ambiser
/
human
mirror of https://github.com/vladmandic/human
9
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
human/dist/human.js

5568 lines
1.5 MiB
Raw Blame History

/*
Human library
homepage: <https://github.com/vladmandic/human>
author: <https://github.com/vladmandic>'
*/
var Human=(()=>{var b3=Object.defineProperty;var qR=e=>b3(e,"__esModule",{value:!0});var Kg=e=>{if(typeof require!="undefined")return require(e);throw new Error('Dynamic require of "'+e+'" is not supported')};var Xg=(e,t)=>{qR(e);for(var n in t)b3(e,n,{get:t[n],enumerable:!0})};var v3=(e,t,n)=>{if(!t.has(e))throw TypeError("Cannot "+n)};var Dn=(e,t,n)=>(v3(e,t,"read from private field"),n?n.call(e):t.get(e)),wr=(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)},es=(e,t,n,r)=>(v3(e,t,"write to private field"),r?r.call(e,n):t.set(e,n),n);var Dwe={};Xg(Dwe,{Human:()=>sR,default:()=>sR});function Ct(e,t){let n=e.endsWith("/")?"":"/",s=t.startsWith(".")||t.startsWith("/")||t.startsWith("http:")||t.startsWith("https:")||t.startsWith("file:")?`${t}`:`${e}${n}${t}`;if(!s.toLocaleLowerCase().includes(".json"))throw new Error(`Human: ModelPath Error: ${s} Expecting JSON file`);return s}function fe(...e){let t=new Date,n=`${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)}var nt=()=>typeof performance!="undefined"?performance.now():parseInt((Number(process.hrtime.bigint())/1e3/1e3).toString());function Fn(...e){let t=n=>n&&typeof n=="object";return e.reduce((n,r)=>(Object.keys(r||{}).forEach(s=>{let a=n[s],o=r[s];Array.isArray(a)&&Array.isArray(o)?n[s]=a.concat(...o):t(a)&&t(o)?n[s]=Fn(a,o):n[s]=o}),n),{})}var w3={backend:"webgl",modelBasePath:"../models/",wasmPath:"../node_modules/@tensorflow/tfjs-backend-wasm/dist/",debug:!0,async:!0,warmup:"full",cacheSensitivity:.75,skipFrame:!1,filter:{enabled:!0,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:15,skipFrames:15,minConfidence:.2,iouThreshold:.1,return:!1},mesh:{enabled:!0,modelPath:"facemesh.json"},iris:{enabled:!0,modelPath:"iris.json"},description:{enabled:!0,modelPath:"faceres.json",skipFrames:11,minConfidence:.1},emotion:{enabled:!0,minConfidence:.1,skipFrames:17,modelPath:"emotion.json"}},body:{enabled:!0,modelPath:"movenet-lightning.json",maxDetected:1,minConfidence:.2,skipFrames:1},hand:{enabled:!0,rotation:!0,skipFrames:18,minConfidence:.1,iouThreshold:.1,maxDetected:2,landmarks:!0,detector:{modelPath:"handdetect.json"},skeleton:{modelPath:"handskeleton.json"}},object:{enabled:!1,modelPath:"mb3-centernet.json",minConfidence:.2,iouThreshold:.4,maxDetected:10,skipFrames:19},segmentation:{enabled:!1,modelPath:"selfie.json"}};function k3(){let e,t;if(typeof navigator!="undefined"){let n=navigator.userAgent.match(/\(([^()]+)\)/g);if(n&&n[0]){let r=n[0].match(/\(([^()]+)\)/g);e=r?r[0].replace(/\(|\)/g,""):"",t=navigator.userAgent.replace(n[0],""),e[1]&&(t=t.replace(n[1],"")),t=t.replace(/ /g," ")}}else typeof process!="undefined"&&(e=`${process.platform} ${process.arch}`,t=`NodeJS ${process.version}`);return{platform:e,agent:t}}var Ah={};Xg(Ah,{Abs:()=>U3,Acos:()=>H3,Acosh:()=>G3,AdadeltaOptimizer:()=>Dp,AdagradOptimizer:()=>Fp,AdamOptimizer:()=>Mp,AdamaxOptimizer:()=>Op,Add:()=>r2,AddN:()=>j3,All:()=>q3,Any:()=>K3,ArgMax:()=>X3,ArgMin:()=>Z3,Asin:()=>Y3,Asinh:()=>J3,Atan:()=>Q3,Atan2:()=>tv,Atanh:()=>ev,AvgPool:()=>nv,AvgPool3D:()=>rv,AvgPool3DGrad:()=>Y_,AvgPoolGrad:()=>Z_,BackendWasm:()=>t$,BatchMatMul:()=>sv,BatchToSpaceND:()=>av,Bincount:()=>ov,BroadcastTo:()=>J_,Callback:()=>I8,CallbackList:()=>hS,Cast:()=>s2,Ceil:()=>iv,ClipByValue:()=>lv,Complex:()=>uv,ComplexAbs:()=>cv,Concat:()=>dv,Conv2D:()=>hv,Conv2DBackpropFilter:()=>pv,Conv2DBackpropInput:()=>fv,Conv3D:()=>mv,Conv3DBackpropFilterV2:()=>Q_,Conv3DBackpropInputV2:()=>gv,Cos:()=>yv,Cosh:()=>Av,CropAndResize:()=>bv,Cumsum:()=>xv,CustomCallback:()=>fS,DataStorage:()=>E_,DenseBincount:()=>vv,DepthToSpace:()=>wv,DepthwiseConv2dNative:()=>kv,DepthwiseConv2dNativeBackpropFilter:()=>Iv,DepthwiseConv2dNativeBackpropInput:()=>Sv,Diag:()=>Tv,Dilation2D:()=>Nv,Dilation2DBackpropFilter:()=>tD,Dilation2DBackpropInput:()=>eD,ENV:()=>kr,EarlyStopping:()=>T8,Einsum:()=>Ev,Elu:()=>$v,EluGrad:()=>nD,Environment:()=>W3,Equal:()=>_v,Erf:()=>Rv,Exp:()=>Dv,ExpandDims:()=>Fv,Expm1:()=>Mv,FFT:()=>Ov,Fill:()=>Pv,FlipLeftRight:()=>zv,Floor:()=>Lv,FloorDiv:()=>Bv,FromPixels:()=>i2,FusedBatchNorm:()=>Wv,FusedConv2D:()=>u2,FusedDepthwiseConv2D:()=>c2,GPGPUContext:()=>Fm,GatherNd:()=>Uv,GatherV2:()=>Vv,GraphModel:()=>sT,Greater:()=>Hv,GreaterEqual:()=>Gv,History:()=>pS,IFFT:()=>jv,Identity:()=>a2,Imag:()=>qv,InputSpec:()=>Qt,IsFinite:()=>Kv,IsInf:()=>Xv,IsNan:()=>Zv,KernelBackend:()=>E3,LRN:()=>ow,LRNGrad:()=>sD,LayerVariable:()=>iS,LayersModel:()=>fa,LeakyRelu:()=>Yv,Less:()=>Jv,LessEqual:()=>Qv,LinSpace:()=>ew,Log:()=>tw,Log1p:()=>nw,LogSoftmax:()=>rD,LogicalAnd:()=>rw,LogicalNot:()=>sw,LogicalOr:()=>aw,MathBackendCPU:()=>x5,MathBackendWebGL:()=>uh,Max:()=>iw,MaxPool:()=>uw,MaxPool3D:()=>cw,MaxPool3DGrad:()=>oD,MaxPoolGrad:()=>aD,MaxPoolWithArgmax:()=>dw,Maximum:()=>lw,Mean:()=>hw,Min:()=>pw,Minimum:()=>fw,MirrorPad:()=>mw,Mod:()=>gw,MomentumOptimizer:()=>Pp,Multinomial:()=>yw,Multiply:()=>Aw,Neg:()=>xw,NonMaxSuppressionV3:()=>vw,NonMaxSuppressionV4:()=>ww,NonMaxSuppressionV5:()=>kw,NotEqual:()=>bw,OP_SCOPE_SUFFIX:()=>B7,OneHot:()=>Sw,OnesLike:()=>Iw,Optimizer:()=>Da,Pack:()=>Tw,PadV2:()=>Nw,Pool:()=>iD,Pow:()=>Cw,Prelu:()=>Ew,Prod:()=>$w,RMSPropOptimizer:()=>zp,RNN:()=>ma,Range:()=>Rw,Rank:()=>m2,Real:()=>_w,RealDiv:()=>Cv,Reciprocal:()=>Dw,Reduction:()=>Pn,Relu:()=>Fw,Relu6:()=>zw,Reshape:()=>Mw,ResizeBilinear:()=>Pw,ResizeBilinearGrad:()=>uD,ResizeNearestNeighbor:()=>Ow,ResizeNearestNeighborGrad:()=>lD,Reverse:()=>Lw,RotateWithOffset:()=>k7,Round:()=>Bw,Rsqrt:()=>Ww,SGDOptimizer:()=>fc,ScatterNd:()=>Vw,Select:()=>Uw,Selu:()=>Hw,Sequential:()=>im,Sigmoid:()=>Xw,Sign:()=>Kw,Sin:()=>jw,Sinh:()=>qw,Slice:()=>Gw,Softmax:()=>t7,Softplus:()=>Zw,SpaceToBatchND:()=>Qw,SparseFillEmptyRows:()=>n7,SparseReshape:()=>r7,SparseSegmentMean:()=>s7,SparseSegmentSum:()=>a7,SparseToDense:()=>o7,SplitV:()=>e7,Sqrt:()=>Yw,Square:()=>cD,SquaredDifference:()=>i7,Step:()=>w7,StridedSlice:()=>l7,StringNGrams:()=>u7,StringSplit:()=>c7,StringToHashBucketFast:()=>d7,Sub:()=>h7,Sum:()=>Jw,SymbolicTensor:()=>hs,Tan:()=>p7,Tanh:()=>f7,Tensor:()=>It,TensorBuffer:()=>lp,Tile:()=>o2,TopK:()=>m7,Transform:()=>g7,Transpose:()=>y7,Unique:()=>A7,Unpack:()=>x7,UnsortedSegmentSum:()=>b7,Variable:()=>nc,ZerosLike:()=>v7,_FusedMatMul:()=>l2,abs:()=>Sr,acos:()=>$M,acosh:()=>_M,add:()=>Me,addN:()=>j2,all:()=>MM,any:()=>PM,argMax:()=>q2,argMin:()=>BM,asin:()=>VM,asinh:()=>HM,atan:()=>jM,atan2:()=>KM,atanh:()=>ZM,avgPool:()=>_k,avgPool3d:()=>oO,backend:()=>wM,backend_util:()=>v4,basicLSTMCell:()=>hO,batchNorm:()=>yp,batchNorm2d:()=>yO,batchNorm3d:()=>xO,batchNorm4d:()=>vO,batchToSpaceND:()=>Dk,bincount:()=>Fk,booleanMaskAsync:()=>$B,broadcastTo:()=>Ap,browser:()=>Br,buffer:()=>Js,callbacks:()=>Kre,cast:()=>Mt,ceil:()=>SO,clipByValue:()=>NO,clone:()=>Qs,complex:()=>go,concat:()=>rn,concat1d:()=>EO,concat2d:()=>ic,concat3d:()=>_O,concat4d:()=>FO,constraints:()=>BI,conv1d:()=>PO,conv2d:()=>xp,conv2dTranspose:()=>BO,conv3d:()=>VO,conv3dTranspose:()=>jO,copyRegisteredKernels:()=>fD,cos:()=>KO,cosh:()=>ZO,cosineWindow:()=>i1,cumsum:()=>JO,customGrad:()=>Ss,data:()=>aT,denseBincount:()=>eP,deprecationWarn:()=>Tk,depthToSpace:()=>nP,depthwiseConv2d:()=>Y2,deregisterOp:()=>Zre,device_util:()=>O7,diag:()=>aP,dilation2d:()=>iP,disableDeprecationWarnings:()=>dM,dispose:()=>We,disposeVariables:()=>hM,div:()=>Je,divNoNan:()=>pP,dot:()=>mP,dropout:()=>UB,einsum:()=>yP,elu:()=>zk,enableDebugMode:()=>cM,enableProdMode:()=>uM,enclosingPowerOfTwo:()=>h4,engine:()=>pM,env:()=>ct,equal:()=>Pk,erf:()=>bP,exp:()=>wo,expandDims:()=>ta,expm1:()=>IP,eye:()=>Lk,fft:()=>s1,fill:()=>vp,findBackend:()=>H2,findBackendFactory:()=>vM,floor:()=>Bk,floorDiv:()=>Ck,forceHalfFloat:()=>uE,fused:()=>p4,gather:()=>Wk,gatherND:()=>BB,gather_util:()=>ik,getBackend:()=>xM,getGradient:()=>d2,getKernel:()=>np,getKernelsForBackend:()=>zi,gpgpu_util:()=>uC,grad:()=>qP,grads:()=>KP,greater:()=>wp,greaterEqual:()=>Vk,ifft:()=>Cp,imag:()=>J2,image:()=>Ze,inTopKAsync:()=>GB,initializers:()=>qI,input:()=>PS,io:()=>Q7,irfft:()=>o4,isFinite:()=>DP,isInf:()=>MP,isNaN:()=>PP,keep:()=>Nk,kernel_impls:()=>S4,layers:()=>sS,leakyRelu:()=>Uk,less:()=>BP,lessEqual:()=>Q2,linalg:()=>$V,linspace:()=>VP,loadGraphModel:()=>Nt,loadLayersModel:()=>ane,localResponseNormalization:()=>HP,log:()=>lc,log1p:()=>Hk,logSigmoid:()=>ez,logSoftmax:()=>az,logSumExp:()=>Kk,logicalAnd:()=>Ip,logicalNot:()=>Xk,logicalOr:()=>Zk,logicalXor:()=>gz,losses:()=>RV,matMul:()=>gt,math:()=>ak,max:()=>_a,maxPool:()=>Yk,maxPool3d:()=>xz,maxPoolWithArgmax:()=>vz,maximum:()=>Jk,mean:()=>Sp,memory:()=>fM,meshgrid:()=>Iz,metrics:()=>v8,min:()=>t1,minimum:()=>Qk,mirrorPad:()=>Cz,mod:()=>$z,model:()=>rne,models:()=>w8,moments:()=>Dz,movingAverage:()=>DB,mul:()=>pe,multiRNNCell:()=>Mz,multinomial:()=>Pz,neg:()=>Ra,nextFrame:()=>OV,norm:()=>o1,notEqual:()=>e4,oneHot:()=>P2,ones:()=>ko,onesLike:()=>Bz,op:()=>H,outerProduct:()=>Vz,pad:()=>cc,pad1d:()=>Gz,pad2d:()=>qz,pad3d:()=>Xz,pad4d:()=>Yz,pool:()=>nL,pow:()=>dc,prelu:()=>n4,print:()=>J7,prod:()=>oL,profile:()=>mM,rand:()=>lL,randomGamma:()=>hL,randomNormal:()=>fL,randomUniform:()=>r4,range:()=>hc,ready:()=>AM,real:()=>Tp,reciprocal:()=>AL,registerBackend:()=>G2,registerCallbackConstructor:()=>one,registerGradient:()=>dD,registerKernel:()=>rp,registerOp:()=>Xre,regularizers:()=>k8,relu:()=>Np,relu6:()=>s4,removeBackend:()=>bM,reshape:()=>le,reverse:()=>Io,reverse1d:()=>kL,reverse2d:()=>SL,reverse3d:()=>NL,reverse4d:()=>EL,rfft:()=>a1,round:()=>a4,rsqrt:()=>_L,scalar:()=>ut,scatterND:()=>MB,scatter_util:()=>uk,selu:()=>FL,separableConv2d:()=>OL,sequential:()=>sne,serialization:()=>vk,setBackend:()=>yM,setPlatform:()=>kM,setWasmPath:()=>Xve,setWasmPaths:()=>Zve,setWebGLContext:()=>Nm,setdiff1dAsync:()=>zL,shared:()=>ET,sigmoid:()=>Is,sign:()=>BL,signal:()=>EV,sin:()=>VL,sinh:()=>HL,slice:()=>Xe,slice1d:()=>jL,slice2d:()=>KL,slice3d:()=>ZL,slice4d:()=>JL,slice_util:()=>B2,softmax:()=>eB,softplus:()=>jk,spaceToBatchND:()=>t4,sparse:()=>_V,sparseToDense:()=>zB,spectral:()=>CV,split:()=>na,sqrt:()=>ra,square:()=>rs,squaredDifference:()=>i4,squeeze:()=>Xn,stack:()=>So,step:()=>l4,stridedSlice:()=>hB,string:()=>DV,sub:()=>Ue,sum:()=>Et,sumOutType:()=>OD,tan:()=>fB,tanh:()=>Z2,tensor:()=>ns,tensor1d:()=>ur,tensor2d:()=>sa,tensor3d:()=>fp,tensor4d:()=>mB,tensor5d:()=>gB,tensor6d:()=>yB,tensor_util:()=>R7,test_util:()=>kk,tidy:()=>Ve,tile:()=>bp,time:()=>gM,topk:()=>xB,train:()=>FV,transpose:()=>pp,truncatedNormal:()=>vB,unique:()=>kB,unregisterGradient:()=>pD,unregisterKernel:()=>hD,unsortedSegmentSum:()=>SB,unstack:()=>pc,upcastType:()=>up,util:()=>I7,valueAndGrad:()=>XP,valueAndGrads:()=>ZP,variable:()=>NB,variableGrads:()=>Gk,version:()=>Qve,version_converter:()=>Qse,version_core:()=>lM,version_cpu:()=>Foe,version_layers:()=>UA,version_wasm:()=>Yve,version_webgl:()=>$fe,webgl:()=>Rfe,webgl_util:()=>ON,where:()=>Hi,whereAsync:()=>c4,zeros:()=>Gi,zerosLike:()=>Tr});var KR=Object.create,Yh=Object.defineProperty,XR=Object.getOwnPropertyDescriptor,ZR=Object.getOwnPropertyNames,YR=Object.getPrototypeOf,JR=Object.prototype.hasOwnProperty,I3=e=>Yh(e,"__esModule",{value:!0}),co=e=>{if(typeof Kg!="undefined")return Kg(e);throw new Error('Dynamic require of "'+e+'" is not supported')},Ft=(e,t)=>function(){return t||(0,e[Object.keys(e)[0]])((t={exports:{}}).exports,t),t.exports},_e=(e,t)=>{I3(e);for(var n in t)Yh(e,n,{get:t[n],enumerable:!0})},QR=(e,t,n)=>{if(t&&typeof t=="object"||typeof t=="function")for(let r of ZR(t))!JR.call(e,r)&&r!=="default"&&Yh(e,r,{get:()=>t[r],enumerable:!(n=XR(t,r))||n.enumerable});return e},Xs=e=>QR(I3(Yh(e!=null?KR(YR(e)):{},"default",e&&e.__esModule&&"default"in e?{get:()=>e.default,enumerable:!0}:{value:e,enumerable:!0})),e),S3=Ft({"node_modules/.pnpm/long@4.0.0/node_modules/long/src/long.js"(e,t){t.exports=r;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(R){}function r(R,N,M){this.low=R|0,this.high=N|0,this.unsigned=!!M}r.prototype.__isLong__,Object.defineProperty(r.prototype,"__isLong__",{value:!0});function s(R){return(R&&R.__isLong__)===!0}r.isLong=s;var a={},o={};function i(R,N){var M,B,q;return N?(R>>>=0,(q=0<=R&&R<256)&&(B=o[R],B)?B:(M=u(R,(R|0)<0?-1:0,!0),q&&(o[R]=M),M)):(R|=0,(q=-128<=R&&R<128)&&(B=a[R],B)?B:(M=u(R,R<0?-1:0,!1),q&&(a[R]=M),M))}r.fromInt=i;function l(R,N){if(isNaN(R))return N?b:x;if(N){if(R<0)return b;if(R>=g)return E}else{if(R<=-y)return D;if(R+1>=y)return S}return R<0?l(-R,N).neg():u(R%m|0,R/m|0,N)}r.fromNumber=l;function u(R,N,M){return new r(R,N,M)}r.fromBits=u;var c=Math.pow;function d(R,N,M){if(R.length===0)throw Error("empty string");if(R==="NaN"||R==="Infinity"||R==="+Infinity"||R==="-Infinity")return x;if(typeof N=="number"?(M=N,N=!1):N=!!N,M=M||10,M<2||36<M)throw RangeError("radix");var B;if((B=R.indexOf("-"))>0)throw Error("interior hyphen");if(B===0)return d(R.substring(1),N,M).neg();for(var q=l(c(M,8)),X=x,J=0;J<R.length;J+=8){var ee=Math.min(8,R.length-J),ae=parseInt(R.substring(J,J+ee),M);if(ee<8){var se=l(c(M,ee));X=X.mul(se).add(l(ae))}else X=X.mul(q),X=X.add(l(ae))}return X.unsigned=N,X}r.fromString=d;function h(R,N){return typeof R=="number"?l(R,N):typeof R=="string"?d(R,N):u(R.low,R.high,typeof N=="boolean"?N:R.unsigned)}r.fromValue=h;var p=1<<16,f=1<<24,m=p*p,g=m*m,y=g/2,A=i(f),x=i(0);r.ZERO=x;var b=i(0,!0);r.UZERO=b;var v=i(1);r.ONE=v;var I=i(1,!0);r.UONE=I;var w=i(-1);r.NEG_ONE=w;var S=u(4294967295|0,2147483647|0,!1);r.MAX_VALUE=S;var E=u(4294967295|0,4294967295|0,!0);r.MAX_UNSIGNED_VALUE=E;var D=u(0,2147483648|0,!1);r.MIN_VALUE=D;var $=r.prototype;$.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(N){if(N=N||10,N<2||36<N)throw RangeError("radix");if(this.isZero())return"0";if(this.isNegative())if(this.eq(D)){var M=l(N),B=this.div(M),q=B.mul(M).sub(this);return B.toString(N)+q.toInt().toString(N)}else return"-"+this.neg().toString(N);for(var X=l(c(N,6),this.unsigned),J=this,ee="";;){var ae=J.div(X),se=J.sub(ae.mul(X)).toInt()>>>0,oe=se.toString(N);if(J=ae,J.isZero())return oe+ee;for(;oe.length<6;)oe="0"+oe;ee=""+oe+ee}},$.getHighBits=function(){return this.high},$.getHighBitsUnsigned=function(){return this.high>>>0},$.getLowBits=function(){return this.low},$.getLowBitsUnsigned=function(){return this.low>>>0},$.getNumBitsAbs=function(){if(this.isNegative())return this.eq(D)?64:this.neg().getNumBitsAbs();for(var N=this.high!=0?this.high:this.low,M=31;M>0&&(N&1<<M)==0;M--);return this.high!=0?M+33:M+1},$.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},$.equals=function(N){return s(N)||(N=h(N)),this.unsigned!==N.unsigned&&this.high>>>31==1&&N.high>>>31==1?!1:this.high===N.high&&this.low===N.low},$.eq=$.equals,$.notEquals=function(N){return!this.eq(N)},$.neq=$.notEquals,$.ne=$.notEquals,$.lessThan=function(N){return this.comp(N)<0},$.lt=$.lessThan,$.lessThanOrEqual=function(N){return this.comp(N)<=0},$.lte=$.lessThanOrEqual,$.le=$.lessThanOrEqual,$.greaterThan=function(N){return this.comp(N)>0},$.gt=$.greaterThan,$.greaterThanOrEqual=function(N){return this.comp(N)>=0},$.gte=$.greaterThanOrEqual,$.ge=$.greaterThanOrEqual,$.compare=function(N){if(s(N)||(N=h(N)),this.eq(N))return 0;var M=this.isNegative(),B=N.isNegative();return M&&!B?-1:!M&&B?1:this.unsigned?N.high>>>0>this.high>>>0||N.high===this.high&&N.low>>>0>this.low>>>0?-1:1:this.sub(N).isNegative()?-1:1},$.comp=$.compare,$.negate=function(){return!this.unsigned&&this.eq(D)?D:this.not().add(v)},$.neg=$.negate,$.add=function(N){s(N)||(N=h(N));var M=this.high>>>16,B=this.high&65535,q=this.low>>>16,X=this.low&65535,J=N.high>>>16,ee=N.high&65535,ae=N.low>>>16,se=N.low&65535,oe=0,ne=0,ce=0,he=0;return he+=X+se,ce+=he>>>16,he&=65535,ce+=q+ae,ne+=ce>>>16,ce&=65535,ne+=B+ee,oe+=ne>>>16,ne&=65535,oe+=M+J,oe&=65535,u(ce<<16|he,oe<<16|ne,this.unsigned)},$.subtract=function(N){return s(N)||(N=h(N)),this.add(N.neg())},$.sub=$.subtract,$.multiply=function(N){if(this.isZero())return x;if(s(N)||(N=h(N)),n){var M=n.mul(this.low,this.high,N.low,N.high);return u(M,n.get_high(),this.unsigned)}if(N.isZero())return x;if(this.eq(D))return N.isOdd()?D:x;if(N.eq(D))return this.isOdd()?D:x;if(this.isNegative())return N.isNegative()?this.neg().mul(N.neg()):this.neg().mul(N).neg();if(N.isNegative())return this.mul(N.neg()).neg();if(this.lt(A)&&N.lt(A))return l(this.toNumber()*N.toNumber(),this.unsigned);var B=this.high>>>16,q=this.high&65535,X=this.low>>>16,J=this.low&65535,ee=N.high>>>16,ae=N.high&65535,se=N.low>>>16,oe=N.low&65535,ne=0,ce=0,he=0,me=0;return me+=J*oe,he+=me>>>16,me&=65535,he+=X*oe,ce+=he>>>16,he&=65535,he+=J*se,ce+=he>>>16,he&=65535,ce+=q*oe,ne+=ce>>>16,ce&=65535,ce+=X*se,ne+=ce>>>16,ce&=65535,ce+=J*ae,ne+=ce>>>16,ce&=65535,ne+=B*oe+q*se+X*ae+J*ee,ne&=65535,u(he<<16|me,ne<<16|ce,this.unsigned)},$.mul=$.multiply,$.divide=function(N){if(s(N)||(N=h(N)),N.isZero())throw Error("division by zero");if(n){if(!this.unsigned&&this.high===-2147483648&&N.low===-1&&N.high===-1)return this;var M=(this.unsigned?n.div_u:n.div_s)(this.low,this.high,N.low,N.high);return u(M,n.get_high(),this.unsigned)}if(this.isZero())return this.unsigned?b:x;var B,q,X;if(this.unsigned){if(N.unsigned||(N=N.toUnsigned()),N.gt(this))return b;if(N.gt(this.shru(1)))return I;X=b}else{if(this.eq(D)){if(N.eq(v)||N.eq(w))return D;if(N.eq(D))return v;var J=this.shr(1);return B=J.div(N).shl(1),B.eq(x)?N.isNegative()?v:w:(q=this.sub(N.mul(B)),X=B.add(q.div(N)),X)}else if(N.eq(D))return this.unsigned?b:x;if(this.isNegative())return N.isNegative()?this.neg().div(N.neg()):this.neg().div(N).neg();if(N.isNegative())return this.div(N.neg()).neg();X=x}for(q=this;q.gte(N);){B=Math.max(1,Math.floor(q.toNumber()/N.toNumber()));for(var ee=Math.ceil(Math.log(B)/Math.LN2),ae=ee<=48?1:c(2,ee-48),se=l(B),oe=se.mul(N);oe.isNegative()||oe.gt(q);)B-=ae,se=l(B,this.unsigned),oe=se.mul(N);se.isZero()&&(se=v),X=X.add(se),q=q.sub(oe)}return X},$.div=$.divide,$.modulo=function(N){if(s(N)||(N=h(N)),n){var M=(this.unsigned?n.rem_u:n.rem_s)(this.low,this.high,N.low,N.high);return u(M,n.get_high(),this.unsigned)}return this.sub(this.div(N).mul(N))},$.mod=$.modulo,$.rem=$.modulo,$.not=function(){return u(~this.low,~this.high,this.unsigned)},$.and=function(N){return s(N)||(N=h(N)),u(this.low&N.low,this.high&N.high,this.unsigned)},$.or=function(N){return s(N)||(N=h(N)),u(this.low|N.low,this.high|N.high,this.unsigned)},$.xor=function(N){return s(N)||(N=h(N)),u(this.low^N.low,this.high^N.high,this.unsigned)},$.shiftLeft=function(N){return s(N)&&(N=N.toInt()),(N&=63)===0?this:N<32?u(this.low<<N,this.high<<N|this.low>>>32-N,this.unsigned):u(0,this.low<<N-32,this.unsigned)},$.shl=$.shiftLeft,$.shiftRight=function(N){return s(N)&&(N=N.toInt()),(N&=63)===0?this:N<32?u(this.low>>>N|this.high<<32-N,this.high>>N,this.unsigned):u(this.high>>N-32,this.high>=0?0:-1,this.unsigned)},$.shr=$.shiftRight,$.shiftRightUnsigned=function(N){if(s(N)&&(N=N.toInt()),N&=63,N===0)return this;var M=this.high;if(N<32){var B=this.low;return u(B>>>N|M<<32-N,M>>>N,this.unsigned)}else return N===32?u(M,0,this.unsigned):u(M>>>N-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)},$.toBytes=function(N){return N?this.toBytesLE():this.toBytesBE()},$.toBytesLE=function(){var N=this.high,M=this.low;return[M&255,M>>>8&255,M>>>16&255,M>>>24,N&255,N>>>8&255,N>>>16&255,N>>>24]},$.toBytesBE=function(){var N=this.high,M=this.low;return[N>>>24,N>>>16&255,N>>>8&255,N&255,M>>>24,M>>>16&255,M>>>8&255,M&255]},r.fromBytes=function(N,M,B){return B?r.fromBytesLE(N,M):r.fromBytesBE(N,M)},r.fromBytesLE=function(N,M){return new r(N[0]|N[1]<<8|N[2]<<16|N[3]<<24,N[4]|N[5]<<8|N[6]<<16|N[7]<<24,M)},r.fromBytesBE=function(N,M){return new r(N[4]<<24|N[5]<<16|N[6]<<8|N[7],N[0]<<24|N[1]<<16|N[2]<<8|N[3],M)}}}),T3=Ft({"(disabled):node_modules/.pnpm/node-fetch@2.6.1/node_modules/node-fetch/browser.js"(){}}),e_=Ft({"node_modules/.pnpm/seedrandom@2.4.3/node_modules/seedrandom/lib/alea.js"(e,t){(function(n,r,s){function a(u){var c=this,d=l();c.next=function(){var h=2091639*c.s0+c.c*23283064365386963e-26;return c.s0=c.s1,c.s1=c.s2,c.s2=h-(c.c=h|0)},c.c=1,c.s0=d(" "),c.s1=d(" "),c.s2=d(" "),c.s0-=d(u),c.s0<0&&(c.s0+=1),c.s1-=d(u),c.s1<0&&(c.s1+=1),c.s2-=d(u),c.s2<0&&(c.s2+=1),d=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 d=new a(u),h=c&&c.state,p=d.next;return p.int32=function(){return d.next()*4294967296|0},p.double=function(){return p()+(p()*2097152|0)*11102230246251565e-32},p.quick=p,h&&(typeof h=="object"&&o(h,d),p.state=function(){return o(d,{})}),p}function l(){var u=4022871197,c=function(d){d=d.toString();for(var h=0;h<d.length;h++){u+=d.charCodeAt(h);var p=.02519603282416938*u;u=p>>>0,p-=u,p*=u,u=p>>>0,p-=u,u+=p*4294967296}return(u>>>0)*23283064365386963e-26};return c}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.alea=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),t_=Ft({"node_modules/.pnpm/seedrandom@2.4.3/node_modules/seedrandom/lib/xor128.js"(e,t){(function(n,r,s){function a(l){var u=this,c="";u.x=0,u.y=0,u.z=0,u.w=0,u.next=function(){var h=u.x^u.x<<11;return u.x=u.y,u.y=u.z,u.z=u.w,u.w^=u.w>>>19^h^h>>>8},l===(l|0)?u.x=l:c+=l;for(var d=0;d<c.length+64;d++)u.x^=c.charCodeAt(d)|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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(typeof d=="object"&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.xor128=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),n_=Ft({"node_modules/.pnpm/seedrandom@2.4.3/node_modules/seedrandom/lib/xorwow.js"(e,t){(function(n,r,s){function a(l){var u=this,c="";u.next=function(){var h=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^(h^h<<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 d=0;d<c.length+64;d++)u.x^=c.charCodeAt(d)|0,d==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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(typeof d=="object"&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.xorwow=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),r_=Ft({"node_modules/.pnpm/seedrandom@2.4.3/node_modules/seedrandom/lib/xorshift7.js"(e,t){(function(n,r,s){function a(l){var u=this;u.next=function(){var d=u.x,h=u.i,p,f,m;return p=d[h],p^=p>>>7,f=p^p<<24,p=d[h+1&7],f^=p^p>>>10,p=d[h+3&7],f^=p^p>>>3,p=d[h+4&7],f^=p^p<<7,p=d[h+7&7],p=p^p<<13,f^=p^p<<9,d[h]=f,u.i=h+1&7,f};function c(d,h){var p,f,m=[];if(h===(h|0))f=m[0]=h;else for(h=""+h,p=0;p<h.length;++p)m[p&7]=m[p&7]<<15^h.charCodeAt(p)+m[p+1&7]<<13;for(;m.length<8;)m.push(0);for(p=0;p<8&&m[p]===0;++p);for(p==8?f=m[7]=-1:f=m[p],d.x=m,d.i=0,p=256;p>0;--p)d.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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(d.x&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.xorshift7=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),s_=Ft({"node_modules/.pnpm/seedrandom@2.4.3/node_modules/seedrandom/lib/xor4096.js"(e,t){(function(n,r,s){function a(l){var u=this;u.next=function(){var d=u.w,h=u.X,p=u.i,f,m;return u.w=d=d+1640531527|0,m=h[p+34&127],f=h[p=p+1&127],m^=m<<13,f^=f<<17,m^=m>>>15,f^=f>>>12,m=h[p]=m^f,u.i=p,m+(d^d>>>16)|0};function c(d,h){var p,f,m,g,y,A=[],x=128;for(h===(h|0)?(f=h,h=null):(h=h+"\0",f=0,x=Math.max(x,h.length)),m=0,g=-32;g<x;++g)h&&(f^=h.charCodeAt((g+32)%h.length)),g===0&&(y=f),f^=f<<10,f^=f>>>15,f^=f<<4,f^=f>>>13,g>=0&&(y=y+1640531527|0,p=A[g&127]^=f+y,m=p==0?m+1:0);for(m>=128&&(A[(h&&h.length||0)&127]=-1),m=127,g=4*128;g>0;--g)f=A[m+34&127],p=A[m=m+1&127],f^=f<<13,p^=p<<17,f^=f>>>15,p^=p>>>12,A[m]=f^p;d.w=y,d.X=A,d.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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(d.X&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.xor4096=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),a_=Ft({"node_modules/.pnpm/seedrandom@2.4.3/node_modules/seedrandom/lib/tychei.js"(e,t){(function(n,r,s){function a(l){var u=this,c="";u.next=function(){var h=u.b,p=u.c,f=u.d,m=u.a;return h=h<<25^h>>>7^p,p=p-f|0,f=f<<24^f>>>8^m,m=m-h|0,u.b=h=h<<20^h>>>12^p,u.c=p=p-f|0,u.d=f<<16^p>>>16^m,u.a=m-h|0},u.a=0,u.b=0,u.c=2654435769|0,u.d=1367130551,l===Math.floor(l)?(u.a=l/4294967296|0,u.b=l|0):c+=l;for(var d=0;d<c.length+20;d++)u.b^=c.charCodeAt(d)|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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(typeof d=="object"&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.tychei=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),N3=Ft({"(disabled):crypto"(){}}),o_=Ft({"node_modules/.pnpm/seedrandom@2.4.3/node_modules/seedrandom/seedrandom.js"(e,t){(function(n,r){var s=this,a=256,o=6,i=52,l="random",u=r.pow(a,o),c=r.pow(2,i),d=c*2,h=a-1,p;function f(v,I,w){var S=[];I=I==!0?{entropy:!0}:I||{};var E=A(y(I.entropy?[v,b(n)]:v==null?x():v,3),S),D=new m(S),$=function(){for(var R=D.g(o),N=u,M=0;R<c;)R=(R+M)*a,N*=a,M=D.g(1);for(;R>=d;)R/=2,N/=2,M>>>=1;return(R+M)/N};return $.int32=function(){return D.g(4)|0},$.quick=function(){return D.g(4)/4294967296},$.double=$,A(b(D.S),n),(I.pass||w||function(R,N,M,B){return B&&(B.S&&g(B,D),R.state=function(){return g(D,{})}),M?(r[l]=R,N):R})($,E,"global"in I?I.global:this==r,I.state)}r["seed"+l]=f;function m(v){var I,w=v.length,S=this,E=0,D=S.i=S.j=0,$=S.S=[];for(w||(v=[w++]);E<a;)$[E]=E++;for(E=0;E<a;E++)$[E]=$[D=h&D+v[E%w]+(I=$[E])],$[D]=I;(S.g=function(R){for(var N,M=0,B=S.i,q=S.j,X=S.S;R--;)N=X[B=h&B+1],M=M*a+X[h&(X[B]=X[q=h&q+N])+(X[q]=N)];return S.i=B,S.j=q,M})(a)}function g(v,I){return I.i=v.i,I.j=v.j,I.S=v.S.slice(),I}function y(v,I){var w=[],S=typeof v,E;if(I&&S=="object")for(E in v)try{w.push(y(v[E],I-1))}catch(D){}return w.length?w:S=="string"?v:v+"\0"}function A(v,I){for(var w=v+"",S,E=0;E<w.length;)I[h&E]=h&(S^=I[h&E]*19)+w.charCodeAt(E++);return b(I)}function x(){try{var v;return p&&(v=p.randomBytes)?v=v(a):(v=new Uint8Array(a),(s.crypto||s.msCrypto).getRandomValues(v)),b(v)}catch(S){var I=s.navigator,w=I&&I.plugins;return[+new Date,s,w,s.screen,b(n)]}}function b(v){return String.fromCharCode.apply(0,v)}if(A(r.random(),n),typeof t=="object"&&t.exports){t.exports=f;try{p=N3()}catch(v){}}else typeof define=="function"&&define.amd&&define(function(){return f})})([],Math)}}),Zg=Ft({"node_modules/.pnpm/seedrandom@2.4.3/node_modules/seedrandom/index.js"(e,t){var n=e_(),r=t_(),s=n_(),a=r_(),o=s_(),i=a_(),l=o_();l.alea=n,l.xor128=r,l.xorwow=s,l.xorshift7=a,l.xor4096=o,l.tychei=i,t.exports=l}}),i_=Ft({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/alea.js"(e,t){(function(n,r,s){function a(u){var c=this,d=l();c.next=function(){var h=2091639*c.s0+c.c*23283064365386963e-26;return c.s0=c.s1,c.s1=c.s2,c.s2=h-(c.c=h|0)},c.c=1,c.s0=d(" "),c.s1=d(" "),c.s2=d(" "),c.s0-=d(u),c.s0<0&&(c.s0+=1),c.s1-=d(u),c.s1<0&&(c.s1+=1),c.s2-=d(u),c.s2<0&&(c.s2+=1),d=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 d=new a(u),h=c&&c.state,p=d.next;return p.int32=function(){return d.next()*4294967296|0},p.double=function(){return p()+(p()*2097152|0)*11102230246251565e-32},p.quick=p,h&&(typeof h=="object"&&o(h,d),p.state=function(){return o(d,{})}),p}function l(){var u=4022871197,c=function(d){d=String(d);for(var h=0;h<d.length;h++){u+=d.charCodeAt(h);var p=.02519603282416938*u;u=p>>>0,p-=u,p*=u,u=p>>>0,p-=u,u+=p*4294967296}return(u>>>0)*23283064365386963e-26};return c}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.alea=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),l_=Ft({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/xor128.js"(e,t){(function(n,r,s){function a(l){var u=this,c="";u.x=0,u.y=0,u.z=0,u.w=0,u.next=function(){var h=u.x^u.x<<11;return u.x=u.y,u.y=u.z,u.z=u.w,u.w^=u.w>>>19^h^h>>>8},l===(l|0)?u.x=l:c+=l;for(var d=0;d<c.length+64;d++)u.x^=c.charCodeAt(d)|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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(typeof d=="object"&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.xor128=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),u_=Ft({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/xorwow.js"(e,t){(function(n,r,s){function a(l){var u=this,c="";u.next=function(){var h=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^(h^h<<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 d=0;d<c.length+64;d++)u.x^=c.charCodeAt(d)|0,d==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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(typeof d=="object"&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.xorwow=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),c_=Ft({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/xorshift7.js"(e,t){(function(n,r,s){function a(l){var u=this;u.next=function(){var d=u.x,h=u.i,p,f,m;return p=d[h],p^=p>>>7,f=p^p<<24,p=d[h+1&7],f^=p^p>>>10,p=d[h+3&7],f^=p^p>>>3,p=d[h+4&7],f^=p^p<<7,p=d[h+7&7],p=p^p<<13,f^=p^p<<9,d[h]=f,u.i=h+1&7,f};function c(d,h){var p,f,m=[];if(h===(h|0))f=m[0]=h;else for(h=""+h,p=0;p<h.length;++p)m[p&7]=m[p&7]<<15^h.charCodeAt(p)+m[p+1&7]<<13;for(;m.length<8;)m.push(0);for(p=0;p<8&&m[p]===0;++p);for(p==8?f=m[7]=-1:f=m[p],d.x=m,d.i=0,p=256;p>0;--p)d.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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(d.x&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.xorshift7=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),d_=Ft({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/xor4096.js"(e,t){(function(n,r,s){function a(l){var u=this;u.next=function(){var d=u.w,h=u.X,p=u.i,f,m;return u.w=d=d+1640531527|0,m=h[p+34&127],f=h[p=p+1&127],m^=m<<13,f^=f<<17,m^=m>>>15,f^=f>>>12,m=h[p]=m^f,u.i=p,m+(d^d>>>16)|0};function c(d,h){var p,f,m,g,y,A=[],x=128;for(h===(h|0)?(f=h,h=null):(h=h+"\0",f=0,x=Math.max(x,h.length)),m=0,g=-32;g<x;++g)h&&(f^=h.charCodeAt((g+32)%h.length)),g===0&&(y=f),f^=f<<10,f^=f>>>15,f^=f<<4,f^=f>>>13,g>=0&&(y=y+1640531527|0,p=A[g&127]^=f+y,m=p==0?m+1:0);for(m>=128&&(A[(h&&h.length||0)&127]=-1),m=127,g=4*128;g>0;--g)f=A[m+34&127],p=A[m=m+1&127],f^=f<<13,p^=p<<17,f^=f>>>15,p^=p>>>12,A[m]=f^p;d.w=y,d.X=A,d.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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(d.X&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.xor4096=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),h_=Ft({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/lib/tychei.js"(e,t){(function(n,r,s){function a(l){var u=this,c="";u.next=function(){var h=u.b,p=u.c,f=u.d,m=u.a;return h=h<<25^h>>>7^p,p=p-f|0,f=f<<24^f>>>8^m,m=m-h|0,u.b=h=h<<20^h>>>12^p,u.c=p=p-f|0,u.d=f<<16^p>>>16^m,u.a=m-h|0},u.a=0,u.b=0,u.c=2654435769|0,u.d=1367130551,l===Math.floor(l)?(u.a=l/4294967296|0,u.b=l|0):c+=l;for(var d=0;d<c.length+20;d++)u.b^=c.charCodeAt(d)|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),d=u&&u.state,h=function(){return(c.next()>>>0)/4294967296};return h.double=function(){do var p=c.next()>>>11,f=(c.next()>>>0)/4294967296,m=(p+f)/(1<<21);while(m===0);return m},h.int32=c.next,h.quick=h,d&&(typeof d=="object"&&o(d,c),h.state=function(){return o(c,{})}),h}r&&r.exports?r.exports=i:s&&s.amd?s(function(){return i}):this.tychei=i})(e,typeof t=="object"&&t,typeof define=="function"&&define)}}),p_=Ft({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/seedrandom.js"(e,t){(function(n,r,s){var a=256,o=6,i=52,l="random",u=s.pow(a,o),c=s.pow(2,i),d=c*2,h=a-1,p;function f(v,I,w){var S=[];I=I==!0?{entropy:!0}:I||{};var E=A(y(I.entropy?[v,b(r)]:v==null?x():v,3),S),D=new m(S),$=function(){for(var R=D.g(o),N=u,M=0;R<c;)R=(R+M)*a,N*=a,M=D.g(1);for(;R>=d;)R/=2,N/=2,M>>>=1;return(R+M)/N};return $.int32=function(){return D.g(4)|0},$.quick=function(){return D.g(4)/4294967296},$.double=$,A(b(D.S),r),(I.pass||w||function(R,N,M,B){return B&&(B.S&&g(B,D),R.state=function(){return g(D,{})}),M?(s[l]=R,N):R})($,E,"global"in I?I.global:this==s,I.state)}function m(v){var I,w=v.length,S=this,E=0,D=S.i=S.j=0,$=S.S=[];for(w||(v=[w++]);E<a;)$[E]=E++;for(E=0;E<a;E++)$[E]=$[D=h&D+v[E%w]+(I=$[E])],$[D]=I;(S.g=function(R){for(var N,M=0,B=S.i,q=S.j,X=S.S;R--;)N=X[B=h&B+1],M=M*a+X[h&(X[B]=X[q=h&q+N])+(X[q]=N)];return S.i=B,S.j=q,M})(a)}function g(v,I){return I.i=v.i,I.j=v.j,I.S=v.S.slice(),I}function y(v,I){var w=[],S=typeof v,E;if(I&&S=="object")for(E in v)try{w.push(y(v[E],I-1))}catch(D){}return w.length?w:S=="string"?v:v+"\0"}function A(v,I){for(var w=v+"",S,E=0;E<w.length;)I[h&E]=h&(S^=I[h&E]*19)+w.charCodeAt(E++);return b(I)}function x(){try{var v;return p&&(v=p.randomBytes)?v=v(a):(v=new Uint8Array(a),(n.crypto||n.msCrypto).getRandomValues(v)),b(v)}catch(S){var I=n.navigator,w=I&&I.plugins;return[+new Date,n,w,n.screen,b(r)]}}function b(v){return String.fromCharCode.apply(0,v)}if(A(s.random(),r),typeof t=="object"&&t.exports){t.exports=f;try{p=N3()}catch(v){}}else typeof define=="function"&&define.amd?define(function(){return f}):s["seed"+l]=f})(typeof self!="undefined"?self:e,[],Math)}}),C3=Ft({"node_modules/.pnpm/seedrandom@3.0.5/node_modules/seedrandom/index.js"(e,t){var n=i_(),r=l_(),s=u_(),a=c_(),o=d_(),i=h_(),l=p_();l.alea=n,l.xor128=r,l.xorwow=s,l.xorshift7=a,l.xor4096=o,l.tychei=i,t.exports=l}}),f_=Ft({"(disabled):node_modules/.pnpm/string_decoder@1.1.1/node_modules/string_decoder/lib/string_decoder.js"(){}}),Gu=Ft({"(disabled):path"(){}}),m_=Ft({"(disabled):worker_threads"(){}}),g_=Ft({"(disabled):perf_hooks"(){}}),y_=Ft({"node_modules/.pnpm/@tensorflow+tfjs-backend-wasm@3.8.0_@tensorflow+tfjs-core@3.8.0/node_modules/@tensorflow/tfjs-backend-wasm/wasm-out/tfjs-backend-wasm-threaded-simd.js"(e,t){var n=function(){var r=typeof document!="undefined"&&document.currentScript?document.currentScript.src:void 0;return typeof __filename!="undefined"&&(r=r||__filename),function(s){s=s||{};function a(){return ne.buffer!=Ye&&vn(ne.buffer),Gn}function o(){return ne.buffer!=Ye&&vn(ne.buffer),Bt}function i(){return ne.buffer!=Ye&&vn(ne.buffer),zr}function l(){return ne.buffer!=Ye&&vn(ne.buffer),Rn}function u(){return ne.buffer!=Ye&&vn(ne.buffer),xr}var c=typeof s!="undefined"?s:{},d,h;c.ready=new Promise(function(C,F){d=C,h=F});var p={},f;for(f in c)c.hasOwnProperty(f)&&(p[f]=c[f]);var m=[],g="./this.program",y=function(C,F){throw F},A=!1,x=!1,b=!1,v=!1;A=typeof window=="object",x=typeof importScripts=="function",b=typeof process=="object"&&typeof process.versions=="object"&&typeof process.versions.node=="string",v=!A&&!b&&!x;var I=c.ENVIRONMENT_IS_PTHREAD||!1;I&&(Ye=c.buffer);var w="";function S(C){return c.locateFile?c.locateFile(C,w):w+C}var E,D,$,R,N,M;if(b){x?w=Gu().dirname(w)+"/":w=__dirname+"/",E=function(F,W){return N||(N=co("fs")),M||(M=Gu()),F=M.normalize(F),N.readFileSync(F,W?null:"utf8")},$=function(F){var W=E(F,!0);return W.buffer||(W=new Uint8Array(W)),be(W.buffer),W},process.argv.length>1&&(g=process.argv[1].replace(/\\/g,"/")),m=process.argv.slice(2),process.on("uncaughtException",function(C){if(!(C instanceof Hu))throw C}),process.on("unhandledRejection",js),y=function(C){process.exit(C)},c.inspect=function(){return"[Emscripten Module object]"};var B;try{B=m_()}catch(C){throw console.error('The "worker_threads" module is not supported in this node.js build - perhaps a newer version is needed?'),C}global.Worker=B.Worker}else v?(typeof read!="undefined"&&(E=function(F){return read(F)}),$=function(F){var W;return typeof readbuffer=="function"?new Uint8Array(readbuffer(F)):(W=read(F,"binary"),be(typeof W=="object"),W)},typeof scriptArgs!="undefined"?m=scriptArgs:typeof arguments!="undefined"&&(m=arguments),typeof quit=="function"&&(y=function(C){quit(C)}),typeof print!="undefined"&&(typeof console=="undefined"&&(console={}),console.log=print,console.warn=console.error=typeof printErr!="undefined"?printErr:print)):(A||x)&&(x?w=self.location.href:typeof document!="undefined"&&document.currentScript&&(w=document.currentScript.src),typeof r!="undefined"&&r&&(w=r),w.indexOf("blob:")!==0?w=w.substr(0,w.lastIndexOf("/")+1):w="",b?(E=function(F,W){return N||(N=co("fs")),M||(M=Gu()),F=M.normalize(F),N.readFileSync(F,W?null:"utf8")},$=function(F){var W=E(F,!0);return W.buffer||(W=new Uint8Array(W)),be(W.buffer),W}):(E=function(C){var F=new XMLHttpRequest;return F.open("GET",C,!1),F.send(null),F.responseText},x&&($=function(C){var F=new XMLHttpRequest;return F.open("GET",C,!1),F.responseType="arraybuffer",F.send(null),new Uint8Array(F.response)}),D=function(C,F,W){var Q=new XMLHttpRequest;Q.open("GET",C,!0),Q.responseType="arraybuffer",Q.onload=function(){if(Q.status==200||Q.status==0&&Q.response){F(Q.response);return}W()},Q.onerror=W,Q.send(null)}),R=function(C){document.title=C});b&&typeof performance=="undefined"&&(global.performance=g_().performance);var q=c.print||console.log.bind(console),X=c.printErr||console.warn.bind(console);for(f in p)p.hasOwnProperty(f)&&(c[f]=p[f]);p=null,c.arguments&&(m=c.arguments),c.thisProgram&&(g=c.thisProgram),c.quit&&(y=c.quit);var J=Atomics.load,ee=Atomics.store,ae=Atomics.compareExchange,se;c.wasmBinary&&(se=c.wasmBinary);var oe=c.noExitRuntime||!0;typeof WebAssembly!="object"&&js("no native wasm support detected");var ne,ce,he=!1,me;function be(C,F){C||js("Assertion failed: "+F)}function Ee(C){var F=c["_"+C];return be(F,"Cannot call unknown function "+C+", make sure it is exported"),F}function $e(C,F,W,Q,Ae){var ge={string:function(_n){var Fi=0;if(_n!=null&&_n!==0){var x3=(_n.length<<2)+1;Fi=Ri(x3),pt(_n,Fi,x3)}return Fi},array:function(_n){var Fi=Ri(_n.length);return dt(_n,Fi),Fi}};function ye(_n){return F==="string"?Be(_n):F==="boolean"?Boolean(_n):_n}var Se=Ee(C),mt=[],hn=0;if(Q)for(var nn=0;nn<Q.length;nn++){var Ia=ge[W[nn]];Ia?(hn===0&&(hn=Uu()),mt[nn]=Ia(Q[nn])):mt[nn]=Q[nn]}var Di=Se.apply(null,mt);return Di=ye(Di),hn!==0&&$i(hn),Di}function Pe(C,F,W,Q){W=W||[];var Ae=W.every(function(ye){return ye==="number"}),ge=F!=="string";return ge&&Ae&&!Q?Ee(C):function(){return $e(C,F,W,arguments,Q)}}function je(C,F,W){for(var Q=F+W,Ae="";!(F>=Q);){var ge=C[F++];if(!ge)return Ae;if(!(ge&128)){Ae+=String.fromCharCode(ge);continue}var ye=C[F++]&63;if((ge&224)==192){Ae+=String.fromCharCode((ge&31)<<6|ye);continue}var Se=C[F++]&63;if((ge&240)==224?ge=(ge&15)<<12|ye<<6|Se:ge=(ge&7)<<18|ye<<12|Se<<6|C[F++]&63,ge<65536)Ae+=String.fromCharCode(ge);else{var mt=ge-65536;Ae+=String.fromCharCode(55296|mt>>10,56320|mt&1023)}}return Ae}function Be(C,F){return C?je(o(),C,F):""}function bt(C,F,W,Q){if(!(Q>0))return 0;for(var Ae=W,ge=W+Q-1,ye=0;ye<C.length;++ye){var Se=C.charCodeAt(ye);if(Se>=55296&&Se<=57343){var mt=C.charCodeAt(++ye);Se=65536+((Se&1023)<<10)|mt&1023}if(Se<=127){if(W>=ge)break;F[W++]=Se}else if(Se<=2047){if(W+1>=ge)break;F[W++]=192|Se>>6,F[W++]=128|Se&63}else if(Se<=65535){if(W+2>=ge)break;F[W++]=224|Se>>12,F[W++]=128|Se>>6&63,F[W++]=128|Se&63}else{if(W+3>=ge)break;F[W++]=240|Se>>18,F[W++]=128|Se>>12&63,F[W++]=128|Se>>6&63,F[W++]=128|Se&63}}return F[W]=0,W-Ae}function pt(C,F,W){return bt(C,o(),F,W)}function ft(C){for(var F=0,W=0;W<C.length;++W){var Q=C.charCodeAt(W);Q>=55296&&Q<=57343&&(Q=65536+((Q&1023)<<10)|C.charCodeAt(++W)&1023),Q<=127?++F:Q<=2047?F+=2:Q<=65535?F+=3:F+=4}return F}function dt(C,F){a().set(C,F)}function xt(C,F){return C%F>0&&(C+=F-C%F),C}var Ye,Gn,Bt,or,bn,zr,Rn,Ar,xr;function vn(C){Ye=C,c.HEAP8=Gn=new Int8Array(C),c.HEAP16=or=new Int16Array(C),c.HEAP32=zr=new Int32Array(C),c.HEAPU8=Bt=new Uint8Array(C),c.HEAPU16=bn=new Uint16Array(C),c.HEAPU32=Rn=new Uint32Array(C),c.HEAPF32=Ar=new Float32Array(C),c.HEAPF64=xr=new Float64Array(C)}var br=c.INITIAL_MEMORY||16777216;if(I)ne=c.wasmMemory,Ye=c.buffer;else if(c.wasmMemory)ne=c.wasmMemory;else if(ne=new WebAssembly.Memory({initial:br/65536,maximum:2147483648/65536,shared:!0}),!(ne.buffer instanceof SharedArrayBuffer))throw X("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"),b&&console.log("(on node you may need: --experimental-wasm-threads --experimental-wasm-bulk-memory and also use a recent version)"),Error("bad memory");ne&&(Ye=ne.buffer),br=Ye.byteLength,vn(Ye);var vr,ir=[],bs=[],Hs=[],xa=[],Ii=[],vs=!1,Eh=!1;I||bs.push({func:function(){Hh()}});function k0(){if(!I){if(c.preRun)for(typeof c.preRun=="function"&&(c.preRun=[c.preRun]);c.preRun.length;)Rh(c.preRun.shift());Ti(ir)}}function Fu(){vs=!0,!I&&Ti(bs)}function I0(){I||Ti(Hs)}function $h(){I||(Eh=!0)}function jn(){if(!I){if(c.postRun)for(typeof c.postRun=="function"&&(c.postRun=[c.postRun]);c.postRun.length;)S0(c.postRun.shift());Ti(Ii)}}function Rh(C){ir.unshift(C)}function S0(C){Ii.unshift(C)}var Gs=0,ba=null,io=null;function T0(C){be(!I,"addRunDependency cannot be used in a pthread worker"),Gs++,c.monitorRunDependencies&&c.monitorRunDependencies(Gs)}function N0(C){if(Gs--,c.monitorRunDependencies&&c.monitorRunDependencies(Gs),Gs==0&&(ba!==null&&(clearInterval(ba),ba=null),io)){var F=io;io=null,F()}}c.preloadedImages={},c.preloadedAudios={};function js(C){c.onAbort&&c.onAbort(C),I&&console.error("Pthread aborting at "+new Error().stack),C+="",X(C),he=!0,me=1,C="abort("+C+"). Build with -s ASSERTIONS=1 for more info.";var F=new WebAssembly.RuntimeError(C);throw h(F),F}function _h(C,F){return String.prototype.startsWith?C.startsWith(F):C.indexOf(F)===0}var Si="data:application/octet-stream;base64,";function Dh(C){return _h(C,Si)}var C0="file://";function Fh(C){return _h(C,C0)}var qn="tfjs-backend-wasm-threaded-simd.wasm";Dh(qn)||(qn=S(qn));function Mh(C){try{if(C==qn&&se)return new Uint8Array(se);if($)return $(C);throw"both async and sync fetching of the wasm failed"}catch(F){js(F)}}function E0(){if(!se&&(A||x)){if(typeof fetch=="function"&&!Fh(qn))return fetch(qn,{credentials:"same-origin"}).then(function(C){if(!C.ok)throw"failed to load wasm binary file at '"+qn+"'";return C.arrayBuffer()}).catch(function(){return Mh(qn)});if(D)return new Promise(function(C,F){D(qn,function(W){C(new Uint8Array(W))},F)})}return Promise.resolve().then(function(){return Mh(qn)})}function $0(){var C={a:bg};function F(ye,Se){var mt=ye.exports;if(c.asm=mt,vr=c.asm.F,ce=Se,!I){var hn=Ce.unusedWorkers.length;Ce.unusedWorkers.forEach(function(nn){Ce.loadWasmModuleToWorker(nn,function(){--hn||N0("wasm-instantiate")})})}}I||T0("wasm-instantiate");function W(ye){F(ye.instance,ye.module)}function Q(ye){return E0().then(function(Se){return WebAssembly.instantiate(Se,C)}).then(ye,function(Se){X("failed to asynchronously prepare wasm: "+Se),js(Se)})}function Ae(){return!se&&typeof WebAssembly.instantiateStreaming=="function"&&!Dh(qn)&&!Fh(qn)&&typeof fetch=="function"?fetch(qn,{credentials:"same-origin"}).then(function(ye){var Se=WebAssembly.instantiateStreaming(ye,C);return Se.then(W,function(mt){return X("wasm streaming compile failed: "+mt),X("falling back to ArrayBuffer instantiation"),Q(W)})}):Q(W)}if(c.instantiateWasm)try{var ge=c.instantiateWasm(C,F);return ge}catch(ye){return X("Module.instantiateWasm callback failed with error: "+ye),!1}return Ae().catch(h),{}}var R0={9832:function(){throw"Canceled!"},9850:function(C,F){setTimeout(function(){p3(C,F)},0)}};function Oh(){Ce.initRuntime()}function Ti(C){for(;C.length>0;){var F=C.shift();if(typeof F=="function"){F(c);continue}var W=F.func;typeof W=="number"?F.arg===void 0?vr.get(W)():vr.get(W)(F.arg):W(F.arg===void 0?null:F.arg)}}function Mu(C,F){if(C<=0||C>a().length||C&!0||F<0)return-28;if(F==0)return 0;F>=2147483647&&(F=1/0);var W=Atomics.load(i(),_i>>2),Q=0;if(W==C){var Ae=Atomics.compareExchange(i(),_i>>2,W,0);if(Ae==W&&(--F,Q=1,F<=0))return 1}var ge=Atomics.notify(i(),C>>2,F);if(ge>=0)return ge+Q;throw"Atomics.notify returned an unexpected value "+ge}c._emscripten_futex_wake=Mu;function _0(C){if(I)throw"Internal Error! killThread() can only ever be called from main application thread!";if(!C)throw"Internal Error! Null pthread_ptr in killThread!";i()[C+12>>2]=0;var F=Ce.pthreads[C];F.worker.terminate(),Ce.freeThreadData(F),Ce.runningWorkers.splice(Ce.runningWorkers.indexOf(F.worker),1),F.worker.pthread=void 0}function D0(C){if(I)throw"Internal Error! cancelThread() can only ever be called from main application thread!";if(!C)throw"Internal Error! Null pthread_ptr in cancelThread!";var F=Ce.pthreads[C];F.worker.postMessage({cmd:"cancel"})}function F0(C){if(I)throw"Internal Error! cleanupThread() can only ever be called from main application thread!";if(!C)throw"Internal Error! Null pthread_ptr in cleanupThread!";var F=Ce.pthreads[C];if(F){i()[C+12>>2]=0;var W=F.worker;Ce.returnWorkerToPool(W)}}var Ce={unusedWorkers:[],runningWorkers:[],initMainThreadBlock:function(){for(var C=Math.min(4,Math.max(1,(navigator.hardwareConcurrency||1)/2)),F=0;F<C;++F)Ce.allocateUnusedWorker()},initRuntime:function(){for(var C=uo(228),F=0;F<228/4;++F)l()[C/4+F]=0;i()[C+12>>2]=C;var W=C+152;i()[W>>2]=W;for(var Q=uo(512),F=0;F<128;++F)l()[Q/4+F]=0;Atomics.store(l(),C+100>>2,Q),Atomics.store(l(),C+40>>2,C),jg(C,!x,1),h3(C)},initWorker:function(){},pthreads:{},threadExitHandlers:[],setThreadStatus:function(){},runExitHandlers:function(){for(;Ce.threadExitHandlers.length>0;)Ce.threadExitHandlers.pop()();I&&Ei()&&d3()},runExitHandlersAndDeinitThread:function(C,F){Atomics.store(l(),C+56>>2,1),Atomics.store(l(),C+60>>2,0),Ce.runExitHandlers(),Atomics.store(l(),C+4>>2,F),Atomics.store(l(),C+0>>2,1),Mu(C+0,2147483647),jg(0,0,0)},threadExit:function(C){var F=Ei();F&&(Ce.runExitHandlersAndDeinitThread(F,C),I&&postMessage({cmd:"exit"}))},threadCancel:function(){Ce.runExitHandlersAndDeinitThread(Ei(),-1),postMessage({cmd:"cancelDone"})},terminateAllThreads:function(){for(var C in Ce.pthreads){var F=Ce.pthreads[C];F&&F.worker&&Ce.returnWorkerToPool(F.worker)}Ce.pthreads={};for(var W=0;W<Ce.unusedWorkers.length;++W){var Q=Ce.unusedWorkers[W];Q.terminate()}Ce.unusedWorkers=[];for(var W=0;W<Ce.runningWorkers.length;++W){var Q=Ce.runningWorkers[W],F=Q.pthread;Ce.freeThreadData(F),Q.terminate()}Ce.runningWorkers=[]},freeThreadData:function(C){if(!!C){if(C.threadInfoStruct){var F=i()[C.threadInfoStruct+100>>2];i()[C.threadInfoStruct+100>>2]=0,Vu(F),Vu(C.threadInfoStruct)}C.threadInfoStruct=0,C.allocatedOwnStack&&C.stackBase&&Vu(C.stackBase),C.stackBase=0,C.worker&&(C.worker.pthread=null)}},returnWorkerToPool:function(C){Ce.runWithoutMainThreadQueuedCalls(function(){delete Ce.pthreads[C.pthread.threadInfoStruct],Ce.unusedWorkers.push(C),Ce.runningWorkers.splice(Ce.runningWorkers.indexOf(C),1),Ce.freeThreadData(C.pthread),C.pthread=void 0})},runWithoutMainThreadQueuedCalls:function(C){i()[A3>>2]=0;try{C()}finally{i()[A3>>2]=1}},receiveObjectTransfer:function(C){},loadWasmModuleToWorker:function(C,F){C.onmessage=function(W){var Q=W.data,Ae=Q.cmd;if(C.pthread&&(Ce.currentProxiedOperationCallerThread=C.pthread.threadInfoStruct),Q.targetThread&&Q.targetThread!=Ei()){var ge=Ce.pthreads[Q.targetThread];ge?ge.worker.postMessage(W.data,Q.transferList):console.error('Internal error! Worker sent a message "'+Ae+'" to target pthread '+Q.targetThread+", but that thread no longer exists!"),Ce.currentProxiedOperationCallerThread=void 0;return}if(Ae==="processQueuedMainThreadWork")Hg();else if(Ae==="spawnThread")Vh(W.data);else if(Ae==="cleanupThread")F0(Q.thread);else if(Ae==="killThread")_0(Q.thread);else if(Ae==="cancelThread")D0(Q.thread);else if(Ae==="loaded")C.loaded=!0,F&&F(C),C.runPthread&&(C.runPthread(),delete C.runPthread);else if(Ae==="print")q("Thread "+Q.threadId+": "+Q.text);else if(Ae==="printErr")X("Thread "+Q.threadId+": "+Q.text);else if(Ae==="alert")alert("Thread "+Q.threadId+": "+Q.text);else if(Ae==="exit"){var ye=C.pthread&&Atomics.load(l(),C.pthread.threadInfoStruct+64>>2);ye&&Ce.returnWorkerToPool(C)}else if(Ae==="exitProcess")try{jR(Q.returnCode)}catch(Se){if(Se instanceof Hu)return;throw Se}else Ae==="cancelDone"?Ce.returnWorkerToPool(C):Ae==="objectTransfer"?Ce.receiveObjectTransfer(W.data):W.data.target==="setimmediate"?C.postMessage(W.data):X("worker sent an unknown command "+Ae);Ce.currentProxiedOperationCallerThread=void 0},C.onerror=function(W){X("pthread sent an error! "+W.filename+":"+W.lineno+": "+W.message)},b&&(C.on("message",function(W){C.onmessage({data:W})}),C.on("error",function(W){C.onerror(W)}),C.on("exit",function(W){})),C.postMessage({cmd:"load",urlOrBlob:c.mainScriptUrlOrBlob||r,wasmMemory:ne,wasmModule:ce})},allocateUnusedWorker:function(){var C=S("tfjs-backend-wasm-threaded-simd.worker.js");Ce.unusedWorkers.push(new Worker(C))},getNewWorker:function(){return Ce.unusedWorkers.length==0&&(Ce.allocateUnusedWorker(),Ce.loadWasmModuleToWorker(Ce.unusedWorkers[0])),Ce.unusedWorkers.length>0?Ce.unusedWorkers.pop():null},busySpinWait:function(C){for(var F=performance.now()+C;performance.now()<F;);}};function M0(C,F){g3(C,F),$i(C)}c.establishStackSpace=M0;function O0(){return oe}c.getNoExitRuntime=O0;function P0(C,F){return vr.get(C)(F)}c.invokeEntryPoint=P0;function z0(C,F,W,Q){js("Assertion failed: "+Be(C)+", at: "+[F?Be(F):"unknown filename",W,Q?Be(Q):"unknown function"])}function L0(C,F){var W=_main(C,F)}var lo;b?lo=function(){var C=process.hrtime();return C[0]*1e3+C[1]/1e6}:I?lo=function(){return performance.now()-c.__performance_now_clock_drift}:typeof dateNow!="undefined"?lo=dateNow:lo=function(){return performance.now()};function B0(C){return i()[u3()>>2]=C,C}function W0(C,F){if(I)return va(1,1,C,F)}function V0(C,F){if(C==F)postMessage({cmd:"processQueuedMainThreadWork"});else if(I)postMessage({targetThread:C,cmd:"processThreadQueue"});else{var W=Ce.pthreads[C],Q=W&&W.worker;if(!Q)return;Q.postMessage({cmd:"processThreadQueue"})}return 1}function U0(){js()}function H0(C,F,W){var Q=X0(F,W);return R0[C].apply(null,Q)}function G0(C,F){}function j0(C,F,W){if(C<=0||C>a().length||C&!0)return-28;if(A){if(Atomics.load(i(),C>>2)!=F)return-6;for(var Ae=performance.now(),ge=Ae+W,ye=Atomics.exchange(i(),_i>>2,C);;){if(Ae=performance.now(),Ae>ge)return ye=Atomics.exchange(i(),_i>>2,0),-73;if(ye=Atomics.exchange(i(),_i>>2,0),ye==0)break;if(Hg(),Atomics.load(i(),C>>2)!=F)return-6;ye=Atomics.exchange(i(),_i>>2,C)}return 0}else{var Q=Atomics.wait(i(),C>>2,F,W);if(Q==="timed-out")return-73;if(Q==="not-equal")return-6;if(Q==="ok")return 0;throw"Atomics.wait returned an unexpected value "+Q}}function q0(C,F,W){o().copyWithin(C,F,F+W)}function K0(){return b?co("os").cpus().length:navigator.hardwareConcurrency}function va(C,F){for(var W=arguments.length-2,Q=Uu(),Ae=W,ge=Ri(Ae*8),ye=ge>>3,Se=0;Se<W;Se++){var mt=arguments[2+Se];u()[ye+Se]=mt}var hn=m3(C,Ae,ge,F);return $i(Q),hn}var Ou=[],Pu=[];function X0(C,F){Pu.length=0;var W;for(F>>=2;W=o()[C++];){var Q=W<105;Q&&F&1&&F++,Pu.push(Q?u()[F++>>1]:i()[F]),++F}return Pu}function Z0(C,F,W){Ou.length=F;for(var Q=W>>3,Ae=0;Ae<F;Ae++)Ou[Ae]=u()[Q+Ae];var ge=C<0,ye=ge?R0[-C-1]:xg[C];return ye.apply(null,Ou)}function Y0(){return o().length}function J0(C){try{return ne.grow(C-Ye.byteLength+65535>>>16),vn(ne.buffer),1}catch(F){}}function Q0(C){var F=Y0();if(C<=F)return!1;var W=2147483648;if(C>W)return!1;for(var Q=1;Q<=4;Q*=2){var Ae=F*(1+.2/Q);Ae=Math.min(Ae,C+100663296);var ge=Math.min(W,xt(Math.max(C,Ae),65536)),ye=J0(ge);if(ye)return!0}return!1}var Ke={inEventHandler:0,removeAllEventListeners:function(){for(var C=Ke.eventHandlers.length-1;C>=0;--C)Ke._removeHandler(C);Ke.eventHandlers=[],Ke.deferredCalls=[]},registerRemoveEventListeners:function(){Ke.removeEventListenersRegistered||(xa.push(Ke.removeAllEventListeners),Ke.removeEventListenersRegistered=!0)},deferredCalls:[],deferCall:function(C,F,W){function Q(ye,Se){if(ye.length!=Se.length)return!1;for(var mt in ye)if(ye[mt]!=Se[mt])return!1;return!0}for(var Ae in Ke.deferredCalls){var ge=Ke.deferredCalls[Ae];if(ge.targetFunction==C&&Q(ge.argsList,W))return}Ke.deferredCalls.push({targetFunction:C,precedence:F,argsList:W}),Ke.deferredCalls.sort(function(ye,Se){return ye.precedence<Se.precedence})},removeDeferredCalls:function(C){for(var F=0;F<Ke.deferredCalls.length;++F)Ke.deferredCalls[F].targetFunction==C&&(Ke.deferredCalls.splice(F,1),--F)},canPerformEventHandlerRequests:function(){return Ke.inEventHandler&&Ke.currentEventHandler.allowsDeferredCalls},runDeferredCalls:function(){if(!!Ke.canPerformEventHandlerRequests())for(var C=0;C<Ke.deferredCalls.length;++C){var F=Ke.deferredCalls[C];Ke.deferredCalls.splice(C,1),--C,F.targetFunction.apply(null,F.argsList)}},eventHandlers:[],removeAllHandlersOnTarget:function(C,F){for(var W=0;W<Ke.eventHandlers.length;++W)Ke.eventHandlers[W].target==C&&(!F||F==Ke.eventHandlers[W].eventTypeString)&&Ke._removeHandler(W--)},_removeHandler:function(C){var F=Ke.eventHandlers[C];F.target.removeEventListener(F.eventTypeString,F.eventListenerFunc,F.useCapture),Ke.eventHandlers.splice(C,1)},registerOrRemoveHandler:function(C){var F=function(Ae){++Ke.inEventHandler,Ke.currentEventHandler=C,Ke.runDeferredCalls(),C.handlerFunc(Ae),Ke.runDeferredCalls(),--Ke.inEventHandler};if(C.callbackfunc)C.eventListenerFunc=F,C.target.addEventListener(C.eventTypeString,F,C.useCapture),Ke.eventHandlers.push(C),Ke.registerRemoveEventListeners();else for(var W=0;W<Ke.eventHandlers.length;++W)Ke.eventHandlers[W].target==C.target&&Ke.eventHandlers[W].eventTypeString==C.eventTypeString&&Ke._removeHandler(W--)},queueEventHandlerOnThread_iiii:function(C,F,W,Q,Ae){var ge=Uu(),ye=Ri(12);i()[ye>>2]=W,i()[ye+4>>2]=Q,i()[ye+8>>2]=Ae,Gg(0,C,637534208,F,Q,ye),$i(ge)},getTargetThreadForEventCallback:function(C){switch(C){case 1:return 0;case 2:return Ce.currentProxiedOperationCallerThread;default:return C}},getNodeNameForTarget:function(C){return C?C==window?"#window":C==screen?"#screen":C&&C.nodeName?C.nodeName:"":""},fullscreenEnabled:function(){return document.fullscreenEnabled||document.webkitFullscreenEnabled}};function eg(C){var F=ft(C)+1,W=uo(F);return pt(C,W,F),W}function tg(C,F,W,Q){var Ae=Uu(),ge=Ri(12),ye=0;F&&(ye=eg(F)),i()[ge>>2]=ye,i()[ge+4>>2]=W,i()[ge+8>>2]=Q,Gg(0,C,657457152,0,ye,ge),$i(Ae)}function ng(C,F,W,Q){F=F?Be(F):"",tg(C,F,W,Q)}function rg(C){return C>2?Be(C):C}var sg=[0,typeof document!="undefined"?document:0,typeof window!="undefined"?window:0];function ag(C){C=rg(C);var F=sg[C]||(typeof document!="undefined"?document.querySelector(C):void 0);return F}function zu(C){return ag(C)}function Ph(C,F,W){var Q=zu(C);if(!Q)return-4;if(Q.canvasSharedPtr&&(i()[Q.canvasSharedPtr>>2]=F,i()[Q.canvasSharedPtr+4>>2]=W),Q.offscreenCanvas||!Q.controlTransferredOffscreen){Q.offscreenCanvas&&(Q=Q.offscreenCanvas);var Ae=!1;if(Q.GLctxObject&&Q.GLctxObject.GLctx){var ge=Q.GLctxObject.GLctx.getParameter(2978);Ae=ge[0]===0&&ge[1]===0&&ge[2]===Q.width&&ge[3]===Q.height}Q.width=F,Q.height=W,Ae&&Q.GLctxObject.GLctx.viewport(0,0,F,W)}else if(Q.canvasSharedPtr){var ye=i()[Q.canvasSharedPtr+8>>2];return ng(ye,C,F,W),1}else return-4;return 0}function zh(C,F,W){return I?va(2,1,C,F,W):Ph(C,F,W)}function og(C,F,W){var Q=zu(C);return Q?Ph(C,F,W):zh(C,F,W)}function ig(C){}function lg(C,F){}function ug(C){var F=C.getExtension("ANGLE_instanced_arrays");if(F)return C.vertexAttribDivisor=function(W,Q){F.vertexAttribDivisorANGLE(W,Q)},C.drawArraysInstanced=function(W,Q,Ae,ge){F.drawArraysInstancedANGLE(W,Q,Ae,ge)},C.drawElementsInstanced=function(W,Q,Ae,ge,ye){F.drawElementsInstancedANGLE(W,Q,Ae,ge,ye)},1}function cg(C){var F=C.getExtension("OES_vertex_array_object");if(F)return C.createVertexArray=function(){return F.createVertexArrayOES()},C.deleteVertexArray=function(W){F.deleteVertexArrayOES(W)},C.bindVertexArray=function(W){F.bindVertexArrayOES(W)},C.isVertexArray=function(W){return F.isVertexArrayOES(W)},1}function dg(C){var F=C.getExtension("WEBGL_draw_buffers");if(F)return C.drawBuffers=function(W,Q){F.drawBuffersWEBGL(W,Q)},1}function hg(C){return!!(C.multiDrawWebgl=C.getExtension("WEBGL_multi_draw"))}var ht={counter:1,buffers:[],programs:[],framebuffers:[],renderbuffers:[],textures:[],uniforms:[],shaders:[],vaos:[],contexts:{},offscreenCanvases:{},timerQueriesEXT:[],programInfos:{},stringCache:{},unpackAlignment:4,recordError:function(F){ht.lastError||(ht.lastError=F)},getNewId:function(C){for(var F=ht.counter++,W=C.length;W<F;W++)C[W]=null;return F},getSource:function(C,F,W,Q){for(var Ae="",ge=0;ge<F;++ge){var ye=Q?i()[Q+ge*4>>2]:-1;Ae+=Be(i()[W+ge*4>>2],ye<0?void 0:ye)}return Ae},createContext:function(C,F){var W=C.getContext("webgl",F);if(!W)return 0;var Q=ht.registerContext(W,F);return Q},registerContext:function(C,F){var W=uo(8);i()[W+4>>2]=Ei();var Q={handle:W,attributes:F,version:F.majorVersion,GLctx:C};return C.canvas&&(C.canvas.GLctxObject=Q),ht.contexts[W]=Q,(typeof F.enableExtensionsByDefault=="undefined"||F.enableExtensionsByDefault)&&ht.initExtensions(Q),W},makeContextCurrent:function(C){return ht.currentContext=ht.contexts[C],c.ctx=wa=ht.currentContext&&ht.currentContext.GLctx,!(C&&!wa)},getContext:function(C){return ht.contexts[C]},deleteContext:function(C){ht.currentContext===ht.contexts[C]&&(ht.currentContext=null),typeof Ke=="object"&&Ke.removeAllHandlersOnTarget(ht.contexts[C].GLctx.canvas),ht.contexts[C]&&ht.contexts[C].GLctx.canvas&&(ht.contexts[C].GLctx.canvas.GLctxObject=void 0),Vu(ht.contexts[C].handle),ht.contexts[C]=null},initExtensions:function(C){if(C||(C=ht.currentContext),!C.initExtensionsDone){C.initExtensionsDone=!0;var F=C.GLctx;ug(F),cg(F),dg(F),F.disjointTimerQueryExt=F.getExtension("EXT_disjoint_timer_query"),hg(F);var W=F.getSupportedExtensions()||[];W.forEach(function(Q){Q.indexOf("lose_context")<0&&Q.indexOf("debug")<0&&F.getExtension(Q)})}},populateUniformTable:function(C){for(var F=ht.programs[C],W=ht.programInfos[C]={uniforms:{},maxUniformLength:0,maxAttributeLength:-1,maxUniformBlockNameLength:-1},Q=W.uniforms,Ae=wa.getProgramParameter(F,35718),ge=0;ge<Ae;++ge){var ye=wa.getActiveUniform(F,ge),Se=ye.name;W.maxUniformLength=Math.max(W.maxUniformLength,Se.length+1),Se.slice(-1)=="]"&&(Se=Se.slice(0,Se.lastIndexOf("[")));var mt=wa.getUniformLocation(F,Se);if(mt){var hn=ht.getNewId(ht.uniforms);Q[Se]=[ye.size,hn],ht.uniforms[hn]=mt;for(var nn=1;nn<ye.size;++nn){var Ia=Se+"["+nn+"]";mt=wa.getUniformLocation(F,Ia),hn=ht.getNewId(ht.uniforms),ht.uniforms[hn]=mt}}}}},pg=["default","low-power","high-performance"];function fg(C,F){var W=F>>2,Q=i()[W+(24>>2)],Ae={alpha:!!i()[W+(0>>2)],depth:!!i()[W+(4>>2)],stencil:!!i()[W+(8>>2)],antialias:!!i()[W+(12>>2)],premultipliedAlpha:!!i()[W+(16>>2)],preserveDrawingBuffer:!!i()[W+(20>>2)],powerPreference:pg[Q],failIfMajorPerformanceCaveat:!!i()[W+(28>>2)],majorVersion:i()[W+(32>>2)],minorVersion:i()[W+(36>>2)],enableExtensionsByDefault:i()[W+(40>>2)],explicitSwapControl:i()[W+(44>>2)],proxyContextToMainThread:i()[W+(48>>2)],renderViaOffscreenBackBuffer:i()[W+(52>>2)]},ge=zu(C);if(!ge||Ae.explicitSwapControl)return 0;var ye=ht.createContext(ge,Ae);return ye}function mg(C,F){return fg(C,F)}var Ni={mappings:{},buffers:[null,[],[]],printChar:function(C,F){var W=Ni.buffers[C];F===0||F===10?((C===1?q:X)(je(W,0)),W.length=0):W.push(F)},varargs:void 0,get:function(){Ni.varargs+=4;var C=i()[Ni.varargs-4>>2];return C},getStr:function(C){var F=Be(C);return F},get64:function(C,F){return C}};function Lh(C){return I?va(3,1,C):0}function Bh(C,F,W,Q,Ae){if(I)return va(4,1,C,F,W,Q,Ae)}function Wh(C,F,W,Q){if(I)return va(5,1,C,F,W,Q);for(var Ae=0,ge=0;ge<W;ge++){for(var ye=i()[F+ge*8>>2],Se=i()[F+(ge*8+4)>>2],mt=0;mt<Se;mt++)Ni.printChar(C,o()[ye+mt]);Ae+=Se}return i()[Q>>2]=Ae,0}function gg(C){var F=Ce.threadExitHandlers.pop();C&&F()}function yg(C,F){Ce.threadExitHandlers.push(function(){vr.get(C)(F)})}function Vh(C){if(I)throw"Internal Error! spawnThread() can only ever be called from main application thread!";var F=Ce.getNewWorker();if(F.pthread!==void 0)throw"Internal error!";if(!C.pthread_ptr)throw"Internal error, no pthread ptr!";Ce.runningWorkers.push(F);for(var W=uo(128*4),Q=0;Q<128;++Q)i()[W+Q*4>>2]=0;var Ae=C.stackBase+C.stackSize,ge=Ce.pthreads[C.pthread_ptr]={worker:F,stackBase:C.stackBase,stackSize:C.stackSize,allocatedOwnStack:C.allocatedOwnStack,threadInfoStruct:C.pthread_ptr},ye=ge.threadInfoStruct>>2;Atomics.store(l(),ye+(64>>2),C.detached),Atomics.store(l(),ye+(100>>2),W),Atomics.store(l(),ye+(40>>2),ge.threadInfoStruct),Atomics.store(l(),ye+(80>>2),C.stackSize),Atomics.store(l(),ye+(76>>2),Ae),Atomics.store(l(),ye+(104>>2),C.stackSize),Atomics.store(l(),ye+(104+8>>2),Ae),Atomics.store(l(),ye+(104+12>>2),C.detached);var Se=c3(),mt=Se+40;Atomics.store(l(),ye+(172>>2),mt),F.pthread=ge;var hn={cmd:"run",start_routine:C.startRoutine,arg:C.arg,threadInfoStruct:C.pthread_ptr,stackBase:C.stackBase,stackSize:C.stackSize};F.runPthread=function(){hn.time=performance.now(),F.postMessage(hn,C.transferList)},F.loaded&&(F.runPthread(),delete F.runPthread)}function Ag(C,F,W,Q){if(typeof SharedArrayBuffer=="undefined")return X("Current environment does not support SharedArrayBuffer, pthreads are not available!"),6;if(!C)return X("pthread_create called with a null thread pointer!"),28;var Ae=[],ge=0;if(I&&(Ae.length===0||ge))return f3(687865856,C,F,W,Q);if(ge)return ge;var ye=0,Se=0,mt=0;F&&F!=-1?(ye=i()[F>>2],ye+=81920,Se=i()[F+8>>2],mt=i()[F+12>>2]!==0):ye=2097152;var hn=Se==0;hn?Se=y3(16,ye):(Se-=ye,be(Se>0));for(var nn=uo(228),Ia=0;Ia<228>>2;++Ia)l()[(nn>>2)+Ia]=0;i()[C>>2]=nn,i()[nn+12>>2]=nn;var Di=nn+152;i()[Di>>2]=Di;var _n={stackBase:Se,stackSize:ye,allocatedOwnStack:hn,detached:mt,startRoutine:W,pthread_ptr:nn,arg:Q,transferList:Ae};return I?(_n.cmd="spawnThread",postMessage(_n,Ae)):Vh(_n),0}function Uh(C){if(I)return va(6,1,C);switch(C){case 30:return 16384;case 85:var F=2147483648;return F/16384;case 132:case 133:case 12:case 137:case 138:case 15:case 235:case 16:case 17:case 18:case 19:case 20:case 149:case 13:case 10:case 236:case 153:case 9:case 21:case 22:case 159:case 154:case 14:case 77:case 78:case 139:case 82:case 68:case 67:case 164:case 11:case 29:case 47:case 48:case 95:case 52:case 51:case 46:return 200809;case 27:case 246:case 127:case 128:case 23:case 24:case 160:case 161:case 181:case 182:case 242:case 183:case 184:case 243:case 244:case 245:case 165:case 178:case 179:case 49:case 50:case 168:case 169:case 175:case 170:case 171:case 172:case 97:case 76:case 32:case 173:case 35:case 80:case 81:case 79:return-1;case 176:case 177:case 7:case 155:case 8:case 157:case 125:case 126:case 92:case 93:case 129:case 130:case 131:case 94:case 91:return 1;case 74:case 60:case 69:case 70:case 4:return 1024;case 31:case 42:case 72:return 32;case 87:case 26:case 33:return 2147483647;case 34:case 1:return 47839;case 38:case 36:return 99;case 43:case 37:return 2048;case 0:return 2097152;case 3:return 65536;case 28:return 32768;case 44:return 32767;case 75:return 16384;case 39:return 1e3;case 89:return 700;case 71:return 256;case 40:return 255;case 2:return 100;case 180:return 64;case 25:return 20;case 5:return 16;case 6:return 6;case 73:return 4;case 84:return typeof navigator=="object"&&navigator.hardwareConcurrency||1}return B0(28),-1}I||Ce.initMainThreadBlock();var wa,xg=[null,W0,zh,Lh,Bh,Wh,Uh],bg={e:z0,r:L0,x:V0,b:U0,y:H0,j:G0,c:j0,d:Mu,f:lo,p:q0,z:K0,u:Z0,q:Q0,v:og,i:ig,t:lg,w:mg,m:Lh,n:Bh,g:Wh,o:Oh,a:ne||c.wasmMemory,k:gg,l:yg,h:Ag,s:Uh},l3=$0(),Hh=c.___wasm_call_ctors=function(){return(Hh=c.___wasm_call_ctors=c.asm.A).apply(null,arguments)},vg=c._init=function(){return(vg=c._init=c.asm.B).apply(null,arguments)},wg=c._register_tensor=function(){return(wg=c._register_tensor=c.asm.C).apply(null,arguments)},kg=c._dispose_data=function(){return(kg=c._dispose_data=c.asm.D).apply(null,arguments)},Ig=c._dispose=function(){return(Ig=c._dispose=c.asm.E).apply(null,arguments)},Sg=c._Abs=function(){return(Sg=c._Abs=c.asm.G).apply(null,arguments)},Tg=c._Add=function(){return(Tg=c._Add=c.asm.H).apply(null,arguments)},Ng=c._AddN=function(){return(Ng=c._AddN=c.asm.I).apply(null,arguments)},Cg=c._All=function(){return(Cg=c._All=c.asm.J).apply(null,arguments)},Eg=c._Any=function(){return(Eg=c._Any=c.asm.K).apply(null,arguments)},$g=c._ArgMax=function(){return($g=c._ArgMax=c.asm.L).apply(null,arguments)},Rg=c._AvgPool=function(){return(Rg=c._AvgPool=c.asm.M).apply(null,arguments)},_g=c._BatchMatMul=function(){return(_g=c._BatchMatMul=c.asm.N).apply(null,arguments)},Dg=c._Ceil=function(){return(Dg=c._Ceil=c.asm.O).apply(null,arguments)},Fg=c._ClipByValue=function(){return(Fg=c._ClipByValue=c.asm.P).apply(null,arguments)},Mg=c._Conv2D=function(){return(Mg=c._Conv2D=c.asm.Q).apply(null,arguments)},Og=c._Conv2DBackpropInput=function(){return(Og=c._Conv2DBackpropInput=c.asm.R).apply(null,arguments)},Pg=c._Cos=function(){return(Pg=c._Cos=c.asm.S).apply(null,arguments)},zg=c._Cosh=function(){return(zg=c._Cosh=c.asm.T).apply(null,arguments)},Lg=c._CropAndResize=function(){return(Lg=c._CropAndResize=c.asm.U).apply(null,arguments)},Bg=c._Cumsum=function(){return(Bg=c._Cumsum=c.asm.V).apply(null,arguments)},Wg=c._DepthToSpace=function(){return(Wg=c._DepthToSpace=c.asm.W).apply(null,arguments)},Vg=c._DepthwiseConv2dNative=function(){return(Vg=c._DepthwiseConv2dNative=c.asm.X).apply(null,arguments)},Gh=c._Equal=function(){return(Gh=c._Equal=c.asm.Y).apply(null,arguments)},jh=c._Exp=function(){return(jh=c._Exp=c.asm.Z).apply(null,arguments)},qh=c._FlipLeftRight=function(){return(qh=c._FlipLeftRight=c.asm._).apply(null,arguments)},Lu=c._Floor=function(){return(Lu=c._Floor=c.asm.$).apply(null,arguments)},Ci=c._FloorDiv=function(){return(Ci=c._FloorDiv=c.asm.aa).apply(null,arguments)},Ug=c._FusedBatchNorm=function(){return(Ug=c._FusedBatchNorm=c.asm.ba).apply(null,arguments)},Bu=c._FusedConv2D=function(){return(Bu=c._FusedConv2D=c.asm.ca).apply(null,arguments)},te=c._FusedDepthwiseConv2D=function(){return(te=c._FusedDepthwiseConv2D=c.asm.da).apply(null,arguments)},ie=c._Gather=function(){return(ie=c._Gather=c.asm.ea).apply(null,arguments)},we=c._GatherNd=function(){return(we=c._GatherNd=c.asm.fa).apply(null,arguments)},lt=c._Greater=function(){return(lt=c._Greater=c.asm.ga).apply(null,arguments)},Ht=c._GreaterEqual=function(){return(Ht=c._GreaterEqual=c.asm.ha).apply(null,arguments)},Dt=c._LeakyRelu=function(){return(Dt=c._LeakyRelu=c.asm.ia).apply(null,arguments)},Qe=c._Less=function(){return(Qe=c._Less=c.asm.ja).apply(null,arguments)},et=c._LessEqual=function(){return(et=c._LessEqual=c.asm.ka).apply(null,arguments)},wn=c._Log=function(){return(wn=c._Log=c.asm.la).apply(null,arguments)},qs=c._LogicalAnd=function(){return(qs=c._LogicalAnd=c.asm.ma).apply(null,arguments)},Ks=c._Max=function(){return(Ks=c._Max=c.asm.na).apply(null,arguments)},Kh=c._MaxPool=function(){return(Kh=c._MaxPool=c.asm.oa).apply(null,arguments)},Wu=c._Maximum=function(){return(Wu=c._Maximum=c.asm.pa).apply(null,arguments)},lr=c._Mean=function(){return(lr=c._Mean=c.asm.qa).apply(null,arguments)},ka=c._Min=function(){return(ka=c._Min=c.asm.ra).apply(null,arguments)},Xh=c._Minimum=function(){return(Xh=c._Minimum=c.asm.sa).apply(null,arguments)},aR=c._MirrorPad=function(){return(aR=c._MirrorPad=c.asm.ta).apply(null,arguments)},oR=c._Multiply=function(){return(oR=c._Multiply=c.asm.ua).apply(null,arguments)},iR=c._Neg=function(){return(iR=c._Neg=c.asm.va).apply(null,arguments)},lR=c._NonMaxSuppressionV3=function(){return(lR=c._NonMaxSuppressionV3=c.asm.wa).apply(null,arguments)},uR=c._NonMaxSuppressionV4=function(){return(uR=c._NonMaxSuppressionV4=c.asm.xa).apply(null,arguments)},cR=c._NonMaxSuppressionV5=function(){return(cR=c._NonMaxSuppressionV5=c.asm.ya).apply(null,arguments)},dR=c._NotEqual=function(){return(dR=c._NotEqual=c.asm.za).apply(null,arguments)},hR=c._OneHot=function(){return(hR=c._OneHot=c.asm.Aa).apply(null,arguments)},pR=c._PadV2=function(){return(pR=c._PadV2=c.asm.Ba).apply(null,arguments)},fR=c._Pow=function(){return(fR=c._Pow=c.asm.Ca).apply(null,arguments)},mR=c._Prelu=function(){return(mR=c._Prelu=c.asm.Da).apply(null,arguments)},gR=c._Prod=function(){return(gR=c._Prod=c.asm.Ea).apply(null,arguments)},yR=c._RealDiv=function(){return(yR=c._RealDiv=c.asm.Fa).apply(null,arguments)},AR=c._Relu=function(){return(AR=c._Relu=c.asm.Ga).apply(null,arguments)},xR=c._Relu6=function(){return(xR=c._Relu6=c.asm.Ha).apply(null,arguments)},bR=c._ResizeBilinear=function(){return(bR=c._ResizeBilinear=c.asm.Ia).apply(null,arguments)},vR=c._Reverse=function(){return(vR=c._Reverse=c.asm.Ja).apply(null,arguments)},wR=c._RotateWithOffset=function(){return(wR=c._RotateWithOffset=c.asm.Ka).apply(null,arguments)},kR=c._Round=function(){return(kR=c._Round=c.asm.La).apply(null,arguments)},IR=c._Rsqrt=function(){return(IR=c._Rsqrt=c.asm.Ma).apply(null,arguments)},SR=c._ScatterNd=function(){return(SR=c._ScatterNd=c.asm.Na).apply(null,arguments)},TR=c._SelectV2=function(){return(TR=c._SelectV2=c.asm.Oa).apply(null,arguments)},NR=c._Sigmoid=function(){return(NR=c._Sigmoid=c.asm.Pa).apply(null,arguments)},CR=c._Sin=function(){return(CR=c._Sin=c.asm.Qa).apply(null,arguments)},ER=c._Softmax=function(){return(ER=c._Softmax=c.asm.Ra).apply(null,arguments)},$R=c._Sqrt=function(){return($R=c._Sqrt=c.asm.Sa).apply(null,arguments)},RR=c._Square=function(){return(RR=c._Square=c.asm.Ta).apply(null,arguments)},_R=c._SquaredDifference=function(){return(_R=c._SquaredDifference=c.asm.Ua).apply(null,arguments)},DR=c._Step=function(){return(DR=c._Step=c.asm.Va).apply(null,arguments)},FR=c._StridedSlice=function(){return(FR=c._StridedSlice=c.asm.Wa).apply(null,arguments)},MR=c._Sub=function(){return(MR=c._Sub=c.asm.Xa).apply(null,arguments)},OR=c._Sum=function(){return(OR=c._Sum=c.asm.Ya).apply(null,arguments)},PR=c._Tan=function(){return(PR=c._Tan=c.asm.Za).apply(null,arguments)},zR=c._Tanh=function(){return(zR=c._Tanh=c.asm._a).apply(null,arguments)},LR=c._Tile=function(){return(LR=c._Tile=c.asm.$a).apply(null,arguments)},BR=c._TopK=function(){return(BR=c._TopK=c.asm.ab).apply(null,arguments)},WR=c._Transform=function(){return(WR=c._Transform=c.asm.bb).apply(null,arguments)},VR=c._Transpose=function(){return(VR=c._Transpose=c.asm.cb).apply(null,arguments)},UR=c.__FusedMatMul=function(){return(UR=c.__FusedMatMul=c.asm.db).apply(null,arguments)},uo=c._malloc=function(){return(uo=c._malloc=c.asm.eb).apply(null,arguments)},Vu=c._free=function(){return(Vu=c._free=c.asm.fb).apply(null,arguments)},u3=c.___errno_location=function(){return(u3=c.___errno_location=c.asm.gb).apply(null,arguments)},c3=c._emscripten_get_global_libc=function(){return(c3=c._emscripten_get_global_libc=c.asm.hb).apply(null,arguments)},Ei=c._pthread_self=function(){return(Ei=c._pthread_self=c.asm.ib).apply(null,arguments)},d3=c.___pthread_tsd_run_dtors=function(){return(d3=c.___pthread_tsd_run_dtors=c.asm.jb).apply(null,arguments)},Hg=c._emscripten_main_thread_process_queued_calls=function(){return(Hg=c._emscripten_main_thread_process_queued_calls=c.asm.kb).apply(null,arguments)},HR=c._emscripten_current_thread_process_queued_calls=function(){return(HR=c._emscripten_current_thread_process_queued_calls=c.asm.lb).apply(null,arguments)},h3=c._emscripten_register_main_browser_thread_id=function(){return(h3=c._emscripten_register_main_browser_thread_id=c.asm.mb).apply(null,arguments)},p3=c.__emscripten_do_dispatch_to_thread=function(){return(p3=c.__emscripten_do_dispatch_to_thread=c.asm.nb).apply(null,arguments)},f3=c._emscripten_sync_run_in_main_thread_4=function(){return(f3=c._emscripten_sync_run_in_main_thread_4=c.asm.ob).apply(null,arguments)},m3=c._emscripten_run_in_main_runtime_thread_js=function(){return(m3=c._emscripten_run_in_main_runtime_thread_js=c.asm.pb).apply(null,arguments)},Gg=c.__emscripten_call_on_thread=function(){return(Gg=c.__emscripten_call_on_thread=c.asm.qb).apply(null,arguments)},GR=c._emscripten_tls_init=function(){return(GR=c._emscripten_tls_init=c.asm.rb).apply(null,arguments)},jg=c.__emscripten_thread_init=function(){return(jg=c.__emscripten_thread_init=c.asm.sb).apply(null,arguments)},Uu=c.stackSave=function(){return(Uu=c.stackSave=c.asm.tb).apply(null,arguments)},$i=c.stackRestore=function(){return($i=c.stackRestore=c.asm.ub).apply(null,arguments)},Ri=c.stackAlloc=function(){return(Ri=c.stackAlloc=c.asm.vb).apply(null,arguments)},g3=c._emscripten_stack_set_limits=function(){return(g3=c._emscripten_stack_set_limits=c.asm.wb).apply(null,arguments)},y3=c._memalign=function(){return(y3=c._memalign=c.asm.xb).apply(null,arguments)},A3=c.__emscripten_allow_main_runtime_queued_calls=9824,_i=c.__emscripten_main_thread_futex=11448;c.cwrap=Pe,c.PThread=Ce,c.PThread=Ce,c.wasmMemory=ne,c.ExitStatus=Hu;var Zh;function Hu(C){this.name="ExitStatus",this.message="Program terminated with exit("+C+")",this.status=C}io=function C(){Zh||qg(),Zh||(io=C)};function qg(C){if(C=C||m,Gs>0)return;if(I){d(c),Fu(),postMessage({cmd:"loaded"});return}if(k0(),Gs>0)return;function F(){Zh||(Zh=!0,c.calledRun=!0,!he&&(Fu(),I0(),d(c),c.onRuntimeInitialized&&c.onRuntimeInitialized(),jn()))}c.setStatus?(c.setStatus("Running..."),setTimeout(function(){setTimeout(function(){c.setStatus("")},1),F()},1)):F()}c.run=qg;function jR(C,F){if(!(F&&oe&&C===0)){if(!F&&I)throw postMessage({cmd:"exitProcess",returnCode:C}),new Hu(C);oe||(Ce.terminateAllThreads(),me=C,$h(),c.onExit&&c.onExit(C),he=!0),y(C,new Hu(C))}}if(c.preInit)for(typeof c.preInit=="function"&&(c.preInit=[c.preInit]);c.preInit.length>0;)c.preInit.pop()();return I&&(oe=!1,Ce.initWorker()),qg(),s.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)}}),A_=Ft({"node_modules/.pnpm/@tensorflow+tfjs-backend-wasm@3.8.0_@tensorflow+tfjs-core@3.8.0/node_modules/@tensorflow/tfjs-backend-wasm/wasm-out/tfjs-backend-wasm.js"(e,t){var n=function(){var r=typeof document!="undefined"&&document.currentScript?document.currentScript.src:void 0;return typeof __filename!="undefined"&&(r=r||__filename),function(s){s=s||{};var a=typeof s!="undefined"?s:{},o,i;a.ready=new Promise(function(te,ie){o=te,i=ie});var l={},u;for(u in a)a.hasOwnProperty(u)&&(l[u]=a[u]);var c=[],d="./this.program",h=function(te,ie){throw ie},p=!1,f=!1,m=!1,g=!1;p=typeof window=="object",f=typeof importScripts=="function",m=typeof process=="object"&&typeof process.versions=="object"&&typeof process.versions.node=="string",g=!p&&!m&&!f;var y="";function A(te){return a.locateFile?a.locateFile(te,y):y+te}var x,b,v,I,w,S;m?(f?y=Gu().dirname(y)+"/":y=__dirname+"/",x=function(ie,we){return w||(w=co("fs")),S||(S=Gu()),ie=S.normalize(ie),w.readFileSync(ie,we?null:"utf8")},v=function(ie){var we=x(ie,!0);return we.buffer||(we=new Uint8Array(we)),q(we.buffer),we},process.argv.length>1&&(d=process.argv[1].replace(/\\/g,"/")),c=process.argv.slice(2),process.on("uncaughtException",function(te){if(!(te instanceof Ug))throw te}),process.on("unhandledRejection",vs),h=function(te){process.exit(te)},a.inspect=function(){return"[Emscripten Module object]"}):g?(typeof read!="undefined"&&(x=function(ie){return read(ie)}),v=function(ie){var we;return typeof readbuffer=="function"?new Uint8Array(readbuffer(ie)):(we=read(ie,"binary"),q(typeof we=="object"),we)},typeof scriptArgs!="undefined"?c=scriptArgs:typeof arguments!="undefined"&&(c=arguments),typeof quit=="function"&&(h=function(te){quit(te)}),typeof print!="undefined"&&(typeof console=="undefined"&&(console={}),console.log=print,console.warn=console.error=typeof printErr!="undefined"?printErr:print)):(p||f)&&(f?y=self.location.href:typeof document!="undefined"&&document.currentScript&&(y=document.currentScript.src),r&&(y=r),y.indexOf("blob:")!==0?y=y.substr(0,y.lastIndexOf("/")+1):y="",x=function(te){var ie=new XMLHttpRequest;return ie.open("GET",te,!1),ie.send(null),ie.responseText},f&&(v=function(te){var ie=new XMLHttpRequest;return ie.open("GET",te,!1),ie.responseType="arraybuffer",ie.send(null),new Uint8Array(ie.response)}),b=function(te,ie,we){var lt=new XMLHttpRequest;lt.open("GET",te,!0),lt.responseType="arraybuffer",lt.onload=function(){if(lt.status==200||lt.status==0&&lt.response){ie(lt.response);return}we()},lt.onerror=we,lt.send(null)},I=function(te){document.title=te});var E=a.print||console.log.bind(console),D=a.printErr||console.warn.bind(console);for(u in l)l.hasOwnProperty(u)&&(a[u]=l[u]);l=null,a.arguments&&(c=a.arguments),a.thisProgram&&(d=a.thisProgram),a.quit&&(h=a.quit);var $;a.wasmBinary&&($=a.wasmBinary);var R=a.noExitRuntime||!0;typeof WebAssembly!="object"&&vs("no native wasm support detected");var N,M=!1,B;function q(te,ie){te||vs("Assertion failed: "+ie)}function X(te){var ie=a["_"+te];return q(ie,"Cannot call unknown function "+te+", make sure it is exported"),ie}function J(te,ie,we,lt,Ht){var Dt={string:function(lr){var ka=0;if(lr!=null&&lr!==0){var Xh=(lr.length<<2)+1;ka=Lu(Xh),ce(lr,ka,Xh)}return ka},array:function(lr){var ka=Lu(lr.length);return he(lr,ka),ka}};function Qe(lr){return ie==="string"?oe(lr):ie==="boolean"?Boolean(lr):lr}var et=X(te),wn=[],qs=0;if(lt)for(var Ks=0;Ks<lt.length;Ks++){var Kh=Dt[we[Ks]];Kh?(qs===0&&(qs=jh()),wn[Ks]=Kh(lt[Ks])):wn[Ks]=lt[Ks]}var Wu=et.apply(null,wn);return Wu=Qe(Wu),qs!==0&&qh(qs),Wu}function ee(te,ie,we,lt){we=we||[];var Ht=we.every(function(Qe){return Qe==="number"}),Dt=ie!=="string";return Dt&&Ht&&!lt?X(te):function(){return J(te,ie,we,arguments,lt)}}var ae=typeof TextDecoder!="undefined"?new TextDecoder("utf8"):void 0;function se(te,ie,we){for(var lt=ie+we,Ht=ie;te[Ht]&&!(Ht>=lt);)++Ht;if(Ht-ie>16&&te.subarray&&ae)return ae.decode(te.subarray(ie,Ht));for(var Dt="";ie<Ht;){var Qe=te[ie++];if(!(Qe&128)){Dt+=String.fromCharCode(Qe);continue}var et=te[ie++]&63;if((Qe&224)==192){Dt+=String.fromCharCode((Qe&31)<<6|et);continue}var wn=te[ie++]&63;if((Qe&240)==224?Qe=(Qe&15)<<12|et<<6|wn:Qe=(Qe&7)<<18|et<<12|wn<<6|te[ie++]&63,Qe<65536)Dt+=String.fromCharCode(Qe);else{var qs=Qe-65536;Dt+=String.fromCharCode(55296|qs>>10,56320|qs&1023)}}return Dt}function oe(te,ie){return te?se($e,te,ie):""}function ne(te,ie,we,lt){if(!(lt>0))return 0;for(var Ht=we,Dt=we+lt-1,Qe=0;Qe<te.length;++Qe){var et=te.charCodeAt(Qe);if(et>=55296&&et<=57343){var wn=te.charCodeAt(++Qe);et=65536+((et&1023)<<10)|wn&1023}if(et<=127){if(we>=Dt)break;ie[we++]=et}else if(et<=2047){if(we+1>=Dt)break;ie[we++]=192|et>>6,ie[we++]=128|et&63}else if(et<=65535){if(we+2>=Dt)break;ie[we++]=224|et>>12,ie[we++]=128|et>>6&63,ie[we++]=128|et&63}else{if(we+3>=Dt)break;ie[we++]=240|et>>18,ie[we++]=128|et>>12&63,ie[we++]=128|et>>6&63,ie[we++]=128|et&63}}return ie[we]=0,we-Ht}function ce(te,ie,we){return ne(te,$e,ie,we)}function he(te,ie){Ee.set(te,ie)}function me(te,ie){return te%ie>0&&(te+=ie-te%ie),te}var be,Ee,$e,Pe,je,Be,bt,pt,ft;function dt(te){be=te,a.HEAP8=Ee=new Int8Array(te),a.HEAP16=Pe=new Int16Array(te),a.HEAP32=Be=new Int32Array(te),a.HEAPU8=$e=new Uint8Array(te),a.HEAPU16=je=new Uint16Array(te),a.HEAPU32=bt=new Uint32Array(te),a.HEAPF32=pt=new Float32Array(te),a.HEAPF64=ft=new Float64Array(te)}var xt=a.INITIAL_MEMORY||16777216,Ye,Gn=[],Bt=[],or=[],bn=[],zr=!1;Bt.push({func:function(){Oh()}});function Rn(){if(a.preRun)for(typeof a.preRun=="function"&&(a.preRun=[a.preRun]);a.preRun.length;)br(a.preRun.shift());ba(Gn)}function Ar(){zr=!0,ba(Bt)}function xr(){ba(or)}function vn(){if(a.postRun)for(typeof a.postRun=="function"&&(a.postRun=[a.postRun]);a.postRun.length;)vr(a.postRun.shift());ba(bn)}function br(te){Gn.unshift(te)}function vr(te){bn.unshift(te)}var ir=0,bs=null,Hs=null;function xa(te){ir++,a.monitorRunDependencies&&a.monitorRunDependencies(ir)}function Ii(te){if(ir--,a.monitorRunDependencies&&a.monitorRunDependencies(ir),ir==0&&(bs!==null&&(clearInterval(bs),bs=null),Hs)){var ie=Hs;Hs=null,ie()}}a.preloadedImages={},a.preloadedAudios={};function vs(te){a.onAbort&&a.onAbort(te),te+="",D(te),M=!0,B=1,te="abort("+te+"). Build with -s ASSERTIONS=1 for more info.";var ie=new WebAssembly.RuntimeError(te);throw i(ie),ie}function Eh(te,ie){return String.prototype.startsWith?te.startsWith(ie):te.indexOf(ie)===0}var k0="data:application/octet-stream;base64,";function Fu(te){return Eh(te,k0)}var I0="file://";function $h(te){return Eh(te,I0)}var jn="tfjs-backend-wasm.wasm";Fu(jn)||(jn=A(jn));function Rh(te){try{if(te==jn&&$)return new Uint8Array($);if(v)return v(te);throw"both async and sync fetching of the wasm failed"}catch(ie){vs(ie)}}function S0(){if(!$&&(p||f)){if(typeof fetch=="function"&&!$h(jn))return fetch(jn,{credentials:"same-origin"}).then(function(te){if(!te.ok)throw"failed to load wasm binary file at '"+jn+"'";return te.arrayBuffer()}).catch(function(){return Rh(jn)});if(b)return new Promise(function(te,ie){b(jn,function(we){te(new Uint8Array(we))},ie)})}return Promise.resolve().then(function(){return Rh(jn)})}function Gs(){var te={a:$0};function ie(Qe,et){var wn=Qe.exports;a.asm=wn,N=a.asm.i,dt(N.buffer),Ye=a.asm.o,Ii("wasm-instantiate")}xa("wasm-instantiate");function we(Qe){ie(Qe.instance)}function lt(Qe){return S0().then(function(et){return WebAssembly.instantiate(et,te)}).then(Qe,function(et){D("failed to asynchronously prepare wasm: "+et),vs(et)})}function Ht(){return!$&&typeof WebAssembly.instantiateStreaming=="function"&&!Fu(jn)&&!$h(jn)&&typeof fetch=="function"?fetch(jn,{credentials:"same-origin"}).then(function(Qe){var et=WebAssembly.instantiateStreaming(Qe,te);return et.then(we,function(wn){return D("wasm streaming compile failed: "+wn),D("falling back to ArrayBuffer instantiation"),lt(we)})}):lt(we)}if(a.instantiateWasm)try{var Dt=a.instantiateWasm(te,ie);return Dt}catch(Qe){return D("Module.instantiateWasm callback failed with error: "+Qe),!1}return Ht().catch(i),{}}function ba(te){for(;te.length>0;){var ie=te.shift();if(typeof ie=="function"){ie(a);continue}var we=ie.func;typeof we=="number"?ie.arg===void 0?Ye.get(we)():Ye.get(we)(ie.arg):we(ie.arg===void 0?null:ie.arg)}}function io(){vs()}function T0(te,ie,we){$e.copyWithin(te,ie,ie+we)}function N0(){return $e.length}function js(te){try{return N.grow(te-be.byteLength+65535>>>16),dt(N.buffer),1}catch(ie){}}function _h(te){var ie=N0(),we=2147483648;if(te>we)return!1;for(var lt=1;lt<=4;lt*=2){var Ht=ie*(1+.2/lt);Ht=Math.min(Ht,te+100663296);var Dt=Math.min(we,me(Math.max(te,Ht),65536)),Qe=js(Dt);if(Qe)return!0}return!1}var Si={mappings:{},buffers:[null,[],[]],printChar:function(te,ie){var we=Si.buffers[te];ie===0||ie===10?