boxRaw=L2(C,e),a.push(C)}ee(S)}else{k.box=z2(A,e),k.boxRaw=L2(A,e),k.score=k.boxScore,k.mesh=A.landmarks.map(S=>[(A.startPoint[0]+A.endPoint[0])/2+(A.endPoint[0]+A.startPoint[0])*S[0]/Ad(),(A.startPoint[1]+A.endPoint[1])/2+(A.endPoint[1]+A.startPoint[1])*S[1]/Ad()]),k.meshRaw=k.mesh.map(S=>[S[0]/(e.shape[2]||0),S[1]/(e.shape[1]||0),(S[2]||0)/hu]);for(let S of Object.keys(iu))k.annotations[S]=[k.mesh[iu[S]]]}k.score>(((y=t.face.detector)==null?void 0:y.minConfidence)||1)?r.push(k):ee(k.tensor)}return ta.boxes=a,r}async function UN(e){var t,n,s,r,a,o;return pe.initial&&(Wn=null),((n=(t=e==null?void 0:e.face)==null?void 0:t.attention)==null?void 0:n.enabled)&&(Wn==null?void 0:Wn.signature)&&Object.keys(((s=Wn==null?void 0:Wn.signature)==null?void 0:s.outputs)||{}).length<6&&(Wn=null),Wn?e.debug&&le("cached model:",Wn.modelUrl):(r=e.face.attention)!=null&&r.enabled?Wn=await Ge((a=e.face.attention)==null?void 0:a.modelPath):Wn=await Ge((o=e.face.mesh)==null?void 0:o.modelPath),hu=Wn.inputs[0].shape?Wn.inputs[0].shape[2]:0,Wn}var GN=lu,HN=Hh;var Bs,Z2=[],jN=0,qN=0,Yb=Number.MAX_SAFE_INTEGER;async function XN(e){var t;return pe.initial&&(Bs=null),Bs?e.debug&&le("cached model:",Bs.modelUrl):Bs=await Ge((t=e.face.description)==null?void 0:t.modelPath),Bs}function Jb(e){let t=e.image||e.tensor||e;if(!(Bs!=null&&Bs.inputs[0].shape))return t;let n=Se.resizeBilinear(t,[Bs.inputs[0].shape[2],Bs.inputs[0].shape[1]],!1),s=L(n,rt.tf255);return ee(n),s}async function Qb(e,t,n,s){var o,i,l,u;if(!Bs)return{age:0,gender:"unknown",genderScore:0,descriptor:[]};let r=Yb<(((o=t.face.description)==null?void 0:o.skipFrames)||0),a=(((i=t.face.description)==null?void 0:i.skipTime)||0)>ue()-jN;return t.skipAllowed&&r&&a&&qN===s&&((l=Z2[n])==null?void 0:l.age)&&((u=Z2[n])==null?void 0:u.age)>0?(Yb++,Z2[n]):(Yb=0,new Promise(async c=>{var d,h;let p={age:0,gender:"unknown",genderScore:0,descriptor:[]};if((d=t.face.description)!=null&&d.enabled){let f=Jb(e),m=Bs==null?void 0:Bs.execute(f);jN=ue(),ee(f);let y=await(await m.find(R=>R.shape[1]===1)).data(),x=Math.trunc(200*Math.abs(y[0]-.5))/100;x>(((h=t.face.description)==null?void 0:h.minConfidence)||0)&&(p.gender=y[0]<=.5?"female":"male",p.genderScore=Math.min(.99,x));let A=Es(m.find(R=>R.shape[1]===100),1),b=(await A.data())[0];ee(A);let k=await m.find(R=>R.shape[1]===100).data();p.age=Math.round(k[b-1]>k[b+1]?10*b-100*k[b-1]:10*b+100*k[b+1])/10;let S=m.find(R=>R.shape[1]===1024),E=S?await S.data():[];p.descriptor=Array.from(E),m.forEach(R=>ee(R))}Z2[n]=p,qN=s,c(p)}))}function Y2(e){return[Math.abs(e.endPoint[0]-e.startPoint[0]),Math.abs(e.endPoint[1]-e.startPoint[1])]}function Kh(e){return[e.startPoint[0]+(e.endPoint[0]-e.startPoint[0])/2,e.startPoint[1]+(e.endPoint[1]-e.startPoint[1])/2]}function YN(e,t,n){let s=t.shape[1],r=t.shape[2],a=[[e.startPoint[1]/s,e.startPoint[0]/r,e.endPoint[1]/s,e.endPoint[0]/r]];return Se.cropAndResize(t,a,[0],n)}function JN(e,t){let n=[e.startPoint[0]*t[0],e.startPoint[1]*t[1]],s=[e.endPoint[0]*t[0],e.endPoint[1]*t[1]],r=e.palmLandmarks.map(a=>[a[0]*t[0],a[1]*t[1]]);return{startPoint:n,endPoint:s,palmLandmarks:r,confidence:e.confidence}}function J2(e,t=1.5){let n=Kh(e),s=Y2(e),r=[t*s[0]/2,t*s[1]/2],a=[n[0]-r[0],n[1]-r[1]],o=[n[0]+r[0],n[1]+r[1]];return{startPoint:a,endPoint:o,palmLandmarks:e.palmLandmarks}}function Q2(e){let t=Kh(e),n=Y2(e),r=Math.max(...n)/2,a=[t[0]-r,t[1]-r],o=[t[0]+r,t[1]+r];return{startPoint:a,endPoint:o,palmLandmarks:e.palmLandmarks}}function $5e(e){return e-2*Math.PI*Math.floor((e+Math.PI)/(2*Math.PI))}function QN(e,t){let n=Math.PI/2-Math.atan2(-(t[1]-e[1]),t[0]-e[0]);return $5e(n)}var KN=(e,t)=>[[1,0,e],[0,1,t],[0,0,1]];function xi(e,t){let n=0;for(let s=0;s<e.length;s++)n+=e[s]*t[s];return n}function P5e(e,t){let n=[];for(let s=0;s<e.length;s++)n.push(e[s][t]);return n}function ZN(e,t){let n=[],s=e.length;for(let r=0;r<s;r++){n.push([]);for(let a=0;a<s;a++)n[r].push(xi(e[r],P5e(t,a)))}return n}function t4(e,t){let n=Math.cos(e),s=Math.sin(e),r=[[n,-s,0],[s,n,0],[0,0,1]],a=KN(t[0],t[1]),o=ZN(a,r),i=KN(-t[0],-t[1]);return ZN(o,i)}function eE(e){let t=[[e[0][0],e[1][0]],[e[0][1],e[1][1]]],n=[e[0][2],e[1][2]],s=[-xi(t[0],n),-xi(t[1],n)];return[t[0].concat(s[0]),t[1].concat(s[1]),[0,0,1]]}function n4(e,t){return[xi(e,t[0]),xi(e,t[1])]}var nE=[{x:.015625,y:.015625},{x:.015625,y:.015625},{x:.046875,y:.015625},{x:.046875,y:.015625},{x:.078125,y:.015625},{x:.078125,y:.015625},{x:.109375,y:.015625},{x:.109375,y:.015625},{x:.140625,y:.015625},{x:.140625,y:.015625},{x:.171875,y:.015625},{x:.171875,y:.015625},{x:.203125,y:.015625},{x:.203125,y:.015625},{x:.234375,y:.015625},{x:.234375,y:.015625},{x:.265625,y:.015625},{x:.265625,y:.015625},{x:.296875,y:.015625},{x:.296875,y:.015625},{x:.328125,y:.015625},{x:.328125,y:.015625},{x:.359375,y:.015625},{x:.359375,y:.015625},{x:.390625,y:.015625},{x:.390625,y:.015625},{x:.421875,y:.015625},{x:.421875,y:.015625},{x:.453125,y:.015625},{x:.453125,y:.015625},{x:.484375,y:.015625},{x:.484375,y:.015625},{x:.515625,y:.015625},{x:.515625,y:.015625},{x:.546875,y:.015625},{x:.546875,y:.015625},{x:.578125,y:.015625},{x:.578125,y:.015625},{x:.609375,y:.015625},{x:.609375,y:.015625},{x:.640625,y:.015625},{x:.640625,y:.015625},{x:.671875,y:.015625},{x:.671875,y:.015625},{x:.703125,y:.015625},{x:.703125,y:.015625},{x:.734375,y:.015625},{x:.734375,y:.015625},{x:.765625,y:.015625},{x:.765625,y:.015625},{x:.796875,y:.015625},{x:.796875,y:.015625},{x:.828125,y:.015625},{x:.828125,y:.015625},{x:.859375,y:.015625},{x:.859375,y:.015625},{x:.890625,y:.015625},{x:.890625,y:.015625},{x:.921875,y:.015625},{x:.921875,y:.015625},{x:.953125,y:.015625},{x:.953125,y:.015625},{x:.984375,y:.015625},{x:.984375,y:.015625},{x:.015625,y:.046875},{x:.015625,y:.046875},{x:.046875,y:.046875},{x:.046875,y:.046875},{x:.078125,y:.046875},{x:.078125,y:.046875},{x:.109375,y:.046875},{x:.109375,y:.046875},{x:.140625,y:.046875},{x:.140625,y:.046875},{x:.171875,y:.046875},{x:.171875,y:.046875},{x:.203125,y:.046875},{x:.203125,y:.046875},{x:.234375,y:.046875},{x:.234375,y:.046875},{x:.265625,y:.046875},{x:.265625,y:.046875},{x:.296875,y:.046875},{x:.296875,y:.046875},{x:.328125,y:.046875},{x:.328125,y:.046875},{x:.359375,y:.046875},{x:.359375,y:.046875},{x:.390625,y:.046875},{x:.390625,y:.046875},{x:.421875,y:.046875},{x:.421875,y:.046875},{x:.453125,y:.046875},{x:.453125,y:.046875},{x:.484375,y:.046875},{x:.484375,y:.046875},{x:.515625,y:.046875},{x:.515625,y:.046875},{x:.546875,y:.046875},{x:.546875,y:.046875},{x:.578125,y:.046875},{x:.578125,y:.046875},{x:.609375,y:.046875},{x:.609375,y:.046875},{x:.640625,y:.046875},{x:.640625,y:.046875},{x:.671875,y:.046875},{x:.671875,y:.046875},{x:.703125,y:.046875},{x:.703125,y:.046875},{x:.734375,y:.046875},{x:.734375,y:.046875},{x:.765625,y:.046875},{x:.765625,y:.046875},{x:.796875,y:.046875},{x:.796875,y:.046875},{x:.828125,y:.046875},{x:.828125,y:.046875},{x:.859375,y:.046875},{x:.859375,y:.046875},{x:.890625,y:.046875},{x:.890625,y:.046875},{x:.921875,y:.046875},{x:.921875,y:.046875},{x:.953125,y:.046875},{x:.953125,y:.046875},{x:.984375,y:.046875},{x:.984375,y:.046875},{x:.015625,y:.078125},{x:.015625,y:.078125},{x:.046875,y:.078125},{x:.046875,y:.078125},{x:.078125,y:.078125},{x:.078125,y:.078125},{x:.109375,y:.078125},{x:.109375,y:.078125},{x:.140625,y:.078125},{x:.140625,y:.078125},{x:.171875,y:.078125},{x:.171875,y:.078125},{x:.203125,y:.078125},{x:.203125,y:.078125},{x:.234375,y:.078125},{x:.234375,y:.078125},{x:.265625,y:.078125},{x:.265625,y:.078125},{x:.296875,y:.078125},{x:.296875,y:.078125},{x:.328125,y:.078125},{x:.328125,y:.078125},{x:.359375,y:.078125},{x:.359375,y:.078125},{x:.390625,y:.078125},{x:.390625,y:.078125},{x:.421875,y:.078125},{x:.421875,y:.078125},{x:.453125,y:.078125},{x:.453125,y:.078125},{x:.484375,y:.078125},{x:.484375,y:.078125},{x:.515625,y:.078125},{x:.515625,y:.078125},{x:.546875,y:.078125},{x:.546875,y:.078125},{x:.578125,y:.078125},{x:.578125,y:.078125},{x:.609375,y:.078125},{x:.609375,y:.078125},{x:.640625,y:.078125},{x:.640625,y:.078125},{x:.671875,y:.078125},{x:.671875,y:.078125},{x:.703125,y:.078125},{x:.703125,y:.078125},{x:.734375,y:.078125},{x:.734375,y:.078125},{x:.765625,y:.078125},{x:.765625,y:.078125},{x:.796875,y:.078125},{x:.796875,y:.078125},{x:.828125,y:.078125},{x:.828125,y:.078125},{x:.859375,y:.078125},{x:.859375,y:.078125},{x:.890625,y:.078125},{x:.890625,y:.078125},{x:.921875,y:.078125},{x:.921875,y:.078125},{x:.953125,y:.078125},{x:.953125,y:.078125},{x:.984375,y:.078125},{x:.984375,y:.078125},{x:.015625,y:.109375},{x:.015625,y:.109375},{x:.046875,y:.109375},{x:.046875,y:.109375},{x:.078125,y:.109375},{x:.078125,y:.109375},{x:.109375,y:.109375},{x:.109375,y:.109375},{x:.140625,y:.109375},{x:.140625,y:.109375},{x:.171875,y:.109375},{x:.171875,y:.109375},{x:.203125,y:.109375},{x:.203125,y:.109375},{x:.234375,y:.109375},{x:.234375,y:.109375},{x:.265625,y:.109375},{x:.265625,y:.109375},{x:.296875,y:.109375},{x:.296875,y:.109375},{x:.328125,y:.109375},{x:.328125,y:.109375},{x:.359375,y:.109375},{x:.359375,y:.109375},{x:.390625,y:.109375},{x:.390625,y:.109375},{x:.421875,y:.109375},{x:.421875,y:.109375},{x:.453125,y:.109375},{x:.453125,y:.109375},{x:.484375,y:.109375},{x:.484375,y:.109375},{x:.515625,y:.109375},{x:.515625,y:.109375},{x:.546875,y:.109375},{x:.546875,y:.109375},{x:.578125,y:.109375},{x:.578125,y:.109375},{x:.609375,y:.109375},{x:.609375,y:.109375},{x:.640625,y:.109375},{x:.640625,y:.109375},{x:.671875,y:.109375},{x:.671875,y:.109375},{x:.703125,y:.109375},{x:.703125,y:.109375},{x:.734375,y:.109375},{x:.734375,y:.109375},{x:.765625,y:.109375},{x:.765625,y:.109375},{x:.796875,y:.109375},{x:.796875,y:.109375},{x:.828125,y:.109375},{x:.828125,y:.109375},{x:.859375,y:.109375},{x:.859375,y:.109375},{x:.890625,y:.109375},{x:.890625,y:.109375},{x:.921875,y:.109375},{x:.921875,y:.109375},{x:.953125,y:.109375},{x:.953125,y:.109375},{x:.984375,y:.109375},{x:.984375,y:.109375},{x:.015625,y:.140625},{x:.015625,y:.140625},{x:.046875,y:.140625},{x:.046875,y:.140625},{x:.078125,y:.140625},{x:.078125,y:.140625},{x:.109375,y:.140625},{x:.109375,y:.140625},{x:.140625,y:.140625},{x:.140625,y:.140625},{x:.171875,y:.140625},{x:.171875,y:.140625},{x:.203125,y:.140625},{x:.203125,y:.140625},{x:.234375,y:.140625},{x:.234375,y:.140625},{x:.265625,y:.140625},{x:.265625,y:.140625},{x:.296875,y:.140625},{x:.296875,y:.140625},{x:.328125,y:.140625},{x:.328125,y:.140625},{x:.359375,y:.140625},{x:.359375,y:.140625},{x:.390625,y:.140625},{x:.390625,y:.140625},{x:.421875,y:.140625},{x:.421875,y:.140625},{x:.453125,y:.140625},{x:.453125,y:.140625},{x:.484375,y:.140625},{x:.484375,y:.140625},{x:.515625,y:.140625},{x:.515625,y:.140625},{x:.546875,y:.140625},{x:.546875,y:.140625},{x:.578125,y:.140625},{x:.578125,y:.140625},{x:.609375,y:.140625},{x:.609375,y:.140625},{x:.640625,y:.140625},{x:.640625,y:.140625},{x:.671875,y:.140625},{x:.671875,y:.140625},{x:.703125,y:.140625},{x:.703125,y:.140625},{x:.734375,y:.140625},{x:.734375,y:.140625},{x:.765625,y:.140625},{x:.765625,y:.140625},{x:.796875,y:.140625},{x:.796875,y:.140625},{x:.828125,y:.140625},{x:.828125,y:.140625},{x:.859375,y:.140625},{x:.859375,y:.140625},{x:.890625,y:.140625},{x:.890625,y:.140625},{x:.921875,y:.140625},{x:.921875,y:.140625},{x:.953125,y:.140625},{x:.953125,y:.140625},{x:.984375,y:.140625},{x:.984375,y:.140625},{x:.015625,y:.171875},{x:.015625,y:.171875},{x:.046875,y:.171875},{x:.046875,y:.171875},{x:.078125,y:.171875},{x:.078125,y:.171875},{x:.109375,y:.171875},{x:.109375,y:.171875},{x:.140625,y:.171875},{x:.140625,y:.171875},{x:.171875,y:.171875},{x:.171875,y:.171875},{x:.203125,y:.171875},{x:.203125,y:.171875},{x:.234375,y:.171875},{x:.234375,y:.171875},{x:.265625,y:.171875},{x:.265625,y:.171875},{x:.296875,y:.171875},{x:.296875,y:.171875},{x:.328125,y:.171875},{x:.328125,y:.171875},{x:.359375,y:.171875},{x:.359375,y:.171875},{x:.390625,y:.171875},{x:.390625,y:.171875},{x:.421875,y:.171875},{x:.421875,y:.171875},{x:.453125,y:.171875},{x:.453125,y:.171875},{x:.484375,y:.171875},{x:.484375,y:.171875},{x:.515625,y:.171875},{x:.515625,y:.171875},{x:.546875,y:.171875},{x:.546875,y:.171875},{x:.578125,y:.171875},{x:.578125,y:.171875},{x:.609375,y:.171875},{x:.609375,y:.171875},{x:.640625,y:.171875},{x:.640625,y:.171875},{x:.671875,y:.171875},{x:.671875,y:.171875},{x:.703125,y:.171875},{x:.703125,y:.171875},{x:.734375,y:.171875},{x:.734375,y:.171875},{x:.765625,y:.171875},{x:.765625,y:.171875},{x:.796875,y:.171875},{x:.796875,y:.171875},{x:.828125,y:.171875},{x:.828125,y:.171875},{x:.859375,y:.171875},{x:.859375,y:.171875},{x:.890625,y:.171875},{x:.890625,y:.171875},{x:.921875,y:.171875},{x:.921875,y:.171875},{x:.953125,y:.171875},{x:.953125,y:.171875},{x:.984375,y:.171875},{x:.984375,y:.171875},{x:.015625,y:.203125},{x:.015625,y:.203125},{x:.046875,y:.203125},{x:.046875,y:.203125},{x:.078125,y:.203125},{x:.078125,y:.203125},{x:.109375,y:.203125},{x:.109375,y:.203125},{x:.140625,y:.203125},{x:.140625,y:.203125},{x:.171875,y:.203125},{x:.171875,y:.203125},{x:.203125,y:.203125},{x:.203125,y:.203125},{x:.234375,y:.203125},{x:.234375,y:.203125},{x:.265625,y:.203125},{x:.265625,y:.203125},{x:.296875,y:.203125},{x:.296875,y:.203125},{x:.328125,y:.203125},{x:.328125,y:.203125},{x:.359375,y:.203125},{x:.359375,y:.203125},{x:.390625,y:.203125},{x:.390625,y:.203125},{x:.421875,y:.203125},{x:.421875,y:.203125},{x:.453125,y:.203125},{x:.453125,y:.203125},{x:.484375,y:.203125},{x:.484375,y:.203125},{x:.515625,y:.203125},{x:.515625,y:.203125},{x:.546875,y:.203125},{x:.546875,y:.203125},{x:.578125,y:.203125},{x:.578125,y:.203125},{x:.609375,y:.203125},{x:.609375,y:.203125},{x:.640625,y:.203125},{x:.640625,y:.203125},{x:.671875,y:.203125},{x:.671875,y:.203125},{x:.703125,y:.203125},{x:.703125,y:.203125},{x:.734375,y:.203125},{x:.734375,y:.203125},{x:.765625,y:.203125},{x:.765625,y:.203125},{x:.796875,y:.203125},{x:.796875,y:.203125},{x:.828125,y:.203125},{x:.828125,y:.203125},{x:.859375,y:.203125},{x:.859375,y:.203125},{x:.890625,y:.203125},{x:.890625,y:.203125},{x:.921875,y:.203125},{x:.921875,y:.203125},{x:.953125,y:.203125},{x:.953125,y:.203125},{x:.984375,y:.203125},{x:.984375,y:.203125},{x:.015625,y:.234375},{x:.015625,y:.234375},{x:.046875,y:.234375},{x:.046875,y:.234375},{x:.078125,y:.234375},{x:.078125,y:.234375},{x:.109375,y:.234375},{x:.109375,y:.234375},{x:.140625,y:.234375},{x:.140625,y:.234375},{x:.171875,y:.234375},{x:.171875,y:.234375},{x:.203125,y:.234375},{x:.203125,y:.234375},{x:.234375,y:.234375},{x:.234375,y:.234375},{x:.265625,y:.234375},{x:.265625,y:.234375},{x:.296875,y:.234375},{x:.296875,y:.234375},{x:.328125,y:.234375},{x:.328125,y:.234375},{x:.359375,y:.234375},{x:.359375,y:.234375},{x:.390625,y:.234375},{x:.390625,y:.234375},{x:.421875,y:.234375},{x:.421875,y:.234375},{x:.453125,y:.234375},{x:.453125,y:.234375},{x:.484375,y:.234375},{x:.484375,y:.234375},{x:.515625,y:.234375},{x:.515625,y:.234375},{x:.546875,y:.234375},{x:.546875,y:.234375},{x:.578125,y:.234375},{x:.578125,y:.234375},{x:.609375,y:.234375},{x:.609375,y:.234375},{x:.640625,y:.234375},{x:.640625,y:.234375},{x:.671875,y:.234375},{x:.671875,y:.234375},{x:.703125,y:.234375},{x:.703125,y:.234375},{x:.734375,y:.234375},{x:.734375,y:.234375},{x:.765625,y:.234375},{x:.765625,y:.234375},{x:.796875,y:.234375},{x:.796875,y:.234375},{x:.828125,y:.234375},{x:.828125,y:.234375},{x:.859375,y:.234375},{x:.859375,y:.234375},{x:.890625,y:.234375},{x:.890625,y:.234375},{x:.921875,y:.234375},{x:.921875,y:.234375},{x:.953125,y:.234375},{x:.953125,y:.234375},{x:.984375,y:.234375},{x:.984375,y:.234375},{x:.015625,y:.265625},{x:.015625,y:.265625},{x:.046875,y:.265625},{x:.046875,y:.265625},{x:.078125,y:.265625},{x:.078125,y:.265625},{x:.109375,y:.265625},{x:.109375,y:.265625},{x:.140625,y:.265625},{x:.140625,y:.265625},{x:.171875,y:.265625},{x:.171875,y:.265625},{x:.203125,y:.265625},{x:.203125,y:.265625},{x:.234375,y:.265625},{x:.234375,y:.265625},{x:.265625,y:.265625},{x:.265625,y:.265625},{x:.296875,y:.265625},{x:.296875,y:.265625},{x:.328125,y:.265625},{x:.328125,y:.265625},{x:.359375,y:.265625},{x:.359375,y:.265625},{x:.390625,y:.265625},{x:.390625,y:.265625},{x:.421875,y:.265625},{x:.421875,y:.265625},{x:.453125,y:.265625},{x:.453125,y:.265625},{x:.484375,y:.265625},{x:.484375,y:.265625},{x:.515625,y:.265625},{x:.515625,y:.265625},{x:.546875,y:.265625},{x:.546875,y:.265625},{x:.578125,y:.265625},{x:.578125,y:.265625},{x:.609375,y:.265625},{x:.609375,y:.265625},{x:.640625,y:.265625},{x:.640625,y:.265625},{x:.671875,y:.265625},{x:.671875,y:.265625},{x:.703125,y:.265625},{x:.703125,y:.265625},{x:.734375,y:.265625},{x:.734375,y:.265625},{x:.765625,y:.265625},{x:.765625,y:.265625},{x:.796875,y:.265625},{x:.796875,y:.265625},{x:.828125,y:.265625},{x:.828125,y:.265625},{x:.859375,y:.265625},{x:.859375,y:.265625},{x:.890625,y:.265625},{x:.890625,y:.265625},{x:.921875,y:.265625},{x:.921875,y:.265625},{x:.953125,y:.265625},{x:.953125,y:.265625},{x:.984375,y:.265625},{x:.984375,y:.265625},{x:.015625,y:.296875},{x:.015625,y:.296875},{x:.046875,y:.296875},{x:.046875,y:.296875},{x:.078125,y:.296875},{x:.078125,y:.296875},{x:.109375,y:.296875},{x:.109375,y:.296875},{x:.140625,y:.296875},{x:.140625,y:.296875},{x:.171875,y:.296875},{x:.171875,y:.296875},{x:.203125,y:.296875},{x:.203125,y:.296875},{x:.234375,y:.296875},{x:.234375,y:.296875},{x:.265625,y:.296875},{x:.265625,y:.296875},{x:.296875,y:.296875},{x:.296875,y:.296875},{x:.328125,y:.296875},{x:.328125,y:.296875},{x:.359375,y:.296875},{x:.359375,y:.296875},{x:.390625,y:.296875},{x:.390625,y:.296875},{x:.421875,y:.296875},{x:.421875,y:.296875},{x:.453125,y:.296875},{x:.453125,y:.296875},{x:.484375,y:.296875},{x:.484375,y:.296875},{x:.515625,y:.296875},{x:.515625,y:.296875},{x:.546875,y:.296875},{x:.546875,y:.296875},{x:.578125,y:.296875},{x:.578125,y:.296875},{x:.609375,y:.296875},{x:.609375,y:.296875},{x:.640625,y:.296875},{x:.640625,y:.296875},{x:.671875,y:.296875},{x:.671875,y:.296875},{x:.703125,y:.296875},{x:.703125,y:.296875},{x:.734375,y:.296875},{x:.734375,y:.296875},{x:.765625,y:.296875},{x:.765625,y:.296875},{x:.796875,y:.296875},{x:.796875,y:.296875},{x:.828125,y:.296875},{x:.828125,y:.296875},{x:.859375,y:.296875},{x:.859375,y:.296875},{x:.890625,y:.296875},{x:.890625,y:.296875},{x:.921875,y:.296875},{x:.921875,y:.296875},{x:.953125,y:.296875},{x:.953125,y:.296875},{x:.984375,y:.296875},{x:.984375,y:.296875},{x:.015625,y:.328125},{x:.015625,y:.328125},{x:.046875,y:.328125},{x:.046875,y:.328125},{x:.078125,y:.328125},{x:.078125,y:.328125},{x:.109375,y:.328125},{x:.109375,y:.328125},{x:.140625,y:.328125},{x:.140625,y:.328125},{x:.171875,y:.328125},{x:.171875,y:.328125},{x:.203125,y:.328125},{x:.203125,y:.328125},{x:.234375,y:.328125},{x:.234375,y:.328125},{x:.265625,y:.328125},{x:.265625,y:.328125},{x:.296875,y:.328125},{x:.296875,y:.328125},{x:.328125,y:.328125},{x:.328125,y:.328125},{x:.359375,y:.328125},{x:.359375,y:.328125},{x:.390625,y:.328125},{x:.390625,y:.328125},{x:.421875,y:.328125},{x:.421875,y:.328125},{x:.453125,y:.328125},{x:.453125,y:.328125},{x:.484375,y:.328125},{x:.484375,y:.328125},{x:.515625,y:.328125},{x:.515625,y:.328125},{x:.546875,y:.328125},{x:.546875,y:.328125},{x:.578125,y:.328125},{x:.578125,y:.328125},{x:.609375,y:.328125},{x:.609375,y:.328125},{x:.640625,y:.328125},{x:.640625,y:.328125},{x:.671875,y:.328125},{x:.671875,y:.328125},{x:.703125,y:.328125},{x:.703125,y:.328125},{x:.734375,y:.328125},{x:.734375,y:.328125},{x:.765625,y:.328125},{x:.765625,y:.328125},{x:.796875,y:.328125},{x:.796875,y:.328125},{x:.828125,y:.328125},{x:.828125,y:.328125},{x:.859375,y:.328125},{x:.859375,y:.328125},{x:.890625,y:.328125},{x:.890625,y:.328125},{x:.921875,y:.328125},{x:.921875,y:.328125},{x:.953125,y:.328125},{x:.953125,y:.328125},{x:.984375,y:.328125},{x:.984375,y:.328125},{x:.015625,y:.359375},{x:.015625,y:.359375},{x:.046875,y:.359375},{x:.046875,y:.359375},{x:.078125,y:.359375},{x:.078125,y:.359375},{x:.109375,y:.359375},{x:.109375,y:.359375},{x:.140625,y:.359375},{x:.140625,y:.359375},{x:.171875,y:.359375},{x:.171875,y:.359375},{x:.203125,y:.359375},{x:.203125,y:.359375},{x:.234375,y:.359375},{x:.234375,y:.359375},{x:.265625,y:.359375},{x:.265625,y:.359375},{x:.296875,y:.359375},{x:.296875,y:.359375},{x:.328125,y:.359375},{x:.328125,y:.359375},{x:.359375,y:.359375},{x:.359375,y:.359375},{x:.390625,y:.359375},{x:.390625,y:.359375},{x:.421875,y:.359375},{x:.421875,y:.359375},{x:.453125,y:.359375},{x:.453125,y:.359375},{x:.484375,y:.359375},{x:.484375,y:.359375},{x:.515625,y:.359375},{x:.515625,y:.359375},{x:.546875,y:.359375},{x:.546875,y:.359375},{x:.578125,y:.359375},{x:.578125,y:.359375},{x:.609375,y:.359375},{x:.609375,y:.359375},{x:.640625,y:.359375},{x:.640625,y:.359375},{x:.671875,y:.359375},{x:.671875,y:.359375},{x:.703125,y:.359375},{x:.703125,y:.359375},{x:.734375,y:.359375},{x:.734375,y:.359375},{x:.765625,y:.359375},{x:.765625,y:.359375},{x:.796875,y:.359375},{x:.796875,y:.359375},{x:.828125,y:.359375},{x:.828125,y:.359375},{x:.859375,y:.359375},{x:.859375,y:.359375},{x:.890625,y:.359375},{x:.890625,y:.359375},{x:.921875,y:.359375},{x:.921875,y:.359375},{x:.953125,y:.359375},{x:.953125,y:.359375},{x:.984375,y:.359375},{x:.984375,y:.359375},{x:.015625,y:.390625},{x:.015625,y:.390625},{x:.046875,y:.390625},{x:.046875,y:.390625},{x:.078125,y:.390625},{x:.078125,y:.390625},{x:.109375,y:.390625},{x:.109375,y:.390625},{x:.140625,y:.390625},{x:.140625,y:.390625},{x:.171875,y:.390625},{x:.171875,y:.390625},{x:.203125,y:.390625},{x:.203125,y:.390625},{x:.234375,y:.390625},{x:.234375,y:.390625},{x:.265625,y:.390625},{x:.265625,y:.390625},{x:.296875,y:.390625},{x:.296875,y:.390625},{x:.328125,y:.390625},{x:.328125,y:.390625},{x:.359375,y:.390625},{x:.359375,y:.390625},{x:.390625,y:.390625},{x:.390625,y:.390625},{x:.421875,y:.390625},{x:.421875,y:.390625},{x:.453125,y:.390625},{x:.453125,y:.390625},{x:.484375,y:.390625},{x:.484375,y:.390625},{x:.515625,y:.390625},{x:.515625,y:.390625},{x:.546875,y:.390625},{x:.546875,y:.390625},{x:.578125,y:.390625},{x:.578125,y:.390625},{x:.609375,y:.390625},{x:.609375,y:.390625},{x:.640625,y:.390625},{x:.640625,y:.390625},{x:.671875,y:.390625},{x:.671875,y:.390625},{x:.703125,y:.390625},{x:.703125,y:.390625},{x:.734375,y:.390625},{x:.734375,y:.390625},{x:.765625,y:.390625},{x:.765625,y:.390625},{x:.796875,y:.390625},{x:.796875,y:.390625},{x:.828125,y:.390625},{x:.828125,y:.390625},{x:.859375,y:.390625},{x:.859375,y:.390625},{x:.890625,y:.390625},{x:.890625,y:.390625},{x:.921875,y:.390625},{x:.921875,y:.390625},{x:.953125,y:.390625},{x:.953125,y:.390625},{x:.984375,y:.390625},{x:.984375,y:.390625},{x:.015625,y:.421875},{x:.015625,y:.421875},{x:.046875,y:.421875},{x:.046875,y:.421875},{x:.078125,y:.421875},{x:.078125,y:.421875},{x:.109375,y:.421875},{x:.109375,y:.421875},{x:.140625,y:.421875},{x:.140625,y:.421875},{x:.171875,y:.421875},{x:.171875,y:.421875},{x:.203125,y:.421875},{x:.203125,y:.421875},{x:.234375,y:.421875},{x:.234375,y:.421875},{x:.265625,y:.421875},{x:.265625,y:.421875},{x:.296875,y:.421875},{x:.296875,y:.421875},{x:.328125,y:.421875},{x:.328125,y:.421875},{x:.359375,y:.421875},{x:.359375,y:.421875},{x:.390625,y:.421875},{x:.390625,y:.421875},{x:.421875,y:.421875},{x:.421875,y:.421875},{x:.453125,y:.421875},{x:.453125,y:.421875},{x:.484375,y:.421875},{x:.484375,y:.421875},{x:.515625,y:.421875},{x:.515625,y:.421875},{x:.546875,y:.421875},{x:.546875,y:.421875},{x:.578125,y:.421875},{x:.578125,y:.421875},{x:.609375,y:.421875},{x:.609375,y:.421875},{x:.640625,y:.421875},{x:.640625,y:.421875},{x:.671875,y:.421875},{x:.671875,y:.421875},{x:.703125,y:.421875},{x:.703125,y:.421875},{x:.734375,y:.421875},{x:.734375,y:.421875},{x:.765625,y:.421875},{x:.765625,y:.421875},{x:.796875,y:.421875},{x:.796875,y:.421875},{x:.828125,y:.421875},{x:.828125,y:.421875},{x:.859375,y:.421875},{x:.859375,y:.421875},{x:.890625,y:.421875},{x:.890625,y:.421875},{x:.921875,y:.421875},{x:.921875,y:.421875},{x:.953125,y:.421875},{x:.953125,y:.421875},{x:.984375,y:.421875},{x:.984375,y:.421875},{x:.015625,y:.453125},{x:.015625,y:.453125},{x:.046875,y:.453125},{x:.046875,y:.453125},{x:.078125,y:.453125},{x:.078125,y:.453125},{x:.109375,y:.453125},{x:.109375,y:.453125},{x:.140625,y:.453125},{x:.140625,y:.453125},{x:.171875,y:.453125},{x:.171875,y:.453125},{x:.203125,y:.453125},{x:.203125,y:.453125},{x:.234375,y:.453125},{x:.234375,y:.453125},{x:.265625,y:.453125},{x:.265625,y:.453125},{x:.296875,y:.453125},{x:.296875,y:.453125},{x:.328125,y:.453125},{x:.328125,y:.453125},{x:.359375,y:.453125},{x:.359375,y:.453125},{x:.390625,y:.453125},{x:.390625,y:.453125},{x:.421875,y:.453125},{x:.421875,y:.453125},{x:.453125,y:.453125},{x:.453125,y:.453125},{x:.484375,y:.453125},{x:.484375,y:.453125},{x:.515625,y:.453125},{x:.515625,y:.453125},{x:.546875,y:.453125},{x:.546875,y:.453125},{x:.578125,y:.453125},{x:.578125,y:.453125},{x:.609375,y:.453125},{x:.609375,y:.453125},{x:.640625,y:.453125},{x:.640625,y:.453125},{x:.671875,y:.453125},{x:.671875,y:.453125},{x:.703125,y:.453125},{x:.703125,y:.453125},{x:.734375,y:.453125},{x:.734375,y:.453125},{x:.765625,y:.453125},{x:.765625,y:.453125},{x:.796875,y:.453125},{x:.796875,y:.453125},{x:.828125,y:.453125},{x:.828125,y:.453125},{x:.859375,y:.453125},{x:.859375,y:.453125},{x:.890625,y:.453125},{x:.890625,y:.453125},{x:.921875,y:.453125},{x:.921875,y:.453125},{x:.953125,y:.453125},{x:.953125,y:.453125},{x:.984375,y:.453125},{x:.984375,y:.453125},{x:.015625,y:.484375},{x:.015625,y:.484375},{x:.046875,y:.484375},{x:.046875,y:.484375},{x:.078125,y:.484375},{x:.078125,y:.484375},{x:.109375,y:.484375},{x:.109375,y:.484375},{x:.140625,y:.484375},{x:.140625,y:.484375},{x:.171875,y:.484375},{x:.171875,y:.484375},{x:.203125,y:.484375},{x:.203125,y:.484375},{x:.234375,y:.484375},{x:.234375,y:.484375},{x:.265625,y:.484375},{x:.265625,y:.484375},{x:.296875,y:.484375},{x:.296875,y:.484375},{x:.328125,y:.484375},{x:.328125,y:.484375},{x:.359375,y:.484375},{x:.359375,y:.484375},{x:.390625,y:.484375},{x:.390625,y:.484375},{x:.421875,y:.484375},{x:.421875,y:.484375},{x:.453125,y:.484375},{x:.453125,y:.484375},{x:.484375,y:.484375},{x:.484375,y:.484375},{x:.515625,y:.484375},{x:.515625,y:.484375},{x:.546875,y:.484375},{x:.546875,y:.484375},{x:.578125,y:.484375},{x:.578125,y:.484375},{x:.609375,y:.484375},{x:.609375,y:.484375},{x:.640625,y:.484375},{x:.640625,y:.484375},{x:.671875,y:.484375},{x:.671875,y:.484375},{x:.703125,y:.484375},{x:.703125,y:.484375},{x:.734375,y:.484375},{x:.734375,y:.484375},{x:.765625,y:.484375},{x:.765625,y:.484375},{x:.796875,y:.484375},{x:.796875,y:.484375},{x:.828125,y:.484375},{x:.828125,y:.484375},{x:.859375,y:.484375},{x:.859375,y:.484375},{x:.890625,y:.484375},{x:.890625,y:.484375},{x:.921875,y:.484375},{x:.921875,y:.484375},{x:.953125,y:.484375},{x:.953125,y:.484375},{x:.984375,y:.484375},{x:.984375,y:.484375},{x:.015625,y:.515625},{x:.015625,y:.515625},{x:.046875,y:.515625},{x:.046875,y:.515625},{x:.078125,y:.515625},{x:.078125,y:.515625},{x:.109375,y:.515625},{x:.109375,y:.515625},{x:.140625,y:.515625},{x:.140625,y:.515625},{x:.171875,y:.515625},{x:.171875,y:.515625},{x:.203125,y:.515625},{x:.203125,y:.515625},{x:.234375,y:.515625},{x:.234375,y:.515625},{x:.265625,y:.515625},{x:.265625,y:.515625},{x:.296875,y:.515625},{x:.296875,y:.515625},{x:.328125,y:.515625},{x:.328125,y:.515625},{x:.359375,y:.515625},{x:.359375,y:.515625},{x:.390625,y:.515625},{x:.390625,y:.515625},{x:.421875,y:.515625},{x:.421875,y:.515625},{x:.453125,y:.515625},{x:.453125,y:.515625},{x:.484375,y:.515625},{x:.484375,y:.515625},{x:.515625,y:.515625},{x:.515625,y:.515625},{x:.546875,y:.515625},{x:.546875,y:.515625},{x:.578125,y:.515625},{x:.578125,y:.515625},{x:.609375,y:.515625},{x:.609375,y:.515625},{x:.640625,y:.515625},{x:.640625,y:.515625},{x:.671875,y:.515625},{x:.671875,y:.515625},{x:.703125,y:.515625},{x:.703125,y:.515625},{x:.734375,y:.515625},{x:.734375,y:.515625},{x:.765625,y:.515625},{x:.765625,y:.515625},{x:.796875,y:.515625},{x:.796875,y:.515625},{x:.828125,y:.515625},{x:.828125,y:.515625},{x:.859375,y:.515625},{x:.859375,y:.515625},{x:.890625,y:.515625},{x:.890625,y:.515625},{x:.921875,y:.515625},{x:.921875,y:.515625},{x:.953125,y:.515625},{x:.953125,y:.515625},{x:.984375,y:.515625},{x:.984375,y:.515625},{x:.015625,y:.546875},{x:.015625,y:.546875},{x:.046875,y:.546875},{x:.046875,y:.546875},{x:.078125,y:.546875},{x:.078125,y:.546875},{x:.109375,y:.546875},{x:.109375,y:.546875},{x:.140625,y:.546875},{x:.140625,y:.546875},{x:.171875,y:.546875},{x:.171875,y:.546875},{x:.203125,y:.546875},{x:.203125,y:.546875},{x:.234375,y:.546875},{x:.234375,y:.546875},{x:.265625,y:.546875},{x:.265625,y:.546875},{x:.296875,y:.546875},{x:.296875,y:.546875},{x:.328125,y:.546875},{x:.328125,y:.546875},{x:.359375,y:.546875},{x:.359375,y:.546875},{x:.390625,y:.546875},{x:.390625,y:.546875},{x:.421875,y:.546875},{x:.421875,y:.546875},{x:.453125,y:.546875},{x:.453125,y:.546875},{x:.484375,y:.546875},{x:.484375,y:.546875},{x:.515625,y:.546875},{x:.515625,y:.546875},{x:.546875,y:.546875},{x:.546875,y:.546875},{x:.578125,y:.546875},{x:.578125,y:.546875},{x:.609375,y:.546875},{x:.609375,y:.546875},{x:.640625,y:.546875},{x:.640625,y:.546875},{x:.671875,y:.546875},{x:.671875,y:.546875},{x:.703125,y:.546875},{x:.703125,y:.546875},{x:.734375,y:.546875},{x:.734375,y:.546875},{x:.765625,y:.546875},{x:.765625,y:.546875},{x:.796875,y:.546875},{x:.796875,y:.546875},{x:.828125,y:.546875},{x:.828125,y:.546875},{x:.859375,y:.546875},{x:.859375,y:.546875},{x:.890625,y:.546875},{x:.890625,y:.546875},{x:.921875,y:.546875},{x:.921875,y:.546875},{x:.953125,y:.546875},{x:.953125,y:.546875},{x:.984375,y:.546875},{x:.984375,y:.546875},{x:.015625,y:.578125},{x:.015625,y:.578125},{x:.046875,y:.578125},{x:.046875,y:.578125},{x:.078125,y:.578125},{x:.078125,y:.578125},{x:.109375,y:.578125},{x:.109375,y:.578125},{x:.140625,y:.578125},{x:.140625,y:.578125},{x:.171875,y:.578125},{x:.171875,y:.578125},{x:.203125,y:.578125},{x:.203125,y:.578125},{x:.234375,y:.578125},{x:.234375,y:.578125},{x:.265625,y:.578125},{x:.265625,y:.578125},{x:.296875,y:.578125},{x:.296875,y:.578125},{x:.328125,y:.578125},{x:.328125,y:.578125},{x:.359375,y:.578125},{x:.359375,y:.578125},{x:.390625,y:.578125},{x:.390625,y:.578125},{x:.421875,y:.578125},{x:.421875,y:.578125},{x:.453125,y:.578125},{x:.453125,y:.578125},{x:.484375,y:.578125},{x:.484375,y:.578125},{x:.515625,y:.578125},{x:.515625,y:.578125},{x:.546875,y:.578125},{x:.546875,y:.578125},{x:.578125,y:.578125},{x:.578125,y:.578125},{x:.609375,y:.578125},{x:.609375,y:.578125},{x:.640625,y:.578125},{x:.640625,y:.578125},{x:.671875,y:.578125},{x:.671875,y:.578125},{x:.703125,y:.578125},{x:.703125,y:.578125},{x:.734375,y:.578125},{x:.734375,y:.578125},{x:.765625,y:.578125},{x:.765625,y:.578125},{x:.796875,y:.578125},{x:.796875,y:.578125},{x:.828125,y:.578125},{x:.828125,y:.578125},{x:.859375,y:.578125},{x:.859375,y:.578125},{x:.890625,y:.578125},{x:.890625,y:.578125},{x:.921875,y:.578125},{x:.921875,y:.578125},{x:.953125,y:.578125},{x:.953125,y:.578125},{x:.984375,y:.578125},{x:.984375,y:.578125},{x:.015625,y:.609375},{x:.015625,y:.609375},{x:.046875,y:.609375},{x:.046875,y:.609375},{x:.078125,y:.609375},{x:.078125,y:.609375},{x:.109375,y:.609375},{x:.109375,y:.609375},{x:.140625,y:.609375},{x:.140625,y:.609375},{x:.171875,y:.609375},{x:.171875,y:.609375},{x:.203125,y:.609375},{x:.203125,y:.609375},{x:.234375,y:.609375},{x:.234375,y:.609375},{x:.265625,y:.609375},{x:.265625,y:.609375},{x:.296875,y:.609375},{x:.296875,y:.609375},{x:.328125,y:.609375},{x:.328125,y:.609375},{x:.359375,y:.609375},{x:.359375,y:.609375},{x:.390625,y:.609375},{x:.390625,y:.609375},{x:.421875,y:.609375},{x:.421875,y:.609375},{x:.453125,y:.609375},{x:.453125,y:.609375},{x:.484375,y:.609375},{x:.484375,y:.609375},{x:.515625,y:.609375},{x:.515625,y:.609375},{x:.546875,y:.609375},{x:.546875,y:.609375},{x:.578125,y:.609375},{x:.578125,y:.609375},{x:.609375,y:.609375},{x:.609375,y:.609375},{x:.640625,y:.609375},{x:.640625,y:.609375},{x:.671875,y:.609375},{x:.671875,y:.609375},{x:.703125,y:.609375},{x:.703125,y:.609375},{x:.734375,y:.609375},{x:.734375,y:.609375},{x:.765625,y:.609375},{x:.765625,y:.609375},{x:.796875,y:.609375},{x:.796875,y:.609375},{x:.828125,y:.609375},{x:.828125,y:.609375},{x:.859375,y:.609375},{x:.859375,y:.609375},{x:.890625,y:.609375},{x:.890625,y:.609375},{x:.921875,y:.609375},{x:.921875,y:.609375},{x:.953125,y:.609375},{x:.953125,y:.609375},{x:.984375,y:.609375},{x:.984375,y:.609375},{x:.015625,y:.640625},{x:.015625,y:.640625},{x:.046875,y:.640625},{x:.046875,y:.640625},{x:.078125,y:.640625},{x:.078125,y:.640625},{x:.109375,y:.640625},{x:.109375,y:.640625},{x:.140625,y:.640625},{x:.140625,y:.640625},{x:.171875,y:.640625},{x:.171875,y:.640625},{x:.203125,y:.640625},{x:.203125,y:.640625},{x:.234375,y:.640625},{x:.234375,y:.640625},{x:.265625,y:.640625},{x:.265625,y:.640625},{x:.296875,y:.640625},{x:.296875,y:.640625},{x:.328125,y:.640625},{x:.328125,y:.640625},{x:.359375,y:.640625},{x:.359375,y:.640625},{x:.390625,y:.640625},{x:.390625,y:.640625},{x:.421875,y:.640625},{x:.421875,y:.640625},{x:.453125,y:.640625},{x:.453125,y:.640625},{x:.484375,y:.640625},{x:.484375,y:.640625},{x:.515625,y:.640625},{x:.515625,y:.640625},{x:.546875,y:.640625},{x:.546875,y:.640625},{x:.578125,y:.640625},{x:.578125,y:.640625},{x:.609375,y:.640625},{x:.609375,y:.640625},{x:.640625,y:.640625},{x:.640625,y:.640625},{x:.671875,y:.640625},{x:.671875,y:.640625},{x:.703125,y:.640625},{x:.703125,y:.640625},{x:.734375,y:.640625},{x:.734375,y:.640625},{x:.765625,y:.640625},{x:.765625,y:.640625},{x:.796875,y:.640625},{x:.796875,y:.640625},{x:.828125,y:.640625},{x:.828125,y:.640625},{x:.859375,y:.640625},{x:.859375,y:.640625},{x:.890625,y:.640625},{x:.890625,y:.640625},{x:.921875,y:.640625},{x:.921875,y:.640625},{x:.953125,y:.640625},{x:.953125,y:.640625},{x:.984375,y:.640625},{x:.984375,y:.640625},{x:.015625,y:.671875},{x:.015625,y:.671875},{x:.046875,y:.671875},{x:.046875,y:.671875},{x:.078125,y:.671875},{x:.078125,y:.671875},{x:.109375,y:.671875},{x:.109375,y:.671875},{x:.140625,y:.671875},{x:.140625,y:.671875},{x:.171875,y:.671875},{x:.171875,y:.671875},{x:.203125,y:.671875},{x:.203125,y:.671875},{x:.234375,y:.671875},{x:.234375,y:.671875},{x:.265625,y:.671875},{x:.265625,y:.671875},{x:.296875,y:.671875},{x:.296875,y:.671875},{x:.328125,y:.671875},{x:.328125,y:.671875},{x:.359375,y:.671875},{x:.359375,y:.671875},{x:.390625,y:.671875},{x:.390625,y:.671875},{x:.421875,y:.671875},{x:.421875,y:.671875},{x:.453125,y:.671875},{x:.453125,y:.671875},{x:.484375,y:.671875},{x:.484375,y:.671875},{x:.515625,y:.671875},{x:.515625,y:.671875},{x:.546875,y:.671875},{x:.546875,y:.671875},{x:.578125,y:.671875},{x:.578125,y:.671875},{x:.609375,y:.671875},{x:.609375,y:.671875},{x:.640625,y:.671875},{x:.640625,y:.671875},{x:.671875,y:.671875},{x:.671875,y:.671875},{x:.703125,y:.671875},{x:.703125,y:.671875},{x:.734375,y:.671875},{x:.734375,y:.671875},{x:.765625,y:.671875},{x:.765625,y:.671875},{x:.796875,y:.671875},{x:.796875,y:.671875},{x:.828125,y:.671875},{x:.828125,y:.671875},{x:.859375,y:.671875},{x:.859375,y:.671875},{x:.890625,y:.671875},{x:.890625,y:.671875},{x:.921875,y:.671875},{x:.921875,y:.671875},{x:.953125,y:.671875},{x:.953125,y:.671875},{x:.984375,y:.671875},{x:.984375,y:.671875},{x:.015625,y:.703125},{x:.015625,y:.703125},{x:.046875,y:.703125},{x:.046875,y:.703125},{x:.078125,y:.703125},{x:.078125,y:.703125},{x:.109375,y:.703125},{x:.109375,y:.703125},{x:.140625,y:.703125},{x:.140625,y:.703125},{x:.171875,y:.703125},{x:.171875,y:.703125},{x:.203125,y:.703125},{x:.203125,y:.703125},{x:.234375,y:.703125},{x:.234375,y:.703125},{x:.265625,y:.703125},{x:.265625,y:.703125},{x:.296875,y:.703125},{x:.296875,y:.703125},{x:.328125,y:.703125},{x:.328125,y:.703125},{x:.359375,y:.703125},{x:.359375,y:.703125},{x:.390625,y:.703125},{x:.390625,y:.703125},{x:.421875,y:.703125},{x:.421875,y:.703125},{x:.453125,y:.703125},{x:.453125,y:.703125},{x:.484375,y:.703125},{x:.484375,y:.703125},{x:.515625,y:.703125},{x:.515625,y:.703125},{x:.546875,y:.703125},{x:.546875,y:.703125},{x:.578125,y:.703125},{x:.578125,y:.703125},{x:.609375,y:.703125},{x:.609375,y:.703125},{x:.640625,y:.703125},{x:.640625,y:.703125},{x:.671875,y:.703125},{x:.671875,y:.703125},{x:.703125,y:.703125},{x:.703125,y:.703125},{x:.734375,y:.703125},{x:.734375,y:.703125},{x:.765625,y:.703125},{x:.765625,y:.703125},{x:.796875,y:.703125},{x:.796875,y:.703125},{x:.828125,y:.703125},{x:.828125,y:.703125},{x:.859375,y:.703125},{x:.859375,y:.703125},{x:.890625,y:.703125},{x:.890625,y:.703125},{x:.921875,y:.703125},{x:.921875,y:.703125},{x:.953125,y:.703125},{x:.953125,y:.703125},{x:.984375,y:.703125},{x:.984375,y:.703125},{x:.015625,y:.734375},{x:.015625,y:.734375},{x:.046875,y:.734375},{x:.046875,y:.734375},{x:.078125,y:.734375},{x:.078125,y:.734375},{x:.109375,y:.734375},{x:.109375,y:.734375},{x:.140625,y:.734375},{x:.140625,y:.734375},{x:.171875,y:.734375},{x:.171875,y:.734375},{x:.203125,y:.734375},{x:.203125,y:.734375},{x:.234375,y:.734375},{x:.234375,y:.734375},{x:.265625,y:.734375},{x:.265625,y:.734375},{x:.296875,y:.734375},{x:.296875,y:.734375},{x:.328125,y:.734375},{x:.328125,y:.734375},{x:.359375,y:.734375},{x:.359375,y:.734375},{x:.390625,y:.734375},{x:.390625,y:.734375},{x:.421875,y:.734375},{x:.421875,y:.734375},{x:.453125,y:.734375},{x:.453125,y:.734375},{x:.484375,y:.734375},{x:.484375,y:.734375},{x:.515625,y:.734375},{x:.515625,y:.734375},{x:.546875,y:.734375},{x:.546875,y:.734375},{x:.578125,y:.734375},{x:.578125,y:.734375},{x:.609375,y:.734375},{x:.609375,y:.734375},{x:.640625,y:.734375},{x:.640625,y:.734375},{x:.671875,y:.734375},{x:.671875,y:.734375},{x:.703125,y:.734375},{x:.703125,y:.734375},{x:.734375,y:.734375},{x:.734375,y:.734375},{x:.765625,y:.734375},{x:.765625,y:.734375},{x:.796875,y:.734375},{x:.796875,y:.734375},{x:.828125,y:.734375},{x:.828125,y:.734375},{x:.859375,y:.734375},{x:.859375,y:.734375},{x:.890625,y:.734375},{x:.890625,y:.734375},{x:.921875,y:.734375},{x:.921875,y:.734375},{x:.953125,y:.734375},{x:.953125,y:.734375},{x:.984375,y:.734375},{x:.984375,y:.734375},{x:.015625,y:.765625},{x:.015625,y:.765625},{x:.046875,y:.765625},{x:.046875,y:.765625},{x:.078125,y:.765625},{x:.078125,y:.765625},{x:.109375,y:.765625},{x:.109375,y:.765625},{x:.140625,y:.765625},{x:.140625,y:.765625},{x:.171875,y:.765625},{x:.171875,y:.765625},{x:.203125,y:.765625},{x:.203125,y:.765625},{x:.234375,y:.765625},{x:.234375,y:.765625},{x:.265625,y:.765625},{x:.265625,y:.765625},{x:.296875,y:.765625},{x:.296875,y:.765625},{x:.328125,y:.765625},{x:.328125,y:.765625},{x:.359375,y:.765625},{x:.359375,y:.765625},{x:.390625,y:.765625},{x:.390625,y:.765625},{x:.421875,y:.765625},{x:.421875,y:.765625},{x:.453125,y:.765625},{x:.453125,y:.765625},{x:.484375,y:.765625},{x:.484375,y:.765625},{x:.515625,y:.765625},{x:.515625,y:.765625},{x:.546875,y:.765625},{x:.546875,y:.765625},{x:.578125,y:.765625},{x:.578125,y:.765625},{x:.609375,y:.765625},{x:.609375,y:.765625},{x:.640625,y:.765625},{x:.640625,y:.765625},{x:.671875,y:.765625},{x:.671875,y:.765625},{x:.703125,y:.765625},{x:.703125,y:.765625},{x:.734375,y:.765625},{x:.734375,y:.765625},{x:.765625,y:.765625},{x:.765625,y:.765625},{x:.796875,y:.765625},{x:.796875,y:.765625},{x:.828125,y:.765625},{x:.828125,y:.765625},{x:.859375,y:.765625},{x:.859375,y:.765625},{x:.890625,y:.765625},{x:.890625,y:.765625},{x:.921875,y:.765625},{x:.921875,y:.765625},{x:.953125,y:.765625},{x:.953125,y:.765625},{x:.984375,y:.765625},{x:.984375,y:.765625},{x:.015625,y:.796875},{x:.015625,y:.796875},{x:.046875,y:.796875},{x:.046875,y:.796875},{x:.078125,y:.796875},{x:.078125,y:.796875},{x:.109375,y:.796875},{x:.109375,y:.796875},{x:.140625,y:.796875},{x:.140625,y:.796875},{x:.171875,y:.796875},{x:.171875,y:.796875},{x:.203125,y:.796875},{x:.203125,y:.796875},{x:.234375,y:.796875},{x:.234375,y:.796875},{x:.265625,y:.796875},{x:.265625,y:.796875},{x:.296875,y:.796875},{x:.296875,y:.796875},{x:.328125,y:.796875},{x:.328125,y:.796875},{x:.359375,y:.796875},{x:.359375,y:.796875},{x:.390625,y:.796875},{x:.390625,y:.796875},{x:.421875,y:.796875},{x:.421875,y:.796875},{x:.453125,y:.796875},{x:.453125,y:.796875},{x:.484375,y:.796875},{x:.484375,y:.796875},{x:.515625,y:.796875},{x:.515625,y:.796875},{x:.546875,y:.796875},{x:.546875,y:.796875},{x:.578125,y:.796875},{x:.578125,y:.796875},{x:.609375,y:.796875},{x:.609375,y:.796875},{x:.640625,y:.796875},{x:.640625,y:.796875},{x:.671875,y:.796875},{x:.671875,y:.796875},{x:.703125,y:.796875},{x:.703125,y:.796875},{x:.734375,y:.796875},{x:.734375,y:.796875},{x:.765625,y:.796875},{x:.765625,y:.796875},{x:.796875,y:.796875},{x:.796875,y:.796875},{x:.828125,y:.796875},{x:.828125,y:.796875},{x:.859375,y:.796875},{x:.859375,y:.796875},{x:.890625,y:.796875},{x:.890625,y:.796875},{x:.921875,y:.796875},{x:.921875,y:.796875},{x:.953125,y:.796875},{x:.953125,y:.796875},{x:.984375,y:.796875},{x:.984375,y:.796875},{x:.015625,y:.828125},{x:.015625,y:.828125},{x:.046875,y:.828125},{x:.046875,y:.828125},{x:.078125,y:.828125},{x:.078125,y:.828125},{x:.109375,y:.828125},{x:.109375,y:.828125},{x:.140625,y:.828125},{x:.140625,y:.828125},{x:.171875,y:.828125},{x:.171875,y:.828125},{x:.203125,y:.828125},{x:.203125,y:.828125},{x:.234375,y:.828125},{x:.234375,y:.828125},{x:.265625,y:.828125},{x:.265625,y:.828125},{x:.296875,y:.828125},{x:.296875,y:.828125},{x:.328125,y:.828125},{x:.328125,y:.828125},{x:.359375,y:.828125},{x:.359375,y:.828125},{x:.390625,y:.828125},{x:.390625,y:.828125},{x:.421875,y:.828125},{x:.421875,y:.828125},{x:.453125,y:.828125},{x:.453125,y:.828125},{x:.484375,y:.828125},{x:.484375,y:.828125},{x:.515625,y:.828125},{x:.515625,y:.828125},{x:.546875,y:.828125},{x:.546875,y:.828125},{x:.578125,y:.828125},{x:.578125,y:.828125},{x:.609375,y:.828125},{x:.609375,y:.828125},{x:.640625,y:.828125},{x:.640625,y:.828125},{x:.671875,y:.828125},{x:.671875,y:.828125},{x:.703125,y:.828125},{x:.703125,y:.828125},{x:.734375,y:.828125},{x:.734375,y:.828125},{x:.765625,y:.828125},{x:.765625,y:.828125},{x:.796875,y:.828125},{x:.796875,y:.828125},{x:.828125,y:.828125},{x:.828125,y:.828125},{x:.859375,y:.828125},{x:.859375,y:.828125},{x:.890625,y:.828125},{x:.890625,y:.828125},{x:.921875,y:.828125},{x:.921875,y:.828125},{x:.953125,y:.828125},{x:.953125,y:.828125},{x:.984375,y:.828125},{x:.984375,y:.828125},{x:.015625,y:.859375},{x:.015625,y:.859375},{x:.046875,y:.859375},{x:.046875,y:.859375},{x:.078125,y:.859375},{x:.078125,y:.859375},{x:.109375,y:.859375},{x:.109375,y:.859375},{x:.140625,y:.859375},{x:.140625,y:.859375},{x:.171875,y:.859375},{x:.171875,y:.859375},{x:.203125,y:.859375},{x:.203125,y:.859375},{x:.234375,y:.859375},{x:.234375,y:.859375},{x:.265625,y:.859375},{x:.265625,y:.859375},{x:.296875,y:.859375},{x:.296875,y:.859375},{x:.328125,y:.859375},{x:.328125,y:.859375},{x:.359375,y:.859375},{x:.359375,y:.859375},{x:.390625,y:.859375},{x:.390625,y:.859375},{x:.421875,y:.859375},{x:.421875,y:.859375},{x:.453125,y:.859375},{x:.453125,y:.859375},{x:.484375,y:.859375},{x:.484375,y:.859375},{x:.515625,y:.859375},{x:.515625,y:.859375},{x:.546875,y:.859375},{x:.546875,y:.859375},{x:.578125,y:.859375},{x:.578125,y:.859375},{x:.609375,y:.859375},{x:.609375,y:.859375},{x:.640625,y:.859375},{x:.640625,y:.859375},{x:.671875,y:.859375},{x:.671875,y:.859375},{x:.703125,y:.859375},{x:.703125,y:.859375},{x:.734375,y:.859375},{x:.734375,y:.859375},{x:.765625,y:.859375},{x:.765625,y:.859375},{x:.796875,y:.859375},{x:.796875,y:.859375},{x:.828125,y:.859375},{x:.828125,y:.859375},{x:.859375,y:.859375},{x:.859375,y:.859375},{x:.890625,y:.859375},{x:.890625,y:.859375},{x:.921875,y:.859375},{x:.921875,y:.859375},{x:.953125,y:.859375},{x:.953125,y:.859375},{x:.984375,y:.859375},{x:.984375,y:.859375},{x:.015625,y:.890625},{x:.015625,y:.890625},{x:.046875,y:.890625},{x:.046875,y:.890625},{x:.078125,y:.890625},{x:.078125,y:.890625},{x:.109375,y:.890625},{x:.109375,y:.890625},{x:.140625,y:.890625},{x:.140625,y:.890625},{x:.171875,y:.890625},{x:.171875,y:.890625},{x:.203125,y:.890625},{x:.203125,y:.890625},{x:.234375,y:.890625},{x:.234375,y:.890625},{x:.265625,y:.890625},{x:.265625,y:.890625},{x:.296875,y:.890625},{x:.296875,y:.890625},{x:.328125,y:.890625},{x:.328125,y:.890625},{x:.359375,y:.890625},{x:.359375,y:.890625},{x:.390625,y:.890625},{x:.390625,y:.890625},{x:.421875,y:.890625},{x:.421875,y:.890625},{x:.453125,y:.890625},{x:.453125,y:.890625},{x:.484375,y:.890625},{x:.484375,y:.890625},{x:.515625,y:.890625},{x:.515625,y:.890625},{x:.546875,y:.890625},{x:.546875,y:.890625},{x:.578125,y:.890625},{x:.578125,y:.890625},{x:.609375,y:.890625},{x:.609375,y:.890625},{x:.640625,y:.890625},{x:.640625,y:.890625},{x:.671875,y:.890625},{x:.671875,y:.890625},{x:.703125,y:.890625},{x:.703125,y:.890625},{x:.734375,y:.890625},{x:.734375,y:.890625},{x:.765625,y:.890625},{x:.765625,y:.890625},{x:.796875,y:.890625},{x:.796875,y:.890625},{x:.828125,y:.890625},{x:.828125,y:.890625},{x:.859375,y:.890625},{x:.859375,y:.890625},{x:.890625,y:.890625},{x:.890625,y:.890625},{x:.921875,y:.890625},{x:.921875,y:.890625},{x:.953125,y:.890625},{x:.953125,y:.890625},{x:.984375,y:.890625},{x:.984375,y:.890625},{x:.015625,y:.921875},{x:.015625,y:.921875},{x:.046875,y:.921875},{x:.046875,y:.921875},{x:.078125,y:.921875},{x:.078125,y:.921875},{x:.109375,y:.921875},{x:.109375,y:.921875},{x:.140625,y:.921875},{x:.140625,y:.921875},{x:.171875,y:.921875},{x:.171875,y:.921875},{x:.203125,y:.921875},{x:.203125,y:.921875},{x:.234375,y:.921875},{x:.234375,y:.921875},{x:.265625,y:.921875},{x:.265625,y:.921875},{x:.296875,y:.921875},{x:.296875,y:.921875},{x:.328125,y:.921875},{x:.328125,y:.921875},{x:.359375,y:.921875},{x:.359375,y:.921875},{x:.390625,y:.921875},{x:.390625,y:.921875},{x:.421875,y:.921875},{x:.421875,y:.921875},{x:.453125,y:.921875},{x:.453125,y:.921875},{x:.484375,y:.921875},{x:.484375,y:.921875},{x:.515625,y:.921875},{x:.515625,y:.921875},{x:.546875,y:.921875},{x:.546875,y:.921875},{x:.578125,y:.921875},{x:.578125,y:.921875},{x:.609375,y:.921875},{x:.609375,y:.921875},{x:.640625,y:.921875},{x:.640625,y:.921875},{x:.671875,y:.921875},{x:.671875,y:.921875},{x:.703125,y:.921875},{x:.703125,y:.921875},{x:.734375,y:.921875},{x:.734375,y:.921875},{x:.765625,y:.921875},{x:.765625,y:.921875},{x:.796875,y:.921875},{x:.796875,y:.921875},{x:.828125,y:.921875},{x:.828125,y:.921875},{x:.859375,y:.921875},{x:.859375,y:.921875},{x:.890625,y:.921875},{x:.890625,y:.921875},{x:.921875,y:.921875},{x:.921875,y:.921875},{x:.953125,y:.921875},{x:.953125,y:.921875},{x:.984375,y:.921875},{x:.984375,y:.921875},{x:.015625,y:.953125},{x:.015625,y:.953125},{x:.046875,y:.953125},{x:.046875,y:.953125},{x:.078125,y:.953125},{x:.078125,y:.953125},{x:.109375,y:.953125},{x:.109375,y:.953125},{x:.140625,y:.953125},{x:.140625,y:.953125},{x:.171875,y:.953125},{x:.171875,y:.953125},{x:.203125,y:.953125},{x:.203125,y:.953125},{x:.234375,y:.953125},{x:.234375,y:.953125},{x:.265625,y:.953125},{x:.265625,y:.953125},{x:.296875,y:.953125},{x:.296875,y:.953125},{x:.328125,y:.953125},{x:.328125,y:.953125},{x:.359375,y:.953125},{x:.359375,y:.953125},{x:.390625,y:.953125},{x:.390625,y:.953125},{x:.421875,y:.953125},{x:.421875,y:.953125},{x:.453125,y:.953125},{x:.453125,y:.953125},{x:.484375,y:.953125},{x:.484375,y:.953125},{x:.515625,y:.953125},{x:.515625,y:.953125},{x:.546875,y:.953125},{x:.546875,y:.953125},{x:.578125,y:.953125},{x:.578125,y:.953125},{x:.609375,y:.953125},{x:.609375,y:.953125},{x:.640625,y:.953125},{x:.640625,y:.953125},{x:.671875,y:.953125},{x:.671875,y:.953125},{x:.703125,y:.953125},{x:.703125,y:.953125},{x:.734375,y:.953125},{x:.734375,y:.953125},{x:.765625,y:.953125},{x:.765625,y:.953125},{x:.796875,y:.953125},{x:.796875,y:.953125},{x:.828125,y:.953125},{x:.828125,y:.953125},{x:.859375,y:.953125},{x:.859375,y:.953125},{x:.890625,y:.953125},{x:.890625,y:.953125},{x:.921875,y:.953125},{x:.921875,y:.953125},{x:.953125,y:.953125},{x:.953125,y:.953125},{x:.984375,y:.953125},{x:.984375,y:.953125},{x:.015625,y:.984375},{x:.015625,y:.984375},{x:.046875,y:.984375},{x:.046875,y:.984375},{x:.078125,y:.984375},{x:.078125,y:.984375},{x:.109375,y:.984375},{x:.109375,y:.984375},{x:.140625,y:.984375},{x:.140625,y:.984375},{x:.171875,y:.984375},{x:.171875,y:.984375},{x:.203125,y:.984375},{x:.203125,y:.984375},{x:.234375,y:.984375},{x:.234375,y:.984375},{x:.265625,y:.984375},{x:.265625,y:.984375},{x:.296875,y:.984375},{x:.296875,y:.984375},{x:.328125,y:.984375},{x:.328125,y:.984375},{x:.359375,y:.984375},{x:.359375,y:.984375},{x:.390625,y:.984375},{x:.390625,y:.984375},{x:.421875,y:.984375},{x:.421875,y:.984375},{x:.453125,y:.984375},{x:.453125,y:.984375},{x:.484375,y:.984375},{x:.484375,y:.984375},{x:.515625,y:.984375},{x:.515625,y:.984375},{x:.546875,y:.984375},{x:.546875,y:.984375},{x:.578125,y:.984375},{x:.578125,y:.984375},{x:.609375,y:.984375},{x:.609375,y:.984375},{x:.640625,y:.984375},{x:.640625,y:.984375},{x:.671875,y:.984375},{x:.671875,y:.984375},{x:.703125,y:.984375},{x:.703125,y:.984375},{x:.734375,y:.984375},{x:.734375,y:.984375},{x:.765625,y:.984375},{x:.765625,y:.984375},{x:.796875,y:.984375},{x:.796875,y:.984375},{x:.828125,y:.984375},{x:.828125,y:.984375},{x:.859375,y:.984375},{x:.859375,y:.984375},{x:.890625,y:.984375},{x:.890625,y:.984375},{x:.921875,y:.984375},{x:.921875,y:.984375},{x:.953125,y:.984375},{x:.953125,y:.984375},{x:.984375,y:.984375},{x:.984375,y:.984375},{x:.03125,y:.03125},{x:.03125,y:.03125},{x:.09375,y:.03125},{x:.09375,y:.03125},{x:.15625,y:.03125},{x:.15625,y:.03125},{x:.21875,y:.03125},{x:.21875,y:.03125},{x:.28125,y:.03125},{x:.28125,y:.03125},{x:.34375,y:.03125},{x:.34375,y:.03125},{x:.40625,y:.03125},{x:.40625,y:.03125},{x:.46875,y:.03125},{x:.46875,y:.03125},{x:.53125,y:.03125},{x:.53125,y:.03125},{x:.59375,y:.03125},{x:.59375,y:.03125},{x:.65625,y:.03125},{x:.65625,y:.03125},{x:.71875,y:.03125},{x:.71875,y:.03125},{x:.78125,y:.03125},{x:.78125,y:.03125},{x:.84375,y:.03125},{x:.84375,y:.03125},{x:.90625,y:.03125},{x:.90625,y:.03125},{x:.96875,y:.03125},{x:.96875,y:.03125},{x:.03125,y:.09375},{x:.03125,y:.09375},{x:.09375,y:.09375},{x:.09375,y:.09375},{x:.15625,y:.09375},{x:.15625,y:.09375},{x:.21875,y:.09375},{x:.21875,y:.09375},{x:.28125,y:.09375},{x:.28125,y:.09375},{x:.34375,y:.09375},{x:.34375,y:.09375},{x:.40625,y:.09375},{x:.40625,y:.09375},{x:.46875,y:.09375},{x:.46875,y:.09375},{x:.53125,y:.09375},{x:.53125,y:.09375},{x:.59375,y:.09375},{x:.59375,y:.09375},{x:.65625,y:.09375},{x:.65625,y:.09375},{x:.71875,y:.09375},{x:.71875,y:.09375},{x:.78125,y:.09375},{x:.78125,y:.09375},{x:.84375,y:.09375},{x:.84375,y:.09375},{x:.90625,y:.09375},{x:.90625,y:.09375},{x:.96875,y:.09375},{x:.96875,y:.09375},{x:.03125,y:.15625},{x:.03125,y:.15625},{x:.09375,y:.15625},{x:.09375,y:.15625},{x:.15625,y:.15625},{x:.15625,y:.15625},{x:.21875,y:.15625},{x:.21875,y:.15625},{x:.28125,y:.15625},{x:.28125,y:.15625},{x:.34375,y:.15625},{x:.34375,y:.15625},{x:.40625,y:.15625},{x:.40625,y:.15625},{x:.46875,y:.15625},{x:.46875,y:.15625},{x:.53125,y:.15625},{x:.53125,y:.15625},{x:.59375,y:.15625},{x:.59375,y:.15625},{x:.65625,y:.15625},{x:.65625,y:.15625},{x:.71875,y:.15625},{x:.71875,y:.15625},{x:.78125,y:.15625},{x:.78125,y:.15625},{x:.84375,y:.15625},{x:.84375,y:.15625},{x:.90625,y:.15625},{x:.90625,y:.15625},{x:.96875,y:.15625},{x:.96875,y:.15625},{x:.03125,y:.21875},{x:.03125,y:.21875},{x:.09375,y:.21875},{x:.09375,y:.21875},{x:.15625,y:.21875},{x:.15625,y:.21875},{x:.21875,y:.21875},{x:.21875,y:.21875},{x:.28125,y:.21875},{x:.28125,y:.21875},{x:.34375,y:.21875},{x:.34375,y:.21875},{x:.40625,y:.21875},{x:.40625,y:.21875},{x:.46875,y:.21875},{x:.46875,y:.21875},{x:.53125,y:.21875},{x:.53125,y:.21875},{x:.59375,y:.21875},{x:.59375,y:.21875},{x:.65625,y:.21875},{x:.65625,y:.21875},{x:.71875,y:.21875},{x:.71875,y:.21875},{x:.78125,y:.21875},{x:.78125,y:.21875},{x:.84375,y:.21875},{x:.84375,y:.21875},{x:.90625,y:.21875},{x:.90625,y:.21875},{x:.96875,y:.21875},{x:.96875,y:.21875},{x:.03125,y:.28125},{x:.03125,y:.28125},{x:.09375,y:.28125},{x:.09375,y:.28125},{x:.15625,y:.28125},{x:.15625,y:.28125},{x:.21875,y:.28125},{x:.21875,y:.28125},{x:.28125,y:.28125},{x:.28125,y:.28125},{x:.34375,y:.28125},{x:.34375,y:.28125},{x:.40625,y:.28125},{x:.40625,y:.28125},{x:.46875,y:.28125},{x:.46875,y:.28125},{x:.53125,y:.28125},{x:.53125,y:.28125},{x:.59375,y:.