((te===1?E:D)(se(we,0)),we.length=0):we.push(ie)},varargs:void 0,get:function(){Si.varargs+=4;var te=Be[Si.varargs-4>>2];return te},getStr:function(te){var ie=oe(te);return ie},get64:function(te,ie){return te}};function Dh(te){return 0}function C0(te,ie,we,lt,Ht){}function Fh(te,ie,we,lt){for(var Ht=0,Dt=0;Dt<we;Dt++){for(var Qe=Be[ie+Dt*8>>2],et=Be[ie+(Dt*8+4)>>2],wn=0;wn<et;wn++)Si.printChar(te,$e[Qe+wn]);Ht+=et}return Be[lt>>2]=Ht,0}function qn(){return 6}function Mh(te){return Be[Gh()>>2]=te,te}function E0(te){switch(te){case 30:return 16384;case 85:var ie=2147483648;return ie/16384;case 132:case 133:case 12:case 137:case 138:case 15:case 235:case 16:case 17:case 18:case 19:case 20:case 149:case 13:case 10:case 236:case 153:case 9:case 21:case 22:case 159:case 154:case 14:case 77:case 78:case 139:case 82:case 68:case 67:case 164:case 11:case 29:case 47:case 48:case 95:case 52:case 51:case 46:return 200809;case 27:case 246:case 127:case 128:case 23:case 24:case 160:case 161:case 181:case 182:case 242:case 183:case 184:case 243:case 244:case 245:case 165:case 178:case 179:case 49:case 50:case 168:case 169:case 175:case 170:case 171:case 172:case 97:case 76:case 32:case 173:case 35:case 80:case 81:case 79:return-1;case 176:case 177:case 7:case 155:case 8:case 157:case 125:case 126:case 92:case 93:case 129:case 130:case 131:case 94:case 91:return 1;case 74:case 60:case 69:case 70:case 4:return 1024;case 31:case 42:case 72:return 32;case 87:case 26:case 33:return 2147483647;case 34:case 1:return 47839;case 38:case 36:return 99;case 43:case 37:return 2048;case 0:return 2097152;case 3:return 65536;case 28:return 32768;case 44:return 32767;case 75:return 16384;case 39:return 1e3;case 89:return 700;case 71:return 256;case 40:return 255;case 2:return 100;case 180:return 64;case 25:return 20;case 5:return 16;case 6:return 6;case 73:return 4;case 84:return typeof navigator=="object"&&navigator.hardwareConcurrency||1}return Mh(28),-1}var $0={a:io,d:T0,e:_h,f:Dh,c:C0,b:Fh,g:qn,h:E0},R0=Gs(),Oh=a.___wasm_call_ctors=function(){return(Oh=a.___wasm_call_ctors=a.asm.j).apply(null,arguments)},Ti=a._init=function(){return(Ti=a._init=a.asm.k).apply(null,arguments)},Mu=a._register_tensor=function(){return(Mu=a._register_tensor=a.asm.l).apply(null,arguments)},_0=a._dispose_data=function(){return(_0=a._dispose_data=a.asm.m).apply(null,arguments)},D0=a._dispose=function(){return(D0=a._dispose=a.asm.n).apply(null,arguments)},F0=a._Abs=function(){return(F0=a._Abs=a.asm.p).apply(null,arguments)},Ce=a._Add=function(){return(Ce=a._Add=a.asm.q).apply(null,arguments)},M0=a._AddN=function(){return(M0=a._AddN=a.asm.r).apply(null,arguments)},O0=a._All=function(){return(O0=a._All=a.asm.s).apply(null,arguments)},P0=a._Any=function(){return(P0=a._Any=a.asm.t).apply(null,arguments)},z0=a._ArgMax=function(){return(z0=a._ArgMax=a.asm.u).apply(null,arguments)},L0=a._AvgPool=function(){return(L0=a._AvgPool=a.asm.v).apply(null,arguments)},lo=a._BatchMatMul=function(){return(lo=a._BatchMatMul=a.asm.w).apply(null,arguments)},B0=a._Ceil=function(){return(B0=a._Ceil=a.asm.x).apply(null,arguments)},W0=a._ClipByValue=function(){return(W0=a._ClipByValue=a.asm.y).apply(null,arguments)},V0=a._Conv2D=function(){return(V0=a._Conv2D=a.asm.z).apply(null,arguments)},U0=a._Conv2DBackpropInput=function(){return(U0=a._Conv2DBackpropInput=a.asm.A).apply(null,arguments)},H0=a._Cos=function(){return(H0=a._Cos=a.asm.B).apply(null,arguments)},G0=a._Cosh=function(){return(G0=a._Cosh=a.asm.C).apply(null,arguments)},j0=a._CropAndResize=function(){return(j0=a._CropAndResize=a.asm.D).apply(null,arguments)},q0=a._Cumsum=function(){return(q0=a._Cumsum=a.asm.E).apply(null,arguments)},K0=a._DepthToSpace=function(){return(K0=a._DepthToSpace=a.asm.F).apply(null,arguments)},va=a._DepthwiseConv2dNative=function(){return(va=a._DepthwiseConv2dNative=a.asm.G).apply(null,arguments)},Ou=a._Equal=function(){return(Ou=a._Equal=a.asm.H).apply(null,arguments)},Pu=a._Exp=function(){return(Pu=a._Exp=a.asm.I).apply(null,arguments)},X0=a._FlipLeftRight=function(){return(X0=a._FlipLeftRight=a.asm.J).apply(null,arguments)},Z0=a._Floor=function(){return(Z0=a._Floor=a.asm.K).apply(null,arguments)},Y0=a._FloorDiv=function(){return(Y0=a._FloorDiv=a.asm.L).apply(null,arguments)},J0=a._FusedBatchNorm=function(){return(J0=a._FusedBatchNorm=a.asm.M).apply(null,arguments)},Q0=a._FusedConv2D=function(){return(Q0=a._FusedConv2D=a.asm.N).apply(null,arguments)},Ke=a._FusedDepthwiseConv2D=function(){return(Ke=a._FusedDepthwiseConv2D=a.asm.O).apply(null,arguments)},eg=a._Gather=function(){return(eg=a._Gather=a.asm.P).apply(null,arguments)},tg=a._GatherNd=function(){return(tg=a._GatherNd=a.asm.Q).apply(null,arguments)},ng=a._Greater=function(){return(ng=a._Greater=a.asm.R).apply(null,arguments)},rg=a._GreaterEqual=function(){return(rg=a._GreaterEqual=a.asm.S).apply(null,arguments)},sg=a._LeakyRelu=function(){return(sg=a._LeakyRelu=a.asm.T).apply(null,arguments)},ag=a._Less=function(){return(ag=a._Less=a.asm.U).apply(null,arguments)},zu=a._LessEqual=function(){return(zu=a._LessEqual=a.asm.V).apply(null,arguments)},Ph=a._Log=function(){return(Ph=a._Log=a.asm.W).apply(null,arguments)},zh=a._LogicalAnd=function(){return(zh=a._LogicalAnd=a.asm.X).apply(null,arguments)},og=a._Max=function(){return(og=a._Max=a.asm.Y).apply(null,arguments)},ig=a._MaxPool=function(){return(ig=a._MaxPool=a.asm.Z).apply(null,arguments)},lg=a._Maximum=function(){return(lg=a._Maximum=a.asm._).apply(null,arguments)},ug=a._Mean=function(){return(ug=a._Mean=a.asm.$).apply(null,arguments)},cg=a._Min=function(){return(cg=a._Min=a.asm.aa).apply(null,arguments)},dg=a._Minimum=function(){return(dg=a._Minimum=a.asm.ba).apply(null,arguments)},hg=a._MirrorPad=function(){return(hg=a._MirrorPad=a.asm.ca).apply(null,arguments)},ht=a._Multiply=function(){return(ht=a._Multiply=a.asm.da).apply(null,arguments)},pg=a._Neg=function(){return(pg=a._Neg=a.asm.ea).apply(null,arguments)},fg=a._NonMaxSuppressionV3=function(){return(fg=a._NonMaxSuppressionV3=a.asm.fa).apply(null,arguments)},mg=a._NonMaxSuppressionV4=function(){return(mg=a._NonMaxSuppressionV4=a.asm.ga).apply(null,arguments)},Ni=a._NonMaxSuppressionV5=function(){return(Ni=a._NonMaxSuppressionV5=a.asm.ha).apply(null,arguments)},Lh=a._NotEqual=function(){return(Lh=a._NotEqual=a.asm.ia).apply(null,arguments)},Bh=a._OneHot=function(){return(Bh=a._OneHot=a.asm.ja).apply(null,arguments)},Wh=a._PadV2=function(){return(Wh=a._PadV2=a.asm.ka).apply(null,arguments)},gg=a._Pow=function(){return(gg=a._Pow=a.asm.la).apply(null,arguments)},yg=a._Prelu=function(){return(yg=a._Prelu=a.asm.ma).apply(null,arguments)},Vh=a._Prod=function(){return(Vh=a._Prod=a.asm.na).apply(null,arguments)},Ag=a._RealDiv=function(){return(Ag=a._RealDiv=a.asm.oa).apply(null,arguments)},Uh=a._Relu=function(){return(Uh=a._Relu=a.asm.pa).apply(null,arguments)},wa=a._Relu6=function(){return(wa=a._Relu6=a.asm.qa).apply(null,arguments)},xg=a._ResizeBilinear=function(){return(xg=a._ResizeBilinear=a.asm.ra).apply(null,arguments)},bg=a._Reverse=function(){return(bg=a._Reverse=a.asm.sa).apply(null,arguments)},l3=a._RotateWithOffset=function(){return(l3=a._RotateWithOffset=a.asm.ta).apply(null,arguments)},Hh=a._Round=function(){return(Hh=a._Round=a.asm.ua).apply(null,arguments)},vg=a._Rsqrt=function(){return(vg=a._Rsqrt=a.asm.va).apply(null,arguments)},wg=a._ScatterNd=function(){return(wg=a._ScatterNd=a.asm.wa).apply(null,arguments)},kg=a._SelectV2=function(){return(kg=a._SelectV2=a.asm.xa).apply(null,arguments)},Ig=a._Sigmoid=function(){return(Ig=a._Sigmoid=a.asm.ya).apply(null,arguments)},Sg=a._Sin=function(){return(Sg=a._Sin=a.asm.za).apply(null,arguments)},Tg=a._Softmax=function(){return(Tg=a._Softmax=a.asm.Aa).apply(null,arguments)},Ng=a._Sqrt=function(){return(Ng=a._Sqrt=a.asm.Ba).apply(null,arguments)},Cg=a._Square=function(){return(Cg=a._Square=a.asm.Ca).apply(null,arguments)},Eg=a._SquaredDifference=function(){return(Eg=a._SquaredDifference=a.asm.Da).apply(null,arguments)},$g=a._Step=function(){return($g=a._Step=a.asm.Ea).apply(null,arguments)},Rg=a._StridedSlice=function(){return(Rg=a._StridedSlice=a.asm.Fa).apply(null,arguments)},_g=a._Sub=function(){return(_g=a._Sub=a.asm.Ga).apply(null,arguments)},Dg=a._Sum=function(){return(Dg=a._Sum=a.asm.Ha).apply(null,arguments)},Fg=a._Tan=function(){return(Fg=a._Tan=a.asm.Ia).apply(null,arguments)},Mg=a._Tanh=function(){return(Mg=a._Tanh=a.asm.Ja).apply(null,arguments)},Og=a._Tile=function(){return(Og=a._Tile=a.asm.Ka).apply(null,arguments)},Pg=a._TopK=function(){return(Pg=a._TopK=a.asm.La).apply(null,arguments)},zg=a._Transform=function(){return(zg=a._Transform=a.asm.Ma).apply(null,arguments)},Lg=a._Transpose=function(){return(Lg=a._Transpose=a.asm.Na).apply(null,arguments)},Bg=a.__FusedMatMul=function(){return(Bg=a.__FusedMatMul=a.asm.Oa).apply(null,arguments)},Wg=a._malloc=function(){return(Wg=a._malloc=a.asm.Pa).apply(null,arguments)},Vg=a._free=function(){return(Vg=a._free=a.asm.Qa).apply(null,arguments)},Gh=a.___errno_location=function(){return(Gh=a.___errno_location=a.asm.Ra).apply(null,arguments)},jh=a.stackSave=function(){return(jh=a.stackSave=a.asm.Sa).apply(null,arguments)},qh=a.stackRestore=function(){return(qh=a.stackRestore=a.asm.Ta).apply(null,arguments)},Lu=a.stackAlloc=function(){return(Lu=a.stackAlloc=a.asm.Ua).apply(null,arguments)};a.cwrap=ee;var Ci;function Ug(te){this.name="ExitStatus",this.message="Program terminated with exit("+te+")",this.status=te}Hs=function te(){Ci||Bu(),Ci||(Hs=te)};function Bu(te){if(te=te||c,ir>0||(Rn(),ir>0))return;function ie(){Ci||(Ci=!0,a.calledRun=!0,!M&&(Ar(),xr(),o(a),a.onRuntimeInitialized&&a.onRuntimeInitialized(),vn()))}a.setStatus?(a.setStatus("Running..."),setTimeout(function(){setTimeout(function(){a.setStatus("")},1),ie()},1)):ie()}if(a.run=Bu,a.preInit)for(typeof a.preInit=="function"&&(a.preInit=[a.preInit]);a.preInit.length>0;)a.preInit.pop()();return Bu(),s.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)}}),x_="3.8.0",b_="3.8.0",v_="3.8.0",w_="3.8.0",k_="3.8.0",I_="3.8.0",S_="3.8.0",T_="3.8.0",N_=1e-7,C_=1e-4,E_=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}},E3=class{refCount(e){return Lr("refCount")}incRef(e){return Lr("incRef")}timerAvailable(){return!0}time(e){return Lr("time")}read(e){return Lr("read")}readSync(e){return Lr("readSync")}numDataIds(){return Lr("numDataIds")}disposeData(e,t){return Lr("disposeData")}write(e,t,n){return Lr("write")}move(e,t,n,r,s){return Lr("move")}memory(){return Lr("memory")}floatPrecision(){return Lr("floatPrecision")}epsilon(){return this.floatPrecision()===32?N_:C_}dispose(){return Lr("dispose")}};function Lr(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 $3(e){let t=e.length,n=0;for(;t>0;)n=Math.random()*t|0,t--,Jh(e,t,n)}function $_(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,r=0;for(;n>0;)r=Math.random()*n|0,n--,Jh(e,n,r),Jh(t,n,r)}function ju(e,t,n){return Math.max(e,Math.min(t,n))}function R_(e){return e%2==0?e:e+1}function Jh(e,t,n){let r=e[t];e[t]=e[n],e[n]=r}function __(e){let t=0;for(let n=0;n<e.length;n++)t+=e[n];return t}function D_(e,t){let n=Math.random();return t*n+(1-n)*e}function F_(e,t){let n=0;for(let r=0;r<e.length;r++){let s=Number(e[r])-Number(t[r]);n+=s*s}return n}function L(e,t){if(!e)throw new Error(typeof t=="string"?t:t())}function Mn(e,t,n=""){L(Zs(e,t),()=>n+` Shapes ${e} and ${t} must match`)}function ho(e){L(e!=null,()=>"The input to the tensor constructor must be a non-null value.")}function po(e,t=[],n=!1){if(t==null&&(t=[]),Array.isArray(e)||Nn(e)&&!n)for(let r=0;r<e.length;++r)po(e[r],t,n);else t.push(e);return t}function Yt(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 M_(e){return e.length===0}function Zs(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 Kn(e){return e%1==0}function O_(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 P_(e){let t=Math.ceil(Math.sqrt(e));return[t,Math.ceil(e/t)]}function z_(e){let t=new Uint32Array(e);for(let n=0;n<e;++n)t[n]=n;return $3(t),t}function qu(e,t){return t<=e.length?e:e+" ".repeat(t-e.length)}function L_(e,t=r=>0,n){return new Promise((r,s)=>{let a=0,o=()=>{if(e()){r();return}a++;let i=t(a);if(n!=null&&a>=n){s();return}setTimeout(o,i)};o()})}function B_(e,t){let n=1,r=-1;for(let a=0;a<e.length;++a)if(e[a]>=0)n*=e[a];else if(e[a]===-1){if(r!==-1)throw Error(`Shapes can only have 1 implicit size. Found -1 at dim ${r} and dim ${a}`);r=a}else if(e[a]<0)throw Error(`Shapes can not be < 0. Found ${e[a]} at dim ${a}`);if(r===-1){if(t>0&&t!==n)throw Error(`Size(${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} / ${n}`);let s=e.slice();return s[r]=t/n,s}function Ku(e,t){let n=t.length;return e=e==null?t.map((r,s)=>s):[].concat(e),L(e.every(r=>r>=-n&&r<n),()=>`All values in axis param must be in range [-${n}, ${n}) but got axis ${e}`),L(e.every(r=>Kn(r)),()=>`All values in axis param must be integers but got axis ${e}`),e.map(r=>r<0?n+r:r)}function R3(e,t){let n=[],r=[],s=t!=null&&Array.isArray(t)&&t.length===0,a=t==null||s?null:Ku(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]),r.push(i)),a[o]<=i&&o++}e[i]!==1&&(n.push(e[i]),r.push(i))}return{newShape:n,keptDims:r}}function _3(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 D3(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 F3(e,t){for(let n=0;n<e.length;n++){let r=e[n];if(isNaN(r)||!isFinite(r))throw Error(`A tensor of type ${t} being uploaded contains ${r}.`)}}function M3(e){return e==="bool"||e==="complex64"||e==="float32"||e==="int32"||e==="string"}function W_(e,t){return!(t==="complex64"||t==="float32"&&e!=="complex64"||t==="int32"&&e!=="float32"&&e!=="complex64"||t==="bool"&&e==="bool")}function Nn(e){return e instanceof Float32Array||e instanceof Int32Array||e instanceof Uint8Array}function Yg(e){if(e==="float32"||e==="int32")return 4;if(e==="complex64")return 8;if(e==="bool")return 1;throw new Error(`Unknown dtype ${e}`)}function O3(e){if(e==null)return 0;let t=0;return e.forEach(n=>t+=n.length),t}function Sa(e){return typeof e=="string"||e instanceof String}function P3(e){return typeof e=="boolean"}function z3(e){return typeof e=="number"}function Qh(e){return Array.isArray(e)?Qh(e[0]):e instanceof Float32Array?"float32":e instanceof Int32Array||e instanceof Uint8Array?"int32":z3(e)?"float32":Sa(e)?"string":P3(e)?"bool":"float32"}function Ta(e){return!!(e&&e.constructor&&e.call&&e.apply)}function ep(e,t){for(let n=t;n<e;++n)if(e%n==0)return n;return e}function Mi(e){let t=e.length;if(t<2)return[];let n=new Array(t-1);n[t-2]=e[t-1];for(let r=t-3;r>=0;--r)n[r]=n[r+1]*e[r+1];return n}function L3(e,t,n,r=!1){let s=new Array;if(t.length===1){let a=t[0]*(r?2:1);for(let o=0;o<a;o++)s[o]=n[e+o]}else{let a=t[0],o=t.slice(1),i=o.reduce((l,u)=>l*u)*(r?2:1);for(let l=0;l<a;l++)s[l]=L3(e+l*i,o,n,r)}return s}function Oi(e,t,n=!1){if(e.length===0)return t[0];let r=e.reduce((s,a)=>s*a)*(n?2:1);if(r===0)return[];if(r!==t.length)throw new Error(`[${e}] does not match the input size ${t.length}${n?" for a complex tensor":""}.`);return L3(0,e,t,n)}function Jg(e,t){let n=tp(e,t);for(let r=0;r<n.length;r++)n[r]=1;return n}function tp(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 V_(e,t){let n=e.reduce((r,s)=>r*s,1);if(t==null||t==="float32")return Oi(e,new Float32Array(n));if(t==="int32")return Oi(e,new Int32Array(n));if(t==="bool")return Oi(e,new Uint8Array(n));throw new Error(`Unknown data type ${t}`)}function Qg(e){e.forEach(t=>{L(Number.isInteger(t)&&t>=0,()=>`Tensor must have a shape comprised of positive integers but got shape [${e}].`)})}function U_(e,t,n){if(t===0)return 0;if(t===1)return e[0];let r=e[e.length-1];for(let s=0;s<e.length-1;++s)r+=n[s]*e[s];return r}function H_(e,t,n){if(t===0)return[];if(t===1)return[e];let r=new Array(t);for(let s=0;s<r.length-1;++s)r[s]=Math.floor(e/n[s]),e-=r[s]*n[s];return r[r.length-1]=e,r}function e2(e){return e&&e.then&&typeof e.then=="function"}var B3="tfjsflags",W3=class{constructor(e){this.global=e,this.flags={},this.flagRegistry={},this.urlFlags={},this.getQueryParams=G_,this.populateURLFlags()}setPlatform(e,t){this.platform!=null&&console.warn(`Platform ${this.platformName} has already been set. Overwriting the platform with ${t}.`),this.platformName=e,this.platform=t}registerFlag(e,t,n){if(this.flagRegistry[e]={evaluationFn:t,setHook:n},this.urlFlags[e]!=null){let r=this.urlFlags[e];console.warn(`Setting feature override from URL ${e}: ${r}.`),this.set(e,r)}}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(e2(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);B3 in e&&e[B3].split(",").forEach(n=>{let[r,s]=n.split(":");this.urlFlags[r]=q_(r,s)})}};function G_(e){let t={};return e.replace(/[?&]([^=?&]+)(?:=([^&]*))?/g,(n,...r)=>(j_(t,r[0],r[1]),r.join("="))),t}function j_(e,t,n){e[decodeURIComponent(t)]=decodeURIComponent(n||"")}function q_(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 ${e}.`)}function ct(){return kr}var kr=null;function K_(e){kr=e}var t2;function V3(){if(t2==null){let e;if(typeof window!="undefined")e=window;else if(typeof global!="undefined")e=global;else if(typeof process!="undefined")e=process;else if(typeof self!="undefined")e=self;else throw new Error("Could not find a global object");t2=e}return t2}function X_(){let e=V3();return e._tfGlobals==null&&(e._tfGlobals=new Map),e._tfGlobals}function n2(e,t){let n=X_();if(n.has(e))return n.get(e);{let r=t();return n.set(e,r),n.get(e)}}var U3="Abs",H3="Acos",G3="Acosh",r2="Add",j3="AddN",q3="All",K3="Any",X3="ArgMax",Z3="ArgMin",Y3="Asin",J3="Asinh",Q3="Atan",ev="Atanh",tv="Atan2",nv="AvgPool",Z_="AvgPoolGrad",rv="AvgPool3D",Y_="AvgPool3DGrad",sv="BatchMatMul",av="BatchToSpaceND",ov="Bincount",J_="BroadcastTo",s2="Cast",iv="Ceil",lv="ClipByValue",uv="Complex",cv="ComplexAbs",dv="Concat",hv="Conv2D",pv="Conv2DBackpropFilter",fv="Conv2DBackpropInput",mv="Conv3D",Q_="Conv3DBackpropFilterV2",gv="Conv3DBackpropInputV2",yv="Cos",Av="Cosh",xv="Cumsum",bv="CropAndResize",vv="DenseBincount",wv="DepthToSpace",kv="DepthwiseConv2dNative",Iv="DepthwiseConv2dNativeBackpropFilter",Sv="DepthwiseConv2dNativeBackpropInput",Tv="Diag",Nv="Dilation2D",eD="Dilation2DBackpropInput",tD="Dilation2DBackpropFilter",Cv="RealDiv",Ev="Einsum",$v="Elu",nD="EluGrad",Rv="Erf",_v="Equal",Dv="Exp",Fv="ExpandDims",Mv="Expm1",Ov="FFT",Pv="Fill",zv="FlipLeftRight",Lv="Floor",Bv="FloorDiv",Wv="FusedBatchNorm",Vv="GatherV2",Uv="GatherNd",Hv="Greater",Gv="GreaterEqual",a2="Identity",jv="IFFT",qv="Imag",Kv="IsFinite",Xv="IsInf",Zv="IsNan",Yv="LeakyRelu",Jv="Less",Qv="LessEqual",ew="LinSpace",tw="Log",nw="Log1p",rw="LogicalAnd",sw="LogicalNot",aw="LogicalOr",rD="LogSoftmax",ow="LRN",sD="LRNGrad",iw="Max",lw="Maximum",uw="MaxPool",aD="MaxPoolGrad",cw="MaxPool3D",oD="MaxPool3DGrad",dw="MaxPoolWithArgmax",hw="Mean",pw="Min",fw="Minimum",mw="MirrorPad",gw="Mod",yw="Multinomial",Aw="Multiply",xw="Neg",bw="NotEqual",vw="NonMaxSuppressionV3",ww="NonMaxSuppressionV4",kw="NonMaxSuppressionV5",Iw="OnesLike",Sw="OneHot",Tw="Pack",Nw="PadV2",iD="Pool",Cw="Pow",Ew="Prelu",$w="Prod",Rw="Range",_w="Real",Dw="Reciprocal",Fw="Relu",Mw="Reshape",Ow="ResizeNearestNeighbor",lD="ResizeNearestNeighborGrad",Pw="ResizeBilinear",uD="ResizeBilinearGrad",zw="Relu6",Lw="Reverse",Bw="Round",Ww="Rsqrt",Vw="ScatterNd",Uw="Select",Hw="Selu",Gw="Slice",jw="Sin",qw="Sinh",Kw="Sign",Xw="Sigmoid",Zw="Softplus",Yw="Sqrt",Jw="Sum",Qw="SpaceToBatchND",e7="SplitV",t7="Softmax",n7="SparseFillEmptyRows",r7="SparseReshape",s7="SparseSegmentMean",a7="SparseSegmentSum",o7="SparseToDense",i7="SquaredDifference",cD="Square",l7="StridedSlice",u7="StringNGrams",c7="StringSplit",d7="StringToHashBucketFast",h7="Sub",p7="Tan",f7="Tanh",o2="Tile",m7="TopK",g7="Transform",y7="Transpose",A7="Unique",x7="Unpack",b7="UnsortedSegmentSum",v7="ZerosLike",w7="Step",i2="FromPixels",k7="RotateWithOffset",l2="_FusedMatMul",u2="FusedConv2D",c2="FusedDepthwiseConv2D",Pi=n2("kernelRegistry",()=>new Map),Xu=n2("gradRegistry",()=>new Map);function np(e,t){let n=h2(e,t);return Pi.get(n)}function d2(e){return Xu.get(e)}function zi(e){let t=Pi.entries(),n=[];for(;;){let{done:r,value:s}=t.next();if(r)break;let[a,o]=s,[i]=a.split("_");i===e&&n.push(o)}return n}function rp(e){let{kernelName:t,backendName:n}=e,r=h2(t,n);Pi.has(r)&&console.warn(`The kernel '${t}' for backend '${n}' is already registered`),Pi.set(r,e)}function dD(e){let{kernelName:t}=e;Xu.has(t)&&ct().getBool("DEBUG")&&console.warn(`Overriding the gradient for '${t}'`),Xu.set(t,e)}function hD(e,t){let n=h2(e,t);if(!Pi.has(n))throw new Error(`The kernel '${e}' for backend '${t}' is not registered`);Pi.delete(n)}function pD(e){if(!Xu.has(e))throw new Error(`The gradient '${e}' for backend is not registered`);Xu.delete(e)}function fD(e,t){zi(e).forEach(r=>{let s=Object.assign({},r,{backendName:t});rp(s)})}function h2(e,t){return`${t}_${e}`}var I7={};_e(I7,{arraysEqual:()=>Zs,assert:()=>L,assertNonNegativeIntegerDimensions:()=>Qg,assertNonNull:()=>ho,assertShapesMatch:()=>Mn,bytesFromStringArray:()=>O3,bytesPerElement:()=>Yg,checkConversionForErrors:()=>F3,clamp:()=>ju,computeStrides:()=>Mi,createScalarValue:()=>bD,createShuffledIndices:()=>z_,decodeString:()=>op,distSquared:()=>F_,encodeString:()=>Ju,fetch:()=>wD,fingerPrint64:()=>xD,flatten:()=>po,getArrayFromDType:()=>D3,getTypedArrayFromDType:()=>_3,hasEncodingLoss:()=>W_,hexToLong:()=>Zu,indexToLoc:()=>H_,inferDtype:()=>Qh,inferFromImplicitShape:()=>B_,isBoolean:()=>P3,isFunction:()=>Ta,isInt:()=>Kn,isNumber:()=>z3,isPromise:()=>e2,isScalarShape:()=>M_,isString:()=>Sa,isTypedArray:()=>Nn,isValidDtype:()=>M3,locToIndex:()=>U_,makeOnesTypedArray:()=>Jg,makeZerosNestedTypedArray:()=>V_,makeZerosTypedArray:()=>tp,nearestDivisor:()=>ep,nearestLargerEven:()=>R_,now:()=>Yu,parseAxisParam:()=>Ku,randUniform:()=>D_,repeatedTry:()=>L_,rightPad:()=>qu,shuffle:()=>$3,shuffleCombo:()=>$_,sizeFromShape:()=>Yt,sizeToSquarishShape:()=>P_,squeezeShape:()=>R3,sum:()=>__,swap:()=>Jh,tanh:()=>O_,toNestedArray:()=>Oi,toTypedArray:()=>ap});var S7=Xs(S3()),fo=S7.default||S7;function Zu(e){return fo.fromString(e,!0,16)}var T7=Zu("c3a5c85c97cb3127"),mo=Zu("b492b66fbe98f273"),On=Zu("9ae16a3b2f90404f");function p2(e){return e.xor(e.shru(47))}function N7(e,t,n){let r=e.slice(t,t+n);return fo.fromBytes(Array.from(r),!0,!0)}function kt(e,t){return N7(e,t,8)}function C7(e,t){return N7(e,t,4)}function pn(e,t){return t===0?e:e.shru(t).or(e.shl(64-t))}function Na(e,t,n=Zu("9ddfea08eb382d69")){let r=e.xor(t).mul(n);r=r.xor(r.shru(47));let s=t.xor(r).mul(n);return s=s.xor(s.shru(47)),s=s.mul(n),s}function mD(e,t,n,r,s,a){s=s.add(e),a=pn(a.add(s).add(r),21);let o=s;return s=s.add(t),s=s.add(n),a=a.add(pn(s,44)),[s.add(r),a.add(o)]}function sp(e,t,n,r){return mD(kt(e,t),kt(e,t+8),kt(e,t+16),kt(e,t+24),n,r)}function gD(e,t=e.length){if(t>=8){let n=On.add(t*2),r=kt(e,0).add(On),s=kt(e,t-8),a=pn(s,37).mul(n).add(r),o=pn(r,25).add(s).mul(n);return Na(a,o,n)}if(t>=4){let n=On.add(t*2),r=C7(e,0);return Na(r.shl(3).add(t),C7(e,t-4),n)}if(t>0){let n=e[0],r=e[t>>1],s=e[t-1],a=n+(r<<8),o=t+(s<<2);return p2(On.mul(a).xor(T7.mul(o))).mul(On)}return On}function yD(e,t=e.length){let n=On.add(t*2),r=kt(e,0).mul(mo),s=kt(e,8),a=kt(e,t-8).mul(n),o=kt(e,t-16).mul(On);return Na(pn(r.add(s),43).add(pn(a,30)).add(o),r.add(pn(s.add(On),18)).add(a),n)}function AD(e,t=e.length){let n=On.add(t*2),r=kt(e,0).mul(On),s=kt(e,8),a=kt(e,t-8).mul(n),o=kt(e,t-16).mul(On),i=pn(r.add(s),43).add(pn(a,30)).add(o),l=Na(i,r.add(pn(s.add(On),18)).add(a),n),u=kt(e,16).mul(n),c=kt(e,24),d=i.add(kt(e,t-32)).mul(n),h=l.add(kt(e,t-24)).mul(n);return Na(pn(u.add(c),43).add(pn(d,30)).add(h),u.add(pn(c.add(r),18)).add(d),n)}function xD(e,t=e.length){let n=fo.fromNumber(81,!0);if(t<=32)return t<=16?gD(e,t):yD(e,t);if(t<=64)return AD(e,t);let r=n,s=n.mul(mo).add(113),a=p2(s.mul(On).add(113)).mul(On),o=[fo.UZERO,fo.UZERO],i=[fo.UZERO,fo.UZERO];r=r.mul(On).add(kt(e,0));let l=0,u=(t-1>>6)*64,c=u+(t-1&63)-63;do r=pn(r.add(s).add(o[0]).add(kt(e,l+8)),37).mul(mo),s=pn(s.add(o[1]).add(kt(e,l+48)),42).mul(mo),r=r.xor(i[1]),s=s.add(o[0]).add(kt(e,l+40)),a=pn(a.add(i[0]),33).mul(mo),o=sp(e,l,o[1].mul(mo),r.add(i[0])),i=sp(e,l+32,a.add(i[1]),s.add(kt(e,l+16))),[a,r]=[r,a],l+=64;while(l!==u);let d=mo.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]),r=pn(r.add(s).add(o[0]).add(kt(e,l+8)),37).mul(d),s=pn(s.add(o[1]).add(kt(e,l+48)),42).mul(d),r=r.xor(i[1].mul(9)),s=s.add(o[0].mul(9).add(kt(e,l+40))),a=pn(a.add(i[0]),33).mul(d),o=sp(e,l,o[1].mul(d),r.add(i[0])),i=sp(e,l+32,a.add(i[1]),s.add(kt(e,l+16))),[a,r]=[r,a],Na(Na(o[0],i[0],d).add(p2(s).mul(T7)).add(a),Na(o[1],i[1],d).add(r),d)}function bD(e,t){return t==="string"?Ju(e):ap([e],t)}function vD(e,t){return e instanceof Float32Array&&t==="float32"||e instanceof Int32Array&&t==="int32"||e instanceof Uint8Array&&t==="bool"}function ap(e,t){if(t==="string")throw new Error("Cannot convert a string[] to a TypedArray");if(Array.isArray(e)&&(e=po(e)),ct().getBool("DEBUG")&&F3(e,t),vD(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 r=0;r<n.length;++r)Math.round(e[r])!==0&&(n[r]=1);return n}else throw new Error(`Unknown data type ${t}`)}function Yu(){return ct().platform.now()}function wD(e,t){return ct().platform.fetch(e,t)}function Ju(e,t="utf-8"){return t=t||"utf-8",ct().platform.encode(e,t)}function op(e,t="utf-8"){return t=t||"utf-8",ct().platform.decode(e,t)}var kD=class{constructor(e,t){this.backendTimer=e,this.logger=t,t==null&&(this.logger=new SD)}profileKernel(e,t,n){let r,s=()=>{r=n()},a,o=Yu();if(this.backendTimer.timerAvailable())a=this.backendTimer.time(s);else{s();for(let l of r)l.dataSync();a=Promise.resolve({kernelMs:Yu()-o})}if(ct().getBool("CHECK_COMPUTATION_FOR_ERRORS"))for(let l=0;l<r.length;l++){let u=r[l];u.data().then(c=>{ID(c,u.dtype,e)})}return{kernelName:e,outputs:r,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:r,inputs:s,extraInfo:a}=e;n.forEach(o=>{Promise.all([o.data(),r,a]).then(i=>{this.logger.logKernelProfile(t,o,i[0],i[1],s,i[2])})})}};function ID(e,t,n){if(t!=="float32")return!1;for(let r=0;r<e.length;r++){let s=e[r];if(isNaN(s)||!isFinite(s))return console.warn(`Found ${s} in the result of '${n}'`),!0}return!1}var SD=class{logKernelProfile(e,t,n,r,s,a){let o=typeof r=="number"?qu(`${r}ms`,9):r.error,i=qu(e,25),l=t.rank,u=t.size,c=qu(t.shape.toString(),14),d="";for(let h in s){let p=s[h];if(p!=null){let f=p.shape||t.shape,m=f.length;d+=`${h}: ${m}D ${m>0?f:""} `}}console.log(`%c${i} %c${o} %c${l}D ${c} %c${u} %c${d} %c${a}`,"font-weight:bold","color:red","color:blue","color: orange","color: green","color: steelblue")}};function TD(e,t,n){let r={},s={};for(let l=0;l<t.length;l++)r[t[l].id]=!0;for(let l=0;l<e.length;l++){let u=e[l],c=u.inputs;for(let d in c){let h=c[d],p=!1;for(let f=0;f<t.length;f++)if(r[h.id]){u.outputs.forEach(m=>r[m.id]=!0),p=!0,s[u.id]=!0;break}if(p)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 d=0;d<u.outputs.length;d++)if(a[u.outputs[d].id]){for(let h in c)a[c[h].id]=!0,o[u.id]=!0;break}}let i=[];for(let l=0;l<e.length;l++){let u=e[l];if(s[u.id]&&o[u.id]){let c={};for(let h in u.inputs){let p=u.inputs[h];r[p.id]&&(c[h]=p)}let d=Object.assign({},u);d.inputs=c,d.outputs=u.outputs,i.push(d)}}return i}function ND(e,t,n,r){for(let s=t.length-1;s>=0;s--){let a=t[s],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(!Zs(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 d=e[c.id];e[c.id]=r(d,u),d.dispose()}}}}var E7=20,Qu=3,f2=7;function CD(e,t,n,r){let s=Mi(t),a=ED(e,t,n,s),o=t.length,i=ip(e,t,n,s,a),l=["Tensor"];return r&&(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 ED(e,t,n,r){let s=Yt(t),a=r[r.length-1],o=new Array(a).fill(0),i=t.length,l=n==="complex64"?tc(e):e;if(i>1)for(let u=0;u<s/a;u++){let c=u*a;for(let d=0;d<a;d++)o[d]=Math.max(o[d],ec(l[c+d],0,n).length)}return o}function ec(e,t,n){let r;return Array.isArray(e)?r=`${parseFloat(e[0].toFixed(f2))} + ${parseFloat(e[1].toFixed(f2))}j`:Sa(e)?r=`'${e}'`:n==="bool"?r=$7(e):r=parseFloat(e.toFixed(f2)).toString(),qu(r,t)}function $7(e){return e===0?"false":"true"}function ip(e,t,n,r,s,a=!0){let o=n==="complex64"?2:1,i=t[0],l=t.length;if(l===0){if(n==="complex64"){let m=tc(e);return[ec(m[0],0,n)]}return n==="bool"?[$7(e[0])]:[e[0].toString()]}if(l===1){if(i>E7){let g=Qu*o,y=Array.from(e.slice(0,g)),A=Array.from(e.slice((i-Qu)*o,i*o));return n==="complex64"&&(y=tc(y),A=tc(A)),["["+y.map((x,b)=>ec(x,s[b],n)).join(", ")+", ..., "+A.map((x,b)=>ec(x,s[i-Qu+b],n)).join(", ")+"]"]}let m=n==="complex64"?tc(e):Array.from(e);return["["+m.map((g,y)=>ec(g,s[y],n)).join(", ")+"]"]}let u=t.slice(1),c=r.slice(1),d=r[0]*o,h=[];if(i>E7){for(let m=0;m<Qu;m++){let g=m*d,y=g+d;h.push(...ip(e.slice(g,y),u,n,c,s,!1))}h.push("...");for(let m=i-Qu;m<i;m++){let g=m*d,y=g+d;h.push(...ip(e.slice(g,y),u,n,c,s,m===i-1))}}else for(let m=0;m<i;m++){let g=m*d,y=g+d;h.push(...ip(e.slice(g,y),u,n,c,s,m===i-1))}let p=l===2?",":"";h[0]="["+h[0]+p;for(let m=1;m<h.length-1;m++)h[m]=" "+h[m]+p;let f=`,
`;for(let m=2;m<l;m++)f+=`
`;return h[h.length-1]=" "+h[h.length-1]+"]"+(a?"":f),h}function tc(e){let t=[];for(let n=0;n<e.length;n+=2)t.push([e[n],e[n+1]]);return t}var lp=class{constructor(e,t,n){if(this.dtype=t,this.shape=e.slice(),this.size=Yt(e),n!=null){let r=n.length;L(r===this.size,()=>`Length of values '${r}' does not match the size inferred by the shape '${this.size}'.`)}if(t==="complex64")throw new Error("complex64 dtype TensorBuffers are not supported. Please create a TensorBuffer for the real and imaginary parts separately and call tf.complex(real, imag).");this.values=n||D3(t,this.size),this.strides=Mi(e)}set(e,...t){t.length===0&&(t=[0]),L(t.length===this.rank,()=>`The number of provided coordinates (${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 r of e){if(r<0||r>=this.shape[t]){let s=`Requested out of range element at ${e}. Buffer shape=${this.shape}`;throw new Error(s)}t++}let n=e[e.length-1];for(let r=0;r<e.length-1;++r)n+=this.strides[r]*e[r];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 ws().makeTensor(this.values,this.shape,this.dtype)}},ws=null,Li=null,$D=null;function RD(e){ws=e}function _D(e){Li=e}function DD(e){$D=e}var It=class{constructor(e,t,n,r){this.kept=!1,this.isDisposedInternal=!1,this.shape=e.slice(),this.dtype=t||"float32",this.size=Yt(e),this.strides=Mi(e),this.dataId=n,this.id=r,this.rankType=this.rank<5?this.rank.toString():"higher"}get rank(){return this.shape.length}async buffer(){let e=await this.data();return Li.buffer(this.shape,this.dtype,e)}bufferSync(){return Li.buffer(this.shape,this.dtype,this.dataSync())}async array(){let e=await this.data();return Oi(this.shape,e,this.dtype==="complex64")}arraySync(){return Oi(this.shape,this.dataSync(),this.dtype==="complex64")}async data(){this.throwIfDisposed();let e=ws().read(this.dataId);if(this.dtype==="string"){let t=await e;try{return t.map(n=>op(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}dataSync(){this.throwIfDisposed();let e=ws().readSync(this.dataId);if(this.dtype==="string")try{return e.map(t=>op(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 ws().read(this.dataId);return this.dtype==="string"?e:new Uint8Array(e.buffer)}dispose(){this.isDisposed||(ws().disposeTensor(this),this.isDisposedInternal=!0)}get isDisposed(){return this.isDisposedInternal}throwIfDisposed(){if(this.isDisposed)throw new Error("Tensor is disposed.")}print(e=!1){return Li.print(this,e)}clone(){return this.throwIfDisposed(),Li.clone(this)}toString(e=!1){let t=this.dataSync();return CD(t,this.shape,this.dtype,e)}cast(e){return this.throwIfDisposed(),Li.cast(this,e)}variable(e=!0,t,n){return this.throwIfDisposed(),ws().makeVariable(this,e,t,n)}};Object.defineProperty(It,Symbol.hasInstance,{value:e=>!!e&&e.data!=null&&e.dataSync!=null&&e.throwIfDisposed!=null});function FD(){return n2("Tensor",()=>It)}FD();var nc=class extends It{constructor(e,t,n,r){super(e.shape,e.dtype,e.dataId,r);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(!Zs(e.shape,this.shape))throw new Error(`shape of the new value (${e.shape}) and previous value (${this.shape}) must match`);ws().disposeTensor(this),this.dataId=e.dataId,ws().incRef(this,null)}dispose(){ws().disposeVariable(this),this.isDisposedInternal=!0}};Object.defineProperty(nc,Symbol.hasInstance,{value:e=>e instanceof It&&e.assign!=null&&e.assign instanceof Function});var R7={};_e(R7,{assertTypesMatch:()=>_7,getTensorsInContainer:()=>b2,isTensorInList:()=>PD,makeTypesMatch:()=>Wt});var m2;(function(e){e.R0="R0",e.R1="R1",e.R2="R2",e.R3="R3",e.R4="R4",e.R5="R5",e.R6="R6"})(m2||(m2={}));var g2;(function(e){e.float32="float32",e.int32="int32",e.bool="int32",e.complex64="complex64"})(g2||(g2={}));var y2;(function(e){e.float32="float32",e.int32="int32",e.bool="bool",e.complex64="complex64"})(y2||(y2={}));var A2;(function(e){e.float32="float32",e.int32="float32",e.bool="float32",e.complex64="complex64"})(A2||(A2={}));var x2;(function(e){e.float32="complex64",e.int32="complex64",e.bool="complex64",e.complex64="complex64"})(x2||(x2={}));var MD={float32:A2,int32:g2,bool:y2,complex64:x2};function up(e,t){if(e==="string"||t==="string"){if(e==="string"&&t==="string")return"string";throw new Error(`Can not upcast ${e} with ${t}`)}return MD[e][t]}function OD(e){return up(e,"int32")}function Wt(e,t){if(e.dtype===t.dtype)return[e,t];let n=up(e.dtype,t.dtype);return[e.cast(n),t.cast(n)]}function _7(e,t){L(e.dtype===t.dtype,()=>`The dtypes of the first(${e.dtype}) and second(${t.dtype}) input must match`)}function PD(e,t){return t.some(n=>n.id===e.id)}function b2(e){let t=[],n=new Set;return D7(e,t,n),t}function D7(e,t,n){if(e==null)return;if(e instanceof It){t.push(e);return}if(!zD(e))return;let r=e;for(let s in r){let a=r[s];n.has(a)||(n.add(a),D7(a,t,n))}}function zD(e){return Array.isArray(e)||typeof e=="object"}function v2(e){return e.kernelName!=null}var F7=class{constructor(){this.registeredVariables={},this.nextTapeNodeId=0,this.numBytes=0,this.numTensors=0,this.numStringTensors=0,this.numDataBuffers=0,this.gradientDepth=0,this.kernelDepth=0,this.scopeStack=[],this.numDataMovesStack=[],this.nextScopeId=0,this.tensorInfo=new WeakMap,this.profiling=!1,this.activeProfile={newBytes:0,newTensors:0,peakBytes:0,kernels:[],result:null,get kernelNames(){return Array.from(new Set(this.kernels.map(e=>e.name)))}}}dispose(){for(let e in this.registeredVariables)this.registeredVariables[e].dispose()}},w2=class{constructor(e){this.ENV=e,this.registry={},this.registryFactory={},this.pendingBackendInitId=0,this.state=new F7}async ready(){if(this.pendingBackendInit!=null)return this.pendingBackendInit.then(()=>{});if(this.backendInstance!=null)return;let e=this.getSortedBackends();for(let t=0;t<e.length;t++){let n=e[t];if(await this.initializeBackend(n).success){await this.setBackend(n);return}}throw new Error("Could not initialize any backends, all backend initializations failed.")}get backend(){if(this.pendingBackendInit!=null)throw new Error(`Backend '${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 '${e}' has not yet been initialized. Make sure to await tf.ready() or await tf.setBackend() before calling other methods`);this.setBackend(e)}return this.backendInstance}backendNames(){return Object.keys(this.registryFactory)}findBackend(e){if(!(e in this.registry))if(e in this.registryFactory){let{asyncInit:t}=this.initializeBackend(e);if(t)return null}else return null;return this.registry[e]}findBackendFactory(e){return e in this.registryFactory?this.registryFactory[e].factory:null}registerBackend(e,t,n=1){return e in this.registryFactory?(console.warn(`${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 kD(this.backendInstance),!0}setupRegisteredKernels(){zi(this.backendName).forEach(t=>{t.setupFunc!=null&&t.setupFunc(this.backendInstance)})}disposeRegisteredKernels(e){zi(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 E3)&&typeof n.then=="function"){let r=++this.pendingBackendInitId,s=n.then(a=>r<this.pendingBackendInitId?!1:(this.registry[e]=a,this.pendingBackendInit=null,!0)).catch(a=>(r<this.pendingBackendInitId||(this.pendingBackendInit=null,console.warn(`Initialization of backend ${e} failed`),console.warn(a.stack||a.message)),!1));return this.pendingBackendInit=s,{success:s,asyncInit:!0}}else return this.registry[e]=n,{success:!0,asyncInit:!1}}catch(n){return console.warn(`Initialization of backend ${e} failed`),console.warn(n.stack||n.message),{success:!1,asyncInit:!1}}}removeBackend(e){if(!(e in this.registryFactory))throw new Error(`${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:r,asyncInit:s}=this.initializeBackend(n);if(s||r)return{name:n,asyncInit:s}}throw new Error("Could not initialize any backends, all backend initializations failed.")}moveData(e,t){let n=this.state.tensorInfo.get(t),r=n.backend,s=this.readSync(t),a=r.refCount(t);r.disposeData(t,!0),n.backend=e,e.move(t,s,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 r;return this.scopedRun(()=>this.startScope(n),()=>this.endScope(r),()=>(r=t(),r instanceof Promise&&console.error("Cannot return a Promise inside of tidy."),r))}scopedRun(e,t,n){e();try{let r=n();return t(),r}catch(r){throw t(),r}}nextTensorId(){return w2.nextTensorId++}nextVariableId(){return w2.nextVariableId++}clone(e){let t=U.runKernel(a2,{x:e}),n={x:e},r=a=>({x:()=>{let o="float32",i={x:a},l={dtype:o};return U.runKernel(s2,i,l)}}),s=[];return this.addTapeNode(this.state.activeScope.name,n,[t],r,s,{}),t}runKernel(e,t,n){if(this.backendName==null&&this.backend,!(np(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 r=this.backend.numDataIds(),s=0;n.forEach(i=>{s+=i.dtype==="complex64"?3:1});let a=this.state.numDataMovesStack[this.state.numDataMovesStack.length-1],o=r-t-s-a;if(o>0)throw new Error(`Backend '${this.backendName}' has an internal memory leak (${o} data ids) after running '${e}'`)}runKernelFunc(e){let t,n=[],r=this.isTapeOn(),s=this.state.numBytes,a=this.state.numTensors;this.shouldCheckForMemLeaks()&&this.state.numDataMovesStack.push(0);let o;this.backendName==null&&this.backend;let i,l=v2(e)?e.kernelName:this.state.activeScope!=null?this.state.activeScope.name:"";if(v2(e)){let{kernelName:p,inputs:f,attrs:m}=e;this.backendName==null&&this.backend;let g=np(p,this.backendName);L(g!=null,()=>`Cannot find registered kernel '${p}' for backend '${this.backendName}'`),o=()=>{let y=this.backend.numDataIds();i=g.kernelFunc({inputs:f,attrs:m,backend:this.backend});let A=Array.isArray(i)?i:[i];this.shouldCheckForMemLeaks()&&this.checkKernelForMemLeak(p,y,A);let x=A.map(b=>{if(b.rank!=null)return b;let{dataId:v,shape:I,dtype:w}=b;return this.makeTensorFromDataId(v,I,w)});if(r){let b=this.getTensorsForGradient(p,f,x);n=this.saveTensorsForBackwardMode(b)}return x}}else{let{forwardFunc:p}=e,f=m=>{!r||(n=m.map(g=>this.keep(this.clone(g))))};o=()=>{let m=this.backend.numDataIds();i=this.tidy(()=>p(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,d=v2(e)?null:e.backwardsFunc,h;return this.scopedRun(()=>this.state.kernelDepth++,()=>this.state.kernelDepth--,()=>{!this.ENV.getBool("DEBUG")&&!this.state.profiling?t=o():(h=this.profiler.profileKernel(l,u,()=>o()),this.ENV.getBool("DEBUG")&&this.profiler.logKernelProfile(h),t=h.outputs)}),r&&this.addTapeNode(l,u,t,d,n,c),this.state.profiling&&this.state.activeProfile.kernels.push({name:l,bytesAdded:this.state.numBytes-s,totalBytesSnapshot:this.state.numBytes,tensorsAdded:this.state.numTensors-a,totalTensorsSnapshot:this.state.numTensors,inputShapes:Object.keys(u).map(p=>u[p]!=null?u[p].shape:null),outputShapes:t.map(p=>p.shape),kernelTimeMs:h.timeMs,extraInfo:h.extraInfo}),Array.isArray(i)?t:t[0]}saveTensorsForBackwardMode(e){return e.map(n=>this.keep(this.clone(n)))}getTensorsForGradient(e,t,n){let r=d2(e);if(r!=null){let s=r.inputsToSave||[],a=r.outputsToSave||[],o;r.saveAllInputs?(L(Array.isArray(t),()=>"saveAllInputs is true, expected inputs to be an array."),o=Object.keys(t).map(l=>t[l])):o=s.map(l=>t[l]);let i=n.filter((l,u)=>a[u]);return o.concat(i)}return[]}makeTensor(e,t,n,r){if(e==null)throw new Error("Values passed to engine.makeTensor() are null");n=n||"float32",r=r||this.backend;let s=e;n==="string"&&Sa(e[0])&&(s=e.map(i=>Ju(i)));let a=r.write(s,t,n),o=new It(t,n,a,this.nextTensorId());if(this.trackTensor(o,r),n==="string"){let i=this.state.tensorInfo.get(a),l=O3(s);this.state.numBytes+=l-i.bytes,i.bytes=l}return o}makeTensorFromDataId(e,t,n,r){n=n||"float32";let s=new It(t,n,e,this.nextTensorId());return this.trackTensor(s,r),s}makeVariable(e,t=!0,n,r){n=n||this.nextVariableId().toString(),r!=null&&r!==e.dtype&&(e=e.cast(r));let s=new nc(e,t,n,this.nextTensorId());if(this.state.registeredVariables[s.name]!=null)throw new Error(`Variable with name ${s.name} was already registered`);return this.state.registeredVariables[s.name]=s,this.incRef(s,this.backend),s}trackTensor(e,t){this.state.numTensors++,e.dtype==="string"&&this.state.numStringTensors++;let n=0;e.dtype!=="complex64"&&e.dtype!=="string"&&(n=e.size*Yg(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 nc||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*Yg(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(r=>r.totalBytesSnapshot)),this.state.activeProfile.newBytes=this.state.numBytes-t,this.state.activeProfile.newTensors=this.state.numTensors-n;for(let r of this.state.activeProfile.kernels)r.kernelTimeMs=await r.kernelTimeMs,r.extraInfo=await r.extraInfo;return this.state.activeProfile}isTapeOn(){return this.state.gradientDepth>0&&this.state.kernelDepth===0}addTapeNode(e,t,n,r,s,a){let o={id:this.state.nextTapeNodeId++,kernelName:e,inputs:t,outputs:n,saved:s},i=d2(e);i!=null&&(r=i.gradFunc),r!=null&&(o.gradient=l=>(l=l.map((u,c)=>{if(u==null){let d=n[c],h=tp(d.size,d.dtype);return this.makeTensor(h,d.shape,d.dtype)}return u}),r(l.length>1?l:l[0],s,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=b2(e),n=new Set(t.map(s=>s.id));for(let s=0;s<this.state.activeScope.track.length;s++){let a=this.state.activeScope.track[s];!a.kept&&!n.has(a.id)&&a.dispose()}let r=this.state.scopeStack.pop();this.state.activeScope=this.state.scopeStack.length===0?null:this.state.scopeStack[this.state.scopeStack.length-1],t.forEach(s=>{!s.kept&&s.scopeId===r.id&&this.track(s)})}gradients(e,t,n,r=!1){if(L(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 s=this.scopedRun(()=>this.startTape(),()=>this.endTape(),()=>this.tidy("forward",e));L(s instanceof It,()=>"The result y returned by f() must be a tensor.");let a=TD(this.state.activeTape,t,s);if(!r&&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[s.id]=n==null?LD(s.shape):n,ND(o,a,l=>this.tidy(l),BD);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:s,grads:i}})}customGrad(e){return L(Ta(e),()=>"The f passed in customGrad(f) must be a function."),(...t)=>{L(t.every(o=>o instanceof It),()=>"The args passed in customGrad(f)(x1, x2,...) must all be tensors");let n,r={};t.forEach((o,i)=>{r[i]=o});let s=(o,i)=>(n=e(...t,i),L(n.value instanceof It,()=>"The function f passed in customGrad(f) must return an object where `obj.value` is a tensor"),L(Ta(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];L(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(...)."),L(u.every(d=>d instanceof It),()=>"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((d,h)=>{c[h]=()=>d}),c};return this.runKernelFunc({forwardFunc:s,backwardsFunc:a,inputs:r})}}readSync(e){return this.state.tensorInfo.get(e).backend.readSync(e)}read(e){return this.state.tensorInfo.get(e).backend.read(e)}async time(e){let t=Yu(),n=await this.backend.time(e);return n.wallMs=Yu()-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 F7;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}},k2=w2;k2.nextTensorId=0;k2.nextVariableId=0;function LD(e){let t=Jg(Yt(e),"float32");return U.makeTensor(t,e,"float32")}function M7(){let e=V3();if(e._tfengine==null){let t=new W3(e);e._tfengine=new k2(t)}return K_(e._tfengine.ENV),RD(()=>e._tfengine),e._tfengine}var U=M7();function BD(e,t){let n={a:e,b:t};return U.runKernel(r2,n)}var O7={};_e(O7,{isBrowser:()=>P7,isMobile:()=>VD});function WD(){return typeof navigator!="undefined"&&navigator!=null}function VD(e){if(e||WD()){if(e||(e=navigator),e.product==="ReactNative")return!0;let t=e.userAgent||e.vendor||window.opera;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 P7(){return typeof window!="undefined"&&window.document!=null||typeof WorkerGlobalScope!="undefined"}var ts=ct();ts.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.")});ts.registerFlag("IS_BROWSER",()=>P7());ts.registerFlag("IS_NODE",()=>typeof process!="undefined"&&typeof process.versions!="undefined"&&typeof process.versions.node!="undefined");ts.registerFlag("IS_CHROME",()=>typeof navigator!="undefined"&&navigator!=null&&navigator.userAgent!=null&&/Chrome/.test(navigator.userAgent)&&/Google Inc/.test(navigator.vendor));ts.registerFlag("PROD",()=>!1);ts.registerFlag("TENSORLIKE_CHECK_SHAPE_CONSISTENCY",()=>ts.getBool("DEBUG"));ts.registerFlag("DEPRECATION_WARNINGS_ENABLED",()=>!0);ts.registerFlag("IS_TEST",()=>!1);ts.registerFlag("CHECK_COMPUTATION_FOR_ERRORS",()=>!0);ts.registerFlag("WRAP_TO_IMAGEBITMAP",()=>!1);function ks(e,t){let n=e;if(Nn(e))return t==="string"?[]:[e.length];if(!Array.isArray(e))return[];let r=[];for(;Array.isArray(n)||Nn(n)&&t!=="string";)r.push(n.length),n=n[0];return Array.isArray(e)&&ct().getBool("TENSORLIKE_CHECK_SHAPE_CONSISTENCY")&&z7(e,r,[]),r}function z7(e,t,n){if(n=n||[],!Array.isArray(e)&&!Nn(e)){L(t.length===0,()=>`Element arr[${n.join("][")}] is a primitive, but should be an array/TypedArray of ${t[0]} elements`);return}L(t.length>0,()=>`Element arr[${n.join("][")}] should be a primitive, but is an array of ${e.length} elements`),L(e.length===t[0],()=>`Element arr[${n.join("][")}] should have ${t[0]} elements, but has ${e.length} elements`);let r=t.slice(1);for(let s=0;s<e.length;++s)z7(e[s],r,n.concat(s))}function L7(e,t,n,r){if(e!=="string_or_numeric"){if(e==null)throw new Error("Expected dtype cannot be null.");if(e!=="numeric"&&e!==t||e==="numeric"&&t==="string")throw new Error(`Argument '${n}' passed to '${r}' must be ${e} tensor, but got ${t} tensor`)}}function P(e,t,n,r="numeric"){if(e instanceof It)return L7(r,e.dtype,t,n),e;let s=Qh(e);if(s!=="string"&&["bool","int32","float32"].indexOf(r)>=0&&(s=r),L7(r,s,t,n),e==null||!Nn(e)&&!Array.isArray(e)&&typeof e!="number"&&typeof e!="boolean"&&typeof e!="string"){let l=e==null?"null":e.constructor.name;throw new Error(`Argument '${t}' passed to '${n}' must be a Tensor or TensorLike, but got '${l}'`)}let a=ks(e,s);!Nn(e)&&!Array.isArray(e)&&(e=[e]);let i=s!=="string"?ap(e,s):po(e,[],!0);return U.makeTensor(i,a,s)}function rc(e,t,n,r="numeric"){if(!Array.isArray(e))throw new Error(`Argument ${t} passed to ${n} must be a \`Tensor[]\` or \`TensorLike[]\``);return e.map((a,o)=>P(a,`${t}[${o}]`,n,r))}var B7="__op";function H(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],r=e[n];n.endsWith("_")&&(n=n.substring(0,n.length-1)),n=n+B7;let s=(...a)=>{U.startScope(n);try{let o=r(...a);return e2(o)&&console.error("Cannot return a Promise inside of tidy."),U.endScope(o),o}catch(o){throw U.endScope(null),o}};return Object.defineProperty(s,"name",{value:n,configurable:!0}),s}function UD(e,t){let n=P(e,"real","complex"),r=P(t,"imag","complex");Mn(n.shape,r.shape,`real and imag shapes, ${n.shape} and ${r.shape}, must match in call to tf.complex().`);let s={real:n,imag:r};return U.runKernel(uv,s)}var go=H({complex_:UD});function Ca(e,t,n,r){if(r==null&&(r=Qh(e)),r==="complex64")throw new Error("Cannot construct a complex64 tensor directly. Please use tf.complex(real, imag).");if(!Nn(e)&&!Array.isArray(e)&&typeof e!="number"&&typeof e!="boolean"&&typeof e!="string")throw new Error("values passed to tensor(values) must be a number/boolean/string or an array of numbers/booleans/strings, or a TypedArray");if(t!=null){Qg(t);let s=Yt(t),a=Yt(n);L(s===a,()=>`Based on the provided shape, [${t}], the tensor should have ${s} values but has ${a}`);for(let o=0;o<n.length;++o){let i=n[o],l=o===n.length-1?i!==Yt(t.slice(o)):!0;L(n[o]===t[o]||!l,()=>`Error creating a new Tensor. Inferred shape (${n}) does not match the provided shape (${t}). `)}}return!Nn(e)&&!Array.isArray(e)&&(e=[e]),t=t||n,e=r!=="string"?ap(e,r):po(e,[],!0),U.makeTensor(e,t,r)}function ns(e,t,n){let r=ks(e,n);return Ca(e,t,r,n)}var I2={float32:4,float16:2,int32:4,uint16:2,uint8:1,bool:1,complex64:8},cp=4;async function HD(e,t){let n=[],r=[],s=Array.isArray(e)?e.map(o=>o.name):Object.keys(e);for(let o=0;o<s.length;++o){let i=s[o],l=Array.isArray(e)?e[o].tensor:e[i];if(l.dtype!=="float32"&&l.dtype!=="int32"&&l.dtype!=="bool"&&l.dtype!=="string"&&l.dtype!=="complex64")throw new Error(`Unsupported dtype in weight '${i}': ${l.dtype}`);let u={name:i,shape:l.shape,dtype:l.dtype};if(l.dtype==="string"){let c=new Promise(async d=>{let h=await l.bytes(),p=h.reduce((g,y)=>g+y.length,0)+cp*h.length,f=new Uint8Array(p),m=0;for(let g=0;g<h.length;g++){let y=h[g],A=new Uint8Array(new Uint32Array([y.length]).buffer);f.set(A,m),m+=cp,f.set(y,m),m+=y.length}d(f)});r.push(c)}else r.push(l.data());t!=null&&(u.group=t),n.push(u)}let a=await Promise.all(r);return{data:GD(a),specs:n}}function W7(e,t){let n={},r,s=0;for(let a of t){let o=a.name,i=a.dtype,l=a.shape,u=Yt(l),c;if("quantization"in a){let d=a.quantization;if(d.dtype==="uint8"||d.dtype==="uint16"){if(!("min"in d&&"scale"in d))throw new Error(`Weight ${a.name} with quantization ${d.dtype} doesn't have corresponding metadata min and scale.`)}else if(d.dtype==="float16"){if(i!=="float32")throw new Error(`Weight ${a.name} is quantized with ${d.dtype} which only supports weights of type float32 not ${i}.`)}else throw new Error(`Weight ${a.name} has unknown quantization dtype ${d.dtype}. Supported quantization dtypes are: 'uint8', 'uint16', and 'float16'.`);let h=I2[d.dtype],p=e.slice(s,s+u*h),f=d.dtype==="uint8"?new Uint8Array(p):new Uint16Array(p);if(i==="float32")if(d.dtype==="uint8"||d.dtype==="uint16"){c=new Float32Array(f.length);for(let m=0;m<f.length;m++){let g=f[m];c[m]=g*d.scale+d.min}}else if(d.dtype==="float16")r===void 0&&(r=YD()),c=r(f);else throw new Error(`Unsupported quantization type ${d.dtype} for weight type float32.`);else if(i==="int32"){if(d.dtype!=="uint8"&&d.dtype!=="uint16")throw new Error(`Unsupported quantization type ${d.dtype} for weight type int32.`);c=new Int32Array(f.length);for(let m=0;m<f.length;m++){let g=f[m];c[m]=Math.round(g*d.scale+d.min)}}else throw new Error(`Unsupported dtype in weight '${o}': ${i}`);s+=u*h}else if(i==="string"){let d=Yt(a.shape);c=[];for(let h=0;h<d;h++){let p=new Uint32Array(e.slice(s,s+cp))[0];s+=cp;let f=new Uint8Array(e.slice(s,s+p));c.push(f),s+=p}}else{let d=I2[i],h=e.slice(s,s+u*d);if(i==="float32")c=new Float32Array(h);else if(i==="int32")c=new Int32Array(h);else if(i==="bool")c=new Uint8Array(h);else if(i==="complex64"){c=new Float32Array(h);let p=new Float32Array(c.length/2),f=new Float32Array(c.length/2);for(let y=0;y<p.length;y++)p[y]=c[y*2],f[y]=c[y*2+1];let m=ns(p,l,"float32"),g=ns(f,l,"float32");n[o]=go(m,g),m.dispose(),g.dispose()}else throw new Error(`Unsupported dtype in weight '${o}': ${i}`);s+=u*d}i!=="complex64"&&(n[o]=ns(c,l,i))}return n}function GD(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 r=new Uint8Array(t),s=0;return n.forEach(a=>{r.set(new Uint8Array(a.buffer),s),s+=a.byteLength}),r.buffer}var S2=typeof Buffer!="undefined"&&(typeof Blob=="undefined"||typeof atob=="undefined"||typeof btoa=="undefined");function V7(e){return S2?Buffer.byteLength(e):new Blob([e]).size}function jD(e){if(S2)return Buffer.from(e).toString("base64");let t=new Uint8Array(e),n="";for(let r=0,s=t.length;r<s;r++)n+=String.fromCharCode(t[r]);return btoa(n)}function qD(e){if(S2){let r=Buffer.from(e,"base64");return r.buffer.slice(r.byteOffset,r.byteOffset+r.byteLength)}let t=atob(e),n=new Uint8Array(t.length);for(let r=0;r<t.length;++r)n.set([t.charCodeAt(r)],r);return n.buffer}function T2(e){if(e.length===1)return e[0];let t=0;e.forEach(s=>{t+=s.byteLength});let n=new Uint8Array(t),r=0;return e.forEach(s=>{n.set(new Uint8Array(s),r),r+=s.byteLength}),n.buffer}function U7(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 H7(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 N2(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[r,s]=await t(e.weightsManifest);n.weightSpecs=r,n.weightData=s}return e.signature!=null&&(n.signature=e.signature),e.userDefinedMetadata!=null&&(n.userDefinedMetadata=e.userDefinedMetadata),e.modelInitializer!=null&&(n.modelInitializer=e.modelInitializer),n}function sc(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:V7(JSON.stringify(e.modelTopology)),weightSpecsBytes:e.weightSpecs==null?0:V7(JSON.stringify(e.weightSpecs)),weightDataBytes:e.weightData==null?0:e.weightData.byteLength}}function KD(){let e=n=>{let r=n<<13,s=0;for(;(r&8388608)==0;)s-=8388608,r<<=1;return r&=~8388608,s+=947912704,r|s},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 XD(){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 ZD(){let e=new Uint32Array(64);for(let t=0;t<64;t++)e[t]=1024;return e[0]=e[32]=0,e}function YD(){let e=KD(),t=XD(),n=ZD();return r=>{let s=new ArrayBuffer(4*r.length),a=new Uint32Array(s);for(let o=0;o<r.length;o++){let i=r[o],l=e[n[i>>10]+(i&1023)]+t[i>>10];a[o]=l}return new Float32Array(s)}}var Gt=class{constructor(){this.saveRouters=[],this.loadRouters=[]}static getInstance(){return Gt.instance==null&&(Gt.instance=new Gt),Gt.instance}static registerSaveRouter(e){Gt.getInstance().saveRouters.push(e)}static registerLoadRouter(e){Gt.getInstance().loadRouters.push(e)}static getSaveHandlers(e){return Gt.getHandlers(e,"save")}static getLoadHandlers(e,t){return Gt.getHandlers(e,"load",t)}static getHandlers(e,t,n){let r=[];return(t==="load"?Gt.getInstance().loadRouters:Gt.getInstance().saveRouters).forEach(a=>{let o=a(e,n);o!==null&&r.push(o)}),r}},JD=e=>Gt.registerSaveRouter(e),QD=e=>Gt.registerLoadRouter(e),eF=e=>Gt.getSaveHandlers(e),tF=(e,t)=>Gt.getLoadHandlers(e,t),C2="tensorflowjs",E2=1,yo="models_store",Ea="model_info_store";function G7(){if(!ct().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 $2(e){let t=e.result;t.createObjectStore(yo,{keyPath:"modelPath"}),t.createObjectStore(Ea,{keyPath:"modelPath"})}var Ao=class{constructor(e){if(this.indexedDB=G7(),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,r)=>{let s=this.indexedDB.open(C2,E2);s.onupgradeneeded=()=>$2(s),s.onsuccess=()=>{let a=s.result;if(t==null){let o=a.transaction(yo,"readonly"),l=o.objectStore(yo).get(this.modelPath);l.onsuccess=()=>{if(l.result==null)return a.close(),r(new Error(`Cannot find model with path '${this.modelPath}' in IndexedDB.`));n(l.result.modelArtifacts)},l.onerror=u=>(a.close(),r(l.error)),o.oncomplete=()=>a.close()}else{let o=sc(t),i=a.transaction(Ea,"readwrite"),l=i.objectStore(Ea),u=l.put({modelPath:this.modelPath,modelArtifactsInfo:o}),c;u.onsuccess=()=>{c=a.transaction(yo,"readwrite");let h=c.objectStore(yo).put({modelPath:this.modelPath,modelArtifacts:t,modelArtifactsInfo:o});h.onsuccess=()=>n({modelArtifactsInfo:o}),h.onerror=p=>{l=i.objectStore(Ea);let f=l.delete(this.modelPath);f.onsuccess=()=>(a.close(),r(h.error)),f.onerror=m=>(a.close(),r(h.error))}},u.onerror=d=>(a.close(),r(u.error)),i.oncomplete=()=>{c==null?a.close():c.oncomplete=()=>a.close()}}},s.onerror=a=>r(s.error)})}};Ao.URL_SCHEME="indexeddb://";var j7=e=>ct().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(Ao.URL_SCHEME)?nF(e.slice(Ao.URL_SCHEME.length)):null;Gt.registerSaveRouter(j7);Gt.registerLoadRouter(j7);function nF(e){return new Ao(e)}function rF(e){return e.startsWith(Ao.URL_SCHEME)?e.slice(Ao.URL_SCHEME.length):e}var sF=class{constructor(){this.indexedDB=G7()}async listModels(){return new Promise((e,t)=>{let n=this.indexedDB.open(C2,E2);n.onupgradeneeded=()=>$2(n),n.onsuccess=()=>{let r=n.result,s=r.transaction(Ea,"readonly"),o=s.objectStore(Ea).getAll();o.onsuccess=()=>{let i={};for(let l of o.result)i[l.modelPath]=l.modelArtifactsInfo;e(i)},o.onerror=i=>(r.close(),t(o.error)),s.oncomplete=()=>r.close()},n.onerror=r=>t(n.error)})}async removeModel(e){return e=rF(e),new Promise((t,n)=>{let r=this.indexedDB.open(C2,E2);r.onupgradeneeded=()=>$2(r),r.onsuccess=()=>{let s=r.result,a=s.transaction(Ea,"readwrite"),o=a.objectStore(Ea),i=o.get(e),l;i.onsuccess=()=>{if(i.result==null)return s.close(),n(new Error(`Cannot find model with path '${e}' in IndexedDB.`));{let u=o.delete(e),c=()=>{l=s.transaction(yo,"readwrite");let h=l.objectStore(yo).delete(e);h.onsuccess=()=>t(i.result.modelArtifactsInfo),h.onerror=p=>n(i.error)};u.onsuccess=c,u.onerror=d=>(c(),s.close(),n(i.error))}},i.onerror=u=>(s.close(),n(i.error)),a.oncomplete=()=>{l==null?s.close():l.oncomplete=()=>s.close()}},r.onerror=s=>n(r.error)})}},Ys="/",Bi="tensorflowjs_models",q7="info",aF="model_topology",oF="weight_specs",iF="weight_data",lF="model_metadata";function K7(e){return{info:[Bi,e,q7].join(Ys),topology:[Bi,e,aF].join(Ys),weightSpecs:[Bi,e,oF].join(Ys),weightData:[Bi,e,iF].join(Ys),modelMetadata:[Bi,e,lF].join(Ys)}}function X7(e){for(let t of Object.values(e))window.localStorage.removeItem(t)}function uF(e){let t=e.split(Ys);if(t.length<3)throw new Error(`Invalid key format: ${e}`);return t.slice(1,t.length-1).join(Ys)}function cF(e){return e.startsWith(xo.URL_SCHEME)?e.slice(xo.URL_SCHEME.length):e}var xo=class{constructor(e){if(!ct().getBool("IS_BROWSER")||typeof window=="undefined"||typeof window.localStorage=="undefined")throw new Error("The current environment does not support local storage.");if(this.LS=window.localStorage,e==null||!e)throw new Error("For local storage, modelPath must not be null, undefined or empty.");this.modelPath=e,this.keys=K7(this.modelPath)}async save(e){if(e.modelTopology instanceof ArrayBuffer)throw new Error("BrowserLocalStorage.save() does not support saving model topology in binary formats yet.");{let t=JSON.stringify(e.modelTopology),n=JSON.stringify(e.weightSpecs),r=sc(e);try{this.LS.setItem(this.keys.info,JSON.stringify(r)),this.LS.setItem(this.keys.topology,t),this.LS.setItem(this.keys.weightSpecs,n),this.LS.setItem(this.keys.weightData,jD(e.weightData));let s={format:e.format,generatedBy:e.generatedBy,convertedBy:e.convertedBy,signature:e.signature!=null?e.signature:void 0,userDefinedMetadata:e.userDefinedMetadata!=null?e.userDefinedMetadata:void 0,modelInitializer:e.modelInitializer!=null?e.modelInitializer:void 0,trainingConfig:e.trainingConfig!=null?e.trainingConfig:void 0};return this.LS.setItem(this.keys.modelMetadata,JSON.stringify(s)),{modelArtifactsInfo:r}}catch(s){throw X7(this.keys),new Error(`Failed to save model '${this.modelPath}' to local storage: size quota being exceeded is a possible cause of this failure: modelTopologyBytes=${r.modelTopologyBytes}, weightSpecsBytes=${r.weightSpecsBytes}, weightDataBytes=${r.weightDataBytes}.`)}}}async load(){let e=JSON.parse(this.LS.getItem(this.keys.info));if(e==null)throw new Error(`In local storage, there is no model with name '${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 '${this.modelPath}' is missing.`);t.modelTopology=n;let r=JSON.parse(this.LS.getItem(this.keys.weightSpecs));if(r==null)throw new Error(`In local storage, the weight specs of model '${this.modelPath}' are missing.`);t.weightSpecs=r;let s=this.LS.getItem(this.keys.modelMetadata);if(s!=null){let o=JSON.parse(s);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 '${this.modelPath}' are missing.`);return t.weightData=qD(a),t}};xo.URL_SCHEME="localstorage://";var Z7=e=>ct().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(xo.URL_SCHEME)?dF(e.slice(xo.URL_SCHEME.length)):null;Gt.registerSaveRouter(Z7);Gt.registerLoadRouter(Z7);function dF(e){return new xo(e)}var hF=class{constructor(){L(ct().getBool("IS_BROWSER"),()=>"Current environment is not a web browser"),L(typeof window=="undefined"||typeof window.localStorage!="undefined",()=>"Current browser does not appear to support localStorage"),this.LS=window.localStorage}async listModels(){let e={},t=Bi+Ys,n=Ys+q7;for(let r=0;r<this.LS.length;++r){let s=this.LS.key(r);if(s.startsWith(t)&&s.endsWith(n)){let a=uF(s);e[a]=JSON.parse(this.LS.getItem(s))}}return e}async removeModel(e){e=cF(e);let t=K7(e);if(this.LS.getItem(t.info)==null)throw new Error(`Cannot find model at path '${e}'`);let n=JSON.parse(this.LS.getItem(t.info));return X7(t),n}},Wi="://",Ir=class{constructor(){this.managers={}}static getInstance(){return Ir.instance==null&&(Ir.instance=new Ir),Ir.instance}static registerManager(e,t){L(e!=null,()=>"scheme must not be undefined or null."),e.endsWith(Wi)&&(e=e.slice(0,e.indexOf(Wi))),L(e.length>0,()=>"scheme must not be an empty string.");let n=Ir.getInstance();L(n.managers[e]==null,()=>`A model store manager is already registered for scheme '${e}'.`),n.managers[e]=t}static getManager(e){let t=this.getInstance().managers[e];if(t==null)throw new Error(`Cannot find model manager for scheme '${e}'`);return t}static getSchemes(){return Object.keys(this.getInstance().managers)}};function dp(e){if(e.indexOf(Wi)===-1)throw new Error(`The url string provided does not contain a scheme. Supported schemes are: ${Ir.getSchemes().join(",")}`);return{scheme:e.split(Wi)[0],path:e.split(Wi)[1]}}async function Y7(e,t,n=!1){L(e!==t,()=>`Old path and new path are the same: '${e}'`);let r=Gt.getLoadHandlers(e);L(r.length>0,()=>`Copying failed because no load handler is found for source URL ${e}.`),L(r.length<2,()=>`Copying failed because more than one (${r.length}) load handlers for source URL ${e}.`);let s=r[0],a=Gt.getSaveHandlers(t);L(a.length>0,()=>`Copying failed because no save handler is found for destination URL ${t}.`),L(a.length<2,()=>`Copying failed because more than one (${r.length}) save handlers for destination URL ${t}.`);let o=a[0],i=dp(e).scheme,l=dp(e).path,u=i===dp(e).scheme,c=await s.load();n&&u&&await Ir.getManager(i).removeModel(l);let d=await o.save(c);return n&&!u&&await Ir.getManager(i).removeModel(l),d.modelArtifactsInfo}async function pF(){let e=Ir.getSchemes(),t={};for(let n of e){let r=await Ir.getManager(n).listModels();for(let s in r){let a=n+Wi+s;t[a]=r[s]}}return t}async function fF(e){let t=dp(e);return Ir.getManager(t.scheme).removeModel(t.path)}async function mF(e,t){return Y7(e,t,!1)}async function gF(e,t){return Y7(e,t,!0)}var yF=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(ct().get("IS_BROWSER")){ct().setPlatform("browser",new yF);try{Ir.registerManager(xo.URL_SCHEME,new hF)}catch(e){}try{Ir.registerManager(Ao.URL_SCHEME,new sF)}catch(e){}}var AF={importFetch:()=>T3()},R2,xF=class{constructor(){this.util=co("util"),this.textEncoder=new this.util.TextEncoder}fetch(e,t){return ct().global.fetch!=null?ct().global.fetch(e,t):(R2==null&&(R2=AF.importFetch()),R2(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)}};ct().get("IS_NODE")&&ct().setPlatform("node",new xF);function Js(e,t="float32",n){return t=t||"float32",Qg(e),new lp(e,t,n)}function bF(e,t){let n=P(e,"x","cast");if(!M3(t))throw new Error(`Failed to cast to unknown dtype ${t}`);if(t==="string"&&n.dtype!=="string"||t!=="string"&&n.dtype==="string")throw new Error("Only strings can be casted to strings");let r={x:n},s={dtype:t};return U.runKernel(s2,r,s)}var Mt=H({cast_:bF});function vF(e){let n={x:P(e,"x","clone","string_or_numeric")};return U.runKernel(a2,n)}var Qs=H({clone_:vF});function J7(e,t=!1){console.log(e.toString(t))}M7();var wF={buffer:Js,cast:Mt,clone:Qs,print:J7};_D(wF);var Q7={};_e(Q7,{browserFiles:()=>EF,browserHTTPRequest:()=>FF,concatenateArrayBuffers:()=>T2,copyModel:()=>mF,decodeWeights:()=>W7,encodeWeights:()=>HD,fromMemory:()=>OF,getLoadHandlers:()=>tF,getModelArtifactsForJSON:()=>N2,getModelArtifactsInfoForJSON:()=>sc,getSaveHandlers:()=>eF,http:()=>M2,isHTTPScheme:()=>F2,listModels:()=>pF,loadWeights:()=>$F,moveModel:()=>gF,registerLoadRouter:()=>QD,registerSaveRouter:()=>JD,removeModel:()=>fF,weightsLoaderFactory:()=>rk,withSaveHandler:()=>PF});var kF="model",IF=".json",SF=".weights.bin";function ek(e){return new Promise(t=>setTimeout(t)).then(e)}var _2=class{constructor(e){if(!ct().getBool("IS_BROWSER"))throw new Error("browserDownloads() cannot proceed because the current environment is not a browser.");e.startsWith(_2.URL_SCHEME)&&(e=e.slice(_2.URL_SCHEME.length)),(e==null||e.length===0)&&(e=kF),this.modelJsonFileName=e+IF,this.weightDataFileName=e+SF}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}],r=H7(e,n),s=window.URL.createObjectURL(new Blob([JSON.stringify(r)],{type:"application/json"})),a=this.modelJsonAnchor==null?document.createElement("a"):this.modelJsonAnchor;if(a.download=this.modelJsonFileName,a.href=s,await ek(()=>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 ek(()=>o.dispatchEvent(new MouseEvent("click")))}return{modelArtifactsInfo:sc(e)}}}},hp=_2;hp.URL_SCHEME="downloads://";var TF=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=r=>{let s=JSON.parse(r.target.result),a=s.modelTopology;if(a==null){t(new Error(`modelTopology field is missing from file ${this.jsonFile.name}`));return}if(s.weightsManifest==null){t(new Error(`weightManifest field is missing from file ${this.jsonFile.name}`));return}if(this.weightsFiles.length===0){e({modelTopology:a});return}let i=N2(s,l=>this.loadWeights(l));e(i)},n.onerror=r=>t(`Failed to read model topology and weights manifest JSON from file '${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 r=this.checkManifestAndWeightFiles(e),s=n.map(a=>this.loadWeightsFile(a,r[a]));return Promise.all(s).then(a=>[t,T2(a)])}loadWeightsFile(e,t){return new Promise((n,r)=>{let s=new FileReader;s.onload=a=>{let o=a.target.result;n(o)},s.onerror=a=>r(`Failed to weights data from file of path '${e}'.`),s.readAsArrayBuffer(t)})}checkManifestAndWeightFiles(e){let t=[],n=this.weightsFiles.map(s=>U7(s.name)),r={};for(let s of e)s.paths.forEach(a=>{let o=U7(a);if(t.indexOf(o)!==-1)throw new Error(`Duplicate file basename found in weights manifest: '${o}'`);if(t.push(o),n.indexOf(o)===-1)throw new Error(`Weight file with basename '${o}' is not provided.`);r[a]=this.weightsFiles[n.indexOf(o)]});if(t.length!==this.weightsFiles.length)throw new Error(`Mismatch in the number of files in weights manifest (${t.length}) and the number of weight files provided (${this.weightsFiles.length}).`);return r}},NF=e=>ct().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(hp.URL_SCHEME)?CF(e.slice(hp.URL_SCHEME.length)):null;Gt.registerSaveRouter(NF);function CF(e="model"){return new hp(e)}function EF(e){return new TF(e)}function tk(e,t,n,r){o(e),n=n==null?0:n,r=r==null?1:r,i(n,r);let s=0,a=l=>(l.then(u=>{let c=n+ ++s/e.length*(r-n);return t(c),u}),l);function o(l){L(l!=null&&Array.isArray(l)&&l.length>0,()=>"promises must be a none empty array")}function i(l,u){L(l>=0&&l<=1,()=>`Progress fraction must be in range [0, 1], but got startFraction ${l}`),L(u>=0&&u<=1,()=>`Progress fraction must be in range [0, 1], but got endFraction ${u}`),L(u>=l,()=>`startFraction must be no more than endFraction, but got startFraction ${l} and endFraction ${u}`)}return Promise.all(e.map(a))}async function nk(e,t){t==null&&(t={});let n=t.fetchFunc==null?ct().platform.fetch:t.fetchFunc,r=e.map(d=>n(d,t.requestInit,{isBinary:!0})),s=0,a=.5,i=(t.onProgress==null?await Promise.all(r):await tk(r,t.onProgress,s,a)).map(d=>d.arrayBuffer()),l=.5,u=1;return t.onProgress==null?await Promise.all(i):await tk(i,t.onProgress,l,u)}async function $F(e,t="",n,r){return rk(o=>nk(o,{requestInit:r}))(e,t,n)}function rk(e){return async(t,n="",r)=>{let s=t.map(()=>!1),a={},o=r!=null?r.map(()=>!1):[],i=[];if(t.forEach((p,f)=>{let m=0;p.weights.forEach(g=>{let y="quantization"in g?g.quantization.dtype:g.dtype,A=I2[y]*Yt(g.shape),x=()=>{s[f]=!0,a[f]==null&&(a[f]=[]),a[f].push({manifestEntry:g,groupOffset:m,sizeBytes:A})};r!=null?r.forEach((b,v)=>{b===g.name&&(x(),o[v]=!0)}):x(),i.push(g.name),m+=A})}),!o.every(p=>p)){let p=r.filter((f,m)=>!o[m]);throw new Error(`Could not find weights in manifest with names: ${p.join(", ")}.
Manifest JSON has weights with names: ${i.join(", ")}.`)}let l=s.reduce((p,f,m)=>(f&&p.push(m),p),[]),u=[];l.forEach(p=>{t[p].paths.forEach(f=>{let m=n+(n.endsWith("/")?"":"/")+f;u.push(m)})});let c=await e(u),d={},h=0;return l.forEach(p=>{let f=t[p].paths.length,m=0;for(let b=0;b<f;b++)m+=c[h+b].byteLength;let g=new ArrayBuffer(m),y=new Uint8Array(g),A=0;for(let b=0;b<f;b++){let v=new Uint8Array(c[h+b]);y.set(v,A),A+=v.byteLength}a[p].forEach(b=>{let v=g.slice(b.groupOffset,b.groupOffset+b.sizeBytes),I=W7(v,[b.manifestEntry]);for(let w in I)d[w]=I[w]}),h+=f}),d}}var RF="application/octet-stream",_F="application/json",D2=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?(L(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=ct().platform.fetch,L(e!=null&&e.length>0,()=>"URL path for http must not be null, undefined or empty."),Array.isArray(e)&&L(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}],r=H7(e,n);t.body.append("model.json",new Blob([JSON.stringify(r)],{type:_F}),"model.json"),e.weightData!=null&&t.body.append("model.weights.bin",new Blob([e.weightData],{type:RF}),"model.weights.bin");let s=await this.fetch(this.path,t);if(s.ok)return{modelArtifactsInfo:sc(e),responses:[s]};throw new Error(`BrowserHTTPRequest.save() failed due to HTTP response status ${s.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(s){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,r=t.weightsManifest;if(n==null&&r==null)throw new Error(`The JSON from HTTP path ${this.path} contains neither model topology or manifest for weights.`);return N2(t,s=>this.loadWeights(s))}async loadWeights(e){let t=Array.isArray(this.path)?this.path[1]:this.path,[n,r]=DF(t),s=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(s+c+r);this.weightUrlConverter&&o.push(...await Promise.all(i));let l=await nk(o,{requestInit:this.requestInit,fetchFunc:this.fetch,onProgress:this.onProgress});return[a,T2(l)]}};D2.URL_SCHEME_REGEX=/^https?:\/\//;function DF(e){let t=e.lastIndexOf("/"),n=e.lastIndexOf("?"),r=e.substring(0,t),s=n>t?e.substring(n):"";return[r+"/",s]}function F2(e){return e.match(D2.URL_SCHEME_REGEX)!=null}var sk=(e,t)=>{if(typeof fetch=="undefined"&&(t==null||t.fetchFunc==null))return null;{let n=!0;if(Array.isArray(e)?n=e.every(r=>F2(r)):n=F2(e),n)return M2(e,t)}return null};Gt.registerSaveRouter(sk);Gt.registerLoadRouter(sk);function M2(e,t){return new D2(e,t)}function FF(e,t){return M2(e,t)}var O2=class{constructor(e){this.modelArtifacts=e}async load(){return this.modelArtifacts}},MF=class{constructor(e){this.saveHandler=e}async save(e){return this.saveHandler(e)}};function OF(e,t,n,r){return arguments.length===1?e.modelTopology!=null||e.weightSpecs!=null?new O2(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 O2({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 O2({modelTopology:e,weightSpecs:t,weightData:n,trainingConfig:r}))}function PF(e){return new MF(e)}var ak={};_e(ak,{confusionMatrix:()=>VF});function zF(e,t,n=!1,r=!1){let s=P(e,"a","matMul"),a=P(t,"b","matMul");[s,a]=Wt(s,a);let o={a:s,b:a},i={transposeA:n,transposeB:r};return U.runKernel(sv,o,i)}var gt=H({matMul_:zF});function LF(e,t,n=1,r=0){if(t<2)throw new Error(`Error in oneHot: depth must be >=2, but it is ${t}`);let a={indices:P(e,"indices","oneHot","int32")},o={depth:t,onValue:n,offValue:r};return U.runKernel(Sw,a,o)}var P2=H({oneHot_:LF});function BF(e,t){let n=P(e,"x","transpose");if(t==null&&(t=n.shape.map((a,o)=>o).reverse()),L(n.rank===t.length,()=>`Error in transpose: rank of input ${n.rank} must match length of perm ${t}.`),t.forEach(a=>{L(a>=0&&a<n.rank,()=>`All entries in 'perm' must be between 0 and ${n.rank-1} but got ${t}`)}),n.rank<=1)return n.clone();let r={x:n},s={perm:t};return U.runKernel(y7,r,s)}var pp=H({transpose_:BF});function WF(e,t,n){let r=P(e,"labels","confusionMatrix"),s=P(t,"predictions","confusionMatrix");L(n==null||n>0&&Number.isInteger(n),()=>`If provided, numClasses must be a positive integer, but got ${n}`),L(r.rank===1,()=>`Expected the rank of labels to be 1, but got ${r.rank}`),L(s.rank===1,()=>`Expected the rank of predictions to be 1, but got ${s.rank}`),L(r.shape[0]===s.shape[0],()=>`Mismatch in the number of examples: ${r.shape[0]} vs. ${s.shape[0]}. Labels and predictions should have the same number of elements.`),L(n>0&&Number.isInteger(n),()=>`numClasses is required to be a positive integer, but got ${n}`);let a=P2(Mt(r,"int32"),n),o=P2(Mt(s,"int32"),n),i=pp(a),l=gt(i,o);return Mt(l,"int32")}var VF=H({confusionMatrix_:WF}),Br={};_e(Br,{fromPixels:()=>XF,fromPixelsAsync:()=>qF,toPixels:()=>KF});function fp(e,t,n){if(ho(e),t!=null&&t.length!==3)throw new Error("tensor3d() requires shape to have three numbers");let r=ks(e,n);if(r.length!==3&&r.length!==1)throw new Error("tensor3d() requires values to be number[][][] or flat/TypedArray");if(r.length===1&&t==null)throw new Error("tensor3d() requires shape to be provided when `values` are a flat array");return Ca(e,t,r,n)}var Vi;function ok(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,r=!1,s=!1,a=!1,o=!1,i=!1;if(e.data instanceof Uint8Array)n=!0;else if(typeof ImageData!="undefined"&&e instanceof ImageData)r=!0;else if(typeof HTMLVideoElement!="undefined"&&e instanceof HTMLVideoElement)s=!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(s){let f=2;if(s&&e.readyState<f)throw new Error("The video element has not loaded data yet. Please wait for `loadeddata` event on the <video> element.")}if(np(i2,U.backendName)!=null){let f={pixels:e},m={numChannels:t};return U.runKernel(i2,f,m)}let[u,c]=s?[e.videoWidth,e.videoHeight]:[e.width,e.height],d;o?d=e.getContext("2d").getImageData(0,0,u,c).data:r||n?d=e.data:(a||s||i)&&(Vi==null&&(Vi=document.createElement("canvas").getContext("2d")),Vi.canvas.width=u,Vi.canvas.height=c,Vi.drawImage(e,0,0,u,c),d=Vi.getImageData(0,0,u,c).data);let h;if(t===4)h=new Int32Array(d);else{let f=u*c;h=new Int32Array(f*t);for(let m=0;m<f;m++)for(let g=0;g<t;++g)h[m*t+g]=d[m*4+g]}return fp(h,[c,u,t],"int32")}function UF(e){return e!=null&&e.data instanceof Uint8Array}function HF(){return typeof window!="undefined"&&typeof ImageBitmap!="undefined"&&window.hasOwnProperty("createImageBitmap")}function GF(e){return e!=null&&e.width!==0&&e.height!==0}function jF(e){return HF()&&!(e instanceof ImageBitmap)&&GF(e)&&!UF(e)}async function qF(e,t=3){let n=null;if(ct().getBool("WRAP_TO_IMAGEBITMAP")&&jF(e)){let r;try{r=await createImageBitmap(e,{premultiplyAlpha:"none"})}catch(s){r=null}r!=null&&r.width===e.width&&r.height===e.height?n=r:n=e}else n=e;return ok(n,t)}async function KF(e,t){let n=P(e,"img","toPixels");if(!(e instanceof It)){let u=n;n=Mt(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[r,s]=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(s*r*4);for(let u=0;u<r*s;++u){let c=[0,0,0,255];for(let h=0;h<a;h++){let p=o[u*a+h];if(n.dtype==="float32"){if(p<0||p>1)throw new Error(`Tensor values for a float32 Tensor must be in the range [0 - 1] but encountered ${p}.`)}else if(n.dtype==="int32"&&(p<0||p>255))throw new Error(`Tensor values for a int32 Tensor must be in the range [0 - 255] but encountered ${p}.`);a===1?(c[0]=p*i,c[1]=p*i,c[2]=p*i):c[h]=p*i}let d=u*4;l[d+0]=Math.round(c[0]),l[d+1]=Math.round(c[1]),l[d+2]=Math.round(c[2]),l[d+3]=Math.round(c[3])}if(t!=null){t.width=s,t.height=r;let u=t.getContext("2d"),c=new ImageData(l,s,r);u.putImageData(c,0,0)}return n!==e&&n.dispose(),l}var XF=H({fromPixels_:ok}),ik={};_e(ik,{prepareAndValidate:()=>lk});function lk(e,t){let n=e.shape.length,r=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(r<1)throw new Error(`tf.gatherND() expects the indices to be rank 1 or higher, but the rank was ${r}.`);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[r-1]>n)throw new Error(`index innermost dimension length must be <= tensor rank; saw: ${t.shape[r-1]} vs. ${n}`);if(Yt(e.shape)===0)throw new Error(`Requested more than 0 entries, but input is empty. Input shape: ${e.shape}.`);let s=t.shape,a=s[s.length-1],o=1;for(let d=0;d<s.length-1;++d)o*=s[d];let i=e.shape,l=s.slice();l.pop();let u=1;for(let d=a;d<n;++d)u*=i[d],l.push(i[d]);let c=[...Mi(e.shape).map(d=>d/u),1].slice(0,a);return[l,o,u,c]}var uk={};_e(uk,{calculateShapes:()=>ck,validateInput:()=>L2,validateUpdateShape:()=>z2});function z2(e,t,n){let r=t.rank>1?t.shape[t.rank-1]:1,s=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: ${r}, and batchDim: ${s}.`;if(n.rank<s)throw new Error(a+` update.rank < ${s}. `);if(e.length<r+(n.rank-s))throw new Error(a+` Output shape length < ${r+(n.rank-s)}`);if(n.rank!==s+e.length-r)throw new Error(a+` update.rank != ${s+e.length-r}`);for(let o=0;o<s;++o)if(n.shape[o]!==t.shape[o])throw new Error(a+` updates.shape[${o}] (${n.shape[o]}) != indices.shape[${o}] (${t.shape[o]}).`);for(let o=0;o<n.rank-s;++o)if(n.shape[o+s]!==e[o+r])throw new Error(a+` updates.shape[${o+s}] (${n.shape[o+s]}) != shape[${o+s}] (${e[o+s]})`)}function L2(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: ${e.shape}`)}z2(n,t,e)}function ck(e,t,n){let r=t.shape.length,s=r>1?t.shape[r-1]:1,a=n.length,o=1;for(let d=s;d<a;++d)o*=n[d];let i=s<1?1:s,l=Yt(t.shape)/i,u=[...Mi(n.slice(0,s)),1],c=Yt(n);return{sliceRank:s,numUpdates:l,sliceSize:o,strides:u,outputSize:c}}var B2={};_e(B2,{assertParamsValid:()=>ZF,computeFlatOffset:()=>JF,computeOutShape:()=>dk,getNormalizedAxes:()=>mk,isSliceContinous:()=>YF,maskToAxes:()=>mp,parseSliceParams:()=>QF,sliceInfo:()=>eM,startForAxis:()=>xk,startIndicesWithElidedDims:()=>gk,stopForAxis:()=>bk,stopIndicesWithElidedDims:()=>yk,stridesForAxis:()=>Ak,stridesWithElidedDims:()=>hk});function ZF(e,t,n){let r=e.shape.length;L(r===t.length,()=>`Error in slice${r}D: Length of begin ${t} must match the rank of the array (${r}).`),L(r===n.length,()=>`Error in slice${r}D: Length of size ${n} must match the rank of the array (${r}).`);for(let s=0;s<r;++s)L(t[s]+n[s]<=e.shape[s],()=>`Error in slice${r}D: begin[${s}] + size[${s}] (${t[s]+n[s]}) would overflow input.shape[${s}] (${e.shape[s]})`)}function mp(e){let t=[],n=0;for(;e>0;)e&1&&t.push(n),e/=2,n++;return t}function dk(e,t,n){let r=[];for(let s=0;s<e.length;s++)r[s]=Math.ceil((t[s]-e[s])/n[s]);return r}function hk(e,t,n,r){let s=[...e];for(let a=s.length;a<r.length;a++)s.push(1);for(let a=0;a<n;a++)a===0?s[t]=1:(s.splice(t,0,1),s.pop());return s}function pk(e,t,n){return n<=e?n:n-(t-1)}function fk(e,t){let n=[];for(let r=0;r<e;r++)n.push(t+r);return n}function mk(e,t,n,r,s,a,o,i,l){let u=e.length,c=new Array(u),d=new Array(u),h=new Array(u);if(t.length&&n>0){let p=t[0],f=n+1;c=gk(o,p,f,r,e),d=yk(i,p,f,s,e),h=hk(a,p,f,e)}else for(let p=0;p<u;p++)c[p]=xk(o,r,a,e,p,l),d[p]=bk(i,s,a,e,p,l),h[p]=Ak(a,p,l);return{begin:c,end:d,strides:h}}function gk(e,t,n,r,s){let a=[...s],o=fk(n,t);for(let i=0;i<a.length;i++)if(o.indexOf(i)>-1)a[i]=0;else{let l=pk(t,n,i),u=r[l];e&1<<l&&(u=0),a[i]=u}return a}function yk(e,t,n,r,s){let a=[...s],o=fk(n,t);for(let i=0;i<a.length;i++)if(o.indexOf(i)>-1)a[i]=Number.MAX_SAFE_INTEGER;else{let l=pk(t,n,i),u=r[l];e&1<<l&&(u=Number.MAX_SAFE_INTEGER),a[i]=u}for(let i=0;i<a.length;i++){let l=s[i];a[i]<0&&(a[i]+=l),a[i]=ju(0,a[i],s[i])}return a}function Ak(e,t,n){let r=e[t];return(n&1<<t||r==null)&&(r=1),r}function xk(e,t,n,r,s,a){let o=t[s],i=n[s]||1;(e&1<<s||a&1<<s||o==null)&&(i>0?o=Number.MIN_SAFE_INTEGER:o=Number.MAX_SAFE_INTEGER);let l=r[s];return o<0&&(o+=l),o=ju(0,o,l-1),o}function bk(e,t,n,r,s,a){let o=t[s],i=n[s]||1;(e&1<<s||a&1<<s||o==null)&&(i>0?o=Number.MAX_SAFE_INTEGER:o=Number.MIN_SAFE_INTEGER);let l=r[s];return o<0&&(o+=l),i>0?o=ju(0,o,l):o=ju(-1,o,l-1),o}function YF(e,t,n){let r=n.length;for(let s=0;s<n.length;s++)if(n[s]>1){r=s;break}for(let s=r+1;s<n.length;s++)if(t[s]>0||n[s]!==e[s])return!1;return!0}function JF(e,t){let n=e.length>0?e[e.length-1]:1;for(let r=0;r<e.length-1;r++)n+=e[r]*t[r];return n}function QF(e,t,n){let r,s=e.shape.length;typeof t=="number"?r=[t,...new Array(s-1).fill(0)]:t.length<s?r=t.concat(new Array(s-t.length).fill(0)):r=t.slice(),r.forEach(o=>{L(o!==-1,()=>"slice() does not support negative begin indexing.")});let a;return n==null?a=new Array(s).fill(-1):typeof n=="number"?a=[n,...new Array(s-1).fill(-1)]:n.length<s?a=n.concat(new Array(s-n.length).fill(-1)):a=n,a=a.map((o,i)=>o>=0?o:(L(o===-1,()=>`Negative size values should be exactly -1 but got ${o} for the slice() size at index ${i}.`),e.shape[i]-r[i])),[r,a]}function eM(e,t,n,r,s,a,o,i,l){let u=t.slice(),c=n.slice(),d=r;r==null&&(d=new Array(u.length));let h=mp(o);if(h.length>1)throw new Error("Multiple ellipses in slice is not allowed.");if(o!==0&&i!==0)throw new Error("Using both ellipsisMask and newAxisMask is not yet supported.");if(o!==0&&l!==0)throw new Error("Using both ellipsisMask and shrinkAxisMask is not yet supported.");let p=e.length-u.length,f=mp(i),m=e.slice();f.forEach(w=>{u[w]=0,c[w]=1,m.splice(w,0,1)});let{begin:g,end:y,strides:A}=mk(m,h,p,u,c,d,s,a,o);u=g,c=y,d=A;let x=mp(l);x.forEach(w=>{c[w]=u[w]+1,d[w]=1});let b=dk(u,c,d),v=b.filter((w,S)=>x.indexOf(S)===-1);return{nonStrided:d.every(w=>w===1),$begin:u,$end:c,$strides:d,size:b,newShape:m,outShape:v}}var vk={};_e(vk,{Serializable:()=>wk,SerializationMap:()=>bo,registerClass:()=>$a});var wk=class{getClassName(){return this.constructor.className}static fromConfig(e,t){return new e(t)}},bo=class{constructor(){this.classNameMap={}}static getMap(){return bo.instance==null&&(bo.instance=new bo),bo.instance}static register(e){bo.getMap().classNameMap[e.className]=[e,e.fromConfig]}};function $a(e){L(e.className!=null,()=>"Class being registered does not have the static className property defined."),L(typeof e.className=="string",()=>"className is required to be a string, but got type "+typeof e.className),L(e.className.length>0,()=>"Class being registered has an empty-string as its className, which is disallowed."),bo.register(e)}var kk={};_e(kk,{TEST_EPSILON_FLOAT16:()=>Ik,encodeStrings:()=>Sk,expectArrayBuffersEqual:()=>iM,expectArraysClose:()=>nM,expectArraysEqual:()=>sM,expectNumbersClose:()=>aM,expectPromiseToFail:()=>rM,expectValuesInRange:()=>oM,testEpsilon:()=>W2});var tM=.001,Ik=.1;function nM(e,t,n){return n==null&&(n=W2()),V2(e,t,(r,s)=>U2(r,s,n))}function W2(){return U.backend.floatPrecision()===32?tM:Ik}function V2(e,t,n){let r=!0;if((Nn(e)||Nn(t))&&(r=!1),Nn(e)&&Nn(t)&&(r=!0),r){let o=e.constructor.name,i=t.constructor.name;if(o!==i)throw new Error(`Arrays are of different type. Actual: ${o}. Expected: ${i}`)}if(Array.isArray(e)&&Array.isArray(t)){let o=ks(e),i=ks(t);if(!Zs(o,i))throw new Error(`Arrays have different shapes. Actual: [${o}]. Expected: [${i}]`)}let s=Nn(e)?e:po(e),a=Nn(t)?t:po(t);if(s.length!==a.length)throw new Error(`Arrays have different lengths actual: ${s.length} vs expected: ${a.length}.
Actual: ${s}.
Expected: ${a}.`);for(let o=0;o<a.length;++o){let i=s[o],l=a[o];if(!n(i,l))throw new Error(`Arrays differ: actual[${o}] = ${i}, expected[${o}] = ${l}.
Actual: ${s}.