28125},{x:.59375,y:.28125},{x:.65625,y:.28125},{x:.65625,y:.28125},{x:.71875,y:.28125},{x:.71875,y:.28125},{x:.78125,y:.28125},{x:.78125,y:.28125},{x:.84375,y:.28125},{x:.84375,y:.28125},{x:.90625,y:.28125},{x:.90625,y:.28125},{x:.96875,y:.28125},{x:.96875,y:.28125},{x:.03125,y:.34375},{x:.03125,y:.34375},{x:.09375,y:.34375},{x:.09375,y:.34375},{x:.15625,y:.34375},{x:.15625,y:.34375},{x:.21875,y:.34375},{x:.21875,y:.34375},{x:.28125,y:.34375},{x:.28125,y:.34375},{x:.34375,y:.34375},{x:.34375,y:.34375},{x:.40625,y:.34375},{x:.40625,y:.34375},{x:.46875,y:.34375},{x:.46875,y:.34375},{x:.53125,y:.34375},{x:.53125,y:.34375},{x:.59375,y:.34375},{x:.59375,y:.34375},{x:.65625,y:.34375},{x:.65625,y:.34375},{x:.71875,y:.34375},{x:.71875,y:.34375},{x:.78125,y:.34375},{x:.78125,y:.34375},{x:.84375,y:.34375},{x:.84375,y:.34375},{x:.90625,y:.34375},{x:.90625,y:.34375},{x:.96875,y:.34375},{x:.96875,y:.34375},{x:.03125,y:.40625},{x:.03125,y:.40625},{x:.09375,y:.40625},{x:.09375,y:.40625},{x:.15625,y:.40625},{x:.15625,y:.40625},{x:.21875,y:.40625},{x:.21875,y:.40625},{x:.28125,y:.40625},{x:.28125,y:.40625},{x:.34375,y:.40625},{x:.34375,y:.40625},{x:.40625,y:.40625},{x:.40625,y:.40625},{x:.46875,y:.40625},{x:.46875,y:.40625},{x:.53125,y:.40625},{x:.53125,y:.40625},{x:.59375,y:.40625},{x:.59375,y:.40625},{x:.65625,y:.40625},{x:.65625,y:.40625},{x:.71875,y:.40625},{x:.71875,y:.40625},{x:.78125,y:.40625},{x:.78125,y:.40625},{x:.84375,y:.40625},{x:.84375,y:.40625},{x:.90625,y:.40625},{x:.90625,y:.40625},{x:.96875,y:.40625},{x:.96875,y:.40625},{x:.03125,y:.46875},{x:.03125,y:.46875},{x:.09375,y:.46875},{x:.09375,y:.46875},{x:.15625,y:.46875},{x:.15625,y:.46875},{x:.21875,y:.46875},{x:.21875,y:.46875},{x:.28125,y:.46875},{x:.28125,y:.46875},{x:.34375,y:.46875},{x:.34375,y:.46875},{x:.40625,y:.46875},{x:.40625,y:.46875},{x:.46875,y:.46875},{x:.46875,y:.46875},{x:.53125,y:.46875},{x:.53125,y:.46875},{x:.59375,y:.46875},{x:.59375,y:.46875},{x:.65625,y:.46875},{x:.65625,y:.46875},{x:.71875,y:.46875},{x:.71875,y:.46875},{x:.78125,y:.46875},{x:.78125,y:.46875},{x:.84375,y:.46875},{x:.84375,y:.46875},{x:.90625,y:.46875},{x:.90625,y:.46875},{x:.96875,y:.46875},{x:.96875,y:.46875},{x:.03125,y:.53125},{x:.03125,y:.53125},{x:.09375,y:.53125},{x:.09375,y:.53125},{x:.15625,y:.53125},{x:.15625,y:.53125},{x:.21875,y:.53125},{x:.21875,y:.53125},{x:.28125,y:.53125},{x:.28125,y:.53125},{x:.34375,y:.53125},{x:.34375,y:.53125},{x:.40625,y:.53125},{x:.40625,y:.53125},{x:.46875,y:.53125},{x:.46875,y:.53125},{x:.53125,y:.53125},{x:.53125,y:.53125},{x:.59375,y:.53125},{x:.59375,y:.53125},{x:.65625,y:.53125},{x:.65625,y:.53125},{x:.71875,y:.53125},{x:.71875,y:.53125},{x:.78125,y:.53125},{x:.78125,y:.53125},{x:.84375,y:.53125},{x:.84375,y:.53125},{x:.90625,y:.53125},{x:.90625,y:.53125},{x:.96875,y:.53125},{x:.96875,y:.53125},{x:.03125,y:.59375},{x:.03125,y:.59375},{x:.09375,y:.59375},{x:.09375,y:.59375},{x:.15625,y:.59375},{x:.15625,y:.59375},{x:.21875,y:.59375},{x:.21875,y:.59375},{x:.28125,y:.59375},{x:.28125,y:.59375},{x:.34375,y:.59375},{x:.34375,y:.59375},{x:.40625,y:.59375},{x:.40625,y:.59375},{x:.46875,y:.59375},{x:.46875,y:.59375},{x:.53125,y:.59375},{x:.53125,y:.59375},{x:.59375,y:.59375},{x:.59375,y:.59375},{x:.65625,y:.59375},{x:.65625,y:.59375},{x:.71875,y:.59375},{x:.71875,y:.59375},{x:.78125,y:.59375},{x:.78125,y:.59375},{x:.84375,y:.59375},{x:.84375,y:.59375},{x:.90625,y:.59375},{x:.90625,y:.59375},{x:.96875,y:.59375},{x:.96875,y:.59375},{x:.03125,y:.65625},{x:.03125,y:.65625},{x:.09375,y:.65625},{x:.09375,y:.65625},{x:.15625,y:.65625},{x:.15625,y:.65625},{x:.21875,y:.65625},{x:.21875,y:.65625},{x:.28125,y:.65625},{x:.28125,y:.65625},{x:.34375,y:.65625},{x:.34375,y:.65625},{x:.40625,y:.65625},{x:.40625,y:.65625},{x:.46875,y:.65625},{x:.46875,y:.65625},{x:.53125,y:.65625},{x:.53125,y:.65625},{x:.59375,y:.65625},{x:.59375,y:.65625},{x:.65625,y:.65625},{x:.65625,y:.65625},{x:.71875,y:.65625},{x:.71875,y:.65625},{x:.78125,y:.65625},{x:.78125,y:.65625},{x:.84375,y:.65625},{x:.84375,y:.65625},{x:.90625,y:.65625},{x:.90625,y:.65625},{x:.96875,y:.65625},{x:.96875,y:.65625},{x:.03125,y:.71875},{x:.03125,y:.71875},{x:.09375,y:.71875},{x:.09375,y:.71875},{x:.15625,y:.71875},{x:.15625,y:.71875},{x:.21875,y:.71875},{x:.21875,y:.71875},{x:.28125,y:.71875},{x:.28125,y:.71875},{x:.34375,y:.71875},{x:.34375,y:.71875},{x:.40625,y:.71875},{x:.40625,y:.71875},{x:.46875,y:.71875},{x:.46875,y:.71875},{x:.53125,y:.71875},{x:.53125,y:.71875},{x:.59375,y:.71875},{x:.59375,y:.71875},{x:.65625,y:.71875},{x:.65625,y:.71875},{x:.71875,y:.71875},{x:.71875,y:.71875},{x:.78125,y:.71875},{x:.78125,y:.71875},{x:.84375,y:.71875},{x:.84375,y:.71875},{x:.90625,y:.71875},{x:.90625,y:.71875},{x:.96875,y:.71875},{x:.96875,y:.71875},{x:.03125,y:.78125},{x:.03125,y:.78125},{x:.09375,y:.78125},{x:.09375,y:.78125},{x:.15625,y:.78125},{x:.15625,y:.78125},{x:.21875,y:.78125},{x:.21875,y:.78125},{x:.28125,y:.78125},{x:.28125,y:.78125},{x:.34375,y:.78125},{x:.34375,y:.78125},{x:.40625,y:.78125},{x:.40625,y:.78125},{x:.46875,y:.78125},{x:.46875,y:.78125},{x:.53125,y:.78125},{x:.53125,y:.78125},{x:.59375,y:.78125},{x:.59375,y:.78125},{x:.65625,y:.78125},{x:.65625,y:.78125},{x:.71875,y:.78125},{x:.71875,y:.78125},{x:.78125,y:.78125},{x:.78125,y:.78125},{x:.84375,y:.78125},{x:.84375,y:.78125},{x:.90625,y:.78125},{x:.90625,y:.78125},{x:.96875,y:.78125},{x:.96875,y:.78125},{x:.03125,y:.84375},{x:.03125,y:.84375},{x:.09375,y:.84375},{x:.09375,y:.84375},{x:.15625,y:.84375},{x:.15625,y:.84375},{x:.21875,y:.84375},{x:.21875,y:.84375},{x:.28125,y:.84375},{x:.28125,y:.84375},{x:.34375,y:.84375},{x:.34375,y:.84375},{x:.40625,y:.84375},{x:.40625,y:.84375},{x:.46875,y:.84375},{x:.46875,y:.84375},{x:.53125,y:.84375},{x:.53125,y:.84375},{x:.59375,y:.84375},{x:.59375,y:.84375},{x:.65625,y:.84375},{x:.65625,y:.84375},{x:.71875,y:.84375},{x:.71875,y:.84375},{x:.78125,y:.84375},{x:.78125,y:.84375},{x:.84375,y:.84375},{x:.84375,y:.84375},{x:.90625,y:.84375},{x:.90625,y:.84375},{x:.96875,y:.84375},{x:.96875,y:.84375},{x:.03125,y:.90625},{x:.03125,y:.90625},{x:.09375,y:.90625},{x:.09375,y:.90625},{x:.15625,y:.90625},{x:.15625,y:.90625},{x:.21875,y:.90625},{x:.21875,y:.90625},{x:.28125,y:.90625},{x:.28125,y:.90625},{x:.34375,y:.90625},{x:.34375,y:.90625},{x:.40625,y:.90625},{x:.40625,y:.90625},{x:.46875,y:.90625},{x:.46875,y:.90625},{x:.53125,y:.90625},{x:.53125,y:.90625},{x:.59375,y:.90625},{x:.59375,y:.90625},{x:.65625,y:.90625},{x:.65625,y:.90625},{x:.71875,y:.90625},{x:.71875,y:.90625},{x:.78125,y:.90625},{x:.78125,y:.90625},{x:.84375,y:.90625},{x:.84375,y:.90625},{x:.90625,y:.90625},{x:.90625,y:.90625},{x:.96875,y:.90625},{x:.96875,y:.90625},{x:.03125,y:.96875},{x:.03125,y:.96875},{x:.09375,y:.96875},{x:.09375,y:.96875},{x:.15625,y:.96875},{x:.15625,y:.96875},{x:.21875,y:.96875},{x:.21875,y:.96875},{x:.28125,y:.96875},{x:.28125,y:.96875},{x:.34375,y:.96875},{x:.34375,y:.96875},{x:.40625,y:.96875},{x:.40625,y:.96875},{x:.46875,y:.96875},{x:.46875,y:.96875},{x:.53125,y:.96875},{x:.53125,y:.96875},{x:.59375,y:.96875},{x:.59375,y:.96875},{x:.65625,y:.96875},{x:.65625,y:.96875},{x:.71875,y:.96875},{x:.71875,y:.96875},{x:.78125,y:.96875},{x:.78125,y:.96875},{x:.84375,y:.96875},{x:.84375,y:.96875},{x:.90625,y:.96875},{x:.90625,y:.96875},{x:.96875,y:.96875},{x:.96875,y:.96875},{x:.0625,y:.0625},{x:.0625,y:.0625},{x:.0625,y:.0625},{x:.0625,y:.0625},{x:.0625,y:.0625},{x:.0625,y:.0625},{x:.1875,y:.0625},{x:.1875,y:.0625},{x:.1875,y:.0625},{x:.1875,y:.0625},{x:.1875,y:.0625},{x:.1875,y:.0625},{x:.3125,y:.0625},{x:.3125,y:.0625},{x:.3125,y:.0625},{x:.3125,y:.0625},{x:.3125,y:.0625},{x:.3125,y:.0625},{x:.4375,y:.0625},{x:.4375,y:.0625},{x:.4375,y:.0625},{x:.4375,y:.0625},{x:.4375,y:.0625},{x:.4375,y:.0625},{x:.5625,y:.0625},{x:.5625,y:.0625},{x:.5625,y:.0625},{x:.5625,y:.0625},{x:.5625,y:.0625},{x:.5625,y:.0625},{x:.6875,y:.0625},{x:.6875,y:.0625},{x:.6875,y:.0625},{x:.6875,y:.0625},{x:.6875,y:.0625},{x:.6875,y:.0625},{x:.8125,y:.0625},{x:.8125,y:.0625},{x:.8125,y:.0625},{x:.8125,y:.0625},{x:.8125,y:.0625},{x:.8125,y:.0625},{x:.9375,y:.0625},{x:.9375,y:.0625},{x:.9375,y:.0625},{x:.9375,y:.0625},{x:.9375,y:.0625},{x:.9375,y:.0625},{x:.0625,y:.1875},{x:.0625,y:.1875},{x:.0625,y:.1875},{x:.0625,y:.1875},{x:.0625,y:.1875},{x:.0625,y:.1875},{x:.1875,y:.1875},{x:.1875,y:.1875},{x:.1875,y:.1875},{x:.1875,y:.1875},{x:.1875,y:.1875},{x:.1875,y:.1875},{x:.3125,y:.1875},{x:.3125,y:.1875},{x:.3125,y:.1875},{x:.3125,y:.1875},{x:.3125,y:.1875},{x:.3125,y:.1875},{x:.4375,y:.1875},{x:.4375,y:.1875},{x:.4375,y:.1875},{x:.4375,y:.1875},{x:.4375,y:.1875},{x:.4375,y:.1875},{x:.5625,y:.1875},{x:.5625,y:.1875},{x:.5625,y:.1875},{x:.5625,y:.1875},{x:.5625,y:.1875},{x:.5625,y:.1875},{x:.6875,y:.1875},{x:.6875,y:.1875},{x:.6875,y:.1875},{x:.6875,y:.1875},{x:.6875,y:.1875},{x:.6875,y:.1875},{x:.8125,y:.1875},{x:.8125,y:.1875},{x:.8125,y:.1875},{x:.8125,y:.1875},{x:.8125,y:.1875},{x:.8125,y:.1875},{x:.9375,y:.1875},{x:.9375,y:.1875},{x:.9375,y:.1875},{x:.9375,y:.1875},{x:.9375,y:.1875},{x:.9375,y:.1875},{x:.0625,y:.3125},{x:.0625,y:.3125},{x:.0625,y:.3125},{x:.0625,y:.3125},{x:.0625,y:.3125},{x:.0625,y:.3125},{x:.1875,y:.3125},{x:.1875,y:.3125},{x:.1875,y:.3125},{x:.1875,y:.3125},{x:.1875,y:.3125},{x:.1875,y:.3125},{x:.3125,y:.3125},{x:.3125,y:.3125},{x:.3125,y:.3125},{x:.3125,y:.3125},{x:.3125,y:.3125},{x:.3125,y:.3125},{x:.4375,y:.3125},{x:.4375,y:.3125},{x:.4375,y:.3125},{x:.4375,y:.3125},{x:.4375,y:.3125},{x:.4375,y:.3125},{x:.5625,y:.3125},{x:.5625,y:.3125},{x:.5625,y:.3125},{x:.5625,y:.3125},{x:.5625,y:.3125},{x:.5625,y:.3125},{x:.6875,y:.3125},{x:.6875,y:.3125},{x:.6875,y:.3125},{x:.6875,y:.3125},{x:.6875,y:.3125},{x:.6875,y:.3125},{x:.8125,y:.3125},{x:.8125,y:.3125},{x:.8125,y:.3125},{x:.8125,y:.3125},{x:.8125,y:.3125},{x:.8125,y:.3125},{x:.9375,y:.3125},{x:.9375,y:.3125},{x:.9375,y:.3125},{x:.9375,y:.3125},{x:.9375,y:.3125},{x:.9375,y:.3125},{x:.0625,y:.4375},{x:.0625,y:.4375},{x:.0625,y:.4375},{x:.0625,y:.4375},{x:.0625,y:.4375},{x:.0625,y:.4375},{x:.1875,y:.4375},{x:.1875,y:.4375},{x:.1875,y:.4375},{x:.1875,y:.4375},{x:.1875,y:.4375},{x:.1875,y:.4375},{x:.3125,y:.4375},{x:.3125,y:.4375},{x:.3125,y:.4375},{x:.3125,y:.4375},{x:.3125,y:.4375},{x:.3125,y:.4375},{x:.4375,y:.4375},{x:.4375,y:.4375},{x:.4375,y:.4375},{x:.4375,y:.4375},{x:.4375,y:.4375},{x:.4375,y:.4375},{x:.5625,y:.4375},{x:.5625,y:.4375},{x:.5625,y:.4375},{x:.5625,y:.4375},{x:.5625,y:.4375},{x:.5625,y:.4375},{x:.6875,y:.4375},{x:.6875,y:.4375},{x:.6875,y:.4375},{x:.6875,y:.4375},{x:.6875,y:.4375},{x:.6875,y:.4375},{x:.8125,y:.4375},{x:.8125,y:.4375},{x:.8125,y:.4375},{x:.8125,y:.4375},{x:.8125,y:.4375},{x:.8125,y:.4375},{x:.9375,y:.4375},{x:.9375,y:.4375},{x:.9375,y:.4375},{x:.9375,y:.4375},{x:.9375,y:.4375},{x:.9375,y:.4375},{x:.0625,y:.5625},{x:.0625,y:.5625},{x:.0625,y:.5625},{x:.0625,y:.5625},{x:.0625,y:.5625},{x:.0625,y:.5625},{x:.1875,y:.5625},{x:.1875,y:.5625},{x:.1875,y:.5625},{x:.1875,y:.5625},{x:.1875,y:.5625},{x:.1875,y:.5625},{x:.3125,y:.5625},{x:.3125,y:.5625},{x:.3125,y:.5625},{x:.3125,y:.5625},{x:.3125,y:.5625},{x:.3125,y:.5625},{x:.4375,y:.5625},{x:.4375,y:.5625},{x:.4375,y:.5625},{x:.4375,y:.5625},{x:.4375,y:.5625},{x:.4375,y:.5625},{x:.5625,y:.5625},{x:.5625,y:.5625},{x:.5625,y:.5625},{x:.5625,y:.5625},{x:.5625,y:.5625},{x:.5625,y:.5625},{x:.6875,y:.5625},{x:.6875,y:.5625},{x:.6875,y:.5625},{x:.6875,y:.5625},{x:.6875,y:.5625},{x:.6875,y:.5625},{x:.8125,y:.5625},{x:.8125,y:.5625},{x:.8125,y:.5625},{x:.8125,y:.5625},{x:.8125,y:.5625},{x:.8125,y:.5625},{x:.9375,y:.5625},{x:.9375,y:.5625},{x:.9375,y:.5625},{x:.9375,y:.5625},{x:.9375,y:.5625},{x:.9375,y:.5625},{x:.0625,y:.6875},{x:.0625,y:.6875},{x:.0625,y:.6875},{x:.0625,y:.6875},{x:.0625,y:.6875},{x:.0625,y:.6875},{x:.1875,y:.6875},{x:.1875,y:.6875},{x:.1875,y:.6875},{x:.1875,y:.6875},{x:.1875,y:.6875},{x:.1875,y:.6875},{x:.3125,y:.6875},{x:.3125,y:.6875},{x:.3125,y:.6875},{x:.3125,y:.6875},{x:.3125,y:.6875},{x:.3125,y:.6875},{x:.4375,y:.6875},{x:.4375,y:.6875},{x:.4375,y:.6875},{x:.4375,y:.6875},{x:.4375,y:.6875},{x:.4375,y:.6875},{x:.5625,y:.6875},{x:.5625,y:.6875},{x:.5625,y:.6875},{x:.5625,y:.6875},{x:.5625,y:.6875},{x:.5625,y:.6875},{x:.6875,y:.6875},{x:.6875,y:.6875},{x:.6875,y:.6875},{x:.6875,y:.6875},{x:.6875,y:.6875},{x:.6875,y:.6875},{x:.8125,y:.6875},{x:.8125,y:.6875},{x:.8125,y:.6875},{x:.8125,y:.6875},{x:.8125,y:.6875},{x:.8125,y:.6875},{x:.9375,y:.6875},{x:.9375,y:.6875},{x:.9375,y:.6875},{x:.9375,y:.6875},{x:.9375,y:.6875},{x:.9375,y:.6875},{x:.0625,y:.8125},{x:.0625,y:.8125},{x:.0625,y:.8125},{x:.0625,y:.8125},{x:.0625,y:.8125},{x:.0625,y:.8125},{x:.1875,y:.8125},{x:.1875,y:.8125},{x:.1875,y:.8125},{x:.1875,y:.8125},{x:.1875,y:.8125},{x:.1875,y:.8125},{x:.3125,y:.8125},{x:.3125,y:.8125},{x:.3125,y:.8125},{x:.3125,y:.8125},{x:.3125,y:.8125},{x:.3125,y:.8125},{x:.4375,y:.8125},{x:.4375,y:.8125},{x:.4375,y:.8125},{x:.4375,y:.8125},{x:.4375,y:.8125},{x:.4375,y:.8125},{x:.5625,y:.8125},{x:.5625,y:.8125},{x:.5625,y:.8125},{x:.5625,y:.8125},{x:.5625,y:.8125},{x:.5625,y:.8125},{x:.6875,y:.8125},{x:.6875,y:.8125},{x:.6875,y:.8125},{x:.6875,y:.8125},{x:.6875,y:.8125},{x:.6875,y:.8125},{x:.8125,y:.8125},{x:.8125,y:.8125},{x:.8125,y:.8125},{x:.8125,y:.8125},{x:.8125,y:.8125},{x:.8125,y:.8125},{x:.9375,y:.8125},{x:.9375,y:.8125},{x:.9375,y:.8125},{x:.9375,y:.8125},{x:.9375,y:.8125},{x:.9375,y:.8125},{x:.0625,y:.9375},{x:.0625,y:.9375},{x:.0625,y:.9375},{x:.0625,y:.9375},{x:.0625,y:.9375},{x:.0625,y:.9375},{x:.1875,y:.9375},{x:.1875,y:.9375},{x:.1875,y:.9375},{x:.1875,y:.9375},{x:.1875,y:.9375},{x:.1875,y:.9375},{x:.3125,y:.9375},{x:.3125,y:.9375},{x:.3125,y:.9375},{x:.3125,y:.9375},{x:.3125,y:.9375},{x:.3125,y:.9375},{x:.4375,y:.9375},{x:.4375,y:.9375},{x:.4375,y:.9375},{x:.4375,y:.9375},{x:.4375,y:.9375},{x:.4375,y:.9375},{x:.5625,y:.9375},{x:.5625,y:.9375},{x:.5625,y:.9375},{x:.5625,y:.9375},{x:.5625,y:.9375},{x:.5625,y:.9375},{x:.6875,y:.9375},{x:.6875,y:.9375},{x:.6875,y:.9375},{x:.6875,y:.9375},{x:.6875,y:.9375},{x:.6875,y:.9375},{x:.8125,y:.9375},{x:.8125,y:.9375},{x:.8125,y:.9375},{x:.8125,y:.9375},{x:.8125,y:.9375},{x:.8125,y:.9375},{x:.9375,y:.9375},{x:.9375,y:.9375},{x:.9375,y:.9375},{x:.9375,y:.9375},{x:.9375,y:.9375},{x:.9375,y:.9375}];var e1=class{constructor(t){ge(this,"model");ge(this,"anchors");ge(this,"anchorsTensor");ge(this,"inputSize");ge(this,"inputSizeTensor");ge(this,"doubleInputSizeTensor");this.model=t,this.anchors=nE.map(n=>[n.x,n.y]),this.anchorsTensor=ir(this.anchors),this.inputSize=this.model&&this.model.inputs&&this.model.inputs[0].shape?this.model.inputs[0].shape[2]:0,this.inputSizeTensor=Ft([this.inputSize,this.inputSize]),this.doubleInputSizeTensor=Ft([this.inputSize*2,this.inputSize*2])}normalizeBoxes(t){let n={};n.boxOffsets=Oe(t,[0,0],[-1,2]),n.boxSizes=Oe(t,[0,2],[-1,2]),n.div=he(n.boxOffsets,this.inputSizeTensor),n.boxCenterPoints=ce(n.div,this.anchorsTensor),n.halfBoxSizes=he(n.boxSizes,this.doubleInputSizeTensor),n.sub=fe(n.boxCenterPoints,n.halfBoxSizes),n.startPoints=L(n.sub,this.inputSizeTensor),n.add=ce(n.boxCenterPoints,n.halfBoxSizes),n.endPoints=L(n.add,this.inputSizeTensor);let s=Kl([n.startPoints,n.endPoints],1);return Object.keys(n).forEach(r=>ee(n[r])),s}normalizeLandmarks(t,n){let s={};s.reshape=U(t,[-1,7,2]),s.div=he(s.reshape,this.inputSizeTensor),s.landmarks=ce(s.div,this.anchors[n]);let r=L(s.landmarks,this.inputSizeTensor);return Object.keys(s).forEach(a=>ee(s[a])),r}async predict(t,n){let s={};s.resize=Se.resizeBilinear(t,[this.inputSize,this.inputSize]),s.div=he(s.resize,rt.tf127),s.image=fe(s.div,rt.tf1),s.batched=this.model.execute(s.image),s.predictions=st(s.batched),s.slice=Oe(s.predictions,[0,0],[-1,1]),s.sigmoid=Cn(s.slice),s.scores=st(s.sigmoid);let r=await s.scores.data();s.boxes=Oe(s.predictions,[0,1],[-1,4]),s.norm=this.normalizeBoxes(s.boxes),s.nms=await Se.nonMaxSuppressionAsync(s.norm,s.scores,3*n.hand.maxDetected,n.hand.iouThreshold,n.hand.minConfidence);let a=await s.nms.array(),o=[];for(let i of a){let l={};l.box=Oe(s.norm,[i,0],[1,-1]),l.slice=Oe(s.predictions,[i,5],[1,14]),l.norm=this.normalizeLandmarks(l.slice,i),l.palmLandmarks=U(l.norm,[-1,2]);let u=await l.box.data(),c=u.slice(0,2),p=u.slice(2,4),d=await l.palmLandmarks.array(),h={startPoint:c,endPoint:p,palmLandmarks:d,confidence:r[i]},f=JN(h,[t.shape[2]/this.inputSize,t.shape[1]/this.inputSize]);o.push(f),Object.keys(l).forEach(m=>ee(l[m]))}return Object.keys(s).forEach(i=>ee(s[i])),o}};var M5e=5,sE=1.65,rE=[0,5,9,13,17,1,2],z5e=0,L5e=2,aE=0,t1=class{constructor(t,n){ge(this,"handDetector");ge(this,"handPoseModel");ge(this,"inputSize");ge(this,"storedBoxes");ge(this,"skipped");ge(this,"detectedHands");this.handDetector=t,this.handPoseModel=n,this.inputSize=this.handPoseModel&&this.handPoseModel.inputs[0].shape?this.handPoseModel.inputs[0].shape[2]:0,this.storedBoxes=[],this.skipped=Number.MAX_SAFE_INTEGER,this.detectedHands=0}calculateLandmarksBoundingBox(t){let n=t.map(o=>o[0]),s=t.map(o=>o[1]),r=[Math.min(...n),Math.min(...s)],a=[Math.max(...n),Math.max(...s)];return{startPoint:r,endPoint:a}}getBoxForPalmLandmarks(t,n){let s=t.map(a=>n4([...a,1],n)),r=this.calculateLandmarksBoundingBox(s);return J2(Q2(r),M5e)}getBoxForHandLandmarks(t){let n=this.calculateLandmarksBoundingBox(t),s=J2(Q2(n),sE);s.palmLandmarks=[];for(let r=0;r<rE.length;r++)s.palmLandmarks.push(t[rE[r]].slice(0,2));return s}transformRawCoords(t,n,s,r){let a=Y2(n),o=[a[0]/this.inputSize,a[1]/this.inputSize,(a[0]+a[1])/this.inputSize/2],i=t.map(h=>[o[0]*(h[0]-this.inputSize/2),o[1]*(h[1]-this.inputSize/2),o[2]*h[2]]),l=t4(s,[0,0]),u=i.map(h=>[...n4(h,l),h[2]]),c=eE(r),p=[...Kh(n),1],d=[xi(p,c[0]),xi(p,c[1])];return u.map(h=>[Math.trunc(h[0]+d[0]),Math.trunc(h[1]+d[1]),Math.trunc(h[2])])}async estimateHands(t,n){let s=!1,r,a=(n.hand.skipTime||0)>ue()-aE,o=this.skipped<(n.hand.skipFrames||0);n.skipAllowed&&a&&o&&(r=await this.handDetector.predict(t,n),this.skipped=0),n.skipAllowed&&this.skipped++,r&&r.length>0&&(r.length!==this.detectedHands&&this.detectedHands!==n.hand.maxDetected||!n.hand.landmarks)&&(this.detectedHands=0,this.storedBoxes=[...r],this.storedBoxes.length>0&&(s=!0));let i=[];for(let l=0;l<this.storedBoxes.length;l++){let u=this.storedBoxes[l];if(!!u)if(n.hand.landmarks){let c=n.hand.rotation?QN(u.palmLandmarks[z5e],u.palmLandmarks[L5e]):0,p=Kh(u),d=[p[0]/t.shape[2],p[1]/t.shape[1]],h=n.hand.rotation&&pe.kernels.includes("rotatewithoffset")?Se.rotateWithOffset(t,c,0,d):t.clone(),f=t4(-c,p),m=s?this.getBoxForPalmLandmarks(u.palmLandmarks,f):u,g=YN(m,h,[this.inputSize,this.inputSize]),y=he(g,rt.tf255);ee(g),ee(h);let[x,A]=this.handPoseModel.execute(y);aE=ue(),ee(y);let b=(await x.data())[0];if(ee(x),b>=n.hand.minConfidence/4){let w=U(A,[-1,3]),k=await w.array();ee(A),ee(w);let S=this.transformRawCoords(k,m,c,f),E=this.getBoxForHandLandmarks(S);this.storedBoxes[l]={...E,confidence:b};let R={landmarks:S,confidence:b,boxConfidence:u.confidence,fingerConfidence:b,box:{topLeft:E.startPoint,bottomRight:E.endPoint}};i.push(R)}else this.storedBoxes[l]=null;ee(A)}else{let c=J2(Q2(u),sE),p={confidence:u.confidence,boxConfidence:u.confidence,fingerConfidence:0,box:{topLeft:c.startPoint,bottomRight:c.endPoint},landmarks:[]};i.push(p)}}return this.storedBoxes=this.storedBoxes.filter(l=>l!==null),this.detectedHands=i.length,i.length>n.hand.maxDetected&&(i.length=n.hand.maxDetected),i}};var cs={thumb:0,index:1,middle:2,ring:3,pinky:4,all:[0,1,2,3,4],nameMapping:{0:"thumb",1:"index",2:"middle",3:"ring",4:"pinky"},pointsMapping:{0:[[0,1],[1,2],[2,3],[3,4]],1:[[0,5],[5,6],[6,7],[7,8]],2:[[0,9],[9,10],[10,11],[11,12]],3:[[0,13],[13,14],[14,15],[15,16]],4:[[0,17],[17,18],[18,19],[19,20]]},getName:e=>cs.nameMapping[e],getPoints:e=>cs.pointsMapping[e]},vi={none:0,half:1,full:2,nameMapping:{0:"none",1:"half",2:"full"},getName:e=>vi.nameMapping[e]},jt={verticalUp:0,verticalDown:1,horizontalLeft:2,horizontalRight:3,diagonalUpRight:4,diagonalUpLeft:5,diagonalDownRight:6,diagonalDownLeft:7,nameMapping:{0:"verticalUp",1:"verticalDown",2:"horizontalLeft",3:"horizontalRight",4:"diagonalUpRight",5:"diagonalUpLeft",6:"diagonalDownRight",7:"diagonalDownLeft"},getName:e=>jt.nameMapping[e]},bi=class{constructor(t){ge(this,"name");ge(this,"curls");ge(this,"directions");ge(this,"weights");ge(this,"weightsRelative");this.name=t,this.curls={},this.directions={},this.weights=[1,1,1,1,1],this.weightsRelative=[1,1,1,1,1]}curl(t,n,s){typeof this.curls[t]=="undefined"&&(this.curls[t]=[]),this.curls[t].push([n,s])}direction(t,n,s){this.directions[t]||(this.directions[t]=[]),this.directions[t].push([n,s])}weight(t,n){this.weights[t]=n;let s=this.weights.reduce((r,a)=>r+a,0);this.weightsRelative=this.weights.map(r=>r*5/s)}matchAgainst(t,n){let s=0;for(let r in t){let a=t[r],o=this.curls[r];if(typeof o=="undefined"){s+=this.weightsRelative[r];continue}for(let[i,l]of o)if(a===i){s+=l*this.weightsRelative[r];break}}for(let r in n){let a=n[r],o=this.directions[r];if(typeof o=="undefined"){s+=this.weightsRelative[r];continue}for(let[i,l]of o)if(a===i){s+=l*this.weightsRelative[r];break}}return s/10}};var{thumb:Mr,index:_a,middle:Da,ring:fu,pinky:mu}=cs,{none:zr,half:W5e,full:Lr}=vi,{verticalUp:kd,verticalDown:Gke,horizontalLeft:s4,horizontalRight:V5e,diagonalUpRight:U5e,diagonalUpLeft:Id,diagonalDownRight:Hke,diagonalDownLeft:jke}=jt,wi=new bi("thumbs up");wi.curl(Mr,zr,1);wi.direction(Mr,kd,1);wi.direction(Mr,Id,.25);wi.direction(Mr,U5e,.25);for(let e of[cs.index,cs.middle,cs.ring,cs.pinky])wi.curl(e,Lr,1),wi.direction(e,s4,1),wi.direction(e,V5e,1);var dn=new bi("victory");dn.curl(Mr,W5e,.5);dn.curl(Mr,zr,.5);dn.direction(Mr,kd,1);dn.direction(Mr,Id,1);dn.curl(_a,zr,1);dn.direction(_a,kd,.75);dn.direction(_a,Id,1);dn.curl(Da,zr,1);dn.direction(Da,kd,1);dn.direction(Da,Id,.75);dn.curl(fu,Lr,1);dn.direction(fu,kd,.2);dn.direction(fu,Id,1);dn.direction(fu,s4,.2);dn.curl(mu,Lr,1);dn.direction(mu,kd,.2);dn.direction(mu,Id,1);dn.direction(mu,s4,.2);dn.weight(_a,2);dn.weight(Da,2);var ki=new bi("point");ki.curl(Mr,Lr,1);ki.curl(_a,zr,.5);ki.curl(Da,Lr,.5);ki.curl(fu,Lr,.5);ki.curl(mu,Lr,.5);ki.weight(_a,2);ki.weight(Da,2);var Ii=new bi("middle finger");Ii.curl(Mr,zr,1);Ii.curl(_a,Lr,.5);Ii.curl(Da,Lr,.5);Ii.curl(fu,Lr,.5);Ii.curl(mu,Lr,.5);Ii.weight(_a,2);Ii.weight(Da,2);var Sd=new bi("open palm");Sd.curl(Mr,zr,.75);Sd.curl(_a,zr,.75);Sd.curl(Da,zr,.75);Sd.curl(fu,zr,.75);Sd.curl(mu,zr,.75);var oE=[wi,dn,ki,Ii,Sd];var G5e=.7,gu={HALF_CURL_START_LIMIT:60,NO_CURL_START_LIMIT:130,DISTANCE_VOTE_POWER:1.1,SINGLE_ANGLE_VOTE_POWER:.9,TOTAL_ANGLE_VOTE_POWER:1.6};function iE(e,t,n,s){let r=(t-s)/(e-n),a=Math.atan(r)*180/Math.PI;return a<=0?a=-a:a>0&&(a=180-a),a}function uE(e,t){if(!e||!t)return[0,0];let n=iE(e[0],e[1],t[0],t[1]);if(e.length===2)return n;let s=iE(e[1],e[2],t[1],t[2]);return[n,s]}function lE(e,t=1){let n=0,s=0,r=0;return e>=75&&e<=105?n=1*t:e>=25&&e<=155?s=1*t:r=1*t,[n,s,r]}function H5e(e,t,n){let s=e[0]-t[0],r=e[0]-n[0],a=t[0]-n[0],o=e[1]-t[1],i=e[1]-n[1],l=t[1]-n[1],u=e[2]-t[2],c=e[2]-n[2],p=t[2]-n[2],d=Math.sqrt(s*s+o*o+u*u),h=Math.sqrt(r*r+i*i+c*c),f=Math.sqrt(a*a+l*l+p*p),m=(f*f+d*d-h*h)/(2*f*d);m>1?m=1:m<-1&&(m=-1);let g=Math.acos(m);g=57.2958*g%180;let y;return g>gu.NO_CURL_START_LIMIT?y=vi.none:g>gu.HALF_CURL_START_LIMIT?y=vi.half:y=vi.full,y}function cE(e,t,n,s){let r;return s===Math.abs(e)?e>0?r=jt.horizontalLeft:r=jt.horizontalRight:s===Math.abs(t)?t>0?r=jt.horizontalLeft:r=jt.horizontalRight:n>0?r=jt.horizontalLeft:r=jt.horizontalRight,r}function dE(e,t,n,s){let r;return s===Math.abs(e)?e<0?r=jt.verticalDown:r=jt.verticalUp:s===Math.abs(t)?t<0?r=jt.verticalDown:r=jt.verticalUp:n<0?r=jt.verticalDown:r=jt.verticalUp,r}function j5e(e,t,n,s,r,a,o,i){let l,u=dE(e,t,n,s),c=cE(r,a,o,i);return u===jt.verticalUp?c===jt.horizontalLeft?l=jt.diagonalUpLeft:l=jt.diagonalUpRight:c===jt.horizontalLeft?l=jt.diagonalDownLeft:l=jt.diagonalDownRight,l}function q5e(e,t,n,s){let r=e[0]-t[0],a=e[0]-n[0],o=t[0]-n[0],i=e[1]-t[1],l=e[1]-n[1],u=t[1]-n[1],c=Math.max(Math.abs(r),Math.abs(a),Math.abs(o)),p=Math.max(Math.abs(i),Math.abs(l),Math.abs(u)),d=0,h=0,f=0,m=p/(c+1e-5);m>1.5?d+=gu.DISTANCE_VOTE_POWER:m>.66?h+=gu.DISTANCE_VOTE_POWER:f+=gu.DISTANCE_VOTE_POWER;let g=Math.sqrt(r*r+i*i),y=Math.sqrt(a*a+l*l),x=Math.sqrt(o*o+u*u),A=Math.max(g,y,x),b=e[0],w=e[1],k=n[0],S=n[1];A===g?(k=n[0],S=n[1]):A===x&&(b=t[0],w=t[1]);let $=uE([b,w],[k,S]),_=lE($,gu.TOTAL_ANGLE_VOTE_POWER);d+=_[0],h+=_[1],f+=_[2];for(let C of s){let F=lE(C,gu.SINGLE_ANGLE_VOTE_POWER);d+=F[0],h+=F[1],f+=F[2]}let P;return d===Math.max(d,h,f)?P=dE(l,i,u,p):f===Math.max(h,f)?P=cE(a,r,o,c):P=j5e(l,i,u,p,a,r,o,c),P}function pE(e){let t=[],n=[],s=[],r=[];if(!e)return{curls:s,directions:r};for(let a of cs.all){let o=cs.getPoints(a),i=[],l=[];for(let u of o){let c=e[u[0]],p=e[u[1]],d=uE(c,p),h=d[0],f=d[1];i.push(h),l.push(f)}t.push(i),n.push(l)}for(let a of cs.all){let o=a===cs.thumb?1:0,i=cs.getPoints(a),l=e[i[o][0]],u=e[i[o+1][1]],c=e[i[3][1]],p=H5e(l,u,c),d=q5e(l,u,c,t[a].slice(o));s[a]=p,r[a]=d}return{curls:s,directions:r}}function n1(e){if(!e||e.length===0)return null;let t=pE(e),n={};for(let s of cs.all)n[cs.getName(s)]={curl:vi.getName(t.curls[s]),direction:jt.getName(t.directions[s])};return n}function hE(e){let t=[];if(!e||e.length===0)return t;let n=pE(e);for(let s of oE){let r=s.matchAgainst(n.curls,n.directions);r>=G5e&&t.push({name:s.name,confidence:r})}return t}var fE={thumb:[1,2,3,4],index:[5,6,7,8],middle:[9,10,11,12],ring:[13,14,15,16],pinky:[17,18,19,20],palm:[0]},Cd,Td,mE;async function a4(e,t){let n=await mE.estimateHands(e,t);if(!n)return[];let s=[];for(let r=0;r<n.length;r++){let a={};if(n[r].landmarks)for(let c of Object.keys(fE))a[c]=fE[c].map(p=>n[r].landmarks[p]);let o=n[r].landmarks,i=[Number.MAX_SAFE_INTEGER,Number.MAX_SAFE_INTEGER,0,0],l=[0,0,0,0];if(o&&o.length>0){for(let c of o)c[0]<i[0]&&(i[0]=c[0]),c[1]<i[1]&&(i[1]=c[1]),c[0]>i[2]&&(i[2]=c[0]),c[1]>i[3]&&(i[3]=c[1]);i[2]-=i[0],i[3]-=i[1],l=[i[0]/(e.shape[2]||0),i[1]/(e.shape[1]||0),i[2]/(e.shape[2]||0),i[3]/(e.shape[1]||0)]}else i=n[r].box?[Math.trunc(Math.max(0,n[r].box.topLeft[0])),Math.trunc(Math.max(0,n[r].box.topLeft[1])),Math.trunc(Math.min(e.shape[2]||0,n[r].box.bottomRight[0])-Math.max(0,n[r].box.topLeft[0])),Math.trunc(Math.min(e.shape[1]||0,n[r].box.bottomRight[1])-Math.max(0,n[r].box.topLeft[1]))]:[0,0,0,0],l=[n[r].box.topLeft[0]/(e.shape[2]||0),n[r].box.topLeft[1]/(e.shape[1]||0),(n[r].box.bottomRight[0]-n[r].box.topLeft[0])/(e.shape[2]||0),(n[r].box.bottomRight[1]-n[r].box.topLeft[1])/(e.shape[1]||0)];let u=n1(o);s.push({id:r,score:Math.round(100*n[r].confidence)/100,boxScore:Math.round(100*n[r].boxConfidence)/100,fingerScore:Math.round(100*n[r].fingerConfidence)/100,label:"hand",box:i,boxRaw:l,keypoints:o,annotations:a,landmarks:u})}return s}async function o4(e){var n,s;pe.initial&&(Cd=null,Td=null),!Cd||!Td?[Cd,Td]=await Promise.all([e.hand.enabled?Ge((n=e.hand.detector)==null?void 0:n.modelPath):null,e.hand.landmarks?Ge((s=e.hand.skeleton)==null?void 0:s.modelPath):null]):(e.debug&&le("cached model:",Cd.modelUrl),e.debug&&le("cached model:",Td.modelUrl));let t=new e1(Cd);return mE=new t1(t,Td),[Cd,Td]}var An=[null,null],X5e=["StatefulPartitionedCall/Postprocessor/Slice","StatefulPartitionedCall/Postprocessor/ExpandDims_1"],Si=[[0,0],[0,0]],K5e=["hand","fist","pinch","point","face","tip","pinchtip"],yE=4,AE=1.6,Z5e=512,Y5e=1.4,s1=Number.MAX_SAFE_INTEGER,i4=0,$a=[0,0],Jt={boxes:[],hands:[]},xE={thumb:[1,2,3,4],index:[5,6,7,8],middle:[9,10,11,12],ring:[13,14,15,16],pinky:[17,18,19,20],base:[0],palm:[0,17,13,9,5,1,0]};async function bE(e){var t;if(pe.initial&&(An[0]=null),An[0])e.debug&&le("cached model:",An[0].modelUrl);else{r1(["tensorlistreserve","enter","tensorlistfromtensor","merge","loopcond","switch","exit","tensorliststack","nextiteration","tensorlistsetitem","tensorlistgetitem","reciprocal","shape","split","where"],e),An[0]=await Ge((t=e.hand.detector)==null?void 0:t.modelPath);let n=Object.values(An[0].modelSignature.inputs);Si[0][0]=Array.isArray(n)?parseInt(n[0].tensorShape.dim[1].size):0,Si[0][1]=Array.isArray(n)?parseInt(n[0].tensorShape.dim[2].size):0}return An[0]}async function vE(e){var t;if(pe.initial&&(An[1]=null),An[1])e.debug&&le("cached model:",An[1].modelUrl);else{An[1]=await Ge((t=e.hand.skeleton)==null?void 0:t.modelPath);let n=Object.values(An[1].modelSignature.inputs);Si[1][0]=Array.isArray(n)?parseInt(n[0].tensorShape.dim[1].size):0,Si[1][1]=Array.isArray(n)?parseInt(n[0].tensorShape.dim[2].size):0}return An[1]}async function J5e(e,t){let n=[];if(!e||!An[0])return n;let s={},r=(e.shape[2]||1)/(e.shape[1]||1),a=Math.min(Math.round((e.shape[1]||0)/8)*8,Z5e),o=Math.round(a*r/8)*8;s.resize=Se.resizeBilinear(e,[a,o]),s.cast=ye(s.resize,"int32"),[s.rawScores,s.rawBoxes]=await An[0].executeAsync(s.cast,X5e),s.boxes=st(s.rawBoxes,[0,2]),s.scores=st(s.rawScores,[0]);let i=En(s.scores,1);ee(i[yE]),i.splice(yE,1),s.filtered=on(i,1),ee(i),s.max=mn(s.filtered,1),s.argmax=Es(s.filtered,1);let l=0;s.nms=await Se.nonMaxSuppressionAsync(s.boxes,s.max,(t.hand.maxDetected||0)+1,t.hand.iouThreshold||0,t.hand.minConfidence||1);let u=await s.nms.data(),c=await s.max.data(),p=await s.argmax.data();for(let d of Array.from(u)){let h=Oe(s.boxes,d,1),f=await h.data();ee(h);let m=[f[1],f[0],f[3]-f[1],f[2]-f[0]],g=G2(m,Y5e),y=[Math.trunc(m[0]*$a[0]),Math.trunc(m[1]*$a[1]),Math.trunc(m[2]*$a[0]),Math.trunc(m[3]*$a[1])],x=c[d],A=K5e[p[d]],b={id:l++,score:x,box:y,boxRaw:g,label:A};n.push(b)}return Object.keys(s).forEach(d=>ee(s[d])),n.sort((d,h)=>h.score-d.score),n.length>(t.hand.maxDetected||1)&&(n.length=t.hand.maxDetected||1),n}async function l4(e,t,n){let s={id:t.id,score:Math.round(100*t.score)/100,boxScore:Math.round(100*t.score)/100,fingerScore:0,box:t.box,boxRaw:t.boxRaw,label:t.label,keypoints:[],landmarks:{},annotations:{}};if(e&&An[1]&&n.hand.landmarks&&t.score>(n.hand.minConfidence||0)){let r={},a=[t.boxRaw[1],t.boxRaw[0],t.boxRaw[3]+t.boxRaw[1],t.boxRaw[2]+t.boxRaw[0]];r.crop=Se.cropAndResize(e,[a],[0],[Si[1][0],Si[1][1]],"bilinear"),r.div=he(r.crop,rt.tf255),[r.score,r.keypoints]=An[1].execute(r.div,["Identity_1","Identity"]);let o=(await r.score.data())[0],i=(100-Math.trunc(100/(1+Math.exp(o))))/100;if(i>=(n.hand.minConfidence||0)){s.fingerScore=i,r.reshaped=U(r.keypoints,[-1,3]);let c=(await r.reshaped.array()).map(p=>[p[0]/Si[1][1],p[1]/Si[1][0],p[2]||0]).map(p=>[p[0]*t.boxRaw[2],p[1]*t.boxRaw[3],p[2]||0]);s.keypoints=c.map(p=>[$a[0]*(p[0]+t.boxRaw[0]),$a[1]*(p[1]+t.boxRaw[1]),p[2]||0]),s.landmarks=n1(s.keypoints);for(let p of Object.keys(xE))s.annotations[p]=xE[p].map(d=>s.landmarks&&s.keypoints[d]?s.keypoints[d]:null)}Object.keys(r).forEach(l=>ee(r[l]))}return s}async function u4(e,t){var r,a;if(!An[0]||!An[1]||!((r=An[0])!=null&&r.inputs[0].shape)||!((a=An[1])!=null&&a.inputs[0].shape))return[];$a=[e.shape[2]||0,e.shape[1]||0],s1++;let n=(t.hand.skipTime||0)>ue()-i4,s=s1<(t.hand.skipFrames||0);return t.skipAllowed&&n&&s?Jt.hands:new Promise(async o=>{let i=3*(t.hand.skipTime||0)>ue()-i4,l=s1<3*(t.hand.skipFrames||0);t.skipAllowed&&Jt.hands.length===t.hand.maxDetected?Jt.hands=await Promise.all(Jt.boxes.map(c=>l4(e,c,t))):t.skipAllowed&&i&&l&&Jt.hands.length>0?Jt.hands=await Promise.all(Jt.boxes.map(c=>l4(e,c,t))):(Jt.boxes=await J5e(e,t),i4=ue(),Jt.hands=await Promise.all(Jt.boxes.map(c=>l4(e,c,t))),s1=0);let u=[...Jt.boxes];if(Jt.boxes.length=0,t.cacheSensitivity>0)for(let c=0;c<Jt.hands.length;c++){let p=iN(Jt.hands[c].keypoints,$a);if(p.box[2]/(e.shape[2]||1)>.05&&p.box[3]/(e.shape[1]||1)>.05&&Jt.hands[c].fingerScore&&Jt.hands[c].fingerScore>(t.hand.minConfidence||0)){let d=G2(p.box,AE),h=G2(p.boxRaw,AE);Jt.boxes.push({...u[c],box:d,boxRaw:h})}}for(let c=0;c<Jt.hands.length;c++){let p=Na(Jt.hands[c].keypoints,$a);Jt.hands[c].box=p.box,Jt.hands[c].boxRaw=p.boxRaw}o(Jt.hands)})}var Vn,a1=[],c4=Number.MAX_SAFE_INTEGER,kE=0,IE=0;async function SE(e){var t;return pe.initial&&(Vn=null),Vn?e.debug&&le("cached model:",Vn.modelUrl):Vn=await Ge((t=e.face.liveness)==null?void 0:t.modelPath),Vn}async function d4(e,t,n,s){var o,i;if(!Vn)return 0;let r=(((o=t.face.liveness)==null?void 0:o.skipTime)||0)>ue()-IE,a=c4<(((i=t.face.liveness)==null?void 0:i.skipFrames)||0);return t.skipAllowed&&r&&a&&kE===s&&a1[n]?(c4++,a1[n]):(c4=0,new Promise(async l=>{let u=Se.resizeBilinear(e,[Vn!=null&&Vn.inputs[0].shape?Vn.inputs[0].shape[2]:0,Vn!=null&&Vn.inputs[0].shape?Vn.inputs[0].shape[1]:0],!1),c=Vn==null?void 0:Vn.execute(u),p=(await c.data())[0];a1[n]=Math.round(100*p)/100,kE=s,IE=ue(),ee([u,c]),l(a1[n])}))}var Zh={};ia(Zh,{connected:()=>i1,horizontal:()=>p4,kpt:()=>o1,relative:()=>f4,vertical:()=>h4});var o1=["nose","leftEye","rightEye","leftEar","rightEar","leftShoulder","rightShoulder","leftElbow","rightElbow","leftWrist","rightWrist","leftHip","rightHip","leftKnee","rightKnee","leftAnkle","rightAnkle"],p4=[["leftEye","rightEye"],["leftEar","rightEar"],["leftShoulder","rightShoulder"],["leftElbow","rightElbow"],["leftWrist","rightWrist"],["leftHip","rightHip"],["leftKnee","rightKnee"],["leftAnkle","rightAnkle"]],h4=[["leftKnee","leftShoulder"],["rightKnee","rightShoulder"],["leftAnkle","leftKnee"],["rightAnkle","rightKnee"]],f4=[[["leftHip","rightHip"],["leftShoulder","rightShoulder"]],[["leftElbow","rightElbow"],["leftShoulder","rightShoulder"]]],i1={leftLeg:["leftHip","leftKnee","leftAnkle"],rightLeg:["rightHip","rightKnee","rightAnkle"],torso:["leftShoulder","rightShoulder","rightHip","leftHip","leftShoulder"],leftArm:["leftShoulder","leftElbow","leftWrist"],rightArm:["rightShoulder","rightElbow","rightWrist"],head:[]};var TE=.005,Ws={keypoints:[],padding:[[0,0],[0,0],[0,0],[0,0]]};function m4(e){for(let t of p4){let n=e.keypoints.findIndex(r=>r.part===t[0]),s=e.keypoints.findIndex(r=>r.part===t[1]);if(e.keypoints[n]&&e.keypoints[s]&&e.keypoints[n].position[0]<e.keypoints[s].position[0]){let r=e.keypoints[n];e.keypoints[n]=e.keypoints[s],e.keypoints[s]=r}}for(let t of h4){let n=e.keypoints.findIndex(r=>r&&r.part===t[0]),s=e.keypoints.findIndex(r=>r&&r.part===t[1]);e.keypoints[n]&&e.keypoints[s]&&e.keypoints[n].position[1]<e.keypoints[s].position[1]&&e.keypoints.splice(n,1)}for(let[t,n]of f4){let s=e.keypoints.findIndex(u=>u&&u.part===t[0]),r=e.keypoints.findIndex(u=>u&&u.part===t[1]),a=e.keypoints.findIndex(u=>u&&u.part===n[0]),o=e.keypoints.findIndex(u=>u&&u.part===n[1]);if(!e.keypoints[a]||!e.keypoints[o])continue;let i=e.keypoints[s]?[Math.abs(e.keypoints[a].position[0]-e.keypoints[s].position[0]),Math.abs(e.keypoints[o].position[0]-e.keypoints[s].position[0])]:[0,0],l=e.keypoints[r]?[Math.abs(e.keypoints[o].position[0]-e.keypoints[r].position[0]),Math.abs(e.keypoints[a].position[0]-e.keypoints[r].position[0])]:[0,0];if(i[0]>i[1]||l[0]>l[1]){let u=e.keypoints[s];e.keypoints[s]=e.keypoints[r],e.keypoints[r]=u}}}function NE(e){for(let t=0;t<e.length;t++)if(e[t]&&Ws.keypoints[t]){let n=[Math.abs(e[t].positionRaw[0]-Ws.keypoints[t].positionRaw[0]),Math.abs(e[t].positionRaw[1]-Ws.keypoints[t].positionRaw[1])];n[0]<TE&&n[1]<TE?e[t]=Ws.keypoints[t]:Ws.keypoints[t]=e[t]}else Ws.keypoints[t]=e[t];return e}function EE(e,t){let n={};if(!e.shape||!e.shape[1]||!e.shape[2])return e;Ws.padding=[[0,0],[e.shape[2]>e.shape[1]?Math.trunc((e.shape[2]-e.shape[1])/2):0,e.shape[2]>e.shape[1]?Math.trunc((e.shape[2]-e.shape[1])/2):0],[e.shape[1]>e.shape[2]?Math.trunc((e.shape[1]-e.shape[2])/2):0,e.shape[1]>e.shape[2]?Math.trunc((e.shape[1]-e.shape[2])/2):0],[0,0]],n.pad=Qs(e,Ws.padding),n.resize=Se.resizeBilinear(n.pad,[t,t]);let s=ye(n.resize,"int32");return Object.keys(n).forEach(r=>ee(n[r])),s}function RE(e,t){e.keypoints=e.keypoints.filter(s=>s&&s.position);for(let s of e.keypoints)s.position=[s.position[0]*(t[0]+Ws.padding[2][0]+Ws.padding[2][1])/t[0]-Ws.padding[2][0],s.position[1]*(t[1]+Ws.padding[1][0]+Ws.padding[1][1])/t[1]-Ws.padding[1][0]],s.positionRaw=[s.position[0]/t[0],s.position[1]/t[1]];let n=Na(e.keypoints.map(s=>s.position),t);return e.box=n.box,e.boxRaw=n.boxRaw,e}var Vs,l1=0,g4=Number.MAX_SAFE_INTEGER,yu={boxes:[],bodies:[],last:0};async function _E(e){return pe.initial&&(Vs=null),Vs?e.debug&&le("cached model:",Vs.modelUrl):(r1(["size"],e),Vs=await Ge(e.body.modelPath)),l1=Vs.inputs[0].shape?Vs.inputs[0].shape[2]:0,l1<64&&(l1=256),Vs}async function exe(e,t,n){let s=e[0][0],r=[],a=0;for(let c=0;c<s.length;c++)if(a=s[c][2],a>t.body.minConfidence){let p=[s[c][1],s[c][0]];r.push({score:Math.round(100*a)/100,part:o1[c],positionRaw:p,position:[Math.round((n.shape[2]||0)*p[0]),Math.round((n.shape[1]||0)*p[1])]})}a=r.reduce((c,p)=>p.score>c?p.score:c,0);let o=[],i=Na(r.map(c=>c.position),[n.shape[2],n.shape[1]]),l={};for(let[c,p]of Object.entries(i1)){let d=[];for(let h=0;h<p.length-1;h++){let f=r.find(g=>g.part===p[h]),m=r.find(g=>g.part===p[h+1]);f&&m&&f.score>(t.body.minConfidence||0)&&m.score>(t.body.minConfidence||0)&&d.push([f.position,m.position])}l[c]=d}let u={id:0,score:a,box:i.box,boxRaw:i.boxRaw,keypoints:r,annotations:l};return m4(u),o.push(u),o}async function txe(e,t,n){let s=[];for(let r=0;r<e[0].length;r++){let a=e[0][r],o=Math.round(100*a[51+4])/100;if(o>t.body.minConfidence){let i=[];for(let p=0;p<17;p++){let d=a[3*p+2];if(d>t.body.minConfidence){let h=[a[3*p+1],a[3*p+0]];i.push({part:o1[p],score:Math.round(100*d)/100,positionRaw:h,position:[Math.round((n.shape[2]||0)*h[0]),Math.round((n.shape[1]||0)*h[1])]})}}let l=Na(i.map(p=>p.position),[n.shape[2],n.shape[1]]),u={};for(let[p,d]of Object.entries(i1)){let h=[];for(let f=0;f<d.length-1;f++){let m=i.find(y=>y.part===d[f]),g=i.find(y=>y.part===d[f+1]);m&&g&&m.score>(t.body.minConfidence||0)&&g.score>(t.body.minConfidence||0)&&h.push([m.position,g.position])}u[p]=h}let c={id:r,score:o,box:l.box,boxRaw:l.boxRaw,keypoints:[...i],annotations:u};m4(c),s.push(c)}}return s.sort((r,a)=>a.score-r.score),s.length>t.body.maxDetected&&(s.length=t.body.maxDetected),s}async function y4(e,t){if(!Vs||!(Vs!=null&&Vs.inputs[0].shape))return[];t.skipAllowed||(yu.boxes.length=0),g4++;let n=(t.body.skipTime||0)>ue()-yu.last,s=g4<(t.body.skipFrames||0);return t.skipAllowed&&n&&s?yu.bodies:new Promise(async r=>{let a={};g4=0,a.input=EE(e,l1),a.res=Vs==null?void 0:Vs.execute(a.input),yu.last=ue();let o=await a.res.array();yu.bodies=a.res.shape[2]===17?await exe(o,t,e):await txe(o,t,e);for(let i of yu.bodies)RE(i,[e.shape[2]||1,e.shape[1]||1]),NE(i.keypoints);Object.keys(a).forEach(i=>ee(a[i])),r(yu.bodies)})}var Nd,u1=[],$E=0,A4=Number.MAX_SAFE_INTEGER,d1=0,c1=2.5;async function PE(e){if(!Nd||pe.initial){Nd=await Ge(e.object.modelPath);let t=Object.values(Nd.modelSignature.inputs);d1=Array.isArray(t)?parseInt(t[0].tensorShape.dim[2].size):0}else e.debug&&le("cached model:",Nd.modelUrl);return Nd}async function nxe(e,t,n){let s=0,r=[];for(let l of[1,2,4]){let u=l*13,c=st(e.find(g=>g.shape[1]===u**2&&(g.shape[2]||0)===bd.length)),p=await c.array(),d=st(e.find(g=>g.shape[1]===u**2&&(g.shape[2]||0)<bd.length)),h=d.reshape([-1,4,d.shape[1]/4]),f=h.argMax(2),m=await f.array();for(let g=0;g<c.shape[0];g++)for(let y=0;y<c.shape[1];y++){let x=p[g][y];if(x>(n.object.minConfidence||0)&&y!==61){let A=(.5+Math.trunc(g%u))/u,b=(.5+Math.trunc(g/u))/u,w=m[g].map(C=>C*(u/l/d1)),[k,S]=[A-c1/l*w[0],b-c1/l*w[1]],[E,R]=[A+c1/l*w[2]-k,b+c1/l*w[3]-S],$=[k,S,E,R];$=$.map(C=>Math.max(0,Math.min(C,1)));let _=[$[0]*t[0],$[1]*t[1],$[2]*t[0],$[3]*t[1]],P={id:s++,score:Math.round(100*x)/100,class:y+1,label:bd[y].label,box:_.map(C=>Math.trunc(C)),boxRaw:$};r.push(P)}}ee([c,d,h,f])}let a=r.map(l=>[l.boxRaw[1],l.boxRaw[0],l.boxRaw[3],l.boxRaw[2]]),o=r.map(l=>l.score),i=[];if(a&&a.length>0){let l=await Se.nonMaxSuppressionAsync(a,o,n.object.maxDetected,n.object.iouThreshold,n.object.minConfidence);i=await l.data(),ee(l)}return r=r.filter((l,u)=>i.includes(u)).sort((l,u)=>u.score-l.score),r}async function x4(e,t){let n=(t.object.skipTime||0)>ue()-$E,s=A4<(t.object.skipFrames||0);return t.skipAllowed&&n&&s&&u1.length>0?(A4++,u1):(A4=0,!pe.kernels.includes("mod")||!pe.kernels.includes("sparsetodense")?u1:new Promise(async r=>{let a=[e.shape[2]||0,e.shape[1]||0],o=Se.resizeBilinear(e,[d1,d1],!1),i=he(o,rt.tf255),l=et(i,[0,3,1,2]),u;t.object.enabled&&(u=Nd.execute(l)),$E=ue();let c=await nxe(u,a,t);u1=c,ee([o,i,l,...u]),r(c)}))}var Jh=["nose","leftEye","rightEye","leftEar","rightEar","leftShoulder","rightShoulder","leftElbow","rightElbow","leftWrist","rightWrist","leftHip","rightHip","leftKnee","rightKnee","leftAnkle","rightAnkle"],sxe=Jh.length,Yh=Jh.reduce((e,t,n)=>(e[t]=n,e),{}),rxe=[["leftHip","leftShoulder"],["leftElbow","leftShoulder"],["leftElbow","leftWrist"],["leftHip","leftKnee"],["leftKnee","leftAnkle"],["rightHip","rightShoulder"],["rightElbow","rightShoulder"],["rightElbow","rightWrist"],["rightHip","rightKnee"],["rightKnee","rightAnkle"],["leftShoulder","rightShoulder"],["leftHip","rightHip"]],gIe=rxe.map(([e,t])=>[Yh[e],Yh[t]]),OE=[["nose","leftEye"],["leftEye","leftEar"],["nose","rightEye"],["rightEye","rightEar"],["nose","leftShoulder"],["leftShoulder","leftElbow"],["leftElbow","leftWrist"],["leftShoulder","leftHip"],["leftHip","leftKnee"],["leftKnee","leftAnkle"],["nose","rightShoulder"],["rightShoulder","rightElbow"],["rightElbow","rightWrist"],["rightShoulder","rightHip"],["rightHip","rightKnee"],["rightKnee","rightAnkle"]];function ME(e){let t=e.reduce(({maxX:n,maxY:s,minX:r,minY:a},{position:{x:o,y:i}})=>({maxX:Math.max(n,o),maxY:Math.max(s,i),minX:Math.min(r,o),minY:Math.min(a,i)}),{maxX:Number.NEGATIVE_INFINITY,maxY:Number.NEGATIVE_INFINITY,minX:Number.POSITIVE_INFINITY,minY:Number.POSITIVE_INFINITY});return[t.minX,t.minY,t.maxX-t.minX,t.maxY-t.minY]}function zE(e,[t,n],[s,r]){let a=t/s,o=n/r,i=(u,c)=>({id:c,score:u.score,boxRaw:[u.box[0]/r,u.box[1]/s,u.box[2]/r,u.box[3]/s],box:[Math.trunc(u.box[0]*o),Math.trunc(u.box[1]*a),Math.trunc(u.box[2]*o),Math.trunc(u.box[3]*a)],keypoints:u.keypoints.map(({score:p,part:d,position:h})=>({score:p,part:d,position:[Math.trunc(h.x*o),Math.trunc(h.y*a)],positionRaw:[h.x/s,h.y/s]})),annotations:{}});return e.map((u,c)=>i(u,c))}var p1=class{constructor(t,n){ge(this,"priorityQueue");ge(this,"numberOfElements");ge(this,"getElementValue");this.priorityQueue=new Array(t),this.numberOfElements=-1,this.getElementValue=n}enqueue(t){this.priorityQueue[++this.numberOfElements]=t,this.swim(this.numberOfElements)}dequeue(){let t=this.priorityQueue[0];return this.exchange(0,this.numberOfElements--),this.sink(0),this.priorityQueue[this.numberOfElements+1]=null,t}empty(){return this.numberOfElements===-1}size(){return this.numberOfElements+1}all(){return this.priorityQueue.slice(0,this.numberOfElements+1)}max(){return this.priorityQueue[0]}swim(t){for(;t>0&&this.less(Math.floor(t/2),t);)this.exchange(t,Math.floor(t/2)),t=Math.floor(t/2)}sink(t){for(;2*t<=this.numberOfElements;){let n=2*t;if(n<this.numberOfElements&&this.less(n,n+1)&&n++,!this.less(t,n))break;this.exchange(t,n),t=n}}getValueAt(t){return this.getElementValue(this.priorityQueue[t])}less(t,n){return this.getValueAt(t)<this.getValueAt(n)}exchange(t,n){let s=this.priorityQueue[t];this.priorityQueue[t]=this.priorityQueue[n],this.priorityQueue[n]=s}};function b4(e,t,n,s){return{y:s.get(e,t,n),x:s.get(e,t,n+sxe)}}function v4(e,t,n){let{heatmapY:s,heatmapX:r,id:a}=e,{y:o,x:i}=b4(s,r,a,n);return{x:e.heatmapX*t+i,y:e.heatmapY*t+o}}function w4(e,t,n){return e<t?t:e>n?n:e}function LE(e,t,n,s){let r=n-e,a=s-t;return r*r+a*a}function k4(e,t){return{x:e.x+t.x,y:e.y+t.y}}var Br,oxe=["MobilenetV1/offset_2/BiasAdd","MobilenetV1/heatmap_2/BiasAdd","MobilenetV1/displacement_fwd_2/BiasAdd","MobilenetV1/displacement_bwd_2/BiasAdd"],h1=1,Ed=16,ixe=50**2;function BE(e,t,n,s,r,a,o=2){let i=y=>({y:a.get(y.y,y.x,e),x:a.get(y.y,y.x,a.shape[2]/2+e)}),l=(y,x,A)=>({y:w4(Math.round(y.y/Ed),0,x-1),x:w4(Math.round(y.x/Ed),0,A-1)}),[u,c]=s.shape,p=l(t.position,u,c),d=i(p),f=k4(t.position,d);for(let y=0;y<o;y++){let x=l(f,u,c),A=b4(x.y,x.x,n,r);f=k4({x:x.x*Ed,y:x.y*Ed},{x:A.x,y:A.y})}let m=l(f,u,c),g=s.get(m.y,m.x,n);return{position:f,part:Jh[n],score:g}}function lxe(e,t,n,s,r){let a=OE.map(([d,h])=>[Yh[d],Yh[h]]),o=a.map(([,d])=>d),i=a.map(([d])=>d),l=t.shape[2],u=o.length,c=new Array(l),p=v4(e.part,Ed,n);c[e.part.id]={score:e.score,part:Jh[e.part.id],position:p};for(let d=u-1;d>=0;--d){let h=o[d],f=i[d];c[h]&&!c[f]&&(c[f]=BE(d,c[h],f,t,n,r))}for(let d=0;d<u;++d){let h=i[d],f=o[d];c[h]&&!c[f]&&(c[f]=BE(d,c[h],f,t,n,s))}return c}function uxe(e,t,n,s,r){let[a,o]=r.shape,i=!0,l=Math.max(n-h1,0),u=Math.min(n+h1+1,a);for(let c=l;c<u;++c){let p=Math.max(s-h1,0),d=Math.min(s+h1+1,o);for(let h=p;h<d;++h)if(r.get(c,h,e)>t){i=!1;break}if(!i)break}return i}function cxe(e,t){let[n,s,r]=t.shape,a=new p1(n*s*r,({score:o})=>o);for(let o=0;o<n;++o)for(let i=0;i<s;++i)for(let l=0;l<r;++l){let u=t.get(o,i,l);u<e||uxe(l,u,o,i,t)&&a.enqueue({score:u,part:{heatmapY:o,heatmapX:i,id:l}})}return a}function WE(e,{x:t,y:n},s){return e.some(({keypoints:r})=>{var o;let a=(o=r[s])==null?void 0:o.position;return a?LE(n,t,a.y,a.x)<=ixe:!1})}function dxe(e,t){return t.reduce((s,{position:r,score:a},o)=>(WE(e,r,o)||(s+=a),s),0)/t.length}function pxe(e,t,n,s,r,a){let o=[],i=cxe(a,t);for(;o.length<r&&!i.empty();){let l=i.dequeue(),u=v4(l.part,Ed,e);if(WE(o,u,l.part.id))continue;let c=lxe(l,t,e,n,s);c=c.filter(h=>h.score>a);let p=dxe(o,c),d=ME(c);p>a&&o.push({keypoints:c,box:d,score:Math.round(100*p)/100})}return o}async function I4(e,t){let n=Y(()=>{if(!Br.inputs[0].shape)return[];let o=Se.resizeBilinear(e,[Br.inputs[0].shape[2],Br.inputs[0].shape[1]]),i=fe(he(ye(o,"float32"),127.5),1),u=Br.execute(i,oxe).map(c=>st(c,[0]));return u[1]=Cn(u[1]),u}),s=await Promise.all(n.map(o=>o.buffer()));for(let o of n)ee(o);let r=await pxe(s[0],s[1],s[2],s[3],t.body.maxDetected,t.body.minConfidence);return Br.inputs[0].shape?zE(r,[e.shape[1],e.shape[2]],[Br.inputs[0].shape[2],Br.inputs[0].shape[1]]):[]}async function VE(e){return!Br||pe.initial?Br=await Ge(e.body.modelPath):e.debug&&le("cached model:",Br.modelUrl),Br}var na,S4=!1;async function C4(e){return!na||pe.initial?na=await Ge(e.segmentation.modelPath):e.debug&&le("cached model:",na.modelUrl),na}async function GE(e,t,n){var m,g;if(S4)return{data:[],canvas:null,alpha:null};S4=!0,na||await C4(n);let s=await gd(e,n),r=((m=s.tensor)==null?void 0:m.shape[2])||0,a=((g=s.tensor)==null?void 0:g.shape[1])||0;if(!s.tensor)return{data:[],canvas:null,alpha:null};let o={};o.resize=Se.resizeBilinear(s.tensor,[na.inputs[0].shape?na.inputs[0].shape[1]:0,na.inputs[0].shape?na.inputs[0].shape[2]:0],!1),ee(s.tensor),o.norm=he(o.resize,rt.tf255),o.res=na.execute(o.norm),o.squeeze=st(o.res,0),o.squeeze.shape[2]===2?(o.softmax=Jl(o.squeeze),[o.bg,o.fg]=En(o.softmax,2),o.expand=Kt(o.fg,2),o.pad=Kt(o.expand,0),o.crop=Se.cropAndResize(o.pad,[[0,0,.5,.5]],[0],[r,a]),o.data=st(o.crop,0)):o.data=Se.resizeBilinear(o.squeeze,[a,r]);let i=Array.from(await o.data.data());if(pe.node&&!pe.Canvas&&typeof ImageData=="undefined")return n.debug&&le("canvas support missing"),Object.keys(o).forEach(y=>ee(o[y])),{data:i,canvas:null,alpha:null};let l=is(r,a);Js&&await Js.toPixels(o.data,l);let u=l.getContext("2d");n.segmentation.blur&&n.segmentation.blur>0&&(u.filter=`blur(${n.segmentation.blur}px)`);let c=u.getImageData(0,0,r,a),p=is(r,a),d=p.getContext("2d");s.canvas&&d.drawImage(s.canvas,0,0),d.globalCompositeOperation="darken",n.segmentation.blur&&n.segmentation.blur>0&&(d.filter=`blur(${n.segmentation.blur}px)`),d.drawImage(l,0,0),d.globalCompositeOperation="source-over",d.filter="none";let h=d.getImageData(0,0,r,a);for(let y=0;y<r*a;y++)h.data[4*y+3]=c.data[4*y+0];d.putImageData(h,0,0);let f=null;if(t&&p){f=is(r,a);let y=await gd(t,n);ee(y.tensor);let x=f.getContext("2d");x.drawImage(y.canvas,0,0,f.width,f.height),x.drawImage(p,0,0)}return Object.keys(o).forEach(y=>ee(o[y])),S4=!1,{data:i,canvas:p,alpha:l}}var Qh=class{constructor(){ge(this,"ssrnetage",null);ge(this,"gear",null);ge(this,"blazeposedetect",null);ge(this,"blazepose",null);ge(this,"centernet",null);ge(this,"efficientpose",null);ge(this,"mobilefacenet",null);ge(this,"insightface",null);ge(this,"emotion",null);ge(this,"facedetect",null);ge(this,"faceiris",null);ge(this,"facemesh",null);ge(this,"faceres",null);ge(this,"ssrnetgender",null);ge(this,"handpose",null);ge(this,"handskeleton",null);ge(this,"handtrack",null);ge(this,"liveness",null);ge(this,"movenet",null);ge(this,"nanodet",null);ge(this,"posenet",null);ge(this,"segmentation",null);ge(this,"antispoof",null)}},T4=e=>{let t=0,n=0,s=0;for(let a of Object.values(Or))t+=a.sizeFromManifest,n+=a.sizeLoadedWeights,s+=a.sizeDesired;let r=s>0?n/s:0;return{numLoadedModels:Object.values(Or).length,numEnabledModels:void 0,numDefinedModels:Object.keys(e.models).length,percentageLoaded:r,totalSizeFromManifest:t,totalSizeWeights:n,totalSizeLoading:s,totalSizeEnabled:void 0,modelStats:Object.values(Or)}};function f1(e){for(let t of Object.keys(e.models))e.models[t]=null}async function N4(e){var t,n,s,r,a,o,i,l,u,c,p,d,h,f,m,g,y,x,A,b,w,k,S,E,R,$,_,P,C,F,G,K;pe.initial&&f1(e),e.config.hand.enabled&&(!e.models.handpose&&((n=(t=e.config.hand.detector)==null?void 0:t.modelPath)==null?void 0:n.includes("handdetect"))&&([e.models.handpose,e.models.handskeleton]=await o4(e.config)),!e.models.handskeleton&&e.config.hand.landmarks&&((r=(s=e.config.hand.detector)==null?void 0:s.modelPath)==null?void 0:r.includes("handdetect"))&&([e.models.handpose,e.models.handskeleton]=await o4(e.config))),e.config.body.enabled&&!e.models.blazepose&&((o=(a=e.config.body)==null?void 0:a.modelPath)==null?void 0:o.includes("blazepose"))&&(e.models.blazepose=pN(e.config)),e.config.body.enabled&&!e.models.blazeposedetect&&e.config.body.detector&&e.config.body.detector.modelPath&&(e.models.blazeposedetect=dN(e.config)),e.config.body.enabled&&!e.models.efficientpose&&((l=(i=e.config.body)==null?void 0:i.modelPath)==null?void 0:l.includes("efficientpose"))&&(e.models.efficientpose=AN(e.config)),e.config.body.enabled&&!e.models.movenet&&((c=(u=e.config.body)==null?void 0:u.modelPath)==null?void 0:c.includes("movenet"))&&(e.models.movenet=_E(e.config)),e.config.body.enabled&&!e.models.posenet&&((d=(p=e.config.body)==null?void 0:p.modelPath)==null?void 0:d.includes("posenet"))&&(e.models.posenet=VE(e.config)),e.config.face.enabled&&!e.models.facedetect&&(e.models.facedetect=nN(e.config)),e.config.face.enabled&&((h=e.config.face.antispoof)==null?void 0:h.enabled)&&!e.models.antispoof&&(e.models.antispoof=UT(e.config)),e.config.face.enabled&&((f=e.config.face.liveness)==null?void 0:f.enabled)&&!e.models.liveness&&(e.models.liveness=SE(e.config)),e.config.face.enabled&&((m=e.config.face.description)==null?void 0:m.enabled)&&!e.models.faceres&&(e.models.faceres=XN(e.config)),e.config.face.enabled&&((g=e.config.face.emotion)==null?void 0:g.enabled)&&!e.models.emotion&&(e.models.emotion=wN(e.config)),e.config.face.enabled&&((y=e.config.face.iris)==null?void 0:y.enabled)&&!((x=e.config.face.attention)!=null&&x.enabled)&&!e.models.faceiris&&(e.models.faceiris=MN(e.config)),e.config.face.enabled&&((A=e.config.face.mesh)==null?void 0:A.enabled)&&!e.models.facemesh&&(e.models.facemesh=UN(e.config)),e.config.face.enabled&&((b=e.config.face.gear)==null?void 0:b.enabled)&&!e.models.gear&&(e.models.gear=RT(e.config)),e.config.face.enabled&&((w=e.config.face.ssrnet)==null?void 0:w.enabled)&&!e.models.ssrnetage&&(e.models.ssrnetage=FT(e.config)),e.config.face.enabled&&((k=e.config.face.ssrnet)==null?void 0:k.enabled)&&!e.models.ssrnetgender&&(e.models.ssrnetgender=LT(e.config)),e.config.face.enabled&&((S=e.config.face.mobilefacenet)==null?void 0:S.enabled)&&!e.models.mobilefacenet&&(e.models.mobilefacenet=TN(e.config)),e.config.face.enabled&&((E=e.config.face.insightface)==null?void 0:E.enabled)&&!e.models.insightface&&(e.models.insightface=DN(e.config)),e.config.hand.enabled&&!e.models.handtrack&&(($=(R=e.config.hand.detector)==null?void 0:R.modelPath)==null?void 0:$.includes("handtrack"))&&(e.models.handtrack=bE(e.config)),e.config.hand.enabled&&e.config.hand.landmarks&&!e.models.handskeleton&&((P=(_=e.config.hand.detector)==null?void 0:_.modelPath)==null?void 0:P.includes("handtrack"))&&(e.models.handskeleton=vE(e.config)),e.config.object.enabled&&!e.models.centernet&&((F=(C=e.config.object)==null?void 0:C.modelPath)==null?void 0:F.includes("centernet"))&&(e.models.centernet=mN(e.config)),e.config.object.enabled&&!e.models.nanodet&&((K=(G=e.config.object)==null?void 0:G.modelPath)==null?void 0:K.includes("nanodet"))&&(e.models.nanodet=PE(e.config)),e.config.segmentation.enabled&&!e.models.segmentation&&(e.models.segmentation=C4(e.config));for await(let z of Object.keys(e.models))e.models[z]&&typeof e.models[z]!