Expected: ${a}.`)}}function rM(e,t){e().then(()=>t.fail(),()=>t())}function sM(e,t){let n=typeof t=="string"||typeof t=="number"||typeof t=="boolean"?[t]:t;return Sa(e)||Sa(e[0])||Sa(t)||Sa(t[0])?V2(e,n,(r,s)=>r==s):V2(e,t,(r,s)=>U2(r,s,0))}function aM(e,t,n){if(n==null&&(n=W2()),!U2(e,t,n))throw new Error(`Numbers differ: actual === ${e}, expected === ${t}`)}function U2(e,t,n){return!isFinite(e)&&!isFinite(t)?!0:!(isNaN(e)||isNaN(t)||Math.abs(e-t)>n)}function oM(e,t,n){for(let r=0;r<e.length;r++)if(e[r]<t||e[r]>n)throw new Error(`Value out of range:${e[r]} low: ${t}, high: ${n}`)}function iM(e,t){expect(new Float32Array(e)).toEqual(new Float32Array(t))}function Sk(e){for(let t=0;t<e.length;t++){let n=e[t];Array.isArray(n)?Sk(n):e[t]=Ju(n)}return e}var lM="3.8.0";function uM(){ct().set("PROD",!0)}function cM(){ct().set("DEBUG",!0)}function dM(){ct().set("DEPRECATION_WARNINGS_ENABLED",!1),console.warn("TensorFlow.js deprecation warnings have been disabled.")}function Tk(e){ct().getBool("DEPRECATION_WARNINGS_ENABLED")&&console.warn(e+" You can disable deprecation warnings with tf.disableDeprecationWarnings().")}DD(Tk);function hM(){U.disposeVariables()}function pM(){return U}function fM(){return U.memory()}function mM(e){return U.profile(e)}function Ve(e,t){return U.tidy(e,t)}function We(e){b2(e).forEach(n=>n.dispose())}function Nk(e){return U.keep(e)}function gM(e){return U.time(e)}function yM(e){return U.setBackend(e)}function AM(){return U.ready()}function xM(){return U.backendName}function bM(e){U.removeBackend(e)}function H2(e){return U.findBackend(e)}function vM(e){return U.findBackendFactory(e)}function G2(e,t,n=1){return U.registerBackend(e,t,n)}function wM(){return U.backend}function kM(e,t){ct().setPlatform(e,t)}function IM(e,t){let n=P(e,"a","add"),r=P(t,"b","add");[n,r]=Wt(n,r);let s={a:n,b:r};return U.runKernel(r2,s)}var Me=H({add_:IM});function SM(e,t){let n=P(e,"a","floorDiv"),r=P(t,"b","floorDiv");[n,r]=Wt(n,r);let s={a:n,b:r};return U.runKernel(Bv,s)}var Ck=H({floorDiv_:SM});function TM(e,t){let n=P(e,"a","div"),r=P(t,"b","div");if([n,r]=Wt(n,r),n.dtype==="int32"&&r.dtype==="int32")return Ck(n,r);let s={a:n,b:r},a={};return U.runKernel(Cv,s,a)}var Je=H({div_:TM});function NM(e,t){let n=P(e,"a","mul"),r=P(t,"b","mul");[n,r]=Wt(n,r);let s={a:n,b:r};return U.runKernel(Aw,s)}var pe=H({mul_:NM});function CM(e){let t=P(e,"x","abs");if(t.dtype==="complex64"){let n={x:t};return U.runKernel(cv,n)}else{let n={x:t};return U.runKernel(U3,n)}}var Sr=H({abs_:CM});function EM(e){let n={x:P(e,"x","acos")};return U.runKernel(H3,n)}var $M=H({acos_:EM});function RM(e){let n={x:P(e,"x","acosh")};return U.runKernel(G3,n)}var _M=H({acosh_:RM});function DM(e){L(Array.isArray(e),()=>"The argument passed to tf.addN() must be a list of tensors"),L(e.length>=1,()=>`Must pass at least one tensor to tf.addN(), but got ${e.length}`);let t=e.map((s,a)=>P(s,`tensors${a}`,"addN")),n=t[0];t.forEach(s=>{if(s.dtype!==n.dtype)throw new Error("All tensors passed to tf.addN() must have the same dtype")}),t.forEach(s=>{if(!Zs(s.shape,n.shape))throw new Error("All tensors passed to tf.addN() must have the same shape")});let r=t;return U.runKernel(j3,r)}var j2=H({addN_:DM});function FM(e,t=null,n=!1){let s={x:P(e,"x","all","bool")},a={axis:t,keepDims:n};return U.runKernel(q3,s,a)}var MM=H({all_:FM});function OM(e,t=null,n=!1){let s={x:P(e,"x","any","bool")},a={axis:t,keepDims:n};return U.runKernel(K3,s,a)}var PM=H({any_:OM});function zM(e,t=0){let r={x:P(e,"x","argMax")},s={axis:t};return U.runKernel(X3,r,s)}var q2=H({argMax_:zM});function LM(e,t=0){let r={x:P(e,"x","argMin")},s={axis:t};return U.runKernel(Z3,r,s)}var BM=H({argMin_:LM});function WM(e){let n={x:P(e,"x","asin")};return U.runKernel(Y3,n)}var VM=H({asin_:WM});function UM(e){let n={x:P(e,"x","asinh")};return U.runKernel(J3,n)}var HM=H({asinh_:UM});function GM(e){let n={x:P(e,"x","atan")};return U.runKernel(Q3,n)}var jM=H({atan_:GM});function qM(e,t){let n=P(e,"a","atan2"),r=P(t,"b","atan2");[n,r]=Wt(n,r);let s={a:n,b:r};return U.runKernel(tv,s)}var KM=H({atan2_:qM});function XM(e){let n={x:P(e,"x","atanh")};return U.runKernel(ev,n)}var ZM=H({atanh_:XM});function YM(e,t,n,r,s="NHWC",a){let o=e[3],i=[...t,o],l=Rk(s);return ac(e,i,n,a,r,null,null,l)}function Ek(e,t,n,r,s,a,o="channelsLast"){let[i,l]=gp(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 ac(e,u,n,r,s,a,!1,o)}function JM(e,t,n,r,s,a,o="NDHWC"){let[i,l,u]=X2(t),c,d;if(o==="NDHWC")d="channelsLast",c=[i,l,u,e[4],e[4]];else if(o==="NCDHW")d="channelsFirst",c=[i,l,u,e[1],e[1]];else throw new Error(`Unknown dataFormat ${o}`);return $k(e,c,n,r,s,!1,d,a)}function ac(e,t,n,r,s,a,o=!1,i="channelsLast"){let[l,u,c,d]=[-1,-1,-1,-1];if(i==="channelsLast")[l,u,c,d]=e;else if(i==="channelsFirst")[l,d,u,c]=e;else throw new Error(`Unknown dataFormat ${i}`);let[h,p,,f]=t,[m,g]=gp(n),[y,A]=gp(r),x=Ui(h,y),b=Ui(p,A),{padInfo:v,outHeight:I,outWidth:w}=tO(s,u,c,m,g,x,b,a,i),S=o?f*d:f,E;return i==="channelsFirst"?E=[l,S,I,w]:i==="channelsLast"&&(E=[l,I,w,S]),{batchSize:l,dataFormat:i,inHeight:u,inWidth:c,inChannels:d,outHeight:I,outWidth:w,outChannels:S,padInfo:v,strideHeight:m,strideWidth:g,filterHeight:h,filterWidth:p,effectiveFilterHeight:x,effectiveFilterWidth:b,dilationHeight:y,dilationWidth:A,inShape:e,outShape:E,filterShape:t}}function $k(e,t,n,r,s,a=!1,o="channelsLast",i){let[l,u,c,d,h]=[-1,-1,-1,-1,-1];if(o==="channelsLast")[l,u,c,d,h]=e;else if(o==="channelsFirst")[l,h,u,c,d]=e;else throw new Error(`Unknown dataFormat ${o}`);let[p,f,m,,g]=t,[y,A,x]=X2(n),[b,v,I]=X2(r),w=Ui(p,b),S=Ui(f,v),E=Ui(m,I),{padInfo:D,outDepth:$,outHeight:R,outWidth:N}=nO(s,u,c,d,y,A,x,w,S,E,i),M=a?g*h:g,B;return o==="channelsFirst"?B=[l,M,$,R,N]:o==="channelsLast"&&(B=[l,$,R,N,M]),{batchSize:l,dataFormat:o,inDepth:u,inHeight:c,inWidth:d,inChannels:h,outDepth:$,outHeight:R,outWidth:N,outChannels:M,padInfo:D,strideDepth:y,strideHeight:A,strideWidth:x,filterDepth:p,filterHeight:f,filterWidth:m,effectiveFilterDepth:w,effectiveFilterHeight:S,effectiveFilterWidth:E,dilationDepth:b,dilationHeight:v,dilationWidth:I,inShape:e,outShape:B,filterShape:t}}function QM(e,t,n,r,s){r==null&&(r=K2(e,t,n));let a=e[0],o=e[1],i=vo((a-t+2*r)/n+1,s),l=vo((o-t+2*r)/n+1,s);return[i,l]}function eO(e,t,n,r,s,a){s==null&&(s=K2(e,t,r));let o=e[0],i=e[1],l=e[2],u=vo((o-t+2*s)/r+1,a),c=vo((i-t+2*s)/r+1,a),d=vo((l-t+2*s)/r+1,a);return[u,c,d,n]}function K2(e,t,n,r=1){let s=Ui(t,r);return Math.floor((e[0]*(n-1)-n+s)/2)}function gp(e){return typeof e=="number"?[e,e,e]:e.length===2?[e[0],e[1],1]:e}function X2(e){return typeof e=="number"?[e,e,e]:e}function Ui(e,t){return t<=1?e:e+(e-1)*(t-1)}function tO(e,t,n,r,s,a,o,i,l){let u,c,d;if(typeof e=="number"){u={top:e,bottom:e,left:e,right:e,type:e===0?"VALID":"NUMBER"};let p=QM([t,n],a,r,e,i);c=p[0],d=p[1]}else if(e==="same"){c=Math.ceil(t/r),d=Math.ceil(n/s);let h=Math.max(0,(c-1)*r+a-t),p=Math.max(0,(d-1)*s+o-n),f=Math.floor(h/2),m=h-f,g=Math.floor(p/2),y=p-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)/r),d=Math.ceil((n-o+1)/s);else if(typeof e=="object"){let h=l==="channelsLast"?e[1][0]:e[2][0],p=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:h,bottom:p,left:f,right:m,type:h===0&&p===0&&f===0&&m===0?"VALID":"EXPLICIT"},c=vo((t-a+h+p)/r+1,i),d=vo((n-o+f+m)/s+1,i)}else throw Error(`Unknown padding parameter: ${e}`);return{padInfo:u,outHeight:c,outWidth:d}}function nO(e,t,n,r,s,a,o,i,l,u,c){let d,h,p,f;if(typeof e=="number"){d={top:e,bottom:e,left:e,right:e,front:e,back:e,type:e===0?"VALID":"NUMBER"};let g=eO([t,n,r,1],i,1,s,e,c);h=g[0],p=g[1],f=g[2]}else if(e==="same"){h=Math.ceil(t/s),p=Math.ceil(n/a),f=Math.ceil(r/o);let m=(h-1)*s+i-t,g=(p-1)*a+l-n,y=(f-1)*o+u-r,A=Math.floor(m/2),x=m-A,b=Math.floor(g/2),v=g-b,I=Math.floor(y/2),w=y-I;d={top:b,bottom:v,left:I,right:w,front:A,back:x,type:"SAME"}}else if(e==="valid")d={top:0,bottom:0,left:0,right:0,front:0,back:0,type:"VALID"},h=Math.ceil((t-i+1)/s),p=Math.ceil((n-l+1)/a),f=Math.ceil((r-u+1)/o);else throw Error(`Unknown padding parameter: ${e}`);return{padInfo:d,outDepth:h,outHeight:p,outWidth:f}}function vo(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 oc(e){let[t,n,r]=gp(e);return t===1&&n===1&&r===1}function ea(e,t){return oc(e)||oc(t)}function Rk(e){if(e==="NHWC")return"channelsLast";if(e==="NCHW")return"channelsFirst";throw new Error(`Unknown dataFormat ${e}`)}function rO(e,t){let r={x:P(e,"x","reshape","string_or_numeric")},s={shape:t};return U.runKernel(Mw,r,s)}var le=H({reshape_:rO});function sO(e,t,n,r,s){let a=P(e,"x","avgPool","float32"),o=1;L(ea(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=le(a,[1,a.shape[0],a.shape[1],a.shape[2]])),L(i.rank===4,()=>`Error in avgPool: x must be rank 4 but got rank ${i.rank}.`),s!=null&&L(Kn(r),()=>`Error in avgPool: pad must be an integer when using, dimRoundingMode ${s} but got pad ${r}.`);let u={x:i},c={filterSize:t,strides:n,pad:r,dimRoundingMode:s},d=U.runKernel(nv,u,c);return d=Mt(d,a.dtype),l?le(d,[d.shape[1],d.shape[2],d.shape[3]]):d}var _k=H({avgPool_:sO});function aO(e,t,n,r,s,a="NDHWC"){let o=P(e,"x","avgPool3d","float32"),i=o,l=!1;o.rank===4&&(l=!0,i=le(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),L(i.rank===5,()=>`Error in avgPool3d: x must be rank 5 but got rank ${i.rank}.`),L(a==="NDHWC",()=>`Error in avgPool3d: Only NDHWC is currently supported, but got dataFormat of ${a}`),s!=null&&L(Kn(r),()=>`Error in avgPool3d: pad must be an integer when using, dimRoundingMode ${s} but got pad ${r}.`);let u={x:i},c={filterSize:t,strides:n,pad:r,dimRoundingMode:s,dataFormat:a},d=U.runKernel(rv,u,c);return d=Mt(d,i.dtype),l?le(d,[d.shape[1],d.shape[2],d.shape[3],d.shape[4]]):d}var oO=H({avgPool3d_:aO});function iO(e,t=0){L(e.length>=1,()=>"Pass at least one tensor to concat");let n=rc(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 Qs(n[0]);let r=n,s={axis:t};return U.runKernel(dv,r,s)}var rn=H({concat_:iO});function lO(e){let n={x:P(e,"x","sigmoid")};return U.runKernel(Xw,n)}var Is=H({sigmoid_:lO});function uO(e,t,n){let r=P(e,"x","slice","string_or_numeric");if(r.rank===0)throw new Error("Slicing scalar is not possible");let s={x:r},a={begin:t,size:n};return U.runKernel(Gw,s,a)}var Xe=H({slice_:uO});function cO(e){let n={x:P(e,"x","tanh")};return U.runKernel(f7,n)}var Z2=H({tanh_:cO});function dO(e,t,n,r,s,a){let o=P(e,"forgetBias","basicLSTMCell"),i=P(t,"lstmKernel","basicLSTMCell"),l=P(n,"lstmBias","basicLSTMCell"),u=P(r,"data","basicLSTMCell"),c=P(s,"c","basicLSTMCell"),d=P(a,"h","basicLSTMCell"),h=rn([u,d],1),p=gt(h,i),f=Me(p,l),m=f.shape[0],g=f.shape[1]/4,y=[m,g],A=Xe(f,[0,0],y),x=Xe(f,[0,g],y),b=Xe(f,[0,g*2],y),v=Xe(f,[0,g*3],y),I=Me(pe(Is(A),Z2(x)),pe(c,Is(Me(o,b)))),w=pe(Z2(I),Is(v));return[I,w]}var hO=H({basicLSTMCell_:dO});function pO(e,t,n){let r=P(e,"x","batchToSpaceND"),s=t.reduce((i,l)=>i*l);L(r.rank>=1+t.length,()=>`input rank is ${r.rank} but should be > than blockShape.length ${t.length}`),L(n.length===t.length,()=>`crops.length is ${n.length} but should be equal to blockShape.length ${t.length}`),L(r.shape[0]%s==0,()=>`input tensor batch is ${r.shape[0]} but is not divisible by the product of the elements of blockShape ${t.join(" * ")} === ${s}`);let a={x:r},o={blockShape:t,crops:n};return U.runKernel(av,a,o)}var Dk=H({batchToSpaceND_:pO});function fO(e){let t;return e.rank===0||e.rank===1?t=le(e,[1,1,1,e.size]):e.rank===2?t=le(e,[1,1,e.shape[0],e.shape[1]]):e.rank===3?t=le(e,[1,e.shape[0],e.shape[1],e.shape[2]]):t=e,t}function mO(e,t,n,r,s,a){a==null&&(a=.001);let o=P(e,"x","batchNorm"),i=P(t,"mean","batchNorm"),l=P(n,"variance","batchNorm"),u;s!=null&&(u=P(s,"scale","batchNorm"));let c;r!=null&&(c=P(r,"offset","batchNorm")),L(i.rank===l.rank,()=>"Batch normalization gradient requires mean and variance to have equal ranks."),L(c==null||i.rank===c.rank,()=>"Batch normalization gradient requires mean and offset to have equal ranks."),L(u==null||i.rank===u.rank,()=>"Batch normalization gradient requires mean and scale to have equal ranks.");let h={x:fO(o),scale:u,offset:c,mean:i,variance:l},p={varianceEpsilon:a},f=U.runKernel(Wv,h,p);return le(f,o.shape)}var yp=H({batchNorm_:mO});function gO(e,t,n,r,s,a){let o=P(e,"x","batchNorm"),i=P(t,"mean","batchNorm"),l=P(n,"variance","batchNorm"),u;s!=null&&(u=P(s,"scale","batchNorm"));let c;return r!=null&&(c=P(r,"offset","batchNorm")),L(o.rank===2,()=>`Error in batchNorm2D: x must be rank 2 but got rank ${o.rank}.`),L(i.rank===2||i.rank===1,()=>`Error in batchNorm2D: mean must be rank 2 or rank 1 but got rank ${i.rank}.`),L(l.rank===2||l.rank===1,()=>`Error in batchNorm2D: variance must be rank 2 or rank 1 but got rank ${l.rank}.`),u!=null&&L(u.rank===2||u.rank===1,()=>`Error in batchNorm2D: scale must be rank 2 or rank 1 but got rank ${u.rank}.`),c!=null&&L(c.rank===2||c.rank===1,()=>`Error in batchNorm2D: offset must be rank 2 or rank 1 but got rank ${c.rank}.`),yp(o,i,l,c,u,a)}var yO=H({batchNorm2d_:gO});function AO(e,t,n,r,s,a){let o=P(e,"x","batchNorm"),i=P(t,"mean","batchNorm"),l=P(n,"variance","batchNorm"),u;s!=null&&(u=P(s,"scale","batchNorm"));let c;return r!=null&&(c=P(r,"offset","batchNorm")),L(o.rank===3,()=>`Error in batchNorm3D: x must be rank 3 but got rank ${o.rank}.`),L(i.rank===3||i.rank===1,()=>`Error in batchNorm3D: mean must be rank 3 or rank 1 but got rank ${i.rank}.`),L(l.rank===3||l.rank===1,()=>`Error in batchNorm3D: variance must be rank 3 or rank 1 but got rank ${l.rank}.`),u!=null&&L(u.rank===3||u.rank===1,()=>`Error in batchNorm3D: scale must be rank 3 or rank 1 but got rank ${u.rank}.`),c!=null&&L(c.rank===3||c.rank===1,()=>`Error in batchNorm3D: offset must be rank 3 or rank 1 but got rank ${c.rank}.`),yp(o,i,l,c,u,a)}var xO=H({batchNorm3d_:AO});function bO(e,t,n,r,s,a){let o=P(e,"x","batchNorm"),i=P(t,"mean","batchNorm"),l=P(n,"variance","batchNorm"),u;s!=null&&(u=P(s,"scale","batchNorm"));let c;return r!=null&&(c=P(r,"offset","batchNorm")),L(o.rank===4,()=>`Error in batchNorm4D: x must be rank 4 but got rank ${o.rank}.`),L(i.rank===4||i.rank===1,()=>`Error in batchNorm4D: mean must be rank 4 or rank 1 but got rank ${i.rank}.`),L(l.rank===4||l.rank===1,()=>`Error in batchNorm4D: variance must be rank 4 or rank 1 but got rank ${l.rank}.`),u!=null&&L(u.rank===4||u.rank===1,()=>`Error in batchNorm4D: scale must be rank 4 or rank 1 but got rank ${u.rank}.`),c!=null&&L(c.rank===4||c.rank===1,()=>`Error in batchNorm4D: offset must be rank 4 or rank 1 but got rank ${c.rank}.`),yp(o,i,l,c,u,a)}var vO=H({batchNorm4d_:bO});function wO(e,t,n){let r=P(e,"x","bincount"),s=P(t,"weights","bincount");L(r.dtype==="int32",()=>`Error in bincount: input dtype must be int32, but got ${r.dtype}`),L(n>=0,()=>`size must be non-negative, but got ${n}.`),L(s.size===r.size||s.size===0,()=>`Error in bincount: weights must have the same size as input or0-length, but got input shape: ${r.shape}, weights shape: ${s.shape}.`);let a={x:r,weights:s},o={size:n};return U.runKernel(ov,a,o)}var Fk=H({bincount_:wO});function kO(e,t){let n=P(e,"broadcastTo","x"),r=n.shape;if(t.some(u=>!(u>0)||u%1!=0))throw new Error(`broadcastTo(): Invalid broadcast shape [${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=le(n,u)}let s=n.shape,a=Array.from(t);for(let u=t.length-1;u>=0;u--)if(s[u]===t[u])a[u]=1;else if(n.shape[u]!==1)throw new Error(`broadcastTo(): [${r}] cannot be broadcast to [${t}].`);if(a.map((u,c)=>u>1?c:-1).filter(u=>u>=0).length===0)return Qs(n);let i={x:n},l={reps:a};return U.runKernel(o2,i,l)}var Ap=H({broadcastTo_:kO});function IO(e){let n={x:P(e,"x","ceil")};return U.runKernel(iv,n)}var SO=H({ceil_:IO});function TO(e,t,n){let r=P(e,"x","clipByValue");L(t<=n,()=>`Error in clip: min (${t}) must be less than or equal to max (${n}).`);let s={x:r},a={clipValueMin:t,clipValueMax:n};return U.runKernel(lv,s,a)}var NO=H({clipByValue_:TO});function CO(e){return rn(e,0)}var EO=H({concat1d_:CO});function $O(e,t){return rn(e,t)}var ic=H({concat2d_:$O});function RO(e,t){return rn(e,t)}var _O=H({concat3d_:RO});function DO(e,t){return rn(e,t)}var FO=H({concat4d_:DO});function MO(e,t,n,r,s="NHWC",a=[1,1],o){let i=P(e,"x","conv2d"),l=P(t,"filter","conv2d"),u=i,c=!1;i.rank===3&&(c=!0,u=le(i,[1,i.shape[0],i.shape[1],i.shape[2]])),L(u.rank===4,()=>`Error in conv2d: input must be rank 4, but got rank ${u.rank}.`),L(l.rank===4,()=>`Error in conv2d: filter must be rank 4, but got rank ${l.rank}.`),o!=null&&L(Kn(r),()=>`Error in conv2d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${r}.`);let d=s==="NHWC"?u.shape[3]:u.shape[1];L(d===l.shape[2],()=>`Error in conv2d: depth of input (${d}) must match input depth for filter ${l.shape[2]}.`),L(ea(n,a),()=>`Error in conv2D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`);let h={x:u,filter:l},p={strides:n,pad:r,dataFormat:s,dilations:a,dimRoundingMode:o},f=U.runKernel(hv,h,p);return c?le(f,[f.shape[1],f.shape[2],f.shape[3]]):f}var xp=H({conv2d_:MO});function OO(e,t,n,r,s="NWC",a=1,o){let i=P(e,"x","conv1d"),l=P(t,"filter","conv1d"),u=i,c=!1;i.rank===2&&(c=!0,u=le(i,[1,i.shape[0],i.shape[1]])),L(u.rank===3,()=>`Error in conv1d: input must be rank 3, but got rank ${u.rank}.`),L(l.rank===3,()=>`Error in conv1d: filter must be rank 3, but got rank ${l.rank}.`),o!=null&&L(Kn(r),()=>`Error in conv1d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${r}.`),L(u.shape[2]===l.shape[1],()=>`Error in conv1d: depth of input (${u.shape[2]}) must match input depth for filter ${l.shape[1]}.`),L(ea(n,a),()=>`Error in conv1D: Either stride or dilation must be 1. Got stride ${n} and dilation '${a}'`),L(s==="NWC",()=>`Error in conv1d: got dataFormat of ${s} but only NWC is currently supported.`);let d=le(l,[1,l.shape[0],l.shape[1],l.shape[2]]),h=le(u,[u.shape[0],1,u.shape[1],u.shape[2]]),g=xp(h,d,[1,n],r,"NHWC",[1,a],o);return c?le(g,[g.shape[2],g.shape[3]]):le(g,[g.shape[0],g.shape[2],g.shape[3]])}var PO=H({conv1d_:OO});function zO(e,t,n,r,s,a="NHWC",o){L(e.length===t.rank,()=>`Length of inShape (${e.length}) and rank of dy (${t.rank}) must match`);let i=e,l=t,u=!1;t.rank===3&&(u=!0,l=le(t,[1,t.shape[0],t.shape[1],t.shape[2]]),i=[1,e[0],e[1],e[2]]),L(i.length===4,()=>`Error in conv2dDerInput: inShape must be length 4, but got length ${i.length}.`),L(l.rank===4,()=>`Error in conv2dDerInput: dy must be rank 4, but got rank ${l.rank}`),L(n.rank===4,()=>`Error in conv2dDerInput: filter must be rank 4, but got rank ${n.rank}`);let c=a==="NHWC"?i[3]:i[1],d=a==="NHWC"?l.shape[3]:l.shape[1];L(c===n.shape[2],()=>`Error in conv2dDerInput: depth of input (${c}) must match input depth for filter ${n.shape[2]}.`),L(d===n.shape[3],()=>`Error in conv2dDerInput: depth of output (${d}) must match output depth for filter ${n.shape[3]}.`),o!=null&&L(Kn(s),()=>`Error in conv2dDerInput: pad must be an integer when using, dimRoundingMode ${o} but got pad ${s}.`);let h={dy:l,filter:n},p={strides:r,pad:s,dataFormat:a,dimRoundingMode:o,inputShape:i},f=U.runKernel(fv,h,p);return u?le(f,[f.shape[1],f.shape[2],f.shape[3]]):f}var Mk=H({conv2DBackpropInput_:zO});function LO(e,t,n,r,s,a){let o=P(e,"x","conv2dTranspose"),i=P(t,"filter","conv2dTranspose");return Mk(n,o,i,r,s,"NHWC",a)}var BO=H({conv2dTranspose_:LO});function WO(e,t,n,r,s="NDHWC",a=[1,1,1]){let o=P(e,"x","conv3d"),i=P(t,"filter","conv3d"),l=o,u=!1;o.rank===4&&(u=!0,l=le(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),L(l.rank===5,()=>`Error in conv3d: input must be rank 5, but got rank ${l.rank}.`),L(i.rank===5,()=>`Error in conv3d: filter must be rank 5, but got rank ${i.rank}.`),L(l.shape[4]===i.shape[3],()=>`Error in conv3d: depth of input (${l.shape[4]}) must match input depth for filter ${i.shape[3]}.`),L(ea(n,a),()=>`Error in conv3D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`),L(s==="NDHWC",()=>`Error in conv3d: got dataFormat of ${s} but only NDHWC is currently supported.`);let c={x:l,filter:i},d={strides:n,pad:r,dataFormat:s,dilations:a},h=U.runKernel(mv,c,d);return u?le(h,[h.shape[1],h.shape[2],h.shape[3],h.shape[4]]):h}var VO=H({conv3d_:WO});function UO(e,t,n,r,s){L(e.length===t.rank,()=>`Length of inShape (${e.length}) and rank of dy (${t.rank}) must match`);let a=e,o=t,i=!1;t.rank===4&&(i=!0,o=le(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];L(a.length===5,()=>`Error in conv3dDerInput: inShape must be length 5, but got length ${a.length}.`),L(o.rank===5,()=>`Error in conv3dDerInput: dy must be rank 5, but got rank ${o.rank}`),L(n.rank===5,()=>`Error in conv3dDerInput: filter must be rank 5, but got rank ${n.rank}`),L(l===n.shape[3],()=>`Error in conv3dDerInput: depth of input (${l}) must match input depth for filter ${n.shape[3]}.`),L(u===n.shape[4],()=>`Error in conv3dDerInput: depth of output (${u}) must match output depth for filter ${n.shape[4]}.`);let c={dy:o,filter:n},d={pad:s,strides:r,inputShape:a},h=U.runKernel(gv,c,d);return i?le(h,[h.shape[1],h.shape[2],h.shape[3],h.shape[4]]):h}var HO=H({conv3DBackpropInput_:UO});function GO(e,t,n,r,s){let a=P(e,"x","conv3dTranspose"),o=P(t,"filter","conv3dTranspose");return HO(n,a,o,r,s)}var jO=H({conv3dTranspose_:GO});function qO(e){let n={x:P(e,"x","cos")};return U.runKernel(yv,n)}var KO=H({cos_:qO});function XO(e){let n={x:P(e,"x","cosh")};return U.runKernel(Av,n)}var ZO=H({cosh_:XO});function YO(e,t=0,n=!1,r=!1){let a={x:P(e,"x","cumsum")},o={axis:t,exclusive:n,reverse:r};return U.runKernel(xv,a,o)}var JO=H({cumsum_:YO});function QO(e,t,n,r=!1){let s=P(e,"x","denseBincount"),a=P(t,"weights","denseBincount");L(s.dtype==="int32",()=>`Error in denseBincount: input dtype must be int32, but got ${s.dtype}`),L(s.rank<=2,()=>`Error in denseBincount: input must be at most rank 2, but got rank ${s.rank}.`),L(n>=0,()=>`size must be non-negative, but got ${n}.`),L(a.size===s.size||a.size===0,()=>`Error in denseBincount: weights must have the same shape as x or 0-length, but got x shape: ${s.shape}, weights shape: ${a.shape}.`);let o={x:s,weights:a},i={size:n,binaryOutput:r};return U.runKernel(vv,o,i)}var eP=H({denseBincount_:QO});function tP(e,t,n="NHWC"){let r=P(e,"x","depthToSpace"),s=n==="NHWC"?r.shape[1]:r.shape[2],a=n==="NHWC"?r.shape[2]:r.shape[3],o=n==="NHWC"?r.shape[3]:r.shape[1];L(s*t>=0,()=>`Negative dimension size caused by overflow when multiplying
${s} and ${t} for depthToSpace with input shape
${r.shape}`),L(a*t>=0,()=>`Negative dimension size caused by overflow when multiplying
${a} and ${t} for depthToSpace with input shape
${r.shape}`),L(o%(t*t)==0,()=>`Dimension size must be evenly divisible by ${t*t} but is ${o} for depthToSpace with input shape ${r.shape}`);let i={x:r},l={blockSize:t,dataFormat:n};return U.runKernel(wv,i,l)}var nP=H({depthToSpace_:tP});function rP(e,t,n,r,s="NHWC",a=[1,1],o){let i=P(e,"x","depthwiseConv2d"),l=P(t,"filter","depthwiseConv2d"),u=i,c=!1;i.rank===3&&(c=!0,u=le(i,[1,i.shape[0],i.shape[1],i.shape[2]])),L(u.rank===4,()=>`Error in depthwiseConv2d: input must be rank 4, but got rank ${u.rank}.`),L(l.rank===4,()=>`Error in depthwiseConv2d: filter must be rank 4, but got rank ${l.rank}.`),L(u.shape[3]===l.shape[2],()=>`Error in depthwiseConv2d: number of input channels (${u.shape[3]}) must match the inChannels dimension in filter ${l.shape[2]}.`),o!=null&&L(Kn(r),()=>`Error in depthwiseConv2d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${r}.`);let d={x:u,filter:l},h={strides:n,pad:r,dataFormat:s,dilations:a,dimRoundingMode:o},p=U.runKernel(kv,d,h);return c?le(p,[p.shape[1],p.shape[2],p.shape[3]]):p}var Y2=H({depthwiseConv2d_:rP});function sP(e){let n={x:P(e,"x","diag")};return U.runKernel(Tv,n)}var aP=H({diag_:sP});function oP(e,t,n,r,s=[1,1],a="NHWC"){let o=P(e,"x","dilation2d"),i=P(t,"filter","dilation2d");L(o.rank===3||o.rank===4,()=>`Error in dilation2d: input must be rank 3 or 4, but got rank ${o.rank}.`),L(i.rank===3,()=>`Error in dilation2d: filter must be rank 3, but got rank ${i.rank}.`),L(a==="NHWC",()=>`Error in dilation2d: Only NHWC is currently supported, but got dataFormat of ${a}`);let l=o,u=!1;o.rank===3&&(l=le(o,[1,o.shape[0],o.shape[1],o.shape[2]]),u=!0);let c={x:l,filter:i},d={strides:n,pad:r,dilations:s},h=U.runKernel(Nv,c,d);return u?le(h,[h.shape[1],h.shape[2],h.shape[3]]):h}var iP=H({dilation2d_:oP});function lP(e,t){let n=e.length,r=[];for(let s=0;s<n;s++){let a=n-1-s,o=e[a]||1;(t[t.length-1-s]||1)>1&&o===1&&r.unshift(a)}return r}function Ok(e,t){let n=[];for(let r=0;r<t.length;r++){let s=e[e.length-r-1],a=t.length-r-1,o=t[a];(s==null||s===1&&o>1)&&n.unshift(a)}return n}function kn(e,t){let n=[],r=Math.max(e.length,t.length);for(let s=0;s<r;s++){let a=e[e.length-s-1];a==null&&(a=1);let o=t[t.length-s-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}function uP(e,t){let n=P(e,"a","equal","string_or_numeric"),r=P(t,"b","equal","string_or_numeric");[n,r]=Wt(n,r),kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(_v,s)}var Pk=H({equal_:uP});function cP(e,t,n){let r=P(t,"a","where"),s=P(n,"b","where"),a=P(e,"condition","where","bool"),o=kn(kn(a.shape,r.shape),s.shape),i=Ap(a,o),l=Ap(r,o),u=Ap(s,o),c={condition:i,t:l,e:u};return U.runKernel(Uw,c)}var Hi=H({where_:cP});function dP(e){let n={x:P(e,"x","zerosLike")};return U.runKernel(v7,n)}var Tr=H({zerosLike_:dP});function hP(e,t){let n=P(e,"a","div"),r=P(t,"b","div");[n,r]=Wt(n,r);let s=Je(n,r),a=Tr(s),o=Pk(r,a);return Hi(o,a,s)}var pP=H({divNoNan_:hP});function fP(e,t){let n=P(e,"t1","dot"),r=P(t,"t2","dot");L((n.rank===1||n.rank===2)&&(r.rank===1||r.rank===2),()=>`Error in dot: inputs must all be rank 1 or 2, but got ranks ${n.rank} and ${r.rank}.`);let s=n.rank===1?n.size:n.shape[1],a=r.rank===1?r.size:r.shape[0];if(L(s===a,()=>`Error in dot: inner dimensions of inputs must match, but got ${s} and ${a}.`),n.rank===1&&r.rank===1){let o=le(n,[1,-1]),i=le(r,[-1,1]),l=gt(o,i);return le(l,[])}else if(n.rank===1&&r.rank===2){let o=le(n,[1,-1]),i=le(r,[r.shape[0],r.shape[1]]),l=gt(o,i);return le(l,[l.size])}else if(n.rank===2&&r.rank===1){let o=le(r,[-1,1]),i=gt(n,o);return le(i,[i.size])}else{let o=le(r,[r.shape[0],r.shape[1]]);return gt(n,o)}}var mP=H({dot_:fP});function gP(e,...t){let n=t.map((s,a)=>P(s,`tensors${a}`,"einsum")),r={equation:e};return U.runKernel(Ev,n,r)}var yP=H({einsum_:gP});function AP(e){let n={x:P(e,"x","elu")};return U.runKernel($v,n)}var zk=H({elu_:AP});function xP(e){let t=P(e,"x","erf");L(t.dtype==="int32"||t.dtype==="float32",()=>"Input dtype must be `int32` or `float32`."),t.dtype==="int32"&&(t=Mt(t,"float32"));let n={x:t};return U.runKernel(Rv,n)}var bP=H({erf_:xP});function vP(e){let n={x:P(e,"x","exp")};return U.runKernel(Dv,n)}var wo=H({exp_:vP});function wP(e,t=0){let n=P(e,"x","expandDims","string_or_numeric");L(t<=n.rank,()=>"Axis must be <= rank of the tensor");let r={input:n},s={dim:t};return U.runKernel(Fv,r,s)}var ta=H({expandDims_:wP});function kP(e){let n={x:P(e,"x","expm1")};return U.runKernel(Mv,n)}var IP=H({expm1_:kP});function SP(e,t){let n=P(e,"x","tile","string_or_numeric");L(n.rank===t.length,()=>`Error in transpose: rank of input ${n.rank} must match length of reps ${t}.`);let r={x:n},s={reps:t};return U.runKernel(o2,r,s)}var bp=H({tile_:SP});function TP(e,t,n,r="float32"){t==null&&(t=e);let s=Js([e,t],r),a=e<=t?e:t;for(let i=0;i<a;++i)s.set(1,i,i);let o=le(s.toTensor(),[e,t]);if(n==null)return o;if(n.length===1)return bp(ta(o,0),[n[0],1,1]);if(n.length===2)return bp(ta(ta(o,0),0),[n[0],n[1],1,1]);if(n.length===3)return bp(ta(ta(ta(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 Lk=H({eye_:TP});function vp(e,t,n){let r={shape:e,value:t,dtype:n};return U.runKernel(Pv,{},r)}function NP(e){let n={x:P(e,"x","floor")};return U.runKernel(Lv,n)}var Bk=H({floor_:NP});function CP(e,t,n=0,r=0){let s=P(e,"x","gather"),a=P(t,"indices","gather","int32"),o={x:s,indices:a},i={axis:n,batchDims:r};return U.runKernel(Vv,o,i)}var Wk=H({gather_:CP});function EP(e,t){let n=P(e,"a","greater","string_or_numeric"),r=P(t,"b","greater","string_or_numeric");[n,r]=Wt(n,r),kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(Hv,s)}var wp=H({greater_:EP});function $P(e,t){let n=P(e,"a","greaterEqual","string_or_numeric"),r=P(t,"b","greaterEqual","string_or_numeric");[n,r]=Wt(n,r),kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(Gv,s)}var Vk=H({greaterEqual_:$P});function RP(e){let n={input:P(e,"input","imag")};return U.runKernel(qv,n)}var J2=H({imag_:RP});function _P(e){let n={x:P(e,"x","isFinite")};return U.runKernel(Kv,n)}var DP=H({isFinite_:_P});function FP(e){let n={x:P(e,"x","isInf")};return U.runKernel(Xv,n)}var MP=H({isInf_:FP});function OP(e){let n={x:P(e,"x","isNaN")};return U.runKernel(Zv,n)}var PP=H({isNaN_:OP});function zP(e,t=.2){let r={x:P(e,"x","leakyRelu")},s={alpha:t};return U.runKernel(Yv,r,s)}var Uk=H({leakyRelu_:zP});function LP(e,t){let n=P(e,"a","less","string_or_numeric"),r=P(t,"b","less","string_or_numeric");[n,r]=Wt(n,r),kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(Jv,s)}var BP=H({less_:LP});function WP(e,t){let n=P(e,"a","lessEqual","string_or_numeric"),r=P(t,"b","lessEqual","string_or_numeric");[n,r]=Wt(n,r),kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(Qv,s)}var Q2=H({lessEqual_:WP});function VP(e,t,n){if(n<=0)throw new Error("The number of values should be positive.");let r={start:e,stop:t,num:n};return U.runKernel(ew,{},r)}function UP(e,t=5,n=1,r=1,s=.5){let a=P(e,"x","localResponseNormalization");L(a.rank===4||a.rank===3,()=>`Error in localResponseNormalization: x must be rank 3 or 4 but got
rank ${a.rank}.`),L(Kn(t),()=>`Error in localResponseNormalization: depthRadius must be an integer but got depthRadius ${t}.`);let o=a,i=!1;a.rank===3&&(i=!0,o=le(a,[1,a.shape[0],a.shape[1],a.shape[2]]));let l={x:o},u={depthRadius:t,bias:n,alpha:r,beta:s},c=U.runKernel(ow,l,u);return i?le(c,[c.shape[1],c.shape[2],c.shape[3]]):c}var HP=H({localResponseNormalization_:UP});function GP(e){let n={x:P(e,"x","log")};return U.runKernel(tw,n)}var lc=H({log_:GP});function jP(e){let n={x:P(e,"x","log1p")};return U.runKernel(nw,n)}var Hk=H({log1p_:jP});function qP(e){return L(Ta(e),()=>"The f passed in grad(f) must be a function"),(t,n)=>{let r=P(t,"x","tf.grad","string_or_numeric"),s=n!=null?P(n,"dy","tf.grad"):null;return U.tidy(()=>{let{value:a,grads:o}=U.gradients(()=>e(r),[r],s);return s!=null&&Mn(a.shape,s.shape,"The shape of dy passed in grad(f)(x, dy) must match the shape returned by f(x)"),kp(o),o[0]})}}function KP(e){return L(Ta(e),()=>"The f passed in grads(f) must be a function"),(t,n)=>{L(Array.isArray(t),()=>"The args passed in grads(f)(args) must be an array of `Tensor`s or `TensorLike`s");let r=rc(t,"args","tf.grads","string_or_numeric"),s=n!=null?P(n,"dy","tf.grads"):null;return U.tidy(()=>{let{value:a,grads:o}=U.gradients(()=>e(...r),r,s);return s!=null&&Mn(a.shape,s.shape,"The shape of dy passed in grads(f)([x1,...], dy) must match the shape returned by f([x1,...])"),kp(o),o})}}function XP(e){return L(Ta(e),()=>"The f passed in valueAndGrad(f) must be a function"),(t,n)=>{L(t instanceof It,()=>"The x passed in valueAndGrad(f)(x) must be a tensor"),L(n==null||n instanceof It,()=>"The dy passed in valueAndGrad(f)(x, dy) must be a tensor");let{grads:r,value:s}=U.gradients(()=>e(t),[t],n);return kp(r),{grad:r[0],value:s}}}function ZP(e){return L(Ta(e),()=>"The f passed in valueAndGrads(f) must be a function"),(t,n)=>{L(Array.isArray(t)&&t.every(s=>s instanceof It),()=>"The args passed in valueAndGrads(f)(args) must be array of tensors"),L(n==null||n instanceof It,()=>"The dy passed in valueAndGrads(f)(args, dy) must be a tensor");let r=U.gradients(()=>e(...t),t,n);return n!=null&&Mn(r.value.shape,n.shape,"The shape of dy passed in valueAndGrads(f)([x1,...], dy) must match the shape returned by f([x1,...])"),kp(r.grads),r}}function Gk(e,t){L(Ta(e),()=>"The f passed in variableGrads(f) must be a function"),L(t==null||Array.isArray(t)&&t.every(u=>u instanceof nc),()=>"The varList passed in variableGrads(f, varList) must be an array of variables");let n=t!=null;if(!n){t=[];for(let u in U.registeredVariables)t.push(U.registeredVariables[u])}let r=n?t.filter(u=>!u.trainable):null,s=t.length;t=t.filter(u=>u.trainable),L(t.length>0,()=>`variableGrads() expects at least one of the input variables to be trainable, but none of the ${s} variables is trainable.`);let a=!0,{value:o,grads:i}=U.gradients(e,t,null,a);L(i.some(u=>u!=null),()=>"Cannot find a connection between any variable and the result of the loss function y=f(x). Please make sure the operations that use variables are inside the function f passed to minimize()."),L(o.rank===0,()=>`The f passed in variableGrads(f) must return a scalar, but it returned a rank-${o.rank} tensor`);let l={};return t.forEach((u,c)=>{i[c]!=null&&(l[u.name]=i[c])}),r!=null&&r.forEach(u=>l[u.name]=null),{value:o,grads:l}}function Ss(e){return U.customGrad(e)}function kp(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 YP(e){let n={x:P(e,"x","neg")};return U.runKernel(xw,n)}var Ra=H({neg_:YP});function JP(e){let n={x:P(e,"x","softplus")};return U.runKernel(Zw,n)}var jk=H({softplus_:JP});function QP(e){let t=P(e,"x","logSigmoid");return Ss(r=>({value:Ra(jk(Ra(r))),gradFunc:o=>pe(o,Is(Ra(r)))}))(t)}var ez=H({logSigmoid_:QP});function tz(e,t=null,n=!1){let s={x:P(e,"x","max")},a={reductionIndices:t,keepDims:n};return U.runKernel(iw,s,a)}var _a=H({max_:tz});function nz(e,t){let n=P(e,"a","sub"),r=P(t,"b","sub");[n,r]=Wt(n,r);let s={a:n,b:r};return U.runKernel(h7,s)}var Ue=H({sub_:nz});function rz(e,t=null,n=!1){let r=P(e,"x","sum");r.dtype==="bool"&&(r=Mt(r,"int32"));let s={x:r},a={axis:t,keepDims:n};return U.runKernel(Jw,s,a)}var Et=H({sum_:rz});function sz(e,t=-1){let n=P(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 ${t}`);return Ss((s,a)=>{let o=!0,i=_a(s,t,!0),l=Ue(s,i),u=Ue(Mt(l,"float32"),lc(Et(wo(l),t,o)));return a([u]),{value:u,gradFunc:(d,h)=>{let[p]=h,f=!0,m=wo(p);return Ue(d,pe(Et(d,t,f),m))}}})(n)}var az=H({logSoftmax_:sz});function e1(e,t){for(let n=0;n<e.length;++n)if(e[e.length-n-1]!==t-1-n)return!1;return!0}function qk(e,t,n){let r=e.length+t.length,s=[],a=0,o=0;for(let i=0;i<r;i++)n.indexOf(i)===-1?s.push(e[a++]):s.push(t[o++]);return s}function oz(e,t){let n=[],r=e.length;for(let a=0;a<r;a++)t.indexOf(a)===-1&&n.push(e[a]);let s=t.map(a=>e[a]);return[n,s]}function uc(e,t){let n=t.map(r=>1);return qk(e,n,t)}function iz(e,t,n){L(e1(t,n),()=>`${e} supports only inner-most axes for now. Got axes ${t} and rank-${n} input.`)}function lz(e,t){if(e1(e,t))return null;let n=[];for(let r=0;r<t;++r)e.indexOf(r)===-1&&n.push(r);return e.forEach(r=>n.push(r)),n}function uz(e){return e.map((t,n)=>[n,t]).sort((t,n)=>t[1]-n[1]).map(t=>t[0])}function cz(e,t){let n=[];for(let r=t-e;r<t;++r)n.push(r);return n}function dz(e,t=null,n=!1){let r=P(e,"x","logSumExp"),s=Ku(t,r.shape),a=_a(r,s,!0),o=Ue(r,a),i=wo(o),l=Et(i,s),u=lc(l),c=Me(le(a,u.shape),u);if(n){let d=uc(c.shape,s);return le(c,d)}return c}var Kk=H({logSumExp_:dz});function hz(e,t){let n=P(e,"a","logicalAnd","bool"),r=P(t,"b","logicalAnd","bool");kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(rw,s)}var Ip=H({logicalAnd_:hz});function pz(e){let n={x:P(e,"x","logicalNot","bool")};return U.runKernel(sw,n)}var Xk=H({logicalNot_:pz});function fz(e,t){let n=P(e,"a","logicalOr","bool"),r=P(t,"b","logicalOr","bool");kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(aw,s)}var Zk=H({logicalOr_:fz});function mz(e,t){let n=P(e,"a","logicalXor","bool"),r=P(t,"b","logicalXor","bool");return kn(n.shape,r.shape),Ip(Zk(e,t),Xk(Ip(e,t)))}var gz=H({logicalXor_:mz});function yz(e,t,n,r,s){let a=P(e,"x","maxPool"),o=1,i=a,l=!1;a.rank===3&&(l=!0,i=le(a,[1,a.shape[0],a.shape[1],a.shape[2]])),L(i.rank===4,()=>`Error in maxPool: input must be rank 4 but got rank ${i.rank}.`),L(ea(n,o),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${n} and dilations '${o}'`),s!=null&&L(Kn(r),()=>`Error in maxPool: pad must be an integer when using, dimRoundingMode ${s} but got pad ${r}.`);let u={x:i},c={filterSize:t,strides:n,pad:r,dimRoundingMode:s},d=U.runKernel(uw,u,c);return l?le(d,[d.shape[1],d.shape[2],d.shape[3]]):d}var Yk=H({maxPool_:yz});function Az(e,t=[1,1,1],n,r,s,a="NDHWC"){let o=P(e,"x","maxPool3d"),i=o,l=!1;o.rank===4&&(l=!0,i=le(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),L(i.rank===5,()=>`Error in maxPool3d: x must be rank 5 but got rank ${i.rank}.`),L(a==="NDHWC",()=>`Error in maxPool3d: Only NDHWC is currently supported, but got dataFormat of ${a}`),s!=null&&L(Kn(r),()=>`Error in maxPool3d: pad must be an integer when using, dimRoundingMode ${s} but got pad ${r}.`);let u={x:i},c={filterSize:t,strides:n,pad:r,dimRoundingMode:s,dataFormat:a},d=U.runKernel(cw,u,c);return l?le(d,[d.shape[1],d.shape[2],d.shape[3],d.shape[4]]):d}var xz=H({maxPool3d_:Az});function bz(e,t,n,r,s=!1){let o={x:P(e,"x","maxPoolWithArgmax")},i={filterSize:t,strides:n,pad:r,includeBatchInIndex:s},l=U.runKernel(dw,o,i);return{result:l[0],indexes:l[1]}}var vz=H({maxPoolWithArgmax_:bz});function wz(e,t){let n=P(e,"a","maximum"),r=P(t,"b","maximum");[n,r]=Wt(n,r),n.dtype==="bool"&&(n=Mt(n,"int32"),r=Mt(r,"int32")),kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(lw,s)}var Jk=H({maximum_:wz});function kz(e,t=null,n=!1){let s={x:P(e,"x","mean")},a={axis:t,keepDims:n};return U.runKernel(hw,s,a)}var Sp=H({mean_:kz});function Gi(e,t="float32"){if(t==="complex64"){let r=Gi(e,"float32"),s=Gi(e,"float32");return go(r,s)}let n=tp(Yt(e),t);return U.makeTensor(n,e,t)}function ko(e,t="float32"){if(t==="complex64"){let r=ko(e,"float32"),s=Gi(e,"float32");return go(r,s)}let n=Jg(Yt(e),t);return U.makeTensor(n,e,t)}function Iz(e,t,{indexing:n="xy"}={}){if(n!=="xy"&&n!=="ij")throw new TypeError(`${n} is not a valid third argument to meshgrid`);if(e===void 0)return[];let r=P(e,"x","meshgrid",e instanceof It?e.dtype:"float32");if(t===void 0)return[r];let s=P(t,"y","meshgrid",t instanceof It?t.dtype:"float32"),a=Yt(r.shape),o=Yt(s.shape);return n==="xy"?(r=le(r,[1,-1]),s=le(s,[-1,1]),[gt(ko([o,1],r.dtype),r),gt(s,ko([1,a],s.dtype))]):(r=le(r,[-1,1]),s=le(s,[1,-1]),[gt(r,ko([1,o],r.dtype)),gt(ko([a,1],s.dtype),s)])}function Sz(e,t=null,n=!1){let s={x:P(e,"x","min")},a={axis:t,keepDims:n};return U.runKernel(pw,s,a)}var t1=H({min_:Sz});function Tz(e,t){let n=P(e,"a","minimum"),r=P(t,"b","minimum");[n,r]=Wt(n,r),n.dtype==="bool"&&(n=Mt(n,"int32"),r=Mt(r,"int32")),kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(fw,s)}var Qk=H({minimum_:Tz});function Nz(e,t,n){L(n==="reflect"||n==="symmetric",()=>`Invalid mode. Mode must be either reflect or symmetric. Got ${n}.`);let r=P(e,"x","mirrorPad");if(r.rank===0)throw new Error("mirrorPad(scalar) is not defined. Pass non-scalar to mirrorPad");L(t.length===r.rank,()=>`Padding doesn't match input. Must be ${r.rank}. Got ${t.length}.`);let s=n==="reflect"?1:0;for(let i=0;i<r.rank;i++)L(t[i].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),L(t[i][0]>=0&&t[i][0]<=r.shape[i]-s&&t[i][1]>=0&&t[i][1]<=r.shape[i]-s,()=>`Padding in dimension ${i} cannot be greater than or equal to ${r.shape[i]-s} or less than 0 for input of shape ${r.shape}`);let a={paddings:t,mode:n},o={x:r};return U.runKernel(mw,o,a)}var Cz=H({mirrorPad_:Nz});function Ez(e,t){let n=P(e,"a","mod"),r=P(t,"b","mod");[n,r]=Wt(n,r);let s={a:n,b:r};return U.runKernel(gw,s)}var $z=H({mod_:Ez});function Rz(e){let t=P(e,"x","square"),n={};return U.runKernel("Square",{x:t},n)}var rs=H({square_:Rz});function _z(e,t=null,n=!1){e=P(e,"x","moments");let r=Ku(t,e.shape),s=Sp(e,r,n),a=s.shape;n||(a=uc(s.shape,r));let o=rs(Ue(Mt(e,"float32"),le(s,a))),i=Sp(o,r,n);return{mean:s,variance:i}}var Dz=H({moments_:_z});function Fz(e,t,n,r){let s=P(t,"data","multiRNNCell"),a=rc(n,"c","multiRNNCell"),o=rc(r,"h","multiRNNCell"),i=s,l=[];for(let d=0;d<e.length;d++){let h=e[d](i,a[d],o[d]);l.push(h[0]),l.push(h[1]),i=h[1]}let u=[],c=[];for(let d=0;d<l.length;d+=2)u.push(l[d]),c.push(l[d+1]);return[u,c]}var Mz=H({multiRNNCell_:Fz});function Oz(e,t,n,r=!1){let s=P(e,"logits","multinomial"),a=s.size,o=s.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?le(s,[1,-1]):s},u={numSamples:t,seed:n,normalized:r},c=U.runKernel(yw,l,u);return o===1?le(c,[c.size]):c}var Pz=H({multinomial_:Oz});function zz(e,t){let n=P(e,"a","notEqual","string_or_numeric"),r=P(t,"b","notEqual","string_or_numeric");[n,r]=Wt(n,r),kn(n.shape,r.shape);let s={a:n,b:r};return U.runKernel(bw,s)}var e4=H({notEqual_:zz});function Lz(e){let n={x:P(e,"x","onesLike")};return U.runKernel(Iw,n)}var Bz=H({onesLike_:Lz});function Wz(e,t){let n=P(e,"v1","outerProduct"),r=P(t,"v2","outerProduct");L(n.rank===1&&r.rank===1,()=>`Error in outerProduct: inputs must be rank 1, but got ranks ${n.rank} and ${r.rank}.`);let s=le(n,[-1,1]),a=le(r,[1,-1]);return gt(s,a)}var Vz=H({outerProduct_:Wz});function Uz(e,t,n=0){let r=P(e,"x","pad");if(r.rank===0)throw new Error("pad(scalar) is not defined. Pass non-scalar to pad");let s={paddings:t,constantValue:n},a={x:r};return U.runKernel(Nw,a,s)}var cc=H({pad_:Uz});function Hz(e,t,n=0){return L(t.length===2,()=>"Invalid number of paddings. Must be length of 2."),cc(e,[t],n)}var Gz=H({pad1d_:Hz});function jz(e,t,n=0){return L(t.length===2&&t[0].length===2&&t[1].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),cc(e,t,n)}var qz=H({pad2d_:jz});function Kz(e,t,n=0){return L(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."),cc(e,t,n)}var Xz=H({pad3d_:Kz});function Zz(e,t,n=0){return L(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."),cc(e,t,n)}var Yz=H({pad4d_:Zz});function Jz(e,t,n){let r=P(e,"x","spaceToBatchND");L(r.rank>=1+t.length,()=>`input rank ${r.rank} should be > than [blockShape] ${t.length}`),L(n.length===t.length,()=>`paddings.shape[0] ${n.length} must be equal to [blockShape] ${t.length}`),L(r.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 ${r.shape.slice(1)} with paddings ${n.toString()} must be divisible by blockShapes ${t.toString()}`);let s={x:r},a={blockShape:t,paddings:n};return U.runKernel(Qw,s,a)}var t4=H({spaceToBatchND_:Jz});function Qz(e,t,n,r,s,a){s==null&&(s=[1,1]),a==null&&(a=1),r===0&&(r="valid");let o=P(e,"x","maxPool"),i=o,l=!1;o.rank===3&&(l=!0,i=le(o,[1,o.shape[0],o.shape[1],o.shape[2]])),L(ea(a,s),()=>`Error in pool: Either strides or dilations must be 1. Got strides ${a} and dilations '${s}'`);let u=Ek(i.shape,t,a,s,r),c=[u.dilationHeight,u.dilationWidth],d;r==="same"?d=tL([u.filterHeight,u.filterWidth],c):d=[[0,0],[0,0]];let h=c[0]===1&&c[1]===1,[p,f]=eL([u.inHeight,u.inWidth],c,d),m=h?r:"valid",g=h?i:t4(i,c,p),A=(n==="avg"?()=>_k(g,t,a,m):()=>Yk(g,t,a,m))(),x=h?A:Dk(A,c,f);return l?le(x,[x.shape[1],x.shape[2],x.shape[3]]):x}function eL(e,t,n){let r=n.map(c=>c[0]),s=n.map(c=>c[1]),a=e.concat(r,s),o=t.map((c,d)=>(c-a[d]%c)%c),i=s.map((c,d)=>c+o[d]),l=t.map((c,d)=>[r[d],i[d]]),u=t.map((c,d)=>[0,o[d]]);return[l,u]}function tL(e,t){let r=e.map((o,i)=>o+(o-1)*(t[i]-1)).map(o=>o-1),s=r.map(o=>Math.floor(o/2)),a=r.map((o,i)=>o-s[i]);return r.map((o,i)=>[s[i],a[i]])}var nL=H({pool_:Qz});function rL(e,t){let n=P(e,"base","pow"),r=P(t,"exp","pow");[n,r]=Wt(n,r);let s={a:n,b:r};return U.runKernel(Cw,s)}var dc=H({pow_:rL});function sL(e,t){let n=P(e,"x","prelu"),r=P(t,"alpha","prelu"),s={x:n,alpha:r};return U.runKernel(Ew,s)}var n4=H({prelu_:sL});function aL(e,t=null,n=!1){let r=P(e,"x","prod");r.dtype==="bool"&&(r=Mt(r,"int32"));let s={x:r},a={axis:t,keepDims:n};return U.runKernel($w,s,a)}var oL=H({prod_:aL});function iL(e,t,n){let r=Yt(e),s=null;if(n==null||n==="float32")s=new Float32Array(r);else if(n==="int32")s=new Int32Array(r);else if(n==="bool")s=new Uint8Array(r);else throw new Error(`Unknown data type ${n}`);for(let a=0;a<r;a++)s[a]=t();return U.makeTensor(s,e,n)}var lL=H({rand_:iL}),n1=Xs(Zg()),r1=class{constructor(e,t,n,r,s){this.mean=e,this.stdDev=t,this.dtype=n,this.nextVal=NaN,this.truncated=r,this.truncated&&(this.upper=this.mean+this.stdDev*2,this.lower=this.mean-this.stdDev*2);let a=s||Math.random();this.random=n1.alea(a.toString())}nextValue(){if(!isNaN(this.nextVal)){let r=this.nextVal;return this.nextVal=NaN,r}let e,t,n=!1;for(;!n;){let r,s,a;do r=2*this.random()-1,s=2*this.random()-1,a=r*r+s*s;while(a>=1||a===0);let o=Math.sqrt(-2*Math.log(a)/a);e=this.mean+this.stdDev*r*o,t=this.mean+this.stdDev*s*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}},uL=class{constructor(e,t,n,r){this.alpha=e,this.beta=1/t,this.dtype=n;let s=r||Math.random();this.randu=n1.alea(s.toString()),this.randn=new r1(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,r,s,a;for(;;){do r=this.randn.nextValue(),a=1+this.c*r;while(a<=0);if(a*=a*a,e=r*r,t=1-.331*e*e,n=.5*e+this.d*(1-a+Math.log(a)),s=this.randu(),s<t||Math.log(s)<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)}},cL=class{constructor(e=0,t=1,n,r){if(this.canReturnFloat=()=>this.dtype==null||this.dtype==="float32",this.min=e,this.range=t-e,this.dtype=n,r==null&&(r=Math.random()),typeof r=="number"&&(r=r.toString()),!this.canReturnFloat()&&this.range<=1)throw new Error(`The difference between ${e} - ${t} <= 1 and dtype is not float`);this.random=n1.alea(r)}convertValue(e){return this.canReturnFloat()?e:Math.round(e)}nextValue(){return this.convertValue(this.min+this.range*this.random())}};function dL(e,t,n=1,r="float32",s){if(n==null&&(n=1),r==null&&(r="float32"),r!=="float32"&&r!=="int32")throw new Error(`Unsupported data type ${r}`);let a=new uL(t,n,r,s),o=Js(e,r);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var hL=H({randomGamma_:dL});function pL(e,t=0,n=1,r,s){if(r!=null&&r==="bool")throw new Error(`Unsupported data type ${r}`);let a=new r1(t,n,r,!1,s),o=Js(e,r);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var fL=H({randomNormal_:pL});function mL(e,t=0,n=1,r="float32",s){let a=Js(e,r),o=new cL(t,n,null,s);for(let i=0;i<a.values.length;i++)a.values[i]=o.nextValue();return a.toTensor()}var r4=H({randomUniform_:mL});function hc(e,t,n=1,r="float32"){if(n===0)throw new Error("Cannot have a step of zero");let s={start:e,stop:t,step:n,dtype:r};return U.runKernel(Rw,{},s)}function gL(e){let n={input:P(e,"input","real")};return U.runKernel(_w,n)}var Tp=H({real_:gL});function yL(e){let n={x:P(e,"x","reciprocal")};return U.runKernel(Dw,n)}var AL=H({reciprocal_:yL});function xL(e){let n={x:P(e,"x","relu")};return U.runKernel(Fw,n)}var Np=H({relu_:xL});function bL(e){let n={x:P(e,"x","relu6")};return U.runKernel(zw,n)}var s4=H({relu6_:bL});function vL(e,t){let r={x:P(e,"x","reverse")},s={dims:t};return U.runKernel(Lw,r,s)}var Io=H({reverse_:vL});function wL(e){let t=P(e,"x","reverse");return L(t.rank===1,()=>`Error in reverse1D: x must be rank 1 but got rank ${t.rank}.`),Io(t,0)}var kL=H({reverse1d_:wL});function IL(e,t){let n=P(e,"x","reverse");return L(n.rank===2,()=>`Error in reverse2D: x must be rank 2 but got rank ${n.rank}.`),Io(n,t)}var SL=H({reverse2d_:IL});function TL(e,t){let n=P(e,"x","reverse");return L(n.rank===3,()=>`Error in reverse3D: x must be rank 3 but got rank ${n.rank}.`),Io(n,t)}var NL=H({reverse3d_:TL});function CL(e,t){let n=P(e,"x","reverse");return L(n.rank===4,()=>`Error in reverse4D: x must be rank 4 but got rank ${n.rank}.`),Io(n,t)}var EL=H({reverse4d_:CL});function $L(e){let n={x:P(e,"x","round")};return U.runKernel(Bw,n)}var a4=H({round_:$L});function RL(e){let n={x:P(e,"x","rsqrt")};return U.runKernel(Ww,n)}var _L=H({rsqrt_:RL});function ut(e,t){if((Nn(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"&&Nn(e)&&!(e instanceof Uint8Array))throw new Error("When making a scalar from encoded string, the value must be `Uint8Array`.");return Ca(e,[],[],t)}function DL(e){let n={x:P(e,"x","selu")};return U.runKernel(Hw,n)}var FL=H({selu_:DL});function ML(e,t,n,r,s,a=[1,1],o="NHWC"){let i=P(e,"x","separableConv2d"),l=P(t,"depthwiseFilter","separableConv2d"),u=P(n,"pointwiseFilter","separableConv2d"),c=i,d=!1;if(i.rank===3&&(d=!0,c=le(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");L(c.rank===4,()=>`Error in separableConv2d: input must be rank 4, but got rank ${c.rank}.`),L(l.rank===4,()=>`Error in separableConv2d: depthwise filter must be rank 4, but got rank ${l.rank}.`),L(u.rank===4,()=>`Error in separableConv2d: pointwise filter must be rank 4, but got rank ${l.rank}.`),L(u.shape[0]===1,()=>`Error in separableConv2d: the first dimension of pointwise filter must be 1, but got ${u.shape[0]}.`),L(u.shape[1]===1,()=>`Error in separableConv2d: the second dimension of pointwise filter must be 1, but got ${u.shape[1]}.`);let h=l.shape[2],p=l.shape[3];L(u.shape[2]===h*p,()=>`Error in separableConv2d: the third dimension of pointwise filter must be ${h*p}, but got ${u.shape[2]}.`);let f=Y2(c,l,r,s,o,a),g=xp(f,u,1,"valid",o);return d?le(g,[g.shape[1],g.shape[2],g.shape[3]]):g}var OL=H({separableConv2d_:ML});async function PL(e,t){let n=P(e,"x","setdiff1d"),r=P(t,"y","setdiff1d");L(n.dtype===r.dtype,()=>`x and y should have the same dtype, but got x (${n.dtype}) and y (${r.dtype}).`),L(n.rank===1,()=>`x should be 1D tensor, but got x (${n.shape}).`),L(r.rank===1,()=>`y should be 1D tensor, but got y (${r.shape}).`);let s=await n.data(),a=await r.data(),o=new Set(a),i=0;for(let c=0;c<s.length;c++)o.has(s[c])||i++;let l=new lp([i],n.dtype),u=new lp([i],"int32");for(let c=0,d=0;c<s.length;c++)o.has(s[c])||(l.values[d]=s[c],u.values[d]=c,d++);return[l.toTensor(),u.toTensor()]}var zL=PL;function LL(e){let n={x:P(e,"x","sign")};return U.runKernel(Kw,n)}var BL=H({sign_:LL});function WL(e){let n={x:P(e,"x","sin")};return U.runKernel(jw,n)}var VL=H({sin_:WL});function UL(e){let n={x:P(e,"x","sinh")};return U.runKernel(qw,n)}var HL=H({sinh_:UL});function GL(e,t,n){let r=P(e,"x","slice1d");return L(r.rank===1,()=>`slice1d expects a rank-1 tensor, but got a rank-${r.rank} tensor`),Xe(r,[t],[n])}var jL=H({slice1d_:GL});function qL(e,t,n){let r=P(e,"x","slice2d");return L(r.rank===2,()=>`slice2d expects a rank-2 tensor, but got a rank-${r.rank} tensor`),Xe(r,t,n)}var KL=H({slice2d_:qL});function XL(e,t,n){let r=P(e,"x","slice3d");return L(r.rank===3,()=>`slice3d expects a rank-3 tensor, but got a rank-${r.rank} tensor`),Xe(r,t,n)}var ZL=H({slice3d_:XL});function YL(e,t,n){let r=P(e,"x","slice4d");return L(r.rank===4,()=>`slice4d expects a rank-4 tensor, but got a rank-${r.rank} tensor`),Xe(r,t,n)}var JL=H({slice4d_:YL});function QL(e,t=-1){let n=P(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 r={logits:n},s={dim:t};return U.runKernel(t7,r,s)}var eB=H({softmax_:QL});function tB(e){L(e.dtype==="complex64",()=>`The dtype for tf.spectral.fft() must be complex64 but got ${e.dtype}.`);let t={input:e};return U.runKernel(Ov,t)}var s1=H({fft_:tB});function nB(e){L(e.dtype==="complex64",()=>`The dtype for tf.spectral.ifft() must be complex64 but got ${e.dtype}.`);let t={input:e};return U.runKernel(jv,t)}var Cp=H({ifft_:nB});function rB(e){let t=e.shape[e.shape.length-1],n=e.size/t,r;if(t<=2){let s=le(e,[n,t]);r=Cp(s)}else{let s=[n,2*(t-1)],a=le(Tp(e),[n,t]),o=le(J2(e),[n,t]),i=Io(Xe(a,[0,1],[n,t-2]),1),l=pe(Io(Xe(o,[0,1],[n,t-2]),1),ut(-1)),u=rn([a,i],1),c=rn([o,l],1),d=le(go(u,c),[s[0],s[1]]);r=Cp(d)}if(r=Tp(r),e.rank===3&&e.shape[0]!==0){let s=r,a=e.shape[0];r=le(r,[a,r.shape[0]/a,r.shape[1]]),s.dispose()}return r}var o4=H({irfft_:rB});function sB(e,t,n=0){let s={x:P(e,"x","split")},a={numOrSizeSplits:t,axis:n};return U.runKernel(e7,s,a)}var na=H({split_:sB});function aB(e,t){L(e.dtype==="float32",()=>`The dtype for rfft() must be real value but got ${e.dtype}`);let n=e.shape[e.shape.length-1],r=e.size/n,s;if(t!=null&&t<n){let f=e.shape.map(g=>0),m=e.shape.map(g=>g);m[e.shape.length-1]=t,s=Xe(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,s=rn([e,Gi(f)],e.shape.length-1),n=t}else s=e;let a=Tr(s),o=le(go(s,a),[r,n]),i=s1(o),l=Math.floor(n/2)+1,u=Tp(i),c=J2(i),d=na(u,[l,n-l],u.shape.length-1),h=na(c,[l,n-l],c.shape.length-1),p=s.shape.slice();return p[s.shape.length-1]=l,le(go(d[0],h[0]),p)}var a1=H({rfft_:aB});function oB(e){let n={x:P(e,"x","sqrt")};return U.runKernel(Yw,n)}var ra=H({sqrt_:oB});function iB(e,t){let n=P(e,"a","squaredDifference"),r=P(t,"b","squaredDifference");[n,r]=Wt(n,r),kn(n.shape,r.shape);let s={a:n,b:r},a={};return U.runKernel(i7,s,a)}var i4=H({squaredDifference_:iB});function lB(e,t){let n=P(e,"x","squeeze");return le(n,R3(n.shape,t).newShape)}var Xn=H({squeeze_:lB});function uB(e,t=0){let n=rc(e,"tensors","stack","string_or_numeric");L(n.length>=1,()=>"Pass at least one tensor to tf.stack"),n.length>0&&L(t<=n[0].rank,()=>"Axis must be <= rank of the tensor");let r=n,s={axis:t};return U.runKernel(Tw,r,s)}var So=H({stack_:uB});function cB(e,t=0){let r={x:P(e,"x","step")},s={alpha:t};return U.runKernel(w7,r,s)}var l4=H({step_:cB});function dB(e,t,n,r,s=0,a=0,o=0,i=0,l=0){let c={x:P(e,"x","stridedSlice","string_or_numeric")},d={begin:t,end:n,strides:r,beginMask:s,endMask:a,ellipsisMask:o,newAxisMask:i,shrinkAxisMask:l};return U.runKernel(l7,c,d)}var hB=H({stridedSlice_:dB});function pB(e){let n={x:P(e,"x","tan")};return U.runKernel(p7,n)}var fB=H({tan_:pB});function ur(e,t){ho(e);let n=ks(e,t);if(n.length!==1)throw new Error("tensor1d() requires values to be a flat/TypedArray");return Ca(e,null,n,t)}function sa(e,t,n){if(ho(e),t!=null&&t.length!==2)throw new Error("tensor2d() requires shape to have two numbers");let r=ks(e,n);if(r.length!==2&&r.length!==1)throw new Error("tensor2d() requires values to be number[][] or flat/TypedArray");if(r.length===1&&t==null)throw new Error("tensor2d() requires shape to be provided when `values` are a flat/TypedArray");return Ca(e,t,r,n)}function mB(e,t,n){if(ho(e),t!=null&&t.length!==4)throw new Error("tensor4d() requires shape to have four numbers");let r=ks(e,n);if(r.length!==4&&r.length!==1)throw new Error("tensor4d() requires values to be number[][][][] or flat/TypedArray");if(r.length===1&&t==null)throw new Error("tensor4d() requires shape to be provided when `values` are a flat array");return Ca(e,t,r,n)}function gB(e,t,n){if(ho(e),t!=null&&t.length!==5)throw new Error("tensor5d() requires shape to have five numbers");let r=ks(e,n);if(r.length!==5&&r.length!==1)throw new Error("tensor5d() requires values to be number[][][][][] or flat/TypedArray");if(r.length===1&&t==null)throw new Error("tensor5d() requires shape to be provided when `values` are a flat array");return Ca(e,t,r,n)}function yB(e,t,n){if(ho(e),t!=null&&t.length!==6)throw new Error("tensor6d() requires shape to have six numbers");let r=ks(e,n);if(r.length!==6&&r.length!==1)throw new Error("tensor6d() requires values to be number[][][][][][] or flat/TypedArray");if(r.length===1&&t==null)throw new Error("tensor6d() requires shape to be provided when `values` are a flat array");return t=t||r,Ca(e,t,r,n)}function AB(e,t=1,n=!0){let r=P(e,"x","topk");if(r.rank===0)throw new Error("topk() expects the input to be of rank 1 or higher");let s=r.shape[r.shape.length-1];if(t<0)throw new Error(`'k' passed to topk() must be >= 0 but got ${t}`);if(t>s)throw new Error(`'k' passed to topk() must be <= the last dimension (${s}) but got ${t}`);let a={x:r},o={k:t,sorted:n},[i,l]=U.runKernel(m7,a,o);return{values:i,indices:l}}var xB=H({topk_:AB});function bB(e,t=0,n=1,r,s){if(r!=null&&r==="bool")throw new Error("Unsupported data type $ { dtype }");let a=new r1(t,n,r,!0,s),o=Js(e,r);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var vB=H({truncatedNormal_:bB});function wB(e,t=0){let n=P(e,"x","unique","string_or_numeric");L(n.rank>0,()=>"The input tensor must be at least 1D");let r={x:n},s={axis:t},[a,o]=U.runKernel(A7,r,s);return{values:a,indices:o}}var kB=H({unique_:wB});function IB(e,t,n){let r=P(e,"x","unsortedSegmentSum"),s=P(t,"segmentIds","unsortedSegmentSum","int32");L(Kn(n),()=>"numSegments must be of dtype int");let a={x:r,segmentIds:s},o={numSegments:n};return U.runKernel(b7,a,o)}var SB=H({unsortedSegmentSum_:IB});function TB(e,t=0){let n=P(e,"x","unstack","string_or_numeric");L(t>=-n.shape.length&&t<n.shape.length,()=>`Axis = ${t} is not in [-${n.shape.length}, ${n.shape.length})`);let r={value:n},s={axis:t};return U.runKernel(x7,r,s)}var pc=H({unstack_:TB});function NB(e,t=!0,n,r){return U.makeVariable(e,t,n,r)}function u4(e,t){let n=[];for(let a=0;a<t.length;a++)t[a]&&n.push(a);let r=Js(e,"int32"),s=Js([n.length,e.length],"int32");for(let a=0;a<n.length;a++){let o=r.indexToLoc(n[a]),i=a*e.length;s.values.set(o,i)}return s.toTensor()}async function CB(e){let t=P(e,"condition","whereAsync","bool"),n=await t.data(),r=u4(t.shape,n);return e!==t&&t.dispose(),r}var c4=CB;async function EB(e,t,n){let r=P(e,"tensor","boolMask"),s=P(t,"mask","boolMask","bool"),a=n==null?0:n,o=s.rank,i=r.shape;L(o>0,()=>"mask cannot be scalar"),Mn(i.slice(a,a+o),s.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=le(r,u),d=le(s,[-1]),h=await c4(d),p=Xn(h,[1]),f=Wk(c,p,a);return e!==r&&r.dispose(),t!==s&&s.dispose(),p.dispose(),c.dispose(),d.dispose(),h.dispose(),f}var $B=EB;function RB(e,t="euclidean",n=null,r=!1){e=P(e,"x","norm");let s=d4(e,t,n),a=s.shape;if(r){let o=Ku(n,e.shape);a=uc(s.shape,o)}return le(s,a)}function d4(e,t,n=null){if(e.rank===0)return Sr(e);if(e.rank!==1&&n===null)return d4(le(e,[-1]),t,n);if(e.rank===1||typeof n=="number"||Array.isArray(n)&&n.length===1){if(t===1)return Et(Sr(e),n);if(t===1/0)return _a(Sr(e),n);if(t===-1/0)return t1(Sr(e),n);if(t==="euclidean"||t===2)return ra(Et(dc(Sr(e),ut(2,"int32")),n));throw new Error(`Error in norm: invalid ord value: ${t}`)}if(Array.isArray(n)&&n.length===2){if(t===1)return _a(Et(Sr(e),n[0]),n[1]-1);if(t===1/0)return _a(Et(Sr(e),n[1]),n[0]);if(t===-1/0)return t1(Et(Sr(e),n[1]),n[0]);if(t==="fro"||t==="euclidean")return ra(Et(rs(e),n));throw new Error(`Error in norm: invalid ord value: ${t}`)}throw new Error(`Error in norm: invalid axis: ${n}`)}var o1=H({norm_:RB});function _B(e,t,n,r,s=!0){let a=P(e,"v","movingAverage"),o=P(t,"x","movingAverage"),i=P(n,"decay","movingAverage");_7(a,o),L(Zs(a.shape,o.shape),()=>"Shape mismatch in v and x");let l=ut(1),u=Ue(l,i),c=pe(Ue(o,a),u);if(s){L(r!=null,()=>"When using zeroDebias: true, step is required.");let d=P(r,"step","movingAverage");c=Je(c,Ue(l,dc(i,d)))}return Me(a,c)}var DB=H({movingAverage_:_B});function FB(e,t,n){let r=P(e,"indices","scatterND","int32"),s=P(t,"updates","scatterND");L2(s,r,n);let a={indices:r,updates:s},o={shape:n};return U.runKernel(Vw,a,o)}var MB=H({scatterND_:FB});function OB(e,t,n,r){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 s=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===s))throw new Error(`sparseValues has incorrect shape ${t.shape}, should be [] or [${s}]`);if(t.dtype!==r.dtype)throw new Error("sparseValues.dtype must match defaultValues.dtype")}function PB(e,t,n,r=0){let s=P(e,"sparseIndices","sparseToDense","int32"),a=P(t,"sparseValues","sparseToDense"),o=P(r,"defaultValue","sparseToDense",a.dtype);OB(s,a,n,o);let i={sparseIndices:s,sparseValues:a,defaultValue:o},l={outputShape:n};return U.runKernel(o7,i,l)}var zB=H({sparseToDense_:PB});function LB(e,t){let n=P(t,"indices","gatherND","int32"),s={params:P(e,"x","gatherND","string_or_numeric"),indices:n};return U.runKernel(Uv,s)}var BB=H({gatherND_:LB});function WB(e,t){if(t==null)return e.shape.slice();if(Zs(e.shape,t))return t;if(e.shape.length===t.length){let n=[];for(let r=0;r<e.shape.length;r++)t[r]==null&&e.shape[r]!=null?n.push(e.shape[r]):n.push(t[r]);return n}return t}function VB(e,t,n,r){let s=P(e,"x","dropout");if(L(s.dtype==="float32",()=>`x has to be a floating point tensor since it's going to be scaled, but got a ${s.dtype} tensor instead.`),L(t>=0&&t<1,()=>`rate must be a float in the range [0, 1), but got ${t}.`),t===0)return e instanceof It?s.clone():s;let a=WB(s,n),o=1-t,i=Je(Bk(Me(r4(a,0,1,"float32",r),o)),o);return pe(s,i)}var UB=H({dropout_:VB});function h4(e){return Math.floor(Math.pow(2,Math.ceil(Math.log(e)/Math.log(2))))}function i1(e,t,n){let r=1-e%2,s=new Float32Array(e);for(let a=0;a<e;++a){let o=2*Math.PI*a/(e+r-1);s[a]=t-n*Math.cos(o)}return ur(s,"float32")}async function HB(e,t,n=1){let r=P(e,"predictions","inTopK"),s=P(t,"targets","inTopK");L(r.rank>1,()=>`inTopK() expects the predictions to be of rank 2 or higher, but got ${r.rank}`),L(r.rank-1===s.rank,()=>`predictions rank should be 1 larger than targets rank, but got predictions rank ${r.rank} and targets rank ${s.rank}`),Mn(r.shape.slice(0,r.shape.length-1),s.shape,"predictions's shape should be align with the targets' shape, except the last dimension.");let a=r.shape[r.shape.length-1];L(n>0&&n<=a,()=>`'k' passed to inTopK() must be > 0 && <= the predictions last dimension (${a}), but got ${n}`);let o=await r.data(),i=await s.data(),[l,u]=[o.length/a,a],c=_3("bool",l);for(let d=0;d<l;d++){let h=d*u,p=o.subarray(h,h+u),f=[];for(let m=0;m<p.length;m++)f.push({value:p[m],index:m});f.sort((m,g)=>g.value-m.value),c[d]=0;for(let m=0;m<n;m++)if(f[m].index===i[d]){c[d]=1;break}}return e!==r&&r.dispose(),t!==s&&s.dispose(),ns(c,s.shape,"bool")}var GB=HB,p4={};_e(p4,{conv2d:()=>XB,depthwiseConv2d:()=>tW,matMul:()=>rW});function jB(e,t,n,r,s,a="NHWC",o){let i=e;e.rank===3&&(i=le(e,[1,e.shape[0],e.shape[1],e.shape[2]]));let l=t;l.rank===3&&(l=le(t,[1,t.shape[0],t.shape[1],t.shape[2]])),L(i.rank===4,()=>`Error in conv2dDerFilter: input must be rank 4, but got shape ${i.shape}.`),L(l.rank===4,()=>`Error in conv2dDerFilter: dy must be rank 4, but got shape ${l.shape}.`),L(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];L(u===n[2],()=>`Error in conv2dDerFilter: depth of input ${u}) must match input depth in filter (${n[2]}.`),L(c===n[3],()=>`Error in conv2dDerFilter: depth of dy (${c}) must match output depth for filter (${n[3]}).`),o!=null&&L(Kn(s),()=>`Error in conv2dDerFilter: pad must be an integer when using, dimRoundingMode ${o} but got pad ${s}.`);let d={x:i,dy:l},h={strides:r,pad:s,dataFormat:a,dimRoundingMode:o,filterShape:n};return U.runKernel(pv,d,h)}var qB=H({conv2DBackpropFilter_:jB});function Ep(e,t,n){if(n==null||n==="linear")return e;if(n==="relu")return pe(e,l4(t));throw new Error(`Cannot compute gradient for fused activation ${n}.`)}function $p(e,t){let n=t,r=Ok(e.shape,t.shape);return r.length>0&&(n=Et(n,r)),le(n,e.shape)}function Rp(e,t,n,r){if(t==="linear")return e;if(t==="relu")return Np(e);if(t==="elu")return zk(e);if(t==="relu6")return s4(e);if(t==="prelu")return n4(e,n);if(t==="leakyrelu")return Uk(e,r);if(t==="sigmoid")return Is(e);throw new Error(`Unknown fused activation ${t}.`)}var _p=(e,t)=>!(e>0)||t==="linear";function KB({x:e,filter:t,strides:n,pad:r,dataFormat:s="NHWC",dilations:a=[1,1],dimRoundingMode:o,bias:i,activation:l="linear",preluActivationWeights:u,leakyreluAlpha:c}){if(l=l||"linear",_p(U.state.gradientDepth,l)===!1){let v=xp(e,t,n,r,s,a,o);return i!=null&&(v=Me(v,i)),Rp(v,l,u,c)}let d=P(e,"x","conv2d"),h=P(t,"filter","conv2d"),p=d,f=!1;d.rank===3&&(f=!0,p=le(d,[1,d.shape[0],d.shape[1],d.shape[2]])),L(p.rank===4,()=>`Error in fused conv2d: input must be rank 4, but got rank ${p.rank}.`),L(h.rank===4,()=>`Error in fused conv2d: filter must be rank 4, but got rank ${h.rank}.`),o!=null&&L(Kn(r),()=>`Error in fused conv2d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${r}.`),L(p.shape[3]===h.shape[2],()=>`Error in conv2d: depth of input (${p.shape[3]}) must match input depth for filter ${h.shape[2]}.`),L(ea(n,a),()=>`Error in conv2D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`),L(s==="NHWC",()=>`Error in conv2d: got dataFormat of ${s} but only NHWC is currently supported.`);let m=ac(p.shape,h.shape,n,a,r,o),g;i!=null&&(g=P(i,"bias","fused conv2d"),[g]=Wt(g,d),kn(m.outShape,g.shape));let y;u!=null&&(y=P(u,"prelu weights","fused conv2d"));let A=(v,I)=>{let[w,S,E,D]=I,$=Ep(v,E,l);L(oc(a),()=>`Error in gradient of fused conv2D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${a}'`);let R=Mk(S.shape,$,w,n,r),N=qB(S,$,w.shape,n,r),M=[R,N];if(D!=null){let B=$p(D,$);M.push(B)}return M},x={x:p,filter:h,bias:g,preluActivationWeights:y},b={strides:n,pad:r,dataFormat:s,dilations:a,dimRoundingMode:o,activation:l,leakyreluAlpha:c};return i==null?Ss((I,w,S)=>{let E=U.runKernel(u2,x,b);return S([w,I,E]),f&&(E=le(E,[E.shape[1],E.shape[2],E.shape[3]])),{value:E,gradFunc:A}})(p,h):Ss((I,w,S,E)=>{let D=U.runKernel(u2,x,b);return E([w,I,D,S]),f&&(D=le(D,[D.shape[1],D.shape[2],D.shape[3]])),{value:D,gradFunc:A}})(p,h,g)}var XB=H({fusedConv2d_:KB});function ZB(e,t,n,r,s,a=[1,1],o){let i=e;e.rank===3&&(i=le(e,[1,e.shape[0],e.shape[1],e.shape[2]]));let l=t;l.rank===3&&(l=le(t,[1,t.shape[0],t.shape[1],t.shape[2]]));let u={x:i,dy:l},c={strides:r,pad:s,dimRoundingMode:o,dilations:a,filterShape:n};return U.runKernel(Iv,u,c)}var YB=H({depthwiseConv2dNativeBackpropFilter_:ZB});function JB(e,t,n,r,s,a=[1,1],o){let i=t,l=!1;t.rank===3&&(l=!0,i=le(t,[1,t.shape[0],t.shape[1],t.shape[2]]));let u={dy:i,filter:n},c={strides:r,pad:s,dimRoundingMode:o,dilations:a,inputShape:e},d=U.runKernel(Sv,u,c);return l?le(d,[d.shape[1],d.shape[2],d.shape[3]]):d}var QB=H({depthwiseConv2dNativeBackpropInput_:JB});function eW({x:e,filter:t,strides:n,pad:r,dataFormat:s="NHWC",dilations:a=[1,1],dimRoundingMode:o,bias:i,activation:l="linear",preluActivationWeights:u,leakyreluAlpha:c}){if(_p(U.state.gradientDepth,l)===!1){let v=Y2(e,t,n,r,s,a,o);return i!=null&&(v=Me(v,i)),Rp(v,l,u,c)}let d=P(e,"x","depthwiseConv2d"),h=P(t,"filter","depthwiseConv2d"),p=d,f=!1;d.rank===3&&(f=!0,p=le(d,[1,d.shape[0],d.shape[1],d.shape[2]])),L(p.rank===4,()=>`Error in fused depthwiseConv2d: input must be rank 4, but got rank ${p.rank}.`),L(h.rank===4,()=>`Error in fused depthwiseConv2d: filter must be rank 4, but got rank ${h.rank}.`),L(p.shape[3]===h.shape[2],()=>`Error in fused depthwiseConv2d: number of input channels (${p.shape[3]}) must match the inChannels dimension in filter ${h.shape[2]}.`),a==null&&(a=[1,1]),L(ea(n,a),()=>`Error in fused depthwiseConv2d: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`),o!=null&&L(Kn(r),()=>`Error in fused depthwiseConv2d: pad must be an integer when using dimRoundingMode ${o} but got pad ${r}.`);let m=ac(p.shape,h.shape,n,a,r,o,!0),g;i!=null&&(g=P(i,"bias","fused conv2d"),[g]=Wt(g,d),kn(m.outShape,g.shape));let y;u!=null&&(y=P(u,"prelu weights","fused depthwiseConv2d"));let A=(v,I)=>{L(oc(a),()=>`Error in gradient of fused depthwiseConv2d: dilation rates greater than 1 are not yet supported. Got dilations '${a}'`);let[w,S,E,D]=I,$=Ep(v,E,l),R=QB(S.shape,$,w,n,r,a,o),N=YB(S,$,w.shape,n,r,a,o);if(D!=null){let M=$p(g,$);return[R,N,M]}return[R,N]},x={x:p,filter:h,bias:g,preluActivationWeights:y},b={strides:n,pad:r,dataFormat:s,dilations:a,dimRoundingMode:o,activation:l,leakyreluAlpha:c};return i==null?Ss((I,w,S)=>{let E=U.runKernel(c2,x,b);return S([w,I,E]),f&&(E=le(E,[E.shape[1],E.shape[2],E.shape[3]])),{value:E,gradFunc:A}})(p,h):Ss((I,w,S,E)=>{let D=U.runKernel(c2,x,b);return E([w,I,D,S]),f&&(D=le(D,[D.shape[1],D.shape[2],D.shape[3]])),{value:D,gradFunc:A}})(p,h,g)}var tW=H({fusedDepthwiseConv2d_:eW});function nW({a:e,b:t,transposeA:n=!1,transposeB:r=!1,bias:s,activation:a="linear",preluActivationWeights:o,leakyreluAlpha:i}){if(_p(U.state.gradientDepth,a)===!1){let D=gt(e,t,n,r);return s!=null&&(D=Me(D,s)),Rp(D,a,o,i)}let l=P(e,"a","fused matMul"),u=P(t,"b","fused matMul");[l,u]=Wt(l,u);let c=n?l.shape[l.rank-2]:l.shape[l.rank-1],d=r?u.shape[u.rank-1]:u.shape[u.rank-2],h=n?l.shape[l.rank-1]:l.shape[l.rank-2],p=r?u.shape[u.rank-2]:u.shape[u.rank-1],f=l.shape.slice(0,-2),m=u.shape.slice(0,-2),g=Yt(f),y=Yt(m);L(l.rank>=2&&u.rank>=2&&l.rank===u.rank,()=>`Error in fused matMul: inputs must have the same rank of at least 2, got ranks ${l.rank} and ${u.rank}.`),L(Zs(f,m),()=>`Error in fused matMul: outer dimensions (${f}) and (${m}) of Tensors with shapes ${l.shape} and ${u.shape} must match.`),L(c===d,()=>`Error in fused matMul: inner shapes (${c}) and (${d}) of Tensors with shapes ${l.shape} and ${u.shape} and transposeA=${n} and transposeB=${r} must match.`);let A=l.shape.slice(0,-2).concat([h,p]),x=n?le(l,[g,c,h]):le(l,[g,h,c]),b=r?le(u,[y,p,d]):le(u,[y,d,p]),v;s!=null&&(v=P(s,"bias","fused matMul"),[v]=Wt(v,l),kn(A,v.shape));let I;o!=null&&(I=P(o,"prelu weights","fused matMul"));let w=(D,$)=>{let[R,N,M,B]=$,q=Ep(le(D,M.shape),M,a),X,J;if(!n&&!r?(X=gt(q,N,!1,!0),J=gt(R,q,!0,!1)):!n&&r?(X=gt(q,N,!1,!1),J=gt(q,R,!0,!1)):n&&!r?(X=gt(N,q,!1,!0),J=gt(R,q,!1,!1)):(X=gt(N,q,!0,!0),J=gt(q,R,!0,!0)),s!=null){let ee=$p(B,q);return[X,J,ee]}else return[X,J]},S={a:x,b,bias:v,preluActivationWeights:I},E={transposeA:n,transposeB:r,activation:a,leakyreluAlpha:i};return s==null?Ss(($,R,N)=>{let M=U.runKernel(l2,S,E);return N([$,R,M]),{value:le(M,A),gradFunc:w}})(x,b):Ss(($,R,N,M)=>{let B=U.runKernel(l2,S,E);return M([$,R,B,N]),{value:le(B,A),gradFunc:w}})(x,b,v)}var rW=H({fusedMatMul_:nW});function sW(e){return i1(e,.54,.46)}var aW=H({hammingWindow_:sW});function oW(e){return i1(e,.5,.5)}var f4=H({hannWindow_:oW});function iW(e,t,n,r=!1,s=0){let a=0,o=[];for(;a+t<=e.size;)o.push(Xe(e,a,t)),a+=n;if(r)for(;a<e.size;){let i=a+t-e.size,l=rn([Xe(e,a,t-i),vp([i],s)]);o.push(l),a+=n}return o.length===0?sa([],[0,t]):le(rn(o),[o.length,t])}var m4=H({frame_:iW});function lW(e,t,n,r,s=f4){r==null&&(r=h4(t));let a=m4(e,t,n),o=pe(a,s(t));return a1(o,r)}var uW=H({stft_:lW});function cW(e,t,n,r,s="bilinear",a=0){let o=P(e,"image","cropAndResize"),i=P(t,"boxes","cropAndResize","float32"),l=P(n,"boxInd","cropAndResize","int32"),u=i.shape[0];L(o.rank===4,()=>`Error in cropAndResize: image must be rank 4,but got rank ${o.rank}.`),L(i.rank===2&&i.shape[1]===4,()=>`Error in cropAndResize: boxes must be have size [${u},4] but had shape ${i.shape}.`),L(l.rank===1&&l.shape[0]===u,()=>`Error in cropAndResize: boxInd must be have size [${u}] but had shape ${i.shape}.`),L(r.length===2,()=>`Error in cropAndResize: cropSize must be of length 2, but got length ${r.length}.`),L(r[0]>=1&&r[1]>=1,()=>`cropSize must be atleast [1,1], but was ${r}`),L(s==="bilinear"||s==="nearest",()=>`method must be bilinear or nearest, but was ${s}`);let c={image:o,boxes:i,boxInd:l},d={method:s,extrapolationValue:a,cropSize:r};return U.runKernel(bv,c,d)}var dW=H({cropAndResize_:cW});function hW(e){let t=P(e,"image","flipLeftRight","float32");L(t.rank===4,()=>`Error in flipLeftRight: image must be rank 4,but got rank ${t.rank}.`);let n={image:t};return U.runKernel(zv,n,{})}var pW=H({flipLeftRight_:hW});function fW(e,t,n=0,r=.5){let s=P(e,"image","rotateWithOffset","float32");L(s.rank===4,()=>`Error in rotateWithOffset: image must be rank 4,but got rank ${s.rank}.`);let a={image:s},o={radians:t,fillValue:n,center:r};return U.runKernel(k7,a,o)}var mW=H({rotateWithOffset_:fW});function ji(e,t,n,r,s,a){r==null&&(r=.5),s==null&&(s=Number.NEGATIVE_INFINITY),a==null&&(a=0);let o=e.shape[0];return n=Math.min(n,o),L(0<=r&&r<=1,()=>`iouThreshold must be in [0, 1], but was '${r}'`),L(e.rank===2,()=>`boxes must be a 2D tensor, but was of rank '${e.rank}'`),L(e.shape[1]===4,()=>`boxes must have 4 columns, but 2nd dimension was ${e.shape[1]}`),L(t.rank===1,()=>"scores must be a 1D tensor"),L(t.shape[0]===o,()=>`scores has incompatible shape with boxes. Expected ${o}, but was ${t.shape[0]}`),L(0<=a&&a<=1,()=>`softNmsSigma must be in [0, 1], but was '${a}'`),{maxOutputSize:n,iouThreshold:r,scoreThreshold:s,softNmsSigma:a}}function gW(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY){let a=P(e,"boxes","nonMaxSuppression"),o=P(t,"scores","nonMaxSuppression"),i=ji(a,o,n,r,s);n=i.maxOutputSize,r=i.iouThreshold,s=i.scoreThreshold;let l={maxOutputSize:n,iouThreshold:r,scoreThreshold:s};return U.runKernel(vw,{boxes:a,scores:o},l)}var yW=H({nonMaxSuppression_:gW});function AW(e,t,n){let r=xW(e,t,n),s=r<0?-(r+1):r;e.splice(s,0,t)}function xW(e,t,n){return vW(e,t,n||bW)}function bW(e,t){return e>t?1:e<t?-1:0}function vW(e,t,n){let r=0,s=e.length,a=0,o=!1;for(;r<s;){a=r+(s-r>>>1);let i=n(t,e[a]);i>0?r=a+1:(s=a,o=!i)}return o?r:-r-1}function g4(e,t,n,r,s){return l1(e,t,n,r,s,0)}function y4(e,t,n,r,s,a){return l1(e,t,n,r,s,0,!1,a,!0)}function A4(e,t,n,r,s,a){return l1(e,t,n,r,s,a,!0)}function l1(e,t,n,r,s,a,o=!1,i=!1,l=!1){let u=[];for(let g=0;g<t.length;g++)t[g]>s&&u.push({score:t[g],boxIndex:g,suppressBeginIndex:0});u.sort(x4);let c=a>0?-.5/a:0,d=[],h=[];for(;d.length<n&&u.length>0;){let g=u.pop(),{score:y,boxIndex:A,suppressBeginIndex:x}=g;if(y<s)break;let b=!1;for(let v=d.length-1;v>=x;--v){let I=wW(e,A,d[v]);if(I>=r){b=!0;break}if(g.score=g.score*kW(r,c,I),g.score<=s)break}g.suppressBeginIndex=d.length,b||(g.score===y?(d.push(A),h.push(g.score)):g.score>s&&AW(u,g,x4))}let p=d.length,f=n-p;i&&f>0&&(d.push(...new Array(f).fill(0)),h.push(...new Array(f).fill(0)));let m={selectedIndices:d};return o&&(m.selectedScores=h),l&&(m.validOutputs=p),m}function wW(e,t,n){let r=e.subarray(t*4,t*4+4),s=e.subarray(n*4,n*4+4),a=Math.min(r[0],r[2]),o=Math.min(r[1],r[3]),i=Math.max(r[0],r[2]),l=Math.max(r[1],r[3]),u=Math.min(s[0],s[2]),c=Math.min(s[1],s[3]),d=Math.max(s[0],s[2]),h=Math.max(s[1],s[3]),p=(i-a)*(l-o),f=(d-u)*(h-c);if(p<=0||f<=0)return 0;let m=Math.max(a,u),g=Math.max(o,c),y=Math.min(i,d),A=Math.min(l,h),x=Math.max(y-m,0)*Math.max(A-g,0);return x/(p+f-x)}function kW(e,t,n){let r=Math.exp(t*n*n);return n<=e?r:0}function x4(e,t){return e.score-t.score||e.score===t.score&&t.boxIndex-e.boxIndex}async function IW(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY){let a=P(e,"boxes","nonMaxSuppressionAsync"),o=P(t,"scores","nonMaxSuppressionAsync"),i=ji(a,o,n,r,s);n=i.maxOutputSize,r=i.iouThreshold,s=i.scoreThreshold;let l=await Promise.all([a.data(),o.data()]),u=l[0],c=l[1],{selectedIndices:d}=g4(u,c,n,r,s);return a!==e&&a.dispose(),o!==t&&o.dispose(),ur(d,"int32")}var SW=IW;function TW(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY,a=0){let o=P(e,"boxes","nonMaxSuppression"),i=P(t,"scores","nonMaxSuppression"),l=ji(o,i,n,r,s,a);n=l.maxOutputSize,r=l.iouThreshold,s=l.scoreThreshold,a=l.softNmsSigma;let u={boxes:o,scores:i},c={maxOutputSize:n,iouThreshold:r,scoreThreshold:s,softNmsSigma:a},d=U.runKernel(kw,u,c);return{selectedIndices:d[0],selectedScores:d[1]}}var NW=H({nonMaxSuppressionWithScore_:TW});async function CW(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY,a=0){let o=P(e,"boxes","nonMaxSuppressionAsync"),i=P(t,"scores","nonMaxSuppressionAsync"),l=ji(o,i,n,r,s,a);n=l.maxOutputSize,r=l.iouThreshold,s=l.scoreThreshold,a=l.softNmsSigma;let u=await Promise.all([o.data(),i.data()]),c=u[0],d=u[1],{selectedIndices:h,selectedScores:p}=A4(c,d,n,r,s,a);return o!==e&&o.dispose(),i!==t&&i.dispose(),{selectedIndices:ur(h,"int32"),selectedScores:ur(p)}}var EW=CW;function $W(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY,a=!1){let o=P(e,"boxes","nonMaxSuppression"),i=P(t,"scores","nonMaxSuppression"),l=ji(o,i,n,r,s,null),u=l.maxOutputSize,c=l.iouThreshold,d=l.scoreThreshold,h={boxes:o,scores:i},p={maxOutputSize:u,iouThreshold:c,scoreThreshold:d,padToMaxOutputSize:a},f=U.runKernel(ww,h,p);return{selectedIndices:f[0],validOutputs:f[1]}}var RW=H({nonMaxSuppressionPadded_:$W});async function _W(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY,a=!1){let o=P(e,"boxes","nonMaxSuppressionAsync"),i=P(t,"scores","nonMaxSuppressionAsync"),l=ji(o,i,n,r,s,null),u=l.maxOutputSize,c=l.iouThreshold,d=l.scoreThreshold,[h,p]=await Promise.all([o.data(),i.data()]),{selectedIndices:f,validOutputs:m}=y4(h,p,u,c,d,a);return o!==e&&o.dispose(),i!==t&&i.dispose(),{selectedIndices:ur(f,"int32"),validOutputs:ut(m,"int32")}}var DW=_W;function FW(e,t,n=!1,r=!1){let s=P(e,"images","resizeBilinear");L(s.rank===3||s.rank===4,()=>`Error in resizeBilinear: x must be rank 3 or 4, but got rank ${s.rank}.`),L(t.length===2,()=>`Error in resizeBilinear: new shape must 2D, but got shape ${t}.`),L(r===!1||n===!1,()=>"Error in resizeBilinear: If halfPixelCenters is true, alignCorners must be false.");let a=s,o=!1;s.rank===3&&(o=!0,a=le(s,[1,s.shape[0],s.shape[1],s.shape[2]]));let[]=t,i={images:a},l={alignCorners:n,halfPixelCenters:r,size:t},u=U.runKernel(Pw,i,l);return o?le(u,[u.shape[1],u.shape[2],u.shape[3]]):u}var MW=H({resizeBilinear_:FW});function OW(e,t,n=!1,r=!1){let s=P(e,"images","resizeNearestNeighbor");L(s.rank===3||s.rank===4,()=>`Error in resizeNearestNeighbor: x must be rank 3 or 4, but got rank ${s.rank}.`),L(t.length===2,()=>`Error in resizeNearestNeighbor: new shape must 2D, but got shape ${t}.`),L(s.dtype==="float32"||s.dtype==="int32",()=>"`images` must have `int32` or `float32` as dtype"),L(r===!1||n===!1,()=>"Error in resizeNearestNeighbor: If halfPixelCenters is true, alignCorners must be false.");let a=s,o=!1;s.rank===3&&(o=!0,a=le(s,[1,s.shape[0],s.shape[1],s.shape[2]]));let[]=t,i={images:a},l={alignCorners:n,halfPixelCenters:r,size:t},u=U.runKernel(Ow,i,l);return o?le(u,[u.shape[1],u.shape[2],u.shape[3]]):u}var PW=H({resizeNearestNeighbor_:OW});function zW(e,t="binary",n=!1,r=.5){let s=P(e,"image","threshold"),a=.2989,o=.587,i=.114,l=s.shape[0]*s.shape[1],u=pe(ur([r]),255),c,d,h,p;if(L(s.rank===3,()=>`Error in threshold: image must be rank 3,but got rank ${s.rank}.`),L(s.shape[2]===3||s.shape[2]===1,()=>`Error in threshold: image color channel must be equal to 3 or 1but got ${s.shape[2]}.`),L(s.dtype==="int32"||s.dtype==="float32",()=>`Error in dtype: image dtype must be int32 or float32,but got dtype ${s.dtype}.`),L(t==="otsu"||t==="binary",()=>`Method must be binary or otsu, but was ${t}`),s.shape[2]===3){[c,d,h]=na(s,[1,1,1],-1);let g=pe(c,a),y=pe(d,o),A=pe(h,i);p=Me(Me(g,y),A)}else p=e;if(t==="otsu"){let g=Fk(Mt(a4(p),"int32"),ns([]),256);u=LW(g,l)}let f=n?Q2(p,u):wp(p,u);return Mt(pe(f,255),"int32")}function LW(e,t){let n=ur([-1]),r=ur([0]),s=ur([0]),a,o,i,l,u,c;for(let d=0;d<e.size-1;d++){a=Xe(e,0,d+1),o=Xe(e,d+1),u=Je(Et(a),t),c=Je(Et(o),t);let h=Et(pe(a,hc(0,a.size)));i=Je(h,Et(a));let p=vp(o.shape,a.size),f=Me(hc(0,o.size),p),m=pe(o,f);l=Je(Et(m),Et(o));let g=Ue(i,l),y=Ue(i,l),A=pe(u,c);s=pe(pe(A,g),y);let x=wp(s,r);r=Hi(x,s,r),n=Hi(x,ur([d]),n)}return n}var BW=H({threshold_:zW});function WW(e,t,n="nearest",r="constant",s=0,a){let o=P(e,"image","transform","float32"),i=P(t,"transforms","transform","float32");L(o.rank===4,()=>`Error in transform: image must be rank 4,but got rank ${o.rank}.`),L(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"),L(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:r,fillValue:s,outputShape:a};return U.runKernel(g7,l,u)}var VW=H({transform_:WW});function UW(e,t,n){L(t%1==0,()=>`bandPart(): numLower must be an integer, got ${t}.`),L(n%1==0,()=>`bandPart(): numUpper must be an integer, got ${n}.`);let r=P(e,"a","bandPart");L(r.rank>=2,()=>`bandPart(): Rank must be at least 2, got ${r.rank}.`);let s=r.shape,[a,o]=r.shape.slice(-2);if(!(t<=a))throw new Error(`bandPart(): numLower (${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 (${o}).`);t<0&&(t=a),n<0&&(n=o);let i=le(hc(0,a,1,"int32"),[-1,1]),l=hc(0,o,1,"int32"),u=Ue(i,l),c=Ip(Q2(u,ut(+t,"int32")),Vk(u,ut(-n,"int32"))),d=Gi([a,o],r.dtype);return le(So(pc(le(r,[-1,a,o])).map(h=>Hi(c,h,d))),s)}var HW=H({bandPart_:UW});function GW(e){let t;if(Array.isArray(e)){t=!1,L(e!=null&&e.length>0,()=>"Gram-Schmidt process: input must not be null, undefined, or empty");let s=e[0].shape[0];for(let a=1;a<e.length;++a)L(e[a].shape[0]===s,()=>`Gram-Schmidt: Non-unique lengths found in the input vectors: (${e[a].shape[0]} vs. ${s})`)}else t=!0,e=na(e,e.shape[0],0).map(s=>Xn(s,[0]));L(e.length<=e[0].shape[0],()=>`Gram-Schmidt: Number of vectors (${e.length}) exceeds number of dimensions (${e[0].shape[0]}).`);let n=[],r=e;for(let s=0;s<e.length;++s)n.push(U.tidy(()=>{let a=r[s];if(s>0)for(let o=0;o<s;++o){let i=pe(Et(pe(n[o],a)),n[o]);a=Ue(a,i)}return Je(a,o1(a,"euclidean"))}));return t?So(n,0):n}var jW=H({gramSchmidt_:GW});function qW(e,t=!1){if(L(e.rank>=2,()=>`qr() requires input tensor to have a rank >= 2, but got rank ${e.rank}`),e.rank===2)return b4(e,t);{let n=e.shape.slice(0,e.shape.length-2).reduce((l,u)=>l*u),r=pc(le(e,[n,e.shape[e.shape.length-2],e.shape[e.shape.length-1]]),0),s=[],a=[];r.forEach(l=>{let[u,c]=b4(l,t);s.push(u),a.push(c)});let o=le(So(s,0),e.shape),i=le(So(a,0),e.shape);return[o,i]}}function b4(e,t=!1){return U.tidy(()=>{L(e.shape.length===2,()=>`qr2d() requires a 2D Tensor, but got a ${e.shape.length}D Tensor.`);let n=e.shape[0],r=e.shape[1],s=Lk(n),a=Qs(e),o=sa([[1]],[1,1]),i=Qs(o),l=n>=r?r:n;for(let u=0;u<l;++u){let c=a,d=i,h=s;[i,a,s]=U.tidy(()=>{let p=Xe(a,[u,u],[n-u,1]),f=o1(p),m=Xe(a,[u,u],[1,1]),g=Hi(wp(m,0),sa([[-1]]),sa([[1]])),y=Ue(m,pe(g,f)),A=Je(p,y);A.shape[0]===1?i=Qs(o):i=rn([o,Xe(A,[1,0],[A.shape[0]-1,A.shape[1]])],0);let x=Ra(Je(gt(g,y),f)),b=Xe(a,[u,0],[n-u,r]),v=pe(x,i),I=pp(i);if(u===0)a=Ue(b,gt(v,gt(I,b)));else{let E=Ue(b,gt(v,gt(I,b)));a=rn([Xe(a,[0,0],[u,r]),E],0)}let w=pp(v),S=Xe(s,[0,u],[n,s.shape[1]-u]);if(u===0)s=Ue(S,gt(gt(S,i),w));else{let E=Ue(S,gt(gt(S,i),w));s=rn([Xe(s,[0,0],[n,u]),E],1)}return[i,a,s]}),We([c,d,h])}return!t&&n>r&&(s=Xe(s,[0,0],[n,r]),a=Xe(a,[0,0],[r,r])),[s,a]})}var KW=H({qr_:qW}),Pn;(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"})(Pn||(Pn={}));function XW(e,t,n=Pn.SUM_BY_NONZERO_WEIGHTS){let r=P(e,"losses","computeWeightedLoss"),s=null;t!=null&&(s=P(t,"weights","computeWeightedLoss"));let a=s==null?r:pe(r,s);if(n===Pn.NONE)return a;if(n===Pn.SUM)return Et(a);if(n===Pn.MEAN){if(s==null)return Sp(a);{let o=r.size/s.size,i=Je(Et(a),Et(s));return o>1?Je(i,ut(o)):i}}if(n===Pn.SUM_BY_NONZERO_WEIGHTS){if(s==null)return Je(Et(a),ut(r.size));{let o=pe(s,ko(r.shape)),i=Mt(Et(e4(o,ut(0))),"float32");return Je(Et(a),i)}}throw Error(`Unknown reduction: ${n}`)}var aa=H({computeWeightedLoss_:XW});function ZW(e,t,n,r=Pn.SUM_BY_NONZERO_WEIGHTS){let s=P(e,"labels","absoluteDifference"),a=P(t,"predictions","absoluteDifference"),o=null;n!=null&&(o=P(n,"weights","absoluteDifference")),Mn(s.shape,a.shape,"Error in absoluteDifference: ");let i=Sr(Ue(s,a));return aa(i,o,r)}var YW=H({absoluteDifference_:ZW});function JW(e,t,n,r,s=Pn.SUM_BY_NONZERO_WEIGHTS){let a=P(e,"labels","cosineDistance"),o=P(t,"predictions","cosineDistance"),i=null;r!=null&&(i=P(r,"weights","cosineDistance")),Mn(a.shape,o.shape,"Error in cosineDistance: ");let l=ut(1),u=Ue(l,Et(pe(a,o),n,!0));return aa(u,i,s)}var QW=H({cosineDistance_:JW});function eV(e,t,n,r=Pn.SUM_BY_NONZERO_WEIGHTS){let s=P(e,"labels","hingeLoss"),a=P(t,"predictions","hingeLoss"),o=null;n!=null&&(o=P(n,"weights","hingeLoss")),Mn(s.shape,a.shape,"Error in hingeLoss: ");let i=ut(1);s=Ue(pe(ut(2),s),i);let l=Np(Ue(i,pe(s,a)));return aa(l,o,r)}var tV=H({hingeLoss_:eV});function nV(e,t,n,r=1,s=Pn.SUM_BY_NONZERO_WEIGHTS){let a=P(e,"labels","huberLoss"),o=P(t,"predictions","huberLoss"),i=null;n!=null&&(i=P(n,"weights","huberLoss")),Mn(a.shape,o.shape,"Error in huberLoss: ");let l=ut(r),u=Sr(Ue(o,a)),c=Qk(u,l),d=Ue(u,c),h=Me(pe(ut(.5),rs(c)),pe(l,d));return aa(h,i,s)}var rV=H({huberLoss_:nV});function sV(e,t,n,r=1e-7,s=Pn.SUM_BY_NONZERO_WEIGHTS){let a=P(e,"labels","logLoss"),o=P(t,"predictions","logLoss"),i=null;n!=null&&(i=P(n,"weights","logLoss")),Mn(a.shape,o.shape,"Error in logLoss: ");let l=ut(1),u=ut(r),c=Ra(pe(a,lc(Me(o,u)))),d=pe(Ue(l,a),lc(Me(Ue(l,o),u))),h=Ue(c,d);return aa(h,i,s)}var aV=H({logLoss_:sV});function oV(e,t,n,r=Pn.SUM_BY_NONZERO_WEIGHTS){let s=P(e,"labels","meanSquaredError"),a=P(t,"predictions","meanSquaredError"),o=null;n!=null&&(o=P(n,"weights","meanSquaredError")),Mn(s.shape,a.shape,"Error in meanSquaredError: ");let i=i4(s,a);return aa(i,o,r)}var iV=H({meanSquaredError_:oV});function lV(e,t){let n=P(e,"labels","sigmoidCrossEntropyWithLogits"),r=P(t,"logits","sigmoidCrossEntropyWithLogits");Mn(n.shape,r.shape,"Error in sigmoidCrossEntropyWithLogits: ");let s=Np(r),a=pe(r,n),o=Hk(wo(Ra(Sr(r))));return Me(Ue(s,a),o)}function uV(e,t,n,r=0,s=Pn.SUM_BY_NONZERO_WEIGHTS){let a=P(e,"multiClassLabels","sigmoidCrossEntropy"),o=P(t,"logits","sigmoidCrossEntropy"),i=null;if(n!=null&&(i=P(n,"weights","sigmoidCrossEntropy")),Mn(a.shape,o.shape,"Error in sigmoidCrossEntropy: "),r>0){let u=ut(r),c=ut(1),d=ut(.5);a=Me(pe(a,Ue(c,u)),pe(d,u))}let l=lV(a,o);return aa(l,i,s)}var cV=H({sigmoidCrossEntropy_:uV});function dV(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 Ss((s,a,o)=>{let l=Kk(a,[n],!0),u=Ue(Mt(a,"float32"),l);o([s,u]);let c=Ra(pe(u,s));return{value:Et(c,[n]),gradFunc:(p,f)=>{let[m,g]=f,y=uc(p.shape,[n]);return[pe(le(p,y),Ue(Mt(m,"float32"),wo(g))),pe(le(p,y),Ue(wo(g),Mt(m,"float32")))]}}})(e,t)}function hV(e,t,n,r=0,s=Pn.SUM_BY_NONZERO_WEIGHTS){let a=P(e,"onehotLabels","softmaxCrossEntropy"),o=P(t,"logits","softmaxCrossEntropy"),i=null;if(n!=null&&(i=P(n,"weights","softmaxCrossEntropy")),Mn(a.shape,o.shape,"Error in softmaxCrossEntropy: "),r>0){let u=ut(r),c=ut(1),d=ut(a.shape[1]);a=Me(pe(a,Ue(c,u)),Je(u,d))}let l=dV(a,o);return aa(l,i,s)}var pV=H({softmaxCrossEntropy_:hV});function fV(e,t,n,r){let s=P(e,"indices","sparseFillEmptyRows"),a=P(t,"values","sparseFillEmptyRows"),o=P(n,"denseShape","sparseFillEmptyRows"),i=P(r,"defaultValue","sparseFillEmptyRows",a.dtype);if(s.rank!==2)throw new Error(`Indices should be Tensor2D but received shape