="undefined"&&(e.models[z]=await e.models[z])}async function E4(e){let t=["const","placeholder","noop","pad","squeeze","add","sub","mul","div"];for(let n of Object.keys(e.models)){let s=e.models[n];if(!s)continue;let r=[],a=s==null?void 0:s.executor;if(a&&a.graph.nodes)for(let i of Object.values(a.graph.nodes)){let l=i.op.toLowerCase();r.includes(l)||r.push(l)}else!a&&e.config.debug&&le("model signature not determined:",n);let o=[];for(let i of r)!t.includes(i)&&!e.env.kernels.includes(i)&&!e.env.kernels.includes(i.replace("_",""))&&!e.env.kernels.includes(i.replace("native",""))&&!e.env.kernels.includes(i.replace("v2",""))&&o.push(i);e.config.debug&&o.length>0&&le("model validation failed:",n,o)}}var Pt={name:"humangl",priority:999,canvas:null,gl:null,extensions:[],webGLattr:{alpha:!1,antialias:!1,premultipliedAlpha:!1,preserveDrawingBuffer:!1,depth:!1,stencil:!1,failIfMajorPerformanceCaveat:!1,desynchronized:!0}};function hxe(){let e=Pt.gl;!e||(Pt.extensions=e.getSupportedExtensions())}async function jE(e){var t;if(e.config.backend==="humangl"&&(Pt.name in sn().registry&&(!Pt.gl||!Pt.gl.getParameter(Pt.gl.VERSION))&&(le("error: humangl backend invalid context"),f1(e)),!My(Pt.name))){try{Pt.canvas=await is(100,100)}catch(s){le("error: cannot create canvas:",s);return}try{if(Pt.gl=(t=Pt.canvas)==null?void 0:t.getContext("webgl2",Pt.webGLattr),!Pt.gl.getParameter(Pt.gl.VERSION).includes("2.0")){le("override: using fallback webgl backend as webgl 2.0 is not detected"),e.config.backend="webgl";return}Pt.canvas&&(Pt.canvas.addEventListener("webglcontextlost",async r=>{throw le("error: humangl:",r.type),le("possible browser memory leak using webgl or conflict with multiple backend registrations"),e.emit("error"),new Error("backend error: webgl context lost")}),Pt.canvas.addEventListener("webglcontextrestored",r=>{le("error: humangl context restored:",r)}),Pt.canvas.addEventListener("webglcontextcreationerror",r=>{le("error: humangl context create:",r)}))}catch(s){le("error: cannot get WebGL context:",s);return}try{x2(2,Pt.gl)}catch(s){le("error: cannot set WebGL context:",s);return}try{let s=new ju(Pt.gl);ql(Pt.name,()=>new id(s),Pt.priority)}catch(s){le("error: cannot register WebGL backend:",s);return}try{qr("webgl").forEach(r=>{let a={...r,backendName:Pt.name};dr(a)})}catch(s){le("error: cannot update WebGL backend registration:",s);return}let n=Ys().getGPGPUContext?Ys().getGPGPUContext().gl:null;if(n)le(`humangl webgl version:${n.getParameter(n.VERSION)} renderer:${n.getParameter(n.RENDERER)}`);else{le("error: no current gl context:",n,Pt.gl);return}try{q().flagRegistry.WEBGL_VERSION&&q().set("WEBGL_VERSION",2)}catch(s){le("error: cannot set WebGL backend flags:",s);return}hxe(),le("backend registered:",Pt.name)}}function fxe(){if(!pe.kernels.includes("mod")){let e={kernelName:"Mod",backendName:ss(),kernelFunc:t=>Y(()=>fe(t.inputs.a,L(he(t.inputs.a,t.inputs.b),t.inputs.b)))};dr(e),pe.kernels.push("mod")}if(!pe.kernels.includes("floormod")){let e={kernelName:"FloorMod",backendName:ss(),kernelFunc:t=>Y(()=>Bc(t.inputs.a/t.inputs.b)*t.inputs.b+Yl(t.inputs.a,t.inputs.b))};dr(e),pe.kernels.push("floormod")}}async function g1(e,t=!1){if(e.state="backend",t||pe.initial||e.config.backend&&e.config.backend.length>0&&ss()!==e.config.backend){let n=ue();if(e.config.backend&&e.config.backend.length>0){if(typeof window=="undefined"&&typeof WorkerGlobalScope!="undefined"&&e.config.debug&&e.config.debug&&le("running inside web worker"),pe.browser&&e.config.backend==="tensorflow"&&(e.config.debug&&le("override: backend set to tensorflow while running in browser"),e.config.backend="humangl"),pe.node&&(e.config.backend==="webgl"||e.config.backend==="humangl")&&(e.config.debug&&le(`override: backend set to ${e.config.backend} while running in nodejs`),e.config.backend="tensorflow"),pe.browser&&e.config.backend==="webgpu")if(typeof navigator=="undefined"||typeof navigator.gpu=="undefined")le("override: backend set to webgpu but browser does not support webgpu"),e.config.backend="humangl";else{let r=await navigator.gpu.requestAdapter();if(e.config.debug&&le("enumerated webgpu adapter:",r),!r)le("override: backend set to webgpu but browser reports no available gpu"),e.config.backend="humangl";else{let a="requestAdapterInfo"in r?await r.requestAdapterInfo():void 0;le("webgpu adapter info:",a)}}e.config.backend==="humangl"&&await jE(e);let s=Object.keys(sn().registryFactory);if(e.config.debug&&le("available backends:",s),s.includes(e.config.backend)||(le(`error: backend ${e.config.backend} not found in registry`),e.config.backend=pe.node?"tensorflow":"webgl",e.config.debug&&le(`override: setting backend ${e.config.backend}`)),e.config.debug&&le("setting backend:",e.config.backend),e.config.backend==="wasm"){if(q().flagRegistry.CANVAS2D_WILL_READ_FREQUENTLY&&q().set("CANVAS2D_WILL_READ_FREQUENTLY",!0),e.config.debug&&le("wasm path:",e.config.wasmPath),typeof(je==null?void 0:je.setWasmPaths)!="undefined")await nb(e.config.wasmPath,e.config.wasmPlatformFetch);else throw new Error("backend error: attempting to use wasm backend but wasm path is not set");let r=await q().getAsync("WASM_HAS_SIMD_SUPPORT"),a=await q().getAsync("WASM_HAS_MULTITHREAD_SUPPORT");e.config.debug&&le(`wasm execution: ${r?"SIMD":"no SIMD"} ${a?"multithreaded":"singlethreaded"}`),e.config.debug&&!r&&le("warning: wasm simd support is not enabled")}try{await Oy(e.config.backend),await Lc(),DT()}catch(r){return le("error: cannot set backend:",e.config.backend,r),!1}}if(ss()==="humangl"&&(q().flagRegistry.CHECK_COMPUTATION_FOR_ERRORS&&q().set("CHECK_COMPUTATION_FOR_ERRORS",!1),q().flagRegistry.WEBGL_CPU_FORWARD&&q().set("WEBGL_CPU_FORWARD",!0),q().flagRegistry.WEBGL_USE_SHAPES_UNIFORMS&&q().set("WEBGL_USE_SHAPES_UNIFORMS",!0),q().flagRegistry.CPU_HANDOFF_SIZE_THRESHOLD&&q().set("CPU_HANDOFF_SIZE_THRESHOLD",256),q().flagRegistry.WEBGL_EXP_CONV&&q().set("WEBGL_EXP_CONV",!0),q().flagRegistry.USE_SETTIMEOUTWPM&&q().set("USE_SETTIMEOUTWPM",!0),typeof e.config.deallocate!="undefined"&&e.config.deallocate&&(le("changing webgl: WEBGL_DELETE_TEXTURE_THRESHOLD:",!0),q().set("WEBGL_DELETE_TEXTURE_THRESHOLD",0)),Ys().getGPGPUContext)){let s=await Ys().getGPGPUContext().gl;e.config.debug&&le(`gl version:${s.getParameter(s.VERSION)} renderer:${s.getParameter(s.RENDERER)}`)}ss(),Py(),await Lc(),e.performance.initBackend=Math.trunc(ue()-n),e.config.backend=ss(),await pe.updateBackend(),fxe()}return!0}function r1(e,t){for(let n of e){let s={kernelName:n,backendName:t.backend,kernelFunc:()=>{t.debug&&le("kernelFunc",n,t.backend)}};dr(s)}pe.kernels=qr(ss()).map(n=>n.kernelName.toLowerCase())}var O4={};ia(O4,{all:()=>F4,body:()=>_d,canvas:()=>P4,face:()=>Rd,gesture:()=>Pd,hand:()=>Dd,object:()=>$d,options:()=>Un,person:()=>$4});var nr=e=>{if(!e)le("draw error: invalid canvas");else if(!e.getContext)le("draw error: canvas context not defined");else{let t=e.getContext("2d");if(!t)le("draw error: cannot get canvas context");else return t}return null},Au=e=>Math.round(e*180/Math.PI),Pa=(e,t)=>{if(!t.useDepth||typeof e=="undefined")return t.color;let n=Uint8ClampedArray.from([127+2*e,127-2*e,255]);return`rgba(${n[0]}, ${n[1]}, ${n[2]}, ${t.alpha})`};function Fa(e,t,n,s,r){e.fillStyle=Pa(s,r),e.beginPath(),e.arc(t,n,r.pointSize,0,2*Math.PI),e.fill()}function sa(e,t,n,s,r,a){if(e.beginPath(),e.lineWidth=a.lineWidth,a.useCurves){let o=(t+t+s)/2,i=(n+n+r)/2;e.ellipse(o,i,s/2,r/2,0,0,2*Math.PI)}else e.moveTo(t+a.roundRect,n),e.lineTo(t+s-a.roundRect,n),e.quadraticCurveTo(t+s,n,t+s,n+a.roundRect),e.lineTo(t+s,n+r-a.roundRect),e.quadraticCurveTo(t+s,n+r,t+s-a.roundRect,n+r),e.lineTo(t+a.roundRect,n+r),e.quadraticCurveTo(t,n+r,t,n+r-a.roundRect),e.lineTo(t,n+a.roundRect),e.quadraticCurveTo(t,n,t+a.roundRect,n),e.closePath();e.stroke()}function R4(e,t,n){if(!(t.length<2)){e.beginPath(),e.moveTo(t[0][0],t[0][1]);for(let s of t)e.strokeStyle=Pa(s[2]||0,n),e.lineTo(Math.trunc(s[0]),Math.trunc(s[1]));e.stroke(),n.fillPolygons&&(e.closePath(),e.fill())}}function XE(e,t,n){if(!(t.length<2)){if(e.lineWidth=n.lineWidth,!n.useCurves||t.length<=2){R4(e,t,n);return}e.moveTo(t[0][0],t[0][1]);for(let s=0;s<t.length-2;s++){let r=(t[s][0]+t[s+1][0])/2,a=(t[s][1]+t[s+1][1])/2;e.quadraticCurveTo(t[s][0],t[s][1],r,a)}e.quadraticCurveTo(t[t.length-2][0],t[t.length-2][1],t[t.length-1][0],t[t.length-1][1]),e.stroke(),n.fillPolygons&&(e.closePath(),e.fill())}}function _4(e,t,n,s=5){let r,a,o;e.beginPath(),e.moveTo(t[0],t[1]),e.lineTo(n[0],n[1]),r=Math.atan2(n[1]-t[1],n[0]-t[0]),a=s*Math.cos(r)+n[0],o=s*Math.sin(r)+n[1],e.moveTo(a,o),r+=1/3*(2*Math.PI),a=s*Math.cos(r)+n[0],o=s*Math.sin(r)+n[1],e.lineTo(a,o),r+=1/3*(2*Math.PI),a=s*Math.cos(r)+n[0],o=s*Math.sin(r)+n[1],e.lineTo(a,o),e.closePath(),e.stroke(),e.fill()}var Un={color:"rgba(173, 216, 230, 0.6)",labelColor:"rgba(173, 216, 230, 1)",shadowColor:"black",alpha:.5,font:'small-caps 16px "Segoe UI"',lineHeight:18,lineWidth:4,pointSize:2,roundRect:8,drawPoints:!1,drawLabels:!0,drawBoxes:!0,drawAttention:!0,drawGestures:!0,drawPolygons:!0,drawGaze:!0,fillPolygons:!1,useDepth:!0,useCurves:!1};var gt;function gxe(e,t){if(gt.drawLabels){let n=[];if(n.push(`face: ${Math.trunc(100*e.score)}%`),e.genderScore&&n.push(`${e.gender||""} ${Math.trunc(100*e.genderScore)}%`),e.age&&n.push(`age: ${e.age||""}`),e.iris&&n.push(`distance: ${e.iris}`),e.real&&n.push(`real: ${Math.trunc(100*e.real)}%`),e.live&&n.push(`live: ${Math.trunc(100*e.live)}%`),e.emotion&&e.emotion.length>0){let s=e.emotion.map(r=>`${Math.trunc(100*r.score)}% ${r.emotion}`);s.length>3&&(s.length=3),n.push(s.join(" "))}e.rotation&&e.rotation.angle&&e.rotation.gaze&&(e.rotation.angle.roll&&n.push(`roll: ${Au(e.rotation.angle.roll)}\xB0 yaw:${Au(e.rotation.angle.yaw)}\xB0 pitch:${Au(e.rotation.angle.pitch)}\xB0`),e.rotation.gaze.bearing&&n.push(`gaze: ${Au(e.rotation.gaze.bearing)}\xB0`)),n.length===0&&n.push("face"),t.fillStyle=gt.color;for(let s=n.length-1;s>=0;s--){let r=Math.max(e.box[0],0),a=s*gt.lineHeight+e.box[1];gt.shadowColor&>.shadowColor!==""&&(t.fillStyle=gt.shadowColor,t.fillText(n[s],r+5,a+16)),t.fillStyle=gt.labelColor,t.fillText(n[s],r+4,a+15)}}}function yxe(e,t){if(e.annotations&&e.annotations.leftEyeIris&&e.annotations.leftEyeIris[0]){t.strokeStyle=gt.useDepth?"rgba(255, 200, 255, 0.3)":gt.color,t.beginPath();let n=Math.abs(e.annotations.leftEyeIris[3][0]-e.annotations.leftEyeIris[1][0])/2,s=Math.abs(e.annotations.leftEyeIris[4][1]-e.annotations.leftEyeIris[2][1])/2;t.ellipse(e.annotations.leftEyeIris[0][0],e.annotations.leftEyeIris[0][1],n,s,0,0,2*Math.PI),t.stroke(),gt.fillPolygons&&(t.fillStyle=gt.useDepth?"rgba(255, 255, 200, 0.3)":gt.color,t.fill())}if(e.annotations&&e.annotations.rightEyeIris&&e.annotations.rightEyeIris[0]){t.strokeStyle=gt.useDepth?"rgba(255, 200, 255, 0.3)":gt.color,t.beginPath();let n=Math.abs(e.annotations.rightEyeIris[3][0]-e.annotations.rightEyeIris[1][0])/2,s=Math.abs(e.annotations.rightEyeIris[4][1]-e.annotations.rightEyeIris[2][1])/2;t.ellipse(e.annotations.rightEyeIris[0][0],e.annotations.rightEyeIris[0][1],n,s,0,0,2*Math.PI),t.stroke(),gt.fillPolygons&&(t.fillStyle=gt.useDepth?"rgba(255, 255, 200, 0.3)":gt.color,t.fill())}}function Axe(e,t){var n;if(gt.drawGaze&&((n=e.rotation)==null?void 0:n.angle)&&typeof Path2D!="undefined"){t.strokeStyle="pink";let s=e.box[0]+e.box[2]/2-e.box[3]*Au(e.rotation.angle.yaw)/90,r=e.box[1]+e.box[3]/2+e.box[2]*Au(e.rotation.angle.pitch)/90,a=new Path2D(`
|
|
M ${e.box[0]+e.box[2]/2} ${e.box[1]}
|
|
C
|
|
${s} ${e.box[1]},
|
|
${s} ${e.box[1]+e.box[3]},
|
|
${e.box[0]+e.box[2]/2} ${e.box[1]+e.box[3]}
|
|
`),o=new Path2D(`
|
|
M ${e.box[0]} ${e.box[1]+e.box[3]/2}
|
|
C
|
|
${e.box[0]} ${r},
|
|
${e.box[0]+e.box[2]} ${r},
|
|
${e.box[0]+e.box[2]} ${e.box[1]+e.box[3]/2}
|
|
`);t.stroke(o),t.stroke(a)}}function xxe(e,t){var n,s,r,a;if(gt.drawGaze&&((s=(n=e.rotation)==null?void 0:n.gaze)==null?void 0:s.strength)&&((a=(r=e.rotation)==null?void 0:r.gaze)==null?void 0:a.bearing)&&e.annotations.leftEyeIris&&e.annotations.rightEyeIris&&e.annotations.leftEyeIris[0]&&e.annotations.rightEyeIris[0]){t.strokeStyle="pink",t.fillStyle="pink";let o=[e.annotations.leftEyeIris[0][0]+Math.sin(e.rotation.gaze.bearing)*e.rotation.gaze.strength*e.box[3],e.annotations.leftEyeIris[0][1]+Math.cos(e.rotation.gaze.bearing)*e.rotation.gaze.strength*e.box[2]];_4(t,[e.annotations.leftEyeIris[0][0],e.annotations.leftEyeIris[0][1]],[o[0],o[1]],4);let i=[e.annotations.rightEyeIris[0][0]+Math.sin(e.rotation.gaze.bearing)*e.rotation.gaze.strength*e.box[3],e.annotations.rightEyeIris[0][1]+Math.cos(e.rotation.gaze.bearing)*e.rotation.gaze.strength*e.box[2]];_4(t,[e.annotations.rightEyeIris[0][0],e.annotations.rightEyeIris[0][1]],[i[0],i[1]],4)}}function bxe(e,t){if(gt.drawPolygons&&e.mesh.length>=468){t.lineWidth=1;for(let n=0;n<lu.length/3;n++){let s=[lu[n*3+0],lu[n*3+1],lu[n*3+2]].map(r=>e.mesh[r]);R4(t,s,gt)}yxe(e,t)}}function vxe(e,t){if(gt.drawPoints&&e.mesh.length>=468)for(let n=0;n<e.mesh.length;n++)Fa(t,e.mesh[n][0],e.mesh[n][1],e.mesh[n][2],gt),gt.drawAttention&&(Xh.includes(n)&&Fa(t,e.mesh[n][0],e.mesh[n][1],e.mesh[n][2]+127,gt),du.includes(n)&&Fa(t,e.mesh[n][0],e.mesh[n][1],e.mesh[n][2]-127,gt),pu.includes(n)&&Fa(t,e.mesh[n][0],e.mesh[n][1],e.mesh[n][2]-127,gt))}function wxe(e,t){gt.drawBoxes&&sa(t,e.box[0],e.box[1],e.box[2],e.box[3],gt)}async function Rd(e,t,n){if(gt=Xt(Un,n),!t||!e)return;let s=nr(e);if(!!s){s.font=gt.font,s.strokeStyle=gt.color,s.fillStyle=gt.color;for(let r of t)wxe(r,s),gxe(r,s),r.mesh&&r.mesh.length>0&&(vxe(r,s),bxe(r,s),Axe(r,s),xxe(r,s))}}async function _d(e,t,n){var a;let s=Xt(Un,n);if(!t||!e)return;let r=nr(e);if(!!r){r.lineJoin="round";for(let o=0;o<t.length;o++){if(r.strokeStyle=s.color,r.fillStyle=s.color,r.lineWidth=s.lineWidth,r.font=s.font,s.drawBoxes&&t[o].box&&((a=t[o].box)==null?void 0:a.length)===4&&(sa(r,t[o].box[0],t[o].box[1],t[o].box[2],t[o].box[3],s),s.drawLabels&&(s.shadowColor&&s.shadowColor!==""&&(r.fillStyle=s.shadowColor,r.fillText(`body ${100*t[o].score}%`,t[o].box[0]+3,1+t[o].box[1]+s.lineHeight,t[o].box[2])),r.fillStyle=s.labelColor,r.fillText(`body ${100*t[o].score}%`,t[o].box[0]+2,0+t[o].box[1]+s.lineHeight,t[o].box[2]))),s.drawPoints&&t[o].keypoints)for(let i=0;i<t[o].keypoints.length;i++)!t[o].keypoints[i].score||t[o].keypoints[i].score===0||(r.fillStyle=Pa(t[o].keypoints[i].position[2],s),Fa(r,t[o].keypoints[i].position[0],t[o].keypoints[i].position[1],0,s));if(s.drawLabels&&t[o].keypoints){r.font=s.font;for(let i of t[o].keypoints)!i.score||i.score===0||(r.fillStyle=Pa(i.position[2],s),r.fillText(`${i.part} ${Math.trunc(100*i.score)}%`,i.position[0]+4,i.position[1]+4))}if(s.drawPolygons&&t[o].keypoints&&t[o].annotations)for(let i of Object.values(t[o].annotations))for(let l of i)XE(r,l,s)}}}async function Dd(e,t,n){let s=Xt(Un,n);if(!t||!e)return;let r=nr(e);if(!!r){r.lineJoin="round",r.font=s.font;for(let a of t){if(s.drawBoxes&&(r.strokeStyle=s.color,r.fillStyle=s.color,sa(r,a.box[0],a.box[1],a.box[2],a.box[3],s),s.drawLabels&&(s.shadowColor&&s.shadowColor!==""&&(r.fillStyle=s.shadowColor,r.fillText(`hand:${Math.trunc(100*a.score)}%`,a.box[0]+3,1+a.box[1]+s.lineHeight,a.box[2])),r.fillStyle=s.labelColor,r.fillText(`hand:${Math.trunc(100*a.score)}%`,a.box[0]+2,0+a.box[1]+s.lineHeight,a.box[2])),r.stroke()),s.drawPoints&&a.keypoints&&a.keypoints.length>0)for(let o of a.keypoints)r.fillStyle=Pa(o[2],s),Fa(r,o[0],o[1],0,s);if(s.drawLabels&&a.annotations){let o=(i,l)=>{if(!i||i.length===0||!i[0])return;let u=i[i.length-1][2]||-256;r.fillStyle=Pa(u,s),r.fillText(l,i[i.length-1][0]+4,i[i.length-1][1]+4)};r.font=s.font,o(a.annotations.index,"index"),o(a.annotations.middle,"middle"),o(a.annotations.ring,"ring"),o(a.annotations.pinky,"pinky"),o(a.annotations.thumb,"thumb"),o(a.annotations.palm,"palm")}if(s.drawPolygons&&a.annotations){let o=i=>{if(!(!i||i.length===0||!i[0]))for(let l=0;l<i.length;l++){r.beginPath();let u=i[l][2]||0;r.strokeStyle=Pa(l*u,s),r.moveTo(i[l>0?l-1:0][0],i[l>0?l-1:0][1]),r.lineTo(i[l][0],i[l][1]),r.stroke()}};r.lineWidth=s.lineWidth,o(a.annotations.index),o(a.annotations.middle),o(a.annotations.ring),o(a.annotations.pinky),o(a.annotations.thumb)}}}}async function $d(e,t,n){let s=Xt(Un,n);if(!t||!e)return;let r=nr(e);if(!!r){r.lineJoin="round",r.font=s.font;for(let a of t)if(s.drawBoxes){if(r.strokeStyle=s.color,r.fillStyle=s.color,sa(r,a.box[0],a.box[1],a.box[2],a.box[3],s),s.drawLabels){let o=`${a.label} ${Math.round(100*a.score)}%`;s.shadowColor&&s.shadowColor!==""&&(r.fillStyle=s.shadowColor,r.fillText(o,a.box[0]+3,1+a.box[1]+s.lineHeight,a.box[2])),r.fillStyle=s.labelColor,r.fillText(o,a.box[0]+2,0+a.box[1]+s.lineHeight,a.box[2])}r.stroke()}}}async function Pd(e,t,n){let s=Xt(Un,n);if(!(!t||!e)&&s.drawGestures){let r=nr(e);if(!r)return;r.font=s.font,r.fillStyle=s.color;let a=1;for(let o=0;o<t.length;o++){let i=[],l=[];if([i,l]=Object.entries(t[o]),l.length>1&&l[1].length>0){let u=i[1]>0?`#${i[1]}`:"",c=`${i[0]} ${u}: ${l[1]}`;s.shadowColor&&s.shadowColor!==""&&(r.fillStyle=s.shadowColor,r.fillText(c,8,2+a*s.lineHeight)),r.fillStyle=s.labelColor,r.fillText(c,6,0+a*s.lineHeight),a+=1}}}}var D4=0;async function $4(e,t,n){let s=Xt(Un,n);if(!t||!e)return;let r=nr(e);if(!!r){r.lineJoin="round",r.font=s.font;for(let a=0;a<t.length;a++)if(s.drawBoxes){if(r.strokeStyle=s.color,r.fillStyle=s.color,sa(r,t[a].box[0],t[a].box[1],t[a].box[2],t[a].box[3],s),s.drawLabels){let o=`person #${a}`;s.shadowColor&&s.shadowColor!==""&&(r.fillStyle=s.shadowColor,r.fillText(o,t[a].box[0]+3,1+t[a].box[1]+s.lineHeight,t[a].box[2])),r.fillStyle=s.labelColor,r.fillText(o,t[a].box[0]+2,0+t[a].box[1]+s.lineHeight,t[a].box[2])}r.stroke()}}}async function P4(e,t){if(!e||!t)return;let n=nr(t);!n||n.drawImage(e,0,0)}async function F4(e,t,n){if(!t||!t.performance||!t||!e)return null;let s=ue(),r=Xt(Un,n),a=Promise.all([Rd(e,t.face,r),_d(e,t.body,r),Dd(e,t.hand,r),$d(e,t.object,r),Pd(e,t.gesture,r)]);return D4=pe.perfadd?D4+Math.round(ue()-s):Math.round(ue()-s),t.performance.draw=D4,a}var Fd=.1,M4=.5;function kxe(e,t,n){let s=!1,r=n.length-1;for(let a=0;a<n.length;r=a++)n[a].y>t!=n[r].y>t&&e<(n[r].x-n[a].x)*(t-n[a].y)/(n[r].y-n[a].y)+n[a].x&&(s=!s);return s}async function KE(e){if(!e.tensor||!e.mesh||e.mesh.length<100)return e.tensor;let t=e.tensor.shape[2]||0,n=e.tensor.shape[1]||0,s=await e.tensor.buffer(),r=[];for(let o of yr.silhouette)r.push({x:(e.mesh[o][0]-e.box[0])/e.box[2],y:(e.mesh[o][1]-e.box[1])/e.box[3]});Fd&&Fd>0&&(r=r.map(o=>({x:o.x>.5?o.x+Fd:o.x-Fd,y:o.y>.5?o.y+Fd:o.y-Fd})));for(let o=0;o<t;o++)for(let i=0;i<n;i++)kxe(o/t,i/t,r)||(s.set(M4*s.get(0,i,o,0),0,i,o,0),s.set(M4*s.get(0,i,o,1),0,i,o,1),s.set(M4*s.get(0,i,o,2),0,i,o,2));let a=s.toTensor();return ee(s),a}var Sxe=e=>{let t=(p,d)=>Math.atan2(p[1]-d[1],p[0]-d[0]);if(!e.annotations.rightEyeIris||!e.annotations.leftEyeIris)return{bearing:0,strength:0};let n=[0,-.1],s=1,r=(e.mesh[33][2]||0)>(e.mesh[263][2]||0),a=r?e.mesh[473]:e.mesh[468],o=r?[(e.mesh[133][0]+e.mesh[33][0])/2,(e.mesh[133][1]+e.mesh[33][1])/2]:[(e.mesh[263][0]+e.mesh[362][0])/2,(e.mesh[263][1]+e.mesh[362][1])/2],i=r?[e.mesh[133][0]-e.mesh[33][0],e.mesh[23][1]-e.mesh[27][1]]:[e.mesh[263][0]-e.mesh[362][0],e.mesh[253][1]-e.mesh[257][1]],l=[(o[0]-a[0])/i[0]-n[0],s*(a[1]-o[1])/i[1]-n[1]],u=Math.sqrt(l[0]*l[0]+l[1]*l[1]);return u=Math.min(u,e.boxRaw[2]/2,e.boxRaw[3]/2),{bearing:(t([0,0],l)+Math.PI/2)%Math.PI,strength:u}},ZE=(e,t)=>{let n=m=>{let g=Math.sqrt(m[0]*m[0]+m[1]*m[1]+m[2]*m[2]);return m[0]/=g,m[1]/=g,m[2]/=g,m},s=(m,g)=>{let y=m[0]-g[0],x=m[1]-g[1],A=m[2]-g[2];return[y,x,A]},r=(m,g)=>{let y=m[1]*g[2]-m[2]*g[1],x=m[2]*g[0]-m[0]*g[2],A=m[0]*g[1]-m[1]*g[0];return[y,x,A]},a=m=>{let[g,y,x,A,b,w,k,S,E]=m,R,$,_;return A<1?A>-1?(_=Math.asin(A),$=Math.atan2(-k,g),R=Math.atan2(-w,b)):(_=-Math.PI/2,$=-Math.atan2(S,E),R=0):(_=Math.PI/2,$=Math.atan2(S,E),R=0),isNaN(R)&&(R=0),isNaN($)&&($=0),isNaN(_)&&(_=0),{pitch:2*-R,yaw:2*-$,roll:2*-_}},o=e.meshRaw;if(!o||o.length<300)return{angle:{pitch:0,yaw:0,roll:0},matrix:[1,0,0,0,1,0,0,0,1],gaze:{bearing:0,strength:0}};let i=Math.max(e.boxRaw[2]*t[0],e.boxRaw[3]*t[1])/1.5,l=[o[10],o[152],o[234],o[454]].map(m=>[m[0]*t[0]/i,m[1]*t[1]/i,m[2]]),u=n(s(l[1],l[0])),c=n(s(l[3],l[2])),p=n(r(c,u));c=r(u,p);let d=[c[0],c[1],c[2],u[0],u[1],u[2],p[0],p[1],p[2]],h=a(d),f=o.length===478?Sxe(e):{bearing:0,strength:0};return{angle:h,matrix:d,gaze:f}};var z4=async(e,t)=>{var f,m,g,y,x,A,b,w,k,S,E,R,$,_,P,C,F,G,K,z,Z,J,te;let n=ue(),s,r,a,o,i,l,u,c,p,d=[];e.state="run:face";let h=await VN(t,e.config);if(e.performance.face=pe.perfadd?(e.performance.face||0)+Math.trunc(ue()-n):Math.trunc(ue()-n),!t.shape||t.shape.length!==4)return[];if(!h)return[];for(let B=0;B<h.length;B++){if(e.analyze("Get Face"),!h[B].tensor||h[B].tensor.isDisposedInternal){le("Face object is disposed:",h[B].tensor);continue}if((f=e.config.face.detector)!=null&&f.mask){let me=await KE(h[B]);ee(h[B].tensor),h[B].tensor=me}let oe=h[B].mesh&&h[B].mesh.length>200?ZE(h[B],[t.shape[2],t.shape[1]]):null;e.analyze("Start Emotion:"),e.config.async?o=(m=e.config.face.emotion)!=null&&m.enabled?Ub(h[B].tensor||ct([]),e.config,B,h.length):[]:(e.state="run:emotion",n=ue(),o=(g=e.config.face.emotion)!=null&&g.enabled?await Ub(h[B].tensor||ct([]),e.config,B,h.length):[],e.performance.emotion=pe.perfadd?(e.performance.emotion||0)+Math.trunc(ue()-n):Math.trunc(ue()-n)),e.analyze("End Emotion:"),e.analyze("Start AntiSpoof:"),e.config.async?u=(y=e.config.face.antispoof)!=null&&y.enabled?Ib(h[B].tensor||ct([]),e.config,B,h.length):0:(e.state="run:antispoof",n=ue(),u=(x=e.config.face.antispoof)!=null&&x.enabled?await Ib(h[B].tensor||ct([]),e.config,B,h.length):0,e.performance.antispoof=pe.perfadd?(e.performance.antispoof||0)+Math.trunc(ue()-n):Math.trunc(ue()-n)),e.analyze("End AntiSpoof:"),e.analyze("Start Liveness:"),e.config.async?c=(A=e.config.face.liveness)!=null&&A.enabled?d4(h[B].tensor||ct([]),e.config,B,h.length):0:(e.state="run:liveness",n=ue(),c=(b=e.config.face.liveness)!=null&&b.enabled?await d4(h[B].tensor||ct([]),e.config,B,h.length):0,e.performance.liveness=pe.perfadd?(e.performance.antispoof||0)+Math.trunc(ue()-n):Math.trunc(ue()-n)),e.analyze("End Liveness:"),e.analyze("Start GEAR:"),e.config.async?r=(w=e.config.face.gear)!=null&&w.enabled?yb(h[B].tensor||ct([]),e.config,B,h.length):null:(e.state="run:gear",n=ue(),r=(k=e.config.face.gear)!=null&&k.enabled?await yb(h[B].tensor||ct([]),e.config,B,h.length):null,e.performance.gear=Math.trunc(ue()-n)),e.analyze("End GEAR:"),e.analyze("Start SSRNet:"),e.config.async?(s=(S=e.config.face.ssrnet)!=null&&S.enabled?xb(h[B].tensor||ct([]),e.config,B,h.length):null,a=(E=e.config.face.ssrnet)!=null&&E.enabled?wb(h[B].tensor||ct([]),e.config,B,h.length):null):(e.state="run:ssrnet",n=ue(),s=(R=e.config.face.ssrnet)!=null&&R.enabled?await xb(h[B].tensor||ct([]),e.config,B,h.length):null,a=($=e.config.face.ssrnet)!=null&&$.enabled?await wb(h[B].tensor||ct([]),e.config,B,h.length):null,e.performance.ssrnet=Math.trunc(ue()-n)),e.analyze("End SSRNet:"),e.analyze("Start MobileFaceNet:"),e.config.async?i=(_=e.config.face.mobilefacenet)!=null&&_.enabled?Hb(h[B].tensor||ct([]),e.config,B,h.length):null:(e.state="run:mobilefacenet",n=ue(),i=(P=e.config.face.mobilefacenet)!=null&&P.enabled?await Hb(h[B].tensor||ct([]),e.config,B,h.length):null,e.performance.mobilefacenet=Math.trunc(ue()-n)),e.analyze("End MobileFaceNet:"),e.analyze("Start InsightFace:"),e.config.async?l=(C=e.config.face.insightface)!=null&&C.enabled?qb(h[B].tensor||ct([]),e.config,B,h.length):null:(e.state="run:mobilefacenet",n=ue(),l=(F=e.config.face.insightface)!=null&&F.enabled?await qb(h[B].tensor||ct([]),e.config,B,h.length):null,e.performance.mobilefacenet=Math.trunc(ue()-n)),e.analyze("End InsightFace:"),e.analyze("Start Description:"),e.config.async?p=Qb(h[B].tensor||ct([]),e.config,B,h.length):(e.state="run:description",n=ue(),p=await Qb(h[B].tensor||ct([]),e.config,B,h.length),e.performance.description=pe.perfadd?(e.performance.description||0)+Math.trunc(ue()-n):Math.trunc(ue()-n)),e.analyze("End Description:"),e.config.async&&([s,a,o,i,l,p,r,u,c]=await Promise.all([s,a,o,i,l,p,r,u,c])),e.analyze("Finish Face:"),((G=e.config.face.ssrnet)==null?void 0:G.enabled)&&s&&a&&(p={...p,age:s.age,gender:a.gender,genderScore:a.genderScore}),((K=e.config.face.gear)==null?void 0:K.enabled)&&r&&(p={...p,age:r.age,gender:r.gender,genderScore:r.genderScore,race:r.race}),((z=e.config.face.mobilefacenet)==null?void 0:z.enabled)&&i&&(p.descriptor=i),((Z=e.config.face.insightface)==null?void 0:Z.enabled)&&l&&(p.descriptor=l),(J=e.config.face.iris)!=null&&J.enabled;let Q=h[B].annotations&&h[B].annotations.leftEyeIris&&h[B].annotations.leftEyeIris[0]&&h[B].annotations.rightEyeIris&&h[B].annotations.rightEyeIris[0]&&h[B].annotations.leftEyeIris.length>0&&h[B].annotations.rightEyeIris.length>0&&h[B].annotations.leftEyeIris[0]!==null&&h[B].annotations.rightEyeIris[0]!==null?Math.max(Math.abs(h[B].annotations.leftEyeIris[3][0]-h[B].annotations.leftEyeIris[1][0]),Math.abs(h[B].annotations.rightEyeIris[4][1]-h[B].annotations.rightEyeIris[2][1]))/t.shape[2]:0,ae=(te=e.config.face.detector)!=null&&te.return?st(h[B].tensor):null;ee(h[B].tensor),h[B].tensor&&delete h[B].tensor;let ie={...h[B],id:B};p!=null&&p.age&&(ie.age=p.age),p!=null&&p.gender&&(ie.gender=p.gender),p!=null&&p.genderScore&&(ie.genderScore=p==null?void 0:p.genderScore),p!=null&&p.descriptor&&(ie.embedding=p==null?void 0:p.descriptor),p!=null&&p.race&&(ie.race=p==null?void 0:p.race),o&&(ie.emotion=o),u&&(ie.real=u),c&&(ie.live=c),Q&&Q!