${s.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:s,values:a,denseShape:o,defaultValue:i},u=U.runKernel(n7,l);return{outputIndices:u[0],outputValues:u[1],emptyRowIndicator:u[2],reverseIndexMap:u[3]}}var mV=H({sparseFillEmptyRows_:fV});function gV(e,t,n){let r=P(e,"inputIndices","sparseReshape"),s=P(t,"inputShape","sparseReshape"),a=P(n,"newShape","sparseReshape");if(r.rank!==2)throw new Error(`Input indices should be Tensor2D but received shape
${r.shape}`);if(s.rank!==1)throw new Error(`Input shape should be Tensor1D but received shape ${s.shape}`);if(a.rank!==1)throw new Error(`New shape should be Tensor1D but received shape ${a.shape}`);let o={inputIndices:r,inputShape:s,newShape:a},i=U.runKernel(r7,o);return{outputIndices:i[0],outputShape:i[1]}}var yV=H({sparseReshape_:gV});function AV(e,t,n){let r=P(e,"data","sparseSegmentMean"),s=P(t,"indices","sparseSegmentMean"),a=P(n,"segmentIds","sparseSegmentMean");if(r.rank<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(s.rank!==1)throw new Error(`Indices should be Tensor1D but received shape
${s.shape}`);if(a.rank!==1)throw new Error(`Segment ids should be Tensor1D but received shape
${a.shape}`);let o={data:r,indices:s,segmentIds:a};return U.runKernel(s7,o)}var xV=H({sparseSegmentMean_:AV});function bV(e,t,n){let r=P(e,"data","sparseSegmentSum"),s=P(t,"indices","sparseSegmentSum"),a=P(n,"segmentIds","sparseSegmentSum");if(r.rank<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(s.rank!==1)throw new Error(`Indices should be Tensor1D but received shape
${s.shape}`);if(a.rank!==1)throw new Error(`Segment ids should be Tensor1D but received shape
${a.shape}`);let o={data:r,indices:s,segmentIds:a};return U.runKernel(a7,o)}var vV=H({sparseSegmentSum_:bV});function wV(e,t,n,r,s,a,o,i){let l=P(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=P(t,"dataSplits","stringNGrams");if(u.dtype!=="int32")throw new Error("Data splits must be of datatype int32");let c={separator:n,nGramWidths:r,leftPad:s,rightPad:a,padWidth:o,preserveShortSequences:i},d={data:l,dataSplits:u},h=U.runKernel(u7,d,c);return{nGrams:h[0],nGramsSplits:h[1]}}var kV=H({stringNGrams_:wV});function IV(e,t,n=!0){let r=P(e,"input","stringSplit","string"),s=P(t,"delimiter","stringSplit","string");if(r.rank!==1)throw new Error(`Input should be Tensor1D but received shape ${r.shape}`);if(s.rank!==0)throw new Error(`Delimiter should be a scalar but received shape ${s.shape}`);let a={skipEmpty:n},o={input:r,delimiter:s},i=U.runKernel(c7,o,a);return{indices:i[0],values:i[1],shape:i[2]}}var SV=H({stringSplit_:IV});function TV(e,t){let n=P(e,"input","stringToHashBucketFast","string"),r={numBuckets:t};if(t<=0)throw new Error("Number of buckets must be at least 1");let s={input:n};return U.runKernel(d7,s,r)}var NV=H({stringToHashBucketFast_:TV}),CV={fft:s1,ifft:Cp,rfft:a1,irfft:o4},EV={hammingWindow:aW,hannWindow:f4,frame:m4,stft:uW},Ze={flipLeftRight:pW,resizeNearestNeighbor:PW,resizeBilinear:MW,rotateWithOffset:mW,cropAndResize:dW,nonMaxSuppression:yW,nonMaxSuppressionAsync:SW,nonMaxSuppressionWithScore:NW,nonMaxSuppressionWithScoreAsync:EW,nonMaxSuppressionPadded:RW,nonMaxSuppressionPaddedAsync:DW,threshold:BW,transform:VW},$V={bandPart:HW,gramSchmidt:jW,qr:KW},RV={absoluteDifference:YW,computeWeightedLoss:aa,cosineDistance:QW,hingeLoss:tV,huberLoss:rV,logLoss:aV,meanSquaredError:iV,sigmoidCrossEntropy:cV,softmaxCrossEntropy:pV},_V={sparseFillEmptyRows:mV,sparseReshape:yV,sparseSegmentMean:xV,sparseSegmentSum:vV},DV={stringNGrams:kV,stringSplit:SV,stringToHashBucketFast:NV},Da=class extends wk{minimize(e,t=!1,n){let{value:r,grads:s}=this.computeGradients(e,n);if(n!=null){let a=n.map(o=>({name:o.name,tensor:s[o.name]}));this.applyGradients(a)}else this.applyGradients(s);return We(s),t?r:(r.dispose(),null)}get iterations(){return this.iterations_==null&&(this.iterations_=0),this.iterations_}incrementIterations(){this.iterations_=this.iterations+1}computeGradients(e,t){return Gk(e,t)}dispose(){this.iterations_!=null&&We(this.iterations_)}async saveIterations(){return this.iterations_==null&&(this.iterations_=0),{name:"iter",tensor:ut(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 ${this.getClassName()}`)}async extractIterations(e){return this.iterations_=(await e[0].tensor.data())[0],e.slice(1)}};Object.defineProperty(Da,Symbol.hasInstance,{value:e=>e.minimize!=null&&e.computeGradients!=null&&e.applyGradients!=null});var Dp=class extends Da{constructor(e,t,n=null){super();this.learningRate=e,this.rho=t,this.epsilon=n,this.accumulatedGrads=[],this.accumulatedUpdates=[],n==null&&(this.epsilon=U.backend.epsilon())}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=U.registeredVariables[n],a=!1;this.accumulatedGrads[r]==null&&(this.accumulatedGrads[r]={originalName:`${n}/accum_grad`,variable:Ve(()=>Tr(s).variable(a))}),this.accumulatedUpdates[r]==null&&(this.accumulatedUpdates[r]={originalName:`${n}/accum_var`,variable:Ve(()=>Tr(s).variable(a))});let o=Array.isArray(e)?e[r].tensor:e[n];if(o==null)return;let i=this.accumulatedGrads[r].variable,l=this.accumulatedUpdates[r].variable;Ve(()=>{let u=Me(pe(i,this.rho),pe(rs(o),1-this.rho)),c=pe(Je(ra(Me(l,this.epsilon)),ra(Me(i,this.epsilon))),o),d=Me(pe(l,this.rho),pe(rs(c),1-this.rho));i.assign(u),l.assign(d);let h=Me(pe(c,-this.learningRate),s);s.assign(h)})}),this.incrementIterations()}dispose(){this.accumulatedUpdates!=null&&(We(this.accumulatedGrads.map(e=>e.variable)),We(this.accumulatedUpdates.map(e=>e.variable)))}async getWeights(){let e=[...this.accumulatedGrads,...this.accumulatedUpdates];return[await this.saveIterations()].concat(e.map(t=>({name:t.originalName,tensor:t.variable})))}async setWeights(e){e=await this.extractIterations(e);let t=e.length/2,n=!1;this.accumulatedGrads=e.slice(0,t).map(r=>({originalName:r.name,variable:r.tensor.variable(n)})),this.accumulatedUpdates=e.slice(t,t*2).map(r=>({originalName:r.name,variable:r.tensor.variable(n)}))}getConfig(){return{learningRate:this.learningRate,rho:this.rho,epsilon:this.epsilon}}static fromConfig(e,t){return new e(t.learningRate,t.rho,t.epsilon)}};Dp.className="Adadelta";$a(Dp);var Fp=class extends Da{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,r)=>{let s=U.registeredVariables[n];if(this.accumulatedGrads[r]==null){let i=!1;this.accumulatedGrads[r]={originalName:`${n}/accumulator`,variable:Ve(()=>vp(s.shape,this.initialAccumulatorValue).variable(i))}}let a=Array.isArray(e)?e[r].tensor:e[n];if(a==null)return;let o=this.accumulatedGrads[r].variable;Ve(()=>{let i=Me(o,rs(a));o.assign(i);let l=Me(pe(Je(a,ra(Me(i,U.backend.epsilon()))),-this.learningRate),s);s.assign(l)})}),this.incrementIterations()}dispose(){this.accumulatedGrads!=null&&We(this.accumulatedGrads.map(e=>e.variable))}async getWeights(){return[await this.saveIterations()].concat(this.accumulatedGrads.map(e=>({name:e.originalName,tensor:e.variable})))}async setWeights(e){e=await this.extractIterations(e);let t=!1;this.accumulatedGrads=e.map(n=>({originalName:n.name,variable:n.tensor.variable(t)}))}getConfig(){return{learningRate:this.learningRate,initialAccumulatorValue:this.initialAccumulatorValue}}static fromConfig(e,t){return new e(t.learningRate,t.initialAccumulatorValue)}};Fp.className="Adagrad";$a(Fp);var Mp=class extends Da{constructor(e,t,n,r=null){super();this.learningRate=e,this.beta1=t,this.beta2=n,this.epsilon=r,this.accumulatedFirstMoment=[],this.accumulatedSecondMoment=[],Ve(()=>{this.accBeta1=ut(t).variable(),this.accBeta2=ut(n).variable()}),r==null&&(this.epsilon=U.backend.epsilon())}applyGradients(e){let t=Array.isArray(e)?e.map(n=>n.name):Object.keys(e);Ve(()=>{let n=Ue(1,this.accBeta1),r=Ue(1,this.accBeta2);t.forEach((s,a)=>{let o=U.registeredVariables[s],i=!1;this.accumulatedFirstMoment[a]==null&&(this.accumulatedFirstMoment[a]={originalName:`${s}/m`,variable:Ve(()=>Tr(o).variable(i))}),this.accumulatedSecondMoment[a]==null&&(this.accumulatedSecondMoment[a]={originalName:`${s}/v`,variable:Ve(()=>Tr(o).variable(i))});let l=Array.isArray(e)?e[a].tensor:e[s];if(l==null)return;let u=this.accumulatedFirstMoment[a].variable,c=this.accumulatedSecondMoment[a].variable,d=Me(pe(u,this.beta1),pe(l,1-this.beta1)),h=Me(pe(c,this.beta2),pe(rs(l),1-this.beta2)),p=Je(d,n),f=Je(h,r);u.assign(d),c.assign(h);let m=Me(pe(Je(p,Me(ra(f),this.epsilon)),-this.learningRate),o);o.assign(m)}),this.accBeta1.assign(pe(this.accBeta1,this.beta1)),this.accBeta2.assign(pe(this.accBeta2,this.beta2))}),this.incrementIterations()}dispose(){this.accBeta1.dispose(),this.accBeta2.dispose(),this.accumulatedFirstMoment!=null&&We(this.accumulatedFirstMoment.map(e=>e.variable)),this.accumulatedSecondMoment!=null&&We(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),Ve(()=>{this.accBeta1.assign(dc(this.beta1,this.iterations_+1)),this.accBeta2.assign(dc(this.beta2,this.iterations_+1))});let t=e.length/2,n=!1;this.accumulatedFirstMoment=e.slice(0,t).map(r=>({originalName:r.name,variable:r.tensor.variable(n)})),this.accumulatedSecondMoment=e.slice(t,t*2).map(r=>({originalName:r.name,variable:r.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)}};Mp.className="Adam";$a(Mp);var Op=class extends Da{constructor(e,t,n,r=null,s=0){super();this.learningRate=e,this.beta1=t,this.beta2=n,this.epsilon=r,this.decay=s,this.accumulatedFirstMoment=[],this.accumulatedWeightedInfNorm=[],Ve(()=>{this.iteration=ut(0).variable(),this.accBeta1=ut(t).variable()}),r==null&&(this.epsilon=U.backend.epsilon())}applyGradients(e){let t=Array.isArray(e)?e.map(n=>n.name):Object.keys(e);Ve(()=>{let n=Ue(1,this.accBeta1),r=Je(-this.learningRate,Me(pe(this.iteration,this.decay),1));t.forEach((s,a)=>{let o=U.registeredVariables[s],i=!1;this.accumulatedFirstMoment[a]==null&&(this.accumulatedFirstMoment[a]={originalName:`${s}/m`,variable:Tr(o).variable(i)}),this.accumulatedWeightedInfNorm[a]==null&&(this.accumulatedWeightedInfNorm[a]={originalName:`${s}/v`,variable:Tr(o).variable(i)});let l=Array.isArray(e)?e[a].tensor:e[s];if(l==null)return;let u=this.accumulatedFirstMoment[a].variable,c=this.accumulatedWeightedInfNorm[a].variable,d=Me(pe(u,this.beta1),pe(l,1-this.beta1)),h=pe(c,this.beta2),p=Sr(l),f=Jk(h,p);u.assign(d),c.assign(f);let m=Me(pe(Je(r,n),Je(d,Me(f,this.epsilon))),o);o.assign(m)}),this.iteration.assign(Me(this.iteration,1)),this.accBeta1.assign(pe(this.accBeta1,this.beta1))}),this.incrementIterations()}dispose(){this.accBeta1.dispose(),this.iteration.dispose(),this.accumulatedFirstMoment!=null&&We(this.accumulatedFirstMoment.map(e=>e.variable)),this.accumulatedWeightedInfNorm!=null&&We(this.accumulatedWeightedInfNorm.map(e=>e.variable))}async getWeights(){throw new Error("getWeights() is not implemented for Adamax yet.")}async setWeights(e){throw new Error("setWeights() is not implemented for Adamax yet.")}getConfig(){return{learningRate:this.learningRate,beta1:this.beta1,beta2:this.beta2,epsilon:this.epsilon,decay:this.decay}}static fromConfig(e,t){return new e(t.learningRate,t.beta1,t.beta2,t.epsilon,t.decay)}};Op.className="Adamax";$a(Op);var fc=class extends Da{constructor(e){super();this.learningRate=e,this.setLearningRate(e)}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=Array.isArray(e)?e[r].tensor:e[n];if(s==null)return;let a=U.registeredVariables[n];Ve(()=>{let o=Me(pe(this.c,s),a);a.assign(o)})}),this.incrementIterations()}setLearningRate(e){this.learningRate=e,this.c!=null&&this.c.dispose(),this.c=Nk(ut(-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)}};fc.className="SGD";$a(fc);var Pp=class extends fc{constructor(e,t,n=!1){super(e);this.learningRate=e,this.momentum=t,this.useNesterov=n,this.accumulations=[],this.m=ut(this.momentum)}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=U.registeredVariables[n];if(this.accumulations[r]==null){let i=!1;this.accumulations[r]={originalName:`${n}/momentum`,variable:Ve(()=>Tr(s).variable(i))}}let a=this.accumulations[r].variable,o=Array.isArray(e)?e[r].tensor:e[n];o!=null&&Ve(()=>{let i,l=Me(pe(this.m,a),o);this.useNesterov?i=Me(pe(this.c,Me(o,pe(l,this.m))),s):i=Me(pe(this.c,l),s),a.assign(l),s.assign(i)})}),this.incrementIterations()}dispose(){this.m.dispose(),this.accumulations!=null&&We(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)}};Pp.className="Momentum";$a(Pp);var zp=class extends Da{constructor(e,t=.9,n=0,r=null,s=!1){super();if(this.learningRate=e,this.decay=t,this.momentum=n,this.epsilon=r,this.accumulatedMeanSquares=[],this.accumulatedMoments=[],this.accumulatedMeanGrads=[],this.centered=s,r==null&&(this.epsilon=U.backend.epsilon()),e==null)throw new Error("learningRate for RMSPropOptimizer must be defined.")}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=U.registeredVariables[n],a=!1;this.accumulatedMeanSquares[r]==null&&(this.accumulatedMeanSquares[r]={originalName:`${n}/rms`,variable:Ve(()=>Tr(s).variable(a))}),this.accumulatedMoments[r]==null&&(this.accumulatedMoments[r]={originalName:`${n}/momentum`,variable:Ve(()=>Tr(s).variable(a))}),this.accumulatedMeanGrads[r]==null&&this.centered&&(this.accumulatedMeanGrads[r]={originalName:`${n}/mg`,variable:Ve(()=>Tr(s).variable(a))});let o=Array.isArray(e)?e[r].tensor:e[n];if(o==null)return;let i=this.accumulatedMeanSquares[r].variable,l=this.accumulatedMoments[r].variable;Ve(()=>{let u=Me(pe(i,this.decay),pe(rs(o),1-this.decay));if(this.centered){let c=this.accumulatedMeanGrads[r].variable,d=Me(pe(c,this.decay),pe(o,1-this.decay)),h=Je(pe(o,this.learningRate),ra(Ue(u,Me(rs(d),this.epsilon)))),p=Me(pe(l,this.momentum),h);i.assign(u),c.assign(d),l.assign(p);let f=Ue(s,p);s.assign(f)}else{let c=Me(pe(i,this.decay),pe(rs(o),1-this.decay)),d=Me(pe(l,this.momentum),Je(pe(o,this.learningRate),ra(Me(c,this.epsilon))));i.assign(c),l.assign(d);let h=Ue(s,d);s.assign(h)}})}),this.incrementIterations()}dispose(){this.accumulatedMeanSquares!=null&&We(this.accumulatedMeanSquares.map(e=>e.variable)),this.accumulatedMeanGrads!=null&&this.centered&&We(this.accumulatedMeanGrads.map(e=>e.variable)),this.accumulatedMoments!=null&&We(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(r=>({originalName:r.name,variable:r.tensor.variable(n)})),this.accumulatedMoments=e.slice(t,t*2).map(r=>({originalName:r.name,variable:r.tensor.variable(n)})),this.centered&&(this.accumulatedMeanGrads=e.slice(t*2,t*3).map(r=>({originalName:r.name,variable:r.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)}};zp.className="RMSProp";$a(zp);var To=class{static sgd(e){return new fc(e)}static momentum(e,t,n=!1){return new Pp(e,t,n)}static rmsprop(e,t=.9,n=0,r=null,s=!1){return new zp(e,t,n,r,s)}static adam(e=.001,t=.9,n=.999,r=null){return new Mp(e,t,n,r)}static adadelta(e=.001,t=.95,n=null){return new Dp(e,t,n)}static adamax(e=.002,t=.9,n=.999,r=null,s=0){return new Op(e,t,n,r,s)}static adagrad(e,t=.1){return new Fp(e,t)}},FV={sgd:To.sgd,momentum:To.momentum,adadelta:To.adadelta,adagrad:To.adagrad,rmsprop:To.rmsprop,adamax:To.adamax,adam:To.adam},MV=(()=>typeof requestAnimationFrame!="undefined"?requestAnimationFrame:typeof setImmediate!="undefined"?setImmediate:e=>e())();function OV(){return new Promise(e=>MV(()=>e()))}var v4={};_e(v4,{ERF_A1:()=>XV,ERF_A2:()=>ZV,ERF_A3:()=>YV,ERF_A4:()=>JV,ERF_A5:()=>QV,ERF_P:()=>KV,PARALLELIZE_THRESHOLD:()=>u1,SELU_SCALE:()=>qV,SELU_SCALEALPHA:()=>jV,applyActivation:()=>Rp,assertAndGetBroadcastShape:()=>kn,assertAxesAreInnerMostDims:()=>iz,assertParamsConsistent:()=>PV,assignToTypedArray:()=>iU,axesAreInnerMostDims:()=>e1,calculateShapes:()=>ck,checkEinsumDimSizes:()=>pU,combineLocations:()=>qk,complexWithEvenIndex:()=>sU,complexWithOddIndex:()=>aU,computeConv2DInfo:()=>ac,computeConv3DInfo:()=>$k,computeDefaultPad:()=>K2,computeDilation2DInfo:()=>YM,computeOptimalWindowSize:()=>LV,computeOutAndReduceShapes:()=>oz,computeOutShape:()=>zV,computePool2DInfo:()=>Ek,computePool3DInfo:()=>JM,convertConv2DDataFormat:()=>Rk,decodeEinsumEquation:()=>dU,eitherStridesOrDilationsAreOne:()=>ea,expandShapeToKeepDim:()=>uc,exponent:()=>uU,exponents:()=>lU,fromStringArrayToUint8:()=>wU,fromUint8ToStringArray:()=>vU,getAxesPermutation:()=>lz,getBroadcastDims:()=>lP,getComplexWithIndex:()=>oU,getEinsumComputePath:()=>fU,getEinsumPermutation:()=>hU,getFusedBiasGradient:()=>$p,getFusedDyActivation:()=>Ep,getImageCenter:()=>BV,getInnerMostAxes:()=>cz,getPermuted:()=>VV,getReductionAxes:()=>Ok,getReshaped:()=>WV,getReshapedPermuted:()=>UV,getSliceBeginCoords:()=>HV,getSliceSize:()=>GV,getUndoAxesPermutation:()=>uz,isIdentityPermutation:()=>mU,log:()=>tU,mergeRealAndImagArrays:()=>nU,prepareAndValidate:()=>lk,prepareSplitSize:()=>yU,segment_util:()=>I4,shouldFuse:()=>_p,slice_util:()=>B2,splitRealAndImagArrays:()=>rU,tupleValuesAreOne:()=>oc,upcastType:()=>up,validateInput:()=>L2,validateUpdateShape:()=>z2,warn:()=>eU});function PV(e,t){let n=e[0].length;e.forEach((s,a)=>{L(s.length===n,()=>`Error in concat${n}D: rank of tensors[${a}] must be the same as the rank of the rest (${n})`)}),L(t>=0&&t<n,()=>`Error in concat${n}D: axis must be between 0 and ${n-1}.`);let r=e[0];e.forEach((s,a)=>{for(let o=0;o<n;o++)L(o===t||s[o]===r[o],()=>`Error in concat${n}D: Shape of tensors[${a}] (${s}) does not match the shape of the rest (${r}) along the non-concatenated axis ${a}.`)})}function zV(e,t){let n=e[0].slice();for(let r=1;r<e.length;r++)n[t]+=e[r][t];return n}var u1=30;function LV(e){return e<=u1?e:ep(e,Math.floor(Math.sqrt(e)))}function BV(e,t,n){let r=n*(typeof e=="number"?e:e[0]),s=t*(typeof e=="number"?e:e[1]);return[r,s]}function WV(e,t,n,r=!0){let s=[];if(r)s=s.concat(t.slice(0)),s.push(e[0]/n),s=s.concat(e.slice(1));else{s=s.concat(e[0]);let a=t.length;for(let o=0;o<a;++o)s=s.concat([e[o+1]/t[o],t[o]]);s=s.concat(e.slice(a+1))}return s}function VV(e,t,n=!0){let r=[];if(n){r.push(t);for(let s=t+1;s<e;++s)s<=2*t?(r.push(s),r.push(s-(t+1))):r.push(s)}else{let s=[],a=[];for(let o=1;o<e;++o)o>=t*2+1||o%2==1?a.push(o):s.push(o);r.push(...s),r.push(0),r.push(...a)}return r}function UV(e,t,n,r=!0){let s=[];r?s.push(e[0]/n):s.push(e[0]*n);for(let a=1;a<e.length;++a)a<=t.length?r?s.push(t[a-1]*e[a]):s.push(e[a]/t[a-1]):s.push(e[a]);return s}function HV(e,t){let n=[0];for(let r=0;r<t;++r)n.push(e[r][0]);return n}function GV(e,t,n){let r=e.slice(0,1);for(let s=0;s<n;++s)r.push(e[s+1]-t[s][0]-t[s][1]);return r}var jV=1.7580993408473768,qV=1.0507009873554805,KV=.3275911,XV=.254829592,ZV=-.284496736,YV=1.421413741,JV=-1.453152027,QV=1.061405429;function eU(...e){ct().getBool("IS_TEST")||console.warn(...e)}function tU(...e){ct().getBool("IS_TEST")||console.log(...e)}function nU(e,t){if(e.length!==t.length)throw new Error(`Cannot merge real and imag arrays of different lengths. real:${e.length}, imag: ${t.length}.`);let n=new Float32Array(e.length*2);for(let r=0;r<n.length;r+=2)n[r]=e[r/2],n[r+1]=t[r/2];return n}function rU(e){let t=new Float32Array(e.length/2),n=new Float32Array(e.length/2);for(let r=0;r<e.length;r+=2)t[r/2]=e[r],n[r/2]=e[r+1];return{real:t,imag:n}}function sU(e){let t=Math.ceil(e.length/4),n=new Float32Array(t),r=new Float32Array(t);for(let s=0;s<e.length;s+=4)n[Math.floor(s/4)]=e[s],r[Math.floor(s/4)]=e[s+1];return{real:n,imag:r}}function aU(e){let t=Math.floor(e.length/4),n=new Float32Array(t),r=new Float32Array(t);for(let s=2;s<e.length;s+=4)n[Math.floor(s/4)]=e[s],r[Math.floor(s/4)]=e[s+1];return{real:n,imag:r}}function oU(e,t){let n=e[t*2],r=e[t*2+1];return{real:n,imag:r}}function iU(e,t,n,r){e[r*2]=t,e[r*2+1]=n}function lU(e,t){let n=new Float32Array(e/2),r=new Float32Array(e/2);for(let s=0;s<Math.ceil(e/2);s++){let a=(t?2:-2)*Math.PI*(s/e);n[s]=Math.cos(a),r[s]=Math.sin(a)}return{real:n,imag:r}}function uU(e,t,n){let r=(n?2:-2)*Math.PI*(e/t),s=Math.cos(r),a=Math.sin(r);return{real:s,imag:a}}var c1="->",cU=/->/g,w4=",",k4="...";function dU(e,t){e=e.replace(/\s/g,"");let n=(e.length-e.replace(cU,"").length)/c1.length;if(n<1)throw new Error("Equations without an arrow are not supported.");if(n>1)throw new Error(`Equation must contain exactly one arrow ("${c1}").`);let[r,s]=e.split(c1);L(r.indexOf(k4)===-1,()=>`The ellipsis notation ("${k4}") is not supported yet.`);let a=r.split(w4),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 h=0;h<s.length;++h){let p=s[h];if(!a.some(f=>f.indexOf(p)!==-1))throw new Error(`Output subscripts contain the label ${p} not present in the input subscripts.`);i.indexOf(p)===-1&&i.push(p)}for(let h=0;h<r.length;++h){let p=r[h];i.indexOf(p)===-1&&p!==w4&&i.push(p)}let l=new Array(a.length);for(let h=0;h<o;++h){if(new Set(a[h].split("")).size!==a[h].length)throw new Error(`Found duplicate axes in input component ${a[h]}. Support for duplicate axes in input is not implemented yet.`);l[h]=[];for(let p=0;p<a[h].length;++p)l[h].push(i.indexOf(a[h][p]))}let u=i.length,c=s.length,d=[];for(let h=c;h<u;++h)d.push(h);return{allDims:i,summedDims:d,idDims:l}}function hU(e,t){let n=new Array(e);n.fill(-1);for(let s=0;s<t.length;++s)n[t[s]]=s;let r=[];for(let s=0;s<e;++s)n[s]===-1&&r.push(s);return n=n.filter(s=>s!==-1),{permutationIndices:n,expandDims:r}}function pU(e,t,n){let r=new Array(e);for(let s=0;s<n.length;++s){let a=n[s].shape;for(let o=0;o<t[s].length;++o)r[t[s][o]]===void 0?r[t[s][o]]=a[o]:L(r[t[s][o]]===a[o],()=>`Expected dimension ${r[t[s][o]]} at axis ${o} of input shaped ${JSON.stringify(a)}, but got dimension ${a[o]}`)}}function fU(e,t){let n=e,r=[],s=0;e.length===0&&n.push(-1),s=e.length+1;for(let o=0;o<s;++o)r.push([]);let a=[];for(let o=0;o<n.length;++o){let i=n[o],l=gU(t,i);for(let u of l)a.indexOf(u)===-1&&(r[o].push(u),a.push(u))}return{path:n,steps:r}}function mU(e){return e.every((t,n)=>t===n)}function gU(e,t){let n=[];for(let r=0;r<e.length;++r)(e[r].length===0||e[r].indexOf(t)!==-1||t===-1)&&n.push(r);return n}function yU(e,t,n=0){let r=[];if(typeof t=="number")L(e.shape[n]%t==0,()=>"Number of splits must evenly divide the axis."),r=new Array(t).fill(e.shape[n]/t);else{let s=t.reduce((o,i)=>(i===-1&&(o+=1),o),0);L(s<=1,()=>"There should be only one negative value in split array.");let a=t.indexOf(-1);if(a!==-1){let o=t.reduce((i,l)=>l>0?i+l:i);t[a]=e.shape[n]-o}L(e.shape[n]===t.reduce((o,i)=>o+i),()=>"The sum of sizes must match the size of the axis dimension."),r=t}return r}var I4={};_e(I4,{collectGatherOpShapeInfo:()=>bU,computeOutShape:()=>xU,segOpComputeOptimalWindowSize:()=>AU});function AU(e,t){let n=!1,r;for(e<=u1?(r=e,n=!0):r=ep(e,Math.floor(Math.sqrt(e)));!n;)r>t||r===e?n=!0:r=ep(e,r+1);return r}function xU(e,t,n){let r=[],s=e.length;for(let a=0;a<s;a++)a!==t?r.push(e[a]):r.push(n);return r}function bU(e,t,n,r){let s=t.shape.length,a=e.shape.length;if(r!==0&&(r<-s||r>s))throw new Error(`Expect batchDims in the range of [-${s}, ${s}], but got ${r}`);if(r<0&&(r+=s),r>a)throw new Error(`batchDims (${r}) must be less than rank(x) (
${a}).`);if(n<r)throw new Error(`batchDims (${r}) must be less than or equal to axis (${n}).`);for(let d=0;d<r;++d)if(e.shape[d]!==t.shape[d])throw new Error(`x.shape[${d}]: ${e.shape[d]} should be equal to indices.shape[${d}]: ${t.shape[d]}.`);let o=e.shape[n],i=[],l=1,u=1,c=1;for(let d=0;d<r;++d)i.push(e.shape[d]),l*=e.shape[d];for(let d=r;d<n;d++)i.push(e.shape[d]),u*=e.shape[d];for(let d=r;d<s;d++)i.push(t.shape[d]);for(let d=n+1;d<a;d++)i.push(e.shape[d]),c*=e.shape[d];return{batchSize:l,sliceSize:c,outerSize:u,dimSize:o,outputShape:i}}function vU(e){try{return e.map(t=>op(t))}catch(t){throw new Error(`Failed to decode encoded string bytes into utf-8, error: ${t}`)}}function wU(e){return e.map(t=>Ju(t))}var S4={};_e(S4,{nonMaxSuppressionV3Impl:()=>g4,nonMaxSuppressionV4Impl:()=>y4,nonMaxSuppressionV5Impl:()=>A4,whereImpl:()=>u4});var kU=1e-7,IU=1e-4,d1=class{constructor(e,t){this.backend=e,this.dataMover=t,this.data=new WeakMap,this.dataIdsCount=0}get(e){return this.data.has(e)||this.dataMover.moveData(this.backend,e),this.data.get(e)}set(e,t){this.dataIdsCount++,this.data.set(e,t)}has(e){return this.data.has(e)}delete(e){return this.dataIdsCount--,this.data.delete(e)}numDataIds(){return this.dataIdsCount}},Lp=class{refCount(e){return Wr("refCount")}incRef(e){return Wr("incRef")}timerAvailable(){return!0}time(e){return Wr("time")}read(e){return Wr("read")}readSync(e){return Wr("readSync")}numDataIds(){return Wr("numDataIds")}disposeData(e,t){return Wr("disposeData")}write(e,t,n){return Wr("write")}move(e,t,n,r,s){return Wr("move")}memory(){return Wr("memory")}floatPrecision(){return Wr("floatPrecision")}epsilon(){return this.floatPrecision()===32?kU:IU}dispose(){return Wr("dispose")}};function Wr(e){throw new Error(`'${e}' not yet implemented or not found in the registry. This kernel may not be supported by the tfjs backend you have chosen`)}function T4(e){let t=e.length,n=0;for(;t>0;)n=Math.random()*t|0,t--,Bp(e,t,n)}function SU(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,r=0;for(;n>0;)r=Math.random()*n|0,n--,Bp(e,n,r),Bp(t,n,r)}function mc(e,t,n){return Math.max(e,Math.min(t,n))}function TU(e){return e%2==0?e:e+1}function Bp(e,t,n){let r=e[t];e[t]=e[n],e[n]=r}function NU(e){let t=0;for(let n=0;n<e.length;n++)t+=e[n];return t}function CU(e,t){let n=Math.random();return t*n+(1-n)*e}function EU(e,t){let n=0;for(let r=0;r<e.length;r++){let s=Number(e[r])-Number(t[r]);n+=s*s}return n}function z(e,t){if(!e)throw new Error(typeof t=="string"?t:t())}function ss(e,t,n=""){z(Fa(e,t),()=>n+` Shapes ${e} and ${t} must match`)}function Wp(e){z(e!=null,()=>"The input to the tensor constructor must be a non-null value.")}function gc(e,t=[],n=!1){if(t==null&&(t=[]),Array.isArray(e)||as(e)&&!n)for(let r=0;r<e.length;++r)gc(e[r],t,n);else t.push(e);return t}function sn(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 $U(e){return e.length===0}function Fa(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 fn(e){return e%1==0}function RU(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 _U(e){let t=Math.ceil(Math.sqrt(e));return[t,Math.ceil(e/t)]}function DU(e){let t=new Uint32Array(e);for(let n=0;n<e;++n)t[n]=n;return T4(t),t}function yc(e,t){return t<=e.length?e:e+" ".repeat(t-e.length)}function FU(e,t=r=>0,n){return new Promise((r,s)=>{let a=0,o=()=>{if(e()){r();return}a++;let i=t(a);if(n!=null&&a>=n){s();return}setTimeout(o,i)};o()})}function MU(e,t){let n=1,r=-1;for(let a=0;a<e.length;++a)if(e[a]>=0)n*=e[a];else if(e[a]===-1){if(r!==-1)throw Error(`Shapes can only have 1 implicit size. Found -1 at dim ${r} and dim ${a}`);r=a}else if(e[a]<0)throw Error(`Shapes can not be < 0. Found ${e[a]} at dim ${a}`);if(r===-1){if(t>0&&t!==n)throw Error(`Size(${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} / ${n}`);let s=e.slice();return s[r]=t/n,s}function Vr(e,t){let n=t.length;return e=e==null?t.map((r,s)=>s):[].concat(e),z(e.every(r=>r>=-n&&r<n),()=>`All values in axis param must be in range [-${n}, ${n}) but got axis ${e}`),z(e.every(r=>fn(r)),()=>`All values in axis param must be integers but got axis ${e}`),e.map(r=>r<0?n+r:r)}function N4(e,t){let n=[],r=[],s=t!=null&&Array.isArray(t)&&t.length===0,a=t==null||s?null:Vr(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]),r.push(i)),a[o]<=i&&o++}e[i]!==1&&(n.push(e[i]),r.push(i))}return{newShape:n,keptDims:r}}function OU(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 C4(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 E4(e,t){for(let n=0;n<e.length;n++){let r=e[n];if(isNaN(r)||!isFinite(r))throw Error(`A tensor of type ${t} being uploaded contains ${r}.`)}}function $4(e){return e==="bool"||e==="complex64"||e==="float32"||e==="int32"||e==="string"}function PU(e,t){return!(t==="complex64"||t==="float32"&&e!=="complex64"||t==="int32"&&e!=="float32"&&e!=="complex64"||t==="bool"&&e==="bool")}function as(e){return e instanceof Float32Array||e instanceof Int32Array||e instanceof Uint8Array}function h1(e){if(e==="float32"||e==="int32")return 4;if(e==="complex64")return 8;if(e==="bool")return 1;throw new Error(`Unknown dtype ${e}`)}function R4(e){if(e==null)return 0;let t=0;return e.forEach(n=>t+=n.length),t}function Vp(e){return typeof e=="string"||e instanceof String}function _4(e){return typeof e=="boolean"}function D4(e){return typeof e=="number"}function Up(e){return Array.isArray(e)?Up(e[0]):e instanceof Float32Array?"float32":e instanceof Int32Array||e instanceof Uint8Array?"int32":D4(e)?"float32":Vp(e)?"string":_4(e)?"bool":"float32"}function Hp(e){return!!(e&&e.constructor&&e.call&&e.apply)}function Gp(e,t){for(let n=t;n<e;++n)if(e%n==0)return n;return e}function qi(e){let t=e.length;if(t<2)return[];let n=new Array(t-1);n[t-2]=e[t-1];for(let r=t-3;r>=0;--r)n[r]=n[r+1]*e[r+1];return n}function F4(e,t,n,r=!1){let s=new Array;if(t.length===1){let a=t[0]*(r?2:1);for(let o=0;o<a;o++)s[o]=n[e+o]}else{let a=t[0],o=t.slice(1),i=o.reduce((l,u)=>l*u)*(r?2:1);for(let l=0;l<a;l++)s[l]=F4(e+l*i,o,n,r)}return s}function Ki(e,t,n=!1){if(e.length===0)return t[0];let r=e.reduce((s,a)=>s*a)*(n?2:1);if(r===0)return[];if(r!==t.length)throw new Error(`[${e}] does not match the input size ${t.length}${n?" for a complex tensor":""}.`);return F4(0,e,t,n)}function p1(e,t){let n=jp(e,t);for(let r=0;r<n.length;r++)n[r]=1;return n}function jp(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 zU(e,t){let n=e.reduce((r,s)=>r*s,1);if(t==null||t==="float32")return Ki(e,new Float32Array(n));if(t==="int32")return Ki(e,new Int32Array(n));if(t==="bool")return Ki(e,new Uint8Array(n));throw new Error(`Unknown data type ${t}`)}function f1(e){e.forEach(t=>{z(Number.isInteger(t)&&t>=0,()=>`Tensor must have a shape comprised of positive integers but got shape [${e}].`)})}function LU(e,t,n){if(t===0)return 0;if(t===1)return e[0];let r=e[e.length-1];for(let s=0;s<e.length-1;++s)r+=n[s]*e[s];return r}function BU(e,t,n){if(t===0)return[];if(t===1)return[e];let r=new Array(t);for(let s=0;s<r.length-1;++s)r[s]=Math.floor(e/n[s]),e-=r[s]*n[s];return r[r.length-1]=e,r}function m1(e){return e&&e.then&&typeof e.then=="function"}var M4="tfjsflags",WU=class{constructor(e){this.global=e,this.flags={},this.flagRegistry={},this.urlFlags={},this.getQueryParams=VU,this.populateURLFlags()}setPlatform(e,t){this.platform!=null&&console.warn(`Platform ${this.platformName} has already been set. Overwriting the platform with ${t}.`),this.platformName=e,this.platform=t}registerFlag(e,t,n){if(this.flagRegistry[e]={evaluationFn:t,setHook:n},this.urlFlags[e]!=null){let r=this.urlFlags[e];console.warn(`Setting feature override from URL ${e}: ${r}.`),this.set(e,r)}}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(m1(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);M4 in e&&e[M4].split(",").forEach(n=>{let[r,s]=n.split(":");this.urlFlags[r]=HU(r,s)})}};function VU(e){let t={};return e.replace(/[?&]([^=?&]+)(?:=([^&]*))?/g,(n,...r)=>(UU(t,r[0],r[1]),r.join("="))),t}function UU(e,t,n){e[decodeURIComponent(t)]=decodeURIComponent(n||"")}function HU(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 ${e}.`)}function re(){return O4}var O4=null;function GU(e){O4=e}var g1;function P4(){if(g1==null){let e;if(typeof window!="undefined")e=window;else if(typeof global!="undefined")e=global;else if(typeof process!="undefined")e=process;else if(typeof self!="undefined")e=self;else throw new Error("Could not find a global object");g1=e}return g1}function jU(){let e=P4();return e._tfGlobals==null&&(e._tfGlobals=new Map),e._tfGlobals}function y1(e,t){let n=jU();if(n.has(e))return n.get(e);{let r=t();return n.set(e,r),n.get(e)}}var Ac="Abs",xc="Acos",bc="Acosh",Ma="Add",Xi="AddN",vc="All",wc="Any",Zi="ArgMax",qp="ArgMin",kc="Asin",Ic="Asinh",Sc="Atan",Tc="Atanh",Nc="Atan2",Yi="AvgPool",A1="AvgPoolGrad",Kp="AvgPool3D",x1="AvgPool3DGrad",Ji="BatchMatMul",Cc="BatchToSpaceND",b1="Bincount",qU="BroadcastTo",Qi="Cast",No="Ceil",Co="ClipByValue",v1="Complex",Xp="ComplexAbs",Ec="Concat",el="Conv2D",w1="Conv2DBackpropFilter",tl="Conv2DBackpropInput",Zp="Conv3D",k1="Conv3DBackpropFilterV2",I1="Conv3DBackpropInputV2",nl="Cos",rl="Cosh",sl="Cumsum",$c="CropAndResize",S1="DenseBincount",Rc="DepthToSpace",al="DepthwiseConv2dNative",T1="DepthwiseConv2dNativeBackpropFilter",N1="DepthwiseConv2dNativeBackpropInput",C1="Diag",Yp="Dilation2D",E1="Dilation2DBackpropInput",$1="Dilation2DBackpropFilter",ol="RealDiv",R1="Einsum",_c="Elu",_1="EluGrad",Dc="Erf",il="Equal",Eo="Exp",Fc="ExpandDims",ll="Expm1",D1="FFT",Jp="Fill",Mc="FlipLeftRight",$o="Floor",ul="FloorDiv",cl="FusedBatchNorm",Oc="GatherV2",Pc="GatherNd",dl="Greater",Ro="GreaterEqual",hl="Identity",F1="IFFT",M1="Imag",zc="IsFinite",Lc="IsInf",Bc="IsNan",pl="LeakyRelu",fl="Less",ml="LessEqual",O1="LinSpace",_o="Log",Wc="Log1p",Vc="LogicalAnd",Qp="LogicalNot",ef="LogicalOr",KU="LogSoftmax",tf="LRN",P1="LRNGrad",gl="Max",Do="Maximum",yl="MaxPool",z1="MaxPoolGrad",nf="MaxPool3D",L1="MaxPool3DGrad",B1="MaxPoolWithArgmax",Al="Mean",xl="Min",Fo="Minimum",bl="MirrorPad",Uc="Mod",W1="Multinomial",Mo="Multiply",Hc="Neg",vl="NotEqual",Gc="NonMaxSuppressionV3",jc="NonMaxSuppressionV4",qc="NonMaxSuppressionV5",Kc="OnesLike",wl="OneHot",Xc="Pack",kl="PadV2",Il="Pow",Sl="Prelu",Zc="Prod",rf="Range",V1="Real",Yc="Reciprocal",Tl="Relu",Jc="Reshape",sf="ResizeNearestNeighbor",U1="ResizeNearestNeighborGrad",Nl="ResizeBilinear",H1="ResizeBilinearGrad",Cl="Relu6",El="Reverse",$l="Round",Oo="Rsqrt",Qc="ScatterNd",ed="Select",td="Selu",nd="Slice",Rl="Sin",rd="Sinh",sd="Sign",_l="Sigmoid",ad="Softplus",Dl="Sqrt",Fl="Sum",od="SpaceToBatchND",id="SplitV",Ml="Softmax",G1="SparseFillEmptyRows",j1="SparseReshape",q1="SparseSegmentMean",K1="SparseSegmentSum",X1="SparseToDense",Po="SquaredDifference",af="Square",ld="StridedSlice",Z1="StringNGrams",Y1="StringSplit",J1="StringToHashBucketFast",zo="Sub",Ol="Tan",Pl="Tanh",Lo="Tile",ud="TopK",cd="Transform",zl="Transpose",Q1="Unique",dd="Unpack",of="UnsortedSegmentSum",hd="ZerosLike",Bo="Step",ey="FromPixels",pd="RotateWithOffset",Ll="_FusedMatMul",Bl="FusedConv2D",Wl="FusedDepthwiseConv2D",lf=y1("kernelRegistry",()=>new Map),ty=y1("gradRegistry",()=>new Map);function ny(e,t){let n=B4(e,t);return lf.get(n)}function z4(e){return ty.get(e)}function L4(e){let t=lf.entries(),n=[];for(;;){let{done:r,value:s}=t.next();if(r)break;let[a,o]=s,[i]=a.split("_");i===e&&n.push(o)}return n}function ry(e){let{kernelName:t,backendName:n}=e,r=B4(t,n);lf.has(r)&&console.warn(`The kernel '${t}' for backend '${n}' is already registered`),lf.set(r,e)}function XU(e){let{kernelName:t}=e;ty.has(t)&&re().getBool("DEBUG")&&console.warn(`Overriding the gradient for '${t}'`),ty.set(t,e)}function B4(e,t){return`${t}_${e}`}var k={};_e(k,{arraysEqual:()=>Fa,assert:()=>z,assertNonNegativeIntegerDimensions:()=>f1,assertNonNull:()=>Wp,assertShapesMatch:()=>ss,bytesFromStringArray:()=>R4,bytesPerElement:()=>h1,checkConversionForErrors:()=>E4,clamp:()=>mc,computeStrides:()=>qi,createScalarValue:()=>tH,createShuffledIndices:()=>DU,decodeString:()=>hf,distSquared:()=>EU,encodeString:()=>df,fetch:()=>rH,fingerPrint64:()=>eH,flatten:()=>gc,getArrayFromDType:()=>C4,getTypedArrayFromDType:()=>OU,hasEncodingLoss:()=>PU,hexToLong:()=>fd,indexToLoc:()=>BU,inferDtype:()=>Up,inferFromImplicitShape:()=>MU,isBoolean:()=>_4,isFunction:()=>Hp,isInt:()=>fn,isNumber:()=>D4,isPromise:()=>m1,isScalarShape:()=>$U,isString:()=>Vp,isTypedArray:()=>as,isValidDtype:()=>$4,locToIndex:()=>LU,makeOnesTypedArray:()=>p1,makeZerosNestedTypedArray:()=>zU,makeZerosTypedArray:()=>jp,nearestDivisor:()=>Gp,nearestLargerEven:()=>TU,now:()=>md,parseAxisParam:()=>Vr,randUniform:()=>CU,repeatedTry:()=>FU,rightPad:()=>yc,shuffle:()=>T4,shuffleCombo:()=>SU,sizeFromShape:()=>sn,sizeToSquarishShape:()=>_U,squeezeShape:()=>N4,sum:()=>NU,swap:()=>Bp,tanh:()=>RU,toNestedArray:()=>Ki,toTypedArray:()=>cf});var W4=Xs(S3()),Wo=W4.default||W4;function fd(e){return Wo.fromString(e,!0,16)}var V4=fd("c3a5c85c97cb3127"),Vo=fd("b492b66fbe98f273"),zn=fd("9ae16a3b2f90404f");function sy(e){return e.xor(e.shru(47))}function U4(e,t,n){let r=e.slice(t,t+n);return Wo.fromBytes(Array.from(r),!0,!0)}function St(e,t){return U4(e,t,8)}function H4(e,t){return U4(e,t,4)}function mn(e,t){return t===0?e:e.shru(t).or(e.shl(64-t))}function Oa(e,t,n=fd("9ddfea08eb382d69")){let r=e.xor(t).mul(n);r=r.xor(r.shru(47));let s=t.xor(r).mul(n);return s=s.xor(s.shru(47)),s=s.mul(n),s}function ZU(e,t,n,r,s,a){s=s.add(e),a=mn(a.add(s).add(r),21);let o=s;return s=s.add(t),s=s.add(n),a=a.add(mn(s,44)),[s.add(r),a.add(o)]}function uf(e,t,n,r){return ZU(St(e,t),St(e,t+8),St(e,t+16),St(e,t+24),n,r)}function YU(e,t=e.length){if(t>=8){let n=zn.add(t*2),r=St(e,0).add(zn),s=St(e,t-8),a=mn(s,37).mul(n).add(r),o=mn(r,25).add(s).mul(n);return Oa(a,o,n)}if(t>=4){let n=zn.add(t*2),r=H4(e,0);return Oa(r.shl(3).add(t),H4(e,t-4),n)}if(t>0){let n=e[0],r=e[t>>1],s=e[t-1],a=n+(r<<8),o=t+(s<<2);return sy(zn.mul(a).xor(V4.mul(o))).mul(zn)}return zn}function JU(e,t=e.length){let n=zn.add(t*2),r=St(e,0).mul(Vo),s=St(e,8),a=St(e,t-8).mul(n),o=St(e,t-16).mul(zn);return Oa(mn(r.add(s),43).add(mn(a,30)).add(o),r.add(mn(s.add(zn),18)).add(a),n)}function QU(e,t=e.length){let n=zn.add(t*2),r=St(e,0).mul(zn),s=St(e,8),a=St(e,t-8).mul(n),o=St(e,t-16).mul(zn),i=mn(r.add(s),43).add(mn(a,30)).add(o),l=Oa(i,r.add(mn(s.add(zn),18)).add(a),n),u=St(e,16).mul(n),c=St(e,24),d=i.add(St(e,t-32)).mul(n),h=l.add(St(e,t-24)).mul(n);return Oa(mn(u.add(c),43).add(mn(d,30)).add(h),u.add(mn(c.add(r),18)).add(d),n)}function eH(e,t=e.length){let n=Wo.fromNumber(81,!0);if(t<=32)return t<=16?YU(e,t):JU(e,t);if(t<=64)return QU(e,t);let r=n,s=n.mul(Vo).add(113),a=sy(s.mul(zn).add(113)).mul(zn),o=[Wo.UZERO,Wo.UZERO],i=[Wo.UZERO,Wo.UZERO];r=r.mul(zn).add(St(e,0));let l=0,u=(t-1>>6)*64,c=u+(t-1&63)-63;do r=mn(r.add(s).add(o[0]).add(St(e,l+8)),37).mul(Vo),s=mn(s.add(o[1]).add(St(e,l+48)),42).mul(Vo),r=r.xor(i[1]),s=s.add(o[0]).add(St(e,l+40)),a=mn(a.add(i[0]),33).mul(Vo),o=uf(e,l,o[1].mul(Vo),r.add(i[0])),i=uf(e,l+32,a.add(i[1]),s.add(St(e,l+16))),[a,r]=[r,a],l+=64;while(l!==u);let d=Vo.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]),r=mn(r.add(s).add(o[0]).add(St(e,l+8)),37).mul(d),s=mn(s.add(o[1]).add(St(e,l+48)),42).mul(d),r=r.xor(i[1].mul(9)),s=s.add(o[0].mul(9).add(St(e,l+40))),a=mn(a.add(i[0]),33).mul(d),o=uf(e,l,o[1].mul(d),r.add(i[0])),i=uf(e,l+32,a.add(i[1]),s.add(St(e,l+16))),[a,r]=[r,a],Oa(Oa(o[0],i[0],d).add(sy(s).mul(V4)).add(a),Oa(o[1],i[1],d).add(r),d)}function tH(e,t){return t==="string"?df(e):cf([e],t)}function nH(e,t){return e instanceof Float32Array&&t==="float32"||e instanceof Int32Array&&t==="int32"||e instanceof Uint8Array&&t==="bool"}function cf(e,t){if(t==="string")throw new Error("Cannot convert a string[] to a TypedArray");if(Array.isArray(e)&&(e=gc(e)),re().getBool("DEBUG")&&E4(e,t),nH(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 r=0;r<n.length;++r)Math.round(e[r])!==0&&(n[r]=1);return n}else throw new Error(`Unknown data type ${t}`)}function md(){return re().platform.now()}function rH(e,t){return re().platform.fetch(e,t)}function df(e,t="utf-8"){return t=t||"utf-8",re().platform.encode(e,t)}function hf(e,t="utf-8"){return t=t||"utf-8",re().platform.decode(e,t)}var sH=class{constructor(e,t){this.backendTimer=e,this.logger=t,t==null&&(this.logger=new oH)}profileKernel(e,t,n){let r,s=()=>{r=n()},a,o=md();if(this.backendTimer.timerAvailable())a=this.backendTimer.time(s);else{s();for(let l of r)l.dataSync();a=Promise.resolve({kernelMs:md()-o})}if(re().getBool("CHECK_COMPUTATION_FOR_ERRORS"))for(let l=0;l<r.length;l++){let u=r[l];u.data().then(c=>{aH(c,u.dtype,e)})}return{kernelName:e,outputs:r,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:r,inputs:s,extraInfo:a}=e;n.forEach(o=>{Promise.all([o.data(),r,a]).then(i=>{this.logger.logKernelProfile(t,o,i[0],i[1],s,i[2])})})}};function aH(e,t,n){if(t!=="float32")return!1;for(let r=0;r<e.length;r++){let s=e[r];if(isNaN(s)||!isFinite(s))return console.warn(`Found ${s} in the result of '${n}'`),!0}return!1}var oH=class{logKernelProfile(e,t,n,r,s,a){let o=typeof r=="number"?yc(`${r}ms`,9):r.error,i=yc(e,25),l=t.rank,u=t.size,c=yc(t.shape.toString(),14),d="";for(let h in s){let p=s[h];if(p!=null){let f=p.shape||t.shape,m=f.length;d+=`${h}: ${m}D ${m>0?f:""} `}}console.log(`%c${i} %c${o} %c${l}D ${c} %c${u} %c${d} %c${a}`,"font-weight:bold","color:red","color:blue","color: orange","color: green","color: steelblue")}};function iH(e,t,n){let r={},s={};for(let l=0;l<t.length;l++)r[t[l].id]=!0;for(let l=0;l<e.length;l++){let u=e[l],c=u.inputs;for(let d in c){let h=c[d],p=!1;for(let f=0;f<t.length;f++)if(r[h.id]){u.outputs.forEach(m=>r[m.id]=!0),p=!0,s[u.id]=!0;break}if(p)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 d=0;d<u.outputs.length;d++)if(a[u.outputs[d].id]){for(let h in c)a[c[h].id]=!0,o[u.id]=!0;break}}let i=[];for(let l=0;l<e.length;l++){let u=e[l];if(s[u.id]&&o[u.id]){let c={};for(let h in u.inputs){let p=u.inputs[h];r[p.id]&&(c[h]=p)}let d=Object.assign({},u);d.inputs=c,d.outputs=u.outputs,i.push(d)}}return i}function lH(e,t,n,r){for(let s=t.length-1;s>=0;s--){let a=t[s],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(!Fa(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 d=e[c.id];e[c.id]=r(d,u),d.dispose()}}}}var G4=20,gd=3,ay=7;function uH(e,t,n,r){let s=qi(t),a=cH(e,t,n,s),o=t.length,i=pf(e,t,n,s,a),l=["Tensor"];return r&&(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 cH(e,t,n,r){let s=sn(t),a=r[r.length-1],o=new Array(a).fill(0),i=t.length,l=n==="complex64"?Ad(e):e;if(i>1)for(let u=0;u<s/a;u++){let c=u*a;for(let d=0;d<a;d++)o[d]=Math.max(o[d],yd(l[c+d],0,n).length)}return o}function yd(e,t,n){let r;return Array.isArray(e)?r=`${parseFloat(e[0].toFixed(ay))} + ${parseFloat(e[1].toFixed(ay))}j`:Vp(e)?r=`'${e}'`:n==="bool"?r=j4(e):r=parseFloat(e.toFixed(ay)).toString(),yc(r,t)}function j4(e){return e===0?"false":"true"}function pf(e,t,n,r,s,a=!0){let o=n==="complex64"?2:1,i=t[0],l=t.length;if(l===0){if(n==="complex64"){let m=Ad(e);return[yd(m[0],0,n)]}return n==="bool"?[j4(e[0])]:[e[0].toString()]}if(l===1){if(i>G4){let g=gd*o,y=Array.from(e.slice(0,g)),A=Array.from(e.slice((i-gd)*o,i*o));return n==="complex64"&&(y=Ad(y),A=Ad(A)),["["+y.map((x,b)=>yd(x,s[b],n)).join(", ")+", ..., "+A.map((x,b)=>yd(x,s[i-gd+b],n)).join(", ")+"]"]}let m=n==="complex64"?Ad(e):Array.from(e);return["["+m.map((g,y)=>yd(g,s[y],n)).join(", ")+"]"]}let u=t.slice(1),c=r.slice(1),d=r[0]*o,h=[];if(i>G4){for(let m=0;m<gd;m++){let g=m*d,y=g+d;h.push(...pf(e.slice(g,y),u,n,c,s,!1))}h.push("...");for(let m=i-gd;m<i;m++){let g=m*d,y=g+d;h.push(...pf(e.slice(g,y),u,n,c,s,m===i-1))}}else for(let m=0;m<i;m++){let g=m*d,y=g+d;h.push(...pf(e.slice(g,y),u,n,c,s,m===i-1))}let p=l===2?",":"";h[0]="["+h[0]+p;for(let m=1;m<h.length-1;m++)h[m]=" "+h[m]+p;let f=`,
`;for(let m=2;m<l;m++)f+=`
`;return h[h.length-1]=" "+h[h.length-1]+"]"+(a?"":f),h}function Ad(e){let t=[];for(let n=0;n<e.length;n+=2)t.push([e[n],e[n+1]]);return t}var Jt=class{constructor(e,t,n){if(this.dtype=t,this.shape=e.slice(),this.size=sn(e),n!=null){let r=n.length;z(r===this.size,()=>`Length of values '${r}' does not match the size inferred by the shape '${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||C4(t,this.size),this.strides=qi(e)}set(e,...t){t.length===0&&(t=[0]),z(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 r of e){if(r<0||r>=this.shape[t]){let s=`Requested out of range element at ${e}. Buffer shape=${this.shape}`;throw new Error(s)}t++}let n=e[e.length-1];for(let r=0;r<e.length-1;++r)n+=this.strides[r]*e[r];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 Ts().makeTensor(this.values,this.shape,this.dtype)}},Ts=null,Vl=null,dH=null;function hH(e){Ts=e}function pH(e){Vl=e}function fH(e){dH=e}var Ot=class{constructor(e,t,n,r){this.kept=!1,this.isDisposedInternal=!1,this.shape=e.slice(),this.dtype=t||"float32",this.size=sn(e),this.strides=qi(e),this.dataId=n,this.id=r,this.rankType=this.rank<5?this.rank.toString():"higher"}get rank(){return this.shape.length}async buffer(){let e=await this.data();return Vl.buffer(this.shape,this.dtype,e)}bufferSync(){return Vl.buffer(this.shape,this.dtype,this.dataSync())}async array(){let e=await this.data();return Ki(this.shape,e,this.dtype==="complex64")}arraySync(){return Ki(this.shape,this.dataSync(),this.dtype==="complex64")}async data(){this.throwIfDisposed();let e=Ts().read(this.dataId);if(this.dtype==="string"){let t=await e;try{return t.map(n=>hf(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}dataSync(){this.throwIfDisposed();let e=Ts().readSync(this.dataId);if(this.dtype==="string")try{return e.map(t=>hf(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 Ts().read(this.dataId);return this.dtype==="string"?e:new Uint8Array(e.buffer)}dispose(){this.isDisposed||(Ts().disposeTensor(this),this.isDisposedInternal=!0)}get isDisposed(){return this.isDisposedInternal}throwIfDisposed(){if(this.isDisposed)throw new Error("Tensor is disposed.")}print(e=!1){return Vl.print(this,e)}clone(){return this.throwIfDisposed(),Vl.clone(this)}toString(e=!1){let t=this.dataSync();return uH(t,this.shape,this.dtype,e)}cast(e){return this.throwIfDisposed(),Vl.cast(this,e)}variable(e=!0,t,n){return this.throwIfDisposed(),Ts().makeVariable(this,e,t,n)}};Object.defineProperty(Ot,Symbol.hasInstance,{value:e=>!!e&&e.data!=null&&e.dataSync!=null&&e.throwIfDisposed!=null});function mH(){return y1("Tensor",()=>Ot)}mH();var ff=class extends Ot{constructor(e,t,n,r){super(e.shape,e.dtype,e.dataId,r);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(!Fa(e.shape,this.shape))throw new Error(`shape of the new value (${e.shape}) and previous value (${this.shape}) must match`);Ts().disposeTensor(this),this.dataId=e.dataId,Ts().incRef(this,null)}dispose(){Ts().disposeVariable(this),this.isDisposedInternal=!0}};Object.defineProperty(ff,Symbol.hasInstance,{value:e=>e instanceof Ot&&e.assign!=null&&e.assign instanceof Function});var Ns={};_e(Ns,{assertTypesMatch:()=>K4,getTensorsInContainer:()=>dy,isTensorInList:()=>yH,makeTypesMatch:()=>Vt});var q4;(function(e){e.R0="R0",e.R1="R1",e.R2="R2",e.R3="R3",e.R4="R4",e.R5="R5",e.R6="R6"})(q4||(q4={}));var oy;(function(e){e.float32="float32",e.int32="int32",e.bool="int32",e.complex64="complex64"})(oy||(oy={}));var iy;(function(e){e.float32="float32",e.int32="int32",e.bool="bool",e.complex64="complex64"})(iy||(iy={}));var ly;(function(e){e.float32="float32",e.int32="float32",e.bool="float32",e.complex64="complex64"})(ly||(ly={}));var uy;(function(e){e.float32="complex64",e.int32="complex64",e.bool="complex64",e.complex64="complex64"})(uy||(uy={}));var gH={float32:ly,int32:oy,bool:iy,complex64:uy};function Ur(e,t){if(e==="string"||t==="string"){if(e==="string"&&t==="string")return"string";throw new Error(`Can not upcast ${e} with ${t}`)}return gH[e][t]}function cy(e){return Ur(e,"int32")}function Vt(e,t){if(e.dtype===t.dtype)return[e,t];let n=Ur(e.dtype,t.dtype);return[e.cast(n),t.cast(n)]}function K4(e,t){z(e.dtype===t.dtype,()=>`The dtypes of the first(${e.dtype}) and second(${t.dtype}) input must match`)}function yH(e,t){return t.some(n=>n.id===e.id)}function dy(e){let t=[],n=new Set;return X4(e,t,n),t}function X4(e,t,n){if(e==null)return;if(e instanceof Ot){t.push(e);return}if(!AH(e))return;let r=e;for(let s in r){let a=r[s];n.has(a)||(n.add(a),X4(a,t,n))}}function AH(e){return Array.isArray(e)||typeof e=="object"}function hy(e){return e.kernelName!=null}var Z4=class{constructor(){this.registeredVariables={},this.nextTapeNodeId=0,this.numBytes=0,this.numTensors=0,this.numStringTensors=0,this.numDataBuffers=0,this.gradientDepth=0,this.kernelDepth=0,this.scopeStack=[],this.numDataMovesStack=[],this.nextScopeId=0,this.tensorInfo=new WeakMap,this.profiling=!1,this.activeProfile={newBytes:0,newTensors:0,peakBytes:0,kernels:[],result:null,get kernelNames(){return Array.from(new Set(this.kernels.map(e=>e.name)))}}}dispose(){for(let e in this.registeredVariables)this.registeredVariables[e].dispose()}},xd=class{constructor(e){this.ENV=e,this.registry={},this.registryFactory={},this.pendingBackendInitId=0,this.state=new Z4}async ready(){if(this.pendingBackendInit!=null)return this.pendingBackendInit.then(()=>{});if(this.backendInstance!=null)return;let e=this.getSortedBackends();for(let t=0;t<e.length;t++){let n=e[t];if(await this.initializeBackend(n).success){await this.setBackend(n);return}}throw new Error("Could not initialize any backends, all backend initializations failed.")}get backend(){if(this.pendingBackendInit!=null)throw new Error(`Backend '${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 '${e}' has not yet been initialized. Make sure to await tf.ready() or await tf.setBackend() before calling other methods`);this.setBackend(e)}return this.backendInstance}backendNames(){return Object.keys(this.registryFactory)}findBackend(e){if(!(e in this.registry))if(e in this.registryFactory){let{asyncInit:t}=this.initializeBackend(e);if(t)return null}else return null;return this.registry[e]}findBackendFactory(e){return e in this.registryFactory?this.registryFactory[e].factory:null}registerBackend(e,t,n=1){return e in this.registryFactory?(console.warn(`${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 sH(this.backendInstance),!0}setupRegisteredKernels(){L4(this.backendName).forEach(t=>{t.setupFunc!=null&&t.setupFunc(this.backendInstance)})}disposeRegisteredKernels(e){L4(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 Lp)&&typeof n.then=="function"){let r=++this.pendingBackendInitId,s=n.then(a=>r<this.pendingBackendInitId?!1:(this.registry[e]=a,this.pendingBackendInit=null,!0)).catch(a=>(r<this.pendingBackendInitId||(this.pendingBackendInit=null,console.warn(`Initialization of backend ${e} failed`),console.warn(a.stack||a.message)),!1));return this.pendingBackendInit=s,{success:s,asyncInit:!0}}else return this.registry[e]=n,{success:!0,asyncInit:!1}}catch(n){return console.warn(`Initialization of backend ${e} failed`),console.warn(n.stack||n.message),{success:!1,asyncInit:!1}}}removeBackend(e){if(!(e in this.registryFactory))throw new Error(`${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:r,asyncInit:s}=this.initializeBackend(n);if(s||r)return{name:n,asyncInit:s}}throw new Error("Could not initialize any backends, all backend initializations failed.")}moveData(e,t){let n=this.state.tensorInfo.get(t),r=n.backend,s=this.readSync(t),a=r.refCount(t);r.disposeData(t,!0),n.backend=e,e.move(t,s,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 r;return this.scopedRun(()=>this.startScope(n),()=>this.endScope(r),()=>(r=t(),r instanceof Promise&&console.error("Cannot return a Promise inside of tidy."),r))}scopedRun(e,t,n){e();try{let r=n();return t(),r}catch(r){throw t(),r}}nextTensorId(){return xd.nextTensorId++}nextVariableId(){return xd.nextVariableId++}clone(e){let t=G.runKernel(hl,{x:e}),n={x:e},r=a=>({x:()=>{let o="float32",i={x:a},l={dtype:o};return G.runKernel(Qi,i,l)}}),s=[];return this.addTapeNode(this.state.activeScope.name,n,[t],r,s,{}),t}runKernel(e,t,n){if(this.backendName==null&&this.backend,!(ny(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 r=this.backend.numDataIds(),s=0;n.forEach(i=>{s+=i.dtype==="complex64"?3:1});let a=this.state.numDataMovesStack[this.state.numDataMovesStack.length-1],o=r-t-s-a;if(o>0)throw new Error(`Backend '${this.backendName}' has an internal memory leak (${o} data ids) after running '${e}'`)}runKernelFunc(e){let t,n=[],r=this.isTapeOn(),s=this.state.numBytes,a=this.state.numTensors;this.shouldCheckForMemLeaks()&&this.state.numDataMovesStack.push(0);let o;this.backendName==null&&this.backend;let i,l=hy(e)?e.kernelName:this.state.activeScope!=null?this.state.activeScope.name:"";if(hy(e)){let{kernelName:p,inputs:f,attrs:m}=e;this.backendName==null&&this.backend;let g=ny(p,this.backendName);z(g!=null,()=>`Cannot find registered kernel '${p}' for backend '${this.backendName}'`),o=()=>{let y=this.backend.numDataIds();i=g.kernelFunc({inputs:f,attrs:m,backend:this.backend});let A=Array.isArray(i)?i:[i];this.shouldCheckForMemLeaks()&&this.checkKernelForMemLeak(p,y,A);let x=A.map(b=>{if(b.rank!=null)return b;let{dataId:v,shape:I,dtype:w}=b;return this.makeTensorFromDataId(v,I,w)});if(r){let b=this.getTensorsForGradient(p,f,x);n=this.saveTensorsForBackwardMode(b)}return x}}else{let{forwardFunc:p}=e,f=m=>{!r||(n=m.map(g=>this.keep(this.clone(g))))};o=()=>{let m=this.backend.numDataIds();i=this.tidy(()=>p(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,d=hy(e)?null:e.backwardsFunc,h;return this.scopedRun(()=>this.state.kernelDepth++,()=>this.state.kernelDepth--,()=>{!this.ENV.getBool("DEBUG")&&!this.state.profiling?t=o():(h=this.profiler.profileKernel(l,u,()=>o()),this.ENV.getBool("DEBUG")&&this.profiler.logKernelProfile(h),t=h.outputs)}),r&&this.addTapeNode(l,u,t,d,n,c),this.state.profiling&&this.state.activeProfile.kernels.push({name:l,bytesAdded:this.state.numBytes-s,totalBytesSnapshot:this.state.numBytes,tensorsAdded:this.state.numTensors-a,totalTensorsSnapshot:this.state.numTensors,inputShapes:Object.keys(u).map(p=>u[p]!=null?u[p].shape:null),outputShapes:t.map(p=>p.shape),kernelTimeMs:h.timeMs,extraInfo:h.extraInfo}),Array.isArray(i)?t:t[0]}saveTensorsForBackwardMode(e){return e.map(n=>this.keep(this.clone(n)))}getTensorsForGradient(e,t,n){let r=z4(e);if(r!=null){let s=r.inputsToSave||[],a=r.outputsToSave||[],o;r.saveAllInputs?(z(Array.isArray(t),()=>"saveAllInputs is true, expected inputs to be an array."),o=Object.keys(t).map(l=>t[l])):o=s.map(l=>t[l]);let i=n.filter((l,u)=>a[u]);return o.concat(i)}return[]}makeTensor(e,t,n,r){if(e==null)throw new Error("Values passed to engine.makeTensor() are null");n=n||"float32",r=r||this.backend;let s=e;n==="string"&&Vp(e[0])&&(s=e.map(i=>df(i)));let a=r.write(s,t,n),o=new Ot(t,n,a,this.nextTensorId());if(this.trackTensor(o,r),n==="string"){let i=this.state.tensorInfo.get(a),l=R4(s);this.state.numBytes+=l-i.bytes,i.bytes=l}return o}makeTensorFromDataId(e,t,n,r){n=n||"float32";let s=new Ot(t,n,e,this.nextTensorId());return this.trackTensor(s,r),s}makeVariable(e,t=!0,n,r){n=n||this.nextVariableId().toString(),r!=null&&r!==e.dtype&&(e=e.cast(r));let s=new ff(e,t,n,this.nextTensorId());if(this.state.registeredVariables[s.name]!=null)throw new Error(`Variable with name ${s.name} was already registered`);return this.state.registeredVariables[s.name]=s,this.incRef(s,this.backend),s}trackTensor(e,t){this.state.numTensors++,e.dtype==="string"&&this.state.numStringTensors++;let n=0;e.dtype!=="complex64"&&e.dtype!=="string"&&(n=e.size*h1(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 ff||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*h1(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(r=>r.totalBytesSnapshot)),this.state.activeProfile.newBytes=this.state.numBytes-t,this.state.activeProfile.newTensors=this.state.numTensors-n;for(let r of this.state.activeProfile.kernels)r.kernelTimeMs=await r.kernelTimeMs,r.extraInfo=await r.extraInfo;return this.state.activeProfile}isTapeOn(){return this.state.gradientDepth>0&&this.state.kernelDepth===0}addTapeNode(e,t,n,r,s,a){let o={id:this.state.nextTapeNodeId++,kernelName:e,inputs:t,outputs:n,saved:s},i=z4(e);i!=null&&(r=i.gradFunc),r!=null&&(o.gradient=l=>(l=l.map((u,c)=>{if(u==null){let d=n[c],h=jp(d.size,d.dtype);return this.makeTensor(h,d.shape,d.dtype)}return u}),r(l.length>1?l:l[0],s,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=dy(e),n=new Set(t.map(s=>s.id));for(let s=0;s<this.state.activeScope.track.length;s++){let a=this.state.activeScope.track[s];!a.kept&&!n.has(a.id)&&a.dispose()}let r=this.state.scopeStack.pop();this.state.activeScope=this.state.scopeStack.length===0?null:this.state.scopeStack[this.state.scopeStack.length-1],t.forEach(s=>{!s.kept&&s.scopeId===r.id&&this.track(s)})}gradients(e,t,n,r=!1){if(z(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 s=this.scopedRun(()=>this.startTape(),()=>this.endTape(),()=>this.tidy("forward",e));z(s instanceof Ot,()=>"The result y returned by f() must be a tensor.");let a=iH(this.state.activeTape,t,s);if(!r&&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[s.id]=n==null?xH(s.shape):n,lH(o,a,l=>this.tidy(l),bH);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:s,grads:i}})}customGrad(e){return z(Hp(e),()=>"The f passed in customGrad(f) must be a function."),(...t)=>{z(t.every(o=>o instanceof Ot),()=>"The args passed in customGrad(f)(x1, x2,...) must all be tensors");let n,r={};t.forEach((o,i)=>{r[i]=o});let s=(o,i)=>(n=e(...t,i),z(n.value instanceof Ot,()=>"The function f passed in customGrad(f) must return an object where `obj.value` is a tensor"),z(Hp(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];z(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(...)."),z(u.every(d=>d instanceof Ot),()=>"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((d,h)=>{c[h]=()=>d}),c};return this.runKernelFunc({forwardFunc:s,backwardsFunc:a,inputs:r})}}readSync(e){return this.state.tensorInfo.get(e).backend.readSync(e)}read(e){return this.state.tensorInfo.get(e).backend.read(e)}async time(e){let t=md(),n=await this.backend.time(e);return n.wallMs=md()-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 Z4;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}};xd.nextTensorId=0;xd.nextVariableId=0;function xH(e){let t=p1(sn(e),"float32");return G.makeTensor(t,e,"float32")}function Y4(){let e=P4();if(e._tfengine==null){let t=new WU(e);e._tfengine=new xd(t)}return GU(e._tfengine.ENV),hH(()=>e._tfengine),e._tfengine}var G=Y4();function bH(e,t){let n={a:e,b:t};return G.runKernel(Ma,n)}var mf={};_e(mf,{isBrowser:()=>J4,isMobile:()=>wH});function vH(){return typeof navigator!="undefined"&&navigator!=null}function wH(e){if(e||vH()){if(e||(e=navigator),e.product==="ReactNative")return!0;let t=e.userAgent||e.vendor||window.opera;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 J4(){return typeof window!="undefined"&&window.document!=null||typeof WorkerGlobalScope!="undefined"}var os=re();os.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.")});os.registerFlag("IS_BROWSER",()=>J4());os.registerFlag("IS_NODE",()=>typeof process!="undefined"&&typeof process.versions!="undefined"&&typeof process.versions.node!="undefined");os.registerFlag("IS_CHROME",()=>typeof navigator!="undefined"&&navigator!=null&&navigator.userAgent!=null&&/Chrome/.test(navigator.userAgent)&&/Google Inc/.test(navigator.vendor));os.registerFlag("PROD",()=>!1);os.registerFlag("TENSORLIKE_CHECK_SHAPE_CONSISTENCY",()=>os.getBool("DEBUG"));os.registerFlag("DEPRECATION_WARNINGS_ENABLED",()=>!0);os.registerFlag("IS_TEST",()=>!1);os.registerFlag("CHECK_COMPUTATION_FOR_ERRORS",()=>!0);os.registerFlag("WRAP_TO_IMAGEBITMAP",()=>!1);function bd(e,t){let n=e;if(as(e))return t==="string"?[]:[e.length];if(!Array.isArray(e))return[];let r=[];for(;Array.isArray(n)||as(n)&&t!=="string";)r.push(n.length),n=n[0];return Array.isArray(e)&&re().getBool("TENSORLIKE_CHECK_SHAPE_CONSISTENCY")&&Q4(e,r,[]),r}function Q4(e,t,n){if(n=n||[],!Array.isArray(e)&&!as(e)){z(t.length===0,()=>`Element arr[${n.join("][")}] is a primitive, but should be an array/TypedArray of ${t[0]} elements`);return}z(t.length>0,()=>`Element arr[${n.join("][")}] should be a primitive, but is an array of ${e.length} elements`),z(e.length===t[0],()=>`Element arr[${n.join("][")}] should have ${t[0]} elements, but has ${e.length} elements`);let r=t.slice(1);for(let s=0;s<e.length;++s)Q4(e[s],r,n.concat(s))}function e6(e,t,n,r){if(e!=="string_or_numeric"){if(e==null)throw new Error("Expected dtype cannot be null.");if(e!=="numeric"&&e!==t||e==="numeric"&&t==="string")throw new Error(`Argument '${n}' passed to '${r}' must be ${e} tensor, but got ${t} tensor`)}}function O(e,t,n,r="numeric"){if(e instanceof Ot)return e6(r,e.dtype,t,n),e;let s=Up(e);if(s!=="string"&&["bool","int32","float32"].indexOf(r)>=0&&(s=r),e6(r,s,t,n),e==null||!as(e)&&!Array.isArray(e)&&typeof e!="number"&&typeof e!="boolean"&&typeof e!="string"){let l=e==null?"null":e.constructor.name;throw new Error(`Argument '${t}' passed to '${n}' must be a Tensor or TensorLike, but got '${l}'`)}let a=bd(e,s);!as(e)&&!Array.isArray(e)&&(e=[e]);let i=s!=="string"?cf(e,s):gc(e,[],!0);return G.makeTensor(i,a,s)}function gf(e,t,n,r="numeric"){if(!Array.isArray(e))throw new Error(`Argument ${t} passed to ${n} must be a \`Tensor[]\` or \`TensorLike[]\``);return e.map((a,o)=>O(a,`${t}[${o}]`,n,r))}var kH="__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],r=e[n];n.endsWith("_")&&(n=n.substring(0,n.length-1)),n=n+kH;let s=(...a)=>{G.startScope(n);try{let o=r(...a);return m1(o)&&console.error("Cannot return a Promise inside of tidy."),G.endScope(o),o}catch(o){throw G.endScope(null),o}};return Object.defineProperty(s,"name",{value:n,configurable:!0}),s}function IH(e,t){let n=O(e,"real","complex"),r=O(t,"imag","complex");ss(n.shape,r.shape,`real and imag shapes, ${n.shape} and ${r.shape}, must match in call to tf.complex().`);let s={real:n,imag:r};return G.runKernel(v1,s)}var Uo=V({complex_:IH});function vd(e,t,n,r){if(r==null&&(r=Up(e)),r==="complex64")throw new Error("Cannot construct a complex64 tensor directly. Please use tf.complex(real, imag).");if(!as(e)&&!Array.isArray(e)&&typeof e!="number"&&typeof e!="boolean"&&typeof e!="string")throw new Error("values passed to tensor(values) must be a number/boolean/string or an array of numbers/booleans/strings, or a TypedArray");if(t!=null){f1(t);let s=sn(t),a=sn(n);z(s===a,()=>`Based on the provided shape, [${t}], the tensor should have ${s} values but has ${a}`);for(let o=0;o<n.length;++o){let i=n[o],l=o===n.length-1?i!==sn(t.slice(o)):!0;z(n[o]===t[o]||!l,()=>`Error creating a new Tensor. Inferred shape (${n}) does not match the provided shape (${t}). `)}}return!as(e)&&!Array.isArray(e)&&(e=[e]),t=t||n,e=r!=="string"?cf(e,r):gc(e,[],!0),G.makeTensor(e,t,r)}function Cs(e,t,n){let r=bd(e,n);return vd(e,t,r,n)}var py={float32:4,float16:2,int32:4,uint16:2,uint8:1,bool:1,complex64:8},yf=4;async function SH(e,t){let n=[],r=[],s=Array.isArray(e)?e.map(o=>o.name):Object.keys(e);for(let o=0;o<s.length;++o){let i=s[o],l=Array.isArray(e)?e[o].tensor:e[i];if(l.dtype!=="float32"&&l.dtype!=="int32"&&l.dtype!=="bool"&&l.dtype!=="string"&&l.dtype!=="complex64")throw new Error(`Unsupported dtype in weight '${i}': ${l.dtype}`);let u={name:i,shape:l.shape,dtype:l.dtype};if(l.dtype==="string"){let c=new Promise(async d=>{let h=await l.bytes(),p=h.reduce((g,y)=>g+y.length,0)+yf*h.length,f=new Uint8Array(p),m=0;for(let g=0;g<h.length;g++){let y=h[g],A=new Uint8Array(new Uint32Array([y.length]).buffer);f.set(A,m),m+=yf,f.set(y,m),m+=y.length}d(f)});r.push(c)}else r.push(l.data());t!=null&&(u.group=t),n.push(u)}let a=await Promise.all(r);return{data:TH(a),specs:n}}function t6(e,t){let n={},r,s=0;for(let a of t){let o=a.name,i=a.dtype,l=a.shape,u=sn(l),c;if("quantization"in a){let d=a.quantization;if(d.dtype==="uint8"||d.dtype==="uint16"){if(!("min"in d&&"scale"in d))throw new Error(`Weight ${a.name} with quantization ${d.dtype} doesn't have corresponding metadata min and scale.`)}else if(d.dtype==="float16"){if(i!=="float32")throw new Error(`Weight ${a.name} is quantized with ${d.dtype} which only supports weights of type float32 not ${i}.`)}else throw new Error(`Weight ${a.name} has unknown quantization dtype ${d.dtype}. Supported quantization dtypes are: 'uint8', 'uint16', and 'float16'.`);let h=py[d.dtype],p=e.slice(s,s+u*h),f=d.dtype==="uint8"?new Uint8Array(p):new Uint16Array(p);if(i==="float32")if(d.dtype==="uint8"||d.dtype==="uint16"){c=new Float32Array(f.length);for(let m=0;m<f.length;m++){let g=f[m];c[m]=g*d.scale+d.min}}else if(d.dtype==="float16")r===void 0&&(r=_H()),c=r(f);else throw new Error(`Unsupported quantization type ${d.dtype} for weight type float32.`);else if(i==="int32"){if(d.dtype!=="uint8"&&d.dtype!=="uint16")throw new Error(`Unsupported quantization type ${d.dtype} for weight type int32.`);c=new Int32Array(f.length);for(let m=0;m<f.length;m++){let g=f[m];c[m]=Math.round(g*d.scale+d.min)}}else throw new Error(`Unsupported dtype in weight '${o}': ${i}`);s+=u*h}else if(i==="string"){let d=sn(a.shape);c=[];for(let h=0;h<d;h++){let p=new Uint32Array(e.slice(s,s+yf))[0];s+=yf;let f=new Uint8Array(e.slice(s,s+p));c.push(f),s+=p}}else{let d=py[i],h=e.slice(s,s+u*d);if(i==="float32")c=new Float32Array(h);else if(i==="int32")c=new Int32Array(h);else if(i==="bool")c=new Uint8Array(h);else if(i==="complex64"){c=new Float32Array(h);let p=new Float32Array(c.length/2),f=new Float32Array(c.length/2);for(let y=0;y<p.length;y++)p[y]=c[y*2],f[y]=c[y*2+1];let m=Cs(p,l,"float32"),g=Cs(f,l,"float32");n[o]=Uo(m,g),m.dispose(),g.dispose()}else throw new Error(`Unsupported dtype in weight '${o}': ${i}`);s+=u*d}i!=="complex64"&&(n[o]=Cs(c,l,i))}return n}function TH(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 r=new Uint8Array(t),s=0;return n.forEach(a=>{r.set(new Uint8Array(a.buffer),s),s+=a.byteLength}),r.buffer}var fy=typeof Buffer!="undefined"&&(typeof Blob=="undefined"||typeof atob=="undefined"||typeof btoa=="undefined");function n6(e){return fy?Buffer.byteLength(e):new Blob([e]).size}function NH(e){if(fy)return Buffer.from(e).toString("base64");let t=new Uint8Array(e),n="";for(let r=0,s=t.length;r<s;r++)n+=String.fromCharCode(t[r]);return btoa(n)}function CH(e){if(fy){let r=Buffer.from(e,"base64");return r.buffer.slice(r.byteOffset,r.byteOffset+r.byteLength)}let t=atob(e),n=new Uint8Array(t.length);for(let r=0;r<t.length;++r)n.set([t.charCodeAt(r)],r);return n.buffer}function my(e){if(e.length===1)return e[0];let t=0;e.forEach(s=>{t+=s.byteLength});let n=new Uint8Array(t),r=0;return e.forEach(s=>{n.set(new Uint8Array(s),r),r+=s.byteLength}),n.buffer}function r6(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 s6(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 gy(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[r,s]=await t(e.weightsManifest);n.weightSpecs=r,n.weightData=s}return e.signature!=null&&(n.signature=e.signature),e.userDefinedMetadata!=null&&(n.userDefinedMetadata=e.userDefinedMetadata),e.modelInitializer!=null&&(n.modelInitializer=e.modelInitializer),n}function wd(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:n6(JSON.stringify(e.modelTopology)),weightSpecsBytes:e.weightSpecs==null?0:n6(JSON.stringify(e.weightSpecs)),weightDataBytes:e.weightData==null?0:e.weightData.byteLength}}function EH(){let e=n=>{let r=n<<13,s=0;for(;(r&8388608)==0;)s-=8388608,r<<=1;return r&=~8388608,s+=947912704,r|s},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 $H(){let e=new Uint32Array(64);e[0]=0,e[31]=1199570944,e[32]=2147483648,e[63]=3347054592;for(let t=1;t<31;t++)e[t]=t<<23;for(let t=33;t<63;t++)e[t]=2147483648+(t-32<<23);return e}function RH(){let e=new Uint32Array(64);for(let t=0;t<64;t++)e[t]=1024;return e[0]=e[32]=0,e}function _H(){let e=EH(),t=$H(),n=RH();return r=>{let s=new ArrayBuffer(4*r.length),a=new Uint32Array(s);for(let o=0;o<r.length;o++){let i=r[o],l=e[n[i>>10]+(i&1023)]+t[i>>10];a[o]=l}return new Float32Array(s)}}var jt=class{constructor(){this.saveRouters=[],this.loadRouters=[]}static getInstance(){return jt.instance==null&&(jt.instance=new jt),jt.instance}static registerSaveRouter(e){jt.getInstance().saveRouters.push(e)}static registerLoadRouter(e){jt.getInstance().loadRouters.push(e)}static getSaveHandlers(e){return jt.getHandlers(e,"save")}static getLoadHandlers(e,t){return jt.getHandlers(e,"load",t)}static getHandlers(e,t,n){let r=[];return(t==="load"?jt.getInstance().loadRouters:jt.getInstance().saveRouters).forEach(a=>{let o=a(e,n);o!==null&&r.push(o)}),r}},DH=e=>jt.registerSaveRouter(e),FH=e=>jt.registerLoadRouter(e),MH=e=>jt.getSaveHandlers(e),OH=(e,t)=>jt.getLoadHandlers(e,t),yy="tensorflowjs",Ay=1,Ho="models_store",Pa="model_info_store";function a6(){if(!re().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 xy(e){let t=e.result;t.createObjectStore(Ho,{keyPath:"modelPath"}),t.createObjectStore(Pa,{keyPath:"modelPath"})}var Go=class{constructor(e){if(this.indexedDB=a6(),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,r)=>{let s=this.indexedDB.open(yy,Ay);s.onupgradeneeded=()=>xy(s),s.onsuccess=()=>{let a=s.result;if(t==null){let o=a.transaction(Ho,"readonly"),l=o.objectStore(Ho).get(this.modelPath);l.onsuccess=()=>{if(l.result==null)return a.close(),r(new Error(`Cannot find model with path '${this.modelPath}' in IndexedDB.`));n(l.result.modelArtifacts)},l.onerror=u=>(a.close(),r(l.error)),o.oncomplete=()=>a.close()}else{let o=wd(t),i=a.transaction(Pa,"readwrite"),l=i.objectStore(Pa),u=l.put({modelPath:this.modelPath,modelArtifactsInfo:o}),c;u.onsuccess=()=>{c=a.transaction(Ho,"readwrite");let h=c.objectStore(Ho).put({modelPath:this.modelPath,modelArtifacts:t,modelArtifactsInfo:o});h.onsuccess=()=>n({modelArtifactsInfo:o}),h.onerror=p=>{l=i.objectStore(Pa);let f=l.delete(this.modelPath);f.onsuccess=()=>(a.close(),r(h.error)),f.onerror=m=>(a.close(),r(h.error))}},u.onerror=d=>(a.close(),r(u.error)),i.oncomplete=()=>{c==null?a.close():c.oncomplete=()=>a.close()}}},s.onerror=a=>r(s.error)})}};Go.URL_SCHEME="indexeddb://";var o6=e=>re().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(Go.URL_SCHEME)?PH(e.slice(Go.URL_SCHEME.length)):null;jt.registerSaveRouter(o6);jt.registerLoadRouter(o6);function PH(e){return new Go(e)}function zH(e){return e.startsWith(Go.URL_SCHEME)?e.slice(Go.URL_SCHEME.length):e}var LH=class{constructor(){this.indexedDB=a6()}async listModels(){return new Promise((e,t)=>{let n=this.indexedDB.open(yy,Ay);n.onupgradeneeded=()=>xy(n),n.onsuccess=()=>{let r=n.result,s=r.transaction(Pa,"readonly"),o=s.objectStore(Pa).getAll();o.onsuccess=()=>{let i={};for(let l of o.result)i[l.modelPath]=l.modelArtifactsInfo;e(i)},o.onerror=i=>(r.close(),t(o.error)),s.oncomplete=()=>r.close()},n.onerror=r=>t(n.error)})}async removeModel(e){return e=zH(e),new Promise((t,n)=>{let r=this.indexedDB.open(yy,Ay);r.onupgradeneeded=()=>xy(r),r.onsuccess=()=>{let s=r.result,a=s.transaction(Pa,"readwrite"),o=a.objectStore(Pa),i=o.get(e),l;i.onsuccess=()=>{if(i.result==null)return s.close(),n(new Error(`Cannot find model with path '${e}' in IndexedDB.`));{let u=o.delete(e),c=()=>{l=s.transaction(Ho,"readwrite");let h=l.objectStore(Ho).delete(e);h.onsuccess=()=>t(i.result.modelArtifactsInfo),h.onerror=p=>n(i.error)};u.onsuccess=c,u.onerror=d=>(c(),s.close(),n(i.error))}},i.onerror=u=>(s.close(),n(i.error)),a.oncomplete=()=>{l==null?s.close():l.oncomplete=()=>s.close()}},r.onerror=s=>n(r.error)})}},oa="/",Ul="tensorflowjs_models",i6="info",BH="model_topology",WH="weight_specs",VH="weight_data",UH="model_metadata";function l6(e){return{info:[Ul,e,i6].join(oa),topology:[Ul,e,BH].join(oa),weightSpecs:[Ul,e,WH].join(oa),weightData:[Ul,e,VH].join(oa),modelMetadata:[Ul,e,UH].join(oa)}}function u6(e){for(let t of Object.values(e))window.localStorage.removeItem(t)}function HH(e){let t=e.split(oa);if(t.length<3)throw new Error(`Invalid key format: ${e}`);return t.slice(1,t.length-1).join(oa)}function GH(e){return e.startsWith(jo.URL_SCHEME)?e.slice(jo.URL_SCHEME.length):e}var jo=class{constructor(e){if(!re().getBool("IS_BROWSER")||typeof window=="undefined"||typeof window.localStorage=="undefined")throw new Error("The current environment does not support local storage.");if(this.LS=window.localStorage,e==null||!e)throw new Error("For local storage, modelPath must not be null, undefined or empty.");this.modelPath=e,this.keys=l6(this.modelPath)}async save(e){if(e.modelTopology instanceof ArrayBuffer)throw new Error("BrowserLocalStorage.save() does not support saving model topology in binary formats yet.");{let t=JSON.stringify(e.modelTopology),n=JSON.stringify(e.weightSpecs),r=wd(e);try{this.LS.setItem(this.keys.info,JSON.stringify(r)),this.LS.setItem(this.keys.topology,t),this.LS.setItem(this.keys.weightSpecs,n),this.LS.setItem(this.keys.weightData,NH(e.weightData));let s={format:e.format,generatedBy:e.generatedBy,convertedBy:e.convertedBy,signature:e.signature!=null?e.signature:void 0,userDefinedMetadata:e.userDefinedMetadata!=null?e.userDefinedMetadata:void 0,modelInitializer:e.modelInitializer!=null?e.modelInitializer:void 0,trainingConfig:e.trainingConfig!=null?e.trainingConfig:void 0};return this.LS.setItem(this.keys.modelMetadata,JSON.stringify(s)),{modelArtifactsInfo:r}}catch(s){throw u6(this.keys),new Error(`Failed to save model '${this.modelPath}' to local storage: size quota being exceeded is a possible cause of this failure: modelTopologyBytes=${r.modelTopologyBytes}, weightSpecsBytes=${r.weightSpecsBytes}, weightDataBytes=${r.weightDataBytes}.`)}}}async load(){let e=JSON.parse(this.LS.getItem(this.keys.info));if(e==null)throw new Error(`In local storage, there is no model with name '${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 '${this.modelPath}' is missing.`);t.modelTopology=n;let r=JSON.parse(this.LS.getItem(this.keys.weightSpecs));if(r==null)throw new Error(`In local storage, the weight specs of model '${this.modelPath}' are missing.`);t.weightSpecs=r;let s=this.LS.getItem(this.keys.modelMetadata);if(s!=null){let o=JSON.parse(s);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 '${this.modelPath}' are missing.`);return t.weightData=CH(a),t}};jo.URL_SCHEME="localstorage://";var c6=e=>re().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(jo.URL_SCHEME)?jH(e.slice(jo.URL_SCHEME.length)):null;jt.registerSaveRouter(c6);jt.registerLoadRouter(c6);function jH(e){return new jo(e)}var qH=class{constructor(){z(re().getBool("IS_BROWSER"),()=>"Current environment is not a web browser"),z(typeof window=="undefined"||typeof window.localStorage!="undefined",()=>"Current browser does not appear to support localStorage"),this.LS=window.localStorage}async listModels(){let e={},t=Ul+oa,n=oa+i6;for(let r=0;r<this.LS.length;++r){let s=this.LS.key(r);if(s.startsWith(t)&&s.endsWith(n)){let a=HH(s);e[a]=JSON.parse(this.LS.getItem(s))}}return e}async removeModel(e){e=GH(e);let t=l6(e);if(this.LS.getItem(t.info)==null)throw new Error(`Cannot find model at path '${e}'`);let n=JSON.parse(this.LS.getItem(t.info));return u6(t),n}},Hl="://",Nr=class{constructor(){this.managers={}}static getInstance(){return Nr.instance==null&&(Nr.instance=new Nr),Nr.instance}static registerManager(e,t){z(e!=null,()=>"scheme must not be undefined or null."),e.endsWith(Hl)&&(e=e.slice(0,e.indexOf(Hl))),z(e.length>0,()=>"scheme must not be an empty string.");let n=Nr.getInstance();z(n.managers[e]==null,()=>`A model store manager is already registered for scheme '${e}'.`),n.managers[e]=t}static getManager(e){let t=this.getInstance().managers[e];if(t==null)throw new Error(`Cannot find model manager for scheme '${e}'`);return t}static getSchemes(){return Object.keys(this.getInstance().managers)}};function Af(e){if(e.indexOf(Hl)===-1)throw new Error(`The url string provided does not contain a scheme. Supported schemes are: ${Nr.getSchemes().join(",")}`);return{scheme:e.split(Hl)[0],path:e.split(Hl)[1]}}async function d6(e,t,n=!1){z(e!==t,()=>`Old path and new path are the same: '${e}'`);let r=jt.getLoadHandlers(e);z(r.length>0,()=>`Copying failed because no load handler is found for source URL ${e}.`),z(r.length<2,()=>`Copying failed because more than one (${r.length}) load handlers for source URL ${e}.`);let s=r[0],a=jt.getSaveHandlers(t);z(a.length>0,()=>`Copying failed because no save handler is found for destination URL ${t}.`),z(a.length<2,()=>`Copying failed because more than one (${r.length}) save handlers for destination URL ${t}.`);let o=a[0],i=Af(e).scheme,l=Af(e).path,u=i===Af(e).scheme,c=await s.load();n&&u&&await Nr.getManager(i).removeModel(l);let d=await o.save(c);return n&&!u&&await Nr.getManager(i).removeModel(l),d.modelArtifactsInfo}async function KH(){let e=Nr.getSchemes(),t={};for(let n of e){let r=await Nr.getManager(n).listModels();for(let s in r){let a=n+Hl+s;t[a]=r[s]}}return t}async function XH(e){let t=Af(e);return Nr.getManager(t.scheme).removeModel(t.path)}async function ZH(e,t){return d6(e,t,!1)}async function YH(e,t){return d6(e,t,!0)}var JH=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(re().get("IS_BROWSER")){re().setPlatform("browser",new JH);try{Nr.registerManager(jo.URL_SCHEME,new qH)}catch(e){}try{Nr.registerManager(Go.URL_SCHEME,new LH)}catch(e){}}var QH={importFetch:()=>T3()},by,eG=class{constructor(){this.util=co("util"),this.textEncoder=new this.util.TextEncoder}fetch(e,t){return re().global.fetch!=null?re().global.fetch(e,t):(by==null&&(by=QH.importFetch()),by(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)}};re().get("IS_NODE")&&re().setPlatform("node",new eG);function ze(e,t="float32",n){return t=t||"float32",f1(e),new Jt(e,t,n)}function tG(e,t){let n=O(e,"x","cast");if(!$4(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 r={x:n},s={dtype:t};return G.runKernel(Qi,r,s)}var xe=V({cast_:tG});function nG(e){let n={x:O(e,"x","clone","string_or_numeric")};return G.runKernel(hl,n)}var za=V({clone_:nG});function rG(e,t=!1){console.log(e.toString(t))}Y4();var sG={buffer:ze,cast:xe,clone:za,print:rG};pH(sG);var cr={};_e(cr,{browserFiles:()=>dG,browserHTTPRequest:()=>gG,concatenateArrayBuffers:()=>my,copyModel:()=>ZH,decodeWeights:()=>t6,encodeWeights:()=>SH,fromMemory:()=>AG,getLoadHandlers:()=>OH,getModelArtifactsForJSON:()=>gy,getModelArtifactsInfoForJSON:()=>wd,getSaveHandlers:()=>MH,http:()=>ky,isHTTPScheme:()=>wy,listModels:()=>KH,loadWeights:()=>hG,moveModel:()=>YH,registerLoadRouter:()=>FH,registerSaveRouter:()=>DH,removeModel:()=>XH,weightsLoaderFactory:()=>m6,withSaveHandler:()=>xG});var aG="model",oG=".json",iG=".weights.bin";function h6(e){return new Promise(t=>setTimeout(t)).then(e)}var Gl=class{constructor(e){if(!re().getBool("IS_BROWSER"))throw new Error("browserDownloads() cannot proceed because the current environment is not a browser.");e.startsWith(Gl.URL_SCHEME)&&(e=e.slice(Gl.URL_SCHEME.length)),(e==null||e.length===0)&&(e=aG),this.modelJsonFileName=e+oG,this.weightDataFileName=e+iG}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}],r=s6(e,n),s=window.URL.createObjectURL(new Blob([JSON.stringify(r)],{type:"application/json"})),a=this.modelJsonAnchor==null?document.createElement("a"):this.modelJsonAnchor;if(a.download=this.modelJsonFileName,a.href=s,await h6(()=>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 h6(()=>o.dispatchEvent(new MouseEvent("click")))}return{modelArtifactsInfo:wd(e)}}}};Gl.URL_SCHEME="downloads://";var lG=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=r=>{let s=JSON.parse(r.target.result),a=s.modelTopology;if(a==null){t(new Error(`modelTopology field is missing from file ${this.jsonFile.name}`));return}if(s.weightsManifest==null){t(new Error(`weightManifest field is missing from file ${this.jsonFile.name}`));return}if(this.weightsFiles.length===0){e({modelTopology:a});return}let i=gy(s,l=>this.loadWeights(l));e(i)},n.onerror=r=>t(`Failed to read model topology and weights manifest JSON from file '${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 r=this.checkManifestAndWeightFiles(e),s=n.map(a=>this.loadWeightsFile(a,r[a]));return Promise.all(s).then(a=>[t,my(a)])}loadWeightsFile(e,t){return new Promise((n,r)=>{let s=new FileReader;s.onload=a=>{let o=a.target.result;n(o)},s.onerror=a=>r(`Failed to weights data from file of path '${e}'.`),s.readAsArrayBuffer(t)})}checkManifestAndWeightFiles(e){let t=[],n=this.weightsFiles.map(s=>r6(s.name)),r={};for(let s of e)s.paths.forEach(a=>{let o=r6(a);if(t.indexOf(o)!==-1)throw new Error(`Duplicate file basename found in weights manifest: '${o}'`);if(t.push(o),n.indexOf(o)===-1)throw new Error(`Weight file with basename '${o}' is not provided.`);r[a]=this.weightsFiles[n.indexOf(o)]});if(t.length!==this.weightsFiles.length)throw new Error(`Mismatch in the number of files in weights manifest (${t.length}) and the number of weight files provided (${this.weightsFiles.length}).`);return r}},uG=e=>re().getBool("IS_BROWSER")&&!Array.isArray(e)&&e.startsWith(Gl.URL_SCHEME)?cG(e.slice(Gl.URL_SCHEME.length)):null;jt.registerSaveRouter(uG);function cG(e="model"){return new Gl(e)}function dG(e){return new lG(e)}function p6(e,t,n,r){o(e),n=n==null?0:n,r=r==null?1:r,i(n,r);let s=0,a=l=>(l.then(u=>{let c=n+ ++s/e.length*(r-n);return t(c),u}),l);function o(l){z(l!=null&&Array.isArray(l)&&l.length>0,()=>"promises must be a none empty array")}function i(l,u){z(l>=0&&l<=1,()=>`Progress fraction must be in range [0, 1], but got startFraction ${l}`),z(u>=0&&u<=1,()=>`Progress fraction must be in range [0, 1], but got endFraction ${u}`),z(u>=l,()=>`startFraction must be no more than endFraction, but got startFraction ${l} and endFraction ${u}`)}return Promise.all(e.map(a))}async function f6(e,t){t==null&&(t={});let n=t.fetchFunc==null?re().platform.fetch:t.fetchFunc,r=e.map(d=>n(d,t.requestInit,{isBinary:!0})),s=0,a=.5,i=(t.onProgress==null?await Promise.all(r):await p6(r,t.onProgress,s,a)).map(d=>d.arrayBuffer()),l=.5,u=1;return t.onProgress==null?await Promise.all(i):await p6(i,t.onProgress,l,u)}async function hG(e,t="",n,r){return m6(o=>f6(o,{requestInit:r}))(e,t,n)}function m6(e){return async(t,n="",r)=>{let s=t.map(()=>!1),a={},o=r!=null?r.map(()=>!1):[],i=[];if(t.forEach((p,f)=>{let m=0;p.weights.forEach(g=>{let y="quantization"in g?g.quantization.dtype:g.dtype,A=py[y]*sn(g.shape),x=()=>{s[f]=!0,a[f]==null&&(a[f]=[]),a[f].push({manifestEntry:g,groupOffset:m,sizeBytes:A})};r!=null?r.forEach((b,v)=>{b===g.name&&(x(),o[v]=!0)}):x(),i.push(g.name),m+=A})}),!o.every(p=>p)){let p=r.filter((f,m)=>!o[m]);throw new Error(`Could not find weights in manifest with names: ${p.join(", ")}.
Manifest JSON has weights with names: ${i.join(", ")}.`)}let l=s.reduce((p,f,m)=>(f&&p.push(m),p),[]),u=[];l.forEach(p=>{t[p].paths.forEach(f=>{let m=n+(n.endsWith("/")?"":"/")+f;u.push(m)})});let c=await e(u),d={},h=0;return l.forEach(p=>{let f=t[p].paths.length,m=0;for(let b=0;b<f;b++)m+=c[h+b].byteLength;let g=new ArrayBuffer(m),y=new Uint8Array(g),A=0;for(let b=0;b<f;b++){let v=new Uint8Array(c[h+b]);y.set(v,A),A+=v.byteLength}a[p].forEach(b=>{let v=g.slice(b.groupOffset,b.groupOffset+b.sizeBytes),I=t6(v,[b.manifestEntry]);for(let w in I)d[w]=I[w]}),h+=f}),d}}var pG="application/octet-stream",fG="application/json",vy=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?(z(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=re().platform.fetch,z(e!=null&&e.length>0,()=>"URL path for http must not be null, undefined or empty."),Array.isArray(e)&&z(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}],r=s6(e,n);t.body.append("model.json",new Blob([JSON.stringify(r)],{type:fG}),"model.json"),e.weightData!=null&&t.body.append("model.weights.bin",new Blob([e.weightData],{type:pG}),"model.weights.bin");let s=await this.fetch(this.path,t);if(s.ok)return{modelArtifactsInfo:wd(e),responses:[s]};throw new Error(`BrowserHTTPRequest.save() failed due to HTTP response status ${s.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(s){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,r=t.weightsManifest;if(n==null&&r==null)throw new Error(`The JSON from HTTP path ${this.path} contains neither model topology or manifest for weights.`);return gy(t,s=>this.loadWeights(s))}async loadWeights(e){let t=Array.isArray(this.path)?this.path[1]:this.path,[n,r]=mG(t),s=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(s+c+r);this.weightUrlConverter&&o.push(...await Promise.all(i));let l=await f6(o,{requestInit:this.requestInit,fetchFunc:this.fetch,onProgress:this.onProgress});return[a,my(l)]}};vy.URL_SCHEME_REGEX=/^https?:\/\//;function mG(e){let t=e.lastIndexOf("/"),n=e.lastIndexOf("?"),r=e.substring(0,t),s=n>t?e.substring(n):"";return[r+"/",s]}function wy(e){return e.match(vy.URL_SCHEME_REGEX)!=null}var g6=(e,t)=>{if(typeof fetch=="undefined"&&(t==null||t.fetchFunc==null))return null;{let n=!0;if(Array.isArray(e)?n=e.every(r=>wy(r)):n=wy(e),n)return ky(e,t)}return null};jt.registerSaveRouter(g6);jt.registerLoadRouter(g6);function ky(e,t){return new vy(e,t)}function gG(e,t){return ky(e,t)}var Iy=class{constructor(e){this.modelArtifacts=e}async load(){return this.modelArtifacts}},yG=class{constructor(e){this.saveHandler=e}async save(e){return this.saveHandler(e)}};function AG(e,t,n,r){return arguments.length===1?e.modelTopology!=null||e.weightSpecs!=null?new Iy(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 Iy({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 Iy({modelTopology:e,weightSpecs:t,weightData:n,trainingConfig:r}))}function xG(e){return new yG(e)}function bG(e,t,n=!1,r=!1){let s=O(e,"a","matMul"),a=O(t,"b","matMul");[s,a]=Vt(s,a);let o={a:s,b:a},i={transposeA:n,transposeB:r};return G.runKernel(Ji,o,i)}var rt=V({matMul_:bG});function vG(e,t,n=1,r=0){if(t<2)throw new Error(`Error in oneHot: depth must be >=2, but it is ${t}`);let a={indices:O(e,"indices","oneHot","int32")},o={depth:t,onValue:n,offValue:r};return G.runKernel(wl,a,o)}var xf=V({oneHot_:vG});function wG(e,t){let n=O(e,"x","transpose");if(t==null&&(t=n.shape.map((a,o)=>o).reverse()),z(n.rank===t.length,()=>`Error in transpose: rank of input ${n.rank} must match length of perm ${t}.`),t.forEach(a=>{z(a>=0&&a<n.rank,()=>`All entries in 'perm' must be between 0 and ${n.rank-1} but got ${t}`)}),n.rank<=1)return n.clone();let r={x:n},s={perm:t};return G.runKernel(zl,r,s)}var st=V({transpose_:wG});function kG(e,t,n){let r=O(e,"labels","confusionMatrix"),s=O(t,"predictions","confusionMatrix");z(n==null||n>0&&Number.isInteger(n),()=>`If provided, numClasses must be a positive integer, but got ${n}`),z(r.rank===1,()=>`Expected the rank of labels to be 1, but got ${r.rank}`),z(s.rank===1,()=>`Expected the rank of predictions to be 1, but got ${s.rank}`),z(r.shape[0]===s.shape[0],()=>`Mismatch in the number of examples: ${r.shape[0]} vs. ${s.shape[0]}. Labels and predictions should have the same number of elements.`),z(n>0&&Number.isInteger(n),()=>`numClasses is required to be a positive integer, but got ${n}`);let a=xf(xe(r,"int32"),n),o=xf(xe(s,"int32"),n),i=st(a),l=rt(i,o);return xe(l,"int32")}var zwe=V({confusionMatrix_:kG}),y6={};_e(y6,{fromPixels:()=>RG,fromPixelsAsync:()=>EG,toPixels:()=>$G});function IG(e,t,n){if(Wp(e),t!=null&&t.length!==3)throw new Error("tensor3d() requires shape to have three numbers");let r=bd(e,n);if(r.length!==3&&r.length!==1)throw new Error("tensor3d() requires values to be number[][][] or flat/TypedArray");if(r.length===1&&t==null)throw new Error("tensor3d() requires shape to be provided when `values` are a flat array");return vd(e,t,r,n)}var jl;function A6(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,r=!1,s=!1,a=!1,o=!1,i=!1;if(e.data instanceof Uint8Array)n=!0;else if(typeof ImageData!="undefined"&&e instanceof ImageData)r=!0;else if(typeof HTMLVideoElement!="undefined"&&e instanceof HTMLVideoElement)s=!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(s){let f=2;if(s&&e.readyState<f)throw new Error("The video element has not loaded data yet. Please wait for `loadeddata` event on the <video> element.")}if(ny(ey,G.backendName)!=null){let f={pixels:e},m={numChannels:t};return G.runKernel(ey,f,m)}let[u,c]=s?[e.videoWidth,e.videoHeight]:[e.width,e.height],d;o?d=e.getContext("2d").getImageData(0,0,u,c).data:r||n?d=e.data:(a||s||i)&&(jl==null&&(jl=document.createElement("canvas").getContext("2d")),jl.canvas.width=u,jl.canvas.height=c,jl.drawImage(e,0,0,u,c),d=jl.getImageData(0,0,u,c).data);let h;if(t===4)h=new Int32Array(d);else{let f=u*c;h=new Int32Array(f*t);for(let m=0;m<f;m++)for(let g=0;g<t;++g)h[m*t+g]=d[m*4+g]}return IG(h,[c,u,t],"int32")}function SG(e){return e!=null&&e.data instanceof Uint8Array}function TG(){return typeof window!="undefined"&&typeof ImageBitmap!="undefined"&&window.hasOwnProperty("createImageBitmap")}function NG(e){return e!=null&&e.width!==0&&e.height!==0}function CG(e){return TG()&&!(e instanceof ImageBitmap)&&NG(e)&&!SG(e)}async function EG(e,t=3){let n=null;if(re().getBool("WRAP_TO_IMAGEBITMAP")&&CG(e)){let r;try{r=await createImageBitmap(e,{premultiplyAlpha:"none"})}catch(s){r=null}r!=null&&r.width===e.width&&r.height===e.height?n=r:n=e}else n=e;return A6(n,t)}async function $G(e,t){let n=O(e,"img","toPixels");if(!(e instanceof Ot)){let u=n;n=xe(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[r,s]=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(s*r*4);for(let u=0;u<r*s;++u){let c=[0,0,0,255];for(let h=0;h<a;h++){let p=o[u*a+h];if(n.dtype==="float32"){if(p<0||p>1)throw new Error(`Tensor values for a float32 Tensor must be in the range [0 - 1] but encountered ${p}.`)}else if(n.dtype==="int32"&&(p<0||p>255))throw new Error(`Tensor values for a int32 Tensor must be in the range [0 - 255] but encountered ${p}.`);a===1?(c[0]=p*i,c[1]=p*i,c[2]=p*i):c[h]=p*i}let d=u*4;l[d+0]=Math.round(c[0]),l[d+1]=Math.round(c[1]),l[d+2]=Math.round(c[2]),l[d+3]=Math.round(c[3])}if(t!=null){t.width=s,t.height=r;let u=t.getContext("2d"),c=new ImageData(l,s,r);u.putImageData(c,0,0)}return n!==e&&n.dispose(),l}var RG=V({fromPixels_:A6}),x6={};_e(x6,{prepareAndValidate:()=>b6});function b6(e,t){let n=e.shape.length,r=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(r<1)throw new Error(`tf.gatherND() expects the indices to be rank 1 or higher, but the rank was ${r}.`);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[r-1]>n)throw new Error(`index innermost dimension length must be <= tensor rank; saw: ${t.shape[r-1]} vs. ${n}`);if(sn(e.shape)===0)throw new Error(`Requested more than 0 entries, but input is empty. Input shape: ${e.shape}.`);let s=t.shape,a=s[s.length-1],o=1;for(let d=0;d<s.length-1;++d)o*=s[d];let i=e.shape,l=s.slice();l.pop();let u=1;for(let d=a;d<n;++d)u*=i[d],l.push(i[d]);let c=[...qi(e.shape).map(d=>d/u),1].slice(0,a);return[l,o,u,c]}var v6={};_e(v6,{calculateShapes:()=>w6,validateInput:()=>Ty,validateUpdateShape:()=>Sy});function Sy(e,t,n){let r=t.rank>1?t.shape[t.rank-1]:1,s=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: ${r}, and batchDim: ${s}.`;if(n.rank<s)throw new Error(a+` update.rank < ${s}. `);if(e.length<r+(n.rank-s))throw new Error(a+` Output shape length < ${r+(n.rank-s)}`);if(n.rank!==s+e.length-r)throw new Error(a+` update.rank != ${s+e.length-r}`);for(let o=0;o<s;++o)if(n.shape[o]!==t.shape[o])throw new Error(a+` updates.shape[${o}] (${n.shape[o]}) != indices.shape[${o}] (${t.shape[o]}).`);for(let o=0;o<n.rank-s;++o)if(n.shape[o+s]!==e[o+r])throw new Error(a+` updates.shape[${o+s}] (${n.shape[o+s]}) != shape[${o+s}] (${e[o+s]})`)}function Ty(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}`)}Sy(n,t,e)}function w6(e,t,n){let r=t.shape.length,s=r>1?t.shape[r-1]:1,a=n.length,o=1;for(let d=s;d<a;++d)o*=n[d];let i=s<1?1:s,l=sn(t.shape)/i,u=[...qi(n.slice(0,s)),1],c=sn(n);return{sliceRank:s,numUpdates:l,sliceSize:o,strides:u,outputSize:c}}var Cn={};_e(Cn,{assertParamsValid:()=>_G,computeFlatOffset:()=>FG,computeOutShape:()=>k6,getNormalizedAxes:()=>N6,isSliceContinous:()=>DG,maskToAxes:()=>bf,parseSliceParams:()=>D6,sliceInfo:()=>MG,startForAxis:()=>R6,startIndicesWithElidedDims:()=>C6,stopForAxis:()=>_6,stopIndicesWithElidedDims:()=>E6,stridesForAxis:()=>$6,stridesWithElidedDims:()=>I6});function _G(e,t,n){let r=e.shape.length;z(r===t.length,()=>`Error in slice${r}D: Length of begin ${t} must match the rank of the array (${r}).`),z(r===n.length,()=>`Error in slice${r}D: Length of size ${n} must match the rank of the array (${r}).`);for(let s=0;s<r;++s)z(t[s]+n[s]<=e.shape[s],()=>`Error in slice${r}D: begin[${s}] + size[${s}] (${t[s]+n[s]}) would overflow input.shape[${s}] (${e.shape[s]})`)}function bf(e){let t=[],n=0;for(;e>0;)e&1&&t.push(n),e/=2,n++;return t}function k6(e,t,n){let r=[];for(let s=0;s<e.length;s++)r[s]=Math.ceil((t[s]-e[s])/n[s]);return r}function I6(e,t,n,r){let s=[...e];for(let a=s.length;a<r.length;a++)s.push(1);for(let a=0;a<n;a++)a===0?s[t]=1:(s.splice(t,0,1),s.pop());return s}function S6(e,t,n){return n<=e?n:n-(t-1)}function T6(e,t){let n=[];for(let r=0;r<e;r++)n.push(t+r);return n}function N6(e,t,n,r,s,a,o,i,l){let u=e.length,c=new Array(u),d=new Array(u),h=new Array(u);if(t.length&&n>0){let p=t[0],f=n+1;c=C6(o,p,f,r,e),d=E6(i,p,f,s,e),h=I6(a,p,f,e)}else for(let p=0;p<u;p++)c[p]=R6(o,r,a,e,p,l),d[p]=_6(i,s,a,e,p,l),h[p]=$6(a,p,l);return{begin:c,end:d,strides:h}}function C6(e,t,n,r,s){let a=[...s],o=T6(n,t);for(let i=0;i<a.length;i++)if(o.indexOf(i)>-1)a[i]=0;else{let l=S6(t,n,i),u=r[l];e&1<<l&&(u=0),a[i]=u}return a}function E6(e,t,n,r,s){let a=[...s],o=T6(n,t);for(let i=0;i<a.length;i++)if(o.indexOf(i)>-1)a[i]=Number.MAX_SAFE_INTEGER;else{let l=S6(t,n,i),u=r[l];e&1<<l&&(u=Number.MAX_SAFE_INTEGER),a[i]=u}for(let i=0;i<a.length;i++){let l=s[i];a[i]<0&&(a[i]+=l),a[i]=mc(0,a[i],s[i])}return a}function $6(e,t,n){let r=e[t];return(n&1<<t||r==null)&&(r=1),r}function R6(e,t,n,r,s,a){let o=t[s],i=n[s]||1;(e&1<<s||a&1<<s||o==null)&&(i>0?o=Number.MIN_SAFE_INTEGER:o=Number.MAX_SAFE_INTEGER);let l=r[s];return o<0&&(o+=l),o=mc(0,o,l-1),o}function _6(e,t,n,r,s,a){let o=t[s],i=n[s]||1;(e&1<<s||a&1<<s||o==null)&&(i>0?o=Number.MAX_SAFE_INTEGER:o=Number.MIN_SAFE_INTEGER);let l=r[s];return o<0&&(o+=l),i>0?o=mc(0,o,l):o=mc(-1,o,l-1),o}function DG(e,t,n){let r=n.length;for(let s=0;s<n.length;s++)if(n[s]>1){r=s;break}for(let s=r+1;s<n.length;s++)if(t[s]>0||n[s]!==e[s])return!1;return!0}function FG(e,t){let n=e.length>0?e[e.length-1]:1;for(let r=0;r<e.length-1;r++)n+=e[r]*t[r];return n}function D6(e,t,n){let r,s=e.shape.length;typeof t=="number"?r=[t,...new Array(s-1).fill(0)]:t.length<s?r=t.concat(new Array(s-t.length).fill(0)):r=t.slice(),r.forEach(o=>{z(o!==-1,()=>"slice() does not support negative begin indexing.")});let a;return n==null?a=new Array(s).fill(-1):typeof n=="number"?a=[n,...new Array(s-1).fill(-1)]:n.length<s?a=n.concat(new Array(s-n.length).fill(-1)):a=n,a=a.map((o,i)=>o>=0?o:(z(o===-1,()=>`Negative size values should be exactly -1 but got ${o} for the slice() size at index ${i}.`),e.shape[i]-r[i])),[r,a]}function MG(e,t,n,r,s,a,o,i,l){let u=t.slice(),c=n.slice(),d=r;r==null&&(d=new Array(u.length));let h=bf(o);if(h.length>1)throw new Error("Multiple ellipses in slice is not allowed.");if(o!==0&&i!==0)throw new Error("Using both ellipsisMask and newAxisMask is not yet supported.");if(o!==0&&l!==0)throw new Error("Using both ellipsisMask and shrinkAxisMask is not yet supported.");let p=e.length-u.length,f=bf(i),m=e.slice();f.forEach(w=>{u[w]=0,c[w]=1,m.splice(w,0,1)});let{begin:g,end:y,strides:A}=N6(m,h,p,u,c,d,s,a,o);u=g,c=y,d=A;let x=bf(l);x.forEach(w=>{c[w]=u[w]+1,d[w]=1});let b=k6(u,c,d),v=b.filter((w,S)=>x.indexOf(S)===-1);return{nonStrided:d.every(w=>w===1),$begin:u,$end:c,$strides:d,size:b,newShape:m,outShape:v}}var ue={};_e(ue,{Serializable:()=>F6,SerializationMap:()=>qo,registerClass:()=>La});var F6=class{getClassName(){return this.constructor.className}static fromConfig(e,t){return new e(t)}},qo=class{constructor(){this.classNameMap={}}static getMap(){return qo.instance==null&&(qo.instance=new qo),qo.instance}static register(e){qo.getMap().classNameMap[e.className]=[e,e.fromConfig]}};function La(e){z(e.className!=null,()=>"Class being registered does not have the static className property defined."),z(typeof e.className=="string",()=>"className is required to be a string, but got type "+typeof e.className),z(e.className.length>0,()=>"Class being registered has an empty-string as its className, which is disallowed."),qo.register(e)}function M6(e){re().getBool("DEPRECATION_WARNINGS_ENABLED")&&console.warn(e+" You can disable deprecation warnings with tf.disableDeprecationWarnings().")}fH(M6);function Ba(){return G}function Ny(){return G.memory()}function Y(e,t){return G.tidy(e,t)}function Ge(e){dy(e).forEach(n=>n.dispose())}function In(e){return G.keep(e)}function Cy(e,t,n=1){return G.registerBackend(e,t,n)}function OG(){return G.backend}function PG(e,t){let n=O(e,"a","add"),r=O(t,"b","add");[n,r]=Vt(n,r);let s={a:n,b:r};return G.runKernel(Ma,s)}var de=V({add_:PG});function zG(e,t){let n=O(e,"a","floorDiv"),r=O(t,"b","floorDiv");[n,r]=Vt(n,r);let s={a:n,b:r};return G.runKernel(ul,s)}var O6=V({floorDiv_:zG});function LG(e,t){let n=O(e,"a","div"),r=O(t,"b","div");if([n,r]=Vt(n,r),n.dtype==="int32"&&r.dtype==="int32")return O6(n,r);let s={a:n,b:r},a={};return G.runKernel(ol,s,a)}var Re=V({div_:LG});function BG(e,t){let n=O(e,"a","mul"),r=O(t,"b","mul");[n,r]=Vt(n,r);let s={a:n,b:r};return G.runKernel(Mo,s)}var j=V({mul_:BG});function WG(e){let t=O(e,"x","abs");if(t.dtype==="complex64"){let n={x:t};return G.runKernel(Xp,n)}else{let n={x:t};return G.runKernel(Ac,n)}}var gn=V({abs_:WG});function VG(e){let n={x:O(e,"x","acos")};return G.runKernel(xc,n)}var UG=V({acos_:VG});function HG(e){let n={x:O(e,"x","acosh")};return G.runKernel(bc,n)}var GG=V({acosh_:HG});function jG(e){z(Array.isArray(e),()=>"The argument passed to tf.addN() must be a list of tensors"),z(e.length>=1,()=>`Must pass at least one tensor to tf.addN(), but got ${e.length}`);let t=e.map((s,a)=>O(s,`tensors${a}`,"addN")),n=t[0];t.forEach(s=>{if(s.dtype!==n.dtype)throw new Error("All tensors passed to tf.addN() must have the same dtype")}),t.forEach(s=>{if(!Fa(s.shape,n.shape))throw new Error("All tensors passed to tf.addN() must have the same shape")});let r=t;return G.runKernel(Xi,r)}var qG=V({addN_:jG});function KG(e,t=null,n=!1){let s={x:O(e,"x","all","bool")},a={axis:t,keepDims:n};return G.runKernel(vc,s,a)}var P6=V({all_:KG});function XG(e,t=null,n=!1){let s={x:O(e,"x","any","bool")},a={axis:t,keepDims:n};return G.runKernel(wc,s,a)}var Ey=V({any_:XG});function ZG(e,t=0){let r={x:O(e,"x","argMax")},s={axis:t};return G.runKernel(Zi,r,s)}var kd=V({argMax_:ZG});function YG(e,t=0){let r={x:O(e,"x","argMin")},s={axis:t};return G.runKernel(qp,r,s)}var JG=V({argMin_:YG});function QG(e){let n={x:O(e,"x","asin")};return G.runKernel(kc,n)}var ej=V({asin_:QG});function tj(e){let n={x:O(e,"x","asinh")};return G.runKernel(Ic,n)}var nj=V({asinh_:tj});function rj(e){let n={x:O(e,"x","atan")};return G.runKernel(Sc,n)}var sj=V({atan_:rj});function aj(e,t){let n=O(e,"a","atan2"),r=O(t,"b","atan2");[n,r]=Vt(n,r);let s={a:n,b:r};return G.runKernel(Nc,s)}var oj=V({atan2_:aj});function ij(e){let n={x:O(e,"x","atanh")};return G.runKernel(Tc,n)}var lj=V({atanh_:ij});function uj(e,t,n,r,s="NHWC",a){let o=e[3],i=[...t,o],l=B6(s);return Id(e,i,n,a,r,null,null,l)}function z6(e,t,n,r,s,a,o="channelsLast"){let[i,l]=vf(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 Id(e,u,n,r,s,a,!1,o)}function cj(e,t,n,r,s,a,o="NDHWC"){let[i,l,u]=Ry(t),c,d;if(o==="NDHWC")d="channelsLast",c=[i,l,u,e[4],e[4]];else if(o==="NCDHW")d="channelsFirst",c=[i,l,u,e[1],e[1]];else throw new Error(`Unknown dataFormat ${o}`);return L6(e,c,n,r,s,!1,d,a)}function Id(e,t,n,r,s,a,o=!1,i="channelsLast"){let[l,u,c,d]=[-1,-1,-1,-1];if(i==="channelsLast")[l,u,c,d]=e;else if(i==="channelsFirst")[l,d,u,c]=e;else throw new Error(`Unknown dataFormat ${i}`);let[h,p,,f]=t,[m,g]=vf(n),[y,A]=vf(r),x=ql(h,y),b=ql(p,A),{padInfo:v,outHeight:I,outWidth:w}=pj(s,u,c,m,g,x,b,a,i),S=o?f*d:f,E;return i==="channelsFirst"?E=[l,S,I,w]:i==="channelsLast"&&(E=[l,I,w,S]),{batchSize:l,dataFormat:i,inHeight:u,inWidth:c,inChannels:d,outHeight:I,outWidth:w,outChannels:S,padInfo:v,strideHeight:m,strideWidth:g,filterHeight:h,filterWidth:p,effectiveFilterHeight:x,effectiveFilterWidth:b,dilationHeight:y,dilationWidth:A,inShape:e,outShape:E,filterShape:t}}function L6(e,t,n,r,s,a=!1,o="channelsLast",i){let[l,u,c,d,h]=[-1,-1,-1,-1,-1];if(o==="channelsLast")[l,u,c,d,h]=e;else if(o==="channelsFirst")[l,h,u,c,d]=e;else throw new Error(`Unknown dataFormat ${o}`);let[p,f,m,,g]=t,[y,A,x]=Ry(n),[b,v,I]=Ry(r),w=ql(p,b),S=ql(f,v),E=ql(m,I),{padInfo:D,outDepth:$,outHeight:R,outWidth:N}=fj(s,u,c,d,y,A,x,w,S,E,i),M=a?g*h:g,B;return o==="channelsFirst"?B=[l,M,$,R,N]:o==="channelsLast"&&(B=[l,$,R,N,M]),{batchSize:l,dataFormat:o,inDepth:u,inHeight:c,inWidth:d,inChannels:h,outDepth:$,outHeight:R,outWidth:N,outChannels:M,padInfo:D,strideDepth:y,strideHeight:A,strideWidth:x,filterDepth:p,filterHeight:f,filterWidth:m,effectiveFilterDepth:w,effectiveFilterHeight:S,effectiveFilterWidth:E,dilationDepth:b,dilationHeight:v,dilationWidth:I,inShape:e,outShape:B,filterShape:t}}function dj(e,t,n,r,s){r==null&&(r=$y(e,t,n));let a=e[0],o=e[1],i=Ko((a-t+2*r)/n+1,s),l=Ko((o-t+2*r)/n+1,s);return[i,l]}function hj(e,t,n,r,s,a){s==null&&(s=$y(e,t,r));let o=e[0],i=e[1],l=e[2],u=Ko((o-t+2*s)/r+1,a),c=Ko((i-t+2*s)/r+1,a),d=Ko((l-t+2*s)/r+1,a);return[u,c,d,n]}function $y(e,t,n,r=1){let s=ql(t,r);return Math.floor((e[0]*(n-1)-n+s)/2)}function vf(e){return typeof e=="number"?[e,e,e]:e.length===2?[e[0],e[1],1]:e}function Ry(e){return typeof e=="number"?[e,e,e]:e}function ql(e,t){return t<=1?e:e+(e-1)*(t-1)}function pj(e,t,n,r,s,a,o,i,l){let u,c,d;if(typeof e=="number"){u={top:e,bottom:e,left:e,right:e,type:e===0?"VALID":"NUMBER"};let p=dj([t,n],a,r,e,i);c=p[0],d=p[1]}else if(e==="same"){c=Math.ceil(t/r),d=Math.ceil(n/s);let h=Math.max(0,(c-1)*r+a-t),p=Math.max(0,(d-1)*s+o-n),f=Math.floor(h/2),m=h-f,g=Math.floor(p/2),y=p-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)/r),d=Math.ceil((n-o+1)/s);else if(typeof e=="object"){let h=l==="channelsLast"?e[1][0]:e[2][0],p=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:h,bottom:p,left:f,right:m,type:h===0&&p===0&&f===0&&m===0?"VALID":"EXPLICIT"},c=Ko((t-a+h+p)/r+1,i),d=Ko((n-o+f+m)/s+1,i)}else throw Error(`Unknown padding parameter: ${e}`);return{padInfo:u,outHeight:c,outWidth:d}}function fj(e,t,n,r,s,a,o,i,l,u,c){let d,h,p,f;if(typeof e=="number"){d={top:e,bottom:e,left:e,right:e,front:e,back:e,type:e===0?"VALID":"NUMBER"};let g=hj([t,n,r,1],i,1,s,e,c);h=g[0],p=g[1],f=g[2]}else if(e==="same"){h=Math.ceil(t/s),p=Math.ceil(n/a),f=Math.ceil(r/o);let m=(h-1)*s+i-t,g=(p-1)*a+l-n,y=(f-1)*o+u-r,A=Math.floor(m/2),x=m-A,b=Math.floor(g/2),v=g-b,I=Math.floor(y/2),w=y-I;d={top:b,bottom:v,left:I,right:w,front:A,back:x,type:"SAME"}}else if(e==="valid")d={top:0,bottom:0,left:0,right:0,front:0,back:0,type:"VALID"},h=Math.ceil((t-i+1)/s),p=Math.ceil((n-l+1)/a),f=Math.ceil((r-u+1)/o);else throw Error(`Unknown padding parameter: ${e}`);return{padInfo:d,outDepth:h,outHeight:p,outWidth:f}}function Ko(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 Wa(e){let[t,n,r]=vf(e);return t===1&&n===1&&r===1}function Es(e,t){return Wa(e)||Wa(t)}function B6(e){if(e==="NHWC")return"channelsLast";if(e==="NCHW")return"channelsFirst";throw new Error(`Unknown dataFormat ${e}`)}function mj(e,t){let r={x:O(e,"x","reshape","string_or_numeric")},s={shape:t};return G.runKernel(Jc,r,s)}var Z=V({reshape_:mj});function gj(e,t,n,r,s){let a=O(e,"x","avgPool","float32"),o=1;z(Es(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=Z(a,[1,a.shape[0],a.shape[1],a.shape[2]])),z(i.rank===4,()=>`Error in avgPool: x must be rank 4 but got rank ${i.rank}.`),s!=null&&z(fn(r),()=>`Error in avgPool: pad must be an integer when using, dimRoundingMode ${s} but got pad ${r}.`);let u={x:i},c={filterSize:t,strides:n,pad:r,dimRoundingMode:s},d=G.runKernel(Yi,u,c);return d=xe(d,a.dtype),l?Z(d,[d.shape[1],d.shape[2],d.shape[3]]):d}var _y=V({avgPool_:gj});function yj(e,t,n,r,s,a="NDHWC"){let o=O(e,"x","avgPool3d","float32"),i=o,l=!1;o.rank===4&&(l=!0,i=Z(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),z(i.rank===5,()=>`Error in avgPool3d: x must be rank 5 but got rank ${i.rank}.`),z(a==="NDHWC",()=>`Error in avgPool3d: Only NDHWC is currently supported, but got dataFormat of ${a}`),s!=null&&z(fn(r),()=>`Error in avgPool3d: pad must be an integer when using, dimRoundingMode ${s} but got pad ${r}.`);let u={x:i},c={filterSize:t,strides:n,pad:r,dimRoundingMode:s,dataFormat:a},d=G.runKernel(Kp,u,c);return d=xe(d,i.dtype),l?Z(d,[d.shape[1],d.shape[2],d.shape[3],d.shape[4]]):d}var W6=V({avgPool3d_:yj});function Aj(e,t=0){z(e.length>=1,()=>"Pass at least one tensor to concat");let n=gf(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 za(n[0]);let r=n,s={axis:t};return G.runKernel(Ec,r,s)}var an=V({concat_:Aj});function xj(e){let n={x:O(e,"x","sigmoid")};return G.runKernel(_l,n)}var ia=V({sigmoid_:xj});function bj(e,t,n){let r=O(e,"x","slice","string_or_numeric");if(r.rank===0)throw new Error("Slicing scalar is not possible");let s={x:r},a={begin:t,size:n};return G.runKernel(nd,s,a)}var at=V({slice_:bj});function vj(e){let n={x:O(e,"x","tanh")};return G.runKernel(Pl,n)}var Sd=V({tanh_:vj});function wj(e,t,n,r,s,a){let o=O(e,"forgetBias","basicLSTMCell"),i=O(t,"lstmKernel","basicLSTMCell"),l=O(n,"lstmBias","basicLSTMCell"),u=O(r,"data","basicLSTMCell"),c=O(s,"c","basicLSTMCell"),d=O(a,"h","basicLSTMCell"),h=an([u,d],1),p=rt(h,i),f=de(p,l),m=f.shape[0],g=f.shape[1]/4,y=[m,g],A=at(f,[0,0],y),x=at(f,[0,g],y),b=at(f,[0,g*2],y),v=at(f,[0,g*3],y),I=de(j(ia(A),Sd(x)),j(c,ia(de(o,b)))),w=j(Sd(I),ia(v));return[I,w]}var Lwe=V({basicLSTMCell_:wj});function kj(e,t,n){let r=O(e,"x","batchToSpaceND"),s=t.reduce((i,l)=>i*l);z(r.rank>=1+t.length,()=>`input rank is ${r.rank} but should be > than blockShape.length ${t.length}`),z(n.length===t.length,()=>`crops.length is ${n.length} but should be equal to blockShape.length ${t.length}`),z(r.shape[0]%s==0,()=>`input tensor batch is ${r.shape[0]} but is not divisible by the product of the elements of blockShape ${t.join(" * ")} === ${s}`);let a={x:r},o={blockShape:t,crops:n};return G.runKernel(Cc,a,o)}var Dy=V({batchToSpaceND_:kj});function Ij(e){let t;return e.rank===0||e.rank===1?t=Z(e,[1,1,1,e.size]):e.rank===2?t=Z(e,[1,1,e.shape[0],e.shape[1]]):e.rank===3?t=Z(e,[1,e.shape[0],e.shape[1],e.shape[2]]):t=e,t}function Sj(e,t,n,r,s,a){a==null&&(a=.001);let o=O(e,"x","batchNorm"),i=O(t,"mean","batchNorm"),l=O(n,"variance","batchNorm"),u;s!=null&&(u=O(s,"scale","batchNorm"));let c;r!=null&&(c=O(r,"offset","batchNorm")),z(i.rank===l.rank,()=>"Batch normalization gradient requires mean and variance to have equal ranks."),z(c==null||i.rank===c.rank,()=>"Batch normalization gradient requires mean and offset to have equal ranks."),z(u==null||i.rank===u.rank,()=>"Batch normalization gradient requires mean and scale to have equal ranks.");let h={x:Ij(o),scale:u,offset:c,mean:i,variance:l},p={varianceEpsilon:a},f=G.runKernel(cl,h,p);return Z(f,o.shape)}var Td=V({batchNorm_:Sj});function Tj(e,t,n,r,s,a){let o=O(e,"x","batchNorm"),i=O(t,"mean","batchNorm"),l=O(n,"variance","batchNorm"),u;s!=null&&(u=O(s,"scale","batchNorm"));let c;return r!=null&&(c=O(r,"offset","batchNorm")),z(o.rank===2,()=>`Error in batchNorm2D: x must be rank 2 but got rank ${o.rank}.`),z(i.rank===2||i.rank===1,()=>`Error in batchNorm2D: mean must be rank 2 or rank 1 but got rank ${i.rank}.`),z(l.rank===2||l.rank===1,()=>`Error in batchNorm2D: variance must be rank 2 or rank 1 but got rank ${l.rank}.`),u!=null&&z(u.rank===2||u.rank===1,()=>`Error in batchNorm2D: scale must be rank 2 or rank 1 but got rank ${u.rank}.`),c!=null&&z(c.rank===2||c.rank===1,()=>`Error in batchNorm2D: offset must be rank 2 or rank 1 but got rank ${c.rank}.`),Td(o,i,l,c,u,a)}var Nj=V({batchNorm2d_:Tj});function Cj(e,t,n,r,s,a){let o=O(e,"x","batchNorm"),i=O(t,"mean","batchNorm"),l=O(n,"variance","batchNorm"),u;s!=null&&(u=O(s,"scale","batchNorm"));let c;return r!=null&&(c=O(r,"offset","batchNorm")),z(o.rank===3,()=>`Error in batchNorm3D: x must be rank 3 but got rank ${o.rank}.`),z(i.rank===3||i.rank===1,()=>`Error in batchNorm3D: mean must be rank 3 or rank 1 but got rank ${i.rank}.`),z(l.rank===3||l.rank===1,()=>`Error in batchNorm3D: variance must be rank 3 or rank 1 but got rank ${l.rank}.`),u!=null&&z(u.rank===3||u.rank===1,()=>`Error in batchNorm3D: scale must be rank 3 or rank 1 but got rank ${u.rank}.`),c!=null&&z(c.rank===3||c.rank===1,()=>`Error in batchNorm3D: offset must be rank 3 or rank 1 but got rank ${c.rank}.`),Td(o,i,l,c,u,a)}var Ej=V({batchNorm3d_:Cj});function $j(e,t,n,r,s,a){let o=O(e,"x","batchNorm"),i=O(t,"mean","batchNorm"),l=O(n,"variance","batchNorm"),u;s!=null&&(u=O(s,"scale","batchNorm"));let c;return r!=null&&(c=O(r,"offset","batchNorm")),z(o.rank===4,()=>`Error in batchNorm4D: x must be rank 4 but got rank ${o.rank}.`),z(i.rank===4||i.rank===1,()=>`Error in batchNorm4D: mean must be rank 4 or rank 1 but got rank ${i.rank}.`),z(l.rank===4||l.rank===1,()=>`Error in batchNorm4D: variance must be rank 4 or rank 1 but got rank ${l.rank}.`),u!=null&&z(u.rank===4||u.rank===1,()=>`Error in batchNorm4D: scale must be rank 4 or rank 1 but got rank ${u.rank}.`),c!=null&&z(c.rank===4||c.rank===1,()=>`Error in batchNorm4D: offset must be rank 4 or rank 1 but got rank ${c.rank}.`),Td(o,i,l,c,u,a)}var Rj=V({batchNorm4d_:$j});function _j(e,t,n){let r=O(e,"x","bincount"),s=O(t,"weights","bincount");z(r.dtype==="int32",()=>`Error in bincount: input dtype must be int32, but got ${r.dtype}`),z(n>=0,()=>`size must be non-negative, but got ${n}.`),z(s.size===r.size||s.size===0,()=>`Error in bincount: weights must have the same size as input or0-length, but got input shape: ${r.shape}, weights shape: ${s.shape}.`);let a={x:r,weights:s},o={size:n};return G.runKernel(b1,a,o)}var V6=V({bincount_:_j});function Dj(e,t){let n=O(e,"broadcastTo","x"),r=n.shape;if(t.some(u=>!(u>0)||u%1!=0))throw new Error(`broadcastTo(): Invalid broadcast shape [${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=Z(n,u)}let s=n.shape,a=Array.from(t);for(let u=t.length-1;u>=0;u--)if(s[u]===t[u])a[u]=1;else if(n.shape[u]!==1)throw new Error(`broadcastTo(): [${r}] cannot be broadcast to [${t}].`);if(a.map((u,c)=>u>1?c:-1).filter(u=>u>=0).length===0)return za(n);let i={x:n},l={reps:a};return G.runKernel(Lo,i,l)}var wf=V({broadcastTo_:Dj});function Fj(e){let n={x:O(e,"x","ceil")};return G.runKernel(No,n)}var Mj=V({ceil_:Fj});function Oj(e,t,n){let r=O(e,"x","clipByValue");z(t<=n,()=>`Error in clip: min (${t}) must be less than or equal to max (${n}).`);let s={x:r},a={clipValueMin:t,clipValueMax:n};return G.runKernel(Co,s,a)}var Cr=V({clipByValue_:Oj});function Pj(e){return an(e,0)}var zj=V({concat1d_:Pj});function Lj(e,t){return an(e,t)}var Bj=V({concat2d_:Lj});function Wj(e,t){return an(e,t)}var Vj=V({concat3d_:Wj});function Uj(e,t){return an(e,t)}var Hj=V({concat4d_:Uj});function Gj(e,t,n,r,s="NHWC",a=[1,1],o){let i=O(e,"x","conv2d"),l=O(t,"filter","conv2d"),u=i,c=!1;i.rank===3&&(c=!0,u=Z(i,[1,i.shape[0],i.shape[1],i.shape[2]])),z(u.rank===4,()=>`Error in conv2d: input must be rank 4, but got rank ${u.rank}.`),z(l.rank===4,()=>`Error in conv2d: filter must be rank 4, but got rank ${l.rank}.`),o!=null&&z(fn(r),()=>`Error in conv2d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${r}.`);let d=s==="NHWC"?u.shape[3]:u.shape[1];z(d===l.shape[2],()=>`Error in conv2d: depth of input (${d}) must match input depth for filter ${l.shape[2]}.`),z(Es(n,a),()=>`Error in conv2D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`);let h={x:u,filter:l},p={strides:n,pad:r,dataFormat:s,dilations:a,dimRoundingMode:o},f=G.runKernel(el,h,p);return c?Z(f,[f.shape[1],f.shape[2],f.shape[3]]):f}var Xo=V({conv2d_:Gj});function jj(e,t,n,r,s="NWC",a=1,o){let i=O(e,"x","conv1d"),l=O(t,"filter","conv1d"),u=i,c=!1;i.rank===2&&(c=!0,u=Z(i,[1,i.shape[0],i.shape[1]])),z(u.rank===3,()=>`Error in conv1d: input must be rank 3, but got rank ${u.rank}.`),z(l.rank===3,()=>`Error in conv1d: filter must be rank 3, but got rank ${l.rank}.`),o!=null&&z(fn(r),()=>`Error in conv1d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${r}.`),z(u.shape[2]===l.shape[1],()=>`Error in conv1d: depth of input (${u.shape[2]}) must match input depth for filter ${l.shape[1]}.`),z(Es(n,a),()=>`Error in conv1D: Either stride or dilation must be 1. Got stride ${n} and dilation '${a}'`),z(s==="NWC",()=>`Error in conv1d: got dataFormat of ${s} but only NWC is currently supported.`);let d=Z(l,[1,l.shape[0],l.shape[1],l.shape[2]]),h=Z(u,[u.shape[0],1,u.shape[1],u.shape[2]]),g=Xo(h,d,[1,n],r,"NHWC",[1,a],o);return c?Z(g,[g.shape[2],g.shape[3]]):Z(g,[g.shape[0],g.shape[2],g.shape[3]])}var U6=V({conv1d_:jj});function qj(e,t,n,r,s,a="NHWC",o){z(e.length===t.rank,()=>`Length of inShape (${e.length}) and rank of dy (${t.rank}) must match`);let i=e,l=t,u=!1;t.rank===3&&(u=!0,l=Z(t,[1,t.shape[0],t.shape[1],t.shape[2]]),i=[1,e[0],e[1],e[2]]),z(i.length===4,()=>`Error in conv2dDerInput: inShape must be length 4, but got length ${i.length}.`),z(l.rank===4,()=>`Error in conv2dDerInput: dy must be rank 4, but got rank ${l.rank}`),z(n.rank===4,()=>`Error in conv2dDerInput: filter must be rank 4, but got rank ${n.rank}`);let c=a==="NHWC"?i[3]:i[1],d=a==="NHWC"?l.shape[3]:l.shape[1];z(c===n.shape[2],()=>`Error in conv2dDerInput: depth of input (${c}) must match input depth for filter ${n.shape[2]}.`),z(d===n.shape[3],()=>`Error in conv2dDerInput: depth of output (${d}) must match output depth for filter ${n.shape[3]}.`),o!=null&&z(fn(s),()=>`Error in conv2dDerInput: pad must be an integer when using, dimRoundingMode ${o} but got pad ${s}.`);let h={dy:l,filter:n},p={strides:r,pad:s,dataFormat:a,dimRoundingMode:o,inputShape:i},f=G.runKernel(tl,h,p);return u?Z(f,[f.shape[1],f.shape[2],f.shape[3]]):f}var Fy=V({conv2DBackpropInput_:qj});function Kj(e,t,n,r,s,a){let o=O(e,"x","conv2dTranspose"),i=O(t,"filter","conv2dTranspose");return Fy(n,o,i,r,s,"NHWC",a)}var H6=V({conv2dTranspose_:Kj});function Xj(e,t,n,r,s="NDHWC",a=[1,1,1]){let o=O(e,"x","conv3d"),i=O(t,"filter","conv3d"),l=o,u=!1;o.rank===4&&(u=!0,l=Z(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),z(l.rank===5,()=>`Error in conv3d: input must be rank 5, but got rank ${l.rank}.`),z(i.rank===5,()=>`Error in conv3d: filter must be rank 5, but got rank ${i.rank}.`),z(l.shape[4]===i.shape[3],()=>`Error in conv3d: depth of input (${l.shape[4]}) must match input depth for filter ${i.shape[3]}.`),z(Es(n,a),()=>`Error in conv3D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`),z(s==="NDHWC",()=>`Error in conv3d: got dataFormat of ${s} but only NDHWC is currently supported.`);let c={x:l,filter:i},d={strides:n,pad:r,dataFormat:s,dilations:a},h=G.runKernel(Zp,c,d);return u?Z(h,[h.shape[1],h.shape[2],h.shape[3],h.shape[4]]):h}var G6=V({conv3d_:Xj});function Zj(e,t,n,r,s){z(e.length===t.rank,()=>`Length of inShape (${e.length}) and rank of dy (${t.rank}) must match`);let a=e,o=t,i=!1;t.rank===4&&(i=!0,o=Z(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];z(a.length===5,()=>`Error in conv3dDerInput: inShape must be length 5, but got length ${a.length}.`),z(o.rank===5,()=>`Error in conv3dDerInput: dy must be rank 5, but got rank ${o.rank}`),z(n.rank===5,()=>`Error in conv3dDerInput: filter must be rank 5, but got rank ${n.rank}`),z(l===n.shape[3],()=>`Error in conv3dDerInput: depth of input (${l}) must match input depth for filter ${n.shape[3]}.`),z(u===n.shape[4],()=>`Error in conv3dDerInput: depth of output (${u}) must match output depth for filter ${n.shape[4]}.`);let c={dy:o,filter:n},d={pad:s,strides:r,inputShape:a},h=G.runKernel(I1,c,d);return i?Z(h,[h.shape[1],h.shape[2],h.shape[3],h.shape[4]]):h}var j6=V({conv3DBackpropInput_:Zj});function Yj(e,t,n,r,s){let a=O(e,"x","conv3dTranspose"),o=O(t,"filter","conv3dTranspose");return j6(n,a,o,r,s)}var Jj=V({conv3dTranspose_:Yj});function Qj(e){let n={x:O(e,"x","cos")};return G.runKernel(nl,n)}var My=V({cos_:Qj});function eq(e){let n={x:O(e,"x","cosh")};return G.runKernel(rl,n)}var q6=V({cosh_:eq});function tq(e,t=0,n=!1,r=!1){let a={x:O(e,"x","cumsum")},o={axis:t,exclusive:n,reverse:r};return G.runKernel(sl,a,o)}var K6=V({cumsum_:tq});function nq(e,t,n,r=!1){let s=O(e,"x","denseBincount"),a=O(t,"weights","denseBincount");z(s.dtype==="int32",()=>`Error in denseBincount: input dtype must be int32, but got ${s.dtype}`),z(s.rank<=2,()=>`Error in denseBincount: input must be at most rank 2, but got rank ${s.rank}.`),z(n>=0,()=>`size must be non-negative, but got ${n}.`),z(a.size===s.size||a.size===0,()=>`Error in denseBincount: weights must have the same shape as x or 0-length, but got x shape: ${s.shape}, weights shape: ${a.shape}.`);let o={x:s,weights:a},i={size:n,binaryOutput:r};return G.runKernel(S1,o,i)}var rq=V({denseBincount_:nq});function sq(e,t,n="NHWC"){let r=O(e,"x","depthToSpace"),s=n==="NHWC"?r.shape[1]:r.shape[2],a=n==="NHWC"?r.shape[2]:r.shape[3],o=n==="NHWC"?r.shape[3]:r.shape[1];z(s*t>=0,()=>`Negative dimension size caused by overflow when multiplying
${s} and ${t} for depthToSpace with input shape
${r.shape}`),z(a*t>=0,()=>`Negative dimension size caused by overflow when multiplying
${a} and ${t} for depthToSpace with input shape
${r.shape}`),z(o%(t*t)==0,()=>`Dimension size must be evenly divisible by ${t*t} but is ${o} for depthToSpace with input shape ${r.shape}`);let i={x:r},l={blockSize:t,dataFormat:n};return G.runKernel(Rc,i,l)}var aq=V({depthToSpace_:sq});function oq(e,t,n,r,s="NHWC",a=[1,1],o){let i=O(e,"x","depthwiseConv2d"),l=O(t,"filter","depthwiseConv2d"),u=i,c=!1;i.rank===3&&(c=!0,u=Z(i,[1,i.shape[0],i.shape[1],i.shape[2]])),z(u.rank===4,()=>`Error in depthwiseConv2d: input must be rank 4, but got rank ${u.rank}.`),z(l.rank===4,()=>`Error in depthwiseConv2d: filter must be rank 4, but got rank ${l.rank}.`),z(u.shape[3]===l.shape[2],()=>`Error in depthwiseConv2d: number of input channels (${u.shape[3]}) must match the inChannels dimension in filter ${l.shape[2]}.`),o!=null&&z(fn(r),()=>`Error in depthwiseConv2d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${r}.`);let d={x:u,filter:l},h={strides:n,pad:r,dataFormat:s,dilations:a,dimRoundingMode:o},p=G.runKernel(al,d,h);return c?Z(p,[p.shape[1],p.shape[2],p.shape[3]]):p}var kf=V({depthwiseConv2d_:oq});function iq(e){let n={x:O(e,"x","diag")};return G.runKernel(C1,n)}var Bwe=V({diag_:iq});function lq(e,t,n,r,s=[1,1],a="NHWC"){let o=O(e,"x","dilation2d"),i=O(t,"filter","dilation2d");z(o.rank===3||o.rank===4,()=>`Error in dilation2d: input must be rank 3 or 4, but got rank ${o.rank}.`),z(i.rank===3,()=>`Error in dilation2d: filter must be rank 3, but got rank ${i.rank}.`),z(a==="NHWC",()=>`Error in dilation2d: Only NHWC is currently supported, but got dataFormat of ${a}`);let l=o,u=!1;o.rank===3&&(l=Z(o,[1,o.shape[0],o.shape[1],o.shape[2]]),u=!0);let c={x:l,filter:i},d={strides:n,pad:r,dilations:s},h=G.runKernel(Yp,c,d);return u?Z(h,[h.shape[1],h.shape[2],h.shape[3]]):h}var uq=V({dilation2d_:lq});function cq(e,t){let n=e.length,r=[];for(let s=0;s<n;s++){let a=n-1-s,o=e[a]||1;(t[t.length-1-s]||1)>1&&o===1&&r.unshift(a)}return r}function on(e,t){let n=[];for(let r=0;r<t.length;r++){let s=e[e.length-r-1],a=t.length-r-1,o=t[a];(s==null||s===1&&o>1)&&n.unshift(a)}return n}function $t(e,t){let n=[],r=Math.max(e.length,t.length);for(let s=0;s<r;s++){let a=e[e.length-s-1];a==null&&(a=1);let o=t[t.length-s-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 ${t}.`;throw Error(i)}else n.unshift(a)}return n}function dq(e,t){let n=O(e,"a","equal","string_or_numeric"),r=O(t,"b","equal","string_or_numeric");[n,r]=Vt(n,r),$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(il,s)}var Hr=V({equal_:dq});function hq(e,t,n){let r=O(t,"a","where"),s=O(n,"b","where"),a=O(e,"condition","where","bool"),o=$t($t(a.shape,r.shape),s.shape),i=wf(a,o),l=wf(r,o),u=wf(s,o),c={condition:i,t:l,e:u};return G.runKernel(ed,c)}var Zn=V({where_:hq});function pq(e){let n={x:O(e,"x","zerosLike")};return G.runKernel(hd,n)}var ot=V({zerosLike_:pq});function fq(e,t){let n=O(e,"a","div"),r=O(t,"b","div");[n,r]=Vt(n,r);let s=Re(n,r),a=ot(s),o=Hr(r,a);return Zn(o,a,s)}var mq=V({divNoNan_:fq});function gq(e,t){let n=O(e,"t1","dot"),r=O(t,"t2","dot");z((n.rank===1||n.rank===2)&&(r.rank===1||r.rank===2),()=>`Error in dot: inputs must all be rank 1 or 2, but got ranks ${n.rank} and ${r.rank}.`);let s=n.rank===1?n.size:n.shape[1],a=r.rank===1?r.size:r.shape[0];if(z(s===a,()=>`Error in dot: inner dimensions of inputs must match, but got ${s} and ${a}.`),n.rank===1&&r.rank===1){let o=Z(n,[1,-1]),i=Z(r,[-1,1]),l=rt(o,i);return Z(l,[])}else if(n.rank===1&&r.rank===2){let o=Z(n,[1,-1]),i=Z(r,[r.shape[0],r.shape[1]]),l=rt(o,i);return Z(l,[l.size])}else if(n.rank===2&&r.rank===1){let o=Z(r,[-1,1]),i=rt(n,o);return Z(i,[i.size])}else{let o=Z(r,[r.shape[0],r.shape[1]]);return rt(n,o)}}var Wwe=V({dot_:gq});function yq(e,...t){let n=t.map((s,a)=>O(s,`tensors${a}`,"einsum")),r={equation:e};return G.runKernel(R1,n,r)}var Aq=V({einsum_:yq});function xq(e){let n={x:O(e,"x","elu")};return G.runKernel(_c,n)}var If=V({elu_:xq});function bq(e){let t=O(e,"x","erf");z(t.dtype==="int32"||t.dtype==="float32",()=>"Input dtype must be `int32` or `float32`."),t.dtype==="int32"&&(t=xe(t,"float32"));let n={x:t};return G.runKernel(Dc,n)}var vq=V({erf_:bq});function wq(e){let n={x:O(e,"x","exp")};return G.runKernel(Eo,n)}var Gr=V({exp_:wq});function kq(e,t=0){let n=O(e,"x","expandDims","string_or_numeric");z(t<=n.rank,()=>"Axis must be <= rank of the tensor");let r={input:n},s={dim:t};return G.runKernel(Fc,r,s)}var Er=V({expandDims_:kq});function Iq(e){let n={x:O(e,"x","expm1")};return G.runKernel(ll,n)}var Sq=V({expm1_:Iq});function Tq(e,t){let n=O(e,"x","tile","string_or_numeric");z(n.rank===t.length,()=>`Error in transpose: rank of input ${n.rank} must match length of reps ${t}.`);let r={x:n},s={reps:t};return G.runKernel(Lo,r,s)}var $s=V({tile_:Tq});function Nq(e,t,n,r="float32"){t==null&&(t=e);let s=ze([e,t],r),a=e<=t?e:t;for(let i=0;i<a;++i)s.set(1,i,i);let o=Z(s.toTensor(),[e,t]);if(n==null)return o;if(n.length===1)return $s(Er(o,0),[n[0],1,1]);if(n.length===2)return $s(Er(Er(o,0),0),[n[0],n[1],1,1]);if(n.length===3)return $s(Er(Er(Er(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 ${n.length}D.`)}var X6=V({eye_:Nq});function Nd(e,t,n){let r={shape:e,value:t,dtype:n};return G.runKernel(Jp,{},r)}function Cq(e){let n={x:O(e,"x","floor")};return G.runKernel($o,n)}var Sf=V({floor_:Cq});function Eq(e,t,n=0,r=0){let s=O(e,"x","gather"),a=O(t,"indices","gather","int32"),o={x:s,indices:a},i={axis:n,batchDims:r};return G.runKernel(Oc,o,i)}var Tf=V({gather_:Eq});function $q(e,t){let n=O(e,"a","greater","string_or_numeric"),r=O(t,"b","greater","string_or_numeric");[n,r]=Vt(n,r),$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(dl,s)}var $r=V({greater_:$q});function Rq(e,t){let n=O(e,"a","greaterEqual","string_or_numeric"),r=O(t,"b","greaterEqual","string_or_numeric");[n,r]=Vt(n,r),$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(Ro,s)}var Kl=V({greaterEqual_:Rq});function _q(e){let n={input:O(e,"input","imag")};return G.runKernel(M1,n)}var Oy=V({imag_:_q});function Dq(e){let n={x:O(e,"x","isFinite")};return G.runKernel(zc,n)}var Vwe=V({isFinite_:Dq});function Fq(e){let n={x:O(e,"x","isInf")};return G.runKernel(Lc,n)}var Uwe=V({isInf_:Fq});function Mq(e){let n={x:O(e,"x","isNaN")};return G.runKernel(Bc,n)}var Oq=V({isNaN_:Mq});function Pq(e,t=.2){let r={x:O(e,"x","leakyRelu")},s={alpha:t};return G.runKernel(pl,r,s)}var Py=V({leakyRelu_:Pq});function zq(e,t){let n=O(e,"a","less","string_or_numeric"),r=O(t,"b","less","string_or_numeric");[n,r]=Vt(n,r),$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(fl,s)}var Z6=V({less_:zq});function Lq(e,t){let n=O(e,"a","lessEqual","string_or_numeric"),r=O(t,"b","lessEqual","string_or_numeric");[n,r]=Vt(n,r),$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(ml,s)}var Xl=V({lessEqual_:Lq});function Bq(e,t,n){if(n<=0)throw new Error("The number of values should be positive.");let r={start:e,stop:t,num:n};return G.runKernel(O1,{},r)}function Wq(e,t=5,n=1,r=1,s=.5){let a=O(e,"x","localResponseNormalization");z(a.rank===4||a.rank===3,()=>`Error in localResponseNormalization: x must be rank 3 or 4 but got
rank ${a.rank}.`),z(fn(t),()=>`Error in localResponseNormalization: depthRadius must be an integer but got depthRadius ${t}.`);let o=a,i=!1;a.rank===3&&(i=!0,o=Z(a,[1,a.shape[0],a.shape[1],a.shape[2]]));let l={x:o},u={depthRadius:t,bias:n,alpha:r,beta:s},c=G.runKernel(tf,l,u);return i?Z(c,[c.shape[1],c.shape[2],c.shape[3]]):c}var Vq=V({localResponseNormalization_:Wq});function Uq(e){let n={x:O(e,"x","log")};return G.runKernel(_o,n)}var jr=V({log_:Uq});function Hq(e){let n={x:O(e,"x","log1p")};return G.runKernel(Wc,n)}var zy=V({log1p_:Hq});function Gq(e,t){z(Hp(e),()=>"The f passed in variableGrads(f) must be a function"),z(t==null||Array.isArray(t)&&t.every(u=>u instanceof ff),()=>"The varList passed in variableGrads(f, varList) must be an array of variables");let n=t!=null;if(!n){t=[];for(let u in G.registeredVariables)t.push(G.registeredVariables[u])}let r=n?t.filter(u=>!u.trainable):null,s=t.length;t=t.filter(u=>u.trainable),z(t.length>0,()=>`variableGrads() expects at least one of the input variables to be trainable, but none of the ${s} variables is trainable.`);let a=!0,{value:o,grads:i}=G.gradients(e,t,null,a);z(i.some(u=>u!=null),()=>"Cannot find a connection between any variable and the result of the loss function y=f(x). Please make sure the operations that use variables are inside the function f passed to minimize()."),z(o.rank===0,()=>`The f passed in variableGrads(f) must return a scalar, but it returned a rank-${o.rank} tensor`);let l={};return t.forEach((u,c)=>{i[c]!=null&&(l[u.name]=i[c])}),r!=null&&r.forEach(u=>l[u.name]=null),{value:o,grads:l}}function la(e){return G.customGrad(e)}function jq(e){let n={x:O(e,"x","neg")};return G.runKernel(Hc,n)}var qt=V({neg_:jq});function qq(e){let n={x:O(e,"x","softplus")};return G.runKernel(ad,n)}var Cd=V({softplus_:qq});function Kq(e){let t=O(e,"x","logSigmoid");return la(r=>({value:qt(Cd(qt(r))),gradFunc:o=>j(o,ia(qt(r)))}))(t)}var Hwe=V({logSigmoid_:Kq});function Xq(e,t=null,n=!1){let s={x:O(e,"x","max")},a={reductionIndices:t,keepDims:n};return G.runKernel(gl,s,a)}var Rs=V({max_:Xq});function Zq(e,t){let n=O(e,"a","sub"),r=O(t,"b","sub");[n,r]=Vt(n,r);let s={a:n,b:r};return G.runKernel(zo,s)}var ke=V({sub_:Zq});function Yq(e,t=null,n=!1){let r=O(e,"x","sum");r.dtype==="bool"&&(r=xe(r,"int32"));let s={x:r},a={axis:t,keepDims:n};return G.runKernel(Fl,s,a)}var Te=V({sum_:Yq});function Jq(e,t=-1){let n=O(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 ${t}`);return la((s,a)=>{let o=!0,i=Rs(s,t,!0),l=ke(s,i),u=ke(xe(l,"float32"),jr(Te(Gr(l),t,o)));return a([u]),{value:u,gradFunc:(d,h)=>{let[p]=h,f=!0,m=Gr(p);return ke(d,j(Te(d,t,f),m))}}})(n)}var Y6=V({logSoftmax_:Jq});function Ly(e,t){for(let n=0;n<e.length;++n)if(e[e.length-n-1]!==t-1-n)return!1;return!0}function J6(e,t,n){let r=e.length+t.length,s=[],a=0,o=0;for(let i=0;i<r;i++)n.indexOf(i)===-1?s.push(e[a++]):s.push(t[o++]);return s}function Q6(e,t){let n=[],r=e.length;for(let a=0;a<r;a++)t.indexOf(a)===-1&&n.push(e[a]);let s=t.map(a=>e[a]);return[n,s]}function Zo(e,t){let n=t.map(r=>1);return J6(e,n,t)}function Qq(e,t,n){z(Ly(t,n),()=>`${e} supports only inner-most axes for now. Got axes ${t} and rank-${n} input.`)}function eI(e,t){if(Ly(e,t))return null;let n=[];for(let r=0;r<t;++r)e.indexOf(r)===-1&&n.push(r);return e.forEach(r=>n.push(r)),n}function By(e){return e.map((t,n)=>[n,t]).sort((t,n)=>t[1]-n[1]).map(t=>t[0])}function eK(e,t){let n=[];for(let r=t-e;r<t;++r)n.push(r);return n}function tK(e,t=null,n=!1){let r=O(e,"x","logSumExp"),s=Vr(t,r.shape),a=Rs(r,s,!0),o=ke(r,a),i=Gr(o),l=Te(i,s),u=jr(l),c=de(Z(a,u.shape),u);if(n){let d=Zo(c.shape,s);return Z(c,d)}return c}var nK=V({logSumExp_:tK});function rK(e,t){let n=O(e,"a","logicalAnd","bool"),r=O(t,"b","logicalAnd","bool");$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(Vc,s)}var _s=V({logicalAnd_:rK});function sK(e){let n={x:O(e,"x","logicalNot","bool")};return G.runKernel(Qp,n)}var Wy=V({logicalNot_:sK});function aK(e,t){let n=O(e,"a","logicalOr","bool"),r=O(t,"b","logicalOr","bool");$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(ef,s)}var tI=V({logicalOr_:aK});function oK(e,t){let n=O(e,"a","logicalXor","bool"),r=O(t,"b","logicalXor","bool");return $t(n.shape,r.shape),_s(tI(e,t),Wy(_s(e,t)))}var Gwe=V({logicalXor_:oK});function iK(e,t,n,r,s){let a=O(e,"x","maxPool"),o=1,i=a,l=!1;a.rank===3&&(l=!0,i=Z(a,[1,a.shape[0],a.shape[1],a.shape[2]])),z(i.rank===4,()=>`Error in maxPool: input must be rank 4 but got rank ${i.rank}.`),z(Es(n,o),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${n} and dilations '${o}'`),s!=null&&z(fn(r),()=>`Error in maxPool: pad must be an integer when using, dimRoundingMode ${s} but got pad ${r}.`);let u={x:i},c={filterSize:t,strides:n,pad:r,dimRoundingMode:s},d=G.runKernel(yl,u,c);return l?Z(d,[d.shape[1],d.shape[2],d.shape[3]]):d}var Vy=V({maxPool_:iK});function lK(e,t=[1,1,1],n,r,s,a="NDHWC"){let o=O(e,"x","maxPool3d"),i=o,l=!1;o.rank===4&&(l=!0,i=Z(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]])),z(i.rank===5,()=>`Error in maxPool3d: x must be rank 5 but got rank ${i.rank}.`),z(a==="NDHWC",()=>`Error in maxPool3d: Only NDHWC is currently supported, but got dataFormat of ${a}`),s!=null&&z(fn(r),()=>`Error in maxPool3d: pad must be an integer when using, dimRoundingMode ${s} but got pad ${r}.`);let u={x:i},c={filterSize:t,strides:n,pad:r,dimRoundingMode:s,dataFormat:a},d=G.runKernel(nf,u,c);return l?Z(d,[d.shape[1],d.shape[2],d.shape[3],d.shape[4]]):d}var nI=V({maxPool3d_:lK});function uK(e,t,n,r,s=!1){let o={x:O(e,"x","maxPoolWithArgmax")},i={filterSize:t,strides:n,pad:r,includeBatchInIndex:s},l=G.runKernel(B1,o,i);return{result:l[0],indexes:l[1]}}var cK=V({maxPoolWithArgmax_:uK});function dK(e,t){let n=O(e,"a","maximum"),r=O(t,"b","maximum");[n,r]=Vt(n,r),n.dtype==="bool"&&(n=xe(n,"int32"),r=xe(r,"int32")),$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(Do,s)}var Va=V({maximum_:dK});function hK(e,t=null,n=!1){let s={x:O(e,"x","mean")},a={axis:t,keepDims:n};return G.runKernel(Al,s,a)}var Xt=V({mean_:hK});function ln(e,t="float32"){if(t==="complex64"){let r=ln(e,"float32"),s=ln(e,"float32");return Uo(r,s)}let n=jp(sn(e),t);return G.makeTensor(n,e,t)}function ua(e,t="float32"){if(t==="complex64"){let r=ua(e,"float32"),s=ln(e,"float32");return Uo(r,s)}let n=p1(sn(e),t);return G.makeTensor(n,e,t)}function pK(e,t=null,n=!1){let s={x:O(e,"x","min")},a={axis:t,keepDims:n};return G.runKernel(xl,s,a)}var Uy=V({min_:pK});function fK(e,t){let n=O(e,"a","minimum"),r=O(t,"b","minimum");[n,r]=Vt(n,r),n.dtype==="bool"&&(n=xe(n,"int32"),r=xe(r,"int32")),$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(Fo,s)}var Nf=V({minimum_:fK});function mK(e,t,n){z(n==="reflect"||n==="symmetric",()=>`Invalid mode. Mode must be either reflect or symmetric. Got ${n}.`);let r=O(e,"x","mirrorPad");if(r.rank===0)throw new Error("mirrorPad(scalar) is not defined. Pass non-scalar to mirrorPad");z(t.length===r.rank,()=>`Padding doesn't match input. Must be ${r.rank}. Got ${t.length}.`);let s=n==="reflect"?1:0;for(let i=0;i<r.rank;i++)z(t[i].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),z(t[i][0]>=0&&t[i][0]<=r.shape[i]-s&&t[i][1]>=0&&t[i][1]<=r.shape[i]-s,()=>`Padding in dimension ${i} cannot be greater than or equal to ${r.shape[i]-s} or less than 0 for input of shape ${r.shape}`);let a={paddings:t,mode:n},o={x:r};return G.runKernel(bl,o,a)}var gK=V({mirrorPad_:mK});function yK(e,t){let n=O(e,"a","mod"),r=O(t,"b","mod");[n,r]=Vt(n,r);let s={a:n,b:r};return G.runKernel(Uc,s)}var AK=V({mod_:yK});function xK(e){let t=O(e,"x","square"),n={};return G.runKernel("Square",{x:t},n)}var Tt=V({square_:xK});function bK(e,t=null,n=!1){e=O(e,"x","moments");let r=Vr(t,e.shape),s=Xt(e,r,n),a=s.shape;n||(a=Zo(s.shape,r));let o=Tt(ke(xe(e,"float32"),Z(s,a))),i=Xt(o,r,n);return{mean:s,variance:i}}var Hy=V({moments_:bK});function vK(e,t,n,r){let s=O(t,"data","multiRNNCell"),a=gf(n,"c","multiRNNCell"),o=gf(r,"h","multiRNNCell"),i=s,l=[];for(let d=0;d<e.length;d++){let h=e[d](i,a[d],o[d]);l.push(h[0]),l.push(h[1]),i=h[1]}let u=[],c=[];for(let d=0;d<l.length;d+=2)u.push(l[d]),c.push(l[d+1]);return[u,c]}var jwe=V({multiRNNCell_:vK});function wK(e,t,n,r=!1){let s=O(e,"logits","multinomial"),a=s.size,o=s.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 ${o}`);n=n||Math.random();let l={logits:o===1?Z(s,[1,-1]):s},u={numSamples:t,seed:n,normalized:r},c=G.runKernel(W1,l,u);return o===1?Z(c,[c.size]):c}var kK=V({multinomial_:wK});function IK(e,t){let n=O(e,"a","notEqual","string_or_numeric"),r=O(t,"b","notEqual","string_or_numeric");[n,r]=Vt(n,r),$t(n.shape,r.shape);let s={a:n,b:r};return G.runKernel(vl,s)}var Ed=V({notEqual_:IK});function SK(e){let n={x:O(e,"x","onesLike")};return G.runKernel(Kc,n)}var qr=V({onesLike_:SK});function TK(e,t){let n=O(e,"v1","outerProduct"),r=O(t,"v2","outerProduct");z(n.rank===1&&r.rank===1,()=>`Error in outerProduct: inputs must be rank 1, but got ranks ${n.rank} and ${r.rank}.`);let s=Z(n,[-1,1]),a=Z(r,[1,-1]);return rt(s,a)}var qwe=V({outerProduct_:TK});function NK(e,t,n=0){let r=O(e,"x","pad");if(r.rank===0)throw new Error("pad(scalar) is not defined. Pass non-scalar to pad");let s={paddings:t,constantValue:n},a={x:r};return G.runKernel(kl,a,s)}var Yo=V({pad_:NK});function CK(e,t,n=0){return z(t.length===2,()=>"Invalid number of paddings. Must be length of 2."),Yo(e,[t],n)}var Kwe=V({pad1d_:CK});function EK(e,t,n=0){return z(t.length===2&&t[0].length===2&&t[1].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),Yo(e,t,n)}var Xwe=V({pad2d_:EK});function $K(e,t,n=0){return z(t.length===3&&t[0].length===2&&t[1].length===2&&t[2].length===2,()=>"Invalid number of paddings. Must be length of 2 each."),Yo(e,t,n)}var Zwe=V({pad3d_:$K});function RK(e,t,n=0){return z(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."),Yo(e,t,n)}var Ywe=V({pad4d_:RK});function _K(e,t,n){let r=O(e,"x","spaceToBatchND");z(r.rank>=1+t.length,()=>`input rank ${r.rank} should be > than [blockShape] ${t.length}`),z(n.length===t.length,()=>`paddings.shape[0] ${n.length} must be equal to [blockShape] ${t.length}`),z(r.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 ${r.shape.slice(1)} with paddings ${n.toString()} must be divisible by blockShapes ${t.toString()}`);let s={x:r},a={blockShape:t,paddings:n};return G.runKernel(od,s,a)}var Gy=V({spaceToBatchND_:_K});function DK(e,t,n,r,s,a){s==null&&(s=[1,1]),a==null&&(a=1),r===0&&(r="valid");let o=O(e,"x","maxPool"),i=o,l=!1;o.rank===3&&(l=!0,i=Z(o,[1,o.shape[0],o.shape[1],o.shape[2]])),z(Es(a,s),()=>`Error in pool: Either strides or dilations must be 1. Got strides ${a} and dilations '${s}'`);let u=z6(i.shape,t,a,s,r),c=[u.dilationHeight,u.dilationWidth],d;r==="same"?d=MK([u.filterHeight,u.filterWidth],c):d=[[0,0],[0,0]];let h=c[0]===1&&c[1]===1,[p,f]=FK([u.inHeight,u.inWidth],c,d),m=h?r:"valid",g=h?i:Gy(i,c,p),A=(n==="avg"?()=>_y(g,t,a,m):()=>Vy(g,t,a,m))(),x=h?A:Dy(A,c,f);return l?Z(x,[x.shape[1],x.shape[2],x.shape[3]]):x}function FK(e,t,n){let r=n.map(c=>c[0]),s=n.map(c=>c[1]),a=e.concat(r,s),o=t.map((c,d)=>(c-a[d]%c)%c),i=s.map((c,d)=>c+o[d]),l=t.map((c,d)=>[r[d],i[d]]),u=t.map((c,d)=>[0,o[d]]);return[l,u]}function MK(e,t){let r=e.map((o,i)=>o+(o-1)*(t[i]-1)).map(o=>o-1),s=r.map(o=>Math.floor(o/2)),a=r.map((o,i)=>o-s[i]);return r.map((o,i)=>[s[i],a[i]])}var Jwe=V({pool_:DK});function OK(e,t){let n=O(e,"base","pow"),r=O(t,"exp","pow");[n,r]=Vt(n,r);let s={a:n,b:r};return G.runKernel(Il,s)}var Jo=V({pow_:OK});function PK(e,t){let n=O(e,"x","prelu"),r=O(t,"alpha","prelu"),s={x:n,alpha:r};return G.runKernel(Sl,s)}var jy=V({prelu_:PK});function zK(e,t=null,n=!1){let r=O(e,"x","prod");r.dtype==="bool"&&(r=xe(r,"int32"));let s={x:r},a={axis:t,keepDims:n};return G.runKernel(Zc,s,a)}var rI=V({prod_:zK});function LK(e,t,n){let r=sn(e),s=null;if(n==null||n==="float32")s=new Float32Array(r);else if(n==="int32")s=new Int32Array(r);else if(n==="bool")s=new Uint8Array(r);else throw new Error(`Unknown data type ${n}`);for(let a=0;a<r;a++)s[a]=t();return G.makeTensor(s,e,n)}var Qwe=V({rand_:LK}),qy=Xs(Zg()),Ky=class{constructor(e,t,n,r,s){this.mean=e,this.stdDev=t,this.dtype=n,this.nextVal=NaN,this.truncated=r,this.truncated&&(this.upper=this.mean+this.stdDev*2,this.lower=this.mean-this.stdDev*2);let a=s||Math.random();this.random=qy.alea(a.toString())}nextValue(){if(!isNaN(this.nextVal)){let r=this.nextVal;return this.nextVal=NaN,r}let e,t,n=!1;for(;!n;){let r,s,a;do r=2*this.random()-1,s=2*this.random()-1,a=r*r+s*s;while(a>=1||a===0);let o=Math.sqrt(-2*Math.log(a)/a);e=this.mean+this.stdDev*r*o,t=this.mean+this.stdDev*s*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}},BK=class{constructor(e,t,n,r){this.alpha=e,this.beta=1/t,this.dtype=n;let s=r||Math.random();this.randu=qy.alea(s.toString()),this.randn=new Ky(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,r,s,a;for(;;){do r=this.randn.nextValue(),a=1+this.c*r;while(a<=0);if(a*=a*a,e=r*r,t=1-.331*e*e,n=.5*e+this.d*(1-a+Math.log(a)),s=this.randu(),s<t||Math.log(s)<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)}},WK=class{constructor(e=0,t=1,n,r){if(this.canReturnFloat=()=>this.dtype==null||this.dtype==="float32",this.min=e,this.range=t-e,this.dtype=n,r==null&&(r=Math.random()),typeof r=="number"&&(r=r.toString()),!this.canReturnFloat()&&this.range<=1)throw new Error(`The difference between ${e} - ${t} <= 1 and dtype is not float`);this.random=qy.alea(r)}convertValue(e){return this.canReturnFloat()?e:Math.round(e)}nextValue(){return this.convertValue(this.min+this.range*this.random())}};function VK(e,t,n=1,r="float32",s){if(n==null&&(n=1),r==null&&(r="float32"),r!=="float32"&&r!=="int32")throw new Error(`Unsupported data type ${r}`);let a=new BK(t,n,r,s),o=ze(e,r);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var e7e=V({randomGamma_:VK});function UK(e,t=0,n=1,r,s){if(r!=null&&r==="bool")throw new Error(`Unsupported data type ${r}`);let a=new Ky(t,n,r,!1,s),o=ze(e,r);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var HK=V({randomNormal_:UK});function GK(e,t=0,n=1,r="float32",s){let a=ze(e,r),o=new WK(t,n,null,s);for(let i=0;i<a.values.length;i++)a.values[i]=o.nextValue();return a.toTensor()}var $d=V({randomUniform_:GK});function Rd(e,t,n=1,r="float32"){if(n===0)throw new Error("Cannot have a step of zero");let s={start:e,stop:t,step:n,dtype:r};return G.runKernel(rf,{},s)}function jK(e){let n={input:O(e,"input","real")};return G.runKernel(V1,n)}var Cf=V({real_:jK});function qK(e){let n={x:O(e,"x","reciprocal")};return G.runKernel(Yc,n)}var KK=V({reciprocal_:qK});function XK(e){let n={x:O(e,"x","relu")};return G.runKernel(Tl,n)}var ca=V({relu_:XK});function ZK(e){let n={x:O(e,"x","relu6")};return G.runKernel(Cl,n)}var sI=V({relu6_:ZK});function YK(e,t){let r={x:O(e,"x","reverse")},s={dims:t};return G.runKernel(El,r,s)}var Kr=V({reverse_:YK});function JK(e){let t=O(e,"x","reverse");return z(t.rank===1,()=>`Error in reverse1D: x must be rank 1 but got rank ${t.rank}.`),Kr(t,0)}var t7e=V({reverse1d_:JK});function QK(e,t){let n=O(e,"x","reverse");return z(n.rank===2,()=>`Error in reverse2D: x must be rank 2 but got rank ${n.rank}.`),Kr(n,t)}var n7e=V({reverse2d_:QK});function eX(e,t){let n=O(e,"x","reverse");return z(n.rank===3,()=>`Error in reverse3D: x must be rank 3 but got rank ${n.rank}.`),Kr(n,t)}var r7e=V({reverse3d_:eX});function tX(e,t){let n=O(e,"x","reverse");return z(n.rank===4,()=>`Error in reverse4D: x must be rank 4 but got rank ${n.rank}.`),Kr(n,t)}var s7e=V({reverse4d_:tX});function nX(e){let n={x:O(e,"x","round")};return G.runKernel($l,n)}var aI=V({round_:nX});function rX(e){let n={x:O(e,"x","rsqrt")};return G.runKernel(Oo,n)}var oI=V({rsqrt_:rX});function De(e,t){if((as(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"&&as(e)&&!(e instanceof Uint8Array))throw new Error("When making a scalar from encoded string, the value must be `Uint8Array`.");return vd(e,[],[],t)}function sX(e){let n={x:O(e,"x","selu")};return G.runKernel(td,n)}var iI=V({selu_:sX});function aX(e,t,n,r,s,a=[1,1],o="NHWC"){let i=O(e,"x","separableConv2d"),l=O(t,"depthwiseFilter","separableConv2d"),u=O(n,"pointwiseFilter","separableConv2d"),c=i,d=!1;if(i.rank===3&&(d=!0,c=Z(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");z(c.rank===4,()=>`Error in separableConv2d: input must be rank 4, but got rank ${c.rank}.`),z(l.rank===4,()=>`Error in separableConv2d: depthwise filter must be rank 4, but got rank ${l.rank}.`),z(u.rank===4,()=>`Error in separableConv2d: pointwise filter must be rank 4, but got rank ${l.rank}.`),z(u.shape[0]===1,()=>`Error in separableConv2d: the first dimension of pointwise filter must be 1, but got ${u.shape[0]}.`),z(u.shape[1]===1,()=>`Error in separableConv2d: the second dimension of pointwise filter must be 1, but got ${u.shape[1]}.`);let h=l.shape[2],p=l.shape[3];z(u.shape[2]===h*p,()=>`Error in separableConv2d: the third dimension of pointwise filter must be ${h*p}, but got ${u.shape[2]}.`);let f=kf(c,l,r,s,o,a),g=Xo(f,u,1,"valid",o);return d?Z(g,[g.shape[1],g.shape[2],g.shape[3]]):g}var oX=V({separableConv2d_:aX});async function iX(e,t){let n=O(e,"x","setdiff1d"),r=O(t,"y","setdiff1d");z(n.dtype===r.dtype,()=>`x and y should have the same dtype, but got x (${n.dtype}) and y (${r.dtype}).`),z(n.rank===1,()=>`x should be 1D tensor, but got x (${n.shape}).`),z(r.rank===1,()=>`y should be 1D tensor, but got y (${r.shape}).`);let s=await n.data(),a=await r.data(),o=new Set(a),i=0;for(let c=0;c<s.length;c++)o.has(s[c])||i++;let l=new Jt([i],n.dtype),u=new Jt([i],"int32");for(let c=0,d=0;c<s.length;c++)o.has(s[c])||(l.values[d]=s[c],u.values[d]=c,d++);return[l.toTensor(),u.toTensor()]}var lX=iX;function uX(e){let n={x:O(e,"x","sign")};return G.runKernel(sd,n)}var cX=V({sign_:uX});function dX(e){let n={x:O(e,"x","sin")};return G.runKernel(Rl,n)}var lI=V({sin_:dX});function hX(e){let n={x:O(e,"x","sinh")};return G.runKernel(rd,n)}var uI=V({sinh_:hX});function pX(e,t,n){let r=O(e,"x","slice1d");return z(r.rank===1,()=>`slice1d expects a rank-1 tensor, but got a rank-${r.rank} tensor`),at(r,[t],[n])}var Xy=V({slice1d_:pX});function fX(e,t,n){let r=O(e,"x","slice2d");return z(r.rank===2,()=>`slice2d expects a rank-2 tensor, but got a rank-${r.rank} tensor`),at(r,t,n)}var cI=V({slice2d_:fX});function mX(e,t,n){let r=O(e,"x","slice3d");return z(r.rank===3,()=>`slice3d expects a rank-3 tensor, but got a rank-${r.rank} tensor`),at(r,t,n)}var Zy=V({slice3d_:mX});function gX(e,t,n){let r=O(e,"x","slice4d");return z(r.rank===4,()=>`slice4d expects a rank-4 tensor, but got a rank-${r.rank} tensor`),at(r,t,n)}var Ef=V({slice4d_:gX});function yX(e,t=-1){let n=O(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 r={logits:n},s={dim:t};return G.runKernel(Ml,r,s)}var Yy=V({softmax_:yX});function AX(e){z(e.dtype==="complex64",()=>`The dtype for tf.spectral.fft() must be complex64 but got ${e.dtype}.`);let t={input:e};return G.runKernel(D1,t)}var dI=V({fft_:AX});function xX(e){z(e.dtype==="complex64",()=>`The dtype for tf.spectral.ifft() must be complex64 but got ${e.dtype}.`);let t={input:e};return G.runKernel(F1,t)}var Jy=V({ifft_:xX});function bX(e){let t=e.shape[e.shape.length-1],n=e.size/t,r;if(t<=2){let s=Z(e,[n,t]);r=Jy(s)}else{let s=[n,2*(t-1)],a=Z(Cf(e),[n,t]),o=Z(Oy(e),[n,t]),i=Kr(at(a,[0,1],[n,t-2]),1),l=j(Kr(at(o,[0,1],[n,t-2]),1),De(-1)),u=an([a,i],1),c=an([o,l],1),d=Z(Uo(u,c),[s[0],s[1]]);r=Jy(d)}if(r=Cf(r),e.rank===3&&e.shape[0]!==0){let s=r,a=e.shape[0];r=Z(r,[a,r.shape[0]/a,r.shape[1]]),s.dispose()}return r}var vX=V({irfft_:bX});function wX(e,t,n=0){let s={x:O(e,"x","split")},a={numOrSizeSplits:t,axis:n};return G.runKernel(id,s,a)}var Rr=V({split_:wX});function kX(e,t){z(e.dtype==="float32",()=>`The dtype for rfft() must be real value but got ${e.dtype}`);let n=e.shape[e.shape.length-1],r=e.size/n,s;if(t!=null&&t<n){let f=e.shape.map(g=>0),m=e.shape.map(g=>g);m[e.shape.length-1]=t,s=at(e,f,m),n=t}else if(t!=null&&t>n){let f=e.shape.map(m=>m);f[e.shape.length-1]=t-n,s=an([e,ln(f)],e.shape.length-1),n=t}else s=e;let a=ot(s),o=Z(Uo(s,a),[r,n]),i=dI(o),l=Math.floor(n/2)+1,u=Cf(i),c=Oy(i),d=Rr(u,[l,n-l],u.shape.length-1),h=Rr(c,[l,n-l],c.shape.length-1),p=s.shape.slice();return p[s.shape.length-1]=l,Z(Uo(d[0],h[0]),p)}var hI=V({rfft_:kX});function IX(e){let n={x:O(e,"x","sqrt")};return G.runKernel(Dl,n)}var Ln=V({sqrt_:IX});function SX(e,t){let n=O(e,"a","squaredDifference"),r=O(t,"b","squaredDifference");[n,r]=Vt(n,r),$t(n.shape,r.shape);let s={a:n,b:r},a={};return G.runKernel(Po,s,a)}var pI=V({squaredDifference_:SX});function TX(e,t){let n=O(e,"x","squeeze");return Z(n,N4(n.shape,t).newShape)}var Qo=V({squeeze_:TX});function NX(e,t=0){let n=gf(e,"tensors","stack","string_or_numeric");z(n.length>=1,()=>"Pass at least one tensor to tf.stack"),n.length>0&&z(t<=n[0].rank,()=>"Axis must be <= rank of the tensor");let r=n,s={axis:t};return G.runKernel(Xc,r,s)}var Xr=V({stack_:NX});function CX(e,t=0){let r={x:O(e,"x","step")},s={alpha:t};return G.runKernel(Bo,r,s)}var $f=V({step_:CX});function EX(e,t,n,r,s=0,a=0,o=0,i=0,l=0){let c={x:O(e,"x","stridedSlice","string_or_numeric")},d={begin:t,end:n,strides:r,beginMask:s,endMask:a,ellipsisMask:o,newAxisMask:i,shrinkAxisMask:l};return G.runKernel(ld,c,d)}var $X=V({stridedSlice_:EX});function RX(e){let n={x:O(e,"x","tan")};return G.runKernel(Ol,n)}var _X=V({tan_:RX});function En(e,t){Wp(e);let n=bd(e,t);if(n.length!==1)throw new Error("tensor1d() requires values to be a flat/TypedArray");return vd(e,null,n,t)}function Zl(e,t,n){if(Wp(e),t!=null&&t.length!==2)throw new Error("tensor2d() requires shape to have two numbers");let r=bd(e,n);if(r.length!==2&&r.length!==1)throw new Error("tensor2d() requires values to be number[][] or flat/TypedArray");if(r.length===1&&t==null)throw new Error("tensor2d() requires shape to be provided when `values` are a flat/TypedArray");return vd(e,t,r,n)}function DX(e,t=1,n=!0){let r=O(e,"x","topk");if(r.rank===0)throw new Error("topk() expects the input to be of rank 1 or higher");let s=r.shape[r.shape.length-1];if(t<0)throw new Error(`'k' passed to topk() must be >= 0 but got ${t}`);if(t>s)throw new Error(`'k' passed to topk() must be <= the last dimension (${s}) but got ${t}`);let a={x:r},o={k:t,sorted:n},[i,l]=G.runKernel(ud,a,o);return{values:i,indices:l}}var FX=V({topk_:DX});function MX(e,t=0,n=1,r,s){if(r!=null&&r==="bool")throw new Error("Unsupported data type $ { dtype }");let a=new Ky(t,n,r,!0,s),o=ze(e,r);for(let i=0;i<o.values.length;i++)o.values[i]=a.nextValue();return o.toTensor()}var Qy=V({truncatedNormal_:MX});function OX(e,t=0){let n=O(e,"x","unique","string_or_numeric");z(n.rank>0,()=>"The input tensor must be at least 1D");let r={x:n},s={axis:t},[a,o]=G.runKernel(Q1,r,s);return{values:a,indices:o}}var fI=V({unique_:OX});function PX(e,t,n){let r=O(e,"x","unsortedSegmentSum"),s=O(t,"segmentIds","unsortedSegmentSum","int32");z(fn(n),()=>"numSegments must be of dtype int");let a={x:r,segmentIds:s},o={numSegments:n};return G.runKernel(of,a,o)}var zX=V({unsortedSegmentSum_:PX});function LX(e,t=0){let n=O(e,"x","unstack","string_or_numeric");z(t>=-n.shape.length&&t<n.shape.length,()=>`Axis = ${t} is not in [-${n.shape.length}, ${n.shape.length})`);let r={value:n},s={axis:t};return G.runKernel(dd,r,s)}var Ds=V({unstack_:LX});function BX(e,t=!0,n,r){return G.makeVariable(e,t,n,r)}function mI(e,t){let n=[];for(let a=0;a<t.length;a++)t[a]&&n.push(a);let r=ze(e,"int32"),s=ze([n.length,e.length],"int32");for(let a=0;a<n.length;a++){let o=r.indexToLoc(n[a]),i=a*e.length;s.values.set(o,i)}return s.toTensor()}async function WX(e){let t=O(e,"condition","whereAsync","bool"),n=await t.data(),r=mI(t.shape,n);return e!==t&&t.dispose(),r}var VX=WX;function UX(e,t="euclidean",n=null,r=!1){e=O(e,"x","norm");let s=gI(e,t,n),a=s.shape;if(r){let o=Vr(n,e.shape);a=Zo(s.shape,o)}return Z(s,a)}function gI(e,t,n=null){if(e.rank===0)return gn(e);if(e.rank!==1&&n===null)return gI(Z(e,[-1]),t,n);if(e.rank===1||typeof n=="number"||Array.isArray(n)&&n.length===1){if(t===1)return Te(gn(e),n);if(t===1/0)return Rs(gn(e),n);if(t===-1/0)return Uy(gn(e),n);if(t==="euclidean"||t===2)return Ln(Te(Jo(gn(e),De(2,"int32")),n));throw new Error(`Error in norm: invalid ord value: ${t}`)}if(Array.isArray(n)&&n.length===2){if(t===1)return Rs(Te(gn(e),n[0]),n[1]-1);if(t===1/0)return Rs(Te(gn(e),n[1]),n[0]);if(t===-1/0)return Uy(Te(gn(e),n[1]),n[0]);if(t==="fro"||t==="euclidean")return Ln(Te(Tt(e),n));throw new Error(`Error in norm: invalid ord value: ${t}`)}throw new Error(`Error in norm: invalid axis: ${n}`)}var yI=V({norm_:UX});function HX(e,t,n,r,s=!0){let a=O(e,"v","movingAverage"),o=O(t,"x","movingAverage"),i=O(n,"decay","movingAverage");K4(a,o),z(Fa(a.shape,o.shape),()=>"Shape mismatch in v and x");let l=De(1),u=ke(l,i),c=j(ke(o,a),u);if(s){z(r!=null,()=>"When using zeroDebias: true, step is required.");let d=O(r,"step","movingAverage");c=Re(c,ke(l,Jo(i,d)))}return de(a,c)}var a7e=V({movingAverage_:HX});function GX(e,t,n){let r=O(e,"indices","scatterND","int32"),s=O(t,"updates","scatterND");Ty(s,r,n);let a={indices:r,updates:s},o={shape:n};return G.runKernel(Qc,a,o)}var jX=V({scatterND_:GX});function qX(e,t,n,r){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 s=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===s))throw new Error(`sparseValues has incorrect shape ${t.shape}, should be [] or [${s}]`);if(t.dtype!==r.dtype)throw new Error("sparseValues.dtype must match defaultValues.dtype")}function KX(e,t,n,r=0){let s=O(e,"sparseIndices","sparseToDense","int32"),a=O(t,"sparseValues","sparseToDense"),o=O(r,"defaultValue","sparseToDense",a.dtype);qX(s,a,n,o);let i={sparseIndices:s,sparseValues:a,defaultValue:o},l={outputShape:n};return G.runKernel(X1,i,l)}var AI=V({sparseToDense_:KX});function XX(e,t){let n=O(t,"indices","gatherND","int32"),s={params:O(e,"x","gatherND","string_or_numeric"),indices:n};return G.runKernel(Pc,s)}var ZX=V({gatherND_:XX});function YX(e,t){if(t==null)return e.shape.slice();if(Fa(e.shape,t))return t;if(e.shape.length===t.length){let n=[];for(let r=0;r<e.shape.length;r++)t[r]==null&&e.shape[r]!=null?n.push(e.shape[r]):n.push(t[r]);return n}return t}function JX(e,t,n,r){let s=O(e,"x","dropout");if(z(s.dtype==="float32",()=>`x has to be a floating point tensor since it's going to be scaled, but got a ${s.dtype} tensor instead.`),z(t>=0&&t<1,()=>`rate must be a float in the range [0, 1), but got ${t}.`),t===0)return e instanceof Ot?s.clone():s;let a=YX(s,n),o=1-t,i=Re(Sf(de($d(a,0,1,"float32",r),o)),o);return j(s,i)}var QX=V({dropout_:JX});function eZ(e){return Math.floor(Math.pow(2,Math.ceil(Math.log(e)/Math.log(2))))}function xI(e,t,n){let r=1-e%2,s=new Float32Array(e);for(let a=0;a<e;++a){let o=2*Math.PI*a/(e+r-1);s[a]=t-n*Math.cos(o)}return En(s,"float32")}var ei={};_e(ei,{conv2d:()=>rZ,depthwiseConv2d:()=>iZ,matMul:()=>uZ});function tZ(e,t,n,r,s,a="NHWC",o){let i=e;e.rank===3&&(i=Z(e,[1,e.shape[0],e.shape[1],e.shape[2]]));let l=t;l.rank===3&&(l=Z(t,[1,t.shape[0],t.shape[1],t.shape[2]])),z(i.rank===4,()=>`Error in conv2dDerFilter: input must be rank 4, but got shape ${i.shape}.`),z(l.rank===4,()=>`Error in conv2dDerFilter: dy must be rank 4, but got shape ${l.shape}.`),z(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];z(u===n[2],()=>`Error in conv2dDerFilter: depth of input ${u}) must match input depth in filter (${n[2]}.`),z(c===n[3],()=>`Error in conv2dDerFilter: depth of dy (${c}) must match output depth for filter (${n[3]}).`),o!=null&&z(fn(s),()=>`Error in conv2dDerFilter: pad must be an integer when using, dimRoundingMode ${o} but got pad ${s}.`);let d={x:i,dy:l},h={strides:r,pad:s,dataFormat:a,dimRoundingMode:o,filterShape:n};return G.runKernel(w1,d,h)}var eA=V({conv2DBackpropFilter_:tZ});function Rf(e,t,n){if(n==null||n==="linear")return e;if(n==="relu")return j(e,$f(t));throw new Error(`Cannot compute gradient for fused activation ${n}.`)}function _f(e,t){let n=t,r=on(e.shape,t.shape);return r.length>0&&(n=Te(n,r)),Z(n,e.shape)}function Df(e,t,n,r){if(t==="linear")return e;if(t==="relu")return ca(e);if(t==="elu")return If(e);if(t==="relu6")return sI(e);if(t==="prelu")return jy(e,n);if(t==="leakyrelu")return Py(e,r);if(t==="sigmoid")return ia(e);throw new Error(`Unknown fused activation ${t}.`)}var Ff=(e,t)=>!(e>0)||t==="linear";function nZ({x:e,filter:t,strides:n,pad:r,dataFormat:s="NHWC",dilations:a=[1,1],dimRoundingMode:o,bias:i,activation:l="linear",preluActivationWeights:u,leakyreluAlpha:c}){if(l=l||"linear",Ff(G.state.gradientDepth,l)===!1){let v=Xo(e,t,n,r,s,a,o);return i!=null&&(v=de(v,i)),Df(v,l,u,c)}let d=O(e,"x","conv2d"),h=O(t,"filter","conv2d"),p=d,f=!1;d.rank===3&&(f=!0,p=Z(d,[1,d.shape[0],d.shape[1],d.shape[2]])),z(p.rank===4,()=>`Error in fused conv2d: input must be rank 4, but got rank ${p.rank}.`),z(h.rank===4,()=>`Error in fused conv2d: filter must be rank 4, but got rank ${h.rank}.`),o!=null&&z(fn(r),()=>`Error in fused conv2d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${r}.`),z(p.shape[3]===h.shape[2],()=>`Error in conv2d: depth of input (${p.shape[3]}) must match input depth for filter ${h.shape[2]}.`),z(Es(n,a),()=>`Error in conv2D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`),z(s==="NHWC",()=>`Error in conv2d: got dataFormat of ${s} but only NHWC is currently supported.`);let m=Id(p.shape,h.shape,n,a,r,o),g;i!=null&&(g=O(i,"bias","fused conv2d"),[g]=Vt(g,d),$t(m.outShape,g.shape));let y;u!=null&&(y=O(u,"prelu weights","fused conv2d"));let A=(v,I)=>{let[w,S,E,D]=I,$=Rf(v,E,l);z(Wa(a),()=>`Error in gradient of fused conv2D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${a}'`);let R=Fy(S.shape,$,w,n,r),N=eA(S,$,w.shape,n,r),M=[R,N];if(D!=null){let B=_f(D,$);M.push(B)}return M},x={x:p,filter:h,bias:g,preluActivationWeights:y},b={strides:n,pad:r,dataFormat:s,dilations:a,dimRoundingMode:o,activation:l,leakyreluAlpha:c};return i==null?la((I,w,S)=>{let E=G.runKernel(Bl,x,b);return S([w,I,E]),f&&(E=Z(E,[E.shape[1],E.shape[2],E.shape[3]])),{value:E,gradFunc:A}})(p,h):la((I,w,S,E)=>{let D=G.runKernel(Bl,x,b);return E([w,I,D,S]),f&&(D=Z(D,[D.shape[1],D.shape[2],D.shape[3]])),{value:D,gradFunc:A}})(p,h,g)}var rZ=V({fusedConv2d_:nZ});function sZ(e,t,n,r,s,a=[1,1],o){let i=e;e.rank===3&&(i=Z(e,[1,e.shape[0],e.shape[1],e.shape[2]]));let l=t;l.rank===3&&(l=Z(t,[1,t.shape[0],t.shape[1],t.shape[2]]));let u={x:i,dy:l},c={strides:r,pad:s,dimRoundingMode:o,dilations:a,filterShape:n};return G.runKernel(T1,u,c)}var bI=V({depthwiseConv2dNativeBackpropFilter_:sZ});function aZ(e,t,n,r,s,a=[1,1],o){let i=t,l=!1;t.rank===3&&(l=!0,i=Z(t,[1,t.shape[0],t.shape[1],t.shape[2]]));let u={dy:i,filter:n},c={strides:r,pad:s,dimRoundingMode:o,dilations:a,inputShape:e},d=G.runKernel(N1,u,c);return l?Z(d,[d.shape[1],d.shape[2],d.shape[3]]):d}var vI=V({depthwiseConv2dNativeBackpropInput_:aZ});function oZ({x:e,filter:t,strides:n,pad:r,dataFormat:s="NHWC",dilations:a=[1,1],dimRoundingMode:o,bias:i,activation:l="linear",preluActivationWeights:u,leakyreluAlpha:c}){if(Ff(G.state.gradientDepth,l)===!1){let v=kf(e,t,n,r,s,a,o);return i!=null&&(v=de(v,i)),Df(v,l,u,c)}let d=O(e,"x","depthwiseConv2d"),h=O(t,"filter","depthwiseConv2d"),p=d,f=!1;d.rank===3&&(f=!0,p=Z(d,[1,d.shape[0],d.shape[1],d.shape[2]])),z(p.rank===4,()=>`Error in fused depthwiseConv2d: input must be rank 4, but got rank ${p.rank}.`),z(h.rank===4,()=>`Error in fused depthwiseConv2d: filter must be rank 4, but got rank ${h.rank}.`),z(p.shape[3]===h.shape[2],()=>`Error in fused depthwiseConv2d: number of input channels (${p.shape[3]}) must match the inChannels dimension in filter ${h.shape[2]}.`),a==null&&(a=[1,1]),z(Es(n,a),()=>`Error in fused depthwiseConv2d: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`),o!=null&&z(fn(r),()=>`Error in fused depthwiseConv2d: pad must be an integer when using dimRoundingMode ${o} but got pad ${r}.`);let m=Id(p.shape,h.shape,n,a,r,o,!0),g;i!=null&&(g=O(i,"bias","fused conv2d"),[g]=Vt(g,d),$t(m.outShape,g.shape));let y;u!=null&&(y=O(u,"prelu weights","fused depthwiseConv2d"));let A=(v,I)=>{z(Wa(a),()=>`Error in gradient of fused depthwiseConv2d: dilation rates greater than 1 are not yet supported. Got dilations '${a}'`);let[w,S,E,D]=I,$=Rf(v,E,l),R=vI(S.shape,$,w,n,r,a,o),N=bI(S,$,w.shape,n,r,a,o);if(D!=null){let M=_f(g,$);return[R,N,M]}return[R,N]},x={x:p,filter:h,bias:g,preluActivationWeights:y},b={strides:n,pad:r,dataFormat:s,dilations:a,dimRoundingMode:o,activation:l,leakyreluAlpha:c};return i==null?la((I,w,S)=>{let E=G.runKernel(Wl,x,b);return S([w,I,E]),f&&(E=Z(E,[E.shape[1],E.shape[2],E.shape[3]])),{value:E,gradFunc:A}})(p,h):la((I,w,S,E)=>{let D=G.runKernel(Wl,x,b);return E([w,I,D,S]),f&&(D=Z(D,[D.shape[1],D.shape[2],D.shape[3]])),{value:D,gradFunc:A}})(p,h,g)}var iZ=V({fusedDepthwiseConv2d_:oZ});function lZ({a:e,b:t,transposeA:n=!1,transposeB:r=!1,bias:s,activation:a="linear",preluActivationWeights:o,leakyreluAlpha:i}){if(Ff(G.state.gradientDepth,a)===!1){let D=rt(e,t,n,r);return s!=null&&(D=de(D,s)),Df(D,a,o,i)}let l=O(e,"a","fused matMul"),u=O(t,"b","fused matMul");[l,u]=Vt(l,u);let c=n?l.shape[l.rank-2]:l.shape[l.rank-1],d=r?u.shape[u.rank-1]:u.shape[u.rank-2],h=n?l.shape[l.rank-1]:l.shape[l.rank-2],p=r?u.shape[u.rank-2]:u.shape[u.rank-1],f=l.shape.slice(0,-2),m=u.shape.slice(0,-2),g=sn(f),y=sn(m);z(l.rank>=2&&u.rank>=2&&l.rank===u.rank,()=>`Error in fused matMul: inputs must have the same rank of at least 2, got ranks ${l.rank} and ${u.rank}.`),z(Fa(f,m),()=>`Error in fused matMul: outer dimensions (${f}) and (${m}) of Tensors with shapes ${l.shape} and ${u.shape} must match.`),z(c===d,()=>`Error in fused matMul: inner shapes (${c}) and (${d}) of Tensors with shapes ${l.shape} and ${u.shape} and transposeA=${n} and transposeB=${r} must match.`);let A=l.shape.slice(0,-2).concat([h,p]),x=n?Z(l,[g,c,h]):Z(l,[g,h,c]),b=r?Z(u,[y,p,d]):Z(u,[y,d,p]),v;s!=null&&(v=O(s,"bias","fused matMul"),[v]=Vt(v,l),$t(A,v.shape));let I;o!=null&&(I=O(o,"prelu weights","fused matMul"));let w=(D,$)=>{let[R,N,M,B]=$,q=Rf(Z(D,M.shape),M,a),X,J;if(!n&&!r?(X=rt(q,N,!1,!0),J=rt(R,q,!0,!1)):!n&&r?(X=rt(q,N,!1,!1),J=rt(q,R,!0,!1)):n&&!r?(X=rt(N,q,!1,!0),J=rt(R,q,!1,!1)):(X=rt(N,q,!0,!0),J=rt(q,R,!0,!0)),s!=null){let ee=_f(B,q);return[X,J,ee]}else return[X,J]},S={a:x,b,bias:v,preluActivationWeights:I},E={transposeA:n,transposeB:r,activation:a,leakyreluAlpha:i};return s==null?la(($,R,N)=>{let M=G.runKernel(Ll,S,E);return N([$,R,M]),{value:Z(M,A),gradFunc:w}})(x,b):la(($,R,N,M)=>{let B=G.runKernel(Ll,S,E);return M([$,R,B,N]),{value:Z(B,A),gradFunc:w}})(x,b,v)}var uZ=V({fusedMatMul_:lZ});function cZ(e){return xI(e,.54,.46)}var o7e=V({hammingWindow_:cZ});function dZ(e){return xI(e,.5,.5)}var hZ=V({hannWindow_:dZ});function pZ(e,t,n,r=!1,s=0){let a=0,o=[];for(;a+t<=e.size;)o.push(at(e,a,t)),a+=n;if(r)for(;a<e.size;){let i=a+t-e.size,l=an([at(e,a,t-i),Nd([i],s)]);o.push(l),a+=n}return o.length===0?Zl([],[0,t]):Z(an(o),[o.length,t])}var fZ=V({frame_:pZ});function mZ(e,t,n,r,s=hZ){r==null&&(r=eZ(t));let a=fZ(e,t,n),o=j(a,s(t));return hI(o,r)}var i7e=V({stft_:mZ});function gZ(e,t,n,r,s="bilinear",a=0){let o=O(e,"image","cropAndResize"),i=O(t,"boxes","cropAndResize","float32"),l=O(n,"boxInd","cropAndResize","int32"),u=i.shape[0];z(o.rank===4,()=>`Error in cropAndResize: image must be rank 4,but got rank ${o.rank}.`),z(i.rank===2&&i.shape[1]===4,()=>`Error in cropAndResize: boxes must be have size [${u},4] but had shape ${i.shape}.`),z(l.rank===1&&l.shape[0]===u,()=>`Error in cropAndResize: boxInd must be have size [${u}] but had shape ${i.shape}.`),z(r.length===2,()=>`Error in cropAndResize: cropSize must be of length 2, but got length ${r.length}.`),z(r[0]>=1&&r[1]>=1,()=>`cropSize must be atleast [1,1], but was ${r}`),z(s==="bilinear"||s==="nearest",()=>`method must be bilinear or nearest, but was ${s}`);let c={image:o,boxes:i,boxInd:l},d={method:s,extrapolationValue:a,cropSize:r};return G.runKernel($c,c,d)}var yZ=V({cropAndResize_:gZ});function AZ(e){let t=O(e,"image","flipLeftRight","float32");z(t.rank===4,()=>`Error in flipLeftRight: image must be rank 4,but got rank ${t.rank}.`);let n={image:t};return G.runKernel(Mc,n,{})}var xZ=V({flipLeftRight_:AZ});function bZ(e,t,n=0,r=.5){let s=O(e,"image","rotateWithOffset","float32");z(s.rank===4,()=>`Error in rotateWithOffset: image must be rank 4,but got rank ${s.rank}.`);let a={image:s},o={radians:t,fillValue:n,center:r};return G.runKernel(pd,a,o)}var vZ=V({rotateWithOffset_:bZ});function Yl(e,t,n,r,s,a){r==null&&(r=.5),s==null&&(s=Number.NEGATIVE_INFINITY),a==null&&(a=0);let o=e.shape[0];return n=Math.min(n,o),z(0<=r&&r<=1,()=>`iouThreshold must be in [0, 1], but was '${r}'`),z(e.rank===2,()=>`boxes must be a 2D tensor, but was of rank '${e.rank}'`),z(e.shape[1]===4,()=>`boxes must have 4 columns, but 2nd dimension was ${e.shape[1]}`),z(t.rank===1,()=>"scores must be a 1D tensor"),z(t.shape[0]===o,()=>`scores has incompatible shape with boxes. Expected ${o}, but was ${t.shape[0]}`),z(0<=a&&a<=1,()=>`softNmsSigma must be in [0, 1], but was '${a}'`),{maxOutputSize:n,iouThreshold:r,scoreThreshold:s,softNmsSigma:a}}function wZ(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY){let a=O(e,"boxes","nonMaxSuppression"),o=O(t,"scores","nonMaxSuppression"),i=Yl(a,o,n,r,s);n=i.maxOutputSize,r=i.iouThreshold,s=i.scoreThreshold;let l={maxOutputSize:n,iouThreshold:r,scoreThreshold:s};return G.runKernel(Gc,{boxes:a,scores:o},l)}var kZ=V({nonMaxSuppression_:wZ});function IZ(e,t,n){let r=SZ(e,t,n),s=r<0?-(r+1):r;e.splice(s,0,t)}function SZ(e,t,n){return NZ(e,t,n||TZ)}function TZ(e,t){return e>t?1:e<t?-1:0}function NZ(e,t,n){let r=0,s=e.length,a=0,o=!1;for(;r<s;){a=r+(s-r>>>1);let i=n(t,e[a]);i>0?r=a+1:(s=a,o=!i)}return o?r:-r-1}function wI(e,t,n,r,s){return tA(e,t,n,r,s,0)}function kI(e,t,n,r,s,a){return tA(e,t,n,r,s,0,!1,a,!0)}function II(e,t,n,r,s,a){return tA(e,t,n,r,s,a,!0)}function tA(e,t,n,r,s,a,o=!1,i=!1,l=!1){let u=[];for(let g=0;g<t.length;g++)t[g]>s&&u.push({score:t[g],boxIndex:g,suppressBeginIndex:0});u.sort(SI);let c=a>0?-.5/a:0,d=[],h=[];for(;d.length<n&&u.length>0;){let g=u.pop(),{score:y,boxIndex:A,suppressBeginIndex:x}=g;if(y<s)break;let b=!1;for(let v=d.length-1;v>=x;--v){let I=CZ(e,A,d[v]);if(I>=r){b=!0;break}if(g.score=g.score*EZ(r,c,I),g.score<=s)break}g.suppressBeginIndex=d.length,b||(g.score===y?(d.push(A),h.push(g.score)):g.score>s&&IZ(u,g,SI))}let p=d.length,f=n-p;i&&f>0&&(d.push(...new Array(f).fill(0)),h.push(...new Array(f).fill(0)));let m={selectedIndices:d};return o&&(m.selectedScores=h),l&&(m.validOutputs=p),m}function CZ(e,t,n){let r=e.subarray(t*4,t*4+4),s=e.subarray(n*4,n*4+4),a=Math.min(r[0],r[2]),o=Math.min(r[1],r[3]),i=Math.max(r[0],r[2]),l=Math.max(r[1],r[3]),u=Math.min(s[0],s[2]),c=Math.min(s[1],s[3]),d=Math.max(s[0],s[2]),h=Math.max(s[1],s[3]),p=(i-a)*(l-o),f=(d-u)*(h-c);if(p<=0||f<=0)return 0;let m=Math.max(a,u),g=Math.max(o,c),y=Math.min(i,d),A=Math.min(l,h),x=Math.max(y-m,0)*Math.max(A-g,0);return x/(p+f-x)}function EZ(e,t,n){let r=Math.exp(t*n*n);return n<=e?r:0}function SI(e,t){return e.score-t.score||e.score===t.score&&t.boxIndex-e.boxIndex}async function $Z(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY){let a=O(e,"boxes","nonMaxSuppressionAsync"),o=O(t,"scores","nonMaxSuppressionAsync"),i=Yl(a,o,n,r,s);n=i.maxOutputSize,r=i.iouThreshold,s=i.scoreThreshold;let l=await Promise.all([a.data(),o.data()]),u=l[0],c=l[1],{selectedIndices:d}=wI(u,c,n,r,s);return a!==e&&a.dispose(),o!==t&&o.dispose(),En(d,"int32")}var RZ=$Z;function _Z(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY,a=0){let o=O(e,"boxes","nonMaxSuppression"),i=O(t,"scores","nonMaxSuppression"),l=Yl(o,i,n,r,s,a);n=l.maxOutputSize,r=l.iouThreshold,s=l.scoreThreshold,a=l.softNmsSigma;let u={boxes:o,scores:i},c={maxOutputSize:n,iouThreshold:r,scoreThreshold:s,softNmsSigma:a},d=G.runKernel(qc,u,c);return{selectedIndices:d[0],selectedScores:d[1]}}var DZ=V({nonMaxSuppressionWithScore_:_Z});async function FZ(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY,a=0){let o=O(e,"boxes","nonMaxSuppressionAsync"),i=O(t,"scores","nonMaxSuppressionAsync"),l=Yl(o,i,n,r,s,a);n=l.maxOutputSize,r=l.iouThreshold,s=l.scoreThreshold,a=l.softNmsSigma;let u=await Promise.all([o.data(),i.data()]),c=u[0],d=u[1],{selectedIndices:h,selectedScores:p}=II(c,d,n,r,s,a);return o!==e&&o.dispose(),i!==t&&i.dispose(),{selectedIndices:En(h,"int32"),selectedScores:En(p)}}var MZ=FZ;function OZ(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY,a=!1){let o=O(e,"boxes","nonMaxSuppression"),i=O(t,"scores","nonMaxSuppression"),l=Yl(o,i,n,r,s,null),u=l.maxOutputSize,c=l.iouThreshold,d=l.scoreThreshold,h={boxes:o,scores:i},p={maxOutputSize:u,iouThreshold:c,scoreThreshold:d,padToMaxOutputSize:a},f=G.runKernel(jc,h,p);return{selectedIndices:f[0],validOutputs:f[1]}}var PZ=V({nonMaxSuppressionPadded_:OZ});async function zZ(e,t,n,r=.5,s=Number.NEGATIVE_INFINITY,a=!1){let o=O(e,"boxes","nonMaxSuppressionAsync"),i=O(t,"scores","nonMaxSuppressionAsync"),l=Yl(o,i,n,r,s,null),u=l.maxOutputSize,c=l.iouThreshold,d=l.scoreThreshold,[h,p]=await Promise.all([o.data(),i.data()]),{selectedIndices:f,validOutputs:m}=kI(h,p,u,c,d,a);return o!==e&&o.dispose(),i!==t&&i.dispose(),{selectedIndices:En(f,"int32"),validOutputs:De(m,"int32")}}var LZ=zZ;function BZ(e,t,n=!1,r=!1){let s=O(e,"images","resizeBilinear");z(s.rank===3||s.rank===4,()=>`Error in resizeBilinear: x must be rank 3 or 4, but got rank ${s.rank}.`),z(t.length===2,()=>`Error in resizeBilinear: new shape must 2D, but got shape ${t}.`),z(r===!1||n===!1,()=>"Error in resizeBilinear: If halfPixelCenters is true, alignCorners must be false.");let a=s,o=!1;s.rank===3&&(o=!0,a=Z(s,[1,s.shape[0],s.shape[1],s.shape[2]]));let[]=t,i={images:a},l={alignCorners:n,halfPixelCenters:r,size:t},u=G.runKernel(Nl,i,l);return o?Z(u,[u.shape[1],u.shape[2],u.shape[3]]):u}var WZ=V({resizeBilinear_:BZ});function VZ(e,t,n=!1,r=!1){let s=O(e,"images","resizeNearestNeighbor");z(s.rank===3||s.rank===4,()=>`Error in resizeNearestNeighbor: x must be rank 3 or 4, but got rank ${s.rank}.`),z(t.length===2,()=>`Error in resizeNearestNeighbor: new shape must 2D, but got shape ${t}.`),z(s.dtype==="float32"||s.dtype==="int32",()=>"`images` must have `int32` or `float32` as dtype"),z(r===!1||n===!1,()=>"Error in resizeNearestNeighbor: If halfPixelCenters is true, alignCorners must be false.");let a=s,o=!1;s.rank===3&&(o=!0,a=Z(s,[1,s.shape[0],s.shape[1],s.shape[2]]));let[]=t,i={images:a},l={alignCorners:n,halfPixelCenters:r,size:t},u=G.runKernel(sf,i,l);return o?Z(u,[u.shape[1],u.shape[2],u.shape[3]]):u}var UZ=V({resizeNearestNeighbor_:VZ});function HZ(e,t="binary",n=!1,r=.5){let s=O(e,"image","threshold"),a=.2989,o=.587,i=.114,l=s.shape[0]*s.shape[1],u=j(En([r]),255),c,d,h,p;if(z(s.rank===3,()=>`Error in threshold: image must be rank 3,but got rank ${s.rank}.`),z(s.shape[2]===3||s.shape[2]===1,()=>`Error in threshold: image color channel must be equal to 3 or 1but got ${s.shape[2]}.`),z(s.dtype==="int32"||s.dtype==="float32",()=>`Error in dtype: image dtype must be int32 or float32,but got dtype ${s.dtype}.`),z(t==="otsu"||t==="binary",()=>`Method must be binary or otsu, but was ${t}`),s.shape[2]===3){[c,d,h]=Rr(s,[1,1,1],-1);let g=j(c,a),y=j(d,o),A=j(h,i);p=de(de(g,y),A)}else p=e;if(t==="otsu"){let g=V6(xe(aI(p),"int32"),Cs([]),256);u=GZ(g,l)}let f=n?Xl(p,u):$r(p,u);return xe(j(f,255),"int32")}function GZ(e,t){let n=En([-1]),r=En([0]),s=En([0]),a,o,i,l,u,c;for(let d=0;d<e.size-1;d++){a=at(e,0,d+1),o=at(e,d+1),u=Re(Te(a),t),c=Re(Te(o),t);let h=Te(j(a,Rd(0,a.size)));i=Re(h,Te(a));let p=Nd(o.shape,a.size),f=de(Rd(0,o.size),p),m=j(o,f);l=Re(Te(m),Te(o));let g=ke(i,l),y=ke(i,l),A=j(u,c);s=j(j(A,g),y);let x=$r(s,r);r=Zn(x,s,r),n=Zn(x,En([d]),n)}return n}var jZ=V({threshold_:HZ});function qZ(e,t,n="nearest",r="constant",s=0,a){let o=O(e,"image","transform","float32"),i=O(t,"transforms","transform","float32");z(o.rank===4,()=>`Error in transform: image must be rank 4,but got rank ${o.rank}.`),z(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"),z(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:r,fillValue:s,outputShape:a};return G.runKernel(cd,l,u)}var KZ=V({transform_:qZ});function XZ(e,t,n){z(t%1==0,()=>`bandPart(): numLower must be an integer, got ${t}.`),z(n%1==0,()=>`bandPart(): numUpper must be an integer, got ${n}.`);let r=O(e,"a","bandPart");z(r.rank>=2,()=>`bandPart(): Rank must be at least 2, got ${r.rank}.`);let s=r.shape,[a,o]=r.shape.slice(-2);if(!(t<=a))throw new Error(`bandPart(): numLower (${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 (${o}).`);t<0&&(t=a),n<0&&(n=o);let i=Z(Rd(0,a,1,"int32"),[-1,1]),l=Rd(0,o,1,"int32"),u=ke(i,l),c=_s(Xl(u,De(+t,"int32")),Kl(u,De(-n,"int32"))),d=ln([a,o],r.dtype);return Z(Xr(Ds(Z(r,[-1,a,o])).map(h=>Zn(c,h,d))),s)}var ZZ=V({bandPart_:XZ});function YZ(e){let t;if(Array.isArray(e)){t=!1,z(e!=null&&e.length>0,()=>"Gram-Schmidt process: input must not be null, undefined, or empty");let s=e[0].shape[0];for(let a=1;a<e.length;++a)z(e[a].shape[0]===s,()=>`Gram-Schmidt: Non-unique lengths found in the input vectors: (${e[a].shape[0]} vs. ${s})`)}else t=!0,e=Rr(e,e.shape[0],0).map(s=>Qo(s,[0]));z(e.length<=e[0].shape[0],()=>`Gram-Schmidt: Number of vectors (${e.length}) exceeds number of dimensions (${e[0].shape[0]}).`);let n=[],r=e;for(let s=0;s<e.length;++s)n.push(G.tidy(()=>{let a=r[s];if(s>0)for(let o=0;o<s;++o){let i=j(Te(j(n[o],a)),n[o]);a=ke(a,i)}return Re(a,yI(a,"euclidean"))}));return t?Xr(n,0):n}var JZ=V({gramSchmidt_:YZ});function QZ(e,t=!1){if(z(e.rank>=2,()=>`qr() requires input tensor to have a rank >= 2, but got rank ${e.rank}`),e.rank===2)return TI(e,t);{let n=e.shape.slice(0,e.shape.length-2).reduce((l,u)=>l*u),r=Ds(Z(e,[n,e.shape[e.shape.length-2],e.shape[e.shape.length-1]]),0),s=[],a=[];r.forEach(l=>{let[u,c]=TI(l,t);s.push(u),a.push(c)});let o=Z(Xr(s,0),e.shape),i=Z(Xr(a,0),e.shape);return[o,i]}}function TI(e,t=!1){return G.tidy(()=>{z(e.shape.length===2,()=>`qr2d() requires a 2D Tensor, but got a ${e.shape.length}D Tensor.`);let n=e.shape[0],r=e.shape[1],s=X6(n),a=za(e),o=Zl([[1]],[1,1]),i=za(o),l=n>=r?r:n;for(let u=0;u<l;++u){let c=a,d=i,h=s;[i,a,s]=G.tidy(()=>{let p=at(a,[u,u],[n-u,1]),f=yI(p),m=at(a,[u,u],[1,1]),g=Zn($r(m,0),Zl([[-1]]),Zl([[1]])),y=ke(m,j(g,f)),A=Re(p,y);A.shape[0]===1?i=za(o):i=an([o,at(A,[1,0],[A.shape[0]-1,A.shape[1]])],0);let x=qt(Re(rt(g,y),f)),b=at(a,[u,0],[n-u,r]),v=j(x,i),I=st(i);if(u===0)a=ke(b,rt(v,rt(I,b)));else{let E=ke(b,rt(v,rt(I,b)));a=an([at(a,[0,0],[u,r]),E],0)}let w=st(v),S=at(s,[0,u],[n,s.shape[1]-u]);if(u===0)s=ke(S,rt(rt(S,i),w));else{let E=ke(S,rt(rt(S,i),w));s=an([at(s,[0,0],[n,u]),E],1)}return[i,a,s]}),Ge([c,d,h])}return!t&&n>r&&(s=at(s,[0,0],[n,r]),a=at(a,[0,0],[r,r])),[s,a]})}var eY=V({qr_:QZ}),Yn;(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"})(Yn||(Yn={}));function tY(e,t,n=Yn.SUM_BY_NONZERO_WEIGHTS){let r=O(e,"losses","computeWeightedLoss"),s=null;t!=null&&(s=O(t,"weights","computeWeightedLoss"));let a=s==null?r:j(r,s);if(n===Yn.NONE)return a;if(n===Yn.SUM)return Te(a);if(n===Yn.MEAN){if(s==null)return Xt(a);{let o=r.size/s.size,i=Re(Te(a),Te(s));return o>1?Re(i,De(o)):i}}if(n===Yn.SUM_BY_NONZERO_WEIGHTS){if(s==null)return Re(Te(a),De(r.size));{let o=j(s,ua(r.shape)),i=xe(Te(Ed(o,De(0))),"float32");return Re(Te(a),i)}}throw Error(`Unknown reduction: ${n}`)}var Ua=V({computeWeightedLoss_:tY});function nY(e,t,n,r=Yn.SUM_BY_NONZERO_WEIGHTS){let s=O(e,"labels","absoluteDifference"),a=O(t,"predictions","absoluteDifference"),o=null;n!=null&&(o=O(n,"weights","absoluteDifference")),ss(s.shape,a.shape,"Error in absoluteDifference: ");let i=gn(ke(s,a));return Ua(i,o,r)}var l7e=V({absoluteDifference_:nY});function rY(e,t,n,r,s=Yn.SUM_BY_NONZERO_WEIGHTS){let a=O(e,"labels","cosineDistance"),o=O(t,"predictions","cosineDistance"),i=null;r!=null&&(i=O(r,"weights","cosineDistance")),ss(a.shape,o.shape,"Error in cosineDistance: ");let l=De(1),u=ke(l,Te(j(a,o),n,!0));return Ua(u,i,s)}var u7e=V({cosineDistance_:rY});function sY(e,t,n,r=Yn.SUM_BY_NONZERO_WEIGHTS){let s=O(e,"labels","hingeLoss"),a=O(t,"predictions","hingeLoss"),o=null;n!=null&&(o=O(n,"weights","hingeLoss")),ss(s.shape,a.shape,"Error in hingeLoss: ");let i=De(1);s=ke(j(De(2),s),i);let l=ca(ke(i,j(s,a)));return Ua(l,o,r)}var c7e=V({hingeLoss_:sY});function aY(e,t,n,r=1,s=Yn.SUM_BY_NONZERO_WEIGHTS){let a=O(e,"labels","huberLoss"),o=O(t,"predictions","huberLoss"),i=null;n!=null&&(i=O(n,"weights","huberLoss")),ss(a.shape,o.shape,"Error in huberLoss: ");let l=De(r),u=gn(ke(o,a)),c=Nf(u,l),d=ke(u,c),h=de(j(De(.5),Tt(c)),j(l,d));return Ua(h,i,s)}var d7e=V({huberLoss_:aY});function oY(e,t,n,r=1e-7,s=Yn.SUM_BY_NONZERO_WEIGHTS){let a=O(e,"labels","logLoss"),o=O(t,"predictions","logLoss"),i=null;n!=null&&(i=O(n,"weights","logLoss")),ss(a.shape,o.shape,"Error in logLoss: ");let l=De(1),u=De(r),c=qt(j(a,jr(de(o,u)))),d=j(ke(l,a),jr(de(ke(l,o),u))),h=ke(c,d);return Ua(h,i,s)}var h7e=V({logLoss_:oY});function iY(e,t,n,r=Yn.SUM_BY_NONZERO_WEIGHTS){let s=O(e,"labels","meanSquaredError"),a=O(t,"predictions","meanSquaredError"),o=null;n!=null&&(o=O(n,"weights","meanSquaredError")),ss(s.shape,a.shape,"Error in meanSquaredError: ");let i=pI(s,a);return Ua(i,o,r)}var p7e=V({meanSquaredError_:iY});function lY(e,t){let n=O(e,"labels","sigmoidCrossEntropyWithLogits"),r=O(t,"logits","sigmoidCrossEntropyWithLogits");ss(n.shape,r.shape,"Error in sigmoidCrossEntropyWithLogits: ");let s=ca(r),a=j(r,n),o=zy(Gr(qt(gn(r))));return de(ke(s,a),o)}function uY(e,t,n,r=0,s=Yn.SUM_BY_NONZERO_WEIGHTS){let a=O(e,"multiClassLabels","sigmoidCrossEntropy"),o=O(t,"logits","sigmoidCrossEntropy"),i=null;if(n!=null&&(i=O(n,"weights","sigmoidCrossEntropy")),ss(a.shape,o.shape,"Error in sigmoidCrossEntropy: "),r>0){let u=De(r),c=De(1),d=De(.5);a=de(j(a,ke(c,u)),j(d,u))}let l=lY(a,o);return Ua(l,i,s)}var f7e=V({sigmoidCrossEntropy_:uY});function cY(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 la((s,a,o)=>{let l=nK(a,[n],!0),u=ke(xe(a,"float32"),l);o([s,u]);let c=qt(j(u,s));return{value:Te(c,[n]),gradFunc:(p,f)=>{let[m,g]=f,y=Zo(p.shape,[n]);return[j(Z(p,y),ke(xe(m,"float32"),Gr(g))),j(Z(p,y),ke(Gr(g),xe(m,"float32")))]}}})(e,t)}function dY(e,t,n,r=0,s=Yn.SUM_BY_NONZERO_WEIGHTS){let a=O(e,"onehotLabels","softmaxCrossEntropy"),o=O(t,"logits","softmaxCrossEntropy"),i=null;if(n!=null&&(i=O(n,"weights","softmaxCrossEntropy")),ss(a.shape,o.shape,"Error in softmaxCrossEntropy: "),r>0){let u=De(r),c=De(1),d=De(a.shape[1]);a=de(j(a,ke(c,u)),Re(u,d))}let l=cY(a,o);return Ua(l,i,s)}var m7e=V({softmaxCrossEntropy_:dY});function hY(e,t,n,r){let s=O(e,"indices","sparseFillEmptyRows"),a=O(t,"values","sparseFillEmptyRows"),o=O(n,"denseShape","sparseFillEmptyRows"),i=O(r,"defaultValue","sparseFillEmptyRows",a.dtype);if(s.rank!==2)throw new Error(`Indices should be Tensor2D but received shape
${s.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:s,values:a,denseShape:o,defaultValue:i},u=G.runKernel(G1,l);return{outputIndices:u[0],outputValues:u[1],emptyRowIndicator:u[2],reverseIndexMap:u[3]}}var pY=V({sparseFillEmptyRows_:hY});function fY(e,t,n){let r=O(e,"inputIndices","sparseReshape"),s=O(t,"inputShape","sparseReshape"),a=O(n,"newShape","sparseReshape");if(r.rank!==2)throw new Error(`Input indices should be Tensor2D but received shape
${r.shape}`);if(s.rank!==1)throw new Error(`Input shape should be Tensor1D but received shape ${s.shape}`);if(a.rank!==1)throw new Error(`New shape should be Tensor1D but received shape ${a.shape}`);let o={inputIndices:r,inputShape:s,newShape:a},i=G.runKernel(j1,o);return{outputIndices:i[0],outputShape:i[1]}}var mY=V({sparseReshape_:fY});function gY(e,t,n){let r=O(e,"data","sparseSegmentMean"),s=O(t,"indices","sparseSegmentMean"),a=O(n,"segmentIds","sparseSegmentMean");if(r.rank<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(s.rank!==1)throw new Error(`Indices should be Tensor1D but received shape
${s.shape}`);if(a.rank!==1)throw new Error(`Segment ids should be Tensor1D but received shape
${a.shape}`);let o={data:r,indices:s,segmentIds:a};return G.runKernel(q1,o)}var yY=V({sparseSegmentMean_:gY});function AY(e,t,n){let r=O(e,"data","sparseSegmentSum"),s=O(t,"indices","sparseSegmentSum"),a=O(n,"segmentIds","sparseSegmentSum");if(r.rank<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(s.rank!==1)throw new Error(`Indices should be Tensor1D but received shape
${s.shape}`);if(a.rank!==1)throw new Error(`Segment ids should be Tensor1D but received shape