==0&&(ie.iris=Math.trunc(500/Q/11.7)/100),oe&&(ie.rotation=oe),ae&&(ie.tensor=ae),d.push(ie),e.analyze("End Face")}return e.analyze("End FaceMesh:"),e.config.async&&(e.performance.face&&delete e.performance.face,e.performance.age&&delete e.performance.age,e.performance.gender&&delete e.performance.gender,e.performance.emotion&&delete e.performance.emotion),d};var YE=e=>{if(!e)return[];let t=[];for(let n=0;n<e.length;n++){let s=e[n].keypoints.find(l=>l.part==="leftWrist"),r=e[n].keypoints.find(l=>l.part==="rightWrist"),a=e[n].keypoints.find(l=>l.part==="nose");a&&s&&r&&s.position[1]<a.position[1]&&r.position[1]<a.position[1]?t.push({body:n,gesture:"i give up"}):a&&s&&s.position[1]<a.position[1]?t.push({body:n,gesture:"raise left hand"}):a&&r&&r.position[1]<a.position[1]&&t.push({body:n,gesture:"raise right hand"});let o=e[n].keypoints.find(l=>l.part==="leftShoulder"),i=e[n].keypoints.find(l=>l.part==="rightShoulder");o&&i&&Math.abs(o.positionRaw[1]-i.positionRaw[1])>.1&&t.push({body:n,gesture:`leaning ${o.position[1]>i.position[1]?"left":"right"}`})}return t},JE=e=>{if(!e)return[];let t=[];for(let n=0;n<e.length;n++)if(e[n].mesh&&e[n].mesh.length>450){let s=(e[n].mesh[33][2]||0)-(e[n].mesh[263][2]||0),r=e[n].mesh[33][0]-e[n].mesh[263][0];Math.abs(s/r)<=.15?t.push({face:n,gesture:"facing center"}):t.push({face:n,gesture:`facing ${s<0?"left":"right"}`}),Math.abs(e[n].mesh[374][1]-e[n].mesh[386][1])/Math.abs(e[n].mesh[443][1]-e[n].mesh[450][1])<.2&&t.push({face:n,gesture:"blink left eye"}),Math.abs(e[n].mesh[145][1]-e[n].mesh[159][1])/Math.abs(e[n].mesh[223][1]-e[n].mesh[230][1])<.2&&t.push({face:n,gesture:"blink right eye"});let i=Math.min(100,500*Math.abs(e[n].mesh[13][1]-e[n].mesh[14][1])/Math.abs(e[n].mesh[10][1]-e[n].mesh[152][1]));i>10&&t.push({face:n,gesture:`mouth ${Math.trunc(i)}% open`});let l=e[n].mesh[152][2]||0;Math.abs(l)>10&&t.push({face:n,gesture:`head ${l<0?"up":"down"}`})}return t},QE=e=>{if(!e)return[];let t=[];for(let n=0;n<e.length;n++){if(!e[n].annotations||!e[n].annotations.leftEyeIris||!e[n].annotations.leftEyeIris[0]||!e[n].annotations.rightEyeIris||!e[n].annotations.rightEyeIris[0])continue;let s=e[n].annotations.leftEyeIris[3][0]-e[n].annotations.leftEyeIris[1][0],r=e[n].annotations.leftEyeIris[4][1]-e[n].annotations.leftEyeIris[2][1],a=Math.abs(s*r),o=e[n].annotations.rightEyeIris[3][0]-e[n].annotations.rightEyeIris[1][0],i=e[n].annotations.rightEyeIris[4][1]-e[n].annotations.rightEyeIris[2][1],l=Math.abs(o*i),u=!1;Math.abs(a-l)/Math.max(a,l)<.25&&(u=!0,t.push({iris:n,gesture:"facing center"}));let p=Math.abs(e[n].mesh[263][0]-e[n].annotations.leftEyeIris[0][0])/e[n].box[2],d=Math.abs(e[n].mesh[33][0]-e[n].annotations.rightEyeIris[0][0])/e[n].box[2];(p>.06||d>.06)&&(u=!1),p>d?p>.05&&t.push({iris:n,gesture:"looking right"}):d>.05&&t.push({iris:n,gesture:"looking left"});let h=Math.abs(e[n].mesh[145][1]-e[n].annotations.rightEyeIris[0][1])/e[n].box[3],f=Math.abs(e[n].mesh[374][1]-e[n].annotations.leftEyeIris[0][1])/e[n].box[3];(f<.01||h<.01||f>.022||h>.022)&&(u=!1),(f<.01||h<.01)&&t.push({iris:n,gesture:"looking down"}),(f>.022||h>.022)&&t.push({iris:n,gesture:"looking up"}),u&&t.push({iris:n,gesture:"looking center"})}return t},eR=e=>{if(!e)return[];let t=[];for(let n=0;n<e.length;n++){let s=[];if(e[n].annotations)for(let[r,a]of Object.entries(e[n].annotations))r!=="palmBase"&&Array.isArray(a)&&a[0]&&s.push({name:r.toLowerCase(),position:a[0]});if(s&&s.length>0){let r=s.reduce((o,i)=>(o.position[2]||0)<(i.position[2]||0)?o:i);t.push({hand:n,gesture:`${r.name} forward`});let a=s.reduce((o,i)=>o.position[1]<i.position[1]?o:i);t.push({hand:n,gesture:`${a.name} up`})}if(e[n].keypoints){let r=hE(e[n].keypoints);for(let a of r)t.push({hand:n,gesture:a.name})}}return t};var Ee={face:[],body:[],hand:[],gesture:[],object:[],persons:[],performance:{},timestamp:0,error:null},L4=0;function tR(e,t){var o,i,l,u,c,p,d,h,f,m,g,y,x,A,b,w,k,S,E,R,$,_,P,C,F,G,K;let n=ue();if(!e)return{face:[],body:[],hand:[],gesture:[],object:[],persons:[],performance:{},timestamp:0,error:null};let s=Date.now()-e.timestamp,r=s<1e3?8-Math.log(s+1):1;if(e.canvas&&(Ee.canvas=e.canvas),e.error&&(Ee.error=e.error),!Ee.body||e.body.length!==Ee.body.length)Ee.body=JSON.parse(JSON.stringify(e.body));else for(let z=0;z<e.body.length;z++){let Z=e.body[z].box.map((Q,ae)=>((r-1)*Ee.body[z].box[ae]+Q)/r),J=e.body[z].boxRaw.map((Q,ae)=>((r-1)*Ee.body[z].boxRaw[ae]+Q)/r),te=e.body[z].keypoints.map((Q,ae)=>{var ie,me,we,Re,_e,Be,He,ot,pt;return{score:Q.score,part:Q.part,position:[Ee.body[z].keypoints[ae]?((r-1)*(Ee.body[z].keypoints[ae].position[0]||0)+(Q.position[0]||0))/r:Q.position[0],Ee.body[z].keypoints[ae]?((r-1)*(Ee.body[z].keypoints[ae].position[1]||0)+(Q.position[1]||0))/r:Q.position[1],Ee.body[z].keypoints[ae]?((r-1)*(Ee.body[z].keypoints[ae].position[2]||0)+(Q.position[2]||0))/r:Q.position[2]],positionRaw:[Ee.body[z].keypoints[ae]?((r-1)*(Ee.body[z].keypoints[ae].positionRaw[0]||0)+(Q.positionRaw[0]||0))/r:Q.positionRaw[0],Ee.body[z].keypoints[ae]?((r-1)*(Ee.body[z].keypoints[ae].positionRaw[1]||0)+(Q.positionRaw[1]||0))/r:Q.positionRaw[1],Ee.body[z].keypoints[ae]?((r-1)*(Ee.body[z].keypoints[ae].positionRaw[2]||0)+(Q.positionRaw[2]||0))/r:Q.positionRaw[2]],distance:[Ee.body[z].keypoints[ae]?((r-1)*(((ie=Ee.body[z].keypoints[ae].distance)==null?void 0:ie[0])||0)+(((me=Q.distance)==null?void 0:me[0])||0))/r:(we=Q.distance)==null?void 0:we[0],Ee.body[z].keypoints[ae]?((r-1)*(((Re=Ee.body[z].keypoints[ae].distance)==null?void 0:Re[1])||0)+(((_e=Q.distance)==null?void 0:_e[1])||0))/r:(Be=Q.distance)==null?void 0:Be[1],Ee.body[z].keypoints[ae]?((r-1)*(((He=Ee.body[z].keypoints[ae].distance)==null?void 0:He[2])||0)+(((ot=Q.distance)==null?void 0:ot[2])||0))/r:(pt=Q.distance)==null?void 0:pt[2]]}}),B={},oe={connected:{}};(i=(o=t.body)==null?void 0:o.modelPath)!=null&&i.includes("efficientpose")?oe=q2:(u=(l=t.body)==null?void 0:l.modelPath)!=null&&u.includes("blazepose")?oe=V2:(p=(c=t.body)==null?void 0:c.modelPath)!=null&&p.includes("movenet")&&(oe=Zh);for(let[Q,ae]of Object.entries(oe.connected)){let ie=[];for(let me=0;me<ae.length-1;me++){let we=te.find(_e=>_e.part===ae[me]),Re=te.find(_e=>_e.part===ae[me+1]);we&&Re&&ie.push([we.position,Re.position])}B[Q]=ie}Ee.body[z]={...e.body[z],box:Z,boxRaw:J,keypoints:te,annotations:B}}if(!Ee.hand||e.hand.length!==Ee.hand.length)Ee.hand=JSON.parse(JSON.stringify(e.hand));else for(let z=0;z<e.hand.length;z++){let Z=e.hand[z].box.map((oe,Q)=>((r-1)*Ee.hand[z].box[Q]+oe)/r),J=e.hand[z].boxRaw.map((oe,Q)=>((r-1)*Ee.hand[z].boxRaw[Q]+oe)/r);Ee.hand[z].keypoints.length!==e.hand[z].keypoints.length&&(Ee.hand[z].keypoints=e.hand[z].keypoints);let te=e.hand[z].keypoints&&e.hand[z].keypoints.length>0?e.hand[z].keypoints.map((oe,Q)=>oe.map((ae,ie)=>((r-1)*(Ee.hand[z].keypoints[Q][ie]||1)+(ae||0))/r)):[],B={};if(Object.keys(Ee.hand[z].annotations).length!==Object.keys(e.hand[z].annotations).length)Ee.hand[z].annotations=e.hand[z].annotations,B=Ee.hand[z].annotations;else if(e.hand[z].annotations)for(let oe of Object.keys(e.hand[z].annotations))B[oe]=e.hand[z].annotations[oe]&&e.hand[z].annotations[oe][0]?e.hand[z].annotations[oe].map((Q,ae)=>Q.map((ie,me)=>((r-1)*Ee.hand[z].annotations[oe][ae][me]+ie)/r)):null;Ee.hand[z]={...e.hand[z],box:Z,boxRaw:J,keypoints:te,annotations:B}}if(!Ee.face||e.face.length!==Ee.face.length)Ee.face=JSON.parse(JSON.stringify(e.face));else for(let z=0;z<e.face.length;z++){let Z=e.face[z].box.map((te,B)=>((r-1)*Ee.face[z].box[B]+te)/r),J=e.face[z].boxRaw.map((te,B)=>((r-1)*Ee.face[z].boxRaw[B]+te)/r);if(e.face[z].rotation){let te={matrix:[0,0,0,0,0,0,0,0,0],angle:{roll:0,yaw:0,pitch:0},gaze:{bearing:0,strength:0}};te.matrix=(d=e.face[z].rotation)==null?void 0:d.matrix,te.angle={roll:((r-1)*(((f=(h=Ee.face[z].rotation)==null?void 0:h.angle)==null?void 0:f.roll)||0)+(((g=(m=e.face[z].rotation)==null?void 0:m.angle)==null?void 0:g.roll)||0))/r,yaw:((r-1)*(((x=(y=Ee.face[z].rotation)==null?void 0:y.angle)==null?void 0:x.yaw)||0)+(((b=(A=e.face[z].rotation)==null?void 0:A.angle)==null?void 0:b.yaw)||0))/r,pitch:((r-1)*(((k=(w=Ee.face[z].rotation)==null?void 0:w.angle)==null?void 0:k.pitch)||0)+(((E=(S=e.face[z].rotation)==null?void 0:S.angle)==null?void 0:E.pitch)||0))/r},te.gaze={bearing:((r-1)*((($=(R=Ee.face[z].rotation)==null?void 0:R.gaze)==null?void 0:$.bearing)||0)+(((P=(_=e.face[z].rotation)==null?void 0:_.gaze)==null?void 0:P.bearing)||0))/r,strength:((r-1)*(((F=(C=Ee.face[z].rotation)==null?void 0:C.gaze)==null?void 0:F.strength)||0)+(((K=(G=e.face[z].rotation)==null?void 0:G.gaze)==null?void 0:K.strength)||0))/r},Ee.face[z]={...e.face[z],rotation:te,box:Z,boxRaw:J}}Ee.face[z]={...e.face[z],box:Z,boxRaw:J}}if(!Ee.object||e.object.length!==Ee.object.length)Ee.object=JSON.parse(JSON.stringify(e.object));else for(let z=0;z<e.object.length;z++){let Z=e.object[z].box.map((te,B)=>((r-1)*Ee.object[z].box[B]+te)/r),J=e.object[z].boxRaw.map((te,B)=>((r-1)*Ee.object[z].boxRaw[B]+te)/r);Ee.object[z]={...e.object[z],box:Z,boxRaw:J}}if(e.persons){let z=e.persons;if(!Ee.persons||z.length!==Ee.persons.length)Ee.persons=JSON.parse(JSON.stringify(z));else for(let Z=0;Z<z.length;Z++)Ee.persons[Z].box=z[Z].box.map((J,te)=>((r-1)*Ee.persons[Z].box[te]+J)/r)}e.gesture&&(Ee.gesture=e.gesture);let a=ue();return L4=pe.perfadd?L4+Math.round(a-n):Math.round(a-n),e.performance&&(Ee.performance={...e.performance,interpolate:L4}),Ee}var V4={};ia(V4,{distance:()=>ef,match:()=>W4,similarity:()=>B4});function ef(e,t,n={order:2,multiplier:25}){let s=0;for(let r=0;r<e.length;r++){let a=!n.order||n.order===2?e[r]-t[r]:Math.abs(e[r]-t[r]);s+=!n.order||n.order===2?a*a:a**n.order}return(n.multiplier||20)*s}var nR=(e,t,n,s)=>{if(e===0)return 1;let r=t===2?Math.sqrt(e):e**(1/t),a=(1-r/100-n)/(s-n);return Math.max(Math.min(a,1),0)};function B4(e,t,n={order:2,multiplier:25,min:.2,max:.8}){let s=ef(e,t,n);return nR(s,n.order||2,n.min||0,n.max||1)}function W4(e,t,n={order:2,multiplier:25,threshold:0,min:.2,max:.8}){if(!Array.isArray(e)||!Array.isArray(t)||e.length<64||t.length===0)return{index:-1,distance:Number.POSITIVE_INFINITY,similarity:0};let s=Number.MAX_SAFE_INTEGER,r=-1;for(let o=0;o<t.length;o++){let i=t[o].length===e.length?ef(e,t[o],n):Number.MAX_SAFE_INTEGER;if(i<s&&(s=i,r=o),s<(n.threshold||0))break}let a=nR(s,n.order||2,n.min||0,n.max||1);return{index:r,distance:s,similarity:a}}function sR(e,t,n,s,r){var i,l,u,c,p,d,h,f,m,g,y,x,A,b,w,k;let a=0,o=[];for(let S of e){let E={id:a++,face:S,body:null,hands:{left:null,right:null},gestures:[],box:[0,0,0,0]};for(let F of t)S.box[0]>F.box[0]&&S.box[0]<F.box[0]+F.box[2]&&S.box[1]+S.box[3]>F.box[1]&&S.box[1]+S.box[3]<F.box[1]+F.box[3]&&(E.body=F);if(E.body)for(let F of n)F.box[0]+F.box[2]>E.body.box[0]&&F.box[0]+F.box[2]<E.body.box[0]+E.body.box[2]&&F.box[1]+F.box[3]>E.body.box[1]&&F.box[1]+F.box[3]<E.body.box[1]+E.body.box[3]&&E.hands&&(E.hands.left=F),F.box[0]<E.body.box[0]+E.body.box[2]&&F.box[0]>E.body.box[0]&&F.box[1]+F.box[3]>E.body.box[1]&&F.box[1]+F.box[3]<E.body.box[1]+E.body.box[3]&&E.hands&&(E.hands.right=F);for(let F of s)F.face!==void 0&&F.face===S.id?(i=E.gestures)==null||i.push(F):F.iris!==void 0&&F.iris===S.id?(l=E.gestures)==null||l.push(F):F.body!==void 0&&F.body===((u=E.body)==null?void 0:u.id)?(c=E.gestures)==null||c.push(F):F.hand!==void 0&&F.hand===((d=(p=E.hands)==null?void 0:p.left)==null?void 0:d.id)?(h=E.gestures)==null||h.push(F):F.hand!==void 0&&F.hand===((m=(f=E.hands)==null?void 0:f.right)==null?void 0:m.id)&&((g=E.gestures)==null||g.push(F));let R=[],$=[],_=F=>{F&&F.length===4&&(R.push(F[0],F[0]+F[2]),$.push(F[1],F[1]+F[3]))};_((y=E.face)==null?void 0:y.box),_((x=E.body)==null?void 0:x.box),_((b=(A=E.hands)==null?void 0:A.left)==null?void 0:b.box),_((k=(w=E.hands)==null?void 0:w.right)==null?void 0:k.box);let P=Math.min(...R),C=Math.min(...$);E.box=[P,C,Math.max(...R)-P,Math.max(...$)-C],r&&r[1]&&r[2]&&(E.boxRaw=[E.box[0]/r[2],E.box[1]/r[1],E.box[2]/r[2],E.box[3]/r[1]]),o.push(E)}return o}var y1=`
|
|
/9j/4AAQSkZJRgABAQEAYABgAAD/4QBoRXhpZgAATU0AKgAAAAgABAEaAAUAAAABAAAAPgEbAAUA
|
|
AAABAAAARgEoAAMAAAABAAIAAAExAAIAAAARAAAATgAAAAAAAABgAAAAAQAAAGAAAAABcGFpbnQu
|
|
bmV0IDQuMi4xMwAA/9sAQwAGBAUGBQQGBgUGBwcGCAoQCgoJCQoUDg8MEBcUGBgXFBYWGh0lHxob
|
|
IxwWFiAsICMmJykqKRkfLTAtKDAlKCko/9sAQwEHBwcKCAoTCgoTKBoWGigoKCgoKCgoKCgoKCgo
|
|
KCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgo/8AAEQgBAAEAAwEhAAIRAQMRAf/E
|
|
AB8AAAEFAQEBAQEBAAAAAAAAAAABAgMEBQYHCAkKC//EALUQAAIBAwMCBAMFBQQEAAABfQECAwAE
|
|
EQUSITFBBhNRYQcicRQygZGhCCNCscEVUtHwJDNicoIJChYXGBkaJSYnKCkqNDU2Nzg5OkNERUZH
|
|
SElKU1RVVldYWVpjZGVmZ2hpanN0dXZ3eHl6g4SFhoeIiYqSk5SVlpeYmZqio6Slpqeoqaqys7S1
|
|
tre4ubrCw8TFxsfIycrS09TV1tfY2drh4uPk5ebn6Onq8fLz9PX29/j5+v/EAB8BAAMBAQEBAQEB
|
|
AQEAAAAAAAABAgMEBQYHCAkKC//EALURAAIBAgQEAwQHBQQEAAECdwABAgMRBAUhMQYSQVEHYXET
|
|
IjKBCBRCkaGxwQkjM1LwFWJy0QoWJDThJfEXGBkaJicoKSo1Njc4OTpDREVGR0hJSlNUVVZXWFla
|
|
Y2RlZmdoaWpzdHV2d3h5eoKDhIWGh4iJipKTlJWWl5iZmqKjpKWmp6ipqrKztLW2t7i5usLDxMXG
|
|
x8jJytLT1NXW19jZ2uLj5OXm5+jp6vLz9PX29/j5+v/aAAwDAQACEQMRAD8A+qaKACigApGOKAML
|
|
Xp8xlF5A7V4X8RtYs7PzfNImnx8sa8Kp9z3q2tEgp6angWs62ZZ5CTGoJ6DArGNz5p+UrID6EUrF
|
|
PUlW1EuN0XNW7PQ2L5j3JnoKXN0KijqNP0eYoqXBdgPuuo+ZPeupisWn2Jd4+0r924XgsQOCff3/
|
|
AJ1FzRKxDqGii6m3siiQ8F1XGfXI6YNWLfRbiRQMkcZI9fpTDluT2/h6Qy8gDPbtmtG38JeY480Z
|
|
5zSLUTZg8M28YwYxjAArXtdPt402qgHbpSaLWhma3o0Uqk7Nx9DWLaaVblgPs6qRyds2M/gRSQp9
|
|
zZOni2iWS2hlQ+kjYz9OMGrdjq89vIPPVhj+8M/lQyDq9P1WOYBlMZz1AOD+VdDaTiReOKulK0jO
|
|
tHmi0WDTlr0TyxRVhT8tJjIX+9SUxHXUV553BRQAVBcPhSBTSuxPY86+IGti0s5I7dsORy9fM3i6
|
|
8e8mfDO5P90ZrWWiJicNPpZZtxV/xrW0jQt4DOv6Vk2dEEdTY6BHuB25rpbPSo0QARjP0qTRI17W
|
|
wA/hFaMWmoQMgflQXYsDS142rU9tpqqenfNA7GgtihxkdKuRW6qMY/GkDZY8sY4Ap4hXbyB+VArk
|
|
EtuH4wPyrk/EGkOm+a3jw3suRQLc5i38SX9hJ9nnY+XnBUdPyNdFY6pa3KkkAE9l6f8AfJ/pSJT6
|
|
GhDmI+Zb4ZRycdv6ium0nUhKFydrelTsNnS2829RnrVgV6NKXNG55lWPLIM81Op+WrZkRMfmNNzT
|
|
A7GivPO4KKAEY4XNYWt3vkwPg4OK0giJdjw/xrqhm87Zs8tc7pX5A+leSajf6aHYJ50kn4AZpTep
|
|
rBWRm2Vobm4BXfyehPFdnpmnBFUY5rI2SN63tlToK0YI+KZpFF+3QdavwoKTLtoW0Toaswpk5pCb
|
|
LCxipAhoIuP2dKevHXoaYDylRyxhlwRQI4nxVoCXWZI1GfpXGtbSWjYPGP73+NIGupt6TqMsLruZ
|
|
ih4xnP5V09mQ+JLd8gn0xSYJnVaVdkook69K34zuUGunDS3Rx4qOzHVIp4rrOMY3NJQI7GivPO8K
|
|
KAILt9kZrz3xlebYiu8KCCWb0XvW0NFch6ysfO3jLVjfXLIn+pQkKorl7WxNxIPl71g2dUUdpo+l
|
|
pBGvHPet23iC8ihFosrxirkHQUFo0IF4FXI1O726CpKLacCrMJoJLYHAPpTwucHpSRJJ5e4AZI9x
|
|
UqpxzVpCuOC8cUpQUMRnXttuB4rjNdsYyeVwfXpmpGmcvcQyafMCFJjPY10eg34BUg4DcZP8jUO4
|
|
HaRq3lLNF+IHet7R7jz7c56rwa2wz9+xhiVeFy/T1PFegeaNPWigDsc0ZrzzvDNIaAM7VpNqdegr
|
|
xL4l6kywyRhseZ19lrdfAZL4jxYg3Fw20d63tJsdrDI5rm3Z3R0R0Mce1eKnQYAplIkWrMJ45oZS
|
|
NO3PHbNXIyfpSGWowSOasxLUiZdjFSqtNEMkUemKlAGKsRJjAppFAiORMjmsTVrNZEO4cfSoZSOD
|
|
1eJ7WXBUzQZ+7nkfSo7e2Ei+ZaMzxntjBX2NSU1Y6/wxqojiEFzkA8KTXYaUoWRyv3W5rSjpNHPX
|
|
+BmpSg8V6J5gUUAdhRXnneFFAGHrTfu5PpXzj8S70/aZtxzztXFbv4DKHxHI+H4GZiz9zxXXW8G3
|
|
GBXMjvLRXAx0oPGPSmMVeOnWrMTYpFI0bcg1fh54xmgovRcD3qxETSIZcRvzp+/BpEkqsBUqsM9K
|
|
q4Em4Gkxk0yRGXrVW6i8yFhkg+tJjRxGsWrxllkUMh9eK5uMz6bcebbnfG33kPcVkay2OntPKuo0
|
|
nhXI67c8qa7Lw3c+adjcEDGK1paSRhVV4s6A0or0jyRRQ1AHX0V553hRQBz+vNtt5z3xXzX8Qbdm
|
|
uic5YnOMdK3l8JnTXvlbwpYl+WySOgrp5YfLOOB9O1c62O7qQkc+9RsKChFPWp4DluOlSykaNruH
|
|
ArUgHShFNF2NT1qxGO3NBmyxGcE1N2560CFzjrUysO9JAPDDjFOVuKoQuSRTWouBkazbCa3cd8cV
|
|
wF7IISQccHBzUSWpV9C3o1x5b5GAjdQD1rs9DjC3kckbEhqKfxIzn8LOupRXqnkPccBSkUAzraK8
|
|
87wooA5rxMSI3HqK8B8bQl9Q8sffY5b/AAraXwkUviNrw9pH2W1ViMMRTdRjw4HpWNtDti9TPc4P
|
|
FQs2M5qdyyMHLcfjV63HTAoBGtap0wK0YxigpsuRDtVhVYd6GQydVwwIqdRnqKCR23I5pCMUW6gD
|
|
YNKuetAEise9KTxQBWuFyhrznxNZkXjFeN3I+tTIZg2OqmzmxNF0PO3vXp/g2+hukVl4zyPanTXv
|
|
JmVR+60dpThXpnlPceopWFAbnV0V553hSGgRynjC5FujOey14Ssp1HxNmTnc+a3kvcIpv37HoEYQ
|
|
QmMdVHSsnVbYJF5jVk0dsNzlruVIsl2wKxbjWrVHILjg1CRbZJb+ILHPzyhfStODWLQgFJFYd+el
|
|
UJM27HUIXxhga1Y5lLVLKLkMnoauxnPPrSEx7ShF+Y/n2qrc6xBbhizDAqkK1zJuvG9nbg8ZA681
|
|
ly/Ei052RO3uKAsZlx8QGd8xxvt9Aa1NH8dK7AXMcip64zigdkdrZX8F7EJLdwwNXMkrz1qRMRly
|
|
CK4TxmpidWI49felPYSOMmi80NIoOV6qRzXYeA5SskYPfirpfEjGr8LPWVHyD6U4CvQPL3ZItOYc
|
|
UDOoNFeed4Uhpks4H4iE/Z5MeleMeGULeLgjds10S+BGdL+Jc9OSBU2Huc5Nc74yvUtrcDBrJnZF
|
|
63PJdXvLy/lKWw46bvQVz82jXhkLO5Y+9ZlsYthcRnbIjY9R3q3awTRkEM3WmJI6C0ea3dGRsr1x
|
|
XY6TqW9FLHnjrUs0izpLK5DDjofSta3ckH09KRUkZuuTvFGdvPauE1Y3U6Mqbssf/rUxHPTaJPK2
|
|
ZmJPbBqzY6DCZh5xJC9s9aBJHU6dpemJjfEmfetJtI0+VPkUr/unFOxdiextHs33W07YHQHk11mk
|
|
Xb3KbZ1xIvcd6LEyWho4Nct41sTPYb16ipexCPPZN+wYGCvH1rrPAEJmvkPoc1VL4kZVvgZ6yFwK
|
|
cBXoHkkqinFaVyzo80GuE7WJRQSziPiGdthK5HQV4x4J/wBI8WPIewNdEvgRNL42emO/yj1UHNef
|
|
eNpRczbC+I17DvWT2OqJxc0sMK4TCisy41q0hfEkqj8aixdwTXNOlwvmqD9anS9tXH7uVG+hosO4
|
|
/wC0oOhrR0+6G4YNIEzsNEuCxAPNdjZruA4xxUmjINSjURksOlcbqFykbnjFA1sYGoassaknCqO5
|
|
rl7rxhGm7yBnBxuJq0rkSlYpw+NLlsfd5P8AerVsvHEqSBHwPVgcgVpyMyVXU3rXxcHYETAk+hru
|
|
/DWti6ZSTyOKzZqndHaxvvUGq2rQ+dYyqR24qWI8dvbr7LqDxyDAzXpvw6FvIxePGSM06Xxoyr/A
|
|
zviKFHNegeX1J41zUhXioGbuaSuM6wpCaBHG/EcA6HN/exxXjXw2jL67cv8A3Qa6H8CFR+NnoWpO
|
|
I4XI44rxLxrqjQzSEsQM1gdSPM9U1uR1YbmWIdXHf2rmpIb67YS28UrRlsLI3c/jW0VZGUpO5pW1
|
|
jfLNOjahawzwReYI5cjzMkDavHJ5/SrVv9uhtPtVxCPLBwzxnlT9KGghLU3tKvvPjHzbl7EGuisJ
|
|
GRxWLOg7nRXJEbDjmvSNK+aFSfSoZr0KutRkphc4NcRrdkVjL9aVio7Hk3iqS8ubhrWzUlsZY9kG
|
|
cZNc5D4aee5MclzJIFTzHAO0MfatqSOWu7bFS1srDUZEis0vIZoUxPvfcC+4/dx2xjr712XiTwXb
|
|
WmlQ6hol3cRhoFd4rlg3zY5wR0GelavQwjq7GD4etdVvSnk2wAB+9v8A8mvcfA2kXiRo0/UdcDis
|
|
ZnTTulqeoWqbUAJqWUb42X1FZlnjfjSwlGrr5S/eNdD4RkvLAAQ4yRyaUZcruVKl7TQ9I0G+mnzH
|
|
ckFwM8VuIK7ac3KF2eXiKapz5UWYxipNtMyNejNch0jSar3cjR27uoyQCRVRWom9DxTx54gu5fMi
|
|
lbKdMVjfCZPNlv5v9rFbVHpYqjGzbOn8SzFI9o715L4u0r7arYzk+lYdTqSujy7U/C0u4vHk+WwO
|
|
xuh9q3J9dgvbdVukMV1EwbDDgn04rZMwlHoZ+orZ6hfQ3RWVnQYCgZAq+8U0ln5NtBsV2yxYcfgK
|
|
JtW0CnB31LlroVwJ1nQLGDjeP7w+lb0dsFxjrWB0tHS6NuWPJ6A16ToUm63T3Gallr4S7cxiTjrX
|
|
PaxaF7dlVeSMUhxZ5jd+H7qCa4eF3DSE5x3zXN3Wk6jbyeaiFWUY6ZyPStYS5SalPmVipFbX0E4c
|
|
W0alvmPHJrag0rVvEE6LdljGpG2NRtQD+tW5XMI0uU9M8NeFo9PiQhecDIIrtrOMIoG3H4VlJm9t
|
|
C6CB06VPGM1IHLeItGS6uw+ORT7e3jsbQvj7gzUNam0JaWE+HN7NqOqX80n3FO1RXo8YzXdS+BHk
|
|
4z+KyzGPapcU2YIv7qQtiuaxvcaWqG4O6FwfSrS1JbPnrxoxkv7qIfejcitj4V2f2exumI+8+aKn
|
|
xHTT+G5d8Txlm4rjLxMsQwzWT3OiK0Mm6sEkVsAcjFc1d+FEmlGwEDPQVopaEuOpr6f4ZWNAu3tW
|
|
vHpAj5ZQcUFIWaDjGMVUMQ3cVDBmvbhY7QAV2nh+T/R1yeKhlrY31+b61FcQK6nIoJMi401WblRi
|
|
qr6PCw5UYq9y+YgOgWzNkRrx3xWjp+nx2v3FQcelAbmko9anQ4GBUNisPHWr1qMrQhS2K11HvmYV
|
|
hamcxSRZ5xRIqluS/DKAQQXZxyXrvo2FdlL4EeZjH+/ZbjNSZpswLNBrE1Gt7VE4ODVIlnh/j61F
|
|
j4lmeTGyUbq6LwdEqWbeX0YbhSqfEddP4Bddj4JIrhL5d8h7VjI6oLQqKNzelWre3yc4/ClFjaL6
|
|
wqBxxUUxwCKu5BmXRA6c+9ZjP83FSBoQuPs4BrsNBlUW659KmRrDY6G1lyQtW3Hy0lqQ1qVJnAbm
|
|
oy3b9KYJCqRj3o4zRctIlhjLHmpSuOBRbQOpLGpPFaES7UqkZzKN1KsEc87/AHUUmvPLTVGv72aQ
|
|
k7WJwKmRrQ3ud74Ltilgz4++2a6iNDXdS0gjyMU71my7GpqTbxSbMki3SViajTTHqkSeR/GeyZmg
|
|
nQHkEE1S+F+oPPavBL96I4/Cia1udVF+4dVrkW+Fq8+v4tjMDWUkdVJ6WM0cNV+F+MVmjUcZgqnP
|
|
1qpNNnkcVRLiZtxIS1UzzIF7mghlxUZpVQdq6nTVdAoAOKzkbQWhvwM6gMM1twOJYx3NOJE11Kt1
|
|
H1/pVVlwBkk+9NocXoOQ45FPj+fkUJFF2NSB700v/hTEty5ZpkjvVyUgcCq6GM9zC14/8Se6GcZQ
|
|
1574Xs5WkI2HBPHFQ1dm1KSSZ7Rotn9l0+KPHIHNacae1dy0Vjxaj5ptlhVp+2s2CJ9ppCKzuWNx
|
|
zSFc1SYrHNeNdIGpaYw25ZeRXmvheyk0jVpEdcLJ0q3ZxNKTa0O3vQHg/DNcHrsJDmsmjspnNzNt
|
|
fFIJ24GazOhC+azDmgZIOOKBsp3J2qSaZodubq58yQ4QAnmhGT3NO18pb7BORmu205LfYpyKVkWp
|
|
Oxr5gKYWoIZWgfGfloFq1qTPLubnGO1RPtxg4P0oBAkY/hBz6VNDDkZ6AU0W2WSdqkdKr9ZOaGSj
|
|
VtcLHmnOcgmmYvcz7mBLy3MbdD1q9ouiRK6bUAVeelOC1InPlidSsWMDFOCEdq3uefykqrinYqGy
|
|
rFvApMVka2DAowKAsMkRXQqwyDXn/iWyitNQ3qPl6itIvRoF8RXinW4tQ6HI6GuW8SIVBPalc6qe
|
|
5x9x97r3qruwTjrWZ0ksZ9TUmcDNAmZ9/wAoao63rR0+w22MLPtAzt6mghmfofiB76LdJBJBIp5D
|
|
d/oa7bSdWLIPnpDi9TM8TeKdas51XTbIyxd3J/pXS+E/EFxqNoFu7do5OmD60maHWrnZyDRkn/69
|
|
MlEyOR0xntVoNx+FUgYjPxg4FLCuWDZyKQr2RoRnP0qO+nEFpJITgAUzLqZnhu6+0rknOTXpOmwJ
|
|
Fbrt5yMmnHYyr6Oxb2ijaKLnPYMClwKQWK3n0hn+lachHOJ9pNNN0apQFzsY10a4v4hXQh0xpieQ
|
|
MA1XLZNjhK80cT8OdV+3Wl3A7ZZJCw+hrR1qLcjZ/CsbnfHRnFXseHJArOYYbrUs1uPhYbuatqFP
|
|
ByfSkMq3UIINYkto+87Tx6GkSxfsDbflGD7CtTw/pk4nzITtPIFMFudsukh4Rxz71paTpKwP5jcn
|
|
0qTRy0NORMDgVCqewoJTJgAoxjntTiTu7fWmFxAcnn1q3EPl+X8KZMi4gKqB1Peob/Tv7Us5bfeU
|
|
yOoq4R5nYxqT5I8xieH9J1DTbvyJELRg8ODwa9Ms5mSFV9BWiptbnNVrKdmif7Q1KLg96XIZc5Is
|
|
pNL5pqeUrmMtZs0jzV08phchaY00zH1p2ZNxjS1g+LdJOt6U9ssmxjyGp2urDjLlaZzng/wUPDqz
|
|
TSTmWeTrjpVjVk3Rvjr2rnqQ5dDvo1XUd2cTqSNk9OKxXGCeKxZ1DAxHTr2q5C/y8GokUhsz54qu
|
|
uCxzSQjQ0+FZblR2ro4bZYiMVQ0dBb7Qi5x0qzuG5QOh71LYErDufpSeWrHnimIXbjkUjLkH1Hem
|
|
gGxryc+tXI19KYmWegq9YLiLJ7mtqS945cS7QsWehqxA9dEjz4krPSxyZqbFFhGxUm6smjRM55Lk
|
|
HvSvNxXTY57kLT+9MNwKdhXGm5FIbkU7Bca1wMEVhaiuQcVhXWiZ14R6tHGanGBI2OtYkqEHjgVy
|
|
s9ErEeo6UBsHipKEZs5qpPdRxcbhx70NCSuybTNWihc5brW9Fq6vjMnFSdEIdDRi8RRKygZbHFbu
|
|
m6nb3RA3gMegNJhOm0jbXGOoxTuCc1Rz3FyoGKawz9KaAVcZqeMgCmIkB4FaUTbYwB6V00Fuzixb
|
|
0SFMuDU8Mlbs4UPeXHeiOXkUrDuXYnyKk3cVk0ap6HMxxketSMhrcwRC0dMMZFMQ3yzSeVQAeUaz
|
|
9Vj8uPd271nVV4m+GdpnHX67pCeKyLtBtNcR6xlk9RVeWTb3qRnO6trgttyIfm71z7ai8j7/AJmN
|
|
DNqUVa5Yi1AnjynHuBV+11YJhWWXcP8AZNSzqgmaEerSsf3NtIQP4mGKtRavdRgMIpVI9KjU0a7n
|
|
R6T43uYQI7qN2Tpkqciu503VVuQGAYZHQjFVc4alPlZrpKGAznpTwxOc9+lWjIlUACnM4XApiLNk
|
|
nmvnsK0NvpXZRVonmYqV52GsmanhXitTmFkSiJTSAvwrxUxXIrJ7miOfjf1pzNWxkRlqYWpgJupu
|
|
6gQbuahvIxPA6eo4pNXVioS5WmefakGhndH4INZs5DJXA10PaTurmLO21uKpSZqGMoXGnRzBiyjd
|
|
9Kx5rcQS428fSkjanLoaOliHGZFB56VswW+mtPufcBsGOAfmxz+tFkd8HpoaUx09FAtFY8DO71qb
|
|
Sms/Nb7RbecG6AEjFLS5c78t+p0djpVs9wsyQiJAdyr1rW+zqjErzSe559Sbk9S3C+MA1bjbgE1S
|
|
MSXzMVG0vNUI2tPKrAuCMnrVzNd0PhR49W/O2xrHmp4TxVMzQshpIzzQBehqesnuaI5VGzT2bitz
|
|
FEbNTC1ADS1JupgG6l3UAc14s04yR/aYRll+8BXCtLncDXFWjys9TCz5oW7GddH5qqNzWDOgQnC8
|
|
VSuo1kHzAGkPYopEY2+RWxV23Vzj5G/Kg3jWaNazhZuqNXS6TaKhB2c0jR1nJWOlhOxRxU4YkCgx
|
|
Y0OQatQyDbyaaFYe8uF4NY3iC9ltbVGj43NTIL3h7WzMihjzXVQXYYDdW9Cf2WcOJpfaRZ3g9KsQ
|
|
mupnCLIabGeaAL0LcVY3cVmzRHIxtUhetzEjZqjLUAIWpN1ArhupwagAfDKQ3Q1594v0c2bm6tx+
|
|
5Y8j+6ayrR5onThp8s7dzkZjuqAAmuBnqC7c0iwgtzSA0rWzjfGRW3ZadDu4AoNYo2rfS4v7orSh
|
|
05UA2r0pDbsTm29KRottBNyJ0wpJ9KhD7f6U0ikNWffIFBz60zVUW52ow4UcUN6EPcx44WsbgOmd
|
|
ua7TT5Bd24KHnFKnLlZFSN4koluLdueRWvp14swweG9DXoxldHlTjYtzGoo25qzEvwtUxas2jRPQ
|
|
5CNqkLVsYoYzUzdQA3dSFqBBmnqaBhuqhriCXTpVIzxUz+Fl03aSPI9QTypW2/dz0qKNw3SvOPZR
|
|
Mqin8VLKRcs3O4Cuk0w/MDjt1NBtHY6O2IIHY1pxgFaETIRwMkjtVSUEk4570MlFW5bap6dKzWm8
|
|
1tqH8aY+hp2FvGoGayNevVt7/ap4xzUvYjqTLtvLPcvJxSaVcyWsxTnFZlnT2t15xHmCtOBYwQy4
|
|
B9q7cPO+jPPxFO2qLEj5HWo42+aus4HpoX4W4FTF+KlotbHII9SFuK0MUNZqiLUDE3UbqBBupwag
|
|
Bc1DefPbyD/ZND2KjujyPWlKzuPesRZjHJXms9lMuw3StjnmphKDSLTJ7OfE3JrpbO4GQc9qlnRA
|
|
3LO82k5NbFvdADkjBoCSHyXIIIzgVQvdRigT7wzjgUzO1jHknlvG7qnp61etYFQDIpCZoqVijzXn
|
|
3iC8EmsOuaCGb/heR/s0ijkVv6fbxy3QMg5xmsnuX0Ldzut3+UYTPWk+2GJSe+M1pFtamcldalmx
|
|
1eO4XaThhWnC+TXqR2PHqL3maUJ4qRjxSEjj42qXdxVmaGs1MJoATfSbqBAG5p6mgAzTJTmNvpQU
|
|
tzzHXY83D/U1zF5FhjgV5r3Pa6FMsV5HWnLe7RhqBRdmTwagN2d2K2rPU1C5LAnPrUs6Iysbdrq6
|
|
f3gK0BrUKj/WClY05iM6xLOcQAj3NT29uznfKSzHuadzNu7NSBFjHNSm5VO9IRnajqoWMhTzXFtA
|
|
bvUfMduSeg702Qz0rS7FbTToQFwzjJqaGTFyfK5PQViyzUuFmuIdgGABya5u/vTaN5cnUHFUmLoZ
|
|
zyskwlgJweSK6zQdUEwVJeGr0aUrxPLxEfe0OrhPAqVjxWhznGRtUwatDK4jNxURbmkAm6jNABup
|
|
6tQAFqhupNtu59qUnZFwV5JHnWsHdIx96w5lz15rzT2uhRmt85xWbcxMnUGmZlB0bdxmrNvFIcfM
|
|
350mWjbs7YkDJY/jW5ZWW4jikWkdNp9mqYJFaJdEHHakUULu/VB1rLn1Ld/FgetMGYd/qWSQmSa0
|
|
/AemS32pfa7piLeLkg9z6UmQtz0W7uQ2cZx0A9BVzR7cAea6j2rPqX0L99KRat5A6Dk1wOoKZ52a
|
|
YfMORTYRLujiGWEq6/NWza2yKQVHNdOHerRy4laJo6TTnbbtb8KuM3Fdh5z3OJjbmpt3FaMxAtUZ
|
|
agBN1GaQBzTwaAAms3VbjERUGsa07RsdeFpuUuY4jUjljWTKK4j02RE4IpJYFk6imQkVl0xWarsO
|
|
mAEcUi0bNnZBR0rWtoguMCkUi21wI161mXuocEKaYXMS4u+pY/hVCSWSY4HT0pEmlouiSahdpEBl
|
|
mOceleiwWcNjClvHgJH97Hc1EmVFFi3Czy7mwIl/WtJbjP7uLgd/apQ2VNVvtsBhiPzdK5S4nAuR
|
|
nqOCaTGi9pcytPlU+XpmumtWII44rah8ZjiNIXRuWeNvvViQ/LXpJWPJbu7nCRvVkNxVsxBmqJmo
|
|
EPiXca0YLMuOlJsuKuPlsSi5IrNuG8s4HWs5VEkbwoOTKsk+FJY4rC1K53k1xTk5O7PSpwVNWRzt
|
|
4cms+WpKICtSLTETQj5q0YeBSGiys23pUguGxQMq3E59ayrm4x3yaAKiRtO2WPHcmhruKFxFajzZ
|
|
ScA44qRHoXhuMaLpxaUg6hcDLMf4F9KlhuDeXGASIl+8azZslYma68y48m1+7nFW5rtbRNhb5z1p
|
|
iMKbUg0zuW4A4rPgb7VdKXOMmpA7HRbMS7nUYiUda0lkQOBngVrS+JGdbWLRt2bAx5BqeQ/LXpnj
|
|
PQ4GJ+ashuK0MhWaoWcA0AaOmASMK7jRNPWYBmHyiuepO2x10qfcv6vYxCzYqoGK4HVYVTJrmb5l
|
|
c6oaM5TUJ8EgGsG4kLNUHT0M64OaqMMikSRsuKbnFMRLG3zVehOaGNE445NNlnVFpDMu6uie9Vo1
|
|
8z5mOAOST2pDK91cNN+5tsrH3PrW54a06KxT7fdrlh/q1Pc+tJ6IUdZGvHPLezMcnBOWbsPap5r3
|
|
ylFtbdT1xUWNWzU0/Zbwlgfmx8zGsHWtRHmMqE59aAMyNifvHPc1f0gtPdqkY5JosJHeNci2tktY
|
|
euPnNY+oXWZEVJNrZ9aun8SIq/CzodHuriIokhDIR1ronbKZr0o6o8ipoz//2Q==`,A1=`
|
|
/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAsICAoIBwsKCQoNDAsNERwSEQ8PESIZGhQcKSQrKigk
|
|
JyctMkA3LTA9MCcnOEw5PUNFSElIKzZPVU5GVEBHSEX/2wBDAQwNDREPESESEiFFLicuRUVFRUVF
|
|
RUVFRUVFRUVFRUVFRUVFRUVFRUVFRUVFRUVFRUVFRUVFRUVFRUVFRUVFRUX/wAARCASwBLADASIA
|
|
AhEBAxEB/8QAGwABAAIDAQEAAAAAAAAAAAAAAAEDAgQFBgf/xABDEAEAAgECBAMECQIDBgUFAQAA
|
|
AQIDBBEFEiExE0FRBiJhcRQjMkJSgZGhsWLBJDNyFSVTY3OSNEPR4fAHFjWCokT/xAAYAQEAAwEA
|
|
AAAAAAAAAAAAAAAAAQIDBP/EACARAQEBAQADAQEBAQEBAAAAAAABAhEDITFBEjJRIhP/2gAMAwEA
|
|
AhEDEQA/APqYAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAKNTq8OkxzfNkisQC8eb1XtRNbzXT4q7eU2nu0MntRq/D8StMccvW29ZmdvgjsTyvZjxOLj
|
|
+s8WLxn8TFPXs6Oj9oct7c14rkxz22nrB2I49KOdTjelmszfmpMeUxv/AA28OqwZ4icWWtt/SUi4
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAmdo3nsPNe0Pt
|
|
Fh09Z0+DNWL7+9O/7A3eJcZppsV5raI27esvH6jX5ddM25p79Ilo59VbUZOe2Tm/PeGvfPfT2iKR
|
|
PLv1+DO678XmW/a97U6TtOyzTbTF538/T9WjTNecm9a7126tqk3rSYxY5ta1plRZqZNXGjyZcPXl
|
|
mZmsx+qjBrsuO16xM7eXRt04JrdTltk5OWJnfaWf0a2lty5MdZnfzSn+WOHiOutFpjHa9e8bQ2fp
|
|
+alYy462pk7zXbuxjPesbRS0f6ZZV1ET1tErzXFLHo+A+1ddZf6NrI8PJHa1vN6iJi0bxMTHwfOa
|
|
zhzd61v1846utwniM6DUdb3nBaNrVmd9vjC/ZVePYirBqMWppz4rxaPgtEAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAItaK1m09ojcHnvarjM8P0vh49+a/eY8ng9D
|
|
h1fGM1rxjtGPfvbzdbjuTJxHX48cTPNltM/KsS9Dw7S49Jp6UpHaGe2vjz1y9J7LYK13vHWe7bj2
|
|
ex1tvM80ekuxW3RnW3Vm6P5jRx8H0+OYmMcb+bapo8GKPdpC6bQwtdHU8JpWkdJ/JweL6e23iU67
|
|
d4dubSqyVi9Zi0bwIs68XGp36TtEq7ZJmZmevzdbifCKWtbJinkt6eTgZPFw32t+sRurbWVzxs1y
|
|
Rv6T8V1NZNPtfq0seTm+Kevr+SZuxXjvaPiV8N4viycto9HseG6+uu08W6Rkj7UPmFck1tE1nlmP
|
|
Ld3eA8V8HVVi1pjq6Ma/pnqce/ERMTETHaUrKgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAADW19+TQ5p/p2bLS4v04Zmt5VjeQeJ4bjnLqsupv+Ka1+ERLv4reTmcNxcuC
|
|
vy3l0qdI2hlr66sT02ot0ZV7qqrInruzrVZLGSZ37JjqgYTG0K5lbaFVhDT1Ub456RPweY4hixWi
|
|
eSdpjvD1eWejz3FNHWYtkpvFo9EIseb3tS3SerOms22rfpPqZKzvvHSYUz70TExG6Gdbs2rljeJ/
|
|
Mx5L0vEzPaelnOi98c9J2bFNTFpit47+a+PVUvx9T9nOIfT+GV5p3yY/ds67wvsXqpxau+G09Lx+
|
|
r3TqrEAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADV4ljnLw3U0jvO
|
|
O0fs2lWqyUw6XLkyfYrWZkHldBEV09eveG3Fq1mI3jd4vPrOIaid8G9MP3Y38k6fNrt/rMk9Ou8s
|
|
tfXXn49rGWInuy8SO/k5Gl1E3rG/fzbOe94wTy99mbRvTrMOOvNfJWsesywniukrG/jU6fF43WYN
|
|
TmtEeJtEQ06aSmK2+bNtEd+qfSO17unF9Hmvy1y13XWyVmN4tExLxVK8PmNq5NrT58zawam+m/yc
|
|
0Xj8NpRYSvQZ7xEOdqI3rPozxayNRXe0ct/ON03jmrKB5nV4q1yTO20Obmv4c+cx8HoeI6WZpNoj
|
|
q83niYmYscU0r8aJ6T1n49zeJ+Meqm1drb9J+Kd5p136StGVem9l9TbHxLDFp7W7+sS+q1nesT6w
|
|
+PcAzVjiGHftzQ+v4f8AJpv6On8jH9ZgIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAABp8VrW/C9TW0ztOO3b5Nxp8VmI4bn37TWYB8f1HFtTfUfR9FWJmsdZ9I7MtJxDX5s
|
|
d8ta1y0xzteaR2277rcuhycP12SceLxMeWNpjttHwlu8I0mfQ1y+D7k5YmJmY36T36Ka43z/AF1t
|
|
cI1ds+qxVj7/AEej19PCw9HJ4NoK4OIU5Y35YmZdzVTGebVZabx5jJS+Tmns81rNLm1Wrzc9rVw4
|
|
Yibbem72mXTTS0w0M3BvEta1bWrM95ie5EanY87wXgNOL6XPfxraXLhra/W28bR/dzYzarBqJxRe
|
|
bzE7Rt5vWU9n8mPHOGmS0Ypnea1naJb+k9ncNLR7u2y/WcxXO4TOoyUrN6zD0FaW5Y3hu49FiwUi
|
|
KxCvLMR0hlW0jn6ukWw3iXjOJzbDlneOj3GaN6zDzfFOH+LE7SRGo83XNSZ2lbG2/WfdlvaT2cy6
|
|
rNFInlrv1mfJ37cK4PwTTxOoidRm2+/2/KFuyMp47XB4LivXiunrH2b2iH2qn2K/J8x4fGDNxTSZ
|
|
9Nh8OviRvTyfT6xtWI+DeXs9MNZubypASqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAOZx6/LoOWPvWiHTcf2hiZ0e8fc2mf1E5+vP/AEeuSd7RC2uKtI6QjHfeINTfwtPf
|
|
Jvty9WPfbt/lucP03gxfJf7d/wBoReYpm97zaNeLb4Ims9Nt94auDjem1Wo5PFi1onylS+1o7l8V
|
|
bxvtupjDMdNkYtXS1+Stt+m63xImEJ4xjHER2ZxMUjeUTO3VRmydBbjLJqPi08mbeVOXJPq1sl5Q
|
|
Vbkz9+rRy35rxHqzmZlVEe/Ez5LRlW5iyfR6zffaIjq1OSNZps2a21rZInafSPJhxGMl9LStLRWM
|
|
lorM/A4dkrWbYfLZC2W/7K6eubX6b4RzT+W76K8b7G6X62cu3Sten59nsm3j+OXz3/0ANGIAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA0OIYfpOHPijvNNo+fdvtXJO18k/
|
|
/OwPFYbz2ls3jx8VqW6xMdWPEdP9D4lkx/dt79flLLHbkxTPwY6nt2512ORTRzE2x4/dpE7cvkme
|
|
E4IrW3hRMxO8THRtU1FKWtvtvK2upx22rzRCtXkqzh2jtF7ZbT122b01ndnpuWuP3Z3+Ky20qDVv
|
|
fauzVy3mejZzNK8dVjqi87KLRLYtXruqvXzkQp7Qoid88R6rcl+WGlW0/Sa22mfhCZOq2x082ix6
|
|
jkm822pO8VrPdr4dNObVeDo8XW3uzMbzK+mvxT7szE27cvnu9j7PcNjSaXx8mOIzZevbrEeic5tN
|
|
+SZnpt8J4fHD9HXHO3PPW0x/DeBtJxx29vaAJQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAKNRim9Z5e89Nl4DzXtVh5babURHrSf7f3ec1+qnDorWrvvt5Pccb0n0zhmWk
|
|
Rvevv1+cPE2rGTFNZU26PFfxwa5dVkjelI2772nZnX6bbrEUq3o0d678u8wmuDL2ittvVjXdneeK
|
|
cGv4jpJ6U56+kS7+j118+GLXpakzHaWlp9NNY3tv+bbiYiNoQy1y30uyZJlrWmZnuym6q1iIJnop
|
|
yW2Te8bdWnnypQqzZOadokiIpSZntWN5lrxki19vNRxrUeBwnNNd+fJEY6/OejXLn3Xe/wDp9wyn
|
|
E8uo4lqqxblv7lJ26T6vpD5X7G8QycKzeBMbzMRM1/FH/wA/h9QwZ6ajDXLitvWzRgsAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAeL45w+dDrZvWv1OWd4+E+j2jX
|
|
12jx67TWw5Y6T2nzifU+rZ1y9eHwzDYxxEy18+DJodXfT5o96vafWPVbjyxDn1OOzHudbM0rt2UW
|
|
iI69mVtRXZq5tREb9VUoy2iIlRbJ0UX1VZ6btTLrI7V6yk62M2oisT1c7JmtkttVMUyZp6x0beDS
|
|
RWOvdKijDimvWd3G9pNRMfRcNfvZOb9Hpb0itJeP47k/3hgjaZnbaP1XxWW3T0movbNS0W645nbf
|
|
0nrMPpXs3xamoxdJiLbe/X1n8Uf3fKsOTw4jbaXo+EarJhtGTHMxeJ6xH7Sti9Zaj6x3HM4NxXFx
|
|
DS1mtoi8dJrv2l011QAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AGjxLhODieOIye7kr9m8d4eM4to9RwjPXFa0ZIvG9bR0fQXmPbDFvTTZPOJmEWS/V8bs9R43NxLL
|
|
G8eFbePg1bajU5/s0l1ceKLx1hbjwRE9mOpx0y2uRTSZsm3PMw2aaKtIjo6kYo9EXpET0hVLXxYK
|
|
xC6MZvyx1lFs0RHfaPiCnU12pLyHGNDbUajBekWma2npWN3p8+opa20e9LSyZLxExTlpM+vdOdcZ
|
|
a9tPS8MyUvFrzWlI6727u1pYxYrbVmb7x+TQx6au3Nqcl7/0rcmW9axGnwZJj1novmxnZXV0fFp4
|
|
ZxLBPgTGK8xzXr5fOH0bFlpmxVyY7Rato3iYfNuG2x56Wrqa8s2jz+7Lu8O12bS6jkwzN6THNNI6
|
|
tvrN68Y4rxlx1vHa0bskAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAA4XtTTm0OKfTJ/aXdcL2pyRGjwU362yb7fkJz9eTxxyZJjyltRXzUZK7TFtl9Lbwy06YzrHwa+
|
|
fJFd/wCVt8m0bQ0eS2qzcm+1K/an+zNZFL5M1pjFXeI72ky48eGnPkvNp27+TPU6nHpMfLXaIjpE
|
|
erk5dRMxOfN1mPeisfshW1ne1a1577Y6x5R3U0zze31FOWI6ze0byU098kRlzbxM9qrMlPDpyRMR
|
|
Md5Vt/Ihp5898mWZm1pjftE91uCt7fCI7dWeHDEW3t723l6rslqxWZnasR+SYhFbzhnfxJ2jyeq9
|
|
lcGXWZcmW0zWKxHLaI7794eJx5fpfEKabT8t8l5isddo3l9S4VjrwrRUwzSJt3tav3pdOL6Y6dXD
|
|
j8HFWm+/KsU4NRXPvtWazHquWVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAa+fXYNP9u8b+kdZBsDkZOO135cWOZn4y5Wu4xqctbe9y19Kp4njt6vi+PDm8DFMWybbzPlV
|
|
5PiGtz67UxbNbeKTtWIjaIXYpnwuaftT5tXJT3vmi1pMsrU5qIrG1V1a+5DCa7b9GFbRr5J6Wnbt
|
|
Cu+Wmk0m8956z8ZWZNorbfzcbX5rZslazPux3hUt41NTntktObJ13+zX1bek01r4/HzVm0bxPXy/
|
|
+bNfDgjVa2uOY92kdfg6ufJOKvLXtttVVSqbcta2vM7zXtHpLQy5ZtMd+vWd+7Zy3mdJHXra3f0c
|
|
vUarw7zFY5rT2hH1Lavnrgx81p3U49Pk4nE5L35MO/StfNRXR5tXnrS8W67WvfyiPSPi7uLHFK1p
|
|
jrtSsbR5Lc4RzsXBaYreP4l45esRD2HD9fnw6evvWvO3Tfr0aGk0U55ra0TFInv6uzgrXFXlx0i0
|
|
77RPlC83Yj+JW7oddqr6vHzTTw9/f6dod+L1t9m0T8pcbFSmPHER3892W0zPuz+jSbVvidkcqmfP
|
|
Sel7bekrI4n4dZnPWIrHeYnZee2Wpy8dEaml4npNZblw5qzb8M9JbYgAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAABEzFYmZnaI7yCXL1XGa0jJXT0571nbee27DiXEprp8nhbxG20W8
|
|
5cbD0ikfnKO+urTPvjoZdXqctdsmTaPSvRpWmsdZ6yztfaGplvv3lWW1tyRlz1x0vkn7Vo5atTNe
|
|
Y0+1o79V2KsZsvX7Ne5mwxnyTNvsx2iGneM/rCdRSuOsTasTt5kRFtpjqmOH4t4nk7estiMNa97R
|
|
Hwhna0iuKTEdmGWa4672nZtRele1N59Zlq6vLOSsYorEc07qcW65euzRvtXvPZy52naZ7ujr6fXV
|
|
rWdukREK8+njHgmZmPc67bq6ivVWhxxgxZLztNrT1mZ/SP4VZs0zaOvfp84WUtNsXLvtv3699+rU
|
|
z7+Jtt5qURqMnPpctaR1rMSw4ZoK57eNk6xHaJRh97Ltt7lo5Z+L1HAPZvVauZ2nFTSzMTzeJEz8
|
|
to6xPfvsZntPZ9rXxabmxzefdrv0j1dXh/BcmstW1qxTHHasR3+b0GPhGl+kWmd64dNEVjf73T7X
|
|
y8vy+Ddx6O3iRakxTH5RXrMw1/lX+3Itw2MFIraN48qRHdZi0cUjmmPen9noox1iO0fNzdXEYrTt
|
|
stcmd9aX0bJ+HePmiKTitO8TMLZ1cVjrMfqpz6ys4pjfrPRWZ9rXXptUit6zO+23VyaRHEc05L1/
|
|
w9J9ys/en1ljqdVbwYw452tlnl3jyjzbmmiMeKtYjpEbLeTXPUU8ee/+qjJpsV5rbkrFqzE1tEbT
|
|
DpYNbW21Mnu29fKWna0KbqTdjXXjld0cvQ63ltGHNPSfs2n+HUbS9c2s2UASqAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAOVxPWe99HpP8ArmP4b+r1EabT3yT3iOkesvMVtN7za07zad5l
|
|
XV5GmM9vVfEstvDx0jtaVVMlq+UJ18b5cMRvPeSuK87bUt+i2Z3PtG7zXpjkzXt6R+TXyTMzvM7t
|
|
ydHqZ+zhv1+Cv/ZuqvPTHMfOYaTMil1a1K2vHSLTELq2v+KWzThGo84rH5rq8JzedqR+ZeI7WnOS
|
|
34pYTafWXR/2Pln/AMyrKOCWnvmiPyR6O1y9585lhWJvl557Q6eo4T4dYiMvW3b3UanhldHpJtGX
|
|
e09unmjsT7eb1l4trI2t0hsZfrdNO0bzy+nzU20/+NmkzO9esz+TZxWis9dttvPv+Tn21jjaW8zn
|
|
26bTG3mp1M/Wzv3t0jyWXiKZJmsTERaZhXXDbNl8WaztWenxZLstPp5pau8frDtVrNMM5cfTfpMf
|
|
3aunxxbes9d/R09Dp8ebJi09ptFr3jtt2WyrW9wy1Jx132mK+Xq9PotT0iIU19ntLtExa3T47T+q
|
|
6nBaYvsZstZ+cT/LeMnUi0TXffo1s2m8Ws2/OIMWk5Jib5L328rS2t94Sh5TV4ppklpW6PT6rh+P
|
|
NbebTHyas8E081mZy5P2W6OFhjxNTE/hr/LoRO0Kvo9dPqctKzMxEx1la5t3tdnjnMs4noievcrO
|
|
yZjeFF1OSnNV0OG62cn1GWffj7Mz5w05joovzY7xes7TE7w0xrjPeex6Ua+j1UarBFu1o6Wj0lsN
|
|
3JfQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACrU5o0+nvlt92P3BxuM6nxNRGCs+7Tv8
|
|
2hToxm1r3m9utrTvMsonqyt7XTmcja0u3O6FMfi5t/u0/lzdJM81p9O3zdvHTwsUR5+bfPqOfX1h
|
|
dqV+3O7bs1+T31oqmI3TEM4rvCdkDGIIhlFd2daboS0NXG2bD6bufxXU1vlmu/u4us/N0+L1tTSx
|
|
kr9qk7w89j1FNZMV3jxLzvaJ8mer+LSOZqK2xZotbvljfr/89U453rXt9lse081xZtNjx7TGKu0t
|
|
DHlrevSevaN5Y6+tJ8c7VRNMt63n3ub+6/R54rERMztDYy4a5omclYmfxKcenrjtHLvtPrCnVmdb
|
|
eFe3JXmjy6eS/DrMuLVYsta9Mdt++6qLxO+0dEc8UmInr18iUfReHcXrqccb9Z27Q61Lb13eJ9nc
|
|
1Z35rTvE9avY4bTkpG8xEfB05vYxqybc07R281naGMREdoT5JQqy9mply7Q3bV3iXG1eXw7TWSka
|
|
c258t7+tpT5/BjT7MfHqndz12Z+M4lMMKyziUJJiN1WSu9fku23RaOgKNJqbaTU1t9yelo+D0cTE
|
|
xEx1iXmM1Nt3W4PqvFweDaffx9vjDbGvxz+TP66QDRiAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAOJxzU73rp6z296zsZMkYsdr2naKxvLyObNOfNfJbvad1dXkaeOdpvsc2yuZVzfbfqybutwu
|
|
s5s8R92J3dvJb3tnO4HSMegtmt3nfZvYp8SZl0z45NfSK7onH1bNcfRFqnUKJr0Y7dVtq7prjEsK
|
|
0XVpEM6028mW20IHK41aPo3J6zs4ODhdcvPnvExFevNXpMOrxi/PlrTee7PLX6Pwa09uaNlKtHg9
|
|
dM3z5d7ReOu02nu0JzZMfblrv5R5uvrcdImZ26T1mYhxs1Os7RH93PZ7axuafNfLitvbaYU3yZYt
|
|
PXs9NwHhui1HBa5LVicsb81onrEuVqNNSuS8Y67dZ6xPZa59Il9uX41vEitImZme3q2Kxbxora0T
|
|
Md/ROSa4Ztkj7c9OafL5LuGYubmyX3iu/TfbdSfVnpvZLT/XZK233+Mbbva1xRXyiPk8pwbH4N6T
|
|
adq5a71n0tD1WDL4tPe6Xr0tDpz8YVnJHWEXYxbqlBedoef4tW0XraO09HdyztSZcbUz43C+ee9b
|
|
SVMaeOfqq7+jGckQ1Yz7+7v2RN/WXPXZPjci2+2yyJaVMuy+uSJlA2d+pNoVRbeDcSxyTE+TDDlt
|
|
pdRXLTynrHrDOyiyZeVFnY9TjvXJjres71tG8MnJ4Nqt4tp7T1jrV1nRL1x2cvABKAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAHJ49qfD09cNZ97JPX5PPw2uI6j6Vrsl/ux7tfk1mWr7dOM8iLdm
|
|
vfebREefRsWldw7SxqNbWbR7lPesrn3Vteo7dYjDpMGCvfbeXQ0uLlxRLRxROfUc34p6fCHYrXlr
|
|
EejqrjY8uzCYW7MZjdVKqK9VlaxCYrsnYExBMRMJRPZA8/xPHtmpP9W2xx76vhWOInvt/C7ike7N
|
|
vwzE9kcapGfhlevTaFbFo8RqJ5vy8/RoW09ek0msxHfp3dzNoLzp4zUmZpMbT8HJyYJi20X2n0lh
|
|
ZY1li/RaidBF4w2mK3jrHaFGp1lN+tptPp5IjBkid5mIp16TKu0abBPv33vPlM7z+iPdFNcWXU5I
|
|
tkrNce/b1W5db1nTaf3ax9q0fxDW1ebNk2phty1mOu09VOm8W19orEz23j1TwfSeERFuEYMddptW
|
|
d43dvBn21eKJ75KbW+cf/JcTgMxXTb3nbljz+TpcPmc2uyZO1KRtVtGVdi0bx07qJnllsRO6rNTe
|
|
N4XVamsy8mnvPwc3R2jPwe8TPbdlxXNOPSZfhWWpwO85OFzv57qrODkzeHntSe8Sn6Rv0a3EZ218
|
|
8nXekfr1a0ZLVnqx19dWb6demXybOO7lYMvNMdW9S/VVLo0us7tPHdtUtEwJiZU3jq2Jhham8CVG
|
|
PNODNTJXvWd3qcWSubFXJWd4tG8PK3pPd1OB6veLaa89Y61/u2xfxh5c/rsgNHOAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAANLimq+i6O0xPv392rdeZ4rq/pOqnlnelOkIt5F8Z7Wj27I2I6sb25YY
|
|
V1ImY3dbQ08LRc23vZp2j5OJG+XJWle9p2h6HHtbJXFT7OOIpX+7TxT31j5rycdTh+Dpz+XaG/sw
|
|
w18PHWseULN2trBE9UcrJKBhFU7JAQi0dEomegNDUYovM7x3jb5tO1ZvpbaTLtzRExWfWPJ08kbT
|
|
Ex5NXWYYyV5omYtHWJieyeDzuizfRs19Jn6TM7Ru1uMcJxZqTkw+5f4ebqa7SV1MR4tdrx2vEfy1
|
|
axqsNOTLjnLXytVXi3Xj8+nmsxTLM16d5npPyUzpekTtSK+U7vS6vQ/SYmK1vWPS1HOn2dvvvvE/
|
|
tDO5XlcO+LbfHSd/W3o6/BdDOXPTnj3Kz38rS6Wm4FNrRyRzTH3p6RH/AKvR8L4dXSzE3jmtHn5I
|
|
mbfqLV+m4dbLSsZInHjr3iI6zLpYaxS01rHuxHRHiT9mv6s67Vj1aqL6326MrWiYa+/Q54BxPaGe
|
|
XRZpj8MquB4+Xg8zPnB7SX30to379GxpK1xcHiKz5IS8xr8PLPixH2bftLTy05o6dHYyVjLhy0t1
|
|
izjZa3pMVv3iO/qz1G2L+NbSajbNyW7xLsY8kTDz+fJXFqKZN4iZnafi6WHL0iYlStI7OO+7axW2
|
|
crFl7dW9jvE9ULN+J3ZbdFGOy+AYWpEqN7afNXLj+1Wd23KrJVMvCzseh0+auow1yU7WhY4fCdV4
|
|
OadPefcvPuz6S7jol649Tl4AJVAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAV581NPhtkvO0R+4NPi2
|
|
r8DB4dJ9+/7Q83Po2NTqLanNbLfvPaPSFDHV66sZ5ET0hRknyW2lTtMyouz0c8usx2n7s7vScKwx
|
|
zc1vu/y85p+maJh6Th+SOWeveXR4/wDLm8v+nX5mUWa9bbrInolmu5jdTNkxYFk2Isr3TuCzeGMz
|
|
+THdEyDDJO9Ja823rt2XWnya946pGvktDXta0ztWu/ybvLE9dkcoOf4GbJPWK1j49VmLh9JtE33v
|
|
Mevb9G7WsW8l1ccREISophiJ2jpDYpijbaOjOuOJ8ujOdqxsgVcsUjaETYvbaFFrgu5lVsm0yUtu
|
|
ryg43H5m+GIj1XcJzePoL4pnrWGtxmfchr8JvfHS1622if3QljzTTLes+qrNjrkiYtCzPMxnm095
|
|
YZJ6boS5teB49Tqscza97VtvWvlv8V/FOF34RrIxTM2xXjelp/eHoeA6XnzReY3ivX/0dfivDcfE
|
|
9HbDbaLx1pb0lOs+jO7K8Lis3cN+0NKcd9PmthzV5clJ2mF9J9GHHVL108dm1SznYr/Ft0tuhLb8
|
|
mNohFbMhLWy0mJ3rPXvDvcO1karBG8/WV6Wj+7kWrvDDBlvpdRGSnbzj1hpjX4z8mOx6UYYstc2O
|
|
uSk71tG7Ns5AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACZ2jeXneJ62dVl5KT9VTt8Z9W9xbWclPo+O
|
|
fft9qfSHEU1pv48ftYST23ZTDC/p0YtlVuvVjMbM5+LCZjYGWGdrTPxiHY4ffaf3cjTxz1v6xMS6
|
|
Olty2iXVj/Dk8n+ndrkhnGRo1v8AFdW3RCrZ5uiYsqrboncSu508yjmZRYQt50TfowYTbYGVrKrT
|
|
uTZjvukQnYhMIGVY2ZxPVWyrHVCWzXpVXkt3TE7Va+W4K7X3jv1auTNy3jdba0RZpamfroQN7Hk3
|
|
6wr1GTaN2OOJiu6Mu98NvgDi8Wy74d/yZ8PiPAiO2zU4nb6qIn1bugjfFE/ASp1ke9u15mbbRDZ1
|
|
Mb823kx0Ontn1OOkedoJCvT8I03gaKsz9q/WW+isRWsVjtHRKyrhe0XCfpWL6Vgr9fjjrEfeh5fF
|
|
feH0V5Dj3DPoOo+k4a/U5J6xH3ZZ7z3228evytOk7NvFbo0cdols47bSybt7HbddHVqUs2aW3Qnq
|
|
xVeu8LILR3SlZw3V/R8nhXn6u0/pLuPMXjeHT4Zruf6jLPvR9mZ8/g1xrvpz+TH7HUAaMAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAABRq9VXSYJyW79qx6yvmdo3l5viGs+maqYrO+OnSvx+KLeLZz2te1rZL2v
|
|
ed7WneZYWnZl5K72YV1xEyxmeqJljzIEWlVkszvbZp5soN3h2SJz3pP3odCnuWmPRxuERfJrZmtZ
|
|
mtY96fR28kbX3dXj/wAuTyf6bmK+9YX1s0cNtm3Sd4LFY2K23W1s16StiUJW7bp22RW3RluBuruz
|
|
mWEgrmCGWyNkoExKE1QlPmsqRDKeyBjaejWy2W3ttDUyz1QKslvehVqKTNosyyTvELabXptIJpaP
|
|
B39Ia2mz+JGpr51jdZefDx2hzuHZObNq58poJaGtjxJ2+LoaKP8ADRPo5+T3skx5OhpOmC0fBNQ0
|
|
5yTbn+bt8A0u9raiY6RHLVwY62mI6zMvaaHBGn0mPHt1iN5+aYVsACBXqMFNTgviyxvW0bSsAeE1
|
|
mkvw7V2w5Ote9besJx2er4rw2nEdNNekZa9aW9JeQjnxZLYskTW9Z2mJY7zz26fHrrdpbZsY7NGt
|
|
mxjvso1b9NmUwpx33XRO4K7VUTE1nmrvEx1bVo2VWiJE/XY4frY1WPlt0y17x6/FuPM0m+HJGTHO
|
|
1qu9pNVXVYt46Xj7VfRtnXXL5MfzexsALsgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHM4jxOMFJphmJv529Dq
|
|
ZLfjDjPEIx450+K3v2+1MeUOHSOWFc3nJkmZnf4yujpVlqunOeFpV2nctLCZUXRM7MJtsWlRkv3Q
|
|
ky5NmpWt9RnrixVm17TtEQnJabXisRMzPSIew9n+CRoccajURvqLx5/chfOest642OGcIpoOG2w7
|
|
ROW9d72+LQvXevyejcPUU5M+SvpLeOataraw2a0dLbLqTtK1G3Es4lVWWUSoldFtmcXUbpidgXzK
|
|
GEW3TuCUSncnsDFMMLSms9EC6J6FpVzbZE5ALy0809ZbFr9GtfrEoFMzuuwz0Ueey3HbaBLDXe7i
|
|
tMOfwWnP9I+NZbuttvhs1uBRtXPb4SDm3iIvf57N7Dbl0VrS5+XrltEd+Z1Jx7cNms9N4TURRw3T
|
|
+PrcO3WszEvZOD7P6aYiMlvu16S7y1QAIAABxOPcLnUY/pWCv1tI96I+9DtgmXl68Biy7/NtUu3+
|
|
O8HnFa2s0tfd75KR5fFyMWTdhrPHVnX9R0cd21S3Rzsdm1iuqs256wrmGcT0RYSx5d047X02SMmO
|
|
esd49YRE9WcdSXhZ2O1p89NRji9J+cei1xMc3wXi+KZj1j1dTTaqmor06WjvWW+ddcu8XK8BZmAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAMMmWmKu952UZ9XFZmuP3revlDTtzWnmvO8q3XGmfHb9ZanV3yxtWeWn7y4es
|
|
vPNtDqZJ6Ts5mppvdl/XXRMyfGvSNlu/RVvtOzLfoipLT1VTKbSpvfogRkvtDVyZOhkyvQcA4Dzz
|
|
XV6yvTvTHMfvK+c9U3rkW+zvA/D21urr789cdZ8vi9KDb45rejl8Rry6iJ/FV1HP4vXbBTJEfYt1
|
|
+UpiHM295bXsqrO9l8QkZ0lZEqqLeyBZHZLGvZkhIndADKJ3TMoqWQMZ6pjsxll2jsCLSrmU2lFY
|
|
36gieyu0LJk3jbsga0wdqzK20QpyztQGprL/AFMrOE05NLkt6qdVWZxNrSe5o9vWBLiUjnzXn0vL
|
|
q555dHt8HOwV928/1z/LpzXxbYccRvzTB+jucOwxh0dI22mY3ltIrHLWIjyjZKyoAAAAACJiJjaY
|
|
3iXleM8InR5J1GniZw2n3oj7s/8Ao9Wi9a3rNbRE1mNpifNFnVs65XhcWTdt47bnFuF24dm8TFEz
|
|
p7T0/pn0a+HJux1OOrOux08d1ndqY7tillVkzExLOk7yd4YxGwluViJhE45raL0na0dtlWO0+bZr
|
|
1TKi+2zptZGTamT3b/tLacvJjiY3XaTWdYxZZ6/dtPm1zrv1z78fPcbwC7EAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABhkyV
|
|
xUm152iAZWtFazNp2iGhm1Vss8uP3aevnKrNntqLdelI7VRHRnrX/HRjx/tZREVjZXeybW6KbWZt
|
|
pCZ6S08tN7Nmbb7zCrJtyoS5145bSx5mWafelr3tsKmS/o08uXyhlly7RPV2+AcBnPNdZrK+53pS
|
|
fP4ytnPVda4y4BwHxOXV6uvu96Unz+MvVxG0bQRG0bR2G0nHLb2gCUDX12LxtFmpHeazt82wT1gH
|
|
mMN4tWs+rcr2aEV8DU5sM/cvO3yb+O0csLUTSdrLphRE8tlkZI7Atr2ZMazDJVKTYSCawi7Ksq7z
|
|
1QERvLK3ZGPrKbyCrbdnMcsbeaa18/RhvvM7oGEwTG0JmYYTIML22a2e28xELM19oURPNO4lOem+
|
|
n3ZY5+prVnMc2GYU4/L4A0a15cNf6rz/AC6fC6+NxCPOuOu/5tHJTbHj+F5/l1+BYumXJMd9o3/d
|
|
MRXYASgAAAAAAABhlxUz4rY8lYtS0bTEvH8R4ffhmo6bzhtPu29Pg9mq1Gnx6rDbFmrzVsizq2df
|
|
zXkMWTeIbNL7tbXaHLwzUctvexWn3bmPL8WFnHVL326VZ91MfFVjvvVlz79kLrcf2m7j7bNHH3bl
|
|
J2SirLQoy4t1++7G0dBC/RanxI8PJPv18/WG241+alovSdrV6w6mDNGfFF4/OPSW2b1zeTPL1aAs
|
|
zAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAVZ9RXBTe3WZ7R6iZOpzZq4ac1p+UermZMl89+a/byj0Ra9815ted59PQ32hlrXXRjH
|
|
DpCLX6ML5NlNsm/ZRqstfdXzbsZt06sLZNvNB1Za8RDWyZdo7q8udq5Mu/mIMt4md2lmy7JzZuWJ
|
|
dHgfBL8RvGo1MTXTxPSPx/8AstJ1XWpIs4BwSdbeNVqq/URPu0n73/s9hEREbRG0QUpWlYrWIisR
|
|
tER5JbSccur2gCUAAAAPM8Sry8Uyz67fwuxbzVPGsE49XGbvF42V4M0TEL33ERnktsxpk3sumK2j
|
|
admFdPFZ33VS2Mdui2J3UU6LYlFSsN2O5NkCyJ6K7T1TEsbAsxdpReerKkTFGMxvYEz0rsqtbbpC
|
|
b2VT1QEzuwtbaGUxspuJU3neWdKoiu8rq12gCI92YatLcublnzbEz1aOptyZqTuDHLfxN6R0+t5X
|
|
qdJhjBp6UiPLeXl9NSMnEKxHa1+bb8nrlvxUAAAAAAAAAAABTqtNj1eC2LLXeto/R43VabJw/VTh
|
|
ydY+7b1h7ho8V4dXiGlmvbJXrS3xRZ1fGv5rzeHN02bEW3cys3xZJx5ImtqztMS3MeTeGFjqlb2O
|
|
8btql3NpbZtYsnSBLeiWfdTjtutid+ghherHS5p0+f3vsX6T8Fkw181d4lMvEWdnHaGnw/UeNh5L
|
|
T7+PpPxbjdyWcvAAQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAo1Oprgr63ntAmTqdRqK4K9etp7Q5d7Wy2m953lNrWyWm953mVd77R0
|
|
Za1104xxlN9lV8qnJl2a9s3xUXX2ybsJyRDWtl3YWydEC+2VRkzeW6q+T4tbJm+KRdfK1cmWZnlr
|
|
vNp7RC/R6HU8SycmCk7ed57Q9ZwvgOn4fEXtHi5/O9o7fJaZ6z1uRyOEezVstq6jiEbV71xevzer
|
|
rWtKxWsRFY6REeSRrJxz22gCUAAAAAANbX6aNVpL0npMRvWfSXlKamsRMVvXm+EvZXjmpaPWHzfL
|
|
oNRjzXicfWJ8phfPxFejx72x7xMzK+sXiNoiXlq+Pi6fWV/VfTNqfLJl/WTg9Pji8R70LqvMV1Gq
|
|
j/zcv6yz+lanzzZP1lWpelTET6S81Gp1P/Gyf90s412rjtnyfqql6asREdWM9+jz9eJ6yP8Az7uh
|
|
odZqMt458tpB1JvEViI3/RhzRt13/R1MNaziiZiJn5K9ZNceKZiIiQcu/WekT+iYrWI3lzdTrs+8
|
|
8uW0fJzcur1Np/zsn6g79phVaIeetqNR/wAXJ/3SwnUaj/i5P+6UD0ldonum161h5mNRqP8Ai5P1
|
|
lNtRqJjacuT9Qd22WN5aGeZyZd/KHJy59RHbLf8AVq31Gp/4uT9ZEvS8Lr/vSs2npzRtL1z53wK+
|
|
oza/HW2XJNd99pmX0Rb8VAAAAAAAAAAAAAAcHj/C5yV+l4I9+v24jzj1cLFk8nu5jeNpeW41wmdL
|
|
knU6ev1Vp96sfdn/ANFdTrXG+eq1q5F2LLtbZoY8m8d11bbSydErsYsm+zZrO/zcnBm226uhiyRK
|
|
EtrvCrJDOJTeu8A1MWX6Lqq5N/dnpb5O5ExMbx2cPNTeJb/DM/iYPDtPvY+nzhri/jDy5/W6AuwA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAa2p1UYo5adbz+xbxMlvqJ1OqjDHLXree0ejmzNrWm953tPmTPWbWneZ7yoy5YhjrXXTjH8s75N
|
|
mtkyxt0VZM2/m175N1V03yTKubMLXVXybeYLLX2VXy7eam+b0bOg4VquJW+rry4/O9uyZOq3UjVm
|
|
9r25axMzPaIdvhns1kzbZddM0p5Y47z8/R2+HcF03Doi1a8+Xzvbv+TotJnjDXkt+K8ODHp8cY8N
|
|
IpSO0RCwF2YAAAAAAAAACvUZYw6fJkntWN3k8dfHz2vLucdz8mkjFE9bz1+UOZosX1UzPm0nqI/W
|
|
MYo9FlcPNklfFGeH/NshLGun+Cz6PtHZtVZWlRLS+jxPkRpIn7rdoupHTdA5s6SI+7H6Mfo+32Y2
|
|
+To3neSIiZ7A0IjPXpXLePlMotGW3272t85datKzHZjbTVnsDj+FG/2Y/RlGP4R+jo20u7H6N1Ql
|
|
o+H8I/REY957R+jpfReiK6eOYHLtj2tttH6KrY/6Y/R2c+kjeJiFVtLG24hxpw7/AHY/RRkw9O37
|
|
O99Hrt1YX0tfOBLjcGp4XF8c+u8fs9c4dcVcGemSI61nd3IneN1orQAAAAAAAAAAAAABFqxes1tE
|
|
TE9JiUgPKcX4RbRXnNgiZwWnrH4XPi28PdXpW9JraImsxtMS8pxXhF9DecuGJtgmf+1TWW2N/la1
|
|
L7N7T5e3Vy6W3hsYcvLbqzbO9jvvCzvDR0+XeO7crO6FmGSvRThy/RtVXJ92elvk2rRvDUzU7pl4
|
|
izsd2J3jeBpcNz+Lg5LT7+Pp+Xk3W7js5eAAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADs0NTrN96Yp6edkW8Wzm6+LNTq4pvTHO9vOfRoWtt
|
|
1mes95YWvs1s2fZldddOczLPLn2ju0MmebT3YZc2/mpm3qqllN1drsbZIhr3yzvtHf4AsvlYYseb
|
|
V5Yx4KTe0+UQ6nDvZ3UazbJqd8OKeu33peq0eh0+hxcmnxxWPOfOfm0mP+steT/ji8N9mKY9suum
|
|
L37+HHaPm9DSlaVitKxWsdohI0Y22gAgAAAAAAAAAABXnyRhw3yT92Nwef4xm8bVzET0rPJH5d12
|
|
CvLhho3rN9RWs9Z23n5y6O21YhrVYbdGOCfrrLPJRpv863zVS6FS09SvZj3lVZZRdPSqmnSWdrIE
|
|
ebOkK4ldTsgW1WKqd1oMZhEVZyRAImOjGI6rJ7IiATNd46qL02bHkiaxaoNGY2n4ImPgtyV2n0Vo
|
|
Gvlx7x2beiyTk08RPevSVUxux00+Fn2n7N+n5rRFb4AAAAAAAAAAAAAAACLVres1tETWekxKQHlu
|
|
L8InR2nPp43wz3j8P/s5dLveWrFqzW0bxPeJeV4xwmdFec+CJnDM9Y/CrY1xv8qvTZ+WYdbDk5oh
|
|
5zHk283U0eo3jaZZ2N5XYjrCnLSJhOK+8d1kxvCqzSwZvousrb7k9LfJ3nB1OLeJdLhufx9LEWn3
|
|
6e7LXN9Ofy5/W4AuxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAETaKxMzO0Qi9646Ta07RDmZ9VbPbaOlI7Qi3i+c3TPUaqcu9adKfy0722ZXvFa9
|
|
XO1OrjrESxt66ZJmcjPUanlidmhkzTZVfLN5VWvsC2b7R3U3yqrZZtO1esz2h2+F+zWTUcuXXTNM
|
|
feKR3n5+iZLVbqRzNJo9TxHLyaekz62ntD1fDOA6fQbZL7Zc/wCKY6R8odLBgxabFGPDSKUjyiFj
|
|
SZkYa3aALKAAAAAAAAAAAAAADQ4pl2pTFH3p3n5Q33E12Tn1eSfKscsLZ+orS00eJqbW+Lfnu1tF
|
|
XaJnZsz3WpCfsyp00fWSvmPdVYOmSUDd8kR3InoQosy7JmUX7MdwZ17ro7KKT1XRPRAsrO0rYndr
|
|
79V1ZBaQiJ6JgCSIJASwrO07MpV2nqBlrv1a1o2bf2qtfLXaQUTO0sb05o3jv3ZXhjS20xEphW5h
|
|
yeJjjf7UdJWNKLziyRePsz0lux1SgAQAAAAAAAAAAAAAADG9K5KTS8Rato2mJZAPIcU4ZbQZuekT
|
|
OC3afT4NXFkmlntc2GmoxWx5K71tG0vHa/RX0GpmlutJ61t6wrY2xr8dXS5uesN+tt4ef0eaa223
|
|
2dnHk3juyreM81OaFGiy/RtZET9jJ7s/2bdutd2jqKeic3iNTsd8a2h1H0jTVtP2o6W+bZbOO+gA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABje9cdJt
|
|
adohGTLXFTmvO0fy52bJfU23t0pHaqLeL5xdK9Rnvqb+cUjtCi94xxvK3JetKuHrdZvaa1ljb10y
|
|
cnIs1Wt3naJc++TmVWvMz1YWybfMGdsm3eWek0mo4jm8PT0mfW3lDf4V7P5tdMZdRviwfvZ6/TaX
|
|
DpMMYsFIpWPTzXmf+steT8jn8L4Dp+HxF77Zc/4pjpHydYGjC3oAAAAAAAAAAAAAAAAADG9opS1p
|
|
7RG7zszN6WtPe0zLua+3Joss/wBOzhzG2OsL5+IrY09dsSyYRijbHEMvOChb7KjF0yS2LQ169Mso
|
|
S24noyrPVXWejNVKbTuw3T3REdQWU6LYlVvsyiUDPfqupPRr79VuOQX1lZEqoZxIMksd0gT2VT0l
|
|
bPZVbuCaW8i8bwr32WxbcGnkjaZa9p2ndv5qbw5+aNugLItF6TEtvTX5sMb969HMpfazc0d9stqe
|
|
vVZDdAQAAAAAAAAAAAAAAAADV1+iprtPOO/2u9bektoB4TJTJpNRbHkja1Z6uto8viVht+0HDvpG
|
|
H6Tjj6zHHvbecONw7Ltfkmeqmo6Ma69DXbbZTkr1mGWO3RneOaGbZRoM30fVzSelMnT83aef1FZ7
|
|
x3h1tBqfpGnjmn369LNc3sc3kzy9bQCzIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAa+q1dNNXr7157VhGp1Xh70x+9f9ocy283m1p5rz3mVbrjXHjt91lz
|
|
5c9+fJ1nyjyhdM8lZlOOIiqrUXikd+kMreunnI5XEdX4dZiZcG+XmtNl/F83PeeWWHDOGanieSKY
|
|
q+5H2rz2hMzWd1Iqx1yajJXHhrNrW6REeb1nCPZumn2z62Ivl7xTyr/6uhwzhGn4Zj2xxzZJ+1kn
|
|
vLoNJnjHW7TbbsAszAAAAAAAAAAAAAAAAAAAAaPFrbaSK/itEOXt0rDf4xb/ACa/GZacRvaF58Q2
|
|
IjasQnzPIhCU92tMbZGzHmotG10C6nZkwpPRmipIllEbMIZIE7solgmJBnCyk9VMM6z1BtVllEqK
|
|
z0WRILYlluriWcSDJVbusV27gwInaSWM9ECyZ3hqamnSWxFmOSOaqRx725bNnSZNs9J+OynVY+WZ
|
|
YYr7TE+nVaIr0Ais81Yn1hKAAAAAAAAAAAAAAAAAABExvG09peU4nov9n66L0j6q/WPg9Y1OJaON
|
|
ZpL0+9HWs/EWzeVz9PbmrEtnyc3h9reHy26TWdnSr2YX6657ijLXpLX0+onSamL/AHJ6W+Tbv2aW
|
|
ekTv16JzeI1Ox6KJiYiY7Slz+E6jxdN4dp3vj6fl5Og2clnKACAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACZ2jeQRMxEbzO0Q08uqtkma4ulfO3r8lefUePMxWf
|
|
cjy9WvlzVxV6T1Z61/x0Y8f7Wc7Ur1lqVy+LqOWJ2hp6rXddon5rOF1tfmz5OkT0qzb8dWbxjp1c
|
|
biuuilJ5Z6r+IcQrixzEy8zl1E6rNt1tMztFY81sztU1eRucN4ffi2p5esRM72n0h7rS6XFo8FcO
|
|
CkVpX082nwXh3+z9FWLxHi36328vg6TZyW9ABAAAAAAAAAAAAAAAAAAAAAADj8Unm1tK/hqppHvw
|
|
y1k8/EMk+m0GOPeafiFpCZYwolnXspvHvLa9mF46gmnZmwozRUiUCBKYYsoBLOFbKAX0llEqqyzi
|
|
QXRLOJVRLOOwLIljZMEgrlhKyYYTAK5nZPN0RZjugUanHzVlz6xtLq361c+9eXItPpXX0dubTU+E
|
|
bL2lw2++O1fSW6m/VYAISAAAAAAAAAAAAAAAAAp1GbwcfTreelYEydcuMcRrM/L9nnlsV6wqpi2r
|
|
tv133mfWVkRyRtEdGFva7MzkYZNoamWN4bV4mYa9qztKIujhVppxGI8r1mJegeZpknBqKZY+7L0t
|
|
LRekWrO8TG8Ns/HJ5ZypAWZAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAADS12fp4VJ6z9qVuq1HgUiI+3bpDl589cOKZmevqprXPTbx477rDJlrhr1nq4+s182tMRP
|
|
RqaziXiZJrWekNG17ZbxWJ336M5LXRbI3dLTJrs07RMY6fan1dHLrowY+X7MVjt6N3R6Kul0EbWm
|
|
s7bz8Z+LnabQX43r7Y53php/mXj+Dnv0f1JO1x/8ZxbUzj02O15mfLtD13AvZqnDds+pmMmo26el
|
|
XX0Wh0/D8EYtNjilY7+s/NstpOOTW7QBKgAAAAAAAAAAAAAAAAAAAAAADG88tLW9I3BwJtz6nNf1
|
|
vK/DHVqYJ3pzT5y3MPZeojOWMQylEKpTVjZnDCwkqzYQyRRICATCITAJZQxhMAshnEq4ZQC2srKq
|
|
qrIBZCWNZZgwswmFloVyCu0dFcx1WyrtCBhv5NTPHXds2U5o3hIz4ffbPt+KHUcTSW5c9Jme0u2v
|
|
VYAKpAAAAAAAAAAAAAAAAYZctcVOa35R6tLrltN795/YvknNqrfhpPLH92V5isd9mWq6fHjk6rn0
|
|
ZxG8KK5Jm/wbVZiYZtqrmkqL023bkxvCiY3lJHNyRG81mHS4Rn5sNsNp64+3yaWaNrzOzHBl+i6q
|
|
mT7s9J+S+ay8mex6EIneN47SNXKAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAImYiJme0JafEs3h6fkidrZOn5eaLeJk7eOdm1Hi2vmtPTry/CHmOJcUvmvOPF1n09Pm
|
|
6HF9ZGm01qxO3R5vSY7XwzmzTy47zzTEd7en5Mfvt2/PURWdo3tvPrPlKymbktFqTtMTvHzbOLDG
|
|
f63JXbFX7FdnoODcDprZpq9TjiMMTvSn4vj8l5fxnrk91saPSa7i2hpOfbTVt5x1m0fLydzR6PDo
|
|
dPGHBXasd585n1lsRERG0dIF5OOe6tAEqgAAAAAAAAAAAAAAAAAAAAAAADX11+TRZrf0y2Gjxe22
|
|
gtH4piP3TPpXKwxtjhuYo9xq442iIblI2pC1RET2ILd9kxCqRjZmwlCSEohIJAQAAJZISDKGUd2M
|
|
MoBnVbVVCyAWVWeSuqyOwIlXZZKue4MJV2WWYT2QKbKL9YlfdRdIo35b7/Hd3KTzUrPrDh27uxpb
|
|
c2mpPwX/ABX9XAKpAAAAAAAAAAAAAACekTIp1eTwtJmv+GkyJn1oafeazbfpMzLR4jq/o8b823zX
|
|
6XNF8ERCvTcNpxLV5LauvPhx9Irv3lhztdtv8TtaWLicXrt03jzjzb2k1nid56ty3s/w+a7Uwzjn
|
|
1raejlarhmbhl/FpbxMO/fzj5p/ixSeXOvTtRfeI280ZI26tfDm3pWe63LaZx7qtGvniJ6tPLvOK
|
|
fOa9WzbJvTbza02jl3n5SSljscK1MajSxWZ96nSW88xw/VfQ9XMT9nfa3yemid43jtLeXsce88qQ
|
|
EqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADia3UTm1l4j7OP3Y/u
|
|
7Vp2rM+kPJW1PhYcmS0+9MzKm/jbwz31weMzbV8UppazPL9q0/BF4rk1GLDSNqxPWPhCnHmnNrtT
|
|
qPKteWPm6U6OdHaZvO+SaRNvhv12Ub/q3FhtrNVj0uKOt56z6R5y9zix1w4qY6RtWsREOJ7L6OKa
|
|
S2rvX6zNM7T6Vh3mmZyOfya7eACzIAAAAAAAAAAAAAAAAAAAAAAAAAAczjVvqMVfW/8AZ03I41bf
|
|
Lp6/OVs/UVrY47NyOzUxd4bUJpEbb3Z7IiOrKIVSjZhMLJYyhKIgmGUQSDESIEbJEgQmCITEAmGU
|
|
IiGUAyhZVhDOoM4Wx2VQtqBKuyyWEgqlhKyyuyBVaGtkbNmvk7A15l1eH2300R6TMORPSXT4ZO+O
|
|
8fFefEX63gEAAAAAAAAAAAAAAAq1WPxdLlp+Kkx+y1Fvsz8gjhaDauGK8sx07y3OE3m1tT6RaP4c
|
|
vU6yMNKUx73zT0ilY3l2eF6a+m0kRl/zbzz3+Ez5M8z26fJruW6wzYq5sV8d43raNpZjRzPPaTmx
|
|
5b6bJ9rHO3zb2WJ8GWPEscY9bgzxH2t62n19GWW0eHOzHU5XbjXZ1x8WTnz2iZ7S2M1IjH2+LX0V
|
|
KTqs8zO9ot0j8nUthi1J3UaOFMTfLFo6xMbS9BwHWTqdHOO8+/hnln5eTjYMFo1WTH5VnePzXcIm
|
|
2k4zlpPSmXy/hfF5eMfJns69OA2cgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAADG/2LfJ874rW845mubliY7bPoto5qzHrDz0+yePNF41OotaJ7RWNtpV1OtfHqZ715fhu
|
|
j8adNpcVfeyzE2/vLuanhOu1nEctIxTTFa/+ZPbZ3eHcF0vDbTfFE2yzG03t32+DokynXl9+leDB
|
|
TTYKYccbUpWIhYCzEAAAAAAAAAAAAAAAAAAAAAAAAAAAAcXjE/4zDH9M/wAu04XF5/3jj/0f3Wz9
|
|
RUYmzDWxS2I7FSyjuzY1ZKpRKEygEwiWUIkGIk2QJNhKQhMIhkCYZQxhlAMoZwwZwgWQshVCyATL
|
|
CWc9ldpBhZXLOVdpQK7NfJPRdaWvknoDVvPvOnwuel4+TlXn3nS4VPvXj4QtEV0wAAAAAAAAAAAA
|
|
AAAAAVV02CmTxK4qRf8AFFeq0AAAanEsfPpZmO9Ji0NDLfkwdOsulrumiyzHlVzJrz4Ovoy26vB8
|
|
cTBa9NffLtMY77Rv8Yegx5ImkKdJoY1HC81Y+3OSbVn0mGGkmbY45u6tnrrTOu2xGO0RxCd+nNVj
|
|
qKxTV1vH2pjaGtnyzXXYdo96ZmGXEMk15b7/AGZiVerWPTYckZcNbx5wzc7hGbnxXxzPWk7x8pdF
|
|
0S9jh1OXgAlUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAcPjEf4/FP9H93ccXjMf4vDP9Mx+62fqKrx+S+GvibEFSsqyYwlVK
|
|
ZYsmIMoRKYJQIPIEiQ2ATCUQygCGUIhMAyhnDCGUIFkLIV1ZxIMpVWWSrsCuyqyyyq09ECq8tfJK
|
|
66jJ2Bp5J6upwn7dv9Lk5J951uE/av8AJaIrqAAAAAAAAAAAAAAAAAAAAAAq1Mc2myxPnWf4cmtu
|
|
XT9fR0tffk0WSe28bfq5Wbamm3326MtunwfK6PCv/AxPraZ/dz9PO97/AOqf5dHhdZrw7Dv3mOb9
|
|
XOxRFM+avpe38mvkPHf/AFWlrKba7Tzt99ZxKkfR7euyNXMTrtPHfa0z+zPiM/UR8Zj+Wbdu8HpN
|
|
M2bfzrV13M4dO2pyR61dNvj44/J/oAWZgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADj8bj63BPzdhyeNx0wz8ZWz9RWri7Nmv
|
|
VrYu0NmqaRZHZlDGGSiwxZSgCEkCBCQSCQBMJRCYgEsoYx3Z17AlMIhlCBnDOGEM4AlhZZKq4KrK
|
|
7LLKrIFN2vdfZReAaObu6/CO9vk5OePR1uEd7fJeIrqAIAAAAAAAAAAAAAAAAAAAAGtxCk5NFliI
|
|
3mI32+XVyNTyZOHTee946PQKPoeDffw4777eW/yVs60xv+ZxOnr4Okx1t05KRv8Ao41Z5q3yed5m
|
|
XY1szXRZ5jvFJ/hxItP0aOSN9q7yrtr4f2tHFM5+KT16Yq/vK/iGSbXw4vO14UcPx5MGfNbPG18m
|
|
1oj4THRsTw7VanPXVYpi3gzMcnrvCnG11JOupwuN8+a3pEQ6jT4divjxWnJExa09pbjbM5HHu90A
|
|
JUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAHM41H1GOf6nTc/jEf4Ws+lls/UX45uGekNujTwdm5RNIthKIZKLDFlsiQIShIC
|
|
EgCUJ7AmGTGO7IDzZQhMSDJMMYZQgZwzhhDOATuqssmVdgVWVWWyqtCBTeVF19lF+wNLNG7q8I+9
|
|
8nLyupwnt+S8RXUAQAAAAAAAAAAAAAAAAAAAAAAItWL1mto3iY2lyrcLyUxzix2ia2nvPeK+jrCL
|
|
OrTVnxpanhuPPemSs8l6RtE7dJj0ldpNP9GwRSZ3neZmV4cR/Vs4AJQAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAANHi1d9H
|
|
M+kt5ra+vPoskfDdOfqK4mn7Q3aNHBPZu0W0RdDOGFWcKLCJZeTGQQlCQSgASBsCYZQxhlAJTAmA
|
|
TsmAgGcM4YQyjsgRLC3VnaVcgwsrt3Z2V2QK7tbJ1bN5a9waeWO7p8Knt8nNyebpcK8vkvlFdQBA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAK9RXmwZI+ErEWjesx6wQeZwejeo0cccuW8
|
|
elpblJaaRGxVnCuss4ZrMvJEgCAASISCQIBlCYYpieoM0wx8k7gzIRueYM4Z79FcSy3QEsLJmWFp
|
|
BjaVVpZWlXMoGNmvkXXlr3kGtknu6XCf7OXkl1OEdl8orqgIAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAHmskcmtzV/rls0U62OXiWX4zErcc9GmkRfWVkSqqziWayxCPIANwBIhIJSxS
|
|
CRG6dwZwlhEs4BluMdzfqgZxLLdXuy3AmVdpZTKuZBjaVVpWWV2QlhZRdfZRcGpl7urwfrzfJy8r
|
|
rcH61vPyWitdMAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHA4nHLxKZ9awnH2ZcY
|
|
jbW459aq8fZpfiI2IZwrqzhmsz3Ebm4JN0AMhCQSIASndiAziWUSriWcAyRujc80DM3RCfIETLCW
|
|
UsZEsJYSslXZAwlTddPZTkBp5e7r8Gj6rJPxhx8k9Xa4PG2C8/FaK10QAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAcfjcbZMFvnDWx9m5x2PqcNvS+zSxT7sNPxH62YZQwqzhRZO6UCB
|
|
KUAJTux3SDIRuAncQAmJZRLBMSgZ7iIAZRKd2DICUSlAljLCYWMLIFVukNfI2bNbIDTyT7zu8Ijb
|
|
Sz/qcG/2nf4T/wCE/wD2WnxWt4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHL9oL
|
|
+Hw2cm28VvEuPptfgyVj6yIn0no7/FtJfW8NzYMe3PaPd39d3iMug1WktNc2C9dvPbeP1aZ9xF+v
|
|
T471tHu2iflK2HkqWmvaZj5Surqc9Ps5bx+alTHqYHm68S1Vf/NmfnC2vGNTXvyT84Ql6A3cSvHM
|
|
sfaxVn5Ssrxyv3sM/lKB1xza8bwT3pePyWV4tpZ+/MfOEjfGrXiGlt2zV/PotrqcN/s5aT/+wLRj
|
|
FontMSlAlKEgndO6IAZQljDIEgeQljLCzOVdkCu/SGrkbF56NPNeKxMzMRHxENe0+89DwuNtHHzl
|
|
5PJr8NcnLW3Pbf7r1nCZm2gpae8zMrz4i/W6AgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAETETG0xukB4HVaeMHEtRi26RedvkyjBSfX9W77QYvC4xz7dMlYlrU7M929dWJLFc6aPK0q
|
|