${a.shape}`);let o={data:r,indices:s,segmentIds:a};return G.runKernel(K1,o)}var xY=V({sparseSegmentSum_:AY});function bY(e,t,n,r,s,a,o,i){let l=O(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=O(t,"dataSplits","stringNGrams");if(u.dtype!=="int32")throw new Error("Data splits must be of datatype int32");let c={separator:n,nGramWidths:r,leftPad:s,rightPad:a,padWidth:o,preserveShortSequences:i},d={data:l,dataSplits:u},h=G.runKernel(Z1,d,c);return{nGrams:h[0],nGramsSplits:h[1]}}var vY=V({stringNGrams_:bY});function wY(e,t,n=!0){let r=O(e,"input","stringSplit","string"),s=O(t,"delimiter","stringSplit","string");if(r.rank!==1)throw new Error(`Input should be Tensor1D but received shape ${r.shape}`);if(s.rank!==0)throw new Error(`Delimiter should be a scalar but received shape ${s.shape}`);let a={skipEmpty:n},o={input:r,delimiter:s},i=G.runKernel(Y1,o,a);return{indices:i[0],values:i[1],shape:i[2]}}var kY=V({stringSplit_:wY});function IY(e,t){let n=O(e,"input","stringToHashBucketFast","string"),r={numBuckets:t};if(t<=0)throw new Error("Number of buckets must be at least 1");let s={input:n};return G.runKernel(J1,s,r)}var SY=V({stringToHashBucketFast_:IY}),is={flipLeftRight:xZ,resizeNearestNeighbor:UZ,resizeBilinear:WZ,rotateWithOffset:vZ,cropAndResize:yZ,nonMaxSuppression:kZ,nonMaxSuppressionAsync:RZ,nonMaxSuppressionWithScore:DZ,nonMaxSuppressionWithScoreAsync:MZ,nonMaxSuppressionPadded:PZ,nonMaxSuppressionPaddedAsync:LZ,threshold:jZ,transform:KZ},TY={bandPart:ZZ,gramSchmidt:JZ,qr:eY},Mf={sparseFillEmptyRows:pY,sparseReshape:mY,sparseSegmentMean:yY,sparseSegmentSum:xY},nA={stringNGrams:vY,stringSplit:kY,stringToHashBucketFast:SY},Ha=class extends F6{minimize(e,t=!1,n){let{value:r,grads:s}=this.computeGradients(e,n);if(n!=null){let a=n.map(o=>({name:o.name,tensor:s[o.name]}));this.applyGradients(a)}else this.applyGradients(s);return Ge(s),t?r:(r.dispose(),null)}get iterations(){return this.iterations_==null&&(this.iterations_=0),this.iterations_}incrementIterations(){this.iterations_=this.iterations+1}computeGradients(e,t){return Gq(e,t)}dispose(){this.iterations_!=null&&Ge(this.iterations_)}async saveIterations(){return this.iterations_==null&&(this.iterations_=0),{name:"iter",tensor:De(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 ${this.getClassName()}`)}async extractIterations(e){return this.iterations_=(await e[0].tensor.data())[0],e.slice(1)}};Object.defineProperty(Ha,Symbol.hasInstance,{value:e=>e.minimize!=null&&e.computeGradients!=null&&e.applyGradients!=null});var rA=class extends Ha{constructor(e,t,n=null){super();this.learningRate=e,this.rho=t,this.epsilon=n,this.accumulatedGrads=[],this.accumulatedUpdates=[],n==null&&(this.epsilon=G.backend.epsilon())}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=G.registeredVariables[n],a=!1;this.accumulatedGrads[r]==null&&(this.accumulatedGrads[r]={originalName:`${n}/accum_grad`,variable:Y(()=>ot(s).variable(a))}),this.accumulatedUpdates[r]==null&&(this.accumulatedUpdates[r]={originalName:`${n}/accum_var`,variable:Y(()=>ot(s).variable(a))});let o=Array.isArray(e)?e[r].tensor:e[n];if(o==null)return;let i=this.accumulatedGrads[r].variable,l=this.accumulatedUpdates[r].variable;Y(()=>{let u=de(j(i,this.rho),j(Tt(o),1-this.rho)),c=j(Re(Ln(de(l,this.epsilon)),Ln(de(i,this.epsilon))),o),d=de(j(l,this.rho),j(Tt(c),1-this.rho));i.assign(u),l.assign(d);let h=de(j(c,-this.learningRate),s);s.assign(h)})}),this.incrementIterations()}dispose(){this.accumulatedUpdates!=null&&(Ge(this.accumulatedGrads.map(e=>e.variable)),Ge(this.accumulatedUpdates.map(e=>e.variable)))}async getWeights(){let e=[...this.accumulatedGrads,...this.accumulatedUpdates];return[await this.saveIterations()].concat(e.map(t=>({name:t.originalName,tensor:t.variable})))}async setWeights(e){e=await this.extractIterations(e);let t=e.length/2,n=!1;this.accumulatedGrads=e.slice(0,t).map(r=>({originalName:r.name,variable:r.tensor.variable(n)})),this.accumulatedUpdates=e.slice(t,t*2).map(r=>({originalName:r.name,variable:r.tensor.variable(n)}))}getConfig(){return{learningRate:this.learningRate,rho:this.rho,epsilon:this.epsilon}}static fromConfig(e,t){return new e(t.learningRate,t.rho,t.epsilon)}};rA.className="Adadelta";La(rA);var sA=class extends Ha{constructor(e,t=.1){super();this.learningRate=e,this.initialAccumulatorValue=t,this.accumulatedGrads=[]}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=G.registeredVariables[n];if(this.accumulatedGrads[r]==null){let i=!1;this.accumulatedGrads[r]={originalName:`${n}/accumulator`,variable:Y(()=>Nd(s.shape,this.initialAccumulatorValue).variable(i))}}let a=Array.isArray(e)?e[r].tensor:e[n];if(a==null)return;let o=this.accumulatedGrads[r].variable;Y(()=>{let i=de(o,Tt(a));o.assign(i);let l=de(j(Re(a,Ln(de(i,G.backend.epsilon()))),-this.learningRate),s);s.assign(l)})}),this.incrementIterations()}dispose(){this.accumulatedGrads!=null&&Ge(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)}};sA.className="Adagrad";La(sA);var aA=class extends Ha{constructor(e,t,n,r=null){super();this.learningRate=e,this.beta1=t,this.beta2=n,this.epsilon=r,this.accumulatedFirstMoment=[],this.accumulatedSecondMoment=[],Y(()=>{this.accBeta1=De(t).variable(),this.accBeta2=De(n).variable()}),r==null&&(this.epsilon=G.backend.epsilon())}applyGradients(e){let t=Array.isArray(e)?e.map(n=>n.name):Object.keys(e);Y(()=>{let n=ke(1,this.accBeta1),r=ke(1,this.accBeta2);t.forEach((s,a)=>{let o=G.registeredVariables[s],i=!1;this.accumulatedFirstMoment[a]==null&&(this.accumulatedFirstMoment[a]={originalName:`${s}/m`,variable:Y(()=>ot(o).variable(i))}),this.accumulatedSecondMoment[a]==null&&(this.accumulatedSecondMoment[a]={originalName:`${s}/v`,variable:Y(()=>ot(o).variable(i))});let l=Array.isArray(e)?e[a].tensor:e[s];if(l==null)return;let u=this.accumulatedFirstMoment[a].variable,c=this.accumulatedSecondMoment[a].variable,d=de(j(u,this.beta1),j(l,1-this.beta1)),h=de(j(c,this.beta2),j(Tt(l),1-this.beta2)),p=Re(d,n),f=Re(h,r);u.assign(d),c.assign(h);let m=de(j(Re(p,de(Ln(f),this.epsilon)),-this.learningRate),o);o.assign(m)}),this.accBeta1.assign(j(this.accBeta1,this.beta1)),this.accBeta2.assign(j(this.accBeta2,this.beta2))}),this.incrementIterations()}dispose(){this.accBeta1.dispose(),this.accBeta2.dispose(),this.accumulatedFirstMoment!=null&&Ge(this.accumulatedFirstMoment.map(e=>e.variable)),this.accumulatedSecondMoment!=null&&Ge(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(Jo(this.beta1,this.iterations_+1)),this.accBeta2.assign(Jo(this.beta2,this.iterations_+1))});let t=e.length/2,n=!1;this.accumulatedFirstMoment=e.slice(0,t).map(r=>({originalName:r.name,variable:r.tensor.variable(n)})),this.accumulatedSecondMoment=e.slice(t,t*2).map(r=>({originalName:r.name,variable:r.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)}};aA.className="Adam";La(aA);var oA=class extends Ha{constructor(e,t,n,r=null,s=0){super();this.learningRate=e,this.beta1=t,this.beta2=n,this.epsilon=r,this.decay=s,this.accumulatedFirstMoment=[],this.accumulatedWeightedInfNorm=[],Y(()=>{this.iteration=De(0).variable(),this.accBeta1=De(t).variable()}),r==null&&(this.epsilon=G.backend.epsilon())}applyGradients(e){let t=Array.isArray(e)?e.map(n=>n.name):Object.keys(e);Y(()=>{let n=ke(1,this.accBeta1),r=Re(-this.learningRate,de(j(this.iteration,this.decay),1));t.forEach((s,a)=>{let o=G.registeredVariables[s],i=!1;this.accumulatedFirstMoment[a]==null&&(this.accumulatedFirstMoment[a]={originalName:`${s}/m`,variable:ot(o).variable(i)}),this.accumulatedWeightedInfNorm[a]==null&&(this.accumulatedWeightedInfNorm[a]={originalName:`${s}/v`,variable:ot(o).variable(i)});let l=Array.isArray(e)?e[a].tensor:e[s];if(l==null)return;let u=this.accumulatedFirstMoment[a].variable,c=this.accumulatedWeightedInfNorm[a].variable,d=de(j(u,this.beta1),j(l,1-this.beta1)),h=j(c,this.beta2),p=gn(l),f=Va(h,p);u.assign(d),c.assign(f);let m=de(j(Re(r,n),Re(d,de(f,this.epsilon))),o);o.assign(m)}),this.iteration.assign(de(this.iteration,1)),this.accBeta1.assign(j(this.accBeta1,this.beta1))}),this.incrementIterations()}dispose(){this.accBeta1.dispose(),this.iteration.dispose(),this.accumulatedFirstMoment!=null&&Ge(this.accumulatedFirstMoment.map(e=>e.variable)),this.accumulatedWeightedInfNorm!=null&&Ge(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)}};oA.className="Adamax";La(oA);var Of=class extends Ha{constructor(e){super();this.learningRate=e,this.setLearningRate(e)}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=Array.isArray(e)?e[r].tensor:e[n];if(s==null)return;let a=G.registeredVariables[n];Y(()=>{let o=de(j(this.c,s),a);a.assign(o)})}),this.incrementIterations()}setLearningRate(e){this.learningRate=e,this.c!=null&&this.c.dispose(),this.c=In(De(-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)}};Of.className="SGD";La(Of);var iA=class extends Of{constructor(e,t,n=!1){super(e);this.learningRate=e,this.momentum=t,this.useNesterov=n,this.accumulations=[],this.m=De(this.momentum)}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=G.registeredVariables[n];if(this.accumulations[r]==null){let i=!1;this.accumulations[r]={originalName:`${n}/momentum`,variable:Y(()=>ot(s).variable(i))}}let a=this.accumulations[r].variable,o=Array.isArray(e)?e[r].tensor:e[n];o!=null&&Y(()=>{let i,l=de(j(this.m,a),o);this.useNesterov?i=de(j(this.c,de(o,j(l,this.m))),s):i=de(j(this.c,l),s),a.assign(l),s.assign(i)})}),this.incrementIterations()}dispose(){this.m.dispose(),this.accumulations!=null&&Ge(this.accumulations.map(e=>e.variable))}setMomentum(e){this.momentum=e}async getWeights(){return[await this.saveIterations()].concat(this.accumulations.map(e=>({name:e.originalName,tensor:e.variable})))}async setWeights(e){e=await this.extractIterations(e);let t=!1;this.accumulations=e.map(n=>({originalName:n.name,variable:n.tensor.variable(t)}))}getConfig(){return{learningRate:this.learningRate,momentum:this.momentum,useNesterov:this.useNesterov}}static fromConfig(e,t){return new e(t.learningRate,t.momentum,t.useNesterov)}};iA.className="Momentum";La(iA);var lA=class extends Ha{constructor(e,t=.9,n=0,r=null,s=!1){super();if(this.learningRate=e,this.decay=t,this.momentum=n,this.epsilon=r,this.accumulatedMeanSquares=[],this.accumulatedMoments=[],this.accumulatedMeanGrads=[],this.centered=s,r==null&&(this.epsilon=G.backend.epsilon()),e==null)throw new Error("learningRate for RMSPropOptimizer must be defined.")}applyGradients(e){(Array.isArray(e)?e.map(n=>n.name):Object.keys(e)).forEach((n,r)=>{let s=G.registeredVariables[n],a=!1;this.accumulatedMeanSquares[r]==null&&(this.accumulatedMeanSquares[r]={originalName:`${n}/rms`,variable:Y(()=>ot(s).variable(a))}),this.accumulatedMoments[r]==null&&(this.accumulatedMoments[r]={originalName:`${n}/momentum`,variable:Y(()=>ot(s).variable(a))}),this.accumulatedMeanGrads[r]==null&&this.centered&&(this.accumulatedMeanGrads[r]={originalName:`${n}/mg`,variable:Y(()=>ot(s).variable(a))});let o=Array.isArray(e)?e[r].tensor:e[n];if(o==null)return;let i=this.accumulatedMeanSquares[r].variable,l=this.accumulatedMoments[r].variable;Y(()=>{let u=de(j(i,this.decay),j(Tt(o),1-this.decay));if(this.centered){let c=this.accumulatedMeanGrads[r].variable,d=de(j(c,this.decay),j(o,1-this.decay)),h=Re(j(o,this.learningRate),Ln(ke(u,de(Tt(d),this.epsilon)))),p=de(j(l,this.momentum),h);i.assign(u),c.assign(d),l.assign(p);let f=ke(s,p);s.assign(f)}else{let c=de(j(i,this.decay),j(Tt(o),1-this.decay)),d=de(j(l,this.momentum),Re(j(o,this.learningRate),Ln(de(c,this.epsilon))));i.assign(c),l.assign(d);let h=ke(s,d);s.assign(h)}})}),this.incrementIterations()}dispose(){this.accumulatedMeanSquares!=null&&Ge(this.accumulatedMeanSquares.map(e=>e.variable)),this.accumulatedMeanGrads!=null&&this.centered&&Ge(this.accumulatedMeanGrads.map(e=>e.variable)),this.accumulatedMoments!=null&&Ge(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(r=>({originalName:r.name,variable:r.tensor.variable(n)})),this.accumulatedMoments=e.slice(t,t*2).map(r=>({originalName:r.name,variable:r.tensor.variable(n)})),this.centered&&(this.accumulatedMeanGrads=e.slice(t*2,t*3).map(r=>({originalName:r.name,variable:r.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)}};lA.className="RMSProp";La(lA);var ti=class{static sgd(e){return new Of(e)}static momentum(e,t,n=!1){return new iA(e,t,n)}static rmsprop(e,t=.9,n=0,r=null,s=!1){return new lA(e,t,n,r,s)}static adam(e=.001,t=.9,n=.999,r=null){return new aA(e,t,n,r)}static adadelta(e=.001,t=.95,n=null){return new rA(e,t,n)}static adamax(e=.002,t=.9,n=.999,r=null,s=0){return new oA(e,t,n,r,s)}static adagrad(e,t=.1){return new sA(e,t)}},Jl={sgd:ti.sgd,momentum:ti.momentum,adadelta:ti.adadelta,adagrad:ti.adagrad,rmsprop:ti.rmsprop,adamax:ti.adamax,adam:ti.adam},NY=(()=>typeof requestAnimationFrame!="undefined"?requestAnimationFrame:typeof setImmediate!="undefined"?setImmediate:e=>e())();function uA(){return new Promise(e=>NY(()=>e()))}var _={};_e(_,{ERF_A1:()=>zY,ERF_A2:()=>LY,ERF_A3:()=>BY,ERF_A4:()=>WY,ERF_A5:()=>VY,ERF_P:()=>PY,PARALLELIZE_THRESHOLD:()=>cA,SELU_SCALE:()=>CI,SELU_SCALEALPHA:()=>NI,applyActivation:()=>Df,assertAndGetBroadcastShape:()=>$t,assertAxesAreInnerMostDims:()=>Qq,assertParamsConsistent:()=>CY,assignToTypedArray:()=>ZY,axesAreInnerMostDims:()=>Ly,calculateShapes:()=>w6,checkEinsumDimSizes:()=>nJ,combineLocations:()=>J6,complexWithEvenIndex:()=>qY,complexWithOddIndex:()=>KY,computeConv2DInfo:()=>Id,computeConv3DInfo:()=>L6,computeDefaultPad:()=>$y,computeDilation2DInfo:()=>uj,computeOptimalWindowSize:()=>$Y,computeOutAndReduceShapes:()=>Q6,computeOutShape:()=>EY,computePool2DInfo:()=>z6,computePool3DInfo:()=>cj,convertConv2DDataFormat:()=>B6,decodeEinsumEquation:()=>eJ,eitherStridesOrDilationsAreOne:()=>Es,expandShapeToKeepDim:()=>Zo,exponent:()=>JY,exponents:()=>YY,fromStringArrayToUint8:()=>dJ,fromUint8ToStringArray:()=>cJ,getAxesPermutation:()=>eI,getBroadcastDims:()=>cq,getComplexWithIndex:()=>XY,getEinsumComputePath:()=>rJ,getEinsumPermutation:()=>tJ,getFusedBiasGradient:()=>_f,getFusedDyActivation:()=>Rf,getImageCenter:()=>RY,getInnerMostAxes:()=>eK,getPermuted:()=>DY,getReductionAxes:()=>on,getReshaped:()=>_Y,getReshapedPermuted:()=>FY,getSliceBeginCoords:()=>MY,getSliceSize:()=>OY,getUndoAxesPermutation:()=>By,isIdentityPermutation:()=>sJ,log:()=>HY,mergeRealAndImagArrays:()=>GY,prepareAndValidate:()=>b6,prepareSplitSize:()=>oJ,segment_util:()=>RI,shouldFuse:()=>Ff,slice_util:()=>Cn,splitRealAndImagArrays:()=>jY,tupleValuesAreOne:()=>Wa,upcastType:()=>Ur,validateInput:()=>Ty,validateUpdateShape:()=>Sy,warn:()=>UY});function CY(e,t){let n=e[0].length;e.forEach((s,a)=>{z(s.length===n,()=>`Error in concat${n}D: rank of tensors[${a}] must be the same as the rank of the rest (${n})`)}),z(t>=0&&t<n,()=>`Error in concat${n}D: axis must be between 0 and ${n-1}.`);let r=e[0];e.forEach((s,a)=>{for(let o=0;o<n;o++)z(o===t||s[o]===r[o],()=>`Error in concat${n}D: Shape of tensors[${a}] (${s}) does not match the shape of the rest (${r}) along the non-concatenated axis ${a}.`)})}function EY(e,t){let n=e[0].slice();for(let r=1;r<e.length;r++)n[t]+=e[r][t];return n}var cA=30;function $Y(e){return e<=cA?e:Gp(e,Math.floor(Math.sqrt(e)))}function RY(e,t,n){let r=n*(typeof e=="number"?e:e[0]),s=t*(typeof e=="number"?e:e[1]);return[r,s]}function _Y(e,t,n,r=!0){let s=[];if(r)s=s.concat(t.slice(0)),s.push(e[0]/n),s=s.concat(e.slice(1));else{s=s.concat(e[0]);let a=t.length;for(let o=0;o<a;++o)s=s.concat([e[o+1]/t[o],t[o]]);s=s.concat(e.slice(a+1))}return s}function DY(e,t,n=!0){let r=[];if(n){r.push(t);for(let s=t+1;s<e;++s)s<=2*t?(r.push(s),r.push(s-(t+1))):r.push(s)}else{let s=[],a=[];for(let o=1;o<e;++o)o>=t*2+1||o%2==1?a.push(o):s.push(o);r.push(...s),r.push(0),r.push(...a)}return r}function FY(e,t,n,r=!0){let s=[];r?s.push(e[0]/n):s.push(e[0]*n);for(let a=1;a<e.length;++a)a<=t.length?r?s.push(t[a-1]*e[a]):s.push(e[a]/t[a-1]):s.push(e[a]);return s}function MY(e,t){let n=[0];for(let r=0;r<t;++r)n.push(e[r][0]);return n}function OY(e,t,n){let r=e.slice(0,1);for(let s=0;s<n;++s)r.push(e[s+1]-t[s][0]-t[s][1]);return r}var NI=1.7580993408473768,CI=1.0507009873554805,PY=.3275911,zY=.254829592,LY=-.284496736,BY=1.421413741,WY=-1.453152027,VY=1.061405429;function UY(...e){re().getBool("IS_TEST")||console.warn(...e)}function HY(...e){re().getBool("IS_TEST")||console.log(...e)}function GY(e,t){if(e.length!==t.length)throw new Error(`Cannot merge real and imag arrays of different lengths. real:${e.length}, imag: ${t.length}.`);let n=new Float32Array(e.length*2);for(let r=0;r<n.length;r+=2)n[r]=e[r/2],n[r+1]=t[r/2];return n}function jY(e){let t=new Float32Array(e.length/2),n=new Float32Array(e.length/2);for(let r=0;r<e.length;r+=2)t[r/2]=e[r],n[r/2]=e[r+1];return{real:t,imag:n}}function qY(e){let t=Math.ceil(e.length/4),n=new Float32Array(t),r=new Float32Array(t);for(let s=0;s<e.length;s+=4)n[Math.floor(s/4)]=e[s],r[Math.floor(s/4)]=e[s+1];return{real:n,imag:r}}function KY(e){let t=Math.floor(e.length/4),n=new Float32Array(t),r=new Float32Array(t);for(let s=2;s<e.length;s+=4)n[Math.floor(s/4)]=e[s],r[Math.floor(s/4)]=e[s+1];return{real:n,imag:r}}function XY(e,t){let n=e[t*2],r=e[t*2+1];return{real:n,imag:r}}function ZY(e,t,n,r){e[r*2]=t,e[r*2+1]=n}function YY(e,t){let n=new Float32Array(e/2),r=new Float32Array(e/2);for(let s=0;s<Math.ceil(e/2);s++){let a=(t?2:-2)*Math.PI*(s/e);n[s]=Math.cos(a),r[s]=Math.sin(a)}return{real:n,imag:r}}function JY(e,t,n){let r=(n?2:-2)*Math.PI*(e/t),s=Math.cos(r),a=Math.sin(r);return{real:s,imag:a}}var dA="->",QY=/->/g,EI=",",$I="...";function eJ(e,t){e=e.replace(/\s/g,"");let n=(e.length-e.replace(QY,"").length)/dA.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 ("${dA}").`);let[r,s]=e.split(dA);z(r.indexOf($I)===-1,()=>`The ellipsis notation ("${$I}") is not supported yet.`);let a=r.split(EI),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 h=0;h<s.length;++h){let p=s[h];if(!a.some(f=>f.indexOf(p)!==-1))throw new Error(`Output subscripts contain the label ${p} not present in the input subscripts.`);i.indexOf(p)===-1&&i.push(p)}for(let h=0;h<r.length;++h){let p=r[h];i.indexOf(p)===-1&&p!==EI&&i.push(p)}let l=new Array(a.length);for(let h=0;h<o;++h){if(new Set(a[h].split("")).size!==a[h].length)throw new Error(`Found duplicate axes in input component ${a[h]}. Support for duplicate axes in input is not implemented yet.`);l[h]=[];for(let p=0;p<a[h].length;++p)l[h].push(i.indexOf(a[h][p]))}let u=i.length,c=s.length,d=[];for(let h=c;h<u;++h)d.push(h);return{allDims:i,summedDims:d,idDims:l}}function tJ(e,t){let n=new Array(e);n.fill(-1);for(let s=0;s<t.length;++s)n[t[s]]=s;let r=[];for(let s=0;s<e;++s)n[s]===-1&&r.push(s);return n=n.filter(s=>s!==-1),{permutationIndices:n,expandDims:r}}function nJ(e,t,n){let r=new Array(e);for(let s=0;s<n.length;++s){let a=n[s].shape;for(let o=0;o<t[s].length;++o)r[t[s][o]]===void 0?r[t[s][o]]=a[o]:z(r[t[s][o]]===a[o],()=>`Expected dimension ${r[t[s][o]]} at axis ${o} of input shaped ${JSON.stringify(a)}, but got dimension ${a[o]}`)}}function rJ(e,t){let n=e,r=[],s=0;e.length===0&&n.push(-1),s=e.length+1;for(let o=0;o<s;++o)r.push([]);let a=[];for(let o=0;o<n.length;++o){let i=n[o],l=aJ(t,i);for(let u of l)a.indexOf(u)===-1&&(r[o].push(u),a.push(u))}return{path:n,steps:r}}function sJ(e){return e.every((t,n)=>t===n)}function aJ(e,t){let n=[];for(let r=0;r<e.length;++r)(e[r].length===0||e[r].indexOf(t)!==-1||t===-1)&&n.push(r);return n}function oJ(e,t,n=0){let r=[];if(typeof t=="number")z(e.shape[n]%t==0,()=>"Number of splits must evenly divide the axis."),r=new Array(t).fill(e.shape[n]/t);else{let s=t.reduce((o,i)=>(i===-1&&(o+=1),o),0);z(s<=1,()=>"There should be only one negative value in split array.");let a=t.indexOf(-1);if(a!==-1){let o=t.reduce((i,l)=>l>0?i+l:i);t[a]=e.shape[n]-o}z(e.shape[n]===t.reduce((o,i)=>o+i),()=>"The sum of sizes must match the size of the axis dimension."),r=t}return r}var RI={};_e(RI,{collectGatherOpShapeInfo:()=>uJ,computeOutShape:()=>lJ,segOpComputeOptimalWindowSize:()=>iJ});function iJ(e,t){let n=!1,r;for(e<=cA?(r=e,n=!0):r=Gp(e,Math.floor(Math.sqrt(e)));!n;)r>t||r===e?n=!0:r=Gp(e,r+1);return r}function lJ(e,t,n){let r=[],s=e.length;for(let a=0;a<s;a++)a!==t?r.push(e[a]):r.push(n);return r}function uJ(e,t,n,r){let s=t.shape.length,a=e.shape.length;if(r!==0&&(r<-s||r>s))throw new Error(`Expect batchDims in the range of [-${s}, ${s}], but got ${r}`);if(r<0&&(r+=s),r>a)throw new Error(`batchDims (${r}) must be less than rank(x) (
${a}).`);if(n<r)throw new Error(`batchDims (${r}) must be less than or equal to axis (${n}).`);for(let d=0;d<r;++d)if(e.shape[d]!==t.shape[d])throw new Error(`x.shape[${d}]: ${e.shape[d]} should be equal to indices.shape[${d}]: ${t.shape[d]}.`);let o=e.shape[n],i=[],l=1,u=1,c=1;for(let d=0;d<r;++d)i.push(e.shape[d]),l*=e.shape[d];for(let d=r;d<n;d++)i.push(e.shape[d]),u*=e.shape[d];for(let d=r;d<s;d++)i.push(t.shape[d]);for(let d=n+1;d<a;d++)i.push(e.shape[d]),c*=e.shape[d];return{batchSize:l,sliceSize:c,outerSize:u,dimSize:o,outputShape:i}}function cJ(e){try{return e.map(t=>hf(t))}catch(t){throw new Error(`Failed to decode encoded string bytes into utf-8, error: ${t}`)}}function dJ(e){return e.map(t=>df(t))}var da={};_e(da,{nonMaxSuppressionV3Impl:()=>wI,nonMaxSuppressionV4Impl:()=>kI,nonMaxSuppressionV5Impl:()=>II,whereImpl:()=>mI});var _I={kernelName:Ac,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(e,$f(xe(n,"float32"),-1))}}},hJ={kernelName:xc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>{let r=Tt(xe(n,"float32")),s=Ln(ke(De(1),r));return qt(Re(e,s))}}}},pJ={kernelName:bc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>{let r=Ln(ke(Tt(xe(n,"float32")),1));return Re(e,r)}}}},fJ={kernelName:Ma,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t,s=$t(n.shape,r.shape);return{a:()=>{let i=e,l=on(n.shape,s);return l.length>0&&(i=Te(i,l)),Z(i,n.shape)},b:()=>{let i=e,l=on(r.shape,s);return l.length>0&&(i=Te(i,l)),Z(i,r.shape)}}}},mJ={kernelName:Xi,saveAllInputs:!0,gradFunc:(e,t)=>{let n={};return t.forEach((r,s)=>{n[s]=()=>e.clone()}),n}},gJ={kernelName:Zi,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>ot(n)}}},yJ={kernelName:qp,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>ot(n)}}},AJ={kernelName:kc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Re(e,Ln(ke(De(1),Tt(xe(n,"float32")))))}}},xJ={kernelName:Ic,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>{let r=Ln(de(De(1),Tt(xe(n,"float32"))));return Re(e,r)}}}},bJ={kernelName:Nc,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t,s=$t(n.shape,r.shape);return{a:()=>{let i=de(Tt(n),Tt(r)),l=j(e,Re(r,i)),u=on(n.shape,s);return u.length>0&&(l=Te(l,u)),Z(l,n.shape)},b:()=>{let i=de(Tt(n),Tt(r)),l=qt(j(e,Re(n,i))),u=on(r.shape,s);return u.length>0&&(l=Te(l,u)),Z(l,r.shape)}}}},vJ={kernelName:Sc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Re(e,de(Tt(xe(n,"float32")),1))}}},wJ={kernelName:Tc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Re(e,ke(De(1),Tt(xe(n,"float32"))))}}};function kJ(e,t,n,r,s,a){let o=O(e,"dy","avgPool3dGrad"),i=O(t,"input","avgPool3dGrad"),l=o,u=i,c=!1;i.rank===4&&(c=!0,l=Z(o,[1,o.shape[0],o.shape[1],o.shape[2],o.shape[3]]),u=Z(i,[1,i.shape[0],i.shape[1],i.shape[2],i.shape[3]])),z(l.rank===5,()=>`Error in avgPool3dGrad: dy must be rank 5 but got rank ${l.rank}.`),z(u.rank===5,()=>`Error in avgPool3dGrad: input must be rank 5 but got rank ${u.rank}.`),a!=null&&z(fn(s),()=>`Error in avgPool3dGrad: pad must be an integer when using, dimRoundingMode ${a} but got pad ${s}.`);let d={dy:l,input:u},h={filterSize:n,strides:r,pad:s,dimRoundingMode:a},p=G.runKernel(x1,d,h);return c?Z(p,[p.shape[1],p.shape[2],p.shape[3],p.shape[4]]):p}var IJ=V({avgPool3dGrad_:kJ}),SJ={kernelName:Kp,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,{filterSize:s,strides:a,pad:o,dimRoundingMode:i}=n;return{x:()=>IJ(e,r,s,a,o,i)}}};function TJ(e,t,n,r,s){let a=O(e,"dy","avgPoolGrad"),o=O(t,"input","avgPoolGrad");z(o.rank===a.rank,()=>`Rank of input (${o.rank}) does not match rank of dy (${a.rank})`);let i=o,l=a,u=!1;o.rank===3&&(u=!0,i=Z(o,[1,o.shape[0],o.shape[1],o.shape[2]]),l=Z(a,[1,a.shape[0],a.shape[1],a.shape[2]])),z(l.rank===4,()=>`Error in avgPoolGrad: dy must be rank 4 but got rank ${l.rank}.`),z(i.rank===4,()=>`Error in avgPoolGrad: input must be rank 4 but got rank ${i.rank}.`);let c={dy:l,input:i},d={filterSize:n,strides:r,pad:s},h=G.runKernel(A1,c,d);return u?Z(h,[h.shape[1],h.shape[2],h.shape[3]]):h}var NJ=V({avgPoolGrad_:TJ}),CJ={kernelName:Yi,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,{filterSize:s,strides:a,pad:o}=n;return{x:()=>NJ(e,r,s,a,o)}}},EJ={kernelName:Ji,inputsToSave:["a","b"],gradFunc:(e,t,n)=>{let[r,s]=t,{transposeA:a,transposeB:o}=n;return!a&&!o?{a:()=>rt(e,s,!1,!0),b:()=>rt(r,e,!0,!1)}:!a&&o?{a:()=>rt(e,s,!1,!1),b:()=>rt(e,r,!0,!1)}:a&&!o?{a:()=>rt(s,e,!1,!0),b:()=>rt(r,e,!1,!1)}:{a:()=>rt(s,e,!0,!0),b:()=>rt(e,r,!0,!0)}}},$J={kernelName:Cc,gradFunc:(e,t,n)=>{let{blockShape:r,crops:s}=n;return{x:()=>Gy(e,r,s)}}},RJ={kernelName:qU,gradFunc:(e,t,n)=>{let r=n,s=r.inputShape,a=r.shape,o=Array.from(a);for(let l=s.length-1;l>=0;l--)if(s[l]===a[l])o[l]=1;else if(s[l]!==1)throw new Error(`broadcastTo(): [${s}] cannot be broadcast to [${a}].`);let i=[];for(let l=0;l<o.length;l++)o[l]>1&&i.push(l);return{x:()=>Te(e,i,!0)}}},_J={kernelName:Qi,gradFunc:e=>({x:()=>e.clone()})},DJ={kernelName:No,gradFunc:e=>({x:()=>ot(e)})},FJ={kernelName:Co,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,{clipValueMin:s,clipValueMax:a}=n;return{x:()=>Zn(_s(Kl(r,s),Xl(r,a)),e,ot(e))}}},MJ={kernelName:Xp,inputsToSave:["x"],gradFunc:_I.gradFunc},OJ={kernelName:Ec,saveAllInputs:!0,gradFunc:(e,t,n)=>{let r=t.map(l=>l.shape),{axis:s}=n,a=Vr(s,t[0].shape)[0],o=r.map(l=>l[a]);return Rr(e,o,a).map(l=>()=>l)}},PJ={kernelName:el,inputsToSave:["x","filter"],gradFunc:(e,t,n)=>{let[r,s]=t,{dilations:a,strides:o,pad:i,dataFormat:l}=n;return z(Wa(a),()=>`Error in gradient of conv2D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${a}'`),{x:()=>Fy(r.shape,e,s,o,i,l),filter:()=>eA(r,e,s.shape,o,i,l)}}},zJ={kernelName:tl,inputsToSave:["dy","filter"],gradFunc:(e,t,n)=>{let[r,s]=t,{strides:a,pad:o,dataFormat:i,dimRoundingMode:l}=n;return{dy:()=>Xo(e,s,a,o,i,1,l),filter:()=>eA(e,r,s.shape,a,o,i,l)}}};function LJ(e,t,n,r,s){let a=e;e.rank===4&&(a=Z(e,[1,e.shape[0],e.shape[1],e.shape[2],e.shape[3]]));let o=t;o.rank===4&&(o=Z(t,[1,t.shape[0],t.shape[1],t.shape[2],t.shape[3]])),z(a.rank===5,()=>`Error in conv3dDerFilter: input must be rank 5, but got shape ${a.shape}.`),z(o.rank===5,()=>`Error in conv3dDerFilter: dy must be rank 5, but got shape ${o.shape}.`),z(n.length===5,()=>`Error in conv3dDerFilter: filterShape must be length 5, but got ${n}.`),z(a.shape[4]===n[3],()=>`Error in conv3dDerFilter: depth of input ${a.shape[4]}) must match input depth in filter (${n[3]}.`),z(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:r,pad:s,filterShape:n};return G.runKernel(k1,i,l)}var BJ=V({conv3DBackpropFilter_:LJ}),WJ={kernelName:Zp,inputsToSave:["x","filter"],gradFunc:(e,t,n)=>{let{dilations:r,strides:s,pad:a}=n;z(Wa(r),()=>`Error in gradient of conv3D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${r}'`);let[o,i]=t;return{x:()=>j6(o.shape,e,i,s,a),filter:()=>BJ(o,e,i.shape,s,a)}}},VJ={kernelName:nl,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(qt(lI(xe(n,"float32"))),e)}}},UJ={kernelName:rl,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(uI(xe(n,"float32")),e)}}},HJ={kernelName:sl,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,{axis:s,exclusive:a,reverse:o}=n;return{x:()=>{let i=eI([s],r.rank),l=K6(e,s,a,!o);return i!=null&&(l=st(l,i)),l}}}},GJ={kernelName:al,inputsToSave:["x","filter"],gradFunc:(e,t,n)=>{let{dilations:r,strides:s,pad:a,dimRoundingMode:o}=n,i=r==null?[1,1]:r;z(Wa(i),()=>`Error in gradient of depthwiseConv2dNative: dilation rates greater than 1 are not yet supported. Got dilations '${i}'`);let[l,u]=t;return z(l.rank===4,()=>`Error in gradient of depthwiseConv2dNative: input must be rank 4, but got rank ${l.rank}.`),z(u.rank===4,()=>`Error in gradient of depthwiseConv2dNative: filter must be rank 4, but got rank ${u.rank}.`),z(l.shape[3]===u.shape[2],()=>`Error in gradient of depthwiseConv2d: number of input channels (${l.shape[3]}) must match the inChannels dimension in filter ${u.shape[2]}.`),z(Es(s,i),()=>`Error in gradient of depthwiseConv2d: Either strides or dilations must be 1. Got strides ${s} and dilations '${i}'.`),o!=null&&z(fn(a),()=>`Error in depthwiseConv2d: pad must be an integer when using, dimRoundingMode ${o} but got pad ${a}.`),{x:()=>vI(l.shape,e,u,s,a,r,o),filter:()=>bI(l,e,u.shape,s,a,r,o)}}},jJ={kernelName:Yp,inputsToSave:["x","filter"],gradFunc:(e,t,n)=>{let[r,s]=t,a={x:r,filter:s,dy:e},o={x:r,filter:s,dy:e};return{x:()=>G.runKernel(E1,a,n),filter:()=>G.runKernel($1,o,n)}}},qJ={kernelName:_c,outputsToSave:[!0],gradFunc:(e,t)=>{let[n]=t,r={dy:e,y:n};return{x:()=>G.runKernel(_1,r)}}},KJ={kernelName:Dc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t,r=j(Gr(qt(Tt(n))),2/Math.sqrt(Math.PI));return{x:()=>j(e,r)}}},XJ={kernelName:Eo,outputsToSave:[!0],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(e,n)}}},ZJ={kernelName:Fc,inputsToSave:["input"],gradFunc:(e,t)=>{let[n]=t;return{input:()=>Z(e,n.shape)}}},YJ={kernelName:ll,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(e,Gr(n))}}},JJ={kernelName:$o,gradFunc:e=>({x:()=>ot(e)})},QJ={kernelName:ul,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t,s=$t(n.shape,r.shape);return{a:()=>{let i=Re(e,xe(r,"float32")),l=on(n.shape,s);return l.length>0?Z(Te(i,l),n.shape):i},b:()=>{let i=j(e,xe(n,"float32")),l=on(r.shape,s);l.length>0&&(i=Z(Te(i,l),r.shape));let u=Tt(r);return qt(Re(i,xe(u,"float32")))}}}},eQ={kernelName:cl,inputsToSave:["x","mean","variance","scale"],gradFunc:(e,t,n)=>{let{varianceEpsilon:r}=n,[s,a,o,i]=t,l=i==null?De(1):i,u=on(a.shape,s.shape),c=[];if(a.rank===1){for(let b=0;b<s.shape.length-1;++b)c.push(s.shape[b]);c.push(1)}let d=ke(s,a),h=j(e,l),p=oI(de(o,De(r))),f=j(j(j(p,p),p),De(-.5));return{x:()=>a.rank===1?Z(j(j(e,$s(Z(p,[1,1,1,a.shape[0]]),c)),l),s.shape):Z(j(j(e,p),l),s.shape),mean:()=>{let b=j(j(p,De(-1)),h);return a.rank===1&&(b=Te(b,u)),Z(b,a.shape)},variance:()=>{let b=j(j(f,d),h);return a.rank===1&&(b=Te(b,u)),Z(b,a.shape)},scale:()=>{let b=j(d,p),v=j(e,b);return a.rank===1&&(v=Te(v,u)),Z(v,a.shape)},offset:()=>{let b=e;return a.rank===1&&(b=Te(b,u)),Z(b,a.shape)}}}},tQ={kernelName:Oc,inputsToSave:["x","indices"],gradFunc:(e,t,n)=>{let[r,s]=t,{axis:a}=n,o=Vr(a,r.shape)[0];return{x:()=>{let l=r.shape,u=s.size,c=l.slice(0,o),d=c.length,h=l.slice(a,l.length).slice(1),p=h.length,f=DI(0,d),m=DI(d+1,d+1+p),g=FI([c,[u],h]),y=Z(e,g),A=Z(s,[u]),x=FI([[d],f,m]),b=st(y,x),v=zX(b,A,r.shape[o]),I=By(x);return v=st(v,I),v},indices:()=>s}}};function DI(e,t){let n=[];for(let r=e;r<t;++r)n.push(r);return n}function FI(e){let t=[];for(let n=0;n<e.length;++n)for(let r=0;r<e[n].length;++r)t.push(e[n][r]);return t}var nQ={kernelName:Ro,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t;return{a:()=>ot(n),b:()=>ot(r)}}},rQ={kernelName:hl,gradFunc:e=>({x:()=>xe(e,"float32")})},sQ={kernelName:zc,gradFunc:e=>({x:()=>ot(e)})},aQ={kernelName:Lc,gradFunc:e=>({x:()=>ot(e)})},oQ={kernelName:Bc,gradFunc:e=>({x:()=>ot(e)})},iQ={kernelName:pl,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,{alpha:s}=n,a=$r(r,0);return{x:()=>Zn(a,e,j(e,s))}}},lQ={kernelName:Wc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Re(e,de(n,1))}}},uQ={kernelName:_o,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Re(e,xe(n,"float32"))}}},cQ={kernelName:KU,inputsToSave:[],outputsToSave:[!0],gradFunc:(e,t,n)=>{let[r]=t,{axis:s}=n;return{logits:()=>{let a=!0,o=Gr(r);return ke(e,j(Te(e,s,a),o))}}}};function dQ(e,t,n,r=5,s=1,a=1,o=.5){let i={x:e,y:t,dy:n},l={depthRadius:r,bias:s,alpha:a,beta:o};return G.runKernel(P1,i,l)}var hQ=V({localResponseNormalizationBackprop_:dQ}),pQ={kernelName:tf,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let[r,s]=t,{depthRadius:a,bias:o,alpha:i,beta:l}=n;return{x:()=>hQ(r,s,e,a,o,i,l)}}};function MI(e,t,n,r){return t.rank<n.rank&&(t=Z(t,Zo(t.shape,r))),e.rank<n.rank&&(e=Z(e,Zo(e.shape,r))),{x:()=>j(e,xe(Hr(n,t),e.dtype))}}var OI={kernelName:gl,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let r=n,{reductionIndices:s}=r,a=t[0],o=t[1],i=Vr(s,a.shape),l=MI(e,o,a,i);return{x:()=>l.x()}}},fQ={kernelName:Do,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t;return{a:()=>j(e,xe(Kl(n,r),"float32")),b:()=>j(e,xe(Z6(n,r),"float32"))}}};function mQ(e,t,n,r,s,a,o){let i=O(e,"dy","maxPool3dGrad"),l=O(t,"input","maxPool3dGrad"),u=O(n,"output","maxPool3dGrad"),c=i,d=l,h=u,p=!1;l.rank===4&&(p=!0,c=Z(i,[1,i.shape[0],i.shape[1],i.shape[2],i.shape[3]]),d=Z(l,[1,l.shape[0],l.shape[1],l.shape[2],l.shape[3]]),h=Z(u,[1,u.shape[0],u.shape[1],u.shape[2],u.shape[3]])),z(c.rank===5,()=>`Error in maxPool3dGrad: dy must be rank 5 but got rank ${c.rank}.`),z(d.rank===5,()=>`Error in maxPool3dGrad: input must be rank 5 but got rank ${d.rank}.`),z(h.rank===5,()=>`Error in maxPool3dGrad: output must be rank 5 but got rank ${h.rank}.`),o!=null&&z(fn(a),()=>`Error in maxPool3dGrad: pad must be an integer when using, dimRoundingMode ${o} but got pad ${a}.`);let f={dy:c,input:d,output:h},m={filterSize:r,strides:s,pad:a,dimRoundingMode:o},g=G.runKernel(L1,f,m);return p?Z(g,[g.shape[1],g.shape[2],g.shape[3],g.shape[4]]):g}var gQ=V({maxPool3dGrad_:mQ}),yQ={kernelName:nf,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let[r,s]=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=n;return{x:()=>gQ(e,r,s,a,o,i,l)}}};function AQ(e,t,n,r,s,a,o){let i=O(e,"dy","maxPoolGrad"),l=O(t,"input","maxPoolGrad"),u=O(n,"output","maxPoolGrad");z(l.rank===i.rank,()=>`Rank of input (${l.rank}) does not match rank of dy (${i.rank})`),z(i.rank===4,()=>`Error in maxPoolGrad: dy must be rank 4 but got rank ${i.rank}.`),z(l.rank===4,()=>`Error in maxPoolGrad: input must be rank 4 but got rank ${l.rank}.`),o!=null&&z(fn(a),()=>`Error in maxPoolGrad: pad must be an integer when using, dimRoundingMode ${o} but got pad ${a}.`);let c={dy:i,input:l,output:u},d={filterSize:r,strides:s,pad:a,dimRoundingMode:o};return G.runKernel(z1,c,d)}var xQ=V({maxPoolGrad_:AQ}),bQ={kernelName:yl,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let[r,s]=t,{filterSize:a,strides:o,pad:i}=n;return{x:()=>xQ(e,r,s,a,o,i)}}},vQ={kernelName:Al,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,{axis:s}=n,a=Vr(s,r.shape),i=Q6(r.shape,a)[1],l=sn(i);return{x:()=>{let c=r.shape.slice();a.forEach(p=>{c[p]=1});let d=Z(e,c);return Re(j(d,ua(r.shape,"float32")),l)}}}},wQ={kernelName:xl,inputsToSave:["x"],outputsToSave:[!0],gradFunc:(e,t,n)=>{let r=n,{axis:s}=r,[a,o]=t,i=Vr(s,a.shape),l=MI(e,o,a,i);return{x:()=>l.x()}}},kQ={kernelName:Fo,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t;return{a:()=>j(e,xe(Xl(n,r),"float32")),b:()=>j(e,xe($r(n,r),"float32"))}}},IQ={kernelName:bl,inputsToSave:["x"],gradFunc:(e,t,n)=>{let r=t[0],{paddings:s}=n,a=s.map(o=>o[0]);return{x:()=>at(e,a,r.shape)}}},SQ={kernelName:Uc,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t,s=$t(n.shape,r.shape);return{a:()=>{let i=on(n.shape,s);return i.length>0?Z(Te(e,i),n.shape):e},b:()=>{let i=j(e,qt(Sf(Re(n,r)))),l=on(r.shape,s);return l.length>0?Z(Te(i,l),r.shape):i}}}},TQ={kernelName:Mo,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t,s=$t(n.shape,r.shape);return{a:()=>{let i=j(e,xe(r,"float32")),l=on(n.shape,s);return l.length>0?Z(Te(i,l),n.shape):i},b:()=>{let i=j(e,xe(n,"float32")),l=on(r.shape,s);return l.length>0?Z(Te(i,l),r.shape):i}}}},NQ={kernelName:Hc,gradFunc:e=>({x:()=>qt(e)})},CQ={kernelName:wl,inputsToSave:["indices"],gradFunc:(e,t)=>{let n=t[0];return{indices:()=>ln(n.shape,"float32")}}},EQ={kernelName:Kc,gradFunc:e=>({x:()=>ot(e)})},$Q={kernelName:Xc,saveAllInputs:!0,gradFunc:(e,t,n)=>{let{axis:r}=n;return Ds(e,r).map(a=>()=>a)}},PI={kernelName:kl,inputsToSave:["x"],gradFunc:(e,t,n)=>{let r=t[0],{paddings:s}=n,a=s.map(o=>o[0]);return{x:()=>at(e,a,r.shape)}}},RQ={kernelName:Il,inputsToSave:["a","b"],outputsToSave:[!0],gradFunc:(e,t)=>{let[n,r,s]=t,a=n,o=r,i=$t(a.shape,o.shape);return{a:()=>{let c=xe(o,"float32"),d=j(e,j(c,Jo(a,ke(c,De(1))))),h=on(a.shape,i);return h.length>0&&(d=Te(d,h)),Z(d,a.shape)},b:()=>{let c=$r(a,0),d=Zn(c,jr(a),ot(a)),h=j(e,j(s,d)),p=on(o.shape,i);return p.length>0&&(h=Te(h,p)),Z(h,o.shape)}}}},_Q={kernelName:Sl,inputsToSave:["x","alpha"],gradFunc:(e,t)=>{let[n,r]=t,s=$r(n,0);return{x:()=>Zn(s,e,j(e,r)),alpha:()=>{let a=Zn(s,ot(e),j(e,n)),o=on(r.shape,e.shape);return o.length>0&&(a=Te(a,o)),Z(a,r.shape)}}}},DQ={kernelName:ol,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t,s=$t(n.shape,r.shape);return{a:()=>{let i=Re(e,xe(r,"float32")),l=on(n.shape,s);return l.length>0?Z(Te(i,l),n.shape):i},b:()=>{let i=j(e,xe(n,"float32")),l=on(r.shape,s);l.length>0&&(i=Z(Te(i,l),r.shape));let u=Tt(r);return qt(Re(i,xe(u,"float32")))}}}},FQ={kernelName:Yc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Re(e,qt(Tt(n)))}}},MQ={kernelName:Cl,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t,r=j(Xl(n,6),$f(n));return{x:()=>j(e,xe(r,"float32"))}}},OQ={kernelName:Tl,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(e,xe($f(n),"float32"))}}},PQ={kernelName:Jc,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Z(e,n.shape)}}},zQ={kernelName:Nl,inputsToSave:["images"],gradFunc:(e,t,n)=>{let[r]=t,s={dy:e,images:r};return{images:()=>G.runKernel(H1,s,n)}}},LQ={kernelName:sf,inputsToSave:["images"],gradFunc:(e,t,n)=>{let[r]=t,s={dy:e,images:r};return{images:()=>G.runKernel(U1,s,n)}}},BQ={kernelName:El,gradFunc:(e,t,n)=>{let{dims:r}=n,s=Vr(r,e.shape);return{x:()=>Kr(e,s)}}},WQ={kernelName:$l,gradFunc:e=>({x:()=>ot(e)})},VQ={kernelName:Oo,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>qt(Re(e,j(Jo(n,1.5),2)))}}},UQ={kernelName:ed,inputsToSave:["condition"],gradFunc:(e,t)=>{let[n]=t;return{condition:()=>xe(ot(n),"float32"),t:()=>j(e,xe(n,e.dtype)),e:()=>j(e,xe(Wy(n),e.dtype))}}},HQ={kernelName:td,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>{let r=$r(n,De(0)),s=De(NI),a=De(CI),o=j(e,a),i=j(j(e,s),Gr(xe(n,"float32")));return Zn(r,o,i)}}}},GQ={kernelName:_l,outputsToSave:[!0],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(e,j(n,ke(De(1),n)))}}},jQ={kernelName:sd,gradFunc:e=>({x:()=>ot(e)})},qQ={kernelName:Rl,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(My(xe(n,"float32")),e)}}},KQ={kernelName:rd,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(q6(xe(n,"float32")),e)}}},XQ={kernelName:nd,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,{begin:s,size:a}=n,o=r.shape,[i,l]=D6(r,s,a),u=[];for(let c=0;c<e.rank;c++)u.push([i[c],o[c]-i[c]-l[c]]);return{x:()=>Yo(e,u)}}},ZQ={kernelName:Ml,outputsToSave:[!0],gradFunc:(e,t,n)=>{let[r]=t,{dim:s}=n,a=!0,o=j(e,r);return{logits:()=>ke(o,j(Te(o,[s],a),r))}}},YQ={kernelName:ad,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(e,ia(n))}}},zI={kernelName:od,gradFunc:(e,t,n)=>{let{blockShape:r,paddings:s}=n;return{x:()=>Dy(e,r,s)}}},LI={kernelName:id,gradFunc:(e,t,n)=>{let{axis:r}=n;return{x:()=>an(e,r)}}},JQ={kernelName:Dl,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Re(e,j(Ln(xe(n,"float32")),2))}}},QQ={kernelName:af,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(e,j(xe(n,"float32"),2))}}},eee={kernelName:Po,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t,s=De(2);return{a:()=>j(e,j(s,ke(n,r))),b:()=>j(e,j(s,ke(r,n)))}}},tee={kernelName:Bo,gradFunc:e=>({x:()=>ot(e)})},nee={kernelName:zo,inputsToSave:["a","b"],gradFunc:(e,t)=>{let[n,r]=t,s=$t(n.shape,r.shape);return{a:()=>{let i=e,l=on(n.shape,s);return l.length>0&&(i=Te(i,l)),Z(i,n.shape)},b:()=>{let i=e,l=on(r.shape,s);return l.length>0&&(i=Te(i,l)),Z(qt(i),r.shape)}}}},ree={kernelName:Fl,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,s=r.shape.slice(),{axis:a}=n;Vr(a,r.shape).forEach(u=>{s[u]=1});let i=Z(e,s),l=j(i,ua(r.shape,"float32"));return{x:()=>l}}},see={kernelName:Ol,inputsToSave:["x"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>Re(e,Tt(My(n)))}}},aee={kernelName:Pl,outputsToSave:[!0],gradFunc:(e,t)=>{let[n]=t;return{x:()=>j(ke(De(1),Tt(n)),e)}}},oee={kernelName:Lo,inputsToSave:["x"],gradFunc:(e,t,n)=>{let[r]=t,{reps:s}=n;return{x:()=>{let o=ot(r);if(r.rank===1)for(let i=0;i<s[0];++i)o=de(o,at(e,[i*r.shape[0]],[r.shape[0]]));else if(r.rank===2)for(let i=0;i<s[0];++i)for(let l=0;l<s[1];++l)o=de(o,at(e,[i*r.shape[0],l*r.shape[1]],[r.shape[0],r.shape[1]]));else if(r.rank===3)for(let i=0;i<s[0];++i)for(let l=0;l<s[1];++l)for(let u=0;u<s[2];++u)o=de(o,at(e,[i*r.shape[0],l*r.shape[1],u*r.shape[2]],[r.shape[0],r.shape[1],r.shape[2]]));else if(r.rank===4)for(let i=0;i<s[0];++i)for(let l=0;l<s[1];++l)for(let u=0;u<s[2];++u)for(let c=0;c<s[3];++c)o=de(o,at(e,[i*r.shape[0],l*r.shape[1],u*r.shape[2],c*r.shape[3]],[r.shape[0],r.shape[1],r.shape[2],r.shape[3]]));else throw new Error(`Gradient for tile operation is not implemented for rank-${r.rank} tensors yet.`);return o}}}},iee={kernelName:zl,gradFunc:(e,t,n)=>{let r=n,{perm:s}=r,a=By(s);return{x:()=>st(e,a)}}},lee={kernelName:dd,gradFunc:(e,t,n)=>{let r=n,{axis:s}=r;return{value:()=>Xr(e,s)}}},uee={kernelName:of,inputsToSave:["segmentIds"],gradFunc:(e,t)=>{let[n]=t;return{x:()=>cee(e,n)}}};function cee(e,t){let n=Va(t,ot(t)),r=Tf(e,n),s=Kl(t,De(0,"int32")),a=r.rank-s.rank;for(let i=0;i<a;++i)s=Er(s,i+1);s=_s(s,ua(r.shape,"bool"));let o=ot(r);return Zn(s,r,o)}var dee={kernelName:hd,gradFunc:e=>({x:()=>ot(e)})},hee=[_I,hJ,pJ,fJ,mJ,gJ,yJ,AJ,xJ,bJ,vJ,wJ,SJ,CJ,EJ,$J,RJ,_J,DJ,FJ,MJ,OJ,zJ,PJ,WJ,VJ,UJ,HJ,GJ,jJ,DQ,qJ,KJ,XJ,ZJ,YJ,QJ,JJ,eQ,tQ,nQ,rQ,sQ,aQ,oQ,iQ,lQ,uQ,cQ,pQ,OI,OI,fQ,yQ,bQ,vQ,wQ,kQ,IQ,SQ,TQ,NQ,CQ,EQ,$Q,PI,PI,RQ,_Q,FQ,MQ,OQ,PQ,zQ,LQ,BQ,WQ,VQ,UQ,HQ,GQ,jQ,qQ,KQ,XQ,ZQ,YQ,zI,zI,LI,LI,JQ,eee,QQ,tee,nee,ree,see,aee,oee,iee,lee,uee,dee];for(let e of hee)XU(e);var BI={};_e(BI,{maxNorm:()=>gee,minMaxNorm:()=>xee,nonNeg:()=>Aee,unitNorm:()=>yee});var hA;function un(){return hA==null&&(hA=OG().epsilon()),hA}function ls(){return"channelsLast"}var ha=class extends Error{constructor(e){super(e);Object.setPrototypeOf(this,ha.prototype)}},us=class extends Error{constructor(e){super(e);Object.setPrototypeOf(this,us.prototype)}},K=class extends Error{constructor(e){super(e);Object.setPrototypeOf(this,K.prototype)}},He=class extends Error{constructor(e){super(e);Object.setPrototypeOf(this,He.prototype)}},WI=class extends Error{constructor(e){super(e);Object.setPrototypeOf(this,WI.prototype)}};function ni(e,t){if(Array.isArray(e)){let n=[];for(let r=0;r<t;r++)n=n.concat(e);return n}else{let n=new Array(t);return n.fill(e),n}}function Fs(e,t){if(!e)throw new WI(t)}function VI(e,t){let n=0;for(let r of e)r===t&&n++;return n}function Jn(e){return e.length===1?e[0]:e}function Rt(e){return Array.isArray(e)?e:[e]}function pa(e){let n=e.replace(/(.)([A-Z][a-z0-9]+)/g,"$1_$2").replace(/([a-z])([A-Z])/g,"$1_$2").toLowerCase();return n[0]!=="_"?n:"private"+n}function ri(e){return e.length<=1||e.indexOf("_")===-1?e:e.replace(/[_]+(\w|$)/g,(t,n)=>n.toUpperCase())}var Zr={};function pA(e){if(e==null)return null;let t={};return t.className=e.getClassName(),t.config=e.getConfig(),t}function fA(e){if(!(e==null||typeof e!="object"))if(Array.isArray(e))e.forEach(t=>fA(t));else{let t=Object.keys(e);for(let n of t){let r=e[n];r!=null&&typeof r=="object"&&(!Array.isArray(r)&&r.type==="ndarray"&&typeof r.value=="number"?e[n]=r.value:fA(r))}}}function _d(e,t={},n={},r="object",s=!1){if(typeof e=="string"){let a=e,o;if(a in n)o=n[a];else if(a in Zr)o=Zr[a];else if(o=t[a],o==null)throw new K(`Unknown ${r}: ${e}. This may be due to one of the following reasons:
1. The ${r} is defined in Python, in which case it needs to be ported to TensorFlow.js or your JavaScript code.
2. The custom ${r} is defined in JavaScript, but is not registered properly with tf.serialization.registerClass().`);return o}else{let a=e;if(a.className==null||a.config==null)throw new K(`${r}: 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 Zr?[i,l]=Zr.className:o in t&&([i,l]=t[o]),i==null)throw new K(`Unknown ${r}: ${o}. This may be due to one of the following reasons:
1. The ${r} is defined in Python, in which case it needs to be ported to TensorFlow.js or your JavaScript code.
2. The custom ${r} is defined in JavaScript, but is not registered properly with tf.serialization.registerClass().`);if(l!=null){let u={};for(let p of Object.keys(Zr))u[p]=Zr[p];for(let p of Object.keys(n))u[p]=n[p];let c=a.config;c.customObjects=u;let d={...Zr};for(let p of Object.keys(n))Zr[p]=n[p];fA(a.config);let h=l(i,a.config,n,s);return Zr={...d},h}else{let u={...Zr};for(let d of Object.keys(n))Zr[d]=n[d];let c=new i(a.config);return Zr={...u},c}}}function pee(e,t){return e<t?-1:e>t?1:0}function Pf(e,t){return-1*pee(e,t)}function Ga(e){if(e==null)return e;let t=[];for(let n of e)t.indexOf(n)===-1&&t.push(n);return t}function fee(e){if(e==null)throw new K(`Invalid value in obj: ${JSON.stringify(e)}`);for(let t in e)if(e.hasOwnProperty(t))return!1;return!0}function si(e,t,n){if(n!=null&&e.indexOf(n)<0)throw new K(`${n} is not a valid ${t}. Valid values are ${e} or null/undefined.`)}function mA(e,t,n=0,r=1/0){return Fs(n>=0),Fs(r>=n),Array.isArray(e)&&e.length>=n&&e.length<=r&&e.every(s=>typeof s===t)}function yn(e,t){Array.isArray(e)?(k.assert(e.length>0,()=>`${t} is unexpectedly an empty array.`),e.forEach((n,r)=>yn(n,`element ${r+1} of ${t}`))):k.assert(Number.isInteger(e)&&e>0,()=>`Expected ${t} to be a positive integer, but got ${UI(e)}.`)}function UI(e){return e===null?"null":Array.isArray(e)?"["+e.map(t=>UI(t)).join(",")+"]":typeof e=="string"?`"${e}"`:`${e}`}function mee(e,t){let n=k.now(),r;return(...a)=>{let o=k.now();return o-n<t||(n=o,r=e(...a)),r}}function HI(e){return e==="relu"?"relu":e==="linear"?"linear":e==="elu"?"elu":null}function gA(e,t){return Y(()=>Ln(Te(j(e,e),t,!0)))}var Dd=class extends ue.Serializable{getConfig(){return{}}},yA=class extends Dd{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=gA(e,this.axis),n=Cr(t,0,this.maxValue);return j(e,Re(n,de(un(),t)))})}getConfig(){return{maxValue:this.maxValue,axis:this.axis}}};yA.className="MaxNorm";ue.registerClass(yA);var AA=class extends Dd{constructor(e){super();this.defaultAxis=0,this.axis=e.axis!=null?e.axis:this.defaultAxis}apply(e){return Y(()=>Re(e,de(un(),gA(e,this.axis))))}getConfig(){return{axis:this.axis}}};AA.className="UnitNorm";ue.registerClass(AA);var xA=class extends Dd{apply(e){return ca(e)}};xA.className="NonNeg";ue.registerClass(xA);var bA=class extends Dd{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=gA(e,this.axis),n=de(j(this.rate,Cr(t,this.minValue,this.maxValue)),j(1-this.rate,t));return j(e,Re(n,de(un(),t)))})}getConfig(){return{minValue:this.minValue,maxValue:this.maxValue,rate:this.rate,axis:this.axis}}};bA.className="MinMaxNorm";ue.registerClass(bA);var GI={maxNorm:"MaxNorm",minMaxNorm:"MinMaxNorm",nonNeg:"NonNeg",unitNorm:"UnitNorm"};function cn(e){return pA(e)}function jI(e,t={}){return _d(e,ue.SerializationMap.getMap().classNameMap,t,"constraint")}function dn(e){if(e==null)return null;if(typeof e=="string"){let n={className:e in GI?GI[e]:e,config:{}};return jI(n)}else return e instanceof Dd?e:jI(e)}function gee(e){return new yA(e)}function yee(e){return new AA(e)}function Aee(){return new xA}function xee(e){return new bA(e)}var qI={};_e(qI,{constant:()=>Vee,glorotNormal:()=>Xee,glorotUniform:()=>Kee,heNormal:()=>Zee,heUniform:()=>Yee,identity:()=>jee,leCunNormal:()=>Jee,leCunUniform:()=>Qee,ones:()=>Wee,orthogonal:()=>ete,randomNormal:()=>Hee,randomUniform:()=>Uee,truncatedNormal:()=>Gee,varianceScaling:()=>qee,zeros:()=>Bee});var bee=["channelsFirst","channelsLast"],vee=["nearest","bilinear"],wee=["valid","same","causal"],kee=["max","avg"],Iee=["sum","mul","concat","ave"],Ql=new Map;function Zt(e){si(bee,"DataFormat",e)}function See(e){si(vee,"InterpolationFormat",e)}function _r(e){si(wee,"PaddingMode",e)}function KI(e){si(kee,"PoolMode",e)}var Fd=[],XI="/";function ai(e,t){Fd.push(e);try{let n=t();return Fd.pop(),n}catch(n){throw Fd.pop(),n}}function Tee(){return Fd.length===0?"":Fd.join(XI)+XI}function ZI(e){if(!JI(e))throw new Error("Not a valid tensor name: '"+e+"'");return Tee()+e}function YI(e){if(!JI(e))throw new Error("Not a valid tensor name: '"+e+"'");Ql.has(e)||Ql.set(e,0);let t=Ql.get(e);if(Ql.set(e,Ql.get(e)+1),t>0){let n=`${e}_${t}`;return Ql.set(n,1),n}else return e}var Nee=new RegExp(/^[A-Za-z0-9][-A-Za-z0-9\._\/]*$/);function JI(e){return!!e.match(Nee)}function Cee(e){return e===parseInt(e.toString(),10)}function ja(e,t,n){t==null&&(t=0),n==null&&(n=e.length);let r=1;for(let s=t;s<n;++s)r*=e[s];return r}function eu(e){if(e.length===0)return Number.NaN;let t=Number.POSITIVE_INFINITY;for(let n=0;n<e.length;n++){let r=e[n];r<t&&(t=r)}return t}function qa(e){if(e.length===0)return Number.NaN;let t=Number.NEGATIVE_INFINITY;for(let n=0;n<e.length;n++){let r=e[n];r>t&&(t=r)}return t}function cs(e,t){if(t<e)throw new K(`end (${t}) < begin (${e}) is forbidden.`);let n=[];for(let r=e;r<t;++r)n.push(r);return n}function zf(e,t){return xe(e,t)}function Md(e,t=-1){let n=e.shape.slice();return t<0&&(t=n.length+t+1),n.splice(t,0,1),Z(e,n)}function Eee(e,t){return Y(()=>{if(e.shape.length!==2)throw new K(`repeat() expects a rank-2 tensor, but received a rank-${e.shape.length} tensor.`);let n=Md(e,1);return kA(n,[1,t,1])})}function $ee(e){let t=[ja(e.shape)];return Z(e,t)}function Ree(e){if(e.rank<=1)throw new K(`batchFlatten requires a minimum rank of 2. Got rank: ${e.rank}.`);let t=[e.shape[0],ja(e.shape,1)];return Z(e,t)}function oi(e,t,n){return Y(()=>{switch(e.rank){case 1:return Xy(e,t,n);case 2:return cI(e,[t,0],[n,e.shape[1]]);case 3:return Zy(e,[t,0,0],[n,e.shape[1],e.shape[2]]);case 4:return Ef(e,[t,0,0,0],[n,e.shape[1],e.shape[2],e.shape[3]]);case 5:return at(e,[t,0,0,0,0],[n,e.shape[1],e.shape[2],e.shape[3],e.shape[4]]);case 6:return at(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 K(`sliceAlongFirstAxis() received an unsupported tensor rank: ${e.rank}`)}})}function vA(e,t,n){return Y(()=>{switch(e.rank){case 1:return Xy(e,t,n);case 2:return cI(e,[0,t],[e.shape[0],n]);case 3:return Zy(e,[0,0,t],[e.shape[0],e.shape[1],n]);case 4:return Ef(e,[0,0,0,t],[e.shape[0],e.shape[1],e.shape[2],n]);default:throw new K(`sliceAlongLastAxis() received an unsupported tensor rank: ${e.rank}`)}})}function Lf(e,t,n,r){return Y(()=>{switch(e.rank){case 1:return Xy(e,t,n);case 2:switch(r){case 1:return oi(e,t,n);case 2:return vA(e,t,n);default:throw new K(`The axis is not within the rank of the tensor ${r}`)}case 3:switch(r){case 1:return oi(e,t,n);case 2:return Zy(e,[0,t,0],[e.shape[0],n,e.shape[2]]);case 3:return vA(e,t,n);default:throw new K(`The axis is not within the rank of the tensor ${r}`)}case 4:switch(r){case 1:return oi(e,t,n);case 2:return Ef(e,[0,t,0,0],[e.shape[0],n,e.shape[2],e.shape[3]]);case 3:return Ef(e,[0,0,t,0],[e.shape[0],e.shape[1],n,e.shape[3]]);case 4:return vA(e,t,n);default:throw new K(`The axis is not within the rank of the tensor ${r}`)}default:throw new K(`sliceAlongLastAxis() received an unsupported tensor rank: ${e.rank}`)}})}function wA(e,t=-1){let n;return t<0&&(n=e[0].rank,n!==0?t=n:t=0),t===e[0].rank&&(t=-1),an(e,t)}function QI(e,t){switch(e.rank){case 1:return zj([e,t]);case 2:return Bj([e,t],0);case 3:return Vj([e,t],0);case 4:return Hj([e,t],0);default:throw new K(`concatAlongFirstAxis() received an unsupported tensor rank: ${e.rank}`)}}function kA(e,t){if(Array.isArray(t)||(t=[t]),e.rank!==t.length)throw new K(`The length of input n (${t.length}) does not match the number of dimensions in input x (${e.rank})`);return $s(e,t)}function Bf(e,t=0,n=1,r,s){return HK(e,t,n,r,s)}function Ms(e,t,n,r){if(e.rank<2||t.rank<2)throw new He(`dot requires both inputs to be rank >= 2 but got x shape = ${e.shape} and y shape = ${t.shape}`);if(t.rank>=3){let s=e.shape.slice(-1)[0],a=t.shape.slice(-2)[0];if(s!==a)throw new He(`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){let s=!1,a=!1;return ei.matMul({a:e,b:t,transposeA:s,transposeB:a,bias:r?IA(e.rank,r,ls()):null,activation:n})}else{let s=e.shape.slice(),a=s.pop();e=Z(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=Z(st(t,c),[l,-1]);let d=[...s,...u],h=!1,p=!1;return Z(ei.matMul({a:e,b:t,transposeA:h,transposeB:p,bias:r?IA(e.rank,r,ls()):null,activation:n}),d)}}function eS(e,t,n){return Y(()=>(Array.isArray(t)?t=En(t,"int32"):t=xe(t,"int32"),Tf(e,t,n)))}function Od(e){return j(e,e)}function IA(e,t,n){let r=t.shape;if(t.rank!==1&&t.rank!==e)throw new K(`Unexpected bias dimensions: ${t.rank}; expected it to be 1 or ${e}`);if(e===5){if(n==="channelsFirst")return r.length===1?Z(t,[1,r[0],1,1,1]):Z(t,[1,r[3],r[0],r[1],r[2]]);if(n==="channelsLast")return r.length===1?Z(t,[1,1,1,1,r[0]]):Z(t,[1].concat(r))}else if(e===4){if(n==="channelsFirst")return r.length===1?Z(t,[1,r[0],1,1]):Z(t,[1,r[2],r[0],r[1]]);if(n==="channelsLast")return r.length===1?Z(t,[1,1,1,r[0]]):Z(t,[1].concat(r))}else if(e===3){if(n==="channelsFirst")return r.length===1?Z(t,[1,r[0],1]):Z(t,[1,r[1],r[0]]);if(n==="channelsLast")return r.length===1?Z(t,[1,1,r[0]]):Z(t,[1].concat(r))}else if(e<3)return t;throw new K(`Unsupported input rank by biasAdd: ${t.rank}`)}function ds(e,t,n){return Y(()=>(n==null&&(n=ls()),Zt(n),de(e,IA(e.rank,t,n))))}function _ee(e,t=1){if(t!==1)throw new He(`Support for alpha values other than 1 (${t}) is not implemented yet.`);return If(e)}function Dee(e){return Y(()=>Re(e,de(gn(e),1)))}function tS(e,t,n,r){return Y(()=>QX(e,t,n,r))}function Fee(e){return Y(()=>{let t=de(.5,j(.2,e));return Cr(t,0,1)})}function Pd(e,t,n=!1){return n?e():t()}var Mee=["fanIn","fanOut","fanAvg"],Oee=["normal","uniform","truncatedNormal"];function Pee(e){si(Mee,"FanMode",e)}function zee(e){si(Oee,"Distribution",e)}var Yr=class extends ue.Serializable{fromConfigUsesCustomObjects(){return!1}getConfig(){return{}}},SA=class extends Yr{apply(e,t){return ln(e,t)}};SA.className="Zeros";ue.registerClass(SA);var Wf=class extends Yr{apply(e,t){return ua(e,t)}};Wf.className="Ones";ue.registerClass(Wf);var TA=class extends Yr{constructor(e){super();if(typeof e!="object")throw new K(`Expected argument of type ConstantConfig but got ${e}`);if(e.value===void 0)throw new K(`config must have value set but got ${e}`);this.value=e.value}apply(e,t){return Y(()=>j(De(this.value),ua(e,t)))}getConfig(){return{value:this.value}}};TA.className="Constant";ue.registerClass(TA);var NA=class extends Yr{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 $d(e,this.minval,this.maxval,t)}getConfig(){return{minval:this.minval,maxval:this.maxval,seed:this.seed}}};NA.className="RandomUniform";ue.registerClass(NA);var CA=class extends Yr{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 He(`randomNormal does not support dType ${t}.`);return Bf(e,this.mean,this.stddev,t,this.seed)}getConfig(){return{mean:this.mean,stddev:this.stddev,seed:this.seed}}};CA.className="RandomNormal";ue.registerClass(CA);var EA=class extends Yr{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 He(`truncatedNormal does not support dType ${t}.`);return Qy(e,this.mean,this.stddev,t,this.seed)}getConfig(){return{mean:this.mean,stddev:this.stddev,seed:this.seed}}};EA.className="TruncatedNormal";ue.registerClass(EA);var $A=class extends Yr{constructor(e){super();this.gain=e.gain!=null?e.gain:1}apply(e,t){return Y(()=>{if(e.length!==2||e[0]!==e[1])throw new K("Identity matrix initializer can only be used for 2D square matrices.");return j(this.gain,X6(e[0]))})}getConfig(){return{gain:this.gain}}};$A.className="Identity";ue.registerClass($A);function Lee(e,t="channelsLast"){let n,r;if(Zt(t),e.length===2)n=e[0],r=e[1];else if([3,4,5].indexOf(e.length)!==-1){if(t==="channelsFirst"){let s=ja(e,2);n=e[1]*s,r=e[0]*s}else if(t==="channelsLast"){let s=ja(e,0,e.length-2);n=e[e.length-2]*s,r=e[e.length-1]*s}}else{let s=ja(e);n=Math.sqrt(s),r=Math.sqrt(s)}return[n,r]}var Qn=class extends Yr{constructor(e){super();if(e.scale<0)throw new K(`scale must be a positive float. Got: ${e.scale}`);this.scale=e.scale==null?1:e.scale,this.mode=e.mode==null?"fanIn":e.mode,Pee(this.mode),this.distribution=e.distribution==null?"normal":e.distribution,zee(this.distribution),this.seed=e.seed}apply(e,t){let n=Lee(e),r=n[0],s=n[1],a=this.scale;if(this.mode==="fanIn"?a/=Math.max(1,r):this.mode==="fanOut"?a/=Math.max(1,s):a/=Math.max(1,(r+s)/2),this.distribution==="normal"){let o=Math.sqrt(a);if(t=t||"float32",t!=="float32"&&t!=="int32")throw new He(`${this.getClassName()} does not support dType ${t}.`);return Qy(e,0,o,t,this.seed)}else{let o=Math.sqrt(3*a);return $d(e,-o,o,t)}}getConfig(){return{scale:this.scale,mode:this.mode,distribution:this.distribution,seed:this.seed}}};Qn.className="VarianceScaling";ue.registerClass(Qn);var Vf=class extends Qn{constructor(e){super({scale:1,mode:"fanAvg",distribution:"uniform",seed:e==null?null:e.seed})}getClassName(){return Qn.className}};Vf.className="GlorotUniform";ue.registerClass(Vf);var Uf=class extends Qn{constructor(e){super({scale:1,mode:"fanAvg",distribution:"normal",seed:e==null?null:e.seed})}getClassName(){return Qn.className}};Uf.className="GlorotNormal";ue.registerClass(Uf);var Hf=class extends Qn{constructor(e){super({scale:2,mode:"fanIn",distribution:"normal",seed:e==null?null:e.seed})}getClassName(){return Qn.className}};Hf.className="HeNormal";ue.registerClass(Hf);var Gf=class extends Qn{constructor(e){super({scale:2,mode:"fanIn",distribution:"uniform",seed:e==null?null:e.seed})}getClassName(){return Qn.className}};Gf.className="HeUniform";ue.registerClass(Gf);var jf=class extends Qn{constructor(e){super({scale:1,mode:"fanIn",distribution:"normal",seed:e==null?null:e.seed})}getClassName(){return Qn.className}};jf.className="LeCunNormal";ue.registerClass(jf);var qf=class extends Qn{constructor(e){super({scale:1,mode:"fanIn",distribution:"uniform",seed:e==null?null:e.seed})}getClassName(){return Qn.className}};qf.className="LeCunNormal";ue.registerClass(qf);var RA=class extends Yr{constructor(e){super();if(this.DEFAULT_GAIN=1,this.gain=e.gain==null?this.DEFAULT_GAIN:e.gain,this.seed=e.seed,this.seed!=null)throw new He("Random seed is not implemented for Orthogonal Initializer yet.")}apply(e,t){return Y(()=>{if(e.length<2)throw new He("Shape must be at least 2D.");e[0]*e[1]>2e3&&console.warn(`Orthogonal initializer is being called on a matrix with more than 2000 (${e[0]*e[1]}) elements: Slowness may result.`);let n=e[0]>e[1]?[e[1],e[0]]:e,r=Bf(n,0,1,"float32"),s=TY.gramSchmidt(r);return e[0]>e[1]&&(s=st(s)),j(this.gain,s)})}getConfig(){return{gain:this.gain,seed:this.seed}}};RA.className="Orthogonal";ue.registerClass(RA);var nS={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 rS(e,t={}){return _d(e,ue.SerializationMap.getMap().classNameMap,t,"initializer")}function Ut(e){return pA(e)}function Pt(e){if(typeof e=="string"){let t=e in nS?nS[e]:e;if(t==="GlorotNormal")return new Uf;if(t==="GlorotUniform")return new Vf;if(t==="HeNormal")return new Hf;if(t==="HeUniform")return new Gf;if(t==="LeCunNormal")return new jf;if(t==="LeCunUniform")return new qf;{let n={};return n.className=t,n.config={},rS(n)}}else return e instanceof Yr?e:rS(e)}function Bee(){return new SA}function Wee(){return new Wf}function Vee(e){return new TA(e)}function Uee(e){return new NA(e)}function Hee(e){return new CA(e)}function Gee(e){return new EA(e)}function jee(e){return new $A(e)}function qee(e){return new Qn(e)}function Kee(e){return new Vf(e)}function Xee(e){return new Uf(e)}function Zee(e){return new Hf(e)}function Yee(e){return new Gf(e)}function Jee(e){return new jf(e)}function Qee(e){return new qf(e)}function ete(e){return new RA(e)}var sS={};_e(sS,{Layer:()=>tt,RNN:()=>ma,RNNCell:()=>Hd,activation:()=>Mne,add:()=>Hne,alphaDropout:()=>Nre,average:()=>Gne,averagePooling1d:()=>Jx,averagePooling2d:()=>Qx,averagePooling3d:()=>e5,avgPool1d:()=>ere,avgPool2d:()=>nre,avgPool3d:()=>sre,avgPooling1d:()=>tre,avgPooling2d:()=>rre,avgPooling3d:()=>are,batchNormalization:()=>Yne,bidirectional:()=>xre,concatenate:()=>jne,conv1d:()=>Tne,conv2d:()=>Nne,conv2dTranspose:()=>Cne,conv3d:()=>Ene,conv3dTranspose:()=>$ne,convLstm2d:()=>mre,convLstm2dCell:()=>gre,cropping2D:()=>_ne,dense:()=>One,depthwiseConv2d:()=>Fne,dot:()=>Zne,dropout:()=>Pne,elu:()=>bne,embedding:()=>Une,flatten:()=>Lne,gaussianDropout:()=>Tre,gaussianNoise:()=>Sre,globalAveragePooling1d:()=>ore,globalAveragePooling2d:()=>ire,globalMaxPool1d:()=>vre,globalMaxPool2d:()=>wre,globalMaxPooling1d:()=>y8,globalMaxPooling2d:()=>A8,gru:()=>ure,gruCell:()=>cre,input:()=>PS,inputLayer:()=>xne,layerNormalization:()=>Jne,leakyReLU:()=>wne,lstm:()=>dre,lstmCell:()=>hre,masking:()=>Cre,maxPool1d:()=>kre,maxPool2d:()=>Ire,maxPooling1d:()=>x8,maxPooling2d:()=>b8,maxPooling3d:()=>lre,maximum:()=>qne,minimum:()=>Kne,multiply:()=>Xne,permute:()=>Vne,prelu:()=>kne,reLU:()=>vne,repeatVector:()=>Bne,reshape:()=>Wne,rnn:()=>yre,separableConv2d:()=>Rne,simpleRNN:()=>pre,simpleRNNCell:()=>fre,softmax:()=>Ine,spatialDropout1d:()=>zne,stackedRNNCells:()=>Are,thresholdedReLU:()=>Sne,timeDistributed:()=>bre,upSampling2d:()=>Dne,zeroPadding2d:()=>Qne});var tte=0;function aS(){return tte++}var Kf={};function Xf(e=""){return e in Kf||(Kf[e]=0),Kf[e]+=1,e+Kf[e].toString()}function _A(e){return Array.isArray(e)&&Array.isArray(e[0])}function Zf(e){return e.length===0?[]:Array.isArray(e[0])?e:[e]}function qe(e){let t;if(Array.isArray(e)){if(e.length!==1)throw new K(`Expected Tensor length to be 1; got ${e.length}`);t=e[0]}else t=e;return t}function yt(e){if(Array.isArray(e)&&Array.isArray(e[0])){if(e.length===1)return e=e,e[0];throw new K(`Expected exactly 1 Shape; got ${e.length}`)}else return e}function Yf(e){let t=0;for(let n of e)n.shape.length===0?t+=1:t+=n.shape.reduce((r,s)=>r*s);return t}var oS="Variable",iS=class{constructor(e,t="float32",n=oS,r=!0,s=null){this.dtype=t==null?"float32":t,this.shape=e.shape,this.id=aS(),n=n==null?oS:n,this.originalName=ZI(n),this.name=YI(this.originalName),this.trainable_=r,this.constraint=s,this.val=BX(e,this.trainable_,this.name,this.dtype)}read(){return this.assertNotDisposed(),this.val}write(e){return this.assertNotDisposed(),nte(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 nte(e,t){if(e.shape.toString()!==t.shape.toString())throw new Error("Shape mismatch: "+JSON.stringify(e.shape)+" vs. "+JSON.stringify(t.shape))}function DA(e){return e.map(t=>t.read())}function FA(e){e.forEach(t=>{t[0].write(t[1])})}var Qt=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||{}}},hs=class{constructor(e,t,n,r,s,a,o){this.dtype=e,this.shape=t,this.sourceLayer=n,this.inputs=r,this.callArgs=s,this.outputTensorIndex=o,this.id=aS(),a!=null&&(this.originalName=ZI(a),this.name=YI(this.originalName)),this.rank=t.length}},rte=0,Jf=class{constructor(e,t){this.callArgs=t,this.id=rte++,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}}},ste=0,tt=class extends ue.Serializable{constructor(e={}){super();this._callHook=null,this._addedWeightNames=[],this._stateful=!1,this.id=ste++,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=pa(n)+"_"+Xf(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 s=null;e.batchSize!=null&&(s=e.batchSize),n=[s].concat(e.inputShape)}this.batchInputShape=n;let r=e.dtype;r==null&&(r=e.inputDType),r==null&&(r="float32"),this.dtype=r}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 us(`The layer has never been called and thus has no defined ${t}.`);if(this.inboundNodes.length<=e)throw new K(`Asked to get ${t} at node ${e}, but the layer has only ${this.inboundNodes.length} inbound nodes.`);return this.inboundNodes[e]}getInputAt(e){return Jn(this.getNodeAtIndex(e,"input").inputTensors)}getOutputAt(e){return Jn(this.getNodeAtIndex(e,"output").outputTensors)}get input(){if(this.inboundNodes.length>1)throw new ha(`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 ha(`Layer ${this.name} is not connected, no input to return.`);return Jn(this.getNodeAtIndex(0,"input").inputTensors)}get output(){if(this.inboundNodes.length===0)throw new ha(`Layer ${this.name} has no inbound nodes.`);if(this.inboundNodes.length>1)throw new ha(`Layer ${this.name} has multiple inbound nodes, hence the notion of "layer output" is ill-defined. Use \`getOutputAt(nodeIndex)\` instead.`);return Jn(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=Rt(e),this.inputSpec==null||this.inputSpec.length===0)return;let t=Rt(this.inputSpec);if(e.length!==t.length)throw new K(`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 r=e[n],s=t[n];if(s==null)continue;let a=r.rank;if(s.ndim!=null&&a!==s.ndim)throw new K(`Input ${n} is incompatible with layer ${this.name}: expected ndim=${s.ndim}, found ndim=${a}`);if(s.maxNDim!=null&&a>s.maxNDim)throw new K(`Input ${n} is incompatible with layer ${this.name}: expected max_ndim=${s.maxNDim}, found ndim=${a}`);if(s.minNDim!=null&&a<s.minNDim)throw new K(`Input ${n} is incompatible with layer ${this.name}: expected min_ndim=${s.minNDim}, found ndim=${a}.`);if(s.dtype!=null&&r.dtype!==s.dtype)throw new K(`Input ${n} is incompatible with layer ${this.name} : expected dtype=${s.dtype}, found dtype=${r.dtype}.`);if(s.axes){let o=r.shape;for(let i in s.axes){let l=Number(i),u=s.axes[i],c=l>=0?o[l]:o[o.length+l];if(u!=null&&[u,null].indexOf(c)===-1)throw new K(`Input ${n} is incompatible with layer ${this.name}: expected axis ${l} of input shape to have value ${u} but got shape ${o}.`)}}if(s.shape!=null)for(let o=0;o<s.shape.length;++o){let i=s.shape[o],l=r.shape[o];if(i!=null&&l!=null&&i!==l)throw new K(`Input ${n} is incompatible with layer ${this.name}: expected shape=${s.shape}, found shape=${r.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=Rt(e),r=!0;for(let a of n)if(!(a instanceof hs)){r=!1;break}let s=!0;for(let a of n)if(a instanceof hs){s=!1;break}if(r===s)throw new K("Arguments to apply() must be all SymbolicTensors or all Tensors");return ai(this.name,()=>{if(!this.built){this.assertInputCompatibility(e);let a=[];for(let o of Rt(e))a.push(o.shape);this.build(Jn(a)),this.built=!0,this.initialWeights&&this.setWeights(this.initialWeights),this._refCount===null&&s&&(this._refCount=1)}if(this.assertInputCompatibility(e),s){let a=this.call(e,t),o=Rt(a),i=[];for(let l of o)n.indexOf(l)!==-1&&(l=l.clone()),i.push(l);if(a=Jn(i),this.activityRegularizer!=null)throw new He("Layer invocation in the presence of activity regularizer(s) is not supported yet.");return a}else{let a=ate(e),o=this.computeOutputShape(a),i,l=ote(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 hs(l,u,this,Rt(e),t,this.name,c)):i=new hs(l,o,this,Rt(e),t,this.name),this.addInboundNode(e,i,null,null,a,o,t),this._refCount++,this.activityRegularizer!=null)throw new He("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 ${this.name}`);else{let t=!1;this.batchInputShape.forEach((n,r)=>{n!=null&&e[r]!=null&&e[r]!==n&&(t=!0)}),t&&console.warn(`The shape of the input tensor (${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 ha(`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 ha(`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 us(`You tried to call countParams() on ${this.name}, but the layer is not built yet. Build it first by calling build(batchInputShape).`);return Yf(this.weights)}build(e){this.built=!0}getWeights(e=!1){return DA(e?this.trainableWeights:this.weights)}setWeights(e){Y(()=>{let t=this.weights;if(t.length!==e.length)throw new K(`You called setWeights(weights) on layer "${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=[],r=DA(t);for(let s=0;s<r.length;++s){let a=r[s],o=t[s],i=e[s];if(!k.arraysEqual(a.shape,i.shape))throw new K(`Layer weight shape ${a.shape} not compatible with provided weight shape ${i.shape}`);n.push([o,i])}FA(n)})}addWeight(e,t,n,r,s,a,o){if(this._addedWeightNames.indexOf(e)!==-1)throw new K(`Duplicate weight name ${e} for layer ${this.name}`);this._addedWeightNames.push(e),n==null&&(n="float32"),this.fastWeightInitDuringBuild&&(r=Pt("zeros"));let i=r.apply(t,n),l=new iS(i,n,e,a,o);return i.dispose(),s!=null&&this.addLoss(()=>s.apply(l.read())),a==null&&(a=!0),a?this._trainableWeights.push(l):this._nonTrainableWeights.push(l),l}setFastWeightInitDuringBuild(e){this.fastWeightInitDuringBuild=e}addLoss(e){e==null||Array.isArray(e)&&e.length===0||(e=Rt(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 ${this.name} does not support masking, but was passed an inputMask.`);return null}return t}addInboundNode(e,t,n,r,s,a,o=null){let i=Rt(e);t=Rt(t),n=Rt(n),r=Rt(r),s=Zf(s),a=Zf(a);let l=[],u=[],c=[];for(let d of i)l.push(d.sourceLayer),u.push(d.nodeIndex),c.push(d.tensorIndex);new Jf({outboundLayer:this,inboundLayers:l,nodeIndices:u,tensorIndices:c,inputTensors:i,outputTensors:t,inputMasks:n,outputMasks:r,inputShapes:s,outputShapes:a},o);for(let d=0;d<t.length;d++)t[d].sourceLayer=this,t[d].nodeIndex=this.inboundNodes.length-1,t[d].tensorIndex=d}getConfig(){let e={name:this.name,trainable:this.trainable};return this.batchInputShape!=null&&(e.batchInputShape=this.batchInputShape),this.dtype!=null&&(e.dtype=this.dtype),e}disposeWeights(){return this.weights.forEach(e=>e.dispose()),this.weights.length}assertNotDisposed(){if(this._refCount===0)throw new Error(`Layer '${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 ate(e){e=Rt(e);let t=[];for(let n of e)t.push(n.shape);return Jn(t)}function ote(e){return"float32"}function lS(e,t,n){if((t==null||n!=null&&n>0)&&(t=e.sourceLayer,n=e.nodeIndex),t.inboundNodes.length===0)return[e];{let r=t.inboundNodes[n];if(r.inboundLayers.length===0)return r.inputTensors;{let s=[];for(let a=0;a<r.inboundLayers.length;a++){let o=r.inputTensors[a],i=r.inboundLayers[a],l=r.nodeIndices[a],u=lS(o,i,l);for(let c of u)s.indexOf(c)===-1&&s.push(c)}return s}}}var tu=class extends tt{constructor(e){super({dtype:e.dtype,name:e.name!=null?e.name:Xf("input").toString()});if(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 K("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 K("An InputLayer should be passed either a `batchInputShape` or an `inputShape`.");t=[e.batchSize].concat(e.inputShape)}else if(e.batchSize!=null)throw new K("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 r=new hs(this.dtype,this.batchInputShape,this,[],{},this.name);r.nodeIndex=0,r.tensorIndex=0,new Jf({outboundLayer:this,inboundLayers:[],nodeIndices:[],tensorIndices:[],inputTensors:[r],outputTensors:[r],inputMasks:[null],outputMasks:[null],inputShapes:[t],outputShapes:[t]})}apply(e,t){throw new K(`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}}};tu.className="InputLayer";ue.registerClass(tu);function uS(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 K("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 tu({batchInputShape:t,name:e.name,dtype:n,sparse:e.sparse}).inboundNodes[0].outputTensors[0]}async function Ka(e){if(e==null)return;let t=[],n=[],r=[];for(let s in e){let a=e[s];if(typeof a!="number"){let o=a;t.push(o.data()),n.push(s),r.push(o)}}if(t.length>0){let s=await Promise.all(t);for(let a=0;a<s.length;++a)e[n[a]]=s[a][0];Ge(r)}}function cS(e){if(e!=null)for(let t in e){let n=e[t];typeof n!="number"&&n.dispose()}}var dS;(function(e){e[e.SILENT=0]="SILENT",e[e.VERBOSE=1]="VERBOSE"})(dS||(dS={}));var ite=125,nu=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){}},hS=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)}},lte=class extends nu{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 r in t){let s=t[r];if(typeof s=="number")this.totals.hasOwnProperty(r)||(this.totals[r]=0),this.totals[r]=this.totals[r]+s*n;else{let a;r in this.totals?a=this.totals[r]:this.totals[r]=0;let o=Y(()=>de(this.totals[r],j(s,n)));this.totals[r]=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 r=j(Re(1,this.seen),this.totals[n]);t[n]=r,this.totals[n].dispose(),In(t[n])}))}},pS=class extends nu{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 s in this.history){let a=this.history[s];for(let o=0;o<a.length;++o)if(typeof a[o]!="number"){let i=a[o];e.push(i.data()),t.push(s),n.push(o)}}let r=await Promise.all(e);for(let s=0;s<r.length;++s)this.history[t[s]][n[s]].dispose(),this.history[t[s]][n[s]]=r[s][0]}},fS=class extends nu{constructor(e,t){super();if(this.currentEpoch=0,this.yieldEvery=t||"auto",this.yieldEvery==="auto"&&(this.yieldEvery=ite),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");k.isNumber(this.yieldEvery)&&(this.maybeWait=mee(this.maybeWait.bind(this),this.yieldEvery)),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 r=[];this.yield!=null&&(await Ka(n),r.push(this.yield(e,t,n))),r.push(uA()),await Promise.all(r)}async onEpochBegin(e,t){this.currentEpoch=e,this.epochBegin!=null&&(await Ka(t),await this.epochBegin(e,t))}async onEpochEnd(e,t){let n=[];this.epochEnd!=null&&(await Ka(t),n.push(this.epochEnd(e,t))),this.yieldEvery==="epoch"&&n.push(uA()),await Promise.all(n)}async onBatchBegin(e,t){this.batchBegin!=null&&(await Ka(t),await this.batchBegin(e,t))}async onBatchEnd(e,t){let n=[];this.batchEnd!=null&&(await Ka(t),n.push(this.batchEnd(e,t))),this.yieldEvery==="batch"?n.push(uA()):k.isNumber(this.yieldEvery)&&n.push(this.maybeWait(this.currentEpoch,e,t)),await Promise.all(n)}async onTrainBegin(e){this.trainBegin!=null&&(await Ka(e),await this.trainBegin(e))}async onTrainEnd(e){this.trainEnd!=null&&(await Ka(e),await this.trainEnd(e))}};function mS(e,t){return e==null&&(e={}),e instanceof nu?[e]:Array.isArray(e)&&e[0]instanceof nu?e:Rt(e).map(r=>new fS(r,t))}var Os=class{constructor(){}static registerCallbackConstructor(e,t){k.assert(e>=0&&Number.isInteger(e),()=>`Verbosity level is expected to be an integer >= 0, but got ${e}`),Os.checkForDuplicate(t),Os.constructors[e]==null&&(Os.constructors[e]=[]),Os.constructors[e].push(t)}static checkForDuplicate(e){for(let t in Os.constructors)Os.constructors[+t].forEach(r=>{if(r===e)throw new K("Duplicate callback constructor.")})}static clear(){Os.constructors={}}static createCallbacks(e){let t=[];for(let n in Os.constructors){let r=+n;e>=r&&t.push(...Os.constructors[r])}return t.map(n=>new n)}},MA=Os;MA.constructors={};function gS(e,t,n,r,s,a,o,i,l){let u=new pS,c=[new lte,...MA.createCallbacks(t)];e!=null&&c.push(...e),c.push(u);let d=new hS(c);return d.setParams({epochs:n,initialEpoch:r,samples:s,steps:a,batchSize:o,verbose:t,doValidation:i,metrics:l}),{callbackList:d,history:u}}function ps(e,t={},n=!1){return _d(e,ue.SerializationMap.getMap().classNameMap,t,"layer",n)}function Qf(e,t){return Y(()=>{e.dtype!=="float32"&&(e=xe(e,"float32"));let n=Te(Od(e),t,!0),r=Nd(n.shape,un()),s=Ln(Va(n,r));return Re(e,s)})}function ii(e,t){return Y(()=>Xt(Od(ke(t,e)),-1))}function em(e,t){return Y(()=>Xt(gn(ke(t,e)),-1))}function ru(e,t){return Y(()=>{let n=ke(e,t),r=Cr(gn(e),un(),Number.MAX_VALUE),s=gn(Re(n,r));return j(100,Xt(s,-1))})}function ute(e,t){return Y(()=>{let n=Cr(t,un(),Number.MAX_VALUE),r=jr(de(1,n)),s=Cr(e,un(),Number.MAX_VALUE),a=jr(de(1,s));return Xt(Od(ke(r,a)),-1)})}function cte(e,t){return Y(()=>{let n=Va(0,ke(1,j(e,t)));return Xt(Od(n),-1)})}function dte(e,t){return Y(()=>{let n=Va(0,ke(1,j(e,t)));return Xt(n,-1)})}function hte(e,t){return Y(()=>{let n=Te(j(e,t),-1),r=Rs(j(ke(1,e),t),-1);return Va(0,de(1,ke(r,n)))})}function pte(e,t){return Y(()=>{let n=Math.log(2),r=ke(t,e),s=ke(de(r,Cd(j(-2,r))),n);return Xt(s,-1)})}function zd(e,t,n=!1){return Y(()=>{if(n)t=Yy(t);else{let r=Te(t,t.shape.length-1,!0);t=Re(t,r)}return t=Cr(t,un(),1-un()),qt(Te(j(xe(e,"float32"),jr(t)),t.shape.length-1))})}function tm(e,t,n=!1){return Y(()=>{let r=xe(Sf($ee(e)),"int32");t=Cr(t,un(),1-un());let s=t.shape,a=Z(xf(r,s[s.length-1]),s);return zd(a,t,n)})}function fte(e,t){if(!k.arraysEqual(e.shape,t.shape))throw new K(`logits and labels must have the same shape, but got shapes ${JSON.stringify(e.shape)} and ${JSON.stringify(t.shape)}`);return Y(()=>{let n=ca(t),r=qt(gn(t));return de(ke(n,j(t,e)),zy(Gr(r)))})}function nm(e,t){return Y(()=>{let n;return n=Cr(t,un(),1-un()),n=jr(Re(n,ke(1,n))),Xt(fte(e,n),-1)})}function mte(e,t){return Y(()=>{let n=Cr(e,un(),1),r=Cr(t,un(),1);return Te(j(e,jr(Re(n,r))),-1)})}function gte(e,t){return Y(()=>{let n=jr(de(un(),t));return Xt(ke(t,j(e,n)),-1)})}function OA(e,t){return Y(()=>{let n=Qf(e,-1),r=Qf(t,-1),s=j(n,r);return qt(Te(s,-1))})}var rm={meanSquaredError:ii,meanAbsoluteError:em,meanAbsolutePercentageError:ru,meanSquaredLogarithmicError:ute,squaredHinge:cte,hinge:dte,categoricalHinge:hte,logcosh:pte,categoricalCrossentropy:zd,sparseCategoricalCrossentropy:tm,binaryCrossentropy:nm,kullbackLeiblerDivergence:mte,poisson:gte,cosineProximity:OA};function PA(e){if(typeof e=="string"){if(e in rm)return rm[e];let t=`Unknown loss ${e}`;throw e.toLowerCase().includes("softmaxcrossentropy")&&(t=`Unknown loss ${e}. Use "categoricalCrossentropy" as the string name for tf.losses.softmaxCrossEntropy`),new K(t)}else return e}function zA(e,t){return Y(()=>{let n=j(.5,qr(t)),r=zf($r(t,n),e.dtype);return Xt(Hr(e,r),-1)})}function LA(e,t){return Y(()=>zf(Hr(kd(e,-1),kd(t,-1)),"float32"))}function yS(e,t){return Y(()=>xe(Te(_s(Hr(e,1),Hr(t,1))),"float32"))}function yte(e,t){return Y(()=>xe(Te(_s(Hr(e,1),Hr(t,0))),"float32"))}function Ate(e,t){return Y(()=>xe(Te(_s(Hr(e,0),Hr(t,1))),"float32"))}function AS(e,t){return Y(()=>{let n=yS(e,t),r=Ate(e,t),s=de(n,r);return xe(Zn($r(s,0),Re(n,s),0),"float32")})}function xte(e,t){return Y(()=>{let n=yS(e,t),r=yte(e,t),s=de(n,r);return xe(Zn($r(s,0),Re(n,s),0),"float32")})}function xS(e,t){return nm(e,t)}function bS(e,t){return e.rank===t.rank&&(e=Qo(e,[e.rank-1])),t=kd(t,-1),t.dtype!==e.dtype&&(t=xe(t,e.dtype)),xe(Hr(e,t),"float32")}var bte=ii,vte=ii,wte=em,kte=em,Ite=ru,Ste=ru,BA=zd,Tte=OA,vS=tm,sm={binaryAccuracy:zA,categoricalAccuracy:LA,precision:AS,categoricalCrossentropy:BA,sparseCategoricalCrossentropy:vS,mse:bte,MSE:vte,mae:wte,MAE:kte,mape:Ite,MAPE:Ste,cosine:Tte};function Nte(e){if(typeof e=="string"&&e in sm)return sm[e];if(typeof e!="string"&&e!=null)return e;throw new K(`Unknown metric ${e}`)}function am(e){if(Fs(e!==null,`Unknown LossOrMetricFn ${e}`),typeof e=="string")return e;{let t;for(let n of Object.keys(rm))if(rm[n]===e){t=n;break}if(t!==void 0)return t;for(let n of Object.keys(sm))if(sm[n]===e){t=n;break}return t!==void 0?t:e.name}}function Cte(e){let t={Adagrad:()=>Jl.adagrad(.01),Adadelta:()=>Jl.adadelta(1,.95,un()),Adam:()=>Jl.adam(.001,.9,.999,un()),Adamax:()=>Jl.adamax(.002,.9,.999,un(),0),RMSProp:()=>Jl.rmsprop(.001,.9,0,un()),SGD:()=>Jl.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 K(`Unknown Optimizer ${e}`)}var wS=1*1024*1024;function kS(e,t,n=!1){if(e==null||typeof e!="object"||Object.getPrototypeOf(e)!==Object.prototype||!WA(e))throw new Error("User-defined metadata is expected to be a JSON object, but is not.");if(n){let r=JSON.stringify(e);r.length>wS&&console.warn(`User-defined metadata of model "${t}" is too large in size (length=${r.length} when serialized). It is not recommended to store such large objects in user-defined metadata. Please make sure its serialized length is <= ${wS}.`)}}function WA(e){if(e===null)return!0;if(typeof e=="object")if(Object.getPrototypeOf(e)===Object.prototype){let t=Object.keys(e);for(let n of t)if(typeof n!="string"||!WA(e[n]))return!1;return!0}else if(Array.isArray(e)){for(let t of e)if(!WA(t))return!1;return!0}else return!1;else{let t=typeof e;return t==="string"||t==="number"||t==="boolean"}}function Ete(e,t,n,r=console.log){let s=Rte(e),a=["Layer (type)","Output shape","Param #"];s?(t=t||65,n=n||[.45,.85,1]):(t=t||98,n=n||[.33,.55,.67,1]),n[n.length-1]<=1&&(n=n.map(c=>Math.floor(t*c)));let o;if(!s){a.push("Receives inputs"),o=[];for(let c in e.nodesByDepth)o.push(...e.nodesByDepth[c])}r("_".repeat(t)),om(a,n,r),r("=".repeat(t));let i=e.layers;for(let c=0;c<i.length;++c)s?_te(i[c],n,r):Dte(i[c],n,o,r),r((c===i.length-1?"=":"_").repeat(t));e.checkTrainableWeightsConsistency();let l=$te(e),u=Yf(e.nonTrainableWeights);r(`Total params: ${l+u}`),r(`Trainable params: ${l}`),r(`Non-trainable params: ${u}`),r("_".repeat(t))}function $te(e){let t;return e.collectedTrainableWeights!=null?t=Yf(e.collectedTrainableWeights):t=Yf(e.trainableWeights),t}function Rte(e){let t=!0,n=[],r=[];for(let s in e.nodesByDepth)n.push(e.nodesByDepth[s]);for(let s of n){if(s.length>1||s.length===1&&s[0].inboundLayers.length>1){t=!1;break}r.push(...s)}if(t)for(let s of e.layers){let a=!1;for(let o of s.inboundNodes)if(r.indexOf(o)!==-1)if(a){t=!1;break}else a=!0;if(!t)break}return t}function om(e,t,n=console.log){let r="";for(let s=0;s<e.length;++s)s>0&&(r=r.slice(0,r.length-1)+" "),r+=e[s],r=r.slice(0,t[s]),r+=" ".repeat(t[s]-r.length);n(r)}function _te(e,t,n){let r;try{r=JSON.stringify(e.outputShape)}catch(i){r="multiple"}let s=e.name,a=e.getClassName(),o=[`${s} (${a})`,r,e.countParams().toString()];om(o,t,n)}function Dte(e,t,n,r){let s;try{s=JSON.stringify(e.outputShape)}catch(c){s="multiple"}let a=[];for(let c of e.inboundNodes)if(!(n!=null&&n.length>0&&n.indexOf(c)===-1))for(let d=0;d<c.inboundLayers.length;++d){let h=c.inboundLayers[d].name,p=c.nodeIndices[d],f=c.tensorIndices[d];a.push(`${h}[${p}][${f}]`)}let o=e.name,i=e.getClassName(),l=a.length===0?"":a[0],u=[`${o} (${i})`,s,e.countParams().toString(),l];om(u,t,r);for(let c=1;c<a.length;++c)om(["","","",a[c]],t,r)}function IS(e,t,n){return(e==="inboundNodes"||e==="outputLayers"||e==="inputLayers")&&t===0&&typeof n=="string"}function Ld(e,t){if(e===null)return null;if(typeof e=="string")return ri(e);if(typeof e=="number"||typeof e=="boolean")return e;if(e instanceof Array){let n=[],r=e.length;for(let s=0;s<r;++s){let a=e[s];IS(t,s,a)?n.push(a):n.push(Ld(a,t))}return n}else{let n={};for(let r of Object.keys(e)){let s=e[r];if(r==="name"&&typeof s=="string")n[r]=s;else{let a=ri(r);n[a]=Ld(s,a)}}return n}}function VA(e,t){if(e==null)return null;if(typeof e=="string")return pa(e);if(typeof e=="number"||typeof e=="boolean")return e;if(e instanceof Array){let n=[],r=e.length;for(let s=0;s<r;++s){let a=e[s];IS(t,s,a)?n.push(a):n.push(VA(a,t))}return n}else{let n={};for(let r of Object.keys(e)){let s=e[r],a=pa(r);(r==="name"||r==="className")&&typeof s=="string"?n[a]=s:n[a]=VA(s,r)}return n}}var UA="3.8.0";function Fte(e,t){if(e.dtype==null||e.dtype===t.dtype)return t;try{return xe(t,e.dtype)}catch(n){throw new K(`The dtype of the feed (${t.dtype}) can not be cast to the dtype of the key '${e.name}' (${e.dtype}).`)}}var li=class{constructor(e){if(this.id2Value={},this.id2Mask={},this.name2Id={},e instanceof li)for(let t in e.id2Value)this.id2Value[t]=e.id2Value[t],t in e.id2Mask&&(this.id2Mask[t]=e.id2Mask[t]);else{if(e==null)return;for(let t of e)this.add(t.key,t.value)}}add(e,t,n){if(this.id2Value[e.id]==null)this.id2Value[e.id]=Fte(e,t),this.name2Id[e.name]=e.id,n!=null&&(this.id2Mask[e.id]=n);else throw new K(`Duplicate key: name=${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 hs){if(this.id2Value[e.id]==null)throw new K(`Nonexistent key: ${e.name}`);return this.id2Value[e.id]}else{let t=this.name2Id[e];if(t==null)throw new K(`Feed dict has no SymbolicTensor name: ${e}`);return this.id2Value[t]}}getMask(e){if(e instanceof hs){if(this.id2Value[e.id]==null)throw new K(`Nonexistent key: ${e.name}`);return this.id2Mask[e.id]}else{let t=this.name2Id[e];if(t==null)throw new K(`Feed dict has no SymbolicTensor name: ${e}`);return this.id2Mask[t]}}disposeMasks(){this.id2Mask!=null&&Ge(this.id2Mask)}},HA={},SS={};function Bd(e,t,n,r){let s=n==null?!1:n.training,a=Array.isArray(e),o=a?e:[e],i=o.map(f=>f.name),l=[],u=t.names();for(let f of i)u.indexOf(f)!==-1?l.push(t.getValue(f)):l.push(null);r!=null&&(r.maxNumTensors=-1/0,r.minNumTensors=1/0);let c=i.join(",")+"|"+t.names().join(","),d,h;if(HA[c]==null){let f=Mte(o,t);d=f.sorted,h=f.recipientCounts,HA[c]=d,SS[c]=h}d=HA[c],h={},s||Object.assign(h,SS[c]);let p=new li(t);for(let f=0;f<d.length;++f){if(r!=null){let E=Ny().numTensors;E>r.maxNumTensors&&(r.maxNumTensors=E),E<r.minNumTensors&&(r.minNumTensors=E)}let m=d[f],g=m.sourceLayer;if(g instanceof tu)continue;let y=[],A=[],x=[],b=!1;for(let E of m.inputs){let D=p.getValue(E),$=p.getMask(E);y.push(D),A.push($),$!=null&&(b=!0),s||(h[E.name]--,h[E.name]===0&&!t.hasKey(E)&&i.indexOf(E.name)===-1&&!D.isDisposed&&E.sourceLayer.stateful!==!0&&x.push(D))}b&&(n=n||{},n.mask=A[0]);let v=Rt(g.apply(y,n)),I=null;g.supportsMasking&&(I=g.computeMask(y,A));let w=Pte(m),S=Array.isArray(w)?w:[w];for(let E=0;E<S.length;++E){p.hasKey(S[E])||p.add(S[E],v[E],Array.isArray(I)?I[0]:I);let D=i.indexOf(S[E].name);D!==-1&&(l[D]=v[E])}s||Ge(x)}return p.disposeMasks(),a?l:l[0]}function Mte(e,t){k.assert(e!=null&&e.length>0,()=>"Expected at least one fetch, got none");let n=[],r={};if(e.length===1){let s=TS(e[0],t);n=s.sorted,r=s.recipientMap}else{let s=new Set;for(let a of e){let{sorted:o,recipientMap:i}=TS(a,t);for(let l of o)s.has(l.name)||(n.push(l),s.add(l.name));for(let l in i)r[l]==null&&(r[l]=new Set),i[l].forEach(u=>r[l].add(u))}}return{sorted:n,recipientCounts:Ote(r)}}function Ote(e){let t={};for(let n in e)t[n]=e[n].size;return t}function TS(e,t){let n=new Set,r=[],s={};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(),r.push(i),n.add(i.name),l&&o.pop();else{o.push(a.length-1);for(let u of i.inputs)s[u.name]==null&&(s[u.name]=new Set),s[u.name].add(i.name),!n.has(u.name)&&a.push(u)}}return{sorted:r,recipientMap:s}}function Pte(e){let t;if(e.sourceLayer.inboundNodes.length===1)t=e.sourceLayer.output;else{let n=null;for(let r=0;r<e.sourceLayer.inboundNodes.length;++r)for(let s of e.sourceLayer.inboundNodes[r].outputTensors)if(s.id===e.id){n=r;break}t=e.sourceLayer.getOutputAt(n)}return t}var Ps=class extends tt{constructor(e){super({});if(this.containerNodes=new Set,this.name=e.name,this.name==null){let y=this.getClassName().toLowerCase();this.name=Xf(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],Ga(this.inputs).length!==this.inputs.length)throw new K(`The list of inputs passed to the model is redundant. All inputs should only appear once. Found: ${this.inputs.map(y=>y.name)}`);Ga(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 A=y.sourceLayer,x=y.nodeIndex,b=y.tensorIndex;this.outputLayers.push(A),this.outputLayersNodeIndices.push(x),this.outputLayersTensorIndices.push(b)}for(let y of this.inputs){let A=y.sourceLayer,x=y.nodeIndex,b=y.tensorIndex;Fs(x===0,"input layer has >1 nodes"),Fs(b===0,"input layer has >1 tensors"),this.inputLayers.push(A),this.inputLayersNodeIndices.push(x),this.inputLayersTensorIndices.push(b)}this.inputNames=[],this.outputNames=[],this.feedInputShapes=[],this.feedInputNames=[],this.feedOutputNames=[];for(let y=0;y<this.inputLayers.length;y++){let A=this.inputLayers[y];if(!(A instanceof tu))throw new TypeError(`Input layers to a LayersModel must be InputLayer objects. Received inputs: ${e.inputs}. Input ${y} (0-based) originates from layer type ${A.getClassName()}.`);this.inputNames.push(A.name),this.feedInputShapes.push(A.batchInputShape),this.feedInputNames.push(A.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={},r={},s={},a={},o=[],i=(y,A,x,b,v,I)=>{(b==null||v==null||I==null)&&(b=y.sourceLayer,v=y.nodeIndex,I=y.tensorIndex);let w=b.inboundNodes[v];if(x.indexOf(w)!==-1)throw new us(`The tensor ${y.name} at layer "${b.name}" is part of a cycle.`);if(A.indexOf(w)!==-1)return;this.containerNodes.add(Ps.nodeKey(b,v)),b.id in a||(a[b.id]=Object.keys(a).length),x.indexOf(w)===-1&&x.push(w);let S=w.inboundLayers.length;for(let E=0;E<S;E++){let D=w.inputTensors[E],$=w.inboundLayers[E],R=w.nodeIndices[E],N=w.tensorIndices[E];i(D,A,x,$,R,N)}for(A.push(w);x.indexOf(w)>=0;)x.splice(x.indexOf(w),1);o.push(w)},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 A=t[y.id],x=r[y.outboundLayer.id]==null?0:r[y.outboundLayer.id];A=Math.max(A,x),r[y.outboundLayer.id]=A,s[y.outboundLayer.id]=y.outboundLayer,t[y.id]=A;for(let b=0;b<y.inboundLayers.length;b++){let v=y.inboundLayers[b],I=y.nodeIndices[b],w=v.inboundNodes[I],S=t[w.id]==null?0:t[w.id];t[w.id]=Math.max(A+1,S),n[w.id]=w}}let d={};for(let y in t){let A=t[y];A in d||(d[A]=[]),d[A].push(n[y])}let h={};for(let y in r){let A=r[y];A in h||(h[A]=[]),h[A].push(s[y])}let p=Object.keys(h).map(y=>parseInt(y,10)).sort(Pf);this.layers=[];for(let y of p){let A=h[y];A.sort((x,b)=>{let v=a[x.id],I=a[b.id];return v<I?-1:v>I?1:0});for(let x of A)x instanceof Ps&&this.internalContainerRefs.push(x),this.layers.push(x)}this.layersByDepth=h,p=Object.keys(d).map(y=>parseInt(y,10)).sort(Pf);let f=this.inputs.slice(),m=[];for(let y of p)for(let A of d[y]){let x=A.outboundLayer;if(x!=null){for(let b of A.inputTensors)if(f.indexOf(b)===-1)throw new us(`Graph disconnected: cannot obtain value for tensor ${b} at layer "${x.name}". The following previous layers were accessed without issue: ${m}`);for(let b of A.outputTensors)f.push(b);m.push(x.name)}}this.nodesByDepth=d;let g=this.layers.map(y=>y.name);for(let y of g){let A=g.filter(x=>x===y).length;if(A!==1)throw new us(`The name "${y}" is used ${A} times in the model. All layer names should be unique. Layer names: `+JSON.stringify(g))}this.outboundNodes=[],this.inboundNodes=[],new Jf({outboundLayer:this,inboundLayers:[],nodeIndices:[],tensorIndices:[],inputTensors:this.inputs,outputTensors:this.outputs,inputMasks:this.inputs.map(y=>null),outputMasks:this.outputs.map(y=>null),inputShapes:this.inputs.map(y=>y.shape),outputShapes:this.outputs.map(y=>y.shape)}),this.built=!0,this._refCount=1}assertNotDisposed(){if(this._refCount===0)throw new Error(`Container '${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 K("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={},r=0;for(let a of this.layers)for(let o of a.weights){if(n[o.originalName]!=null)throw new K(`Duplicate weight name: ${o.originalName}`);n[o.originalName]=o,r++}let s=[];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)s.push([n[o],e[a]]);else if(t)throw new K(`Provided weight data has no target variable: ${a}`);delete n[o]}if(t){let a=[];for(let o in n)a.push(o);if(a.length>0)throw new K(`${a.length} of ${r} weights are not set: ${a}`)}FA(s)}updatedConfig(){let e=this.getConfig(),t={};return t.className=this.getClassName(),t.config=e,t.kerasVersion=`tfjs-layers ${UA}`,t.backend="TensorFlow.js",t}toJSON(e,t=!0){let n=VA(this.updatedConfig());return t?JSON.stringify(n):n}call(e,t){return Y(()=>{e=Rt(e);let n=new li;for(let r=0;r<this.inputs.length;++r)n.add(this.inputs[r],e[r]);return Bd(this.outputs,n,t)})}computeMask(e,t){return Y(()=>{e=Rt(e);let n;return t==null?n=ni(null,e.length):n=Rt(t),this.runInternalGraph(e,n)[1]})}computeOutputShape(e){let t=Zf(e);if(t.length!==this.inputLayers.length)throw new K(`Invalid inputShape argument ${e}: model has ${this.inputLayers.length} tensor inputs.`);let n={};for(let o=0;o<t.length;o++){let i=this.inputLayers[o],l=t[o],u=i.name+"_0_0";n[u]=l}let r=Object.keys(this.nodesByDepth).map(o=>parseInt(o,10)).sort(Pf);if(r.length>1)for(let o of r){let i=this.nodesByDepth[o];for(let l of i){let u=l.outboundLayer;if(this.inputLayers.map(f=>f.id).indexOf(u.id)!==-1)continue;let c=[];for(let f=0;f<l.inboundLayers.length;f++){let m=l.inboundLayers[f],g=l.nodeIndices[f],y=l.tensorIndices[f],A=`${m.name}_${g}_${y}`,x=n[A];c.push(x)}let d=u.computeOutputShape(Jn(c)),h=Zf(d),p=u.inboundNodes.indexOf(l);for(let f=0;f<h.length;f++){let m=`${u.name}_${p}_${f}`;n[m]=h[f]}}}let s=[],a=[];for(let o=0;o<this.outputLayers.length;o++){let i=this.outputLayers[o],l=this.outputLayersNodeIndices[o],u=this.outputLayersTensorIndices[o],c=`${i.name}_${l}_${u}`;a.push(c)}for(let o=0;o<a.length;o++){let i=a[o];Fs(i in n),s.push(n[i])}return Jn(s)}runInternalGraph(e,t){t==null&&(t=ni(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 r=Object.keys(this.nodesByDepth).map(i=>parseInt(i,10)).sort(Pf);for(let i of r){let l=this.nodesByDepth[i];for(let u of l){let c=u.outboundLayer,d=u.inputTensors,h=u.outputTensors,p=new Array;for(let f of d)f.id in n&&p.push(n[f.id]);if(p.length===d.length){let f={},m,g,y,A;if(u.callArgs!=null&&(f=u.callArgs),p.length===1){let[x,b]=p[0];f.mask==null&&(f.mask=b),y=Rt(c.call(x,f)),A=Rt(c.computeMask(x,b)),m=[x],g=[b]}else m=p.map(x=>x[0]),g=p.map(x=>x[1]),f.mask==null&&(f.mask=g),y=Rt(c.call(m,f)),A=Rt(c.computeMask(m,g));if(c.activityRegularizer)throw new He("LayersModel invocation with concrete Tensor value(s) in the presence of activity regularizer(s) is not supported yet.");for(let x=0;x<h.length;++x){let b=h[x],v=y[x],I=A[x];n[b.id]=[v,I]}}}}let s=[],a=[],o=[];for(let i of this.outputs){Fs(i.id in n,`Could not compute output ${i.name} : ${i.id}`);let[l,u]=n[i.id];o.push(l.shape),s.push(l),a.push(u)}return[s,a,o]}buildNodeConversionMap(e){let t={},n;for(let r of this.layers){n=r instanceof Ps?1:0;for(let s=0;s<r.inboundNodes.length;s++){let a=Ps.nodeKey(r,s);this.containerNodes.has(a)&&(t[a]=n,n+=1)}}return t}getLayer(e,t){if(t!=null){if(this.layers.length<=t)throw new K(`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 K("Provide either a layer name or layer index");for(let n of this.layers)if(n.name===e)return n;throw new K(`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 r=Ps.nodeKey(t,n);this.containerNodes.has(r)&&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 d=a.inboundNodes[c],h=Ps.nodeKey(a,c),p={};if(this.containerNodes.has(h)){if(d.callArgs)try{JSON.stringify(d.callArgs),p=d.callArgs}catch(f){console.warn(`Layer ${a.name} was passed non-serializable keyword arguments: ${d.callArgs}. They will not be included in the serialized model (and thus will be missing at deserialization time).`),p={}}if(d.inboundLayers.length>0){let f=[];for(let m=0;m<d.inboundLayers.length;m++){let g=d.inboundLayers[m],y=d.nodeIndices[m],A=d.tensorIndices[m],x=Ps.nodeKey(g,y),b=t[x];b==null&&(b=0),f.push([g.name,b,A,p])}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 r=[];for(let a=0;a<this.inputLayers.length;a++){let o=this.inputLayers[a],i=this.inputLayersNodeIndices[a],l=Ps.nodeKey(o,i);if(!this.containerNodes.has(l))continue;let u=t[l];u==null&&(u=0);let c=this.inputLayersTensorIndices[a];r.push([o.name,u,c])}e.inputLayers=r;let s=[];for(let a=0;a<this.outputLayers.length;a++){let o=this.outputLayers[a],i=this.outputLayersNodeIndices[a],l=Ps.nodeKey(o,i);if(!this.containerNodes.has(l))continue;let u=t[l];u==null&&(u=0);let c=this.outputLayersTensorIndices[a];s.push([o.name,u,c])}return e.outputLayers=s,e}static fromConfig(e,t,n={},r=!1){let s={},a={};function o(m,g){m.name in a?a[m.name].push(g):a[m.name]=[g]}function i(m,g){let y=[],A;for(let x of g){let b=x[0],v=x[1],I=x[2];if(A=x[3]==null?{}:x[3],!(b in s)){o(m,g);return}let w=s[b];if(w.inboundNodes.length<=v){o(m,g);return}let S=w.inboundNodes[v];y.push(S.outputTensors[I])}y.length>0&&m.apply(Jn(y),A)}function l(m){let g=m.name,y=ps(m,t.customObjects!=null?t.customObjects:{});y.setFastWeightInitDuringBuild(r),s[g]=y,m.inboundNodes.forEach(x=>{if(!(x instanceof Array))throw new K(`Corrupted configuration, expected array for nodeData: ${x}`);o(y,x)})}let u=t.name,c=t.layers;for(let m of c)l(m);for(;!fee(a);)for(let m of c){let g=s[m.name];if(g.name in a){let y=a[g.name];delete a[g.name];for(let A of y)i(g,A)}}let d=[],h=[],p=t.inputLayers;for(let m of p){let g=m[0],y=m[1],A=m[2];Fs(g in s);let b=s[g].inboundNodes[y].outputTensors;d.push(b[A])}let f=t.outputLayers;for(let m of f){let g=m[0],y=m[1],A=m[2];Fs(g in s);let b=s[g].inboundNodes[y].outputTensors;h.push(b[A])}return new e({inputs:d,outputs:h,name:u})}get stateful(){if(this._stateful)throw new K("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 zte(e,t,n){let r=t.length;if(e==null||Array.isArray(e)&&e.length===0)return t.map(s=>null);if(r===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!==r)throw new Error(`Provided ${n} is an array of ${e.length} element(s), but the model has ${r} outputs. Make sure a set of weights is provided for each model output.`);return e}else if(typeof e=="object"&&Object.keys(e).length>0&&typeof e[Object.keys(e)[0]]=="object"){let s=[];return t.forEach(a=>{a in e?s.push(e[a]):s.push(null)}),s}else throw new Error(`The model has multiple (${r}) outputs, so ${n} must be either an array with ${r} elements or an object with ${t} keys. Provided ${n} not understood: ${JSON.stringify(e)}`)}function NS(e,t){return zte(e,t,"classWeight")}async function CS(e,t,n,r){if(t!=null||r!=null)throw new Error("Support sampleWeight is not implemented yet");if(n!=null){let s=Y(()=>{if(e.shape.length===1)return za(e);if(e.shape.length===2){if(e.shape[1]>1)return kd(e,1);if(e.shape[1]===1)return Z(e,[e.shape[0]]);throw new Error(`Encountered unexpected last-dimension size (${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 s.data());Ge(s);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])}),En(o,"float32")}else return null}function Lte(e,t){return j(e,t)}var Bte=32;function ES(e,t){let n,r,s=t;n=s.xs,r=s.ys,k.assert(n!=null&&r!=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 ${t}`);let a=$S("input",e.inputNames,n),o=$S("output",e.outputNames,r),i=a[0].shape[0];k.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)})`),k.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++)k.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++)k.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 $S(e,t,n){if(n instanceof Ot)return[n];if(Array.isArray(n))return k.assert(n.length===t.length,()=>`Received an array of ${n.length} Tensors, but expected ${t.length} to match the ${e} keys ${t}.`),n;{let r=[];for(let s of t){if(n[s]==null)throw new K(`The feature data generated by the dataset lacks the required ${e} key '${s}'.`);r.push(n[s])}return r}}function Wte(e){if(e.length===3)throw new He("Validation with sample weights is not implemented yet.");return{xs:e[0],ys:e[1]}}async function Vte(e,t,n){let r=n.batchesPerEpoch!=null;if(k.assert(e.optimizer!=null,()=>"You must compile a model before training/testing. Use LayersModel.compile(modelCompileConfig)."),k.assert(n!=null,()=>"For fitDataset(), the 2nd argument (config) is required, but it is not provided in this call."),k.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}`),k.assert(!r||n.batchesPerEpoch>0&&Number.isInteger(n.batchesPerEpoch),()=>`For fitDataset(), config.batchesPerEpoch is expected to be a positive integer if specified, but got ${n.batchesPerEpoch}`),k.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 s=n.validationData!=null,a,o;if(s)if(RS(n.validationData))k.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=Wte(n.validationData);a=g.xs,o=g.ys}let i=e.makeTrainFunction(),l=e.getDedupedMetricsNames(),u;s?u=l.slice().concat(l.map(g=>"val_"+g)):u=l.slice();let c=mS(n.callbacks,n.yieldEvery),d=n.verbose==null?1:n.verbose,{callbackList:h,history:p}=gS(c,d,n.epochs,null,null,Ute(t,n),null,s,u);h.setModel(e),e.history=p,await h.onTrainBegin(),e.stopTraining_=!1;let f=n.initialEpoch==null?0:n.initialEpoch,m=await t.iterator();for(;f<n.epochs;){let g={};await h.onEpochBegin(f);let y=0,A=0;for(r||(m=await t.iterator());r?y<n.batchesPerEpoch:!0;){let x=await m.next();if(r&&x.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(x.value!=null){let{xs:b,ys:v}=ES(e,x.value),I={};I.batch=A,I.size=b[0].shape[0],await h.onBatchBegin(A,I);let w=[];if(n.classWeight!=null){let D=NS(n.classWeight,e.outputNames);for(let $=0;$<D.length;++$)w.push(await CS(v[$],null,D[$]))}let S=b.concat(v).concat(w),E=i(S);Ge(S);for(let D=0;D<l.length;++D){let $=l[D],R=E[D];I[$]=R,In(R)}await h.onBatchEnd(A,I),cS(I),A++,y++}if(r?y>=n.batchesPerEpoch:x.done){if(s){let b;RS(n.validationData)?b=Rt(await e.evaluateDataset(n.validationData,{batches:n.validationBatches})):b=Rt(e.evaluate(a,o,{batchSize:n.validationBatchSize==null?Bte:n.validationBatchSize,verbose:0}));for(let v=0;v<e.metricsNames.length;++v)g[`val_${e.metricsNames[v]}`]=b[v]}break}if(e.stopTraining_)break}if(await h.onEpochEnd(f,g),f++,e.stopTraining_)break}return await h.onTrainEnd(),await e.history.syncData(),e.history}finally{e.isTraining=!1}}function Ute(e,t){let n=null;return t.batchesPerEpoch!=null?n=t.batchesPerEpoch:Number.isFinite(e.size)&&(n=e.size),n}function RS(e){return typeof e.iterator=="function"}function Hte(e){return typeof e.next=="function"}async function Gte(e,t,n){n=n||{};let r=n.batches!=null,s=e.testFunction,a=[];if(n.verbose>0)throw new He("Verbose mode is not implemented yet.");k.assert(!r||n.batches>0&&Number.isInteger(n.batches),()=>`Test loop expects \`batches\` to be a positive integer, but received ${JSON.stringify(n.batches)}`);let o=Hte(t)?t:await t.iterator(),i=0,l=0;for(;r?l<n.batches:!0;){let u=await o.next();if(a=Y(()=>{if(u.value){let{xs:c,ys:d}=ES(e,u.value),h=c.concat(d),p=Y(()=>s(h));if(Ge(h),l===0)for(let m=0;m<p.length;++m)a.push(De(0));let f=h[0].shape[0];for(let m=0;m<p.length;++m){let g=p[m],y=a[m];a[m]=Y(()=>de(a[m],j(f,g))),l>0&&Ge(y)}Ge(p),i+=f,++l}return a}),u.done){r&&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]=Re(a[u],i),Ge(c)}return Jn(a)}function GA(e){k.assert(e>0&&Number.isInteger(e),()=>`batchSize is required to be a positive integer, but got ${e}`)}function Wd(e,t,n){return e==null?[null]:Array.isArray(e)?e.map(r=>oi(r,t,n-t)):oi(e,t,n-t)}function jA(e,t){return Y(()=>e==null?null:Array.isArray(e)?e.map(n=>jA(n,t)):eS(e,t.dtype==="int32"?t:xe(t,"int32")))}function qA(e,t){let n=[],r=0,s=null;for(;r<e;)s=r+t,s>=e&&(s=e),n.push([r,s]),r=s;return n}async function jte(e,t,n,r,s,a,o,i,l,u,c,d,h,p,f){s==null&&(s=32),a==null&&(a=1),c==null&&(c=!0),h==null&&(h=0);let m=!1;if(l!=null&&u!=null&&(m=!0),f!=null&&(m=!0,p==null))throw new K("Can only use `validationSteps` when doing step-wise training, i.e., `stepsPerEpoch` must be set.");let g=e.checkNumSamples(n,s,p,"steps_per_epoch"),y;g!=null&&(y=cs(0,g)),o==null&&(o=1);let{callbackList:A,history:x}=gS(i,o,a,h,g,p,s,m,d);A.setModel(e),e.history=x,await A.onTrainBegin(),e.stopTraining_=!1;for(let b=h;b<a;++b){await A.onEpochBegin(b);let v={};if(p!=null)throw new He("stepsPerEpoch mode is not implemented yet.");{if(c==="batch")throw new He("batch shuffling is not implemneted yet");c&&k.shuffle(y);let I=En(y),w=qA(g,s);for(let S=0;S<w.length;++S){let E={};if(await A.onBatchBegin(S,E),Y(()=>{let D=w[S][0],$=w[S][1],R=oi(I,D,$-D);E.batch=S,E.size=$-D;let N=jA(n,R),M=t(N);for(let B=0;B<r.length;++B){let q=r[B],X=M[B];E[q]=X,In(X)}if(S===w.length-1&&m){let B=e.testLoop(l,u,s);for(let q=0;q<r.length;++q){let X=r[q],J=B[q];In(J),v["val_"+X]=J}}}),await A.onBatchEnd(S,E),cS(E),e.stopTraining_)break}I.dispose()}if(await A.onEpochEnd(b,v),e.stopTraining_)break}return await A.onTrainEnd(),await e.history.syncData(),e.history}async function qte(e,t,n,r={}){if(e.isTraining)throw new Error("Cannot start training because another fit() call is ongoing.");e.isTraining=!0;let s,a,o,i,l,u,c;try{let d=r.batchSize==null?32:r.batchSize;GA(d);let h=!1,p=await e.standardizeUserData(t,n,r.sampleWeight,r.classWeight,h,d);s=p[0],a=p[1],c=p[2];let f=!1,m;if(r.validationData!=null&&r.validationData.length>0){if(f=!0,r.validationData.length===2)o=r.validationData[0],i=r.validationData[1];else throw r.validationData.length===3?new He("validationData including sample weights is not supported yet."):new K(`When passing validation data, it must contain 2 (valX, valY) or 3 (valX, valY, valSampleWeight) items; ${r.validationData} is invalid.`);let w=!0,S=await e.standardizeUserData(o,i,null,null,w,d);l=S[0],u=S[1],m=l.concat(u)}else if(r.validationSplit!=null&&r.validationSplit>0&&r.validationSplit<1){f=!0;let w=Math.floor(s[0].shape[0]*(1-r.validationSplit)),S=s[0].shape[0];l=Wd(s,w,S),s=Wd(s,0,w),u=Wd(a,w,S),a=Wd(a,0,w),m=l.concat(u)}else r.validationSteps!=null&&(f=!0);let g=s.concat(a).concat(c);e.checkTrainableWeightsConsistency();let y=e.makeTrainFunction(),A=e.getDedupedMetricsNames(),x,b;f?(e.makeTestFunction(),x=e.testFunction,b=A.slice().concat(A.map(w=>"val_"+w))):(x=null,m=[],b=A.slice());let v=mS(r.callbacks,r.yieldEvery);return await jte(e,y,g,A,d,r.epochs,r.verbose,v,x,m,r.shuffle,b,r.initialEpoch,null,null)}finally{e.isTraining=!1,ui(s,t),ui(a,n),ui(l,o),ui(u,i),c!=null&&Ge(c)}}function _S(e){let t=[];e instanceof Ot&&(e=[e]);for(let n=0;n<e.length;++n){let r=e[n];if(r.rank===1)t.push(Md(r,1));else{if(r.rank===0)throw new Error("Expected tensor to be at least 1D, but received a 0D tensor (scalar).");t.push(r)}}return t}function ui(e,t){if(e==null)return;let n=[];if(t instanceof Ot)n.push(t.id);else if(Array.isArray(t))t.forEach(s=>n.push(s.id));else if(t!=null)for(let s in t){let a=t[s];n.push(a.id)}let r=[];if(e instanceof Ot)n.indexOf(e.id)===-1&&r.push(e);else if(Array.isArray(e))e.forEach(s=>{n.indexOf(s.id)===-1&&r.push(s)});else if(e!=null)for(let s in e){let a=e[s];n.indexOf(a.id)===-1&&r.push(a)}r.forEach(s=>{s.isDisposed||s.dispose()})}function Kte(e){return e instanceof Ot}function KA(e){return Array.isArray(e)}function DS(e){return!Kte(e)&&!KA(e)}function FS(e,t,n,r=!0,s=""){if(t==null||t.length===0){if(e!=null){let o=!1;if(KA(e)&&e.length>0)o=!0;else if(DS(e)){for(let i in e)if(e.hasOwnProperty(i)){o=!0;break}}else o=!0;if(o)throw new K(`Error when checking model ${s} expected no data, but got ${e}`)}return[]}if(e==null)return t.map(o=>null);let a;if(DS(e)){e=e,a=[];for(let o of t){if(e[o]==null)throw new K(`No data provided for "${o}". Need data for each key in: ${t}`);a.push(e[o])}}else if(KA(e)){if(e=e,e.length!==t.length)throw new K(`Error when checking model ${s}: 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 K(`The model ${s} expects ${t.length} Tensor(s), but only received one Tensor. Found: Tensor with shape ${e.shape}`);a=[e]}if(a=_S(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 K(`Error when checking ${s}: 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&&!r)continue;let u=i.shape[l],c=n[o][l];if(c!=null&&c>=0&&u!==c)throw new K(`Error when checking ${s}: expected ${t[o]} to have shape [${n[o]}], but got array with shape [${i.shape}].`)}}return a}function Xte(e,t,n){let r=Ga(e.map(a=>a.shape[0]));r.sort();let s=Ga(t.map(a=>a.shape[0]));if(s.sort(),r.length>1)throw new K(`All input Tensors (x) should have the same number of samples. Got array shapes: ${JSON.stringify(e.map(a=>a.shape))}`);if(s.length>1)throw new K(`All target Tensors (y) should have the same number of samples. Got array shapes: ${JSON.stringify(t.map(a=>a.shape))}`);if(r.length>0&&s.length>0&&!k.arraysEqual(r,s))throw new K(`Input Tensors should have the same number of samples as target Tensors. Found ${r[0]} input sample(s) and ${s[0]} target sample(s).`)}function Zte(e,t,n){let r=[ii,nm,zd];for(let s=0;s<e.length;++s){let a=e[s],o=t[s],i=n[s];if(o!=null){if(o===zd&&a.shape[a.shape.length-1]===1)throw new K(`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(r.indexOf(o)!==-1){let l=a.shape.slice(1),u=i.slice(1);for(let c=0;c<l.length;++c){let d=l[c],h=u[c];if(h!=null&&d!==h)throw new K(`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 MS(e,t,n,r=!0,s=""){let a;if(Array.isArray(e)){if(e.length!==t.length)throw new K(`Error when checking model ${s}: 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 K(`The model expects ${t.length} ${s} 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 K(`Error when checking ${s}: 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&&!r)continue;let u=i.shape[l],c=n[o][l];if(c!=null&&c!==u)throw new K(`Error when checking ${s}: expected ${t[o]} to have shape ${JSON.stringify(n[o])} but got array with shape ${JSON.stringify(i.shape)}.`)}}}function Yte(e,t){if(e==null||Array.isArray(e)&&e.length===0)return t.map(r=>[]);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: ${e}`);if(Array.isArray(n))return t.map(r=>n);{let r=[];for(let s of t){let a=n.hasOwnProperty(s)?n[s]:[];Array.isArray(a)||(a=[a]),r.push(a)}return r}}var Jte="layers-model",fa=class extends Ps{constructor(e){super(e);this.isTraining=!1}summary(e,t,n=console.log){if(!this.built)throw new K("This model has never been called, thus its weights have not been created yet. So no summary can be displayed. Build the model first (e.g., by calling it on some test data).");Ete(this,e,t,n)}compile(e){if(e.loss==null&&(e.loss=[]),this.loss=e.loss,typeof e.optimizer=="string")this.optimizer_=Cte(e.optimizer),this.isOptimizerOwned=!0;else{if(!(e.optimizer instanceof Ha))throw new K("User-defined optimizer must be an instance of tf.Optimizer.");this.optimizer_=e.optimizer,this.isOptimizerOwned=!1}let t=[];if(!Array.isArray(e.loss)&&typeof e.loss!="string"&&typeof e.loss!="function"){e.loss=e.loss;for(let a in e.loss)if(this.outputNames.indexOf(a)===-1)throw new K(`Unknown entry in loss dictionary: "${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(PA(e.loss[a]))}else if(Array.isArray(e.loss)){if(e.loss.length!==this.outputs.length)throw new K(`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=>PA(o))}else{let a=PA(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=[],ai("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 r=Yte(e.metrics,this.outputNames),s=(a,o,i)=>{this.outputNames.length>1&&(o=this.outputNames[a]+"_"+o),this.metricsNames.push(o),this.metricsTensors.push([i,a])};ai("metric",()=>{for(let a=0;a<this.outputs.length;++a){if(n.indexOf(a)!==-1)continue;let o=r[a];(l=>{let u="",c,d,h;for(let p of l){if(typeof p=="string"&&["accuracy","acc","crossentropy","ce"].indexOf(p)!==-1){let m=this.internalOutputShapes[a];m[m.length-1]===1||this.lossFunctions[a]===nm?["accuracy","acc"].indexOf(p)!==-1?d=zA:["crossentropy","ce"].indexOf(p)!==-1&&(d=xS):this.lossFunctions[a]===tm?["accuracy","acc"].indexOf(p)!==-1?d=bS:["crossentropy","ce"].indexOf(p)!==-1&&(d=vS):["accuracy","acc"].indexOf(p)!==-1?d=LA:["crossentropy","ce"].indexOf(p)!==-1&&(d=BA);let g;["accuracy","acc"].indexOf(p)!==-1?g="acc":["crossentropy","ce"].indexOf(p)!==-1&&(g="ce"),h=d,c=u+g}else h=Nte(p),c=u+am(p);let f;ai(c,()=>{f=h}),s(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 r=n.batchSize==null?32:n.batchSize;GA(r);let s=!0,a=this.standardizeUserDataXY(e,t,s,r);try{let o=a[0].concat(a[1]);this.makeTestFunction();let i=this.testFunction,l=this.testLoop(i,o,r,n.verbose,n.steps);return Jn(l)}finally{ui(a[0],e),ui(a[1],t)}}async evaluateDataset(e,t){return this.makeTestFunction(),Gte(this,e,t)}checkNumSamples(e,t,n,r="steps"){let s;if(n!=null){if(s=null,t!=null)throw new K(`If ${r} is set, batchSize must be null or undefined.Got batchSize = ${t}`)}else if(e!=null)Array.isArray(e)?s=e[0].shape[0]:s=e.shape[0];else throw new K(`Either the input data should have a defined shape, or ${r} shoud be specified.`);return s}execute(e,t){if(Array.isArray(t)&&t.length===0)throw new K("`outputs` is an empty Array, which is not allowed.");let n=Array.isArray(t),r=n?t:[t],s=this.retrieveSymbolicTensors(r),a=new li;if(e instanceof Ot&&(e=[e]),Array.isArray(e)){if(e.length!==this.inputs.length)throw new K(`The number of inputs provided (${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 K(`No value is provided for the model's input ${i.name}`);a.add(i,l)}let o=Bd(s,a);return n?o:o[0]}retrieveSymbolicTensors(e){let t=ni(null,e.length),n=e.length;for(let r of this.layers){let s=Array.isArray(r.output)?r.output:[r.output],a=s.map(o=>o.name);for(let o=0;o<e.length;++o){let i=a.indexOf(e[o]);if(i!==-1&&(t[o]=s[i],n--),n===0)break}if(n===0)break}if(n>0){let r=[];throw t.forEach((s,a)=>{s==null&&r.push(e[a])}),new K(`Cannot find SymbolicTensors for output name(s): ${JSON.stringify(r)}`)}return t}predictLoop(e,t=32,n=!1){return Y(()=>{let r=this.checkNumSamples(e);if(n)throw new He("Verbose predictLoop() is not implemented yet.");let s=qA(r,t),a=this.outputs.map(o=>[]);for(let o=0;o<s.length;++o)Y(()=>{let l=s[o][0],u=s[o][1],c=Wd(e,l,u),d=[];if(Array.isArray(c))for(let p=0;p<c.length;++p)d.push({key:this.inputs[p],value:c[p]});else d.push({key:this.inputs[0],value:c});let h=new li(d);return Bd(this.outputs,h)}).forEach((l,u)=>a[u].push(l));return Jn(a.map(o=>an(o,0)))})}predict(e,t={}){let n=_S(e);MS(n,this.inputNames,this.feedInputShapes,!1);try{let r=t.batchSize==null?32:t.batchSize;return GA(r),this.predictLoop(n,r)}finally{ui(n,e)}}predictOnBatch(e){MS(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,r){if(this.optimizer_==null)throw new us("You must compile a model before training/testing. Use LayersModel.compile(modelCompileArgs).");let s=[];for(let a=0;a<this.feedOutputShapes.length;++a){let o=this.feedOutputShapes[a];this.feedLossFns[a]===tm?s.push(o.slice(0,o.length-1).concat([1])):s.push(o)}if(e=FS(e,this.feedInputNames,this.feedInputShapes,!1,"input"),t=FS(t,this.feedOutputNames,s,!1,"target"),Xte(e,t,null),Zte(t,this.feedLossFns,this.feedOutputShapes),this.stateful&&r!=null&&r>0&&e[0].shape[0]%r!=0)throw new K(`In a stateful network, you should only pass inputs with a number of samples that is divisible by the batch size ${r}. Found: ${e[0].shape[0]} sample(s).`);return[e,t]}async standardizeUserData(e,t,n,r,s=!0,a){let[o,i]=this.standardizeUserDataXY(e,t,s,a);if(n!=null)throw new Error("sample weight is not supported yet.");let l=null;if(r!=null){let u=NS(r,this.outputNames);l=[];for(let c=0;c<u.length;++c)l.push(await CS(i[c],null,u[c]))}return[o,i,l]}testLoop(e,t,n,r=0,s){return Y(()=>{let a=this.checkNumSamples(t,n,s,"steps"),o=[];if(r>0)throw new He("Verbose mode is not implemented yet.");if(s!=null)throw new He("steps mode in testLoop() is not implemented yet");{let i=qA(a,n),l=En(cs(0,a));for(let u=0;u<i.length;++u){let c=i[u][0],d=i[u][1],h=oi(l,c,d-c),p=jA(t,h),f=e(p);if(u===0)for(let m=0;m<f.length;++m)o.push(De(0));for(let m=0;m<f.length;++m){let g=f[m];o[m]=de(o[m],j(d-c,g))}}for(let u=0;u<o.length;++u)o[u]=Re(o[u],a)}return o})}getDedupedMetricsNames(){let e=this.metricsNames,t=[];for(let n=0;n<e.length;++n){let r=e[n],s=r;VI(e,r)>1&&(s+=`_${VI(e.slice(0,n),r)}`),t.push(s)}return t}makeTrainFunction(){return e=>{let t=[],n=e.slice(0,this.inputs.length),r=e.slice(this.inputs.length,this.inputs.length+this.outputs.length),s=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 d=new li(c),h=Bd(this.outputs,d,{training:!0}),p;for(let f=0;f<this.lossFunctions.length;++f){let g=this.lossFunctions[f](r[f],h[f]);s[f]!=null&&(g=Lte(g,s[f]));let y=Xt(g);t.push(y),f===0?p=g:p=de(p,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=Xt(g(r[y],h[y]))}In(m),a.push(m)}return p=Xt(p),this.calculateLosses().forEach(f=>{p=de(p,f)}),p},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,r=e.slice(0,this.inputs.length),s=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:r[l]});let o=new li(a),i=Bd(this.outputs,o);for(let l=0;l<this.lossFunctions.length;++l){let u=this.lossFunctions[l],c=Xt(u(s[l],i[l]));l===0?n=c:n=de(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],d=Xt(u(s[c],i[c]));t.push(d)}return t})}async fit(e,t,n={}){return qte(this,e,t,n)}async fitDataset(e,t){return Vte(this,e,t)}async trainOnBatch(e,t){let n=await this.standardizeUserData(e,t),r=n[0],s=n[1],o=this.makeTrainFunction()(r.concat(s)),i=[];for(let l of o){let u=await l.data();i.push(u[0])}return Ge(o),Jn(i)}getNamedWeights(e){let t=[],n=e!=null&&e.trainableOnly,r=n?this.trainableWeights:this.weights,s=this.getWeights(n);for(let a=0;a<r.length;++a)n&&!r[a].trainable||t.push({name:r[a].originalName,tensor:s[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=Ny().numTensors;this.optimizer_.dispose(),e.numDisposedVariables+=t-Ny().numTensors}return e}getLossIdentifiers(){let e;if(typeof this.loss=="string")e=pa(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=>pa(t))}else{let t=Object.keys(this.loss);e={};let n=this.loss;for(let r of t)if(typeof n[r]=="string")e[r]=pa(n[r]);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[pa(am(this.metrics))];if(Array.isArray(this.metrics))return this.metrics.map(e=>pa(am(e)));{let e={};for(let t in this.metrics)e[t]=pa(am(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=Ld(e.optimizer_config),n=ps(t),r;if(typeof e.loss=="string")r=ri(e.loss);else if(Array.isArray(e.loss))r=e.loss.map(a=>ri(a));else if(e.loss!=null){r={};for(let a in e.loss)r[a]=ri(e.loss[a])}let s;if(Array.isArray(e.metrics))s=e.metrics.map(a=>ri(a));else if(e.metrics!=null){s={};for(let a in e.metrics)s[a]=ri(e.metrics[a])}this.compile({loss:r,metrics:s,optimizer:n})}async save(e,t){if(typeof e=="string"){let l=cr.getSaveHandlers(e);if(l.length===0)throw new K(`Cannot find any save handlers for URL '${e}'`);if(l.length>1)throw new K(`Found more than one (${l.length}) save handlers for URL '${e}'`);e=l[0]}if(e.save==null)throw new K("LayersModel.save() cannot proceed because the IOHandler provided does not have the `save` attribute defined.");let n=await cr.encodeWeights(this.getNamedWeights(t)),r=!1,s=null,o={modelTopology:this.toJSON(s,r),format:Jte,generatedBy:`TensorFlow.js tfjs-layers v${UA}`,convertedBy:null};if((t==null?!1:t.includeOptimizer)&&this.optimizer!=null){o.trainingConfig=this.getTrainingConfig();let l="optimizer",{data:u,specs:c}=await cr.encodeWeights(await this.optimizer.getWeights(),l);n.specs.push(...c),n.data=cr.concatenateArrayBuffers([n.data,u])}if(this.userDefinedMetadata!=null){let l=!0;kS(this.userDefinedMetadata,this.name,l),o.userDefinedMetadata=this.userDefinedMetadata}return o.weightData=n.data,o.weightSpecs=n.specs,e.save(o)}setUserDefinedMetadata(e){kS(e,this.name),this.userDefinedMetadata=e}getUserDefinedMetadata(){return this.userDefinedMetadata}};fa.className="Model";ue.registerClass(fa);var OS=class extends fa{};OS.className="Functional";ue.registerClass(OS);async function Qte(e,t){"modelTopology"in e||(e={modelTopology:e}),e=e;let n=e.modelTopology;n.model_config!=null&&(n=n.model_config);let r=Ld(n),s=ps(r,t);if(e.weightsManifest!=null){let a=await cr.loadWeights(e.weightsManifest,e.pathPrefix,s.weights.map(i=>i.originalName)),o={};for(let i of s.weights)o[i.originalName]=a[i.originalName];s.loadWeights(o),Ge(a)}return s}async function ene(e,t){if(t==null&&(t={}),typeof e=="string"){let n=cr.getLoadHandlers(e,t);if(n.length===0)n.push(cr.browserHTTPRequest(e,t));else if(n.length>1)throw new K(`Found more than one (${n.length}) load handlers for URL '${e}'`);e=n[0]}return tne(e,void 0,t)}async function tne(e,t,n){if(n==null&&(n={}),e.load==null)throw new K("Cannot proceed with model loading because the IOHandler provided does not have the `load` method implemented.");let r=await e.load(),s=r.modelTopology;s.model_config!=null&&(s=s.model_config);let a=n.strict==null?!0:n.strict,o=r.weightData!=null&&r.weightSpecs!=null&&a,i=ps(Ld(s),t,o),l=r.trainingConfig;if(l!=null&&i.loadTrainingConfig(l),r.userDefinedMetadata!=null&&i.setUserDefinedMetadata(r.userDefinedMetadata),r.weightData!=null){if(r.weightSpecs==null)throw new K("LayersModel artifacts contains weight data, but not weight specs. Therefore loading of weights cannot proceed.");let{modelWeights:u,optimizerWeights:c}=nne(r.weightData,r.weightSpecs);i.loadWeights(u,a),i.optimizer!=null&&c.length>0&&await i.optimizer.setWeights(c),Ge(u),Ge(c.map(d=>d.tensor))}return i}function nne(e,t){let n=cr.decodeWeights(e,t),r={},s=[];return t.forEach(a=>{a.group==="optimizer"?s.push({name:a.name,tensor:n[a.name]}):r[a.name]=n[a.name]}),{modelWeights:r,optimizerWeights:s}}var XA=class extends fa{constructor(e){super({inputs:[],outputs:[]});if(e=e||{},this.trainable=!0,this.built=!1,this.name=e.name!=null?e.name:Xf("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 K(`Negative dimension size caused by adding layer ${e.name} with input shape [${e.inboundNodes[0].inputTensors[0].shape}]`)}add(e){let t=e instanceof XA||e instanceof fa,n;if(t){if(n=e,n.outputs.length!==1)throw new K("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 K("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 K("The first layer in a Sequential model must get an `inputShape` or `batchInputShape` argument.");let r=uS({batchShape:e.batchInputShape,dtype:e.dtype,name:e.name+"_input"});e.apply(r)}if(t)this.outputs=n.outputs,this.inputs=n.inputs;else{if(e.inboundNodes.length!==1)throw new K(`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 K("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=lS(this.outputs[0])}this.inboundNodes=[],new Jf({outboundLayer:this,inboundLayers:[],nodeIndices:[],tensorIndices:[],inputTensors:this.inputs,outputTensors:this.outputs,inputMasks:ni(null,this.inputs.length),outputMasks:[null],inputShapes:this.inputs.map(r=>r.shape),outputShapes:this.outputs[0].shape})}else{let r=e.apply(this.outputs[0]);if(Array.isArray(r))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=[r],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(yt(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 fa({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 us("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 us("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 us("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 us("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={},r=!1){let s,a={};if(t instanceof Array){if(t[0].className==null||t[0].className==="Merge")throw new K("Legacy serialization format not supported yet.");s=t}else k.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."),s=t.layers,delete t.layers,a=t;let o=new e(a);if(!(o instanceof XA))throw new He(`Sequential.fromConfig called on non-Sequential input: ${o}`);for(let i of s){let u=ps(i,void 0,r);r&&u.setFastWeightInitDuringBuild(!0),o.add(u)}return o}set stopTraining(e){if(this.model==null)throw new K("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 K("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}}},im=XA;im.className="Sequential";ue.registerClass(im);function rne(e){return new fa(e)}function sne(e){return new im(e)}function ane(e,t){return t==null&&(t={}),ene(e,t)}function PS(e){return uS(e)}function one(e,t){MA.registerCallbackConstructor(e,t)}var er=class extends ue.Serializable{getConfig(){return{}}},zS=class extends er{apply(e,t=1){return _ee(e,t)}};zS.className="elu";ue.registerClass(zS);var LS=class extends er{apply(e){return iI(e)}};LS.className="selu";ue.registerClass(LS);var BS=class extends er{apply(e){return ca(e)}};BS.className="relu";ue.registerClass(BS);var WS=class extends er{apply(e){return Y(()=>Nf(6,ca(e)))}};WS.className="relu6";ue.registerClass(WS);var VS=class extends er{apply(e){return e}};VS.className="linear";ue.registerClass(VS);var US=class extends er{apply(e){return ia(e)}};US.className="sigmoid";ue.registerClass(US);var HS=class extends er{apply(e){return Fee(e)}};HS.className="hardSigmoid";ue.registerClass(HS);var GS=class extends er{apply(e){return Cd(e)}};GS.className="softplus";ue.registerClass(GS);var jS=class extends er{apply(e){return Dee(e)}};jS.className="softsign";ue.registerClass(jS);var qS=class extends er{apply(e){return Sd(e)}};qS.className="tanh";ue.registerClass(qS);var ZA=class extends er{apply(e,t=-1){return Yy(e,t)}};ZA.className="softmax";ue.registerClass(ZA);var KS=class extends er{apply(e,t=-1){return Y6(e,t)}};KS.className="logSoftmax";ue.registerClass(KS);var XS=class extends er{apply(e,t=1){return Y(()=>j(ia(j(e,t)),e))}};XS.className="swish";ue.registerClass(XS);var ZS=class extends er{apply(e){return Y(()=>j(e,Sd(Cd(e))))}};ZS.className="mish";ue.registerClass(ZS);function Xa(e){return e.getClassName()}function YA(e,t={}){return _d(e,ue.SerializationMap.getMap().classNameMap,t,"activation")}function Za(e){if(e==null){let t={};return t.className="linear",t.config={},YA(t)}if(typeof e=="string"){let t={};return t.className=e,t.config={},YA(t)}else return e instanceof er?e:YA(e)}function JA(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 YS=class extends ue.Serializable{},Vd=class extends YS{constructor(e){super();JA(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=ln([1]);return this.hasL1&&(t=de(t,Te(j(this.l1,gn(e))))),this.hasL2&&(t=de(t,Te(j(this.l2,Od(e))))),Z(t,[])})}getConfig(){return{l1:this.l1,l2:this.l2}}static fromConfig(e,t){return new e({l1:t.l1,l2:t.l2})}};Vd.className="L1L2";ue.registerClass(Vd);function ine(e){return JA(e),new Vd({l1:e!=null?e.l1:null,l2:0})}function lne(e){return JA(e),new Vd({l2:e!=null?e.l2:null,l1:0})}var JS={l1l2:"L1L2"};function vt(e){return pA(e)}function QS(e,t={}){return _d(e,ue.SerializationMap.getMap().classNameMap,t,"regularizer")}function zt(e){if(e==null)return null;if(typeof e=="string"){let n={className:e in JS?JS[e]:e,config:{}};return QS(n)}else return e instanceof YS?e:QS(e)}var QA=class extends tt{constructor(e){super(e==null?{}:e);this.supportsMasking=!0,e!=null&&(this.maxValue=e.maxValue)}call(e,t){e=qe(e);let n=ca(e);return this.maxValue!=null&&(n=Cr(n,0,this.maxValue)),n}computeOutputShape(e){return e}getConfig(){let e={maxValue:this.maxValue},t=super.getConfig();return Object.assign(e,t),e}};QA.className="ReLU";ue.registerClass(QA);var ex=class extends tt{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=qe(e);return Py(n,this.alpha)}computeOutputShape(e){return e}getConfig(){let e={alpha:this.alpha},t=super.getConfig();return Object.assign(e,t),e}};ex.className="LeakyReLU";ue.registerClass(ex);var tx=class extends tt{constructor(e){super(e==null?{}:e);if(this.DEFAULT_ALPHA_INITIALIZER="zeros",e==null&&(e={}),this.supportsMasking=!0,this.alphaInitializer=Pt(e.alphaInitializer||this.DEFAULT_ALPHA_INITIALIZER),this.alphaRegularizer=zt(e.alphaRegularizer),this.alphaConstraint=dn(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 K(`Expected sharedAxes to be a number or an array of numbers, but got ${e.sharedAxes}`)}build(e){e=yt(e);let t=e.slice(1);if(this.sharedAxes!=null)for(let r of this.sharedAxes)t[r-1]=1;this.alpha=this.addWeight("alpha",t,"float32",this.alphaInitializer,this.alphaRegularizer,!0,this.alphaConstraint);let n={};if(this.sharedAxes!=null)for(let r=1;r<e.length;++r)n[r]=e[r];this.inputSpec=[new Qt({ndim:e.length,axes:n})],this.built=!0}call(e,t){return e=qe(e),jy(e,this.alpha.read())}getConfig(){let e={alphaInitializer:Ut(this.alphaInitializer),alphaRegularizer:vt(this.alphaRegularizer),alphaConstraint:cn(this.alphaConstraint),sharedAxes:this.sharedAxes},t=super.getConfig();return Object.assign(e,t),e}};tx.className="PReLU";ue.registerClass(tx);var nx=class extends tt{constructor(e){super(e==null?{}:e);if(this.DEFAULT_ALPHA=1,e==null&&(e={}),e.alpha!=null&&e.alpha!==this.DEFAULT_ALPHA)throw new He(`Non-default alpha value (${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=qe(e);return If(n)}computeOutputShape(e){return e}getConfig(){let e={alpha:this.alpha},t=super.getConfig();return Object.assign(e,t),e}};nx.className="ELU";ue.registerClass(nx);var rx=class extends tt{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=qe(e);return j(n,xe($r(n,this.theta),"float32"))}computeOutputShape(e){return e}getConfig(){let e={theta:this.theta},t=super.getConfig();return Object.assign(e,t),e}};rx.className="ThresholdedReLU";ue.registerClass(rx);var sx=class extends tt{constructor(e){super(e==null?{}:e);this.DEFAULT_AXIS=1,e==null&&(e={}),this.softmax=new ZA().apply,this.axis=e.axis==null?this.DEFAULT_AXIS:e.axis}call(e,t){let n=qe(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}};sx.className="Softmax";ue.registerClass(sx);function su(e,t,n){if(typeof e=="number")return ni(e,t);if(e.length!==t)throw new K(`The ${n} argument must be an integer or tuple of ${t} integers. Received: ${e.length} elements.`);for(let r=0;r<t;++r){let s=e[r];if(!Cee(s))throw new K(`The ${n} argument must be an integer or tuple of ${t} integers. Received: ${JSON.stringify(e)} including a non-integer number ${s}`)}return e}function fs(e,t,n,r,s=1){if(e==null)return e;let a=t+(t-1)*(s-1),o;return n==="same"?o=e:o=e-a+1,Math.floor((o+r-1)/r)}function zs(e,t,n,r){if(e==null)return null;if(r==="valid")e=e*t+qa([n-t,0]);else if(r==="same")e=e*t;else throw new K(`Unsupport padding mode: ${r}.`);return e}function ax(e,t){return Y(()=>(Zt(t),t==="channelsFirst"?st(e,[0,2,3,1]):e))}function e8(e,t){return Y(()=>(Zt(t),t==="channelsFirst"?st(e,[0,2,3,4,1]):e))}function une(e,t,n,r=1,s="valid",a,o=1){return Y(()=>{if(a==null&&(a=ls()),Zt(a),e.shape.length!==3)throw new K(`The input of a conv1dWithBias operation should be 3, but is ${e.shape.length} instead.`);if(t.shape.length!==3)throw new K(`The kernel for a conv1dWithBias operation should be 3, but is ${t.shape.length} instead`);if(n!=null&&n.shape.length!==1)throw new K(`The bias for a conv1dWithBias operation should be 1, but is ${t.shape.length} instead`);if(a==="channelsFirst"&&(e=st(e,[0,2,1])),s==="causal")throw new He("The support for CAUSAL padding mode in conv1dWithBias is not implemented yet.");let i=U6(e,t,r,s==="same"?"same":"valid","NWC",o);return n!=null&&(i=ds(i,n)),i})}function t8(e,t,n,r=[1,1],s="valid",a,o,i=null){return Y(()=>{if(a==null&&(a=ls()),Zt(a),e.rank!==3&&e.rank!==4)throw new K(`conv2dWithBiasActivation expects input to be of rank 3 or 4, but received ${e.rank}.`);if(t.rank!==3&&t.rank!==4)throw new K(`conv2dWithBiasActivation expects kernel to be of rank 3 or 4, but received ${e.rank}.`);let l=ax(e,a);if(s==="causal")throw new He("The support for CAUSAL padding mode in conv1dWithBias is not implemented yet.");return l=ei.conv2d({x:l,filter:t,strides:r,pad:s==="same"?"same":"valid",dilations:o,dataFormat:"NHWC",bias:n,activation:i}),a==="channelsFirst"&&(l=st(l,[0,3,1,2])),l})}function cne(e,t,n,r=[1,1,1],s="valid",a,o){return Y(()=>{if(a==null&&(a=ls()),Zt(a),e.rank!==4&&e.rank!==5)throw new K(`conv3dWithBias expects input to be of rank 4 or 5, but received ${e.rank}.`);if(t.rank!==4&&t.rank!==5)throw new K(`conv3dWithBias expects kernel to be of rank 4 or 5, but received ${e.rank}.`);let i=e8(e,a);if(s==="causal")throw new He("The support for CAUSAL padding mode in conv3dWithBias is not implemented yet.");return i=G6(i,t,r,s==="same"?"same":"valid","NDHWC",o),n!=null&&(i=ds(i,n)),a==="channelsFirst"&&(i=st(i,[0,4,1,2,3])),i})}var ox=class extends tt{constructor(e,t){super(t);if(this.bias=null,this.DEFAULT_KERNEL_INITIALIZER="glorotNormal",this.DEFAULT_BIAS_INITIALIZER="zeros",ox.verifyArgs(t),this.rank=e,yn(this.rank,"rank"),this.rank!==1&&this.rank!==2&&this.rank!==3)throw new He(`Convolution layer for rank other than 1, 2, or 3 (${this.rank}) is not implemented yet.`);if(this.kernelSize=su(t.kernelSize,e,"kernelSize"),this.strides=su(t.strides==null?1:t.strides,e,"strides"),this.padding=t.padding==null?"valid":t.padding,_r(this.padding),this.dataFormat=t.dataFormat==null?"channelsLast":t.dataFormat,Zt(this.dataFormat),this.activation=Za(t.activation),this.useBias=t.useBias==null?!0:t.useBias,this.biasInitializer=Pt(t.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.biasConstraint=dn(t.biasConstraint),this.biasRegularizer=zt(t.biasRegularizer),this.activityRegularizer=zt(t.activityRegularizer),this.dilationRate=su(t.dilationRate==null?1:t.dilationRate,e,"dilationRate"),this.rank===1&&Array.isArray(this.dilationRate)&&this.dilationRate.length!==1)throw new K(`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 K(`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 K(`dilationRate must be a number or array of three numbers for 3D convolution, but received ${JSON.stringify(this.dilationRate)}`)}}static verifyArgs(e){if(Fs("kernelSize"in e,"required key 'kernelSize' not in config"),typeof e.kernelSize!="number"&&!mA(e.kernelSize,"number",1,3))throw new K(`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:Xa(this.activation),useBias:this.useBias,biasInitializer:Ut(this.biasInitializer),biasRegularizer:vt(this.biasRegularizer),activityRegularizer:vt(this.activityRegularizer),biasConstraint:cn(this.biasConstraint)},t=super.getConfig();return Object.assign(e,t),e}},Ud=class extends ox{constructor(e,t){super(e,t);this.kernel=null,Ud.verifyArgs(t),this.filters=t.filters,yn(this.filters,"filters"),this.kernelInitializer=Pt(t.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.kernelConstraint=dn(t.kernelConstraint),this.kernelRegularizer=zt(t.kernelRegularizer)}build(e){e=yt(e);let t=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[t]==null)throw new K(`The channel dimension of the input should be defined. Found ${e[t]}`);let n=e[t],r=this.kernelSize.concat([n,this.filters]);this.kernel=this.addWeight("kernel",r,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=qe(e);let n,r=this.bias==null?null:this.bias.read(),s=HI(this.activation.getClassName());if(s!=null&&this.rank===2)n=t8(e,this.kernel.read(),r,this.strides,this.padding,this.dataFormat,this.dilationRate,s);else{if(this.rank===1)n=une(e,this.kernel.read(),r,this.strides[0],this.padding,this.dataFormat,this.dilationRate[0]);else if(this.rank===2)n=t8(e,this.kernel.read(),r,this.strides,this.padding,this.dataFormat,this.dilationRate);else if(this.rank===3)n=cne(e,this.kernel.read(),r,this.strides,this.padding,this.dataFormat,this.dilationRate);else throw new He("convolutions greater than 3D are not implemented yet.");this.activation!=null&&(n=this.activation.apply(n))}return n})}computeOutputShape(e){e=yt(e);let t=[],n=this.dataFormat==="channelsLast"?e.slice(1,e.length-1):e.slice(2);for(let s=0;s<n.length;++s){let a=fs(n[s],this.kernelSize[s],this.padding,this.strides[s],typeof this.dilationRate=="number"?this.dilationRate:this.dilationRate[s]);t.push(a)}let r=[e[0]];return this.dataFormat==="channelsLast"?(r=r.concat(t),r.push(this.filters)):(r.push(this.filters),r=r.concat(t)),r}getConfig(){let e={filters:this.filters,kernelInitializer:Ut(this.kernelInitializer),kernelRegularizer:vt(this.kernelRegularizer),kernelConstraint:cn(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 K(`Convolution layer expected config.filters to be a 'number' > 0 but got ${JSON.stringify(e.filters)}`)}},n8=class extends Ud{constructor(e){super(2,e);n8.verifyArgs(e)}getConfig(){let e=super.getConfig();return delete e.rank,e}static verifyArgs(e){if(typeof e.kernelSize!="number"&&!mA(e.kernelSize,"number",1,2))throw new K(`Conv2D expects config.kernelSize to be number or number[] with length 1 or 2, but received ${JSON.stringify(e.kernelSize)}.`)}},lm=n8;lm.className="Conv2D";ue.registerClass(lm);var r8=class extends Ud{constructor(e){super(3,e);r8.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 K(`Conv3D expects config.kernelSize to be number or [number, number, number], but received ${JSON.stringify(e.kernelSize)}.`)}},um=r8;um.className="Conv3D";ue.registerClass(um);var ix=class extends lm{constructor(e){super(e);if(this.inputSpec=[new Qt({ndim:4})],this.padding!=="same"&&this.padding!=="valid")throw new K(`Conv2DTranspose currently supports only padding modes 'same' and 'valid', but received padding mode ${this.padding}`)}build(e){if(e=yt(e),e.length!==4)throw new K("Input should have rank 4; Received input shape: "+JSON.stringify(e));let t=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[t]==null)throw new K("The channel dimension of the inputs should be defined. Found `None`.");let n=e[t],r=this.kernelSize.concat([this.filters,n]);this.kernel=this.addWeight("kernel",r,"float32",this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.useBias&&(this.bias=this.addWeight("bias",[this.filters],"float32",this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint)),this.inputSpec=[new Qt({ndim:4,axes:{[t]:n}})],this.built=!0}call(e,t){return Y(()=>{let n=qe(e);if(n.shape.length!==4)throw new K(`Conv2DTranspose.call() expects input tensor to be rank-4, but received a tensor of rank-${n.shape.length}`);let r=n.shape,s=r[0],a,o;this.dataFormat==="channelsFirst"?(a=2,o=3):(a=1,o=2);let i=r[a],l=r[o],u=this.kernelSize[0],c=this.kernelSize[1],d=this.strides[0],h=this.strides[1],p=zs(i,d,u,this.padding),f=zs(l,h,c,this.padding),m=[s,p,f,this.filters];this.dataFormat!=="channelsLast"&&(n=st(n,[0,2,3,1]));let g=H6(n,this.kernel.read(),m,this.strides,this.padding);return this.dataFormat!=="channelsLast"&&(g=st(g,[0,3,1,2])),this.bias!=null&&(g=ds(g,this.bias.read(),this.dataFormat)),this.activation!=null&&(g=this.activation.apply(g)),g})}computeOutputShape(e){e=yt(e);let t=e.slice(),n,r,s;this.dataFormat==="channelsFirst"?(n=1,r=2,s=3):(n=3,r=1,s=2);let a=this.kernelSize[0],o=this.kernelSize[1],i=this.strides[0],l=this.strides[1];return t[n]=this.filters,t[r]=zs(t[r],i,a,this.padding),t[s]=zs(t[s],l,o,this.padding),t}getConfig(){let e=super.getConfig();return delete e.dilationRate,e}};ix.className="Conv2DTranspose";ue.registerClass(ix);var lx=class extends um{constructor(e){super(e);if(this.inputSpec=[new Qt({ndim:5})],this.padding!=="same"&&this.padding!=="valid")throw new K(`Conv3DTranspose currently supports only padding modes 'same' and 'valid', but received padding mode ${this.padding}`)}build(e){if(e=yt(e),e.length!==5)throw new K("Input should have rank 5; Received input shape: "+JSON.stringify(e));let t=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[t]==null)throw new K("The channel dimension of the inputs should be defined. Found `None`.");let n=e[t],r=this.kernelSize.concat([this.filters,n]);this.kernel=this.addWeight("kernel",r,"float32",this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint),this.useBias&&(this.bias=this.addWeight("bias",[this.filters],"float32",this.biasInitializer,this.biasRegularizer,!0,this.biasConstraint)),this.inputSpec=[new Qt({ndim:5,axes:{[t]:n}})],this.built=!0}call(e,t){return Y(()=>{let n=qe(e);if(n.shape.length!==5)throw new K(`Conv3DTranspose.call() expects input tensor to be rank-4, but received a tensor of rank-${n.shape.length}`);let r=n.shape,s=r[0],a,o,i;this.dataFormat==="channelsFirst"?(i=2,a=3,o=4):(i=1,a=2,o=3);let l=r[i],u=r[a],c=r[o],d=this.kernelSize[0],h=this.kernelSize[1],p=this.kernelSize[2],f=this.strides[0],m=this.strides[1],g=this.strides[2],y=zs(l,f,d,this.padding),A=zs(u,m,h,this.padding),x=zs(c,g,p,this.padding),b=[s,y,A,x,this.filters];this.dataFormat!=="channelsLast"&&(n=st(n,[0,2,3,4,1]));let v=Jj(n,this.kernel.read(),b,this.strides,this.padding);return this.dataFormat!=="channelsLast"&&(v=st(v,[0,4,1,2,3])),this.bias!==null&&(v=ds(v,this.bias.read(),this.dataFormat)),this.activation!==null&&(v=this.activation.apply(v)),v})}computeOutputShape(e){e=yt(e);let t=e.slice(),n,r,s,a;this.dataFormat==="channelsFirst"?(n=1,r=2,s=3,a=4):(n=4,r=1,s=2,a=3);let o=this.kernelSize[0],i=this.kernelSize[1],l=this.kernelSize[2],u=this.strides[0],c=this.strides[1],d=this.strides[2];return t[n]=this.filters,t[r]=zs(t[r],u,o,this.padding),t[s]=zs(t[s],c,i,this.padding),t[a]=zs(t[a],d,l,this.padding),t}getConfig(){let e=super.getConfig();return delete e.dilationRate,e}};lx.className="Conv3DTranspose";ue.registerClass(lx);var s8=class extends Ud{constructor(e,t){super(e,t);if(this.DEFAULT_DEPTHWISE_INITIALIZER="glorotUniform",this.DEFAULT_POINTWISE_INITIALIZER="glorotUniform",this.depthwiseKernel=null,this.pointwiseKernel=null,t.filters==null)throw new K("The `filters` configuration field is required by SeparableConv, but is unspecified.");if(t.kernelInitializer!=null||t.kernelRegularizer!=null||t.kernelConstraint!=null)throw new K("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 K(`SeparableConv${this.rank}D supports only padding modes: 'same' and 'valid', but received ${JSON.stringify(t.padding)}`);this.depthMultiplier=t.depthMultiplier==null?1:t.depthMultiplier,this.depthwiseInitializer=Pt(t.depthwiseInitializer||this.DEFAULT_DEPTHWISE_INITIALIZER),this.depthwiseRegularizer=zt(t.depthwiseRegularizer),this.depthwiseConstraint=dn(t.depthwiseConstraint),this.pointwiseInitializer=Pt(t.depthwiseInitializer||this.DEFAULT_POINTWISE_INITIALIZER),this.pointwiseRegularizer=zt(t.pointwiseRegularizer),this.pointwiseConstraint=dn(t.pointwiseConstraint)}build(e){if(e=yt(e),e.length<this.rank+2)throw new K(`Inputs to SeparableConv${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 K(`The channel dimension of the inputs should be defined, but found ${JSON.stringify(e[t])}`);let n=e[t],r=this.kernelSize.concat([n,this.depthMultiplier]),s=[];for(let o=0;o<this.rank;++o)s.push(1);s.push(n*this.depthMultiplier,this.filters);let a=!0;this.depthwiseKernel=this.addWeight("depthwise_kernel",r,"float32",this.depthwiseInitializer,this.depthwiseRegularizer,a,this.depthwiseConstraint),this.pointwiseKernel=this.addWeight("pointwise_kernel",s,"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 Qt({ndim:this.rank+2,axes:{[t]:n}})],this.built=!0}call(e,t){return Y(()=>{e=qe(e);let n;if(this.rank===1)throw new He("1D separable convolution is not implemented yet.");return this.rank===2&&(this.dataFormat==="channelsFirst"&&(e=st(e,[0,2,3,1])),n=oX(e,this.depthwiseKernel.read(),this.pointwiseKernel.read(),this.strides,this.padding,this.dilationRate,"NHWC")),this.useBias&&(n=ds(n,this.bias.read(),this.dataFormat)),this.activation!=null&&(n=this.activation.apply(n)),this.dataFormat==="channelsFirst"&&(n=st(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=Ut(this.depthwiseInitializer),e.pointwiseInitializer=Ut(this.pointwiseInitializer),e.depthwiseRegularizer=vt(this.depthwiseRegularizer),e.pointwiseRegularizer=vt(this.pointwiseRegularizer),e.depthwiseConstraint=cn(this.depthwiseConstraint),e.pointwiseConstraint=cn(this.pointwiseConstraint),e}};s8.className="SeparableConv";var ux=class extends s8{constructor(e){super(2,e)}};ux.className="SeparableConv2D";ue.registerClass(ux);var a8=class extends Ud{constructor(e){super(1,e);a8.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"&&!mA(e.kernelSize,"number",1,1))throw new K(`Conv1D expects config.kernelSize to be number or number[] with length 1, but received ${JSON.stringify(e.kernelSize)}.`)}},cx=a8;cx.className="Conv1D";ue.registerClass(cx);var dx=class extends tt{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=qe(e),this.dataFormat==="channelsLast"){let n=Lf(e,this.cropping[0][0],e.shape[1]-this.cropping[0][0]-this.cropping[0][1],2);return Lf(n,this.cropping[1][0],e.shape[2]-this.cropping[1][1]-this.cropping[1][0],3)}else{let n=Lf(e,this.cropping[0][0],e.shape[2]-this.cropping[0][0]-this.cropping[0][1],3);return Lf(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}};dx.className="Cropping2D";ue.registerClass(dx);var hx=class extends tt{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,Zt(this.dataFormat),this.interpolation=e.interpolation==null?"nearest":e.interpolation,See(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=qe(e),r=n.shape;if(this.dataFormat==="channelsFirst"){n=st(n,[0,2,3,1]);let s=this.size[0]*r[2],a=this.size[1]*r[3],o=this.interpolation==="nearest"?is.resizeNearestNeighbor(n,[s,a]):is.resizeBilinear(n,[s,a]);return st(o,[0,3,1,2])}else{let s=this.size[0]*r[1],a=this.size[1]*r[2];return this.interpolation==="nearest"?is.resizeNearestNeighbor(n,[s,a]):is.resizeBilinear(n,[s,a])}})}getConfig(){let e={size:this.size,dataFormat:this.dataFormat},t=super.getConfig();return Object.assign(e,t),e}};hx.className="UpSampling2D";ue.registerClass(hx);function dne(e,t,n=[1,1],r="valid",s,a){return Y(()=>{s==null&&(s=ls()),Zt(s);let o=ax(e,s);if(e.rank!==4)throw new K(`Input for depthwiseConv2d is required to be 4-D, but is instead ${e.rank}-D`);if(t.rank!==4)throw new K(`depthwiseKernel is required to be 4-D, but is instead ${t.rank}-D`);return o=kf(o,t,n,r==="same"?"same":"valid","NHWC",a),s==="channelsFirst"&&(o=st(o,[0,3,1,2])),o})}var px=class extends ox{constructor(e){super(2,e);this.depthwiseKernel=null,this.depthMultiplier=e.depthMultiplier==null?1:e.depthMultiplier,this.depthwiseInitializer=Pt(e.depthwiseInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.depthwiseConstraint=dn(e.depthwiseConstraint),this.depthwiseRegularizer=zt(e.depthwiseRegularizer)}build(e){if(e=yt(e),e.length<4)throw new K(`Inputs to DepthwiseConv2D should have rank 4. Received input shape: ${JSON.stringify(e)}.`);let t=this.dataFormat==="channelsFirst"?1:3;if(e[t]==null||e[t]<0)throw new K(`The channel dimension of the inputs to DepthwiseConv2D should be defined, but is not (${e[t]}).`);let n=e[t],r=[this.kernelSize[0],this.kernelSize[1],n,this.depthMultiplier];this.depthwiseKernel=this.addWeight("depthwise_kernel",r,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=qe(e);let n=dne(e,this.depthwiseKernel.read(),this.strides,this.padding,this.dataFormat,null);return this.useBias&&(n=ds(n,this.bias.read(),this.dataFormat)),this.activation!=null&&(n=this.activation.apply(n)),n})}computeOutputShape(e){e=yt(e);let t=this.dataFormat==="channelsFirst"?e[2]:e[1],n=this.dataFormat==="channelsFirst"?e[3]:e[2],r=this.dataFormat==="channelsFirst"?e[1]*this.depthMultiplier:e[3]*this.depthMultiplier,s=fs(t,this.kernelSize[0],this.padding,this.strides[0]),a=fs(n,this.kernelSize[1],this.padding,this.strides[1]);return this.dataFormat==="channelsFirst"?[e[0],r,s,a]:[e[0],s,a,r]}getConfig(){let e=super.getConfig();return e.depthMultiplier=this.depthMultiplier,e.depthwiseInitializer=Ut(this.depthwiseInitializer),e.depthwiseRegularizer=vt(this.depthwiseRegularizer),e.depthwiseConstraint=cn(this.depthwiseRegularizer),e}};px.className="DepthwiseConv2D";ue.registerClass(px);function o8(e,t,n,r){if(Array.isArray(e)){if(t!=null||n!=null)throw new K("When inputs is an array, neither initialState or constants should be provided");r!=null&&(n=e.slice(e.length-r,e.length),e=e.slice(0,e.length-r)),e.length>1&&(t=e.slice(1,e.length)),e=e[0]}function s(a){return a==null||Array.isArray(a)?a:[a]}return t=s(t),n=s(n),{inputs:e,initialState:t,constants:n}}function i8(e,t,n,r=!1,s,a,o=!1,i=!1){return Y(()=>{let l=t.shape.length;if(l<3)throw new K(`Input should be at least 3D, but is ${l}D.`);let u=[1,0].concat(cs(2,l));if(t=st(t,u),a!=null)throw new He("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."),s!=null&&(s=xe(xe(s,"bool"),"float32"),s.rank===l-1&&(s=Er(s,-1)),s=st(s,u)),r&&(t=Kr(t,0),s!=null&&(s=Kr(s,0)));let c=[],d,h=n,p=t.shape[0],f=Ds(t),m;s!=null&&(m=Ds(s));for(let y=0;y<p;++y){let A=f[y],x=Y(()=>e(A,h));if(s==null)d=x[0],h=x[1];else{let b=Y(()=>{let v=m[y],I=ke(qr(v),v),w=de(j(x[0],v),j(h[0],I)),S=h.map((E,D)=>de(j(x[1][D],v),j(E,I)));return{output:w,newStates:S}});d=b.output,h=b.newStates}i&&c.push(d)}let g;return i&&(g=Xr(c,1)),[d,g,h]})}var l8=class extends tt{constructor(e){super(e);let t;if(e.cell==null)throw new K("cell property is missing for the constructor of RNN.");if(Array.isArray(e.cell)?t=new hm({cells:e.cell}):t=e.cell,t.stateSize==null)throw new K("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 Qt({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 cs(0,e).map(t=>null)}else return this.states_}setStates(e){this.states_=e}computeOutputShape(e){_A(e)&&(e=e[0]),e=e;let t=this.cell.stateSize;Array.isArray(t)||(t=[t]);let n=t[0],r;if(this.returnSequences?r=[e[0],e[1],n]:r=[e[0],n],this.returnState){let s=[];for(let a of t)s.push([e[0],a]);return[r].concat(s)}else return r}computeMask(e,t){return Y(()=>{Array.isArray(t)&&(t=t[0]);let n=this.returnSequences?t:null;if(this.returnState){let r=this.states.map(s=>null);return[n].concat(r)}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){let t=null;if(this.numConstants!=null)throw new He("Constants support is not implemented in RNN yet.");_A(e)&&(e=e[0]),e=e;let n=this.stateful?e[0]:null,r=e.slice(2);this.inputSpec[0]=new Qt({shape:[n,null,...r]});let s=[e[0]].concat(e.slice(2));if(t!=null)throw new He("Constants support is not implemented in RNN yet.");this.cell.build(s);let a;if(Array.isArray(this.cell.stateSize)?a=this.cell.stateSize:a=[this.cell.stateSize],this.stateSpec!=null){if(!k.arraysEqual(this.stateSpec.map(o=>o.shape[o.shape.length-1]),a))throw new K(`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 Qt({shape:[null,o]}));this.stateful&&this.resetStates()}resetStates(e,t=!1){Y(()=>{if(!this.stateful)throw new ha("Cannot call resetStates() on an RNN Layer that is not stateful.");let n=this.inputSpec[0].shape[0];if(n==null)throw new K("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(r=>ln([n,r])):this.states_=[ln([n,this.cell.stateSize])];else if(e==null)Ge(this.states_),this.keptStates!=null&&(Ge(this.keptStates),this.keptStates=[]),Array.isArray(this.cell.stateSize)?this.states_=this.cell.stateSize.map(r=>ln([n,r])):this.states_[0]=ln([n,this.cell.stateSize]);else{if(Array.isArray(e)||(e=[e]),e.length!==this.states_.length)throw new K(`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()):Ge(this.states_);for(let r=0;r<this.states_.length;++r){let s=e[r],a=Array.isArray(this.cell.stateSize)?this.cell.stateSize[r]:this.cell.stateSize,o=[n,a];if(!k.arraysEqual(s.shape,o))throw new K(`State ${r} is incompatible with layer ${this.name}: expected shape=${o}, received shape=${s.shape}`);this.states_[r]=s}}this.states_=this.states_.map(r=>In(r.clone()))})}apply(e,t){let n=t==null?null:t.initialState,r=t==null?null:t.constants;t==null&&(t={});let s=o8(e,n,r,this.numConstants);e=s.inputs,n=s.initialState,r=s.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 Qt({shape:l.shape}));o=o.concat(this.stateSpec)}if(r!=null&&(t.constants=r,a=a.concat(r),this.numConstants=r.length),a[0]instanceof hs){let l=[e].concat(a),u=this.inputSpec.concat(o),c=this.inputSpec;this.inputSpec=u;let d=super.apply(l,t);return this.inputSpec=c,d}else return super.apply(e,t)}call(e,t){return Y(()=>{let n=t==null?null:t.mask,r=t==null?null:t.training,s=t==null?null:t.initialState;e=qe(e),s==null&&(this.stateful?s=this.states_:s=this.getInitialState(e));let a=Array.isArray(this.cell.stateSize)?this.cell.stateSize.length:1;if(s.length!==a)throw new K(`RNN Layer has ${a} state(s) but was passed ${s.length} initial state(s).`);this.unroll&&console.warn("Ignoring unroll = true for RNN layer, due to imperative backend.");let o={training:r},l=i8((p,f)=>{let m=this.cell.call([p].concat(f),o);return[m[0],m.slice(1)]},e,s,this.goBackwards,n,null,this.unroll,this.returnSequences),u=l[0],c=l[1],d=l[2];this.stateful&&this.resetStates(d,r);let h=this.returnSequences?c:u;return this.returnState?[h].concat(d):h})}getInitialState(e){return Y(()=>{let t=ln(e.shape);return t=Te(t,[1,2]),t=Md(t),Array.isArray(this.cell.stateSize)?this.cell.stateSize.map(n=>n>1?kA(t,[1,n]):t):this.cell.stateSize>1?[kA(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()===l8.className&&(t.cell={className:this.cell.getClassName(),config:n}),{...n,...e,...t}}static fromConfig(e,t,n={}){let r=t.cell,s=ps(r,n);return new e(Object.assign(t,{cell:s}))}},ma=l8;ma.className="RNN";ue.registerClass(ma);var Hd=class extends tt{},cm=class extends Hd{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,yn(this.units,"units"),this.activation=Za(e.activation==null?this.DEFAULT_ACTIVATION:e.activation),this.useBias=e.useBias==null?!0:e.useBias,this.kernelInitializer=Pt(e.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.recurrentInitializer=Pt(e.recurrentInitializer||this.DEFAULT_RECURRENT_INITIALIZER),this.biasInitializer=Pt(e.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.kernelRegularizer=zt(e.kernelRegularizer),this.recurrentRegularizer=zt(e.recurrentRegularizer),this.biasRegularizer=zt(e.biasRegularizer),this.kernelConstraint=dn(e.kernelConstraint),this.recurrentConstraint=dn(e.recurrentConstraint),this.biasConstraint=dn(e.biasConstraint),this.dropout=eu([1,qa([0,e.dropout==null?0:e.dropout])]),this.recurrentDropout=eu([1,qa([0,e.recurrentDropout==null?0:e.recurrentDropout])]),this.stateSize=this.units,this.dropoutMask=null,this.recurrentDropoutMask=null}build(e){e=yt(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 K(`SimpleRNNCell expects 2 input Tensors, got ${e.length}.`);let n=e[1];e=e[0];let r=t.training==null?!1:t.training;0<this.dropout&&this.dropout<1&&this.dropoutMask==null&&(this.dropoutMask=Ya({ones:()=>qr(e),rate:this.dropout,training:r})),0<this.recurrentDropout&&this.recurrentDropout<1&&this.recurrentDropoutMask==null&&(this.recurrentDropoutMask=Ya({ones:()=>qr(n),rate:this.recurrentDropout,training:r}));let s,a=this.dropoutMask,o=this.recurrentDropoutMask;a!=null?s=Ms(j(e,a),this.kernel.read()):s=Ms(e,this.kernel.read()),this.bias!=null&&(s=ds(s,this.bias.read())),o!=null&&(n=j(n,o));let i=de(s,Ms(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:Xa(this.activation),useBias:this.useBias,kernelInitializer:Ut(this.kernelInitializer),recurrentInitializer:Ut(this.recurrentInitializer),biasInitializer:Ut(this.biasInitializer),kernelRegularizer:vt(this.kernelRegularizer),recurrentRegularizer:vt(this.recurrentRegularizer),biasRegularizer:vt(this.biasRegularizer),activityRegularizer:vt(this.activityRegularizer),kernelConstraint:cn(this.kernelConstraint),recurrentConstraint:cn(this.recurrentConstraint),biasConstraint:cn(this.biasConstraint),dropout:this.dropout,recurrentDropout:this.recurrentDropout};return{...e,...t}}};cm.className="SimpleRNNCell";ue.registerClass(cm);var fx=class extends ma{constructor(e){e.cell=new cm(e),super(e)}call(e,t){return Y(()=>{this.cell.dropoutMask!=null&&(Ge(this.cell.dropoutMask),this.cell.dropoutMask=null),this.cell.recurrentDropoutMask!=null&&(Ge(this.cell.recurrentDropoutMask),this.cell.recurrentDropoutMask=null);let n=t==null?null:t.mask,r=t==null?null:t.training,s=t==null?null:t.initialState;return super.call(e,{mask:n,training:r,initialState:s})})}static fromConfig(e,t){return new e(t)}};fx.className="SimpleRNN";ue.registerClass(fx);var dm=class extends Hd{constructor(e){super(e);if(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 K("GRUCell does not support reset_after parameter set to true.");this.units=e.units,yn(this.units,"units"),this.activation=Za(e.activation===void 0?this.DEFAULT_ACTIVATION:e.activation),this.recurrentActivation=Za(e.recurrentActivation===void 0?this.DEFAULT_RECURRENT_ACTIVATION:e.recurrentActivation),this.useBias=e.useBias==null?!0:e.useBias,this.kernelInitializer=Pt(e.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.recurrentInitializer=Pt(e.recurrentInitializer||this.DEFAULT_RECURRENT_INITIALIZER),this.biasInitializer=Pt(e.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.kernelRegularizer=zt(e.kernelRegularizer),this.recurrentRegularizer=zt(e.recurrentRegularizer),this.biasRegularizer=zt(e.biasRegularizer),this.kernelConstraint=dn(e.kernelConstraint),this.recurrentConstraint=dn(e.recurrentConstraint),this.biasConstraint=dn(e.biasConstraint),this.dropout=eu([1,qa([0,e.dropout==null?0:e.dropout])]),this.recurrentDropout=eu([1,qa([0,e.recurrentDropout==null?0:e.recurrentDropout])]),this.implementation=e.implementation,this.stateSize=this.units,this.dropoutMask=null,this.recurrentDropoutMask=null}build(e){e=yt(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 K(`GRUCell expects 2 input Tensors (inputs, h, c), got ${e.length}.`);let n=t.training==null?!1:t.training,r=e[1];e=e[0],0<this.dropout&&this.dropout<1&&this.dropoutMask==null&&(this.dropoutMask=Ya({ones:()=>qr(e),rate:this.dropout,training:n,count:3})),0<this.recurrentDropout&&this.recurrentDropout<1&&this.recurrentDropoutMask==null&&(this.recurrentDropoutMask=Ya({ones:()=>qr(r),rate:this.recurrentDropout,training:n,count:3}));let s=this.dropoutMask,a=this.recurrentDropoutMask,o,i,l;0<this.dropout&&this.dropout<1&&(e=j(e,s[0]));let u=Ms(e,this.kernel.read());this.useBias&&(u=ds(u,this.bias.read())),0<this.recurrentDropout&&this.recurrentDropout<1&&(r=j(r,a[0]));let c=this.recurrentKernel.read(),[d,h]=Rr(c,[2*this.units,this.units],c.rank-1),p=Ms(r,d),[f,m,g]=Rr(u,3,u.rank-1),[y,A]=Rr(p,2,p.rank-1);o=this.recurrentActivation.apply(de(f,y)),i=this.recurrentActivation.apply(de(m,A));let x=Ms(j(i,r),h);l=this.activation.apply(de(g,x));let b=de(j(o,r),j(de(1,qt(o)),l));return[b,b]})}getConfig(){let e=super.getConfig(),t={units:this.units,activation:Xa(this.activation),recurrentActivation:Xa(this.recurrentActivation),useBias:this.useBias,kernelInitializer:Ut(this.kernelInitializer),recurrentInitializer:Ut(this.recurrentInitializer),biasInitializer:Ut(this.biasInitializer),kernelRegularizer:vt(this.kernelRegularizer),recurrentRegularizer:vt(this.recurrentRegularizer),biasRegularizer:vt(this.biasRegularizer),activityRegularizer:vt(this.activityRegularizer),kernelConstraint:cn(this.kernelConstraint),recurrentConstraint:cn(this.recurrentConstraint),biasConstraint:cn(this.biasConstraint),dropout:this.dropout,recurrentDropout:this.recurrentDropout,implementation:this.implementation,resetAfter:!1};return{...e,...t}}};dm.className="GRUCell";ue.registerClass(dm);var mx=class extends ma{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 dm(e),super(e)}call(e,t){return Y(()=>{this.cell.dropoutMask!=null&&(Ge(this.cell.dropoutMask),this.cell.dropoutMask=null),this.cell.recurrentDropoutMask!=null&&(Ge(this.cell.recurrentDropoutMask),this.cell.recurrentDropoutMask=null);let n=t==null?null:t.mask,r=t==null?null:t.training,s=t==null?null:t.initialState;return super.call(e,{mask:n,training:r,initialState:s})})}static fromConfig(e,t){return t.implmentation===0&&(t.implementation=1),new e(t)}};mx.className="GRU";ue.registerClass(mx);var Gd=class extends Hd{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,yn(this.units,"units"),this.activation=Za(e.activation===void 0?this.DEFAULT_ACTIVATION:e.activation),this.recurrentActivation=Za(e.recurrentActivation===void 0?this.DEFAULT_RECURRENT_ACTIVATION:e.recurrentActivation),this.useBias=e.useBias==null?!0:e.useBias,this.kernelInitializer=Pt(e.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.recurrentInitializer=Pt(e.recurrentInitializer||this.DEFAULT_RECURRENT_INITIALIZER),this.biasInitializer=Pt(e.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.unitForgetBias=e.unitForgetBias,this.kernelRegularizer=zt(e.kernelRegularizer),this.recurrentRegularizer=zt(e.recurrentRegularizer),this.biasRegularizer=zt(e.biasRegularizer),this.kernelConstraint=dn(e.kernelConstraint),this.recurrentConstraint=dn(e.recurrentConstraint),this.biasConstraint=dn(e.biasConstraint),this.dropout=eu([1,qa([0,e.dropout==null?0:e.dropout])]),this.recurrentDropout=eu([1,qa([0,e.recurrentDropout==null?0:e.recurrentDropout])]),this.implementation=e.implementation,this.stateSize=[this.units,this.units],this.dropoutMask=null,this.recurrentDropoutMask=null}build(e){var t;e=yt(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 r;if(this.useBias){if(this.unitForgetBias){let s=this.biasInitializer,a=this.units;r=new(t=class extends Yr{apply(o,i){let l=s.apply([a]),u=new Wf().apply([a]),c=s.apply([a*2]);return QI(QI(l,u),c)}},t.className="CustomInit",t)}else r=this.biasInitializer;this.bias=this.addWeight("bias",[this.units*4],null,r,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 K(`LSTMCell expects 3 input Tensors (inputs, h, c), got ${e.length}.`);let r=e[1],s=e[2];e=e[0],0<this.dropout&&this.dropout<1&&this.dropoutMask==null&&(this.dropoutMask=Ya({ones:()=>qr(e),rate:this.dropout,training:n,count:4})),0<this.recurrentDropout&&this.recurrentDropout<1&&this.recurrentDropoutMask==null&&(this.recurrentDropoutMask=Ya({ones:()=>qr(r),rate:this.recurrentDropout,training:n,count:4}));let a=this.dropoutMask,o=this.recurrentDropoutMask,i,l,u,c;0<this.dropout&&this.dropout<1&&(e=j(e,a[0]));let d=Ms(e,this.kernel.read());0<this.recurrentDropout&&this.recurrentDropout<1&&(r=j(r,o[0])),d=de(d,Ms(r,this.recurrentKernel.read())),this.useBias&&(d=ds(d,this.bias.read()));let[h,p,f,m]=Rr(d,4,d.rank-1);i=this.recurrentActivation.apply(h),l=this.recurrentActivation.apply(p),u=de(j(l,s),j(i,this.activation.apply(f))),c=this.recurrentActivation.apply(m);let g=j(c,this.activation.apply(u));return[g,g,u]})}getConfig(){let e=super.getConfig(),t={units:this.units,activation:Xa(this.activation),recurrentActivation:Xa(this.recurrentActivation),useBias:this.useBias,kernelInitializer:Ut(this.kernelInitializer),recurrentInitializer:Ut(this.recurrentInitializer),biasInitializer:Ut(this.biasInitializer),unitForgetBias:this.unitForgetBias,kernelRegularizer:vt(this.kernelRegularizer),recurrentRegularizer:vt(this.recurrentRegularizer),biasRegularizer:vt(this.biasRegularizer),activityRegularizer:vt(this.activityRegularizer),kernelConstraint:cn(this.kernelConstraint),recurrentConstraint:cn(this.recurrentConstraint),biasConstraint:cn(this.biasConstraint),dropout:this.dropout,recurrentDropout:this.recurrentDropout,implementation:this.implementation};return{...e,...t}}};Gd.className="LSTMCell";ue.registerClass(Gd);var gx=class extends ma{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 Gd(e),super(e)}call(e,t){return Y(()=>{this.cell.dropoutMask!=null&&(Ge(this.cell.dropoutMask),this.cell.dropoutMask=null),this.cell.recurrentDropoutMask!=null&&(Ge(this.cell.recurrentDropoutMask),this.cell.recurrentDropoutMask=null);let n=t==null?null:t.mask,r=t==null?null:t.training,s=t==null?null:t.initialState;return super.call(e,{mask:n,training:r,initialState:s})})}static fromConfig(e,t){return t.implmentation===0&&(t.implementation=1),new e(t)}};gx.className="LSTM";ue.registerClass(gx);var hm=class extends Hd{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),r=[];for(let o of this.cells.slice().reverse())Array.isArray(o.stateSize)?r.push(n.splice(0,o.stateSize.length)):r.push(n.splice(0,1));r.reverse();let s=[],a;for(let o=0;o<this.cells.length;++o){let i=this.cells[o];n=r[o],o===0?a=[e[0]].concat(n):a=[a[0]].concat(n),a=i.call(a,t),s.push(a.slice(1))}n=[];for(let o of s.slice().reverse())n.push(...o);return[a[0]].concat(n)})}build(e){_A(e)&&(e=e[0]),e=e;let t;this.cells.forEach((n,r)=>{ai(`RNNCell_${r}`,()=>{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=s=>({className:s.getClassName(),config:s.getConfig()}),r={cells:this.cells.map(t)};return{...e,...r}}static fromConfig(e,t,n={}){let r=[];for(let s of t.cells)r.push(ps(s,n));return new e({cells:r})}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 DA(e)}setWeights(e){let t=[];for(let n of this.cells){let r=n.weights.length,s=e.splice(r);for(let a=0;a<n.weights.length;++a)t.push([n.weights[a],s[a]])}FA(t)}};hm.className="StackedRNNCells";ue.registerClass(hm);function Ya(e){let{ones:t,rate:n,training:r=!1,count:s=1}=e,a=()=>tS(t(),n),o=()=>Pd(a,t,r);return!s||s<=1?In(o().clone()):Array(s).fill(void 0).map(o).map(l=>In(l.clone()))}var u8=class extends ma{constructor(e){if(e.unroll)throw new He("Unrolling is not possible with convolutional RNNs.");if(Array.isArray(e.cell))throw new He("It is not possible at the moment to stack convolutional cells.");super(e);this.inputSpec=[new Qt({ndim:5})]}call(e,t){return Y(()=>{if(this.cell.dropoutMask!=null&&(Ge(this.cell.dropoutMask),this.cell.dropoutMask=null),this.cell.recurrentDropoutMask!=null&&(Ge(this.cell.recurrentDropoutMask),this.cell.recurrentDropoutMask=null),t&&t.constants)throw new K("ConvRNN2D cell does not support constants");let n=t==null?null:t.mask,r=t==null?null:t.training,s=t==null?null:t.initialState;return super.call(e,{mask:n,training:r,initialState:s})})}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,r=this.computeSingleOutputShape(n),s=[r[0],...r.slice(2)],a=ln(s);return Array.isArray(t)?Array(t.length).fill(a):[a]})}resetStates(e,t=!1){Y(()=>{if(!this.stateful)throw new ha("Cannot call resetStates() on an RNN Layer that is not stateful.");let n=this.inputSpec[0].shape,r=this.computeSingleOutputShape(n),s=[r[0],...r.slice(2)];if(n[0]==null)throw new K("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(()=>ln(s)):this.states_=[ln(s)];else if(e==null)Ge(this.states_),this.keptStates!=null&&(Ge(this.keptStates),this.keptStates=[]),Array.isArray(this.cell.stateSize)?this.states_=this.cell.stateSize.map(()=>ln(s)):this.states_[0]=ln(s);else{if(Array.isArray(e)||(e=[e]),e.length!==this.states_.length)throw new K(`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()):Ge(this.states_);for(let o=0;o<this.states_.length;++o){let i=e[o],l=s;if(!k.arraysEqual(i.shape,l))throw new K(`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=>In(o.clone()))})}computeSingleOutputShape(e){let{dataFormat:t,filters:n,kernelSize:r,padding:s,strides:a,dilationRate:o}=this.cell,i=t==="channelsFirst",l=e[i?3:2],u=e[i?4:3],c=fs(l,r[0],s,a[0],o[0]),d=fs(u,r[1],s,a[1],o[1]);return[...e.slice(0,2),...i?[n,c,d]:[c,d,n]]}};u8.className="ConvRNN2D";var pm=class extends Gd{constructor(e){let{filters:t,kernelSize:n,strides:r,padding:s,dataFormat:a,dilationRate:o}=e;super({...e,units:t});this.filters=t,yn(this.filters,"filters"),this.kernelSize=su(n,2,"kernelSize"),this.kernelSize.forEach(i=>yn(i,"kernelSize")),this.strides=su(r||1,2,"strides"),this.strides.forEach(i=>yn(i,"strides")),this.padding=s||"valid",_r(this.padding),this.dataFormat=a||"channelsLast",Zt(this.dataFormat),this.dilationRate=su(o||1,2,"dilationRate"),this.dilationRate.forEach(i=>yn(i,"dilationRate"))}build(e){var t;e=yt(e);let n=this.dataFormat==="channelsFirst"?1:e.length-1;if(e[n]==null)throw new K(`The channel dimension of the input should be defined. Found ${e[n]}`);let r=e[n],s=4,a=this.kernelSize.concat([r,this.filters*s]);this.kernel=this.addWeight("kernel",a,null,this.kernelInitializer,this.kernelRegularizer,!0,this.kernelConstraint);let o=this.kernelSize.concat([this.filters,this.filters*s]);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 Yr{apply(c,d){let h=l.apply([u]),p=ua([u]),f=l.apply([u*2]);return wA([h,p,f])}},t.className="CustomInit",t)}else i=this.biasInitializer;this.bias=this.addWeight("bias",[this.filters*s],null,i,this.biasRegularizer,!0,this.biasConstraint)}this.built=!0}call(e,t){return Y(()=>{if(e.length!==3)throw new K(`ConvLSTM2DCell expects 3 input Tensors (inputs, h, c), got ${e.length}.`);let n=t.training||!1,r=e[0],s=e[1],a=e[2],o=4;0<this.dropout&&this.dropout<1&&this.dropoutMask==null&&(this.dropoutMask=Ya({ones:()=>qr(r),rate:this.dropout,training:n,count:o}));let i=this.dropoutMask,l=(ee,ae,se)=>!ae||!ae[se]?ee:j(ae[se],ee),u=l(r,i,0),c=l(r,i,1),d=l(r,i,2),h=l(r,i,3);0<this.recurrentDropout&&this.recurrentDropout<1&&this.recurrentDropoutMask==null&&(this.recurrentDropoutMask=Ya({ones:()=>qr(s),rate:this.recurrentDropout,training:n,count:o}));let p=this.recurrentDropoutMask,f=l(s,p,0),m=l(s,p,1),g=l(s,p,2),y=l(s,p,3),A=3,[x,b,v,I]=Rr(this.kernel.read(),o,A),[w,S,E,D]=this.useBias?Rr(this.bias.read(),o):[null,null,null,null];u=this.inputConv(u,x,w,this.padding),c=this.inputConv(c,b,S,this.padding),d=this.inputConv(d,v,E,this.padding),h=this.inputConv(h,I,D,this.padding);let[$,R,N,M]=Rr(this.recurrentKernel.read(),o,A);f=this.recurrentConv(f,$),m=this.recurrentConv(m,R),g=this.recurrentConv(g,N),y=this.recurrentConv(y,M);let B=this.recurrentActivation.apply(de(u,f)),q=this.recurrentActivation.apply(de(c,m)),X=de(j(q,a),j(B,this.activation.apply(de(d,g)))),J=j(this.recurrentActivation.apply(de(h,y)),this.activation.apply(X));return[J,J,X]})}getConfig(){let{units:e,...t}=super.getConfig(),n={filters:this.filters,kernelSize:this.kernelSize,padding:this.padding,dataFormat:this.dataFormat,dilationRate:this.dilationRate,strides:this.strides};return{...t,...n}}inputConv(e,t,n,r){let s=Xo(e,t,this.strides,r||"valid",this.dataFormat==="channelsFirst"?"NCHW":"NHWC",this.dilationRate);return n?ds(s,n,this.dataFormat):s}recurrentConv(e,t){return Xo(e,t,1,"same",this.dataFormat==="channelsFirst"?"NCHW":"NHWC")}};pm.className="ConvLSTM2DCell";ue.registerClass(pm);var yx=class extends u8{constructor(e){let t=new pm(e);super({...e,cell:t})}static fromConfig(e,t){return new e(t)}};yx.className="ConvLSTM2D";ue.registerClass(yx);var fm=class extends tt{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 r=0;r<this.noiseShape.length;++r)n.push(this.noiseShape[r]==null?t[r]:this.noiseShape[r]);return n}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=qe(e);if(0<this.rate&&this.rate<1){let r=t.training==null?!1:t.training,s=this.getNoiseShape(n);return Pd(()=>tS(n,this.rate,s,this.seed),()=>n,r)}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()}};fm.className="Dropout";ue.registerClass(fm);var Ax=class extends fm{constructor(e){super(e);this.inputSpec=[{ndim:3}]}getNoiseShape(e){let t=e.shape;return[t[0],1,t[2]]}};Ax.className="SpatialDropout1D";ue.registerClass(Ax);var xx=class extends tt{constructor(e){super(e);if(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,yn(this.units,"units"),this.activation=Za(e.activation),e.useBias!=null&&(this.useBias=e.useBias),this.kernelInitializer=Pt(e.kernelInitializer||this.DEFAULT_KERNEL_INITIALIZER),this.biasInitializer=Pt(e.biasInitializer||this.DEFAULT_BIAS_INITIALIZER),this.kernelConstraint=dn(e.kernelConstraint),this.biasConstraint=dn(e.biasConstraint),this.kernelRegularizer=zt(e.kernelRegularizer),this.biasRegularizer=zt(e.biasRegularizer),this.activityRegularizer=zt(e.activityRegularizer),this.supportsMasking=!0,this.inputSpec=[{minNDim:2}]}build(e){e=yt(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=yt(e);let t=e.slice();return t[t.length-1]=this.units,t}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=qe(e),r=HI(this.activation.getClassName()),s;return r!=null?s=Ms(n,this.kernel.read(),r,this.bias?this.bias.read():null):(s=Ms(n,this.kernel.read()),this.bias!=null&&(s=ds(s,this.bias.read())),this.activation!=null&&(s=this.activation.apply(s))),s})}getConfig(){let e={units:this.units,activation:Xa(this.activation),useBias:this.useBias,kernelInitializer:Ut(this.kernelInitializer),biasInitializer:Ut(this.biasInitializer),kernelRegularizer:vt(this.kernelRegularizer),biasRegularizer:vt(this.biasRegularizer),activityRegularizer:vt(this.activityRegularizer),kernelConstraint:cn(this.kernelConstraint),biasConstraint:cn(this.biasConstraint)},t=super.getConfig();return Object.assign(e,t),e}};xx.className="Dense";ue.registerClass(xx);var bx=class extends tt{constructor(e){e=e||{},super(e),this.inputSpec=[{minNDim:3}],this.dataFormat=e.dataFormat}computeOutputShape(e){e=yt(e);for(let t of e.slice(1))if(t==null)throw new K(`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],ja(e,1)]}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=qe(e);if(this.dataFormat==="channelsFirst"&&n.rank>1){let r=[0];for(let s=2;s<n.rank;++s)r.push(s);r.push(1),n=st(n,r)}return Ree(n)})}getConfig(){let e={};this.dataFormat!=null&&(e.dataFormat=this.dataFormat);let t=super.getConfig();return Object.assign(e,t),e}};bx.className="Flatten";ue.registerClass(bx);var vx=class extends tt{constructor(e){super(e);this.supportsMasking=!0,this.activation=Za(e.activation)}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=qe(e);return this.activation.apply(n)})}getConfig(){let e={activation:Xa(this.activation)},t=super.getConfig();return Object.assign(e,t),e}};vx.className="Activation";ue.registerClass(vx);var wx=class extends tt{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=qe(e),Eee(e,this.n)))}getConfig(){let e={n:this.n},t=super.getConfig();return Object.assign(e,t),e}};wx.className="RepeatVector";ue.registerClass(wx);var kx=class extends tt{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.",r=t.slice(),s=1,a=null;for(let i=0;i<r.length;++i){let l=r[i];if(this.isUnknown(l))if(a===null)a=i;else throw new K("Can only specifiy one unknown dimension.");else s*=l}let o=ja(e);if(a!==null){if(s===0||o%s!=0)throw new K(n);r[a]=o/s}else if(o!==s)throw new K(n);return r}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=qe(e),r=n.shape,s=r.slice(0,1).concat(this.fixUnknownDimension(r.slice(1),this.targetShape));return Z(n,s)})}getConfig(){let e={targetShape:this.targetShape},t=super.getConfig();return Object.assign(e,t),e}};kx.className="Reshape";ue.registerClass(kx);var Ix=class extends tt{constructor(e){super(e);if(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=cs(1,e.dims.length+1);if(!k.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 Qt({ndim:this.dims.length+1})]}computeOutputShape(e){e=yt(e);let t=e.slice();return this.dims.forEach((n,r)=>{t[r+1]=e[n]}),t}call(e,t){return st(qe(e),this.dimsIncludingBatch)}getConfig(){let e={dims:this.dims},t=super.getConfig();return Object.assign(e,t),e}};Ix.className="Permute";ue.registerClass(Ix);var Sx=class extends tt{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=qe(e),r=-1;return Ey(Ed(n,this.maskValue),r)}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=qe(e),r=-1,s=!0,a=Ey(Ed(n,this.maskValue),r,s);return j(n,xe(a,n.dtype))})}};Sx.className="Masking";ue.registerClass(Sx);var Tx=class extends tt{constructor(e){super(e);if(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(Rt(e.inputLength))}this.inputDim=e.inputDim,yn(this.inputDim,"inputDim"),this.outputDim=e.outputDim,yn(this.outputDim,"outputDim"),this.embeddingsInitializer=Pt(e.embeddingsInitializer||this.DEFAULT_EMBEDDINGS_INITIALIZER),this.embeddingsRegularizer=zt(e.embeddingsRegularizer),this.activityRegularizer=zt(e.activityRegularizer),this.embeddingsConstraint=dn(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=qe(e),Ed(e,ot(e))):null)}computeOutputShape(e){if(e=yt(e),this.inputLength==null)return[...e,this.outputDim];let t=Rt(this.inputLength);if(t.length!==e.length-1)throw new K(`"inputLength" is ${this.inputLength}, but received input shape has shape ${e}`);{let n=0;for(let r=0;r<t.length;++r){let s=t[r],a=e[r+1];if(s!=null&&a!=null&&s!==a)throw new K(`"inputLength" is ${this.inputLength}, but received input shape has shape ${e}`);s==null&&(t[n]=a),n++}}return[e[0],...t,this.outputDim]}call(e,t){return Y(()=>{this.invokeCallHook(e,t);let n=qe(e);n.dtype!=="int32"&&(n=zf(n,"int32"));let r=eS(this.embeddings.read(),Z(n,[n.size]));return Z(r,yt(this.computeOutputShape(n.shape)))})}getConfig(){let e={inputDim:this.inputDim,outputDim:this.outputDim,embeddingsInitializer:Ut(this.embeddingsInitializer),embeddingsRegularizer:vt(this.embeddingsRegularizer),activityRegularizer:vt(this.activityRegularizer),embeddingsConstraint:cn(this.embeddingsConstraint),maskZero:this.maskZero,inputLength:this.inputLength},t=super.getConfig();return Object.assign(e,t),e}};Tx.className="Embedding";ue.registerClass(Tx);var ci=class extends tt{constructor(e){super(e||{});this.supportsMasking=!0}mergeFunction(e){throw new He}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 r=0;r<t.length;++r){let s=e[e.length-t.length+r],a=t[r];if(s==null||a==null||s<0||a<0)n.push(null);else if(s===1)n.push(a);else if(a===1)n.push(s);else{if(s!==a)throw new K("Operands could not be broadcast together with shapes "+JSON.stringify(e)+" "+JSON.stringify(t));n.push(s)}}return n}build(e){if(Array.isArray(e)&&!Array.isArray(e[0])&&(e=[yt(e)]),e=e,e.length<2)throw new K(`A merge layer should be called on an Array of at least 2 inputs. Got ${e.length} input(s).`);let t=[];for(let s of e)s!=null&&s[0]!==null&&t.push(s[0]);if(t=Ga(t),t.length>1)throw new K(`Can not merge tensors with different batch sizes. Got tensors with shapes: ${JSON.stringify(e)}.`);let n=e[0]==null?null:e[0].slice(1);for(let s=1;s<e.length;++s){let a=e[s]==null?null:e[s].slice(1);n=this.computeElementwiseOpOutputShape(n,a)}let r=e.map(s=>s.length);e.indexOf(null)===-1&&Ga(r).length===1?this.reshapeRequired=!1:this.reshapeRequired=!0}call(e,t){return Y(()=>{if(e=e,this.reshapeRequired){let n=[],r=e.map(s=>s.rank);if(r.indexOf(null)===-1){let s=qa(r);for(let a of e){let o=a.rank;for(let i=0;i<s-o;++i)a=Md(a,1);n.push(a)}return this.mergeFunction(n)}else{let s=!1;for(let i of e){let l=i.rank;if(l==null){let u=i.shape,c=u[0],d=u.slice(1).concat([c]),h=Z(i,[c].concat(ja(u.slice(1))));h=st(h,[1,0]),h=Z(h,d),n.push(h),s=!0}else if(l>1){let u=cs(1,l).concat([0]);n.push(st(i,u)),s=!0}else n.push(i)}let a=this.mergeFunction(n),o=a.rank;if(s){if(o==null){let i=a.shape,l=i.length,u=i[l-1],c=[u].concat(i.slice(0,i.length-1));a=Z(st(Z(a,[-1,u]),[1,0]),c)}else if(o>1){let i=[o-1].concat(cs(0,o-1));a=st(a,i)}}return a}}else return this.mergeFunction(e)})}computeOutputShape(e){e=e;let t;e[0]==null?t=null:t=e[0].slice(1);for(let r=1;r<e.length;++r){let s=e[r]==null?null:e[r].slice(1);t=this.computeElementwiseOpOutputShape(t,s)}let n=[];for(let r of e)r!=null&&r[0]!==null&&n.push(r[0]);return n=Ga(n),n.length===1?t=n.concat(t):t=[null].concat(t),t}computeMask(e,t){return Y(()=>{if(t==null)return null;if(!Array.isArray(t))throw new K("`mask` should be an Array");if(!Array.isArray(e))throw new K("`inputs` should be an Array");if(t.length!==e.length)throw new K(`The Array 'inputs' and 'mask' are expected to have the same length, but have different lengths (${e.length} vs ${t.length})`);if(t.every(r=>r==null))return null;t=t.map(r=>r==null?r:Er(r,0));let n=t[0];for(let r=1;r<t.length-1;++r)n=_s(n,t[r]);return n})}},Nx=class extends ci{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0].clone();for(let n=1;n<e.length;++n)t=de(t,e[n]);return t})}};Nx.className="Add";ue.registerClass(Nx);var Cx=class extends ci{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0].clone();for(let n=1;n<e.length;++n)t=j(t,e[n]);return t})}};Cx.className="Multiply";ue.registerClass(Cx);var Ex=class extends ci{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0].clone();for(let n=1;n<e.length;++n)t=de(t,e[n]);return j(1/e.length,t)})}};Ex.className="Average";ue.registerClass(Ex);var $x=class extends ci{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0];for(let n=1;n<e.length;++n)t=Va(t,e[n]);return t})}};$x.className="Maximum";ue.registerClass($x);var Rx=class extends ci{constructor(e){super(e)}mergeFunction(e){return Y(()=>{let t=e[0];for(let n=1;n<e.length;++n)t=Nf(t,e[n]);return t})}};Rx.className="Minimum";ue.registerClass(Rx);var _x=class extends ci{constructor(e){super(e);this.DEFAULT_AXIS=-1,e==null&&(e={}),this.axis=e.axis==null?this.DEFAULT_AXIS:e.axis,this.supportsMasking=!0,this.reshapeRequired=!1}build(e){if(!(Array.isArray(e)&&Array.isArray(e[0]))||e.length===1)throw new K("A `Concatenate` layer should be called on a list of at least 2 inputs");e=e;let t=!0;for(let r of e)if(r!=null){t=!1;break}if(t)return;let n=[];for(let r=0;r<e.length;++r){let s=e[r].slice();s.splice(this.axis,1);let a=!1;for(let o of n)if(k.arraysEqual(o,s)){a=!0;break}a||n.push(s)}if(n.length>1)throw new K("A `Concatenate` layer requires inputs with matching shapes except for the concat axis. Got input shapes: "+JSON.stringify(e))}mergeFunction(e){return Y(()=>wA(e,this.axis))}computeOutputShape(e){if(!(Array.isArray(e)&&Array.isArray(e[0])))throw new K("A `Concatenate` layer should be called on a list of inputs.");let t=e,n=t[0].slice(),r=this.axis<0?n.length+this.axis:this.axis;for(let s of t.slice(1)){if(n[r]==null||s[r]==null){n[r]=null;break}n[r]+=s[r]}return n}computeMask(e,t){if(t==null)return null;if(!Array.isArray(t))throw new K("`mask` should be an array for Concatenate");if(!Array.isArray(e))throw new K("`inputs` should be an array for Concatenate");if(t.length!==e.length)throw new K(`Mismatch in the length of mask (${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 r=[];for(let a=0;a<e.length;++a)t[a]==null?r.push(xe(qr(e[a]),"bool")):t[a].rank<e[a].rank?r.push(Er(t[a],-1)):r.push(t[a]);let s=an(r,this.axis);return P6(s,-1,!1)})}getConfig(){let e={axis:this.axis},t=super.getConfig();return Object.assign(e,t),e}};_x.className="Concatenate";ue.registerClass(_x);function jd(e,t){for(;e<0;)e+=t;return e}function hne(e,t,n){if(e.shape.length>3||t.shape.length>3)throw new He("batchDot is not implemented for tensors of 4D or higher rank yet");if(k.assert(e.shape.length>=2,()=>`batchDot requires the rank of x to be >= 2, but got ${e.shape.length}`),k.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 He("batchDot is not implemented for complex64-type Tensors yet.");let r=e.shape.length,s=t.shape.length;n==null&&(n=[r-1,s-2]);let a=n;return Y(()=>{let o;if(r>s){o=r-s;let l=[];for(let u=0;u<o;++u)l.push(1);t=Z(t,t.shape.concat(l))}else if(s>r){o=s-r;let l=[];for(let u=0;u<o;++u)l.push(1);e=Z(e,e.shape.concat(l))}else o=0;let i;if(e.shape.length===2&&t.shape.length===2)a[0]===a[1]?i=Te(j(e,t),a[0]):i=Te(j(st(e,[1,0]),t),a[1]);else{let l=a[0]!==e.shape.length-1,u=a[1]===t.shape.length-1;i=rt(e,t,l,u)}if(o>0){let l;r>s?l=r+s-3:l=r-1;let u=[];for(let c=l;c<l+o;++c)u.push(c);i=Qo(i,u)}return i.shape.length===1&&(i=Er(i,1)),i})}var Dx=class extends ci{constructor(e){super(e);this.axes=e.axes,this.normalize=e.normalize==null?!1:e.normalize,this.supportsMasking=!0,this.reshapeRequired=!1}build(e){k.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 He("Dot layer does not support tensors of 4D or higher rank yet.");let r=this.interpretAxes(t,n);if(t[r[0]]!==n[r[1]])throw new K(`Dimension incompatibility: ${t[r[0]]} !== ${n[r[1]]}`)}mergeFunction(e){if(e.length!==2)throw new K(`A \`Dot\` layer must be called on exactly 2 inputs, but received ${e.length} input(s).`);let t=e[0],n=e[1],r;return Array.isArray(this.axes)?r=this.axes.map((s,a)=>jd(s,e[a].shape.length)):r=[jd(this.axes,t.shape.length),jd(this.axes,n.shape.length)],this.normalize&&(t=Qf(t,r[0]),n=Qf(n,r[1])),hne(t,n,r)}interpretAxes(e,t){let n;return Array.isArray(this.axes)?n=this.axes:n=[jd(this.axes,e.length),jd(this.axes,t.length)],n}computeOutputShape(e){k.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 He("Dot layer does not support tensors of 4D or higher rank yet.");let r=this.interpretAxes(t,n);t.splice(r[0],1),n.splice(r[1],1),n.splice(0,1);let s=t.concat(n);return s.length===1&&s.push(1),s}computeMask(e,t){return null}getConfig(){let e={axes:this.axes,normalize:this.normalize},t=super.getConfig();return Object.assign(e,t),e}};Dx.className="Dot";ue.registerClass(Dx);var Fx=class extends tt{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=qe(e);return Pd(()=>de(Bf(n.shape,0,this.stddev),n),()=>n,t.training||!1)})}};Fx.className="GaussianNoise";ue.registerClass(Fx);var Mx=class extends tt{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=qe(e);return this.rate>0&&this.rate<1?Pd(()=>{let s=Math.sqrt(this.rate/(1-this.rate));return j(n,Bf(n.shape,1,s))},()=>n,t.training||!1):n})}};Mx.className="GaussianDropout";ue.registerClass(Mx);var Ox=class extends tt{constructor(e){super(e);this.supportsMasking=!0,this.rate=e.rate,this.noiseShape=e.noiseShape}_getNoiseShape(e){return this.noiseShape||qe(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 Pd(()=>{let s=qe(e),a=1.6732632423543772,o=1.0507009873554805,i=-a*o,l=Kl($d(n),this.rate);l=zf(l,"float32");let u=((1-this.rate)*(1+this.rate*i**2))**-.5,c=-u*i*this.rate,d=de(j(s,l),j(de(l,-1),i));return de(j(d,u),c)},()=>qe(e),t.training||!1)}return e})}};Ox.className="AlphaDropout";ue.registerClass(Ox);function qd(e,t,n,r,s,a=.001){let o;if(e.rank===2)o=Nj(e,t,n,r,s,a);else if(e.rank===3)o=Ej(e,t,n,r,s,a);else if(e.rank===4)o=Rj(e,t,n,r,s,a);else throw new He(`batchNormalization is not implemented for array of rank ${e.rank} yet`);return o}function pne(e,t,n,r,s=.001){return Y(()=>{let a=Hy(e,r),o=a.mean,i=a.variance;return[qd(e,o,i,n,t,s),o,i]})}function fne(e,t,n,r,s=.001){return Y(()=>{let a=Hy(e,r),o=a.mean,i=a.variance,l=[];for(let f of cs(0,e.rank))r.indexOf(f)!==-1?l.push(1):l.push(e.shape[f]);let u=Z(o,l),c=Z(i,l),d=t==null?null:Z(t,l),h=n==null?null:Z(n,l);return[qd(e,u,c,h,d,s),o,i]})}function mne(e,t,n,r,s=.001){return k.arraysEqual(r.slice().sort(),cs(0,e.rank-1))?pne(e,t,n,r,s):fne(e,t,n,r,s)}var Px=class extends tt{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=Pt(e.betaInitializer||"zeros"),this.gammaInitializer=Pt(e.gammaInitializer||"ones"),this.movingMeanInitializer=Pt(e.movingMeanInitializer||"zeros"),this.movingVarianceInitializer=Pt(e.movingVarianceInitializer||"ones"),this.betaConstraint=dn(e.betaConstraint),this.gammaConstraint=dn(e.gammaConstraint),this.betaRegularizer=zt(e.betaRegularizer),this.gammaRegularizer=zt(e.gammaRegularizer)}build(e){e=yt(e);let t=this.axis>=0?this.axis:this.axis+e.length,n=e[t];if(n==null)throw new K(`Axis ${t} of input tensor should have a defined dimension but the layer received an input with shape ${JSON.stringify(e)}.`);this.inputSpec=[new Qt({ndim:e.length,axes:{[t]:n}})];let r=[n];this.scale&&(this.gamma=this.addWeight("gamma",r,null,this.gammaInitializer,this.gammaRegularizer,!0,this.gammaConstraint)),this.center&&(this.beta=this.addWeight("beta",r,null,this.betaInitializer,this.betaRegularizer,!0,this.betaConstraint)),this.movingMean=this.addWeight("moving_mean",r,null,this.movingMeanInitializer,null,!1),this.movingVariance=this.addWeight("moving_variance",r,null,this.movingVarianceInitializer,null,!1),this.built=!0}call(e,t){return Y(()=>{let n=t.training==null?!1:t.training,r=qe(e),s=r.shape,a=s.length,o=cs(0,a),i=this.axis>=0?this.axis:this.axis+a;o.splice(i,1);let l=ni(1,a);l[i]=s[i];let u=o.slice();u.sort();let c=!k.arraysEqual(u,cs(0,a).slice(0,a-1)),d=()=>{if(c){let y=Z(this.movingMean.read(),l),A=Z(this.movingVariance.read(),l),x=this.center?Z(this.beta.read(),l):null,b=this.scale?Z(this.gamma.read(),l):null;return qd(r,y,A,x,b,this.epsilon)}else return qd(r,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 d();let[h,p,f]=mne(r,this.gamma.read(),this.beta.read(),o,this.epsilon),m=(y,A,x)=>{Y(()=>{let b=1-x,v=y.read(),I=j(ke(v,A),b);y.write(ke(v,I))})};return(()=>{m(this.movingMean,p,this.momentum),m(this.movingVariance,f,this.momentum)})(),h})}getConfig(){let e={axis:this.axis,momentum:this.momentum,epsilon:this.epsilon,center:this.center,scale:this.scale,betaInitializer:Ut(this.betaInitializer),gammaInitializer:Ut(this.gammaInitializer),movingMeanInitializer:Ut(this.movingMeanInitializer),movingVarianceInitializer:Ut(this.movingVarianceInitializer),betaRegularizer:vt(this.betaRegularizer),gammaRegularizer:vt(this.gammaRegularizer),betaConstraint:cn(this.betaConstraint),gammaConstraint:cn(this.gammaConstraint)},t=super.getConfig();return Object.assign(e,t),e}};Px.className="BatchNormalization";ue.registerClass(Px);var zx=class extends tt{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=Pt(e.betaInitializer||"zeros"),this.gammaInitializer=Pt(e.gammaInitializer||"ones"),this.betaRegularizer=zt(e.betaRegularizer),this.gammaRegularizer=zt(e.gammaRegularizer),this.supportsMasking=!0}build(e){e=yt(e);let t=e.length;typeof this.axis=="number"&&(this.axis=[this.axis]);for(let s=0;s<this.axis.length;++s)this.axis[s]<0&&(this.axis[s]+=t);for(let s of this.axis)if(s<0||s>=t)throw new Error(`Invalid axis: ${s}`);if(this.axis.length!==Ga(this.axis).length)throw new Error(`Found duplicate axes in: ${this.axis}`);let n=this.axis.map(s=>e[s]),r=!0;this.scale?this.gamma=this.addWeight("gamma",n,"float32",this.gammaInitializer,this.gammaRegularizer,r):this.gamma=null,this.center?this.beta=this.addWeight("beta",n,"float32",this.betaInitializer,this.betaRegularizer,r):this.beta=null,this.built=!0}call(e,t){let n=qe(e),r=n.shape,s=r.length;return Y(()=>{let a=!0,{mean:o,variance:i}=Hy(n,this.axis,a),l=ni(1,s);for(let f of this.axis)l[f]=r[f];let u=f=>f!=null&&f.shape.length!==s&&this.axis!==[s-1]?Z(f,l):f,c=u(this.gamma.read()),d=u(this.beta.read()),h=[],p=[];for(let f=0;f<s;++f)this.axis.indexOf(f)!==-1?(h.push(r[f]),p.push(1)):(h.push(1),p.push(r[f]));return o=$s(o,h),i=$s(i,h),c=$s(c,p),d=$s(d,p),qd(n,o,i,d,c,this.epsilon)})}getConfig(){let e={axis:this.axis,epsilon:this.epsilon,center:this.center,scale:this.scale,betaInitializer:Ut(this.betaInitializer),gammaInitializer:Ut(this.gammaInitializer),betaRegularizer:vt(this.betaRegularizer),gammaRegularizer:vt(this.gammaRegularizer)},t=super.getConfig();return Object.assign(e,t),e}};zx.className="LayerNormalization";ue.registerClass(zx);function gne(e,t,n){return Y(()=>{if(e.rank!==4)throw new K(`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 K("spatial2dPadding expects `padding` to be an Array of two Arrays, each of which is an Array of two integers.");if(n==null&&(n=ls()),n!=="channelsLast"&&n!=="channelsFirst")throw new K(`Unknown data format: ${n}. Supported data formats are 'channelsLast' and 'channelsFirst.`);let r;return n==="channelsFirst"?r=[[0,0],[0,0],t[0],t[1]]:r=[[0,0],t[0],t[1],[0,0]],Yo(e,r)})}var Lx=class extends tt{constructor(e){if(e==null&&(e={}),super(e),this.dataFormat=e.dataFormat==null?ls():e.dataFormat,e.padding==null)this.padding=[[1,1],[1,1]];else if(typeof e.padding=="number")this.padding=[[e.padding,e.padding],[e.padding,e.padding]];else{if(e.padding=e.padding,e.padding.length!==2)throw new K(`ZeroPadding2D expects padding to be a length-2 array, but received a length-${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 K(`ZeroPadding2D expects height padding to be a length-2 array, but received a length-${e.padding[0].length} array.`);if(t=e.padding[0],e.padding[1].length!==2)throw new K(`ZeroPadding2D expects width padding to be a length-2 array, but received a length-${e.padding[1].length} array.`);n=e.padding[1]}this.padding=[t,n]}this.inputSpec=[new Qt({ndim:4})]}computeOutputShape(e){e=yt(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(()=>gne(qe(e),this.padding,this.dataFormat))}getConfig(){let e={padding:this.padding,dataFormat:this.dataFormat},t=super.getConfig();return Object.assign(e,t),e}};Lx.className="ZeroPadding2D";ue.registerClass(Lx);function mm(e,t,n,r,s,a){return Y(()=>{Zt(s),KI(a),_r(r),n==null&&(n=[1,1]),r==null&&(r="valid"),s==null&&(s=ls()),a==null&&(a="max"),e=ax(e,s);let o,i=r==="same"?"same":"valid";return a==="max"?o=Vy(e,t,n,i):o=_y(e,t,n,i),s==="channelsFirst"&&(o=st(o,[0,3,1,2])),o})}function c8(e,t,n,r,s,a){return Y(()=>{Zt(s),KI(a),_r(r),n==null&&(n=[1,1,1]),r==null&&(r="valid"),s==null&&(s=ls()),a==null&&(a="max"),e=e8(e,s);let o,i=r==="same"?"same":"valid";return a==="max"?o=nI(e,t,n,i):o=W6(e,t,n,i),s==="channelsFirst"&&(o=st(o,[0,4,1,2,3])),o})}var d8=class extends tt{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 K(`poolSize for 1D convolutional layer must be a number or an Array of a single number, but received ${JSON.stringify(e.poolSize)}`);if(yn(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 K(`strides for 1D convolutional layer must be a number or an Array of a single number, but received ${JSON.stringify(e.strides)}`);yn(this.strides,"strides"),this.padding=e.padding==null?"valid":e.padding,_r(this.padding),this.inputSpec=[new Qt({ndim:3})]}computeOutputShape(e){e=yt(e);let t=fs(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=Md(qe(e),2);let n=this.poolingFunction(qe(e),[this.poolSize[0],1],[this.strides[0],1],this.padding,"channelsLast");return Qo(n,[2])})}getConfig(){let e={poolSize:this.poolSize,padding:this.padding,strides:this.strides},t=super.getConfig();return Object.assign(e,t),e}},Bx=class extends d8{constructor(e){super(e)}poolingFunction(e,t,n,r,s){return Zt(s),_r(r),mm(e,t,n,r,s,"max")}};Bx.className="MaxPooling1D";ue.registerClass(Bx);var Wx=class extends d8{constructor(e){super(e)}poolingFunction(e,t,n,r,s){return Zt(s),_r(r),mm(e,t,n,r,s,"avg")}};Wx.className="AveragePooling1D";ue.registerClass(Wx);var h8=class extends tt{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 K(`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];yn(this.poolSize,"poolSize"),yn(this.strides,"strides"),this.padding=e.padding==null?"valid":e.padding,this.dataFormat=e.dataFormat==null?"channelsLast":e.dataFormat,Zt(this.dataFormat),_r(this.padding),this.inputSpec=[new Qt({ndim:4})]}computeOutputShape(e){e=yt(e);let t=this.dataFormat==="channelsFirst"?e[2]:e[1],n=this.dataFormat==="channelsFirst"?e[3]:e[2];return t=fs(t,this.poolSize[0],this.padding,this.strides[0]),n=fs(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(qe(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}},Vx=class extends h8{constructor(e){super(e)}poolingFunction(e,t,n,r,s){return Zt(s),_r(r),mm(e,t,n,r,s,"max")}};Vx.className="MaxPooling2D";ue.registerClass(Vx);var Ux=class extends h8{constructor(e){super(e)}poolingFunction(e,t,n,r,s){return Zt(s),_r(r),mm(e,t,n,r,s,"avg")}};Ux.className="AveragePooling2D";ue.registerClass(Ux);var p8=class extends tt{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 K(`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];yn(this.poolSize,"poolSize"),yn(this.strides,"strides"),this.padding=e.padding==null?"valid":e.padding,this.dataFormat=e.dataFormat==null?"channelsLast":e.dataFormat,Zt(this.dataFormat),_r(this.padding),this.inputSpec=[new Qt({ndim:5})]}computeOutputShape(e){e=yt(e);let t=this.dataFormat==="channelsFirst"?e[2]:e[1],n=this.dataFormat==="channelsFirst"?e[3]:e[2],r=this.dataFormat==="channelsFirst"?e[4]:e[3];return t=fs(t,this.poolSize[0],this.padding,this.strides[0]),n=fs(n,this.poolSize[1],this.padding,this.strides[1]),r=fs(r,this.poolSize[2],this.padding,this.strides[2]),this.dataFormat==="channelsFirst"?