7YLxPS0S22FlP6q38Zac0yR92s/KVc3tHfFf8tpbcsLRvB/dR/8ALLVnU0r9uL1+dZI1mnmdvGpv
|
|
6TOy6ym+Oto2tWJ+cJ/tW+KLK5KW+zes/KU7tG+h01p64qx8Y6NXNo6Y+uPJlp8rLf0rfG7MXtHa
|
|
0x8pZxqs9e2a8f8A7Oj7HaTHn0+f6RWM23LETfr6vRW4PoL99NT8ui7F4+vEdXXtnt+fVbXjGsr/
|
|
AOZE/OsPS29nuH27YrV+VpeV9pdPXhOtw49NG9Mld55+vXcTPd42I47qo7xSfyWV9oM8d8VJ/VxM
|
|
d8l46xWF9cV7en6o/qLfxp2I9ob+eCv/AHMo9op89P8A/wBORGmyT5R+qfo2X8P7n9Q/jTsx7RR5
|
|
6ef+4/8AuHftg/8A6cWcOSO9J/WEbWr3pY7Efzp2Lcfv5YK/9zWy8d1E/ZpSv5Oba1/+Hb9lc+LP
|
|
bFt87I7E/wAabWbiurvEx4nL/pjZzc2bJkn372t85ZXx55/BX85lucC0vPxnTxlnnjm32mOiZqUu
|
|
LJ2p4TwnVavNWaYbRTfre0bQ99pcH0bT0xb78vmtiIiNojaErMwAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAHnfarF7umzRHaZrLjYrdIen9ocPi8JyTt1xzF4eUw23rCm3R4r6bMy
|
|
wt6kdTaWLdjswmNoZontsCm0K5XWjopnuDC0dGpqG5bs08/daKV672MjbSaif6oh6Z5f2LtvptRX
|
|
0tEvUN3Jfo8f7cYve0eX4zV7B5z20xc/C8eSPuZIRficfXlcPaG7ino08HWIbePpLF2NuiyOyrHK
|
|
3fZFSwuovHVfaVF4QK5YWTM9UT0EKry6Ps1Tn4zjn8NZn9nOtLseydObiWW34cf918fWfk+PYANn
|
|
KAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAq1WKM+ly4p+/WYeBxTNd6zG0xO0
|
|
vobw3FcP0bi2em20Tbmj5Srr418V9sa2Z7qKyzi07MXUylhaU7yjqhLCeiq3ddaFNxFYW7NLNG8t
|
|
zya+WO6Va9J7FW66mvwidnrXiPY3Ny8RyUn71Jj9Ht3RPjk19HK9pMHj8D1ER3rHN+jqqtTjjNps
|
|
uOe16zAifXzfTz7kNyndpYazS9qT0mszDdoxrsi6m8LazMq6zDOsq1ZEyrt1WWlXaUCqyq0rbKbi
|
|
Fdp6PReyFd8uqv8ACsfy83aXrPZHHto89/xX2/SP/dpj6y8vx6EBq5gAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAB5n2q03LfDqqx39y39npmlxbS/TOG5se29tuavzgWzeV4mtui2
|
|
O3RRSY2hdVhqO2MvI36iu9lUsrSrvDHn6spnmSiq5jooyV6tq1VV69RC32byTh43h8otMx+r6I+Z
|
|
aK/g8TwX7bXh9Mid4iW+fjl8n1ICWb57xLBOm4zqse20Tbmj8+qKdnS9q8PhcTw5tumSm0/OHMxz
|
|
0Za+uzx3sX1t0Zxurr1ZxvspWiZYWZbsbT0QK7KLrZVZJFaqt5vbezNOTg9J/FaZeJns93wCvLwb
|
|
T/GJn92uGHldIBowAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAADuAPA67F9H4l
|
|
qMW20VvO3yRWW97T4fC4rXJHSMtI/WGhVlue3b473K2KzMML4+62tujG9pnozXaOSOVFMnVbmq1t
|
|
trJRW5E7wwvUxTvCyY6CHOt7moxz6Wh9PxTzYaT61h8x1MbZK/OH0zTf+Fxf6I/htj45vL9WgLMn
|
|
mvbPFvocGWO9L7fq85p5maw9d7VYvE4JkmPu2if3eW0+PasdFNOnxfF1Y2hlykRsmY+LJ0MZjZXa
|
|
eq2eyi8oQTO0KLdZWzPRjWu6VaqtHR73g0bcI0sf0Q8Nkq93wqNuFaWP+XDTDDytwBowAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAef9q8HNpcGaI60vtPyl56k9Iew49j8ThGe
|
|
PwxFv0l4zH2U26fDfTYiyJljvsjf4sm6vJ1hrXjq2MkqLdZEVbgbMx0auGdmzNt6iHN1Ub5af6of
|
|
TdPG2nxx6Vj+HzaaTm1+nx/iyVj930ysbViPRrj45vL9SAuyc7j1efguqj+jd4/T33rD3HEcPj8O
|
|
1GP8WOY/Z4TTT7sKadHhbcsZnaCJ3TPZk6VdrKbTutmP0U2nqgrGOsr8deiuI2X09EqKM1dt3uuG
|
|
f/jdN/06/wAPE546S9rwud+Gaaf+XH8NMMPK2wGjAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAABrcRp4nDtRWPPHP8PCYusPoWSvNjtX1iYfPuWaXtX8MzCuvjfw32siu8ptXoxi
|
|
0wy5t4YulReqmazu2skbquURWFInddM7VYRGyL291KFnCcfj8e0le/Lbmn8n0N4b2Ur4nHLWmPsY
|
|
5e5a5+OXyXugBZmiY3iY9Xz7NjnTa3Ph/BeYj5PoTxftFg8Hjk2iOmWkW/Psrr418V5WrWd2faFc
|
|
V2jdnEMXWxntupmN7NiYU27iWML6dVMVnddjgVqMsdHr+CW5uE6f4Rt+7yuSsTDv+zWXn0WTHP3L
|
|
/tK+GHl+O0A1c4AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA8Dn93W56/wDM
|
|
t/L3z59qp24jn+OS38lnpr4r7ZxHQ2TEstt3PXUrt27K57rr1VT0BjKnJPRbMqMs7QlV2fYvHvrd
|
|
VknyrEfu9m8f7FZI8fVU85iJewbT45NfQBKo817W4eulzxHaZrL0rje09ItwqbfhtBVs3leai8RD
|
|
KLw1sduesL606dWFdsZT1jdhNeq6K9DlhCVUU6s4jZnt1YzAhnM71dH2bycmszY/K1d/0c6OzY4R
|
|
fwuK4p8rTstn6z8k7HrwGzkAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHz3
|
|
Vxvr80/8y38voTwGpj/F5/8AqT/JfjTx/WVeyY6FPspc9dZPVXaOq2WEwIUTVRmjo2rNfLHRI3vZ
|
|
DJycXtX8dZh7t879nsnhcbwz23tt+r6I2nxyb+gCVBzuPY/E4PqI9K7ui19fTxNBnp60n+Aj5/pJ
|
|
3jZu1aOnnltMNussdfXbm+l3ZM9URHREdZVXTuT1Nk7boQiOkJw28PU47/htEp5eivJPLMTCZ9Vv
|
|
x7mJ3iJ9UqNHk8XR4b+tIXuhxAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD
|
|
weqjbWZ4/wCZP8vePCaz/wDIaiP+Zb+UX408f0r9lOxWOifJhXWjfyYWllPRXYQxnrCrJHRd3YZI
|
|
6A1NJecHEsN/S0T+76bE7xE+r5dk93LW3pL6ZpMni6PDf8VIn9m2fjm8s9rgFmQxvHNS0esbMiew
|
|
PnHLyai9fS0w2aNfUTtrs3+uf5bGPqy068fF227KtSsdFlKqNGMV6myyY6sbdIQI8tlOWOi6Jhhk
|
|
j3RD0vA8nicMx9etZmHRcT2Zyb6XNT8N9/2dt0T449T2AJVAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAHhdfG3E9TH9cvdPEcXjk4zqI/q3L8aeP6xr2TsxpLOekMK6mFo6qpXSrm
|
|
OqBixvHSVmzC4OfqK7S9/wAByeLwbTW9K7fo8Fqo6Paeyl+fglI/Da0NcMPK7QC7AAB8313TiOf/
|
|
AKk/y2MHWrX4jG3E9R/1Lfyv0/aFNOrHxuU7LI7MMayGTVlHWUXhNe6Z6wIUsb9d1m20q7dkDpez
|
|
N9tRqKT5xEvRvKez9+Xis1/FSYerb5+OTyf6AFlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAB43j9eXjN/jWJ/Z7J5L2mry8Upb8VIF8f6aGOey2eynHvOy7bowrrYSxZSwQJ2YXZ
|
|
92N4BoanrEvVexmTm4blr+HJ/aHltRHSXofYm/1Wrp5RaJaYY+X49WA0c4AD51xONuKan/qW/lbp
|
|
+0MOLRtxbU/9SU4J7KadWPjep2WQrr2WRPRk1TvsndXMpiRCb9FNu0rbTuqvKBscCjfi9PhWZeue
|
|
V9n434rafTHL1TfPxy+T/QAszAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHmv
|
|
avHtfTZfnV6VxPajHzcNrf8ABeJFs/XnMcr4no18c+6vr2YadkY2YM57sEDLyY37Mo7MMnYGlqO0
|
|
vQ+xNfqNVb1tEfs87qZ2rL0/sVX/AHdnt65P7Q0wx8vx6UBo5wAHz/jUbcX1PT78qtO2vaCnJxjP
|
|
8Zif2amnnspp04+OjWejKJ6MKdmcMmyJn4m5ZHzEVPMwtJv0VZLbQDqezcb8RzT6Y/7vUPM+ytZt
|
|
n1OTyiIh6Ztn45N/6AFlAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABocbxeLw
|
|
nUR5xXm/Rvq8+OMuDJjntaswEeBxT0bNZ6NatZpNqz3rO0rqsdO3PxlaWEMpY+aqWXkryT0ZT2V3
|
|
7A0dVPuy9f7G124NM/iyT/Z4zWT7sw957MYfB4Fp4/FE2/WWmGHldcBowAAeM9qKcvFeb8VIly9P
|
|
0nq7ntbTbVYL+tJj93CwT76unR4/jo0nozhhTsy3Y1sWljM9Ce7HyQIm3RRlttVbaWrnt0Sh6n2U
|
|
x8vD8mSfv3/h3XN4Bi8Lg2nj8Uc36y6TeOPXugCUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAPD8RxeBxXUU26Tbmj8+quro+02Lw+I4ssdslNvzhzazvDPbq8d7GW7Dfqz2VzG
|
|
0s2qd+iu/Zn5Ksk9BVztX1mI8930zh2LwOHabH+HHWP2fNYp4+vwYvxXiP3fUqxtWIjyjZtj45/L
|
|
faQFmQADzftfj3w6fJ6WmHmsP23rvaqnNwqLfhvEvIYZ+sV038bo0noy36MK9oZQxrdMyrlnMbMZ
|
|
QKrS1M07zEestq/RRjr4utwY/wAV4j91p9V18fQdJj8LR4ccfdpEfsuREbREJbuMAAAAAAAAAAAA
|
|
BAJAAAAEAJEAJQAJQAJEAJQAJQAJEACUJAQlAJEAJQAJQJAAAEAJEAJBAAAJAABAJEJAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABwvanDzaPFmjvjv8A
|
|
tLztJ3h7HjGHx+FainnFeaPnHV4vFbeIU038VbHeGF+kso7Mb9mTdhKnLK3dRm7SIrHhGPxeP6Sv
|
|
9cT/AHfSnz72Zx+J7Q45/BWZ/Z9BbZ+OXyfQBZQABzeP4/E4NqI9Ii36S8Ng/wAx9C4jTxOH6ivr
|
|
jn+Hz3B/mQi/GvjdCnWNlsdI2V07LIlg6USrt2ZzZXMoFV+zPhGLxeOaavpbm/RVltEN72Yx+Jxm
|
|
b7dKUmf7L5+s9/HtRA2cqRACRACRACRACUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAACQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAQCQQCRACRACRCQBCQBCQB
|
|
ACRACRACRACRACL1i9LVntMbPATTwdRkxT3pea/u+gPE8Xx+DxrPHlaYt+qNfGvjvtXXsi0dOrKk
|
|
dEXjZg6VMtbP2bMtXUdpEV0/Y2nNxbNf8OP+727xvsXH+N1U/wBEfy9k3nxyb+gCVQAGOWvNivX1
|
|
rMPnGGOXNNfOJ2fSZ6w+dZKeHxDPX8N7R+6L8a+L63KdoZ7q6zvEMpnowdKJ6ywmWUyqvIKM0vQ+
|
|
x+D6rU55+9aKx+TzWa36vbezmDwODYenW+95/Nphj5L6dQBo5wAAAAAAAAAAAAAAAAAAAAAAAAAA
|
|
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACAEiAAAEoA
|
|
AAAAAAAAAAAAAEAkEAkRuAkQbgkQAkQAkQAkQAl5T2nx8nEMOT8dNv0l6pwfarHvpcGWPu32/WCr
|
|
YvK4mOem6b9mGKd4Z3idmFdka0y1c892zfpMtLPaNpEV6D2Kj/Eauf6YeweQ9ieuTVz8K/3evbT4
|
|
5NfQBKoAA8FxCvJxrUx/XMvevD8Zry8fz/Haf2RfjTx/6RSOnRMyypHu9kXjowrqVSrvPRnZVl6V
|
|
kK0775MsUjvadn0nT4ow6bFijtSsVfPuFYvpPGtNTy54mfy6vorXDm8l9pEC7JIgBIgBIgBIgBIg
|
|
BIgBIhIAgBIhIAgBIgBIIBIAAhIAhIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJAAAAAAAAAAAAAAA
|
|
AAAAAAAAABAJQkAEAAAAAAAAAAjc3BIjdG4Mkbo5kcwMjdhzHMDPc3V8xzAs3N1fMjmBZubq+Y5g
|
|
Wbm6vmOYFm5ur5jmBZubq+Y5gWbm6vmOYFm5ur5jmBZubq+Y5gWbm6vmTzAz3N2HMnmBlu5ftFTx
|
|
OEZJ/DMW/d0t2rxKni8N1FPWkiZ9eS08e7Cy8dGGn6UhZaJljXZGnmc3UT3dPP2cnUT78xCIV6j2
|
|
H/8A9c/6f7vXPI+w8bU1U+vL/d63du5NfUiDcVSIAS8b7RV5eOb/AIqRL2TyXtNX/e2KfXH/AHlF
|
|
+NPH/pr4+2xcxx0hFpY11K7R16KM32ZWz3UaidqSgrc9kcPicWyZJjfw6T+727y3sXh2xarN+K0V
|
|
h6lvPjj3e0ASqAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJQAAAAAkQAkQAkAAAAAAAAAAAAAAA
|
|
EgAAAAAAAAAAAAAAAAAAAAAgAAABKDcAN0bgkY8xzAyRux5kcwM9zdXNkTcFm6OZXzMeYFvMibKu
|
|
ZHMC2bo51U2RuC2bom6rc3BZzom6sBZzI52ADPnOdggFnMc6skFnMc6rc3BbznOp3RzAv50c6nml
|
|
HMC/nOf4qOY5wX85zqOc5wbHOc7X5znBsc6edr85zg2ec52vzpi4NjmY5bROG+/bllVzsNTk5dLl
|
|
n0pP8BHmMHWNmzt0aum8obm08vVjfrtnxztR0mXHzTvaZdjVRMTLkZo6yiFen9iZ2pqY/wBP93rN
|
|
3kPY+/LfPX1rE/u9XzN3HfqzdO6vmTuIZ7m7Hc3Bnu8t7TR/vHBP9E/y9Pu837SV31umn+if5Rfi
|
|
/j/01MMb1hjkrtKzBG0bMsmOZY11tOYamr6Und0LUc7XT7u3rJPqL8er9lcPhcFpbzyWm39v7O00
|
|
+FYvA4Zpsc94xxu227jv1IAgAAAAAAAAABKAAAASgASgBIgBIgBIgBIhIAAAAAAAAAAAAAAAAAAC
|
|
UACUJAAAAAAAAAAAABIAAAAAAAAAAAAAAAAAAAAg3AEbomQZbo3YzLGbAz3RNlc3YzcFs2YzdVN2
|
|
M2Bdzom6nmNwW86JurTAMuY3REJ2BB1ZRVMVBhsbSsiqeUFXLucq3lTygp5TlXcpygp5TlXcpygp
|
|
5TlXcqOUFXKjlXcrGYBXysdlswiYBVMdUTCyY6sZBWxlnMMZgGLGZZSwkDdHMiWO4MuY5mEyjcFn
|
|
N1OdVzHMC3nTzqeY5gX85zqOZPMC+Lqdbk20eb/RKOZr8QybaK/XvtH7iZ9aGlp2luzT3fg19NHS
|
|
OjbmPcYX67XH1XSZ9XIzRvMuzrK7zLkZYmYnciunb9lZ5dTk+OP+71cXeP8AZnJ/ip2nf3J/l6iL
|
|
/Fu5L9bMWZczXi6YuIbEWTzKIuyiwLt3nuO25uI4a/hx7/rLuczg8TicvFLbfdpEK6+NPH/phhjo
|
|
stLGkctUWnoxrrU3j1cnWTzZq1jzl1clo5Zcu8c+txR63iP3Tn6pv4+g4o5cVI9IiGe7CJ2iE7t3
|
|
GyN2O6dwSINwSISAlAAlACRAAlAAlACRACRCQAAAAAAAAAASgASISAAAAAAAAAAAAACQAAAAAAAA
|
|
AAAAAASAAAAAAAAAAAAAAAAIAAAQCAJljuljsCJlhMs9mOwMJYys5TkBVsjZdyHICrZPKt5E8oK4
|
|
qmKrOVOwMIqyirPY2Bjyp2ZbAI2NmSARsbMgEbI2ZAMdjZICNkbMkSCNmOzJEgx2YyzljMAwlhKy
|
|
WEwCuWErJhhMArlhLOWEgxljMpljIImWMyTKJA3N0IBO5vux3NwZbnMx3NwZczT4jf3MdPW27a3a
|
|
fJOq1XNP2KdIRfi+J2trSYfcjeF+Wm1OicVeWIiN9kai8xjY12ORqultnI1Ecsujq79XP1FovWYI
|
|
rTgeq+j8QrWZ+3Mx+r2UXeC0WG2Ti2kiN5mL807eUREvbzbaejefHJv62Iv8WUXa0WTFhVtRdlF2
|
|
rz9WUXBtc7jR9dqc2T1ttHyhvZMvJitb0jdq6XHNcNenWVN3028U99WRj6Kb02be3Tq18/SN2Lpc
|
|
3UdN9nOmZrqKX/DaJ/d0svvTLRzV3jomK6+Pd1vvWJj0ZczT0mXxNJht60hfFnQ4qu3N1cWTEgs3
|
|
Tur5k7gz3N2O5uDM3Y7m4MtxBuCQASIASIASAAAAAAACRCQAAAAAAAAEoSAAAAAAAAAAAlAAlCQA
|
|
AAAAAAAAAAASAAAAAAAAAAAAIASgAAAEJAQJQCNkbMgGOyOVnsAw5TlZ7GwMOVPKy2NgY7GzIBGx
|
|
skA2AAAAAAAAAAQkBAEghEskAxYzDPZGwK5hjMLJhjMAqmGEwumrCagomFcw2JqqtUFEsLLrV82F
|
|
o7gqljKyYYTGwMZRKUSCAQAboJnaN5Bjkneu0d5W4ccViIiOzHFWbTzNumP1Zarr8eeRMbxDW1Mx
|
|
NO67NbkhzNVnmInqzaOZrL93JyZeV0M1++7S02jvxDWxhxx033tPpC8Z6rrezWjmZyazJG2/u03h
|
|
2vFibTHoqvamiwVwY+nLGzV0+SZ1Mx8G0/45tOhzJ5lXMc3UVXRdlF1HP+iYsDPLPPy49/tz1+Te
|
|
pSIr0ho6ak5Ms5J8o2q6NImOrHV7XX488ypzTtHXo0s9t6zG7c1G1qz6ubeZiZ3UatXJG3yauSO7
|
|
cvMTEx5tPLb3prPRMVr0HB8vicNxf0+7+kt+LOJwTJyY/Bnz3tH93X36N58cWvq6LSyiyndMSlC7
|
|
mZcymLJiwLosmJVRLKLAtiU7q4lMSCzc3YxJuDMRuAlKAEgAAAlAkAAAAAABKAEgAAAAAJAAAAAA
|
|
AAAAAAAEgAAAAAAAAAAAAAkAAAAAAAAEAAAAAAAAAAAAAAAAAAAAAhIAAACAAAASgAAAAAAEAAAA
|
|
hGzJAImGMwzQDDZjNVuyNgUTVhNGxysZqDVmiu1G5NN2M4waM0+DCaN2cbGcQNGaMZq3JxMJxA1J
|
|
qx2bU4kU09slorWNwa20z02RXHbJbl26QvtFovbHWkxEdJt5y2MOHlr2U1W3jx+1hiw8vSO63lmI
|
|
XRTaEWmtY6snRHO1VpmJ+DjavpSZl2s8b7y4HFcnh0n0gha5ebJN55KRM2mdoiPN6fh+kpwXh0Wy
|
|
RHj5Otp/s5Ps1p62y31+em9aTMYt/OfVfxTiPjZ52naI7fBrI5t66xz5+a1rW7yx0eSL6iZjtEOX
|
|
qNbSletom3lENjh2fbHzbbWt3iVozruc+5ztWubf4M4ybpQ2Oboyrva0Vjza8WdDR4OkXt3n9ldX
|
|
kaePP9VtYqctYhdvt5oivTeCZ2YOxXk6ubqMfV0b9mrljfqlFcq88k7z2U5axeItDa1OPessuC8P
|
|
ya7XRWYnwqdbT/ZMilvIu4dpslNdixXja8Y5tt85djZdbDWnGOesRtXFtuw6T27No5Kx2OrKYQlC
|
|
ExKJgBnEpiyvdlEgsizKLKollFgWxLKJVRLKJBbEp3VxLKJBnuMWQJEbpBIAAAJAAAABIAAAAAAA
|
|
lAJAAAAAAAAAAAAAASAAAAAAAAAAAAAJAAAABAJABAlAAAAAAAAAAAAAAAAAAAAAAAAIAAAAAAAA
|
|
AAABAJQAAAAgAABAAI2EoBGyJhkgGPKxmqxAKpownHC+YRMdN5BrTj67R3bOn01o7p01Iv71u89o
|
|
b9a7LfBTfS1vWI2jf12VfQPSW8KX2mas+NC2iv6xMNfJpMnLtEbuuxtMRCtzF55NR5rPps1N/ctP
|
|
y6uHreE6nXZ4pak48X3rT06fB7fNeI33cbX6mI32R/MWu7XF116aDSRhxbRERs8f499bkyZeeKae
|
|
kzE2mdon81/tfxDLGOunwbzlzbx08oaHBvZHJlx48mrvaa94pu04y617576rNGLRRM0397JEd/lu
|
|
9Dw/S3x4qxffo6mm4NjwUiKY4iI9Ib1dHFY6QIaNabbrYrLfrpJtaK1rMzPZb/s+05IpP59OyLeJ
|
|
k7eNfRaOc1ue32I7fGXYpi5Y77M8OGMeOKxHSFsU3Y29deZMzirl6dlVvhLatCjJHeYQv1rXnps1
|
|
8k9/VsW6qLVmZIi1rzitlvFKRvaZ2h6TSaenC9FFY+3brM+sqeG8Prp4+kZ+lvuxPkr1mqm95nfp
|
|
DXM459676a2q1dsV7XietvNno78+CJn1cjX6mOeIm0bR33dfRU5NJjidt9t5afjG/V6JZ7I2QMNh
|
|
nyo2BhsMuVG3wAhMSbbQRAMolnE+iuGUSCyJZRKuGUSCyJZK4llEgyZMYTuCUsYSCQASISAAAlCQ
|
|
AAAAAAEoASCASAAAAAAAAAAAAlACRACQAAAAAAAAAEgCEoASCAAAAAAAAAAAAAAAAAAAAAAABAAA
|
|
AAAAAAAISAIAAAAAAQAAACASgAAAQJAQAAhIDHZhln3do7z0WS18mWsajHjmes7pg3dNi5aRMNqO
|
|
yvDHTpPRaigHZhN4hHRlaVN59JY3zRENLUavaO+yq0iNVlitJ6vNcR1MVi0zO0era1/Ea0rPvbz5
|
|
PM5MWp45qvo2GZrhmfrsnpHpHzTCseEcM/2vrr8Q1Eb4qzy44nziPN63HpYiIiI7LNHoqabBTFii
|
|
IpSNohuVxrKtWMEejPwY9G1FFmHB4mWJn7MdfnIM9JpIx15to5pbUaas/a6rqViI7MxPxqX0UT1r
|
|
O3wVzpbR2hviP5i03Y5s6a879FNtHljydhExCv8AMTPJXBnRZbz0iG5ptFjwe/l96zctMVamTJtE
|
|
yTMibu1VrdTzRMR0j0ed4lr64MVpm0RERvMz5NvX62uOJ69XhOKX1HH9bHDtFvNYnfJeOy0Z2ojX
|
|
6jjnEq6fRUmccTvN/J9H0eKcOnx45neaxEbubwHgOHg+milI3vP2resu3Wu0JQmITsmISDHZHKz2
|
|
JgFc1RMLJhGwK9iIZ7MZgEdgmAEwyiWCdwWRLKJVxKYsC2JTuriWUSDNlEsIlMAySx3SCRCQSIAS
|
|
AAACRACQAAAAAAASIASAAAAAAAAAAAAAAACRACRACQASIAAAAAAAAAAAAAAAAAAAAAAAAQCUAAAA
|
|
AAAAAAIAAAAAAAAQAAAAAACBICBICAAEJAQJQCJcLjuS2ny6fPG/LWdpd1o8T0X07SXx/e7wCdJx
|
|
Wa0jmneHQpxPDMdZmJfNtZm49weZrh0/j4o7VtSZ2+Uw0/8A7o49k92vBLc/ntFohFW9PqGXimOI
|
|
6Tu1L8T3eCx6r2t1O3JwvHjifO99v7t/Bwf2l1PXU6rS6eJ8qUm8x+so5TsekzcSjbvs4mt4rzW5
|
|
K2mbT0itesy2cHsvbvqtbmyz5xERWP2jd1tJwrTaONsOKtZ8585+cnDrzmn4Rq+IZObUROHD32n7
|
|
Vv8A0ej0uhxaXFGPFSK1j0bkY4jyZRVZVXFGUVWbGwKsk8mObekNrSW3pWf1a2aYjHbm7bNnQ1id
|
|
PW0TvuDdhJEbQABMsLW2R0ZTMQrvfbz2YWzVhpanUxEd0dWkW5c8R5uXxDX1w4pnfr5Q19XxKuOJ
|
|
2neXltVqtVxbV/RdJ715+1bypANfiOu1HENV9C0MTfNeesx2rD1PAeBYuE6aKx72W3W9/WVnBuB4
|
|
eF4dqRzZbdb5J72l160WVK02ZxCYhOwI23TsnY2BGxsnYBjsiYZsZBjMMZZSgGEolMsQDdG6NwZ7
|
|
piVe6YkFsSziVMWZRILolMSriWUSCyJTuwhMSDMRCQSI3SAlACRCQAAEoAEoASAAAAAAAAACUACR
|
|
ACQAAAAAAAAAAAAASAAAAAAAAAAAAAAAAAAACAAAAAAAAAAAAAABAAAAAAAAAAAAACBKAAAAAAAQ
|
|
JQAAAhICEbJAYTWJ7wx8KvpC0BV4ceieWGewDHlNmWwCNjZICNhIDmcZredBecdpiY69FXCOLW+i
|
|
UiZidukulmxxlx2paN4mNng+K4+I8Hy2yaTfl37TXetoCPfRxfp1qi3F48ofKMvtvxak8s6LDv61
|
|
rZji9rPaLUf5PC+bfttS0q8q3p9W/wBrRMdpUZuKdN99nzvFqPbTVz7nD8OKs+do2/mW3h4D7Xaq
|
|
ZnPrtNpqz35aRaYOHY9Zk4pNt9rR+rl6zi+OnS+WN57Rv1lXp/YrNaYtruL6zNPnGO3hxP6O5w/2
|
|
f0HDuun09Yv55Le9afznqcOvO4tBreMTHu30unnva0bWt8on+70nDuE4OHYYx4Kbesz3tPrMuhGO
|
|
IjpDOKrK9YVpsyiGUQnYGOyUgI2SlAIEmwMWMs9kTAMJYzDOYRMArmGErZhhMArlHmzmGMwDE3Ts
|
|
bAbs4swj5pgFkSziVcM4BZEsolXDKAZwyhjCYBkACQhIAAAAAAAJAAAAAAAAAAAAAAAAAAAShIAA
|
|
AAAAAAJAAAAAAAAAAAAAABAJEAAAAAAAAAAAAAAAIEoBKAAAAAAAAAAAAAAABAlAAAAAAAIAAAAA
|
|
BAkBAkBAkBAlACEgMZjdjbFW8bWrEx8YWANb6Fp+bfwab+vLDKMFK9qxH5L0bAr8OPRPKz2AY7J2
|
|
SbAjYZAI2E7AIEgIEgIEgMdkSy2NgY7MdlmyNoBXsxmFuyNgVTVjNV3KjlBRNTlXTVHKCrlIqt5T
|
|
lBhEMohlFerLlBjEMohMVTEARDKCITsAk2AEgAAAkAAAAAAAAAAAAAAAAAAAAAAAASAAAAAAAAD/
|
|
2Q==`;async function _xe(e){let t=(r,a="application/octet-stream")=>fetch(`data:${a};base64,${r}`).then(o=>o.blob()),n,s;switch(e.config.warmup){case"face":n=await t(y1);break;case"body":case"full":n=await t(A1);break;default:n=null}if(n){let r=await createImageBitmap(n);s=await e.detect(r,e.config),r.close()}return s}async function Dxe(e){return new Promise(t=>{let n;switch(e.config.warmup){case"face":n="data:image/jpeg;base64,"+y1;break;case"full":case"body":n="data:image/jpeg;base64,"+A1;break;default:n=null}let s;if(typeof Image!="undefined")s=new Image;else if(pe.Image)s=new pe.Image;else return;s.onload=async()=>{let r=is(s.naturalWidth,s.naturalHeight);if(!r)le("Warmup: Canvas not found"),t(void 0);else{let a=r.getContext("2d");a&&a.drawImage(s,0,0);let o=await e.image(r),i=await e.detect(o.tensor,e.config);t(i)}},n?s.src=n:t(void 0)})}async function $xe(e){let t=r=>Buffer.from(r,"base64"),n;e.config.warmup==="face"?n=t(y1):n=t(A1);let s;if("node"in je){let r=(void 0).decodeJpeg(n),a=r.expandDims(0);e.tf.dispose(r),s=await e.detect(a,e.config),e.tf.dispose(a)}else e.config.debug&&le("Warmup tfjs-node not loaded");return s}async function Pxe(e){let t;return typeof createImageBitmap=="function"?t=await _xe(e):typeof Image!="undefined"||pe.Canvas!==void 0?t=await Dxe(e):t=await $xe(e),t}async function Fxe(e){if(!q().flagRegistry.ENGINE_COMPILE_ONLY)return;let t=ss(),n=Ys();if(t!=="webgl"&&t!=="humangl"||!n||!n.checkCompileCompletion)return;q().set("ENGINE_COMPILE_ONLY",!0);let s=sn().state.numTensors,r=[];for(let[i,l]of Object.entries(e).filter(([u,c])=>u!==null&&c!==null)){let u=l.inputs&&l.inputs[0]&&l.inputs[0].shape?[...l.inputs[0].shape]:[1,64,64,3],c=l.inputs&&l.inputs[0]&&l.inputs[0].dtype?l.inputs[0].dtype:"float32";for(let d=0;d<u.length;d++)u[d]===-1&&(u[d]=d===0?1:64);let p=Wt(u,c);try{let d=l.execute(p);r.push(i),Array.isArray(d)?d.forEach(h=>ee(h)):ee(d)}catch(d){le("compile fail model:",i)}ee(p)}let a=await n.checkCompileCompletionAsync();n.getUniformLocations(),le("compile pass models:",r),le("compile pass kernels:",a.length),q().set("ENGINE_COMPILE_ONLY",!1);let o=sn().state.numTensors;o-s>0&&le("tensor leak:",o-s)}async function rR(e,t){let n=ue();return e.state="warmup",t&&(e.config=Xt(e.config,t)),!e.config.warmup||e.config.warmup.length===0||e.config.warmup==="none"?{face:[],body:[],hand:[],gesture:[],object:[],performance:e.performance,timestamp:ue(),persons:[],error:null}:new Promise(async s=>{await Fxe(e.models);let r=await Pxe(e),a=ue();e.config.debug&&le("warmup",e.config.warmup,Math.round(a-n),"ms"),e.emit("warmup"),s(r)})}var Od,tf,nf,x1,U4=class{constructor(t){ge(this,"version");ge(this,"config");ge(this,"result");ge(this,"state");ge(this,"process");ge(this,"tf");ge(this,"env");ge(this,"draw");ge(this,"models");ge(this,"events");ge(this,"faceTriangulation");ge(this,"faceUVMap");ge(this,"performance");Yd(this,Od,void 0);Yd(this,tf,void 0);Yd(this,nf,void 0);ge(this,"gl");ge(this,"analyze",(...t)=>{if(!Zd(this,tf))return;let n=this.tf.engine().state.numTensors,s=Zd(this,Od);Jd(this,Od,n);let r=n-s;r!==0&&le(...t,r)});Yd(this,x1,t=>{if(!Zd(this,nf))return null;if(!t)return"input is not defined";if(this.env.node&&!(t instanceof nt))return"input must be a tensor";try{this.tf.getBackend()}catch(n){return"backend not loaded"}return null});ge(this,"similarity",B4);ge(this,"distance",ef);ge(this,"match",W4);ge(this,"emit",t=>{var n;this.events&&this.events.dispatchEvent&&((n=this.events)==null||n.dispatchEvent(new Event(t)))});var s;this.env=pe;let n=(((s=Uh)==null?void 0:s.tfjs)||Hy).replace(/-(.*)/,"");Ba.wasmPath=`https://cdn.jsdelivr.net/npm/@tensorflow/tfjs-backend-wasm@${n}/dist/`,Ba.modelBasePath=pe.browser?"../models/":"file://models/",Ba.backend=pe.browser?"humangl":"tensorflow",this.version=fb,Object.defineProperty(this,"version",{value:fb}),this.config=JSON.parse(JSON.stringify(Ba)),Object.seal(this.config),this.config.cacheModels=typeof indexedDB!="undefined",t&&(this.config=Xt(this.config,t)),TT(this.config),this.tf=je,this.state="idle",Jd(this,Od,0),Jd(this,tf,!1),Jd(this,nf,!1),this.performance={},this.events=typeof EventTarget!="undefined"?new EventTarget:void 0,this.models=new Qh,this.draw={options:Un,canvas:(r,a)=>P4(r,a),face:(r,a,o)=>Rd(r,a,o),body:(r,a,o)=>_d(r,a,o),hand:(r,a,o)=>Dd(r,a,o),gesture:(r,a,o)=>Pd(r,a,o),object:(r,a,o)=>$d(r,a,o),person:(r,a,o)=>$4(r,a,o),all:(r,a,o)=>F4(r,a,o)},this.result={face:[],body:[],hand:[],gesture:[],object:[],performance:{},timestamp:0,persons:[],error:null},this.process={tensor:null,canvas:null},this.faceTriangulation=GN,this.faceUVMap=HN,this.gl=Pt,this.emit("create")}reset(){let t=this.config.backend;this.config=JSON.parse(JSON.stringify(Ba)),this.config.backend=t}validate(t){return t3(Ba,t||this.config)}now(){return ue()}image(t,n=!0){return gd(t,this.config,n)}async segmentation(t,n){return GE(t,n,this.config)}enhance(t){return Jb(t)}compare(t,n){return CT(this.config,t,n)}async init(){await g1(this,!0),await this.tf.ready()}async load(t){this.state="load";let n=ue(),s=Object.values(this.models).filter(o=>o).length;t&&(this.config=Xt(this.config,t)),this.env.initial&&(this.config.debug&&le(`version: ${this.version}`),this.config.debug&&le(`tfjs version: ${this.tf.version["tfjs-core"]}`),await g1(this)||le("error: backend check failed"),await Lc(),this.env.browser&&(this.config.debug&&le("configuration:",this.config),this.config.debug&&le("environment:",this.env),this.config.debug&&le("tf flags:",this.tf.ENV.flags))),await N4(this),this.env.initial&&this.config.debug&&le("tf engine state:",this.tf.engine().state.numBytes,"bytes",this.tf.engine().state.numTensors,"tensors"),this.env.initial=!1,Object.values(this.models).filter(o=>o).length!==s&&(await E4(this),this.emit("load"));let a=Math.trunc(ue()-n);a>(this.performance.loadModels||0)&&(this.performance.loadModels=this.env.perfadd?(this.performance.loadModels||0)+a:a)}next(t=this.result){return tR(t,this.config)}getModelStats(){return T4(this)}async warmup(t){let n=ue(),s=await rR(this,t),r=ue();return this.performance.warmup=Math.trunc(r-n),s}async profile(t,n){let s=await this.tf.profile(()=>this.detect(t,n)),r={},a=0;for(let i of s.kernels)r[i.name]?r[i.name]+=i.kernelTimeMs:r[i.name]=i.kernelTimeMs,a+=i.kernelTimeMs;let o=[];Object.entries(r).forEach(i=>o.push({kernel:i[0],time:i[1],perc:0}));for(let i of o)i.perc=Math.round(1e3*i.time/a)/1e3,i.time=Math.round(1e3*i.time)/1e3;return o.sort((i,l)=>l.time-i.time),o.length=20,o}async detect(t,n){return this.state="detect",new Promise(async s=>{var g,y,x,A,b,w,k,S,E,R,$,_,P,C,F,G,K,z,Z,J,te,B;this.state="config";let r;this.config=Xt(this.config,n),this.state="check";let a=Zd(this,x1).call(this,t);a&&(le(a,t),this.emit("error"),s({face:[],body:[],hand:[],gesture:[],object:[],performance:this.performance,timestamp:ue(),persons:[],error:a}));let o=ue();await g1(this),await this.load(),r=ue(),this.state="image";let i=await gd(t,this.config);if(this.process=i,this.performance.inputProcess=this.env.perfadd?(this.performance.inputProcess||0)+Math.trunc(ue()-r):Math.trunc(ue()-r),this.analyze("Get Image:"),!i.tensor){this.config.debug&&le("could not convert input to tensor"),this.emit("error"),s({face:[],body:[],hand:[],gesture:[],object:[],performance:this.performance,timestamp:ue(),persons:[],error:"could not convert input to tensor"});return}this.emit("image"),r=ue(),this.config.skipAllowed=await ST(this.config,i.tensor),this.performance.totalFrames||(this.performance.totalFrames=0),this.performance.cachedFrames||(this.performance.cachedFrames=0),this.performance.totalFrames++,this.config.skipAllowed&&this.performance.cachedFrames++,this.performance.cacheCheck=this.env.perfadd?(this.performance.cacheCheck||0)+Math.trunc(ue()-r):Math.trunc(ue()-r),this.analyze("Check Changed:");let l=[],u=[],c=[],p=[];this.state="detect:face",this.config.async?(l=this.config.face.enabled?z4(this,i.tensor):[],this.performance.face&&delete this.performance.face):(r=ue(),l=this.config.face.enabled?await z4(this,i.tensor):[],this.performance.face=this.env.perfadd?(this.performance.face||0)+Math.trunc(ue()-r):Math.trunc(ue()-r)),this.config.async&&(this.config.body.maxDetected===-1||this.config.hand.maxDetected===-1)&&(l=await l),this.analyze("Start Body:"),this.state="detect:body";let d=this.config.body.maxDetected===-1?Xt(this.config,{body:{maxDetected:this.config.face.enabled?1*l.length:1}}):this.config;this.config.async?((g=this.config.body.modelPath)!=null&&g.includes("posenet")?u=this.config.body.enabled?I4(i.tensor,d):[]:(y=this.config.body.modelPath)!=null&&y.includes("blazepose")?u=this.config.body.enabled?Pb(i.tensor,d):[]:(x=this.config.body.modelPath)!=null&&x.includes("efficientpose")?u=this.config.body.enabled?Wb(i.tensor,d):[]:(A=this.config.body.modelPath)!=null&&A.includes("movenet")&&(u=this.config.body.enabled?y4(i.tensor,d):[]),this.performance.body&&delete this.performance.body):(r=ue(),(b=this.config.body.modelPath)!=null&&b.includes("posenet")?u=this.config.body.enabled?await I4(i.tensor,d):[]:(w=this.config.body.modelPath)!=null&&w.includes("blazepose")?u=this.config.body.enabled?await Pb(i.tensor,d):[]:(k=this.config.body.modelPath)!=null&&k.includes("efficientpose")?u=this.config.body.enabled?await Wb(i.tensor,d):[]:(S=this.config.body.modelPath)!=null&&S.includes("movenet")&&(u=this.config.body.enabled?await y4(i.tensor,d):[]),this.performance.body=this.env.perfadd?(this.performance.body||0)+Math.trunc(ue()-r):Math.trunc(ue()-r)),this.analyze("End Body:"),this.analyze("Start Hand:"),this.state="detect:hand";let h=this.config.hand.maxDetected===-1?Xt(this.config,{hand:{maxDetected:this.config.face.enabled?2*l.length:1}}):this.config;this.config.async?((R=(E=this.config.hand.detector)==null?void 0:E.modelPath)!=null&&R.includes("handdetect")?c=this.config.hand.enabled?a4(i.tensor,h):[]:(_=($=this.config.hand.detector)==null?void 0:$.modelPath)!=null&&_.includes("handtrack")&&(c=this.config.hand.enabled?u4(i.tensor,h):[]),this.performance.hand&&delete this.performance.hand):(r=ue(),(C=(P=this.config.hand.detector)==null?void 0:P.modelPath)!=null&&C.includes("handdetect")?c=this.config.hand.enabled?await a4(i.tensor,h):[]:(G=(F=this.config.hand.detector)==null?void 0:F.modelPath)!=null&&G.includes("handtrack")&&(c=this.config.hand.enabled?await u4(i.tensor,h):[]),this.performance.hand=this.env.perfadd?(this.performance.hand||0)+Math.trunc(ue()-r):Math.trunc(ue()-r)),this.analyze("End Hand:"),this.analyze("Start Object:"),this.state="detect:object",this.config.async?((K=this.config.object.modelPath)!=null&&K.includes("nanodet")?p=this.config.object.enabled?x4(i.tensor,this.config):[]:(z=this.config.object.modelPath)!=null&&z.includes("centernet")&&(p=this.config.object.enabled?Mb(i.tensor,this.config):[]),this.performance.object&&delete this.performance.object):(r=ue(),(Z=this.config.object.modelPath)!=null&&Z.includes("nanodet")?p=this.config.object.enabled?await x4(i.tensor,this.config):[]:(J=this.config.object.modelPath)!=null&&J.includes("centernet")&&(p=this.config.object.enabled?await Mb(i.tensor,this.config):[]),this.performance.object=this.env.perfadd?(this.performance.object||0)+Math.trunc(ue()-r):Math.trunc(ue()-r)),this.analyze("End Object:"),this.state="detect:await",this.config.async&&([l,u,c,p]=await Promise.all([l,u,c,p])),this.state="detect:gesture";let f=[];this.config.gesture.enabled&&(r=ue(),f=[...JE(l),...YE(u),...eR(c),...QE(l)],this.config.async?this.performance.gesture&&delete this.performance.gesture:this.performance.gesture=this.env.perfadd?(this.performance.gesture||0)+Math.trunc(ue()-r):Math.trunc(ue()-r)),this.performance.total=this.env.perfadd?(this.performance.total||0)+Math.trunc(ue()-o):Math.trunc(ue()-o);let m=((B=(te=this.process)==null?void 0:te.tensor)==null?void 0:B.shape)||[];this.result={face:l,body:u,hand:c,gesture:f,object:p,performance:this.performance,canvas:this.process.canvas,timestamp:Date.now(),error:null,get persons(){return sR(l,u,c,f,m)}},ee(i.tensor),this.emit("detect"),this.state="idle",s(this.result)})}};Od=new WeakMap,tf=new WeakMap,nf=new WeakMap,x1=new WeakMap;return s_(Mxe);})();
|