[e[0],e[1],t,n,r]:[e[0],t,n,r,e[4]]}call(e,t){return Y(()=>(this.invokeCallHook(e,t),this.poolingFunction(qe(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}},Hx=class extends p8{constructor(e){super(e)}poolingFunction(e,t,n,r,s){return Zt(s),_r(r),c8(e,t,n,r,s,"max")}};Hx.className="MaxPooling3D";ue.registerClass(Hx);var Gx=class extends p8{constructor(e){super(e)}poolingFunction(e,t,n,r,s){return Zt(s),_r(r),c8(e,t,n,r,s,"avg")}};Gx.className="AveragePooling3D";ue.registerClass(Gx);var f8=class extends tt{constructor(e){super(e);this.inputSpec=[new Qt({ndim:3})]}computeOutputShape(e){return[e[0],e[2]]}call(e,t){throw new He}},jx=class extends f8{constructor(e){super(e||{})}call(e,t){return Y(()=>{let n=qe(e);return Xt(n,1)})}};jx.className="GlobalAveragePooling1D";ue.registerClass(jx);var qx=class extends f8{constructor(e){super(e||{})}call(e,t){return Y(()=>{let n=qe(e);return Rs(n,1)})}};qx.className="GlobalMaxPooling1D";ue.registerClass(qx);var m8=class extends tt{constructor(e){super(e);this.dataFormat=e.dataFormat==null?"channelsLast":e.dataFormat,Zt(this.dataFormat),this.inputSpec=[new Qt({ndim:4})]}computeOutputShape(e){return e=e,this.dataFormat==="channelsLast"?[e[0],e[3]]:[e[0],e[1]]}call(e,t){throw new He}getConfig(){let e={dataFormat:this.dataFormat},t=super.getConfig();return Object.assign(e,t),e}},Kx=class extends m8{call(e,t){return Y(()=>{let n=qe(e);return this.dataFormat==="channelsLast"?Xt(n,[1,2]):Xt(n,[2,3])})}};Kx.className="GlobalAveragePooling2D";ue.registerClass(Kx);var Xx=class extends m8{call(e,t){return Y(()=>{let n=qe(e);return this.dataFormat==="channelsLast"?Rs(n,[1,2]):Rs(n,[2,3])})}};Xx.className="GlobalMaxPooling2D";ue.registerClass(Xx);var g8=class extends tt{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 r=t.layer,s=ps(r,n);delete t.layer;let a={layer:s};return Object.assign(a,t),new e(a)}},Zx=class extends g8{constructor(e){super(e);this.supportsMasking=!0}build(e){if(e=yt(e),e.length<3)throw new K(`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=yt(e);let t=[e[0]].concat(e.slice(2)),n=this.layer.computeOutputShape(t),r=e[1];return[n[0],r].concat(n.slice(1))}call(e,t){return Y(()=>(e=qe(e),i8((a,o)=>[qe(this.layer.call(a,t)),[]],e,[],!1,null,null,!1,!0)[1]))}};Zx.className="TimeDistributed";ue.registerClass(Zx);function yne(e){si(Iee,"BidirectionalMergeMode",e)}var Ane="concat",Yx=class extends g8{constructor(e){super(e);let t=e.layer.getConfig(),n={};n.className=e.layer.getClassName(),n.config=t,this.forwardLayer=ps(n),t.goBackwards=t.goBackwards!==!0;let r={};if(r.className=e.layer.getClassName(),r.config=t,this.backwardLayer=ps(r),this.forwardLayer.name="forward_"+this.forwardLayer.name,this.backwardLayer.name="backward_"+this.backwardLayer.name,this.mergeMode=e.mergeMode===void 0?Ane:e.mergeMode,yne(this.mergeMode),e.weights)throw new He("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,r,s;return this.returnState&&(s=t.slice(1)),n=t[0],n=n,this.mergeMode==="concat"?(n[n.length-1]*=2,r=[n]):this.mergeMode==null?r=[n,n.slice()]:r=[n],this.returnState?this.mergeMode==null?r.concat(s).concat(s.slice()):[n].concat(s).concat(s.slice()):Jn(r)}apply(e,t){let n=t==null?null:t.initialState,r=t==null?null:t.constants;t==null&&(t={});let s=o8(e,n,r,this.numConstants);if(e=s.inputs,n=s.initialState,r=s.constants,Array.isArray(e)&&(n=e.slice(1),e=e[0]),(n==null||n.length===0)&&r==null)return super.apply(e,t);let a=[],o=[];if(n!=null){let l=n.length;if(l%2>0)throw new K("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 Qt({shape:c.shape}));this.forwardLayer.stateSpec=u.slice(0,l/2),this.backwardLayer.stateSpec=u.slice(l/2),o.push(...u)}if(r!=null)throw new He("Support for constants in Bidirectional layers is not implemented yet.");let i=a[0]instanceof hs;for(let l of a)if(l instanceof hs!==i)throw new K("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 d=super.apply(l,t);return this.inputSpec=c,d}else return super.apply(e,t)}call(e,t){return Y(()=>{let n=t.initialState,r,s;if(n==null)r=this.forwardLayer.call(e,t),s=this.backwardLayer.call(e,t);else{let i=n.slice(0,n.length/2),l=n.slice(n.length/2);r=this.forwardLayer.call(e,Object.assign(t,{initialState:i})),s=this.backwardLayer.call(e,Object.assign(t,{initialState:l}))}let a;this.returnState&&(Array.isArray(r)&&(a=r.slice(1).concat(s.slice(1))),r=r[0],s=s[0]),this.returnSequences&&(s=Kr(s,1));let o;return this.mergeMode==="concat"?o=wA([r,s]):this.mergeMode==="sum"?o=de(r,s):this.mergeMode==="ave"?o=j(.5,de(r,s)):this.mergeMode==="mul"?o=j(r,s):this.mergeMode==null&&(o=[r,s]),this.returnState?this.mergeMode==null?o.concat(a):[o].concat(a):o})}resetStates(e){this.forwardLayer.resetStates(),this.backwardLayer.resetStates()}build(e){ai(this.forwardLayer.name,()=>{this.forwardLayer.build(e)}),ai(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 s=this.forwardLayer.states.map(a=>null);return Array.isArray(n)?n.concat(s).concat(s):[n].concat(s).concat(s)}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=ps(t.layer);if(delete t.layer,t.numConstants!=null)throw new He("Deserialization of a Bidirectional layer with numConstants present is not supported yet.");let r=t;return r.layer=n,new e(r)}};Yx.className="Bidirectional";ue.registerClass(Yx);function xne(e){return new tu(e)}function bne(e){return new nx(e)}function vne(e){return new QA(e)}function wne(e){return new ex(e)}function kne(e){return new tx(e)}function Ine(e){return new sx(e)}function Sne(e){return new rx(e)}function Tne(e){return new cx(e)}function Nne(e){return new lm(e)}function Cne(e){return new ix(e)}function Ene(e){return new um(e)}function $ne(e){return new lx(e)}function Rne(e){return new ux(e)}function _ne(e){return new dx(e)}function Dne(e){return new hx(e)}function Fne(e){return new px(e)}function Mne(e){return new vx(e)}function One(e){return new xx(e)}function Pne(e){return new fm(e)}function zne(e){return new Ax(e)}function Lne(e){return new bx(e)}function Bne(e){return new wx(e)}function Wne(e){return new kx(e)}function Vne(e){return new Ix(e)}function Une(e){return new Tx(e)}function Hne(e){return new Nx(e)}function Gne(e){return new Ex(e)}function jne(e){return new _x(e)}function qne(e){return new $x(e)}function Kne(e){return new Rx(e)}function Xne(e){return new Cx(e)}function Zne(e){return new Dx(e)}function Yne(e){return new Px(e)}function Jne(e){return new zx(e)}function Qne(e){return new Lx(e)}function Jx(e){return new Wx(e)}function ere(e){return Jx(e)}function tre(e){return Jx(e)}function Qx(e){return new Ux(e)}function nre(e){return Qx(e)}function rre(e){return Qx(e)}function e5(e){return new Gx(e)}function sre(e){return e5(e)}function are(e){return e5(e)}function ore(e){return new jx(e)}function ire(e){return new Kx(e)}function y8(e){return new qx(e)}function A8(e){return new Xx(e)}function x8(e){return new Bx(e)}function b8(e){return new Vx(e)}function lre(e){return new Hx(e)}function ure(e){return new mx(e)}function cre(e){return new dm(e)}function dre(e){return new gx(e)}function hre(e){return new Gd(e)}function pre(e){return new fx(e)}function fre(e){return new cm(e)}function mre(e){return new yx(e)}function gre(e){return new pm(e)}function yre(e){return new ma(e)}function Are(e){return new hm(e)}function xre(e){return new Yx(e)}function bre(e){return new Zx(e)}var vre=y8,wre=A8,kre=x8,Ire=b8;function Sre(e){return new Fx(e)}function Tre(e){return new Mx(e)}function Nre(e){return new Ox(e)}function Cre(e){return new Sx(e)}var v8={};_e(v8,{MAPE:()=>Lre,MSE:()=>Vre,binaryAccuracy:()=>Ere,binaryCrossentropy:()=>$re,categoricalAccuracy:()=>_re,categoricalCrossentropy:()=>Dre,cosineProximity:()=>Ore,mape:()=>Bre,meanAbsoluteError:()=>Pre,meanAbsolutePercentageError:()=>zre,meanSquaredError:()=>Wre,mse:()=>Ure,precision:()=>Fre,recall:()=>Mre,sparseCategoricalAccuracy:()=>Rre});function Ere(e,t){return zA(e,t)}function $re(e,t){return xS(e,t)}function Rre(e,t){return bS(e,t)}function _re(e,t){return LA(e,t)}function Dre(e,t){return BA(e,t)}function Fre(e,t){return AS(e,t)}function Mre(e,t){return xte(e,t)}function Ore(e,t){return OA(e,t)}function Pre(e,t){return em(e,t)}function zre(e,t){return ru(e,t)}function Lre(e,t){return ru(e,t)}function Bre(e,t){return ru(e,t)}function Wre(e,t){return ii(e,t)}function Vre(e,t){return ii(e,t)}function Ure(e,t){return ii(e,t)}var w8={};_e(w8,{modelFromJSON:()=>Qte});var k8={};_e(k8,{l1:()=>Gre,l1l2:()=>Hre,l2:()=>jre});function Hre(e){return new Vd(e)}function Gre(e){return ine(e)}function jre(e){return lne(e)}var I8=class extends nu{constructor(){super(...arguments);this.model=null}setModel(e){if(!(e instanceof fa))throw new Error("model must be a LayersModel, not some other Container");this.model=e}};function gm(e,t){return e<t}function S8(e,t){return e>t}var T8=class extends I8{constructor(e){super();if(e==null&&(e={}),e.restoreBestWeights)throw new He("restoreBestWeights = True is not implemented in EarlyStopping yet.");this.monitor=e.monitor||"val_loss",this.minDelta=Math.abs(e.minDelta||0),this.patience=e.patience||0,this.verbose=e.verbose||0,this.mode=e.mode||"auto",this.baseline=e.baseline,["auto","min","max"].indexOf(this.mode)===-1&&(console.warn(`EarlyStopping mode '${this.mode}' is invalid. Falling back to mode 'auto'.`),this.mode="auto"),this.mode==="min"?this.monitorFunc=gm:this.mode==="max"?this.monitorFunc=S8:this.monitor.indexOf("acc")!==-1?this.monitorFunc=S8:this.monitorFunc=gm,this.monitorFunc===gm&&(this.minDelta*=-1)}async onTrainBegin(e){this.wait=0,this.stoppedEpoch=0,this.baseline!=null?this.best=this.baseline:this.best=this.monitorFunc===gm?1/0:-1/0}async onEpochEnd(e,t){await Ka(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 qre(e){return new T8(e)}var Kre={earlyStopping:qre},ms;(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_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"})(ms||(ms={}));var N8;(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={}))})(N8||(N8={}));var t5={};function Xre(e,t){let n={tfOpName:e,category:"custom",inputs:[],attrs:[],customExecutor:t};t5[e]=n}function C8(e){return t5[e]}function Zre(e){delete t5[e]}function T(e,t,n,r,s){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 Bn(t.inputNames[a.inputIndexStart],n,r,s);if(a.type==="tensors")return t.inputNames.slice(i,l).map(h=>Bn(h,n,r,s));let u=Bn(t.inputNames.slice(i)[0],n,r,s),c=u.dataSync();return a.type==="number"?c[0]:k.toNestedArray(u.shape,c)}let o=t.attrParams[e];return o&&o.value}function Bn(e,t,n,r){let[s,a]=dr(e);if(r!=null){let i=r.getHashTableHandleByName(s);if(i!=null)return i}let o=n.currentContextIds.find(i=>!!t[ym(s,i)]);return o!==void 0?t[ym(s,o)][a]:void 0}function Yre(e,t,n){return t[ym(e,n.currentContextId)]}function ga(e,t){let[n,r,s]=dr(e);return[ym(n,t&&t.currentContextId),r,s]}function ym(e,t){return t?`${e}-${t}`:e}function dr(e){let t=e.split(":");if(t.length===1)return[e,0,void 0];let n=t[0],r=t.length===3?t[1]:void 0,s=Number(t[t.length-1]);return[n,s,r]}function Am(e,t,n){let r=T("pad",e,t,n);if(r==="explicit"){r=T("explicitPaddings",e,t,n);let s=[[0,0],[0,0],[0,0],[0,0]];for(let a=0;a<4;a++)s[a][0]=r[a*2],s[a][1]=r[a*2+1];return s}return r}function ya(e){return e.kept?e:za(e)}var E8={};_e(E8,{json:()=>Jre});var Jre=[{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}]}],$8={};_e($8,{json:()=>Qre});var Qre=[{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}]}],R8={};_e(R8,{json:()=>ese});var ese=[{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:"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"}]}],_8={};_e(_8,{json:()=>tse});var tse=[{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"}]},{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"}]}],D8={};_e(D8,{json:()=>nse});var nse=[{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:"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"}]}],F8={};_e(F8,{json:()=>rse});var rse=[{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}]}],M8={};_e(M8,{json:()=>sse});var sse=[{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:"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"}]}],O8={};_e(O8,{json:()=>ase});var ase=[{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"}]}],P8={};_e(P8,{json:()=>ose});var ose=[{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"}]}],z8={};_e(z8,{json:()=>ise});var ise=[{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"}]}],L8={};_e(L8,{json:()=>lse});var lse=[{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}]}],B8={};_e(B8,{json:()=>use});var use=[{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:"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"}]}],W8={};_e(W8,{json:()=>cse});var cse=[{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}]}],V8={};_e(V8,{json:()=>dse});var dse=[{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:"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"}]}],U8={};_e(U8,{json:()=>hse});var hse=[{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}]}],H8={};_e(H8,{json:()=>pse});var pse=[{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"}]}],G8={};_e(G8,{json:()=>fse});var fse=[{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}]}],j8={};_e(j8,{json:()=>mse});var mse=[{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"}]}],q8={};_e(q8,{json:()=>gse});var gse=[{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:[]}],K8=class{static get Instance(){return this._instance||(this._instance=new this)}constructor(){let e=[E8,$8,R8,_8,D8,F8,M8,O8,P8,z8,L8,B8,W8,V8,U8,H8,G8,j8,q8],t=[].concat(...e.map(n=>n.json));this.opMappers=t.reduce((n,r)=>(n[r.tfOpName]=r,n),{})}transformGraph(e,t={}){let n=e.node,r=[],s=[],a=[],o=n.reduce((f,m)=>(f[m.name]=this.mapNode(m),m.op.startsWith("Placeholder")?r.push(f[m.name]):m.op==="Const"?s.push(f[m.name]):(m.input==null||m.input.length===0)&&a.push(f[m.name]),f),{}),i=[],l=[],u={},c={};t!=null&&(u=this.mapSignatureEntries(t.inputs),c=this.mapSignatureEntries(t.outputs));let d=Object.keys(o);d.forEach(f=>{let m=o[f];m.inputNames.forEach((g,y)=>{let[A,,x]=ga(g),b=o[A];if(b.outputs!=null){let v=b.outputs.indexOf(x);if(v!==-1){let I=`${A}:${v}`;m.inputNames[y]=I}}m.inputs.push(b),b.children.push(m)})}),Object.keys(c).length===0?d.forEach(f=>{let m=o[f];m.children.length===0&&l.push(m)}):Object.keys(c).forEach(f=>{let[m]=ga(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]=ga(f),g=o[m];g&&(g.signatureKey=u[f],i.push(g))}):i=r;let h={};e.library!=null&&e.library.function!=null&&(h=e.library.function.reduce((f,m)=>(f[m.signature.name]=this.mapFunction(m),f),{}));let p={nodes:o,inputs:i,outputs:l,weights:s,placeholders:r,signature:t,functions:h};return a.length>0&&(p.initNodes=a),p}mapSignatureEntries(e){return Object.keys(e||{}).reduce((t,n)=>(t[e[n].name]=n,t),{})}mapNode(e){let t=C8(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(r=>r.startsWith("^")?r.substr(1):r),inputs:[],children:[],inputParams:{},attrParams:{},rawAttrs:e.attr,outputs:t.outputs};return t.inputs!=null&&(n.inputParams=t.inputs.reduce((r,s)=>(r[s.name]={type:s.type,inputIndexStart:s.start,inputIndexEnd:s.end},r),{})),t.attrs!=null&&(n.attrParams=t.attrs.reduce((r,s)=>{let a=s.type,o;switch(s.type){case"string":o=n5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=n5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"string[]":o=c5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=c5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"number":o=s5(e.attr,s.tfName,s.defaultValue||0),o===void 0&&!!s.tfDeprecatedName&&(o=s5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"number[]":o=u5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=u5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"bool":o=r5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=r5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"bool[]":o=h5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=h5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"shape":o=l5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=l5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"shape[]":o=d5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=d5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"dtype":o=o5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=o5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"dtype[]":o=i5(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=i5(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"func":o=Z8(e.attr,s.tfName,s.defaultValue),o===void 0&&!!s.tfDeprecatedName&&(o=Z8(e.attr,s.tfDeprecatedName,s.defaultValue));break;case"tensor":case"tensors":break;default:throw new Error(`Unsupported param type: ${s.type} for op: ${e.op}`)}return r[s.name]={value:o,type:a},r},{})),n}mapFunction(e){let t=e.nodeDef,n=[],r=[],s={};t!=null&&(s=t.reduce((c,d)=>(c[d.name]=this.mapNode(d),d.op==="Const"&&r.push(c[d.name]),c),{}));let a=[],o=[];e.signature.inputArg.forEach(c=>{let[d]=ga(c.name),h={name:d,op:"Placeholder",inputs:[],inputNames:[],category:"graph",inputParams:{},attrParams:{dtype:{value:a5(c.type),type:"dtype"}},children:[]};h.signatureKey=c.name,a.push(h),s[d]=h}),Object.keys(s).forEach(c=>{let d=s[c];d.inputNames.forEach((h,p)=>{let[f,,m]=ga(h),g=s[f];if(g.outputs!=null){let y=g.outputs.indexOf(m);if(y!==-1){let A=`${f}:${y}`;d.inputNames[p]=A}}d.inputs.push(g),g.children.push(d)})});let l=e.ret;e.signature.outputArg.forEach(c=>{let[d,h]=ga(l[c.name]),p=s[d];p!=null&&(p.defaultOutput=h,o.push(p))});let u=this.mapArgsToSignature(e);return{nodes:s,inputs:a,outputs:o,weights:r,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 yse(e){let t=re().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 X8(e,t){let n=Array.isArray(e)?String.fromCharCode.apply(null,e):yse(e);return t?n:n.toLowerCase()}function n5(e,t,n,r=!1){let s=e[t];return s!=null?X8(s.s,r):n}function r5(e,t,n){let r=e[t];return r?r.b:n}function s5(e,t,n){let r=e[t]||{},s=r.i!=null?r.i:r.f!=null?r.f:n;return typeof s=="number"?s:parseInt(s,10)}function a5(e){switch(typeof e=="string"&&(e=ms[e]),e){case ms.DT_FLOAT:return"float32";case ms.DT_INT32:case ms.DT_INT64:case ms.DT_INT8:case ms.DT_UINT8:return"int32";case ms.DT_BOOL:return"bool";case ms.DT_DOUBLE:return"float32";case ms.DT_STRING:return"string";default:return null}}function Z8(e,t,n){let r=e[t];return r&&r.func?r.func.name:n}function o5(e,t,n){let r=e[t];return r&&r.type?a5(r.type):n}function i5(e,t,n){let r=e[t];return r&&r.list&&r.list.type?r.list.type.map(s=>a5(s)):n}function Y8(e){if(!e.unknownRank)return e.dim!=null?e.dim.map(t=>typeof t.size=="number"?t.size:parseInt(t.size,10)):[]}function l5(e,t,n){let r=e[t];return r&&r.shape?Y8(r.shape):n}function u5(e,t,n){let r=e[t];return r?((r.list.f&&r.list.f.length?r.list.f:r.list.i)||[]).map(s=>typeof s=="number"?s:parseInt(s,10)):n}function c5(e,t,n,r=!1){let s=e[t];return s&&s.list&&s.list.s?s.list.s.map(a=>X8(a,r)):n}function d5(e,t,n){let r=e[t];return r&&r.list&&r.list.shape?r.list.shape.map(s=>Y8(s)):n}function h5(e,t,n){let r=e[t];return r&&r.list&&r.list.b?r.list.b:n}var Ase=class{constructor(e,t,n){this.node=e,this.tensorMap=t,this.context=n,this.inputs=[],this.attrs={},this.inputs=e.inputNames.map(r=>this.getInput(r)),e.rawAttrs!=null&&(this.attrs=Object.keys(e.rawAttrs).reduce((r,s)=>(r[s]=this.getAttr(s),r),{}))}getInput(e){return Bn(e,this.tensorMap,this.context)}getAttr(e,t){let n=this.node.rawAttrs[e];if(n.tensor!=null)return Bn(e,this.tensorMap,this.context);if(n.i!=null||n.f!=null)return s5(this.node.rawAttrs,e,t);if(n.s!=null)return n5(this.node.rawAttrs,e,t);if(n.b!=null)return r5(this.node.rawAttrs,e,t);if(n.shape!=null)return l5(this.node.rawAttrs,e,t);if(n.type!=null)return o5(this.node.rawAttrs,e,t);if(n.list!=null){if(n.list.i!=null||n.list.f!=null)return u5(this.node.rawAttrs,e,t);if(n.list.s!=null)return c5(this.node.rawAttrs,e,t);if(n.list.shape!=null)return d5(this.node.rawAttrs,e,t);if(n.list.b!=null)return h5(this.node.rawAttrs,e,t);if(n.list.type!=null)return i5(this.node.rawAttrs,e,t)}return t}},xse=(e,t,n)=>{switch(e.op){case"BiasAdd":case"AddV2":case"Add":return[de(T("a",e,t,n),T("b",e,t,n))];case"AddN":return[qG(T("tensors",e,t,n))];case"FloorMod":case"Mod":return[AK(T("a",e,t,n),T("b",e,t,n))];case"Mul":return[j(T("a",e,t,n),T("b",e,t,n))];case"RealDiv":case"Div":return[Re(T("a",e,t,n),T("b",e,t,n))];case"DivNoNan":return[mq(T("a",e,t,n),T("b",e,t,n))];case"FloorDiv":return[O6(T("a",e,t,n),T("b",e,t,n))];case"Sub":return[ke(T("a",e,t,n),T("b",e,t,n))];case"Minimum":return[Nf(T("a",e,t,n),T("b",e,t,n))];case"Maximum":return[Va(T("a",e,t,n),T("b",e,t,n))];case"Pow":return[Jo(T("a",e,t,n),T("b",e,t,n))];case"SquaredDifference":return[pI(T("a",e,t,n),T("b",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},bse=(e,t,n)=>{switch(e.op){case"Abs":case"ComplexAbs":return[gn(T("x",e,t,n))];case"Acos":return[UG(T("x",e,t,n))];case"Acosh":return[GG(T("x",e,t,n))];case"Asin":return[ej(T("x",e,t,n))];case"Asinh":return[nj(T("x",e,t,n))];case"Atan":return[sj(T("x",e,t,n))];case"Atan2":return[oj(T("x",e,t,n),T("y",e,t,n))];case"Atanh":return[lj(T("x",e,t,n))];case"Ceil":return[Mj(T("x",e,t,n))];case"Complex":return[Uo(T("real",e,t,n),T("imag",e,t,n))];case"Cos":return[My(T("x",e,t,n))];case"Cosh":return[q6(T("x",e,t,n))];case"Elu":return[If(T("x",e,t,n))];case"Erf":return[vq(T("x",e,t,n))];case"Exp":return[Gr(T("x",e,t,n))];case"Expm1":return[Sq(T("x",e,t,n))];case"Floor":return[Sf(T("x",e,t,n))];case"Log":return[jr(T("x",e,t,n))];case"Log1p":return[zy(T("x",e,t,n))];case"Imag":return[Oy(T("x",e,t,n))];case"Neg":return[qt(T("x",e,t,n))];case"Reciprocal":return[KK(T("x",e,t,n))];case"Real":return[Cf(T("x",e,t,n))];case"Relu":return[ca(T("x",e,t,n))];case"Round":return[aI(T("x",e,t,n))];case"Selu":return[iI(T("x",e,t,n))];case"Sigmoid":return[ia(T("x",e,t,n))];case"Sin":return[lI(T("x",e,t,n))];case"Sign":return[cX(T("x",e,t,n))];case"Sinh":return[uI(T("x",e,t,n))];case"Softplus":return[Cd(T("x",e,t,n))];case"Sqrt":return[Ln(T("x",e,t,n))];case"Square":return[Tt(T("x",e,t,n))];case"Tanh":return[Sd(T("x",e,t,n))];case"Tan":return[_X(T("x",e,t,n))];case"ClipByValue":return[Cr(T("x",e,t,n),T("clipValueMin",e,t,n),T("clipValueMax",e,t,n))];case"Relu6":return[sI(T("x",e,t,n))];case"Rsqrt":return[oI(Bn(e.inputNames[0],t,n))];case"Prod":return[rI(T("x",e,t,n),T("axes",e,t,n))];case"LeakyRelu":return[Py(T("x",e,t,n),T("alpha",e,t,n))];case"Prelu":return[jy(T("x",e,t,n),T("alpha",e,t,n))];case"IsNan":return[Oq(Bn(e.inputNames[0],t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}};function Jr(e,t,n=""){if(!(typeof e=="number"||typeof t=="number")){k.assert(e.length===t.length,()=>n+` Shapes ${e} and ${t} must match`);for(let r=0;r<e.length;r++){let s=e[r],a=t[r];k.assert(s<0||a<0||s===a,()=>n+` Shapes ${e} and ${t} must match`)}}}function J8(e){return!(typeof e=="number"||e.some(t=>t<0))}function Kd(e,t,n){let r=p5(e,n),s=!J8(r);if(s&&t.length===0)throw new Error(`Tried to calculate elements of an empty list with non-fully-defined elementShape: ${r}`);if(s&&t.forEach(a=>{r=p5(a.shape,r)}),!J8(r))throw new Error(`Non-fully-defined elementShape: ${r}`);return r}function p5(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 r=0;r<e.length;++r){let s=e[r],a=t[r];if(s>=0&&a>=0&&s!==a)throw new Error(`Incompatible shape during merge: ${e} vs. ${t}`);n[r]=s>=0?s:a}return n}var vse=class{constructor(e,t,n,r,s,a,o){this.name=e,this.dtype=t,this.maxSize=n,this.elementShape=r,this.identicalElementShapes=s,this.dynamicSize=a,this.clearAfterRead=o,this.tensors=[],this.closed_=!1,this.idTensor=De(0),In(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),Jr(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,In(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,r)=>this.write(n,t[r]))}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 r=0;r<this.size();r++)e.push(r)}if(e.length===0)return Cs([],[0].concat(this.elementShape));let n=this.readMany(e);return Jr(this.elementShape,n[0].shape,"TensorArray shape mismatch: "),Xr(n,0)}concat(e){if(!!e&&e!==this.dtype)throw new Error(`TensorArray dtype is ${this.dtype} but concat requested dtype ${e}`);if(this.size()===0)return Cs([],[0].concat(this.elementShape));let t=[];for(let r=0;r<this.size();r++)t.push(r);let n=this.readMany(t);return Jr(this.elementShape,n[0].shape,`TensorArray shape mismatch: tensor array shape (${this.elementShape}) vs first tensor shape (${n[0].shape})`),an(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,Ds(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,r=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 s=n===0?0:t.size/n,a=[];Y(()=>{t=Z(t,[1,n,s]);for(let i=0;i<e.length;++i){let l=i===0?0:r[i-1],u=[0,l,0],c=[1,e[i],s];a[i]=Z(at(t,u,c),this.elementShape)}return a});let o=[];for(let i=0;i<e.length;i++)o[i]=i;this.writeMany(o,a)}},Xd=class{constructor(e,t,n,r=-1){this.tensors=e,this.elementShape=t,this.elementDtype=n,e!=null&&e.forEach(s=>{if(n!==s.dtype)throw new Error(`Invalid data types; op elements ${n}, but list elements ${s.dtype}`);Jr(t,s.shape,"TensorList shape mismatch: "),In(s)}),this.idTensor=De(0),this.maxNumElements=r,In(this.idTensor)}get id(){return this.idTensor.id}copy(){return new Xd([...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.`);Jr(e,this.elementShape,"TensorList shape mismatch: ");let r=Kd(this.elementShape,this.tensors,e);return Y(()=>{let s=this.tensors.map(a=>Z(a,r));return Xr(s,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=Kd(this.elementShape,this.tensors,e),r=this.tensors.pop();return Jr(r.shape,e,"TensorList shape mismatch: "),Z(r,n)}pushBack(e){if(e.dtype!==this.elementDtype)throw new Error(`Invalid data types; op elements ${e.dtype}, but list elements ${this.elementDtype}`);if(Jr(e.shape,this.elementShape,"TensorList shape mismatch: "),this.maxNumElements===this.size())throw new Error("Trying to push element into a full list.");In(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}.`);this.tensors.length=e}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.`);Jr(this.tensors[e].shape,t,"TensorList shape mismatch: ");let r=Kd(this.elementShape,this.tensors,t);return Z(this.tensors[e],r)}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.`);Jr(this.elementShape,t.shape,"TensorList shape mismatch: "),In(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}`);Jr(this.elementShape,n,"TensorList shape mismatch: "),e=e.slice(0,this.size());let r=Kd(this.elementShape,this.tensors,n);return e.length===0?Cs([],[0].concat(r)):Y(()=>{let s=e.map(a=>Z(this.tensors[a],r));return Xr(s,0)})}concat(e,t){if(!!e&&e!==this.elementDtype)throw new Error(`TensorList dtype is ${this.elementDtype} but concat requested dtype ${e}`);Jr(this.elementShape,t,"TensorList shape mismatch: ");let n=Kd(this.elementShape,this.tensors,t);return this.size()===0?Cs([],[0].concat(n)):Y(()=>{let r=this.tensors.map(s=>Z(s,n));return an(r,0)})}};function wse(e,t,n){let r=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 s=e.shape.slice(1);Jr(s,t,"TensorList shape mismatch: ");let a=Ds(e);return new Xd(a,t,r)}function kse(e,t,n){return new Xd([],e,t,n)}function Ise(e,t,n,r){if(t.length!==e.shape[0])throw new Error(`Expected len(indices) == tensor.shape[0], but saw: ${t.length} vs. ${e.shape[0]}`);let s=Math.max(...t);if(r!=null&&r!==-1&&s>=r)throw new Error(`Max index must be < array size (${s} vs. ${r})`);let a=new Xd([],n,e.dtype,r),o=Ds(e,0);return t.forEach((i,l)=>{a.setItem(i,o[l])}),a}function Sse(e,t,n){let r=0,s=t.map(c=>(r+=c,r));if(r!==e.shape[0])throw new Error(`Expected sum of lengths to be equal to
tensor.shape[0], but sum of lengths is
${r}, and tensor's shape is: ${e.shape}`);let a=e.shape.slice(1),o=p5(a,n),i=r===0?0:e.size/r,l=Y(()=>{let c=[];e=Z(e,[1,r,i]);for(let d=0;d<t.length;++d){let h=d===0?0:s[d-1],p=[0,h,0],f=[1,t[d],i];c[d]=Z(at(e,p,f),o)}return e.dispose(),c}),u=new Xd([],n,e.dtype,t.length);for(let c=0;c<l.length;c++)u.setItem(c,l[c]);return u}var Tse=async(e,t,n)=>{switch(e.op){case"If":case"StatelessIf":{let r=T("thenBranch",e,t,n),s=T("elseBranch",e,t,n),a=T("cond",e,t,n),o=T("args",e,t,n);return(await a.data())[0]?n.functionMap[r].executeFunctionAsync(o,n.tensorArrayMap,n.tensorListMap):n.functionMap[s].executeFunctionAsync(o,n.tensorArrayMap,n.tensorListMap)}case"While":case"StatelessWhile":{let r=T("body",e,t,n),s=T("cond",e,t,n),a=T("args",e,t,n),o=await n.functionMap[s].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[r].executeFunctionAsync(u,n.tensorArrayMap,n.tensorListMap);let d=u.map(p=>p.id);c.forEach(p=>{!p.kept&&i.indexOf(p.id)===-1&&d.indexOf(p.id)===-1&&p.dispose()});let h=await n.functionMap[s].executeFunctionAsync(u,n.tensorArrayMap,n.tensorListMap);l=await h[0].data(),h.forEach(p=>{!p.kept&&i.indexOf(p.id)===-1&&d.indexOf(p.id)===-1&&p.dispose()})}return u}case"LoopCond":{let r=T("pred",e,t,n);return[ya(r)]}case"Switch":{let r=T("pred",e,t,n),s=T("data",e,t,n);return s.kept||(s=ya(s)),(await r.data())[0]?[void 0,s]:[s,void 0]}case"Merge":{let r=e.inputNames.find(s=>Bn(s,t,n)!==void 0);if(r){let s=Bn(r,t,n);return[ya(s)]}return}case"Enter":{let r=T("frameName",e,t,n),s=T("tensor",e,t,n);return n.enterFrame(r),[ya(s)]}case"Exit":{let r=T("tensor",e,t,n);return n.exitFrame(),[ya(r)]}case"NextIteration":{let r=T("tensor",e,t,n);return n.nextIteration(),[ya(r)]}case"TensorArrayV3":{let r=T("size",e,t,n),s=T("dtype",e,t,n),a=T("elementShape",e,t,n),o=T("dynamicSize",e,t,n),i=T("clearAfterRead",e,t,n),l=T("identicalElementShapes",e,t,n),u=T("name",e,t,n),c=new vse(u,s,r,a,l,o,i);return n.addTensorArray(c),[c.idTensor,De(1)]}case"TensorArrayWriteV3":{let r=T("tensorArrayId",e,t,n),s=T("index",e,t,n),a=T("tensor",e,t,n),o=n.getTensorArray(r.id);return o.write(s,a),[o.idTensor]}case"TensorArrayReadV3":{let r=T("tensorArrayId",e,t,n),s=T("index",e,t,n);return[n.getTensorArray(r.id).read(s)]}case"TensorArrayGatherV3":{let r=T("tensorArrayId",e,t,n),s=T("indices",e,t,n),a=T("dtype",e,t,n);return[n.getTensorArray(r.id).gather(s,a)]}case"TensorArrayScatterV3":{let r=T("tensorArrayId",e,t,n),s=T("indices",e,t,n),a=T("tensor",e,t,n),o=n.getTensorArray(r.id);return o.scatter(s,a),[o.idTensor]}case"TensorArrayConcatV3":{let r=T("tensorArrayId",e,t,n),s=n.getTensorArray(r.id),a=T("dtype",e,t,n);return[s.concat(a)]}case"TensorArraySplitV3":{let r=T("tensorArrayId",e,t,n),s=T("tensor",e,t,n),a=T("lengths",e,t,n),o=n.getTensorArray(r.id);return o.split(a,s),[o.idTensor]}case"TensorArraySizeV3":{let r=T("tensorArrayId",e,t,n),s=n.getTensorArray(r.id);return[De(s.size(),"int32")]}case"TensorArrayCloseV3":{let r=T("tensorArrayId",e,t,n),s=n.getTensorArray(r.id);return s.clearAndClose(),[s.idTensor]}case"TensorListSetItem":{let r=T("tensorListId",e,t,n),s=T("index",e,t,n),a=T("tensor",e,t,n),o=n.getTensorList(r.id);return o.setItem(s,a),[o.idTensor]}case"TensorListGetItem":{let r=T("tensorListId",e,t,n),s=T("index",e,t,n),a=T("elementShape",e,t,n),o=T("elementDType",e,t,n);return[n.getTensorList(r.id).getItem(s,a,o)]}case"TensorListScatterV2":case"TensorListScatter":{let r=T("indices",e,t,n),s=T("tensor",e,t,n),a=T("elementShape",e,t,n),o=T("numElements",e,t,n),i=Ise(s,r,a,o);return n.addTensorList(i),[i.idTensor]}case"TensorListReserve":case"EmptyTensorList":{let r=T("elementShape",e,t,n),s=T("elementDType",e,t,n),a;e.op==="TensorListReserve"?a="numElements":a="maxNumElements";let o=T(a,e,t,n),i=kse(r,s,o);return n.addTensorList(i),[i.idTensor]}case"TensorListGather":{let r=T("tensorListId",e,t,n),s=T("indices",e,t,n),a=T("elementShape",e,t,n),o=T("elementDType",e,t,n);return[n.getTensorList(r.id).gather(s,o,a)]}case"TensorListStack":{let r=T("tensorListId",e,t,n),s=T("elementShape",e,t,n),a=T("elementDType",e,t,n),o=T("numElements",e,t,n);return[n.getTensorList(r.id).stack(s,a,o)]}case"TensorListFromTensor":{let r=T("tensor",e,t,n),s=T("elementShape",e,t,n),a=T("elementDType",e,t,n),o=wse(r,s,a);return n.addTensorList(o),[o.idTensor]}case"TensorListConcat":{let r=T("tensorListId",e,t,n),s=n.getTensorList(r.id),a=T("dtype",e,t,n),o=T("elementShape",e,t,n);return[s.concat(a,o)]}case"TensorListPushBack":{let r=T("tensorListId",e,t,n),s=T("tensor",e,t,n),a=n.getTensorList(r.id);return a.pushBack(s),[a.idTensor]}case"TensorListPopBack":{let r=T("tensorListId",e,t,n),s=T("elementShape",e,t,n),a=T("elementDType",e,t,n);return[n.getTensorList(r.id).popBack(s,a)]}case"TensorListSplit":{let r=T("tensor",e,t,n),s=T("elementShape",e,t,n),a=T("lengths",e,t,n),o=Sse(r,a,s);return n.addTensorList(o),[o.idTensor]}default:throw TypeError(`Node type ${e.op} is not implemented`)}};function Q8(e,t,n){let[r,s]=T("fusedOps",e,t,n),a=r==="biasadd",o=!a,i=s==="prelu",l=r==="fusedbatchnorm",u=T("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=T("strides",e,t,n),d=Am(e,t,n),h=T("dataFormat",e,t,n).toUpperCase(),p=T("dilations",e,t,n),[f,m]=T("args",e,t,n);o&&(m=f,f=void 0);let g=T("leakyreluAlpha",e,t,n);return{stride:c,pad:d,dataFormat:h,dilations:p,biasArg:f,preluArg:m,activationFunc:s,leakyreluAlpha:g}}var Nse=(e,t,n)=>{switch(e.op){case"Conv1D":{let r=T("stride",e,t,n),s=T("pad",e,t,n),a=T("dataFormat",e,t,n).toUpperCase(),o=T("dilation",e,t,n);return[U6(T("x",e,t,n),T("filter",e,t,n),r,s,a,o)]}case"Conv2D":{let r=T("strides",e,t,n),s=Am(e,t,n),a=T("dataFormat",e,t,n).toUpperCase(),o=T("dilations",e,t,n);return[Xo(T("x",e,t,n),T("filter",e,t,n),[r[1],r[2]],s,a,[o[1],o[2]])]}case"_FusedConv2D":{let{stride:r,pad:s,dataFormat:a,dilations:o,biasArg:i,preluArg:l,activationFunc:u,leakyreluAlpha:c}=Q8(e,t,n);return[ei.conv2d({x:T("x",e,t,n),filter:T("filter",e,t,n),strides:[r[1],r[2]],pad:s,dataFormat:a,dilations:[o[1],o[2]],bias:i,activation:u,preluActivationWeights:l,leakyreluAlpha:c})]}case"FusedDepthwiseConv2dNative":{let{stride:r,pad:s,dataFormat:a,dilations:o,biasArg:i,preluArg:l,activationFunc:u,leakyreluAlpha:c}=Q8(e,t,n);return[ei.depthwiseConv2d({x:T("x",e,t,n),filter:T("filter",e,t,n),strides:[r[1],r[2]],pad:s,dataFormat:a,dilations:[o[1],o[2]],bias:i,activation:u,preluActivationWeights:l,leakyreluAlpha:c})]}case"Conv2DBackpropInput":case"Conv2dTranspose":{let r=T("outputShape",e,t,n),s=T("strides",e,t,n),a=Am(e,t,n);return[H6(T("x",e,t,n),T("filter",e,t,n),r,[s[1],s[2]],a)]}case"DepthwiseConv2dNative":case"DepthwiseConv2d":{let r=T("strides",e,t,n),s=Am(e,t,n),a=T("dilations",e,t,n),o=T("dataFormat",e,t,n).toUpperCase();return[kf(T("input",e,t,n),T("filter",e,t,n),[r[1],r[2]],s,o,[a[1],a[2]])]}case"Conv3D":{let r=T("strides",e,t,n),s=T("pad",e,t,n),a=T("dataFormat",e,t,n).toUpperCase(),o=T("dilations",e,t,n);return[G6(T("x",e,t,n),T("filter",e,t,n),[r[1],r[2],r[3]],s,a,[o[1],o[2],o[3]])]}case"AvgPool":{let r=T("strides",e,t,n),s=T("pad",e,t,n),a=T("kernelSize",e,t,n);return[_y(T("x",e,t,n),[a[1],a[2]],[r[1],r[2]],s)]}case"MaxPool":{let r=T("strides",e,t,n),s=T("pad",e,t,n),a=T("kernelSize",e,t,n);return[Vy(T("x",e,t,n),[a[1],a[2]],[r[1],r[2]],s)]}case"MaxPoolWithArgmax":{let r=T("strides",e,t,n),s=T("pad",e,t,n),a=T("kernelSize",e,t,n),o=T("includeBatchInIndex",e,t,n),{result:i,indexes:l}=cK(T("x",e,t,n),[a[1],a[2]],[r[1],r[2]],s,o);return[i,l]}case"AvgPool3D":{let r=T("strides",e,t,n),s=T("pad",e,t,n),a=T("kernelSize",e,t,n);return[W6(T("x",e,t,n),[a[1],a[2],a[3]],[r[1],r[2],r[3]],s)]}case"MaxPool3D":{let r=T("strides",e,t,n),s=T("pad",e,t,n),a=T("kernelSize",e,t,n);return[nI(T("x",e,t,n),[a[1],a[2],a[3]],[r[1],r[2],r[3]],s)]}case"Dilation2D":{let r=T("strides",e,t,n),s=T("pad",e,t,n),a=T("dilations",e,t,n),o=r[1],i=r[2],l=a[1],u=a[2];return[uq(T("x",e,t,n),T("filter",e,t,n),[o,i],s,[l,u],"NHWC")]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Cse=(e,t,n)=>{switch(e.op){case"Fill":{let r=T("shape",e,t,n),s=T("dtype",e,t,n),a=T("value",e,t,n);return[Nd(r,a,s)]}case"LinSpace":{let r=T("start",e,t,n),s=T("stop",e,t,n),a=T("num",e,t,n);return[Bq(r,s,a)]}case"Multinomial":{let r=T("logits",e,t,n),s=T("numSamples",e,t,n),a=T("seed",e,t,n);return[kK(r,s,a)]}case"OneHot":{let r=T("indices",e,t,n),s=T("depth",e,t,n),a=T("onValue",e,t,n),o=T("offValue",e,t,n);return[xf(r,s,a,o)]}case"Ones":return[ua(T("shape",e,t,n),T("dtype",e,t,n))];case"OnesLike":return[qr(T("x",e,t,n))];case"RandomUniform":return[$d(T("shape",e,t,n),T("minval",e,t,n),T("maxval",e,t,n),T("dtype",e,t,n))];case"Range":{let r=T("start",e,t,n),s=T("stop",e,t,n),a=T("step",e,t,n);return[Rd(r,s,a,T("dtype",e,t,n))]}case"TruncatedNormal":{let r=T("shape",e,t,n),s=T("mean",e,t,n),a=T("stdDev",e,t,n),o=T("seed",e,t,n);return[Qy(r,s,a,T("dtype",e,t,n),o)]}case"Zeros":return[ln(T("shape",e,t,n),T("dtype",e,t,n))];case"ZerosLike":return[ot(T("x",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}};function f5(e,t,n){let r=T("boxes",e,t,n),s=T("scores",e,t,n),a=T("maxOutputSize",e,t,n),o=T("iouThreshold",e,t,n),i=T("scoreThreshold",e,t,n),l=T("softNmsSigma",e,t,n);return{boxes:r,scores:s,maxOutputSize:a,iouThreshold:o,scoreThreshold:i,softNmsSigma:l}}var Ese=async(e,t,n)=>{switch(e.op){case"NonMaxSuppressionV5":{let{boxes:r,scores:s,maxOutputSize:a,iouThreshold:o,scoreThreshold:i,softNmsSigma:l}=f5(e,t,n),u=await is.nonMaxSuppressionWithScoreAsync(r,s,a,o,i,l);return[u.selectedIndices,u.selectedScores]}case"NonMaxSuppressionV4":{let{boxes:r,scores:s,maxOutputSize:a,iouThreshold:o,scoreThreshold:i}=f5(e,t,n),l=T("padToMaxOutputSize",e,t,n),u=await is.nonMaxSuppressionPaddedAsync(r,s,a,o,i,l);return[u.selectedIndices,u.validOutputs]}case"NonMaxSuppressionV3":case"NonMaxSuppressionV2":{let{boxes:r,scores:s,maxOutputSize:a,iouThreshold:o,scoreThreshold:i}=f5(e,t,n);return[await is.nonMaxSuppressionAsync(r,s,a,o,i)]}case"Where":{let r=xe(T("condition",e,t,n),"bool"),s=[await VX(r)];return r.dispose(),s}case"ListDiff":return lX(T("x",e,t,n),T("y",e,t,n));default:throw TypeError(`Node type ${e.op} is not implemented`)}},$se=(e,t,n)=>{switch(e.op){case"TopKV2":{let r=T("x",e,t,n),s=T("k",e,t,n),a=T("sorted",e,t,n),o=FX(r,s,a);return[o.values,o.indices]}case"Unique":{let r=T("x",e,t,n),s=fI(r);return[s.values,s.indices]}case"UniqueV2":{let r=T("x",e,t,n),s=T("axis",e,t,n),a=fI(r,s);return[a.values,a.indices]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Rse=(e,t,n)=>{switch(e.op){case"Const":return t[e.name];case"PlaceholderWithDefault":let r=T("default",e,t,n);return[Bn(e.name,t,n)||r];case"Placeholder":return[Bn(e.name,t,n)];case"Identity":case"StopGradient":case"FakeQuantWithMinMaxVars":{let u=T("x",e,t,n);return[ya(u)]}case"IdentityN":return T("x",e,t,n).map(u=>ya(u));case"Snapshot":let s=T("x",e,t,n);return[ya(s)];case"Shape":return[En(T("x",e,t,n).shape,"int32")];case"ShapeN":return T("x",e,t,n).map(u=>En(u.shape));case"Size":return[De(T("x",e,t,n).size,"int32")];case"Rank":return[De(T("x",e,t,n).rank,"int32")];case"NoOp":return[De(1)];case"Print":let a=T("x",e,t,n),o=T("data",e,t,n),i=T("message",e,t,n),l=T("summarize",e,t,n);console.warn("The graph has a tf.print() operation,usually used for debugging, which slows down performance."),console.log(i);for(let u=0;u<o.length;u++)console.log(Array.prototype.slice.call(o[u].dataSync()).slice(0,l));return[a];default:throw TypeError(`Node type ${e.op} is not implemented`)}},_se=class{constructor(e,t){this.keyDType=e,this.valueDType=t,this.handle=De(0),this.tensorMap=new Map,In(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 De(this.size(),"int32")}async import(e,t){this.checkKeyAndValueTensor(e,t);let n=await e.data();return this.tensorMap.forEach(r=>r.dispose()),this.tensorMap.clear(),Y(()=>{let r=Ds(t),s=n.length,a=r.length;k.assert(s===a,()=>`The number of elements doesn't match, keys has ${s} elements, the values has ${a} elements.`);for(let o=0;o<s;o++){let i=n[o],l=r[o];In(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 r=[];for(let s=0;s<n.length;s++){let a=n[s],o=this.findWithDefault(a,t);r.push(o)}return Xr(r)})}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}`)}},Dse=async(e,t,n,r)=>{switch(e.op){case"HashTable":case"HashTableV2":{let s=T("keyDType",e,t,n),a=T("valueDType",e,t,n),o=new _se(s,a);return r.addHashTable(e.name,o),[o.handle]}case"LookupTableImport":case"LookupTableImportV2":{let s=T("tableHandle",e,t,n,r),a=T("keys",e,t,n),o=T("values",e,t,n);return[await r.getHashTableById(s.id).import(a,o)]}case"LookupTableFind":case"LookupTableFindV2":{let s=T("tableHandle",e,t,n,r),a=T("keys",e,t,n),o=T("defaultValue",e,t,n);return[await r.getHashTableById(s.id).find(a,o)]}case"LookupTableSize":case"LookupTableSizeV2":{let s=T("tableHandle",e,t,n,r);return[r.getHashTableById(s.id).tensorSize()]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Fse=(e,t,n)=>{switch(e.op){case"ResizeBilinear":{let r=T("images",e,t,n),s=T("size",e,t,n),a=T("alignCorners",e,t,n),o=T("halfPixelCenters",e,t,n);return[is.resizeBilinear(r,[s[0],s[1]],a,o)]}case"ResizeNearestNeighbor":{let r=T("images",e,t,n),s=T("size",e,t,n),a=T("alignCorners",e,t,n),o=T("halfPixelCenters",e,t,n);return[is.resizeNearestNeighbor(r,[s[0],s[1]],a,o)]}case"CropAndResize":{let r=T("image",e,t,n),s=T("boxes",e,t,n),a=T("boxInd",e,t,n),o=T("cropSize",e,t,n),i=T("method",e,t,n),l=T("extrapolationValue",e,t,n);return[is.cropAndResize(r,s,a,o,i,l)]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Mse=(e,t,n)=>{switch(e.op){case"Equal":return[Hr(T("a",e,t,n),T("b",e,t,n))];case"NotEqual":return[Ed(T("a",e,t,n),T("b",e,t,n))];case"Greater":return[$r(T("a",e,t,n),T("b",e,t,n))];case"GreaterEqual":return[Kl(T("a",e,t,n),T("b",e,t,n))];case"Less":return[Z6(T("a",e,t,n),T("b",e,t,n))];case"LessEqual":return[Xl(T("a",e,t,n),T("b",e,t,n))];case"LogicalAnd":return[_s(T("a",e,t,n),T("b",e,t,n))];case"LogicalNot":return[Wy(T("a",e,t,n))];case"LogicalOr":return[tI(T("a",e,t,n),T("b",e,t,n))];case"Select":case"SelectV2":return[Zn(T("condition",e,t,n),T("a",e,t,n),T("b",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Ose=(e,t,n)=>{switch(e.op){case"BatchMatMul":case"BatchMatMulV2":case"MatMul":return[rt(T("a",e,t,n),T("b",e,t,n),T("transposeA",e,t,n),T("transposeB",e,t,n))];case"Einsum":return[Aq(T("equation",e,t,n),...T("tensors",e,t,n))];case"Transpose":return[st(T("x",e,t,n),T("perm",e,t,n))];case"_FusedMatMul":let[r,s]=T("fusedOps",e,t,n),a=r==="biasadd",o=s==="prelu",i=T("numArgs",e,t,n),l=T("leakyreluAlpha",e,t,n);if(a){if(o&&i!==2)throw new Error("Fused MatMul with BiasAdd and Prelu must have two extra arguments: bias and alpha.");if(!o&&i!==1)throw new Error("Fused MatMul with BiasAdd must have one extra argument: bias.")}let[u,c]=T("args",e,t,n);return[ei.matMul({a:T("a",e,t,n),b:T("b",e,t,n),transposeA:T("transposeA",e,t,n),transposeB:T("transposeB",e,t,n),bias:u,activation:s,preluActivationWeights:c,leakyreluAlpha:l})];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Pse=(e,t,n)=>{switch(e.op){case"FusedBatchNorm":case"FusedBatchNormV2":return[Td(T("x",e,t,n),T("mean",e,t,n),T("variance",e,t,n),T("offset",e,t,n),T("scale",e,t,n),T("epsilon",e,t,n))];case"FusedBatchNormV3":return[Td(T("x",e,t,n),T("mean",e,t,n),T("variance",e,t,n),T("offset",e,t,n),T("scale",e,t,n),T("epsilon",e,t,n))];case"LRN":return[Vq(T("x",e,t,n),T("radius",e,t,n),T("bias",e,t,n),T("alpha",e,t,n),T("beta",e,t,n))];case"Softmax":return[Yy(T("x",e,t,n))];case"LogSoftmax":return[Y6(T("x",e,t,n))];case"SparseToDense":return[AI(T("sparseIndices",e,t,n),T("outputShape",e,t,n),T("sparseValues",e,t,n),T("defaultValue",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},zse=(e,t,n)=>{switch(e.op){case"Max":{let o=T("axis",e,t,n),i=T("keepDims",e,t,n);return[Rs(T("x",e,t,n),o,i)]}case"Mean":{let o=T("axis",e,t,n),i=T("keepDims",e,t,n);return[Xt(T("x",e,t,n),o,i)]}case"Min":{let o=T("axis",e,t,n),i=T("keepDims",e,t,n);return[Uy(T("x",e,t,n),o,i)]}case"Sum":{let o=T("axis",e,t,n),i=T("keepDims",e,t,n);return[Te(T("x",e,t,n),o,i)]}case"All":{let o=T("axis",e,t,n),i=T("keepDims",e,t,n);return[P6(T("x",e,t,n),o,i)]}case"Any":{let o=T("axis",e,t,n),i=T("keepDims",e,t,n);return[Ey(T("x",e,t,n),o,i)]}case"ArgMax":{let o=T("axis",e,t,n);return[kd(T("x",e,t,n),o)]}case"ArgMin":{let o=T("axis",e,t,n);return[JG(T("x",e,t,n),o)]}case"Prod":{let o=T("axis",e,t,n),i=T("keepDims",e,t,n);return[rI(T("x",e,t,n),o,i)]}case"Cumsum":{let o=T("axis",e,t,n),i=T("exclusive",e,t,n),l=T("reverse",e,t,n);return[K6(T("x",e,t,n),o,i,l)]}case"Bincount":let r=T("x",e,t,n),s=T("weights",e,t,n),a=T("size",e,t,n);return[V6(r,s,a)];case"DenseBincount":{let o=T("x",e,t,n),i=T("weights",e,t,n),l=T("size",e,t,n),u=T("binaryOutput",e,t,n);return[rq(o,i,l,u)]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Lse=(e,t,n)=>{switch(e.op){case"ConcatV2":case"Concat":{let r=T("n",e,t,n),s=T("axis",e,t,n),a=T("tensors",e,t,n);return a=a.slice(0,r),[an(a,s)]}case"Gather":{let r=T("x",e,t,n),s=T("indices",e,t,n);return[Tf(r,xe(s,"int32"),0)]}case"GatherV2":{let r=T("axis",e,t,n),s=T("batchDims",e,t,n),a=T("x",e,t,n),o=T("indices",e,t,n);return[Tf(a,xe(o,"int32"),r,s)]}case"Reverse":{let r=T("dims",e,t,n),s=[];for(let o=0;o<r.length;o++)r[o]&&s.push(o);let a=T("x",e,t,n);return[Kr(a,s)]}case"ReverseV2":{let r=T("axis",e,t,n),s=T("x",e,t,n);return[Kr(s,r)]}case"Slice":{let r=T("begin",e,t,n),s=T("size",e,t,n);return[at(T("x",e,t,n),r,s)]}case"StridedSlice":{let r=T("begin",e,t,n),s=T("end",e,t,n),a=T("strides",e,t,n),o=T("beginMask",e,t,n),i=T("endMask",e,t,n),l=T("ellipsisMask",e,t,n),u=T("newAxisMask",e,t,n),c=T("shrinkAxisMask",e,t,n),d=T("x",e,t,n);return[$X(d,r,s,a,o,i,l,u,c)]}case"Pack":return Y(()=>{let r=T("axis",e,t,n),s=T("tensors",e,t,n),a=s[0].shape,o=Qo(s[0]).shape,i=s.map(l=>{let u=k.arraysEqual(l.shape,a);if(!u&&!k.arraysEqual(Qo(l).shape,o))throw new Error("the input tensors shape does not match");return u?l:Z(l,a)});return[Xr(i,r)]});case"Unpack":{let r=T("axis",e,t,n),s=T("tensor",e,t,n);return Ds(s,r)}case"Tile":{let r=T("reps",e,t,n);return[$s(T("x",e,t,n),r)]}case"Split":case"SplitV":{let r=T("axis",e,t,n),s=T("numOrSizeSplits",e,t,n),a=T("x",e,t,n);return Rr(a,s,r)}case"ScatterNd":{let r=T("indices",e,t,n),s=T("values",e,t,n),a=T("shape",e,t,n);return[jX(r,s,a)]}case"GatherNd":{let r=T("x",e,t,n),s=T("indices",e,t,n);return[ZX(r,s)]}case"SparseToDense":{let r=T("sparseIndices",e,t,n),s=T("outputShape",e,t,n),a=T("sparseValues",e,t,n),o=T("defaultValue",e,t,n);return[AI(r,a,s,a.dtype===o.dtype?o:xe(o,a.dtype))]}default:throw TypeError(`Node type ${e.op} is not implemented`)}},Bse=(e,t,n)=>{switch(e.op){case"SparseFillEmptyRows":{let{outputIndices:r,outputValues:s,emptyRowIndicator:a,reverseIndexMap:o}=Mf.sparseFillEmptyRows(T("indices",e,t,n),T("values",e,t,n),T("denseShape",e,t,n),T("defaultValue",e,t,n));return[r,s,a,o]}case"SparseReshape":{let{outputIndices:r,outputShape:s}=Mf.sparseReshape(T("inputIndices",e,t,n),T("inputShape",e,t,n),T("newShape",e,t,n));return[r,s]}case"SparseSegmentMean":return[Mf.sparseSegmentMean(T("data",e,t,n),T("indices",e,t,n),T("segmentIds",e,t,n))];case"SparseSegmentSum":return[Mf.sparseSegmentSum(T("data",e,t,n),T("indices",e,t,n),T("segmentIds",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Wse=(e,t,n)=>{switch(e.op){case"FFT":return[dI(T("x",e,t,n))];case"IFFT":return[Jy(T("x",e,t,n))];case"RFFT":return[hI(T("x",e,t,n))];case"IRFFT":return[vX(T("x",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Vse=(e,t,n)=>{switch(e.op){case"StringNGrams":{let{nGrams:r,nGramsSplits:s}=nA.stringNGrams(T("data",e,t,n),T("dataSplits",e,t,n),T("separator",e,t,n),T("nGramWidths",e,t,n),T("leftPad",e,t,n),T("rightPad",e,t,n),T("padWidth",e,t,n),T("preserveShortSequences",e,t,n));return[r,s]}case"StringSplit":{let{indices:r,values:s,shape:a}=nA.stringSplit(T("input",e,t,n),T("delimiter",e,t,n),T("skipEmpty",e,t,n));return[r,s,a]}case"StringToHashBucketFast":return[nA.stringToHashBucketFast(T("input",e,t,n),T("numBuckets",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}},Use=(e,t,n)=>{switch(e.op){case"Cast":return[xe(T("x",e,t,n),T("dtype",e,t,n))];case"ExpandDims":{let r=T("axis",e,t,n);return[Er(T("x",e,t,n),r)]}case"Squeeze":{let r=T("axis",e,t,n);return[Qo(T("x",e,t,n),r)]}case"Reshape":return[Z(T("x",e,t,n),T("shape",e,t,n))];case"MirrorPad":return[gK(T("x",e,t,n),T("padding",e,t,n),T("mode",e,t,n))];case"PadV2":case"Pad":return[Yo(T("x",e,t,n),T("padding",e,t,n),T("constantValue",e,t,n))];case"SpaceToBatchND":{let r=T("blockShape",e,t,n),s=T("paddings",e,t,n);return[Gy(T("x",e,t,n),r,s)]}case"BatchToSpaceND":{let r=T("blockShape",e,t,n),s=T("crops",e,t,n);return[Dy(T("x",e,t,n),r,s)]}case"DepthToSpace":{let r=T("blockSize",e,t,n),s=T("dataFormat",e,t,n).toUpperCase();return[aq(T("x",e,t,n),r,s)]}case"BroadcastTo":return[wf(T("x",e,t,n),T("shape",e,t,n))];default:throw TypeError(`Node type ${e.op} is not implemented`)}};function eT(e,t,n,r){let s=((a,o,i)=>{switch(a.category){case"arithmetic":return Y(()=>xse(a,o,i));case"basic_math":return Y(()=>bse(a,o,i));case"control":return Tse(a,o,i);case"convolution":return Y(()=>Nse(a,o,i));case"creation":return Y(()=>Cse(a,o,i));case"dynamic":return Ese(a,o,i);case"evaluation":return Y(()=>$se(a,o,i));case"image":return Y(()=>Fse(a,o,i));case"graph":return Y(()=>Rse(a,o,i));case"logical":return Y(()=>Mse(a,o,i));case"matrices":return Y(()=>Ose(a,o,i));case"normalization":return Y(()=>Pse(a,o,i));case"reduction":return Y(()=>zse(a,o,i));case"slice_join":return Y(()=>Lse(a,o,i));case"sparse":return Y(()=>Bse(a,o,i));case"spectral":return Y(()=>Wse(a,o,i));case"string":return Y(()=>Vse(a,o,i));case"transformation":return Y(()=>Use(a,o,i));case"hash_table":return Dse(a,o,i,r);case"custom":let l=C8(a.op);if(l&&l.customExecutor)return l.customExecutor(new Ase(a,o,i));throw TypeError(`Custom op ${a.op} is not registered.`);default:throw TypeError(`Unknown op '${a.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 k.isPromise(s)?s.then(a=>[].concat(a)):[].concat(s)}var tT=class{constructor(e={},t={},n={},r={}){this.weightMap=e,this.tensorArrayMap=t,this.tensorListMap=n,this.functionMap=r,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 nT(e,t,n,r){let s=new Set,a=[],o=null,i=null,l=new Set,u=Object.keys(e).map(h=>dr(h)[0]),c=[];r!=null&&(c=r.map(h=>dr(h.name)[0]));let d=[...t];for(;d.length>0;){let h=d.pop();if((rT(h)||Kse(h)||Xse(h))&&o==null&&(o=h,i=o.children.map(p=>p.name).filter(p=>s.has(p))),s.add(h.name),n[h.name]==null&&u.indexOf(h.name)===-1&&c.indexOf(h.name)===-1){if(h.inputs.length===0){a.push(h.name);continue}h.inputs.forEach(p=>{l.has(p.name)||(l.add(p.name),d.push(p))})}}return{inputs:e,outputs:t,usedNodes:s,missingInputs:a,dynamicNode:o,syncInputs:i}}function Hse(e,t,n){let{usedNodes:r,inputs:s}=n,a=[],o=Object.keys(s).map(c=>dr(c)[0]).map(c=>e.nodes[c]),i=e.initNodes;o.forEach(c=>{r.has(c.name)&&a.push(c)}),e.weights.forEach(c=>{r.has(c.name)&&a.push(c)}),i!=null&&i.forEach(c=>{r.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(d=>{!l.has(d.name)&&r.has(d.name)&&d.inputs.every(h=>l.has(h.name))&&a.push(d)})}return u}var Gse=["Switch","Merge","Enter","Exit","NextIteration","StatelessIf","StatelessWhile","if","While"],jse=["NonMaxSuppressionV2","NonMaxSuppressionV3","NonMaxSuppressionV5","Where"],qse=["HashTable","HashTableV2","LookupTableImport","LookupTableImportV2","LookupTableFind","LookupTableFindV2","LookupTableSize","LookupTableSizeV2"];function rT(e){return Gse.indexOf(e.op)>=0}function Kse(e){return jse.indexOf(e.op)>=0}function Xse(e){return qse.indexOf(e.op)>=0}var m5=class{constructor(e,t){this.graph=e,this.parent=t,this.compiledMap=new Map,this._weightMap={},this.SEPERATOR=",",this._functions={},this._functionExecutorMap={},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 m5(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(r=>r.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(s=>s.name).sort(),r=t.map(s=>s.name).sort();return n.join(this.SEPERATOR)+"--"+r.join(this.SEPERATOR)}compile(e,t){let n=nT(e,t,this.weightMap,this._initNodes),{missingInputs:r,dynamicNode:s,syncInputs:a}=n;if(s!=null)throw new Error(`This execution contains the node '${s.name}', which has the dynamic op '${s.op}'. Please use model.executeAsync() instead. Alternatively, to avoid the dynamic ops, specify the inputs [${a}]`);if(r.length>0){let o=t.map(l=>l.name),i=Object.keys(e);throw new Error(`Cannot compute the outputs [${o}] from the provided inputs [${i}]. Missing the following inputs: [${r}]`)}return Hse(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 r=n.map(c=>this.graph.nodes[dr(c)[0]]),s=t.map(c=>dr(c)[0]),a=s.map(c=>this.graph.nodes[c]);a.length===0&&(a=this._outputs);let o=this.getCompilationKey(r,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 tT(this.weightMap,l,u,this.functionExecutorMap),d={...this.weightMap};Object.keys(e).forEach(f=>{let[m,g]=dr(f),y=[];y[g]=e[f],d[m]=y});let h=this.getFrozenTensorIds(d),p={};for(let f=0;f<i.length;f++){let m=i[f];if(!d[m.name]){let g=eT(m,d,c,this._resourceManager);if(k.isPromise(g))throw new Error(`The execution of the op '${m.op}' returned a promise. Please use model.executeAsync() instead.`);d[m.name]=g,this.checkTensorForDisposal(m.name,m,d,c,h,s,p)}}return this.parent==null&&c.dispose(h),t.map(f=>Bn(f,d,c))})}getFrozenTensorIds(e){let t=[].concat.apply([],Object.keys(e).map(n=>e[n]).map(n=>n.map(r=>r.id)));return new Set(t)}checkTensorForDisposal(e,t,n,r,s,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=Yre(i.name,n,r);l!=null&&l.forEach(u=>{if(u&&!u.kept&&!s.has(u.id)){let c=o[u.id];c===1?(u.dispose(),delete o[u.id]):c!=null&&o[u.id]--}})}}))}async executeAsync(e,t){return this._executeAsync(e,t)}async _executeAsync(e,t,n=!1,r={},s={}){n||(e=this.mapInputs(e),this.checkInputs(e),this.checkInputShapeAndType(e),t=this.mapOutputs(t),this.checkOutputs(t));let a=new tT(this.weightMap,r,s,this.functionExecutorMap),o=await this.executeWithControlFlow(e,a,t,n),i=t.map(d=>Bn(d,o,a)),l=i.map(d=>d.id),u=Object.keys(e).map(d=>e[d].id),c=new Set([...l,...u,...this.weightIds]);return Object.keys(o).forEach(d=>{o[d].forEach(p=>{p&&!p.kept&&!p.isDisposed&&!c.has(p.id)&&p.dispose()})}),this.parent==null&&a.dispose(c),i}async executeFunctionAsync(e,t,n){let r=e.reduce((s,a,o)=>(s[this.inputs[o].name]=a,s),{});return this._executeAsync(r,this.outputNodes,!0,t,n)}async executeWithControlFlow(e,t,n,r){let s=Object.keys(e),a=s.map(A=>this.graph.nodes[dr(A)[0]]),o=n.map(A=>dr(A)[0]),i=o.map(A=>this.graph.nodes[A]);i.length===0&&(i=this._outputs);let{usedNodes:l,missingInputs:u,dynamicNode:c,syncInputs:d}=nT(e,i,this.weightMap,this._initNodes),h=[...a,...this.graph.weights,...this._initNodes||[]].map(A=>({node:A,contexts:t.currentContext})),p={...this.weightMap};Object.keys(e).forEach(A=>{let[x,b]=dr(A),v=[];v[b]=e[A],p[x]=v});let f={},m=this.getFrozenTensorIds(p),g={};for(;h.length>0;){let A=this.processStack(a,h,t,p,g,m,o,f,l);await Promise.all(A)}c==null&&!r&&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(A=>!rT(A)&&!Bn(A.name,p,t)).map(A=>A.name);if(y.length>0){let A="";throw c!=null&&(A=`Alternatively, to avoid the dynamic ops, use model.execute() and specify the inputs [${d}]`),new Error(`Cannot compute the outputs [${y}] from the provided inputs [${s}]. Consider providing the following inputs: [${u}]. ${A}`)}return p}processStack(e,t,n,r,s,a,o,i,l){let u=[];for(;t.length>0;){let c=t.pop();n.currentContext=c.contexts;let d="";if(c.node.op==="Enter"&&T("isConstant",c.node,r,n)&&([d]=ga(c.node.name,n)),r[c.node.name]==null){let h=eT(c.node,r,n,this._resourceManager);d||([d]=ga(c.node.name,n));let p=n.currentContext;k.isPromise(h)?u.push(h.then(f=>(r[d]=f,n.currentContext=p,this.checkTensorForDisposal(d,c.node,r,n,a,o,i),this.processChildNodes(c.node,t,n,r,s,l),f))):(r[d]=h,this.checkTensorForDisposal(d,c.node,r,n,a,o,i),this.processChildNodes(c.node,t,n,r,s,l))}else this.processChildNodes(c.node,t,n,r,s,l)}return u}processChildNodes(e,t,n,r,s,a){e.children.forEach(o=>{let[i]=ga(o.name,n);s[i]||!a.has(o.name)||(o.op==="Merge"?o.inputNames.some(l=>!!Bn(l,r,n))&&(s[i]=!0,t.push({contexts:n.currentContext,node:o})):o.inputNames.every(l=>!!Bn(l,r,n))&&(s[i]=!0,t.push({contexts:n.currentContext,node:o})))})}dispose(){Object.keys(this.weightMap).forEach(e=>this.weightMap[e].forEach(t=>t.dispose()))}checkInputShapeAndType(e){Object.keys(e).forEach(t=>{let n=e[t],[r]=dr(t),s=this.graph.nodes[r];if(s.attrParams.shape&&s.attrParams.shape.value){let a=s.attrParams.shape.value,o=a.length===n.shape.length&&n.shape.every((i,l)=>a[l]===-1||a[l]===i);k.assert(o,()=>`The shape of dict['${s.name}'] provided in model.execute(dict) must be [${a}], but was [${n.shape}]`)}s.attrParams.dtype&&s.attrParams.dtype.value&&k.assert(n.dtype===s.attrParams.dtype.value,()=>`The dtype of dict['${s.name}'] provided in model.execute(dict) must be ${s.attrParams.dtype.value}, but was ${n.dtype}`)})}mapInputs(e){let t={};for(let n in e)if(this._signature!=null&&this._signature.inputs!=null&&this._signature.inputs[n]!=null){let r=this._signature.inputs[n];t[r.name]=e[n]}else t[n]=e[n];return t}checkInputs(e){let t=Object.keys(e).filter(n=>{let[r]=dr(n);return this.graph.nodes[r]==null});if(t.length>0)throw new Error(`The dict provided in model.execute(dict) has keys: [${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]=dr(t);if(!this.graph.nodes[n])throw new Error(`The output '${t}' is not found in the graph`)})}},Zse=class{constructor(e={},t={}){this.hashTableNameToHandle=e,this.hashTableMap=t}addHashTable(e,t){this.hashTableNameToHandle[e]=t.handle,this.hashTableMap[t.id]=t}getHashTableHandleByName(e){return this.hashTableNameToHandle[e]}getHashTableById(e){return this.hashTableMap[e]}dispose(){for(let e in this.hashTableMap)this.hashTableMap[e].clearAndClose(),delete this.hashTableMap[e];for(let e in this.hashTableNameToHandle)this.hashTableNameToHandle[e].dispose(),delete this.hashTableNameToHandle[e]}},Yse="?tfjs-format=file",Jse="model.json",sT=class{constructor(e,t={}){this.modelUrl=e,this.loadOptions=t,this.version="n/a",t==null&&(this.loadOptions={}),this.resourceManager=new Zse}get modelVersion(){return this.version}get inputNodes(){return this.executor.inputNodes}get outputNodes(){return this.executor.outputNodes}get inputs(){return this.executor.inputs}get outputs(){return this.executor.outputs}get weights(){return this.executor.weightMap}get metadata(){return this.artifacts.userDefinedMetadata}get modelSignature(){return this.signature}findIOHandler(){let e=this.modelUrl;if(e.load!=null)this.handler=e;else if(this.loadOptions.requestInit!=null)this.handler=cr.browserHTTPRequest(e,this.loadOptions);else{let t=cr.getLoadHandlers(e,this.loadOptions);if(t.length===0)t.push(cr.browserHTTPRequest(e,this.loadOptions));else if(t.length>1)throw new Error(`Found more than one (${t.length}) load handlers for URL '${[e]}'`);this.handler=t[0]}}async load(){if(this.findIOHandler(),this.handler.load==null)throw new Error("Cannot proceed with model loading because the IOHandler provided does not have the `load` method implemented.");let e=await this.handler.load();return this.loadSync(e)}loadSync(e){this.artifacts=e;let t=this.artifacts.modelTopology,n;this.artifacts.userDefinedMetadata!=null&&this.artifacts.userDefinedMetadata.signature!=null?n=this.artifacts.userDefinedMetadata.signature:n=this.artifacts.signature,this.signature=n,this.version=`${t.versions.producer}.${t.versions.minConsumer}`;let r=cr.decodeWeights(this.artifacts.weightData,this.artifacts.weightSpecs);if(this.executor=new m5(K8.Instance.transformGraph(t,this.signature)),this.executor.weightMap=this.convertTensorMapToTensorsMap(r),this.executor.resourceManager=this.resourceManager,e.modelInitializer!=null&&e.modelInitializer.node!=null){let s=K8.Instance.transformGraph(e.modelInitializer);this.initializer=new m5(s),this.initializer.weightMap=this.executor.weightMap,this.initializer.resourceManager=this.resourceManager,this.initializer.executeAsync({},[])}return!0}async save(e,t){if(typeof e=="string"){let n=cr.getSaveHandlers(e);if(n.length===0)throw new Error(`Cannot find any save handlers for URL '${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){return this.execute(e,this.outputNodes)}normalizeInputs(e){if(!(e instanceof Ot)&&!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 ${e.length} input tensors.`);return this.inputNodes.reduce((t,n,r)=>(t[n]=e[r],t),{})}normalizeOutputs(e){return e=e||this.outputNodes,Array.isArray(e)?e:[e]}execute(e,t){e=this.normalizeInputs(e),t=this.normalizeOutputs(t);let n=this.executor.execute(e,t);return n.length>1?n:n[0]}async executeAsync(e,t){e=this.normalizeInputs(e),t=this.normalizeOutputs(t);let n=await this.executor.executeAsync(e,t);return n.length>1?n:n[0]}convertTensorMapToTensorsMap(e){return Object.keys(e).reduce((t,n)=>(t[n]=[e[n]],t),{})}dispose(){this.executor.dispose(),this.initializer&&this.initializer.dispose(),this.resourceManager.dispose()}};async function Nt(e,t={}){if(e==null)throw new Error("modelUrl in loadGraphModel() cannot be null. Please provide a url or an IOHandler that loads the model");t==null&&(t={}),t.fromTFHub&&e.load==null&&(e.endsWith("/")||(e=e+"/"),e=`${e}${Jse}${Yse}`);let n=new sT(e,t);return await n.load(),n}var Qse="3.8.0",aT={};_e(aT,{CSVDataset:()=>xT,Dataset:()=>ou,FileDataSource:()=>TT,TextLineDataset:()=>gT,URLDataSource:()=>NT,array:()=>wae,csv:()=>Dae,func:()=>Fae,generator:()=>Mae,microphone:()=>Pae,version_data:()=>zae,webcam:()=>Oae,zip:()=>kae});var eae=Xs(C3()),tae=Xs(C3());function nae(e,t){return xm(e,t)}function xm(e,t,n=new Map,r=new Set){if(e==null)return null;if(r.has(e))throw new Error("Circular references are not supported.");if(n.has(e))return n.get(e);let s=t(e);if(s.recurse&&s.value!==null)throw new Error("A deep map function may not return both a value and recurse=true.");if(s.recurse)if(au(e)){let a=Array.isArray(e)?[]:{};r.add(e);for(let o in e){let i=e[o],l=xm(i,t,n,r);a[o]=l}return r.delete(e),a}else throw new Error(`Can't recurse into non-iterable type: ${e}`);else return n.set(e,s.value),s.value}function rae(e,t=iT){return oT(e,t)}function oT(e,t,n=new Set){let r=e[0];if(n.has(r))throw new Error("Circular references are not supported.");let s=t(e);if(s.recurse&&s.value!==null)throw new Error("A deep zip function may not return both a value and recurse=true.");if(s.recurse)if(au(r)){let a=Array.isArray(r)?[]:{};n.add(r);for(let o in r){let i=e.map(u=>u[o]),l=oT(i,t,n);a[o]=l}return n.delete(r),a}else throw new Error(`Can't recurse into non-iterable type: ${r}`);else return s.value}function iT(e){return e===null?null:au(e[0])?{value:null,recurse:!0}:{value:e,recurse:!1}}async function lT(e,t){let n=new Map;xm(e,t,n);for(let s of Array.from(n.keys())){let a=n.get(s);if(k.isPromise(a)){let o=await a;n.set(s,o)}}return xm(e,t,n)}function au(e){return e!=null&&!ArrayBuffer.isView(e)&&(Array.isArray(e)||typeof e=="object"&&!(e instanceof Ot))}function sae(e){return e==null||aae(e)||Array.isArray(e)||typeof e=="object"&&e instanceof Ot||k.isTypedArray(e)}function aae(e){return e===null||typeof e!="object"&&typeof e!="function"}function oae(e){return nae(e,iae)}function iae(e){return e instanceof Ot?{value:e.clone(),recurse:!1}:au(e)?{value:null,recurse:!0}:{value:e,recurse:!1}}var uT=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}},cT=class extends uT{constructor(){super(cT.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 r=0;r<n;r++)t[r]=this.get(this.wrap(this.begin+r));this.data=t,this.capacity=e,this.doubledCapacity=2*this.capacity,this.begin=0,this.end=n}},dT=cT;dT.INITIAL_CAPACITY=32;function hT(e){return new cae(e)}function g5(e){return new dae(e)}function lae(e,t){return new fT(e,t)}function uae(e,t=bm.FAIL){return new bae(e,t)}var An=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 Aae(this,e)}filter(e){return new gae(this,e)}map(e){return new yae(this,e)}mapAsync(e){return new pT(this,e)}serialMapAsync(e){return new pT(this,e).serial()}flatmap(e){return new xae(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 mae(this,e,t)}columnMajorBatch(e,t=!0,n=iT){return this.rowMajorBatch(e,t).map(s=>rae(s,n))}concatenate(e,t){return new fT(hT([this,e]),t)}take(e){return e<0||e==null?this:new fae(this,e)}skip(e){return e<0||e==null?this:new pae(this,e)}prefetch(e){return new mT(this,e)}shuffle(e,t){return new vae(this,e,t)}serial(){return new hae(this)}},cae=class extends An{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:oae(e),done:!1}}},dae=class extends An{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: ${e.message}`,e}}},hae=class extends An{constructor(e){super();this.upstream=e,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${this.upstream.summary()} -> Serial`}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){return this.upstream.next()}},pae=class extends An{constructor(e,t){super();this.upstream=e,this.maxCount=t,this.count=0,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${this.upstream.summary()} -> Skip`}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){for(;this.count++<this.maxCount;){let e=await this.upstream.next();if(e.done)return e;Ge(e.value)}return this.upstream.next()}},fae=class extends An{constructor(e,t){super();this.upstream=e,this.maxCount=t,this.count=0}summary(){return`${this.upstream.summary()} -> Take`}async next(){return this.count++>=this.maxCount?{value:null,done:!0}:this.upstream.next()}},mae=class extends An{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`${this.upstream.summary()} -> RowMajorBatch`}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){let e=[];for(;e.length<this.batchSize;){let t=await this.upstream.next();if(t.done)return this.enableSmallLastBatch&&e.length>0?{value:e,done:!1}:{value:null,done:!0};e.push(t.value)}return{value:e,done:!1}}},gae=class extends An{constructor(e,t){super();this.upstream=e,this.predicate=t,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${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;Ge(e.value)}}},yae=class extends An{constructor(e,t){super();this.upstream=e,this.transform=t}summary(){return`${this.upstream.summary()} -> Map`}async next(){let e=await this.upstream.next();if(e.done)return{value:null,done:!0};let t=Ns.getTensorsInContainer(e.value),n=this.transform(e.value),r=Ns.getTensorsInContainer(n);for(let s of t)Ns.isTensorInList(s,r)||s.dispose();return{value:n,done:!1}}},Aae=class extends An{constructor(e,t){super();this.upstream=e,this.handler=t,this.count=0,this.lastRead=Promise.resolve({value:null,done:!1})}summary(){return`${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}}}},pT=class extends An{constructor(e,t){super();this.upstream=e,this.transform=t}summary(){return`${this.upstream.summary()} -> AsyncMap`}async next(){let e=await this.upstream.next();if(e.done)return{value:null,done:!0};let t=Ns.getTensorsInContainer(e.value),n=await this.transform(e.value),r=Ns.getTensorsInContainer(n);for(let s of t)Ns.isTensorInList(s,r)||s.dispose();return{value:n,done:!1}}},y5=class extends An{constructor(){super();this.outputQueue=new dT,this.lastRead=Promise.resolve({value:null,done:!1})}async next(){return this.lastRead=this.lastRead.then(()=>this.serialNext()),this.lastRead}async serialNext(){for(;this.outputQueue.length()===0;)if(!await this.pump())return{value:null,done:!0};return{value:this.outputQueue.shift(),done:!1}}},xae=class extends y5{constructor(e,t){super();this.upstream=e,this.transform=t}summary(){return`${this.upstream.summary()} -> Flatmap`}async pump(){let e=await this.upstream.next();if(e.done)return!1;let t=Ns.getTensorsInContainer(e.value),n=this.transform(e.value),r=Ns.getTensorsInContainer(n);this.outputQueue.pushAll(n);for(let s of t)Ns.isTensorInList(s,r)||s.dispose();return!0}},fT=class extends An{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}},bm;(function(e){e[e.FAIL=0]="FAIL",e[e.SHORTEST=1]="SHORTEST",e[e.LONGEST=2]="LONGEST"})(bm||(bm={}));var bae=class extends An{constructor(e,t=0){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 r(a){return a instanceof An?{value:a.next().then(i=>(t++,i.done&&n++,i.value)),recurse:!1}:{value:null,recurse:!0}}let s=await lT(this.iterators,r);if(t===n)return{value:null,done:!0};if(n>0)switch(this.mismatchMode){case 0:throw new Error(`Zipped streams should have the same length. Mismatched at element ${this.count}.`);case 1:return{value:null,done:!0};case 2:default:}return this.count++,{value:s,done:!1}}async next(){return this.currentPromise=this.nextState(this.currentPromise),this.currentPromise}},mT=class extends An{constructor(e,t){super();this.upstream=e,this.bufferSize=t,this.buffer=new uT(t)}summary(){return`${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()}},vae=class extends mT{constructor(e,t,n){super(e,t);this.upstream=e,this.windowSize=t,this.upstreamExhausted=!1,this.random=tae.alea(n||k.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}}},ou=class{constructor(){this.size=null}batch(e,t=!0){let n=this;k.assert(e>0,()=>`batchSize needs to be positive, but it is
${e}`);let r;return this.size===1/0||this.size==null?r=this.size:t?r=Math.ceil(this.size/e):r=Math.floor(this.size/e),hr(async()=>(await n.iterator()).columnMajorBatch(e,t,Iae),r)}concatenate(e){let t=this,n;return this.size===1/0||e.size===1/0?n=1/0:this.size!=null&&e.size!=null?n=this.size+e.size:n=null,hr(async()=>(await t.iterator()).concatenate(await e.iterator()),n)}filter(e){let t=this,n;return this.size===1/0?n=1/0:n=null,hr(async()=>(await t.iterator()).filter(r=>Y(()=>e(r))),n)}async forEachAsync(e){return(await this.iterator()).forEachAsync(e)}map(e){let t=this;return hr(async()=>(await t.iterator()).map(n=>Y(()=>e(n))),this.size)}mapAsync(e){let t=this;return hr(async()=>(await t.iterator()).mapAsync(e),this.size)}prefetch(e){if(e==null)throw new RangeError("`Dataset.prefetch()` requires bufferSize to be specified.");let t=this;return hr(async()=>(await t.iterator()).prefetch(e),this.size)}repeat(e){let t=this,n;return this.size!=null&&e>0?n=this.size*e:e===0?n=0:this.size!=null&&(e===void 0||e<0)?n=1/0:n=null,hr(async()=>{let r=g5(async()=>({value:await t.iterator(),done:!1}));return lae(r.take(e))},n)}skip(e){let t=this,n;return this.size!=null&&e>=0&&this.size>=e?n=this.size-e:this.size!=null&&(this.size<e||e===void 0||e<0)?n=0:n=null,hr(async()=>(await t.iterator()).skip(e),n)}shuffle(e,t,n=!0){if(e==null||e<0)throw this.size==null?new RangeError("`Dataset.shuffle()` requires bufferSize to be specified."):new RangeError(`\`Dataset.shuffle()\` requires bufferSize to be specified. If your data fits in main memory (for regular JS objects), and/or GPU memory (for \`tf.Tensor\`s), consider setting bufferSize to the dataset size (${this.size} elements)`);let r=this,s=eae.alea(t||k.now().toString());return hr(async()=>{let a=s.int32();return n&&(a+=s.int32()),(await r.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,hr(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()}};ou.MAX_BUFFER_SIZE=1e4;function hr(e,t=null){return new class extends ou{constructor(){super(...arguments);this.size=t}async iterator(){return e()}}}function wae(e){return hr(async()=>hT(e),e.length)}function kae(e){if(!au(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 hr(async()=>{let n=await lT(e,r=>{if(r instanceof ou)return{value:r.iterator(),recurse:!1};if(au(r))return{value:null,recurse:!0};throw new Error("Leaves of the structure passed to zip() must be Datasets, not primitives.")});return uae(n,bm.SHORTEST)},t)}function Iae(e){if(e===null)return null;let t=e[0];return sae(t)?{value:Sae(e),recurse:!1}:{value:null,recurse:!0}}function Sae(e){if(e.length===0)throw new Error("Can't make a batch of zero elements.");return e[0]instanceof Ot?Xr(e):Cs(e)}var gT=class extends ou{constructor(e){super();this.input=e}async iterator(){return(await this.input.iterator()).decodeUTF8().split(`
`).map(r=>(r.endsWith("\r")&&(r=r.slice(0,-1)),r))}},vm='"',Zd=Symbol("out"),yT=Symbol("field"),wm=Symbol("quote"),A5=Symbol("quoteafterquote"),AT=Symbol("quoteinquote"),xT=class extends ou{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 gT(e),t||(t={}),this.hasHeader=t.hasHeader!==!1,this.fullColumnNames=t.columnNames,this.columnConfigs=t.columnConfigs,this.configuredColumnsOnly=t.configuredColumnsOnly,t.delimWhitespace?(k.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&&k.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((r,s)=>(r[s]=r[s]+1||1,r),{}),n=Object.keys(t).filter(r=>t[r]>1);if(k.assert(n.length===0,()=>"Duplicate column names found: "+n.toString()),this.columnConfigs){for(let r of Object.keys(this.columnConfigs))if(this.fullColumnNames.indexOf(r)===-1)throw new Error('The key "'+r+'" 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={},r={};for(let s=0;s<this.fullColumnNames.length;s++){let a=this.fullColumnNames[s],o=this.columnConfigs?this.columnConfigs[a]:null;if(!(this.configuredColumnsOnly&&!o)){let i=t[s],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?r[a]=l:n[a]=l}}return Object.keys(r).length===0?n:{xs:n,ys:r}}getBoolean(e){return e==="1"||e.toLowerCase()==="true"?1:0}parseRow(e,t=!0){let n=[],r=0,s=e.length,a=Zd;for(let o=0;o<s;o++)switch(a){case Zd:switch(e.charAt(o)){case vm:r=o+1,a=wm;break;case this.delimiter:if(r=o+1,this.delimiter===" "&&this.delimWhitespace)break;n.push(""),a=Zd;break;default:a=yT,r=o;break}break;case yT:switch(e.charAt(o)){case this.delimiter:n.push(e.substring(r,o)),a=Zd,r=o+1;break;default:}break;case wm:switch(e.charAt(o)){case vm:a=A5;break;default:}break;case A5:switch(e.charAt(o)){case this.delimiter:n.push(e.substring(r,o-1)),a=Zd,r=o+1;break;case vm:a=wm;break;default:a=AT;break}break;case AT:switch(e.charAt(o)){case vm:a=wm;break;default:}break;default:}if(a===A5?n.push(e.substring(r,s-1)):n.push(e.substring(r)),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}},bT=class extends An{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(re().get("IS_NODE"))throw new Error("microphone API is only supported in browser environment.");let t=new bT(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 r=this.flattenQueue(n.freqDataQueue);e=this.getTensorFromAudioDataArray(r,[this.numFrames,this.columnTruncateLength,1])}if(this.includeWaveform){let r=this.flattenQueue(n.timeDataQueue);t=this.getTensorFromAudioDataArray(r,[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(r=>{let s=setInterval(()=>{this.includeSpectrogram&&(this.analyser.getFloatFrequencyData(this.freqData),this.freqData[0]===-1/0&&r({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(s),r({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((r,s)=>n.set(r,s*t)),n}getTensorFromAudioDataArray(e,t){let n=new Float32Array(k.sizeFromShape(t));return n.set(e,n.length-e.length),Cs(n,t)}},vT=class extends An{constructor(e,t){super();if(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=En([0],"int32"),this.webcamConfig.centerCrop){let n=this.webcamConfig.resizeWidth*1/this.webcamVideoElement.width,r=this.webcamConfig.resizeHeight*1/this.webcamVideoElement.height,s=(1-n)/2,a=(1-r)/2,o=s+n,i=r+a;this.cropBox=Zl([a,s,i,o],[1,4])}else this.cropBox=Zl([0,0,1,1],[1,4])}summary(){return"webcam"}static async create(e,t={}){if(re().get("IS_NODE"))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 vT(e,t);return await n.start(),n}async start(){this.webcamConfig.facingMode&&k.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=y6.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: ${t.message}`)}finally{e.dispose()}else return{value:e,done:!1}}needToResize(){return!!(this.webcamConfig.resizeWidth&&this.webcamConfig.resizeHeight&&(this.webcamVideoElement.width!==this.webcamConfig.resizeWidth||this.webcamVideoElement.height!==this.webcamConfig.resizeHeight))}cropAndResizeFrame(e){return Y(()=>{let t=Er(xe(e,"float32"),0),n;n=is.cropAndResize(t,this.cropBox,this.cropBoxInd,this.cropSize,"bilinear");let r=n.shape;return Z(n,r.slice(1))})}async capture(){return(await this.next()).value}stop(){this.stream.getTracks().forEach(t=>t.stop());try{this.webcamVideoElement.srcObject=null}catch(t){console.log(t),this.webcamVideoElement.src=null}this.isClosed=!0}toArray(){throw new Error("Can not convert infinite video stream to array.")}},wT=class{},kT=class extends An{split(e){return new Tae(this,e)}},Tae=class extends kT{constructor(e,t){super();this.upstream=e,this.impl=new Nae(e,t)}summary(){return this.impl.summary()}async next(){return this.impl.next()}},Nae=class extends y5{constructor(e,t){super();this.upstream=e,this.separator=t,this.carryover=""}summary(){return`${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}},Cae=class extends An{decodeUTF8(){return new Eae(this)}},Eae=class extends kT{constructor(e){super();this.upstream=e,this.impl=new $ae(e)}summary(){return this.impl.summary()}async next(){return this.impl.next()}},$ae=class extends y5{constructor(e){super();if(this.upstream=e,re().get("IS_BROWSER"))this.decoder=new TextDecoder("utf-8");else{let{StringDecoder:t}=f_();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 re().get("IS_BROWSER")?n=this.decoder.decode(t,{stream:!0}):n=this.decoder.write(Buffer.from(t.buffer)),this.outputQueue.push(n),!0}},IT=class extends Cae{constructor(e,t={}){super();this.file=e,this.options=t,k.assert(e instanceof Uint8Array||(re().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 ${this.file}`}async next(){return this.offset>=(this.file instanceof Uint8Array?this.file.byteLength:this.file.size)?{value:null,done:!0}:{value:await new Promise((t,n)=>{let r=this.offset+this.chunkSize;if(this.file instanceof Uint8Array)t(new Uint8Array(this.file.slice(this.offset,r)));else{let s=new FileReader;s.onload=o=>{let i=s.result;if(i instanceof ArrayBuffer&&(i=new Uint8Array(i)),!(i instanceof Uint8Array))return n(new TypeError("FileReader returned unknown type."));t(i)},s.onabort=o=>n(new Error("Aborted")),s.onerror=o=>n(new Error(o.type));let a=this.file.slice(this.offset,r);s.readAsArrayBuffer(a)}this.offset=r}),done:!1}}};async function Rae(e,t={}){let n,r;typeof e=="string"?n=e:(n=e.url,r=_ae(e));let s=await k.fetch(n,r);if(s.ok){let a=new Uint8Array(await s.arrayBuffer());return new IT(a,t)}else throw new Error(s.statusText)}var _ae=e=>({method:e.method,headers:e.headers,body:e.body,mode:e.mode,credentials:e.credentials,cache:e.cache,redirect:e.redirect,referrer:e.referrer,integrity:e.integrity});function ST(e){return typeof e=="string"&&e.substr(0,7)==="file://"}var TT=class extends wT{constructor(e,t={}){super();this.input=e,this.options=t}async iterator(){if(ST(this.input)&&re().get("IS_NODE")){let e=co("fs");this.input=e.readFileSync(this.input.substr(7))}return new IT(this.input,this.options)}},NT=class extends wT{constructor(e,t={}){super();this.url=e,this.fileOptions=t}async iterator(){return ST(this.url)?new TT(this.url,this.fileOptions).iterator():Rae(this.url,this.fileOptions)}};function Dae(e,t={}){return new xT(new NT(e),t)}function Fae(e){let t=g5(e);return hr(async()=>t)}function Mae(e){return hr(async()=>{let t=await e();return g5(()=>t.next())})}async function Oae(e,t){return vT.create(e,t)}async function Pae(e){return bT.create(e)}var zae="3.8.0";function Ne(e,t){Array.isArray(e)||(e=[e]),e.forEach(n=>{n!=null&&k.assert(n.dtype!=="complex64",()=>`${t} does not support complex64 tensors in the CPU backend.`)})}var Lae=da.whereImpl,CT=class extends Lp{constructor(){super();this.blockSize=48,this.firstUse=!0,this.data=new d1(this,Ba())}nextDataId(){return CT.nextDataId++}write(e,t,n){this.firstUse&&(this.firstUse=!1,re().get("IS_NODE")&&_.warn(`
============================
Hi there \u{1F44B}. Looks like you are running TensorFlow.js in Node.js. To speed things up dramatically, install our node backend, which binds to TensorFlow C++, by running npm i @tensorflow/tfjs-node, or npm i @tensorflow/tfjs-node-gpu if you have CUDA. Then call require('@tensorflow/tfjs-node'); (-gpu suffix for CUDA) at the start of your program. Visit https://github.com/tensorflow/tfjs-node for more details.
============================`));let r={id:this.nextDataId()};return this.data.set(r,{values:e,dtype:n,refCount:1}),r}makeTensorInfo(e,t,n){let r;if(t==="string"&&n!=null&&n.length>0&&k.isString(n[0])){let s=n.map(a=>k.encodeString(a));r=this.write(s,e,t)}else r=this.write(n,e,t);return{dataId:r,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,r,s){this.data.set(e,{values:t,dtype:r,refCount:s})}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 r=this.readSync(n.real.dataId),s=this.readSync(n.imag.dataId);return _.mergeRealAndImagArrays(r,s)}return this.data.get(e).values}bufferSync(e){let t=this.readSync(e.dataId),n=t;if(e.dtype==="string")try{n=t.map(r=>k.decodeString(r))}catch(r){throw new Error("Failed to decode encoded string bytes into utf-8")}return ze(e.shape,e.dtype,n)}makeOutput(e,t,n){let r=this.write(e,t,n);return Ba().makeTensorFromDataId(r,t,n,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=k.now();return e(),{kernelMs:k.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){Ne([e],"where");let t=this.readSync(e.dataId);return Lae(e.shape,t)}dispose(){}floatPrecision(){return 32}epsilon(){return super.epsilon()}},x5=CT;x5.nextDataId=0;var ET={};_e(ET,{addImpl:()=>RT,bincountImpl:()=>v5,bincountReduceImpl:()=>_T,ceilImpl:()=>DT,concatImpl:()=>FT,equalImpl:()=>MT,expImpl:()=>PT,expm1Impl:()=>LT,floorImpl:()=>BT,gatherNdImpl:()=>WT,gatherV2Impl:()=>VT,greaterEqualImpl:()=>HT,greaterImpl:()=>UT,lessEqualImpl:()=>jT,lessImpl:()=>GT,linSpaceImpl:()=>qT,logImpl:()=>KT,maxImpl:()=>XT,maximumImpl:()=>ZT,minimumImpl:()=>YT,multiplyImpl:()=>w5,negImpl:()=>JT,notEqualImpl:()=>QT,prodImpl:()=>eN,rangeImpl:()=>tN,rsqrtImpl:()=>nN,simpleAbsImpl:()=>$T,sliceImpl:()=>rN,sparseFillEmptyRowsImpl:()=>sN,sparseReshapeImpl:()=>aN,sparseSegmentReductionImpl:()=>I5,squaredDifferenceImpl:()=>oN,stridedSliceImpl:()=>iN,stringNGramsImpl:()=>lN,stringSplitImpl:()=>uN,stringToHashBucketFastImpl:()=>cN,subImpl:()=>dN,tileImpl:()=>hN,topKImpl:()=>fN,transposeImpl:()=>k5,uniqueImpl:()=>mN});function $T(e){let t=new Float32Array(e.length);for(let n=0;n<e.length;++n)t[n]=Math.abs(e[n]);return t}var Bae=e=>{let{x:t}=e.inputs,n=e.backend;Ne(t,"abs");let r=new Float32Array(k.sizeFromShape(t.shape)),s=n.data.get(t.dataId).values;return r=$T(s),n.makeOutput(r,t.shape,"float32")},Wae={kernelName:Ac,backendName:"cpu",kernelFunc:Bae};function en(e){return(t,n,r,s,a)=>{let o=_.assertAndGetBroadcastShape(t,n),i=o.length,l=k.computeStrides(o),u=k.sizeFromShape(o),c=k.getTypedArrayFromDType(a,u),d=t.length,h=n.length,p=k.computeStrides(t),f=k.computeStrides(n),m=_.getBroadcastDims(t,o),g=_.getBroadcastDims(n,o);if(m.length+g.length===0)for(let y=0;y<c.length;++y)c[y]=e(r[y%r.length],s[y%s.length]);else for(let y=0;y<c.length;++y){let A=k.indexToLoc(y,i,l),x=A.slice(-d);m.forEach(w=>x[w]=0);let b=k.locToIndex(x,d,p),v=A.slice(-h);g.forEach(w=>v[w]=0);let I=k.locToIndex(v,h,f);c[y]=e(r[b],s[I])}return[c,o]}}function pr(e){let{inputs:t,backend:n}=e,{real:r,imag:s}=t,a=n.data.get(r.dataId).values,o=n.data.get(s.dataId).values,i=n.makeTensorInfo(r.shape,"complex64"),l=n.data.get(i.dataId);return l.complexTensorInfos={real:n.makeTensorInfo(r.shape,"float32",a),imag:n.makeTensorInfo(s.shape,"float32",o)},i}var Vae={kernelName:v1,backendName:"cpu",kernelFunc:pr};function km(e,t,n="float32"){if(n==="complex64"){let s=km(e,t,"float32"),a=km(e,t,"float32");return pr({inputs:{real:s,imag:a},backend:e})}let r=k.makeZerosTypedArray(k.sizeFromShape(t),n);return e.makeTensorInfo(t,n,r)}function Ls(e){let{inputs:t,backend:n}=e,{x:r}=t;return n.incRef(r.dataId),{dataId:r.dataId,shape:r.shape,dtype:r.dtype}}var Uae={kernelName:hl,backendName:"cpu",kernelFunc:Ls};function di(e){let{inputs:t,backend:n}=e,{input:r}=t,s=n.data.get(r.dataId).complexTensorInfos.real,a=n.data.get(s.dataId).values;return n.makeTensorInfo(s.shape,s.dtype,a)}var Hae={kernelName:V1,backendName:"cpu",kernelFunc:di};function Ja(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{dtype:a}=r;if(a==="complex64"){if(s.dtype==="complex64")return Ls({inputs:{x:s},backend:n});let o=km(n,s.shape,s.dtype),i=Ja({inputs:{x:s},backend:n,attrs:{dtype:"float32"}}),l=pr({inputs:{real:i,imag:o},backend:n});return n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}if(s.dtype==="complex64"){let o=di({inputs:{input:s},backend:n}),i=Ja({inputs:{x:o},backend:n,attrs:{dtype:a}});return n.disposeIntermediateTensorInfo(o),i}if(!k.hasEncodingLoss(s.dtype,a)){let o=Ls({inputs:{x:s},backend:n});return{dataId:o.dataId,shape:o.shape,dtype:a}}if(a==="int32"){let o=n.data.get(s.dataId).values,i=Int32Array.from(o);return n.makeTensorInfo(s.shape,"int32",i)}if(a==="bool"){let o=n.data.get(s.dataId).values,i=k.toTypedArray([0],s.dtype),[l,u]=en((c,d)=>c!==d?1:0)(s.shape,[],o,i,"bool");return n.makeTensorInfo(u,"bool",l)}throw new Error(`Error in Cast: failed to cast ${s.dtype} to ${a}`)}var Gae={kernelName:Qi,backendName:"cpu",kernelFunc:Ja};function xn(e,t,n,r){return n==null?({inputs:s,backend:a})=>{let{a:o,b:i}=s,l=a;Ne([o,i],e);let u=l.data.get(o.dataId).values,c=l.data.get(i.dataId).values,d=o.dtype==="string"?_.fromUint8ToStringArray(u):u,h=o.dtype==="string"?_.fromUint8ToStringArray(c):c,p=r||o.dtype,[f,m]=t(o.shape,i.shape,d,h,p);return l.makeTensorInfo(m,p,f)}:({inputs:s,backend:a})=>{let{a:o,b:i}=s,l=a;if(o.dtype==="complex64"||i.dtype==="complex64"){let u=Ja({inputs:{x:o},backend:l,attrs:{dtype:"complex64"}}),c=l.data.get(u.dataId),d=c.complexTensorInfos.real,h=c.complexTensorInfos.imag,p=l.data.get(d.dataId).values,f=l.data.get(h.dataId).values,m=Ja({inputs:{x:i},backend:l,attrs:{dtype:"complex64"}}),g=l.data.get(m.dataId),y=g.complexTensorInfos.real,A=g.complexTensorInfos.imag,x=l.data.get(y.dataId).values,b=l.data.get(A.dataId).values,[v,I,w]=n(o.shape,i.shape,p,f,x,b),S=l.makeTensorInfo(w,"float32",v),E=l.makeTensorInfo(w,"float32",I),D=pr({inputs:{real:S,imag:E},backend:l});return l.disposeIntermediateTensorInfo(u),l.disposeIntermediateTensorInfo(m),l.disposeIntermediateTensorInfo(S),l.disposeIntermediateTensorInfo(E),D}else{let u=l.data.get(o.dataId).values,c=l.data.get(i.dataId).values,d=r||o.dtype,[h,p]=t(o.shape,i.shape,u,c,d);return l.makeTensorInfo(p,d,h)}}}function b5(e){return(t,n,r,s,a,o)=>{let i=_.assertAndGetBroadcastShape(t,n),l=k.sizeFromShape(i),u=i.length,c=k.computeStrides(i),d=k.getTypedArrayFromDType("float32",l),h=k.getTypedArrayFromDType("float32",l),p=_.getBroadcastDims(t,i),f=_.getBroadcastDims(n,i),m=_.mergeRealAndImagArrays(r,s),g=_.mergeRealAndImagArrays(a,o),y=t.length,A=k.computeStrides(t),x=n.length,b=k.computeStrides(n);if(p.length+f.length===0)for(let v=0;v<d.length;v++){let I=v%m.length,w=v%g.length,S=e(m[I*2],m[I*2+1],g[w*2],g[w*2+1]);d[v]=S.real,h[v]=S.imag}else for(let v=0;v<d.length;v++){let I=k.indexToLoc(v,u,c),w=I.slice(-y);p.forEach(R=>w[R]=0);let S=k.locToIndex(w,y,A),E=I.slice(-x);f.forEach(R=>E[R]=0);let D=k.locToIndex(E,x,b),$=e(m[S*2],m[S*2+1],g[D*2],g[D*2+1]);d[v]=$.real,h[v]=$.imag}return[d,h,i]}}var RT=en((e,t)=>e+t),jae=b5((e,t,n,r)=>({real:e+n,imag:t+r})),Yd=xn(Ma,RT,jae),qae={kernelName:Ma,backendName:"cpu",kernelFunc:Yd};function v5(e,t,n,r,s){let a=k.sizeFromShape(r),o=k.makeZerosTypedArray(s,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>=s||(a>0?o[l]+=t[i]:o[l]+=1)}return o}function _T(e,t,n,r=!1){let s=e.shape[0],a=e.shape[1],o=ze([s,n],t.dtype);for(let i=0;i<s;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||(r?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 iu(e){return(t,n,r)=>{let s=k.getTypedArrayFromDType(n,t.length);for(let a=0;a<t.length;++a)s[a]=e(t[a],r);return s}}function At(e,t,n){return({inputs:r,attrs:s,backend:a})=>{let{x:o}=r;if(Ne(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=k.sizeFromShape(o.shape),c=n||o.dtype,d=k.getArrayFromDType(c,u);for(let h=0;h<u;++h)d[h]=t(l[h],s);return i.makeTensorInfo(o.shape,c,d)}}function lu(e,t,n){return({inputs:r,attrs:s,backend:a})=>{let{x:o}=r;if(Ne(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,s);return i.makeTensorInfo(o.shape,u,c)}}var DT=iu(e=>Math.ceil(e)),Kae=lu(No,DT),Xae={kernelName:No,backendName:"cpu",kernelFunc:Kae};function FT(e,t,n,r){let s=k.getArrayFromDType(n,k.sizeFromShape(t));if(r&&n!=="string"){let a=0;e.forEach(o=>{let i=k.sizeFromShape(o.shape);s.set(o.vals,a),a+=i})}else{let a=0;e.forEach(o=>{let i=n==="string"?_.fromUint8ToStringArray(o.vals):o.vals,l=0;for(let u=0;u<o.shape[0];++u){let c=u*t[1]+a;for(let d=0;d<o.shape[1];++d)s[c+d]=i[l++]}a+=o.shape[1]})}return s}var MT=en((e,t)=>e===t?1:0),OT=xn(il,MT,null,"bool"),Zae={kernelName:il,backendName:"cpu",kernelFunc:OT},PT=iu(e=>Math.exp(e)),zT=lu(Eo,PT),Yae={kernelName:Eo,backendName:"cpu",kernelFunc:zT},LT=iu(e=>Math.expm1(e)),Jae=lu(ll,LT),Qae={kernelName:ll,backendName:"cpu",kernelFunc:Jae},BT=iu(e=>Math.floor(e)),eoe=lu($o,BT),toe={kernelName:$o,backendName:"cpu",kernelFunc:eoe};function WT(e,t,n,r,s,a,o,i,l){let u=ze([r,a],n);for(let c=0;c<r;c++){let d=[],h=0;for(let p=0;p<s;p++){let f=e[c*s+p];h+=f*o[p],d.push(f)}if(h<0||h>=l/a)throw new Error(`Invalid indices: ${d} does not index into ${i}`);for(let p=0;p<a;p++)u.values[c*a+p]=t.get(...t.indexToLoc(h*a+p))}return u}function VT(e,t,n){let r=ze(n,e.dtype);for(let s=0;s<r.size;++s){let o=r.indexToLoc(s).slice(),i=o[0],l=o[2],u=t.locToIndex([i,l]);o[2]=t.values[u];let c=e.locToIndex(o);r.values[s]=e.values[c]}return r}var UT=en((e,t)=>e>t?1:0),noe=xn(dl,UT,null,"bool"),roe={kernelName:dl,backendName:"cpu",kernelFunc:noe},HT=en((e,t)=>e>=t?1:0),soe=xn(Ro,HT,null,"bool"),aoe={kernelName:Ro,backendName:"cpu",kernelFunc:soe},GT=en((e,t)=>e<t?1:0),ooe=xn(fl,GT,null,"bool"),ioe={kernelName:fl,backendName:"cpu",kernelFunc:ooe},jT=en((e,t)=>e<=t?1:0),loe=xn(ml,jT,null,"bool"),uoe={kernelName:ml,backendName:"cpu",kernelFunc:loe};function qT(e,t,n){let r=(t-e)/(n-1),s=k.makeZerosTypedArray(n,"float32");s[0]=e;for(let a=1;a<s.length;a++)s[a]=s[a-1]+r;return s}var KT=iu(e=>Math.log(e)),coe=lu(_o,KT),doe={kernelName:_o,backendName:"cpu",kernelFunc:coe};function XT(e,t,n,r){let s=k.getTypedArrayFromDType(r,k.sizeFromShape(n));for(let a=0;a<s.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)}s[a]=i}return s}var ZT=en((e,t)=>Math.max(e,t)),hoe=xn(Do,ZT),poe={kernelName:Do,backendName:"cpu",kernelFunc:hoe},YT=en((e,t)=>Math.min(e,t)),foe=xn(Fo,YT),moe={kernelName:Fo,backendName:"cpu",kernelFunc:foe},w5=en((e,t)=>e*t),goe=b5((e,t,n,r)=>({real:e*n-t*r,imag:e*r+t*n})),Im=xn(Mo,w5,goe),yoe={kernelName:Mo,backendName:"cpu",kernelFunc:Im};function JT(e,t,n){let r=k.createScalarValue(-1,n);return w5([],t,r,e,n)}function Aoe(e){let{inputs:t,backend:n}=e,{x:r}=t;Ne(r,"neg");let s=n.data.get(r.dataId).values,[a,o]=JT(s,r.shape,r.dtype);return n.makeTensorInfo(o,r.dtype,a)}var xoe={kernelName:Hc,backendName:"cpu",kernelFunc:Aoe},QT=en((e,t)=>e!==t?1:0),boe=xn(vl,QT,null,"bool"),voe={kernelName:vl,backendName:"cpu",kernelFunc:boe};function k5(e,t,n,r,s){let a=t.length,o=k.sizeFromShape(t),i=k.computeStrides(t),l=k.computeStrides(s),u=k.getTypedArrayFromDType(n,k.sizeFromShape(s));for(let c=0;c<o;++c){let d=k.indexToLoc(c,a,i),h=new Array(d.length);for(let f=0;f<h.length;f++)h[f]=d[r[f]];let p=k.locToIndex(h,a,l);u[p]=e[c]}return u}function Dr(e){let{inputs:t,attrs:n,backend:r}=e,{x:s}=t,{perm:a}=n;Ne(s,"transpose");let o=s.shape.length,i=new Array(o);for(let d=0;d<i.length;d++)i[d]=s.shape[a[d]];let l=r.data.get(s.dataId).values,u=k5(l,s.shape,s.dtype,a,i);return{dataId:r.write(u,i,s.dtype),shape:i,dtype:s.dtype}}var woe={kernelName:zl,backendName:"cpu",kernelFunc:Dr};function eN(e,t,n,r){let[s,a]=_.computeOutAndReduceShapes(e,r),o=Ur(t,"int32"),i=k.makeZerosTypedArray(k.sizeFromShape(s),o),l=k.sizeFromShape(a);for(let u=0;u<i.length;++u){let c=u*l,d=1;for(let h=0;h<l;++h)d*=n[c+h];i[u]=d}return{outVals:i,outShape:s,outDtype:o}}function koe(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r;Ne(s,"prod");let i=s.shape.length,l=k.parseAxisParam(a,s.shape),u=_.getAxesPermutation(l,i),c=l,d=s,h=[];u!=null&&(d=Dr({inputs:{x:s},backend:n,attrs:{perm:u}}),h.push(d),c=_.getInnerMostAxes(c.length,i));let p=n.data.get(d.dataId).values,{outVals:f,outShape:m,outDtype:g}=eN(d.shape,d.dtype,p,c),y=m;return o&&(y=_.expandShapeToKeepDim(m,l)),h.forEach(A=>n.disposeIntermediateTensorInfo(A)),n.makeTensorInfo(y,g,f)}var Ioe={kernelName:Zc,backendName:"cpu",kernelFunc:koe};function tN(e,t,n,r){let s=e===t,a=e<t&&n<0,o=t<e&&n>1;if(s||a||o)return k.makeZerosTypedArray(0,r);let i=Math.abs(Math.ceil((t-e)/n)),l=k.makeZerosTypedArray(i,r);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 nN=iu(e=>1/Math.sqrt(e)),Soe=lu(Oo,nN),Toe={kernelName:Oo,backendName:"cpu",kernelFunc:Soe};function rN(e,t,n,r,s){let a=Cn.isSliceContinous(r,t,n),o=k.sizeFromShape(n),i=k.computeStrides(r);if(a){let d=Cn.computeFlatOffset(t,i);return s==="string"?e.slice(d,d+o):e.subarray(d,d+o)}let l=s==="string"?_.fromUint8ToStringArray(e):e,u=ze(r,s,l),c=ze(n,s);for(let d=0;d<c.size;++d){let h=c.indexToLoc(d),p=h.map((f,m)=>f+t[m]);c.set(u.get(...p),...h)}return s==="string"?_.fromStringArrayToUint8(c.values):c.values}function hi(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{begin:a,size:o}=r;Ne(s,"slice");let[i,l]=Cn.parseSliceParams(s,a,o);Cn.assertParamsValid(s,i,l);let u=n.data.get(s.dataId).values,c=rN(u,i,l,s.shape,s.dtype);return n.makeTensorInfo(l,s.dtype,c)}var Noe={kernelName:nd,backendName:"cpu",kernelFunc:hi};function sN(e,t,n,r,s,a,o){let i=t[0],l=a[0],u=new Array(l),c=new Array(i),d=t[1];if(l===0){if(i!==0)throw new Error(`Received SparseTensor with denseShape[0] = 0 but
indices.shape[0] = ${i}`);let g=k.getArrayFromDType(n,0),y=k.getArrayFromDType(s,0);return[g,[0,d],y,u,c]}let h=!0,p=0,f=new Array(l).fill(0);for(let g=0;g<i;++g){let y=e[g*d];if(y<0)throw new Error(`indices(${g}, 0) is invalid: ${y} < 0`);if(y>=l)throw new Error(`indices(${g}, 0) is invalid: ${y} >= ${l}`);++f[y],h=h&&y>=p,p=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&&h){let g=e,y=r;for(let A=0;A<i;++A)c[A]=A;return[g,[i,d],y,u,c]}else{let g=f[l-1],y=k.getArrayFromDType(n,g*d),A=k.getArrayFromDType(s,g),x=new Array(l).fill(0);for(let b=0;b<i;++b){let v=e[b*d],I=x[v],w=(v===0?0:f[v-1])+I;x[v]++;for(let S=0;S<d;++S)y[w*d+S]=e[b*d+S];A[w]=r[b],c[b]=w}for(let b=0;b<l;++b)if(x[b]===0){let I=b===0?0:f[b-1];y[I*d+0]=b;for(let w=1;w<d;++w)y[I*d+w]=0;A[I]=o}return[y,[g,d],A,u,c]}}function aN(e,t,n,r,s){let a=k.sizeFromShape(r),o=t[0],i=s.length,l=[],u=1,c=-1;for(let g=0;g<i;++g){let y=s[g];if(y===-1){if(c!==-1)throw new Error(`only one output dimension may be -1, not both ${c} and ${g}`);c=g,l.push(1)}else{if(y<0)throw new Error(`size ${g} must be non-negative, not ${y}`);u*=y,l.push(y)}}if(c!==-1){if(u<=0)throw new Error("reshape cannot infer the missing input size for an empty tensor unless all specified input sizes are non-zero");let g=Math.trunc(a/u);if(u*g!==a)throw new Error(`Input to reshape is a SparseTensor with ${a}
dense values, but the requested shape requires a multiple of ${u}. inputShape=${r} outputShape= ${l}`);l[c]=g}let d=k.sizeFromShape(l);if(d!==a)throw new Error(`Input to reshape is a tensor with ${a} dense values, but the requested shape has ${d}. inputShape=${r} outputShape=${l}`);let h=r.length,p=[];if(h>0){p[h-1]=1;for(let g=h-2;g>=0;--g)p[g]=p[g+1]*r[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=k.getArrayFromDType(n,o*i);for(let g=0;g<o;++g){let y=0;for(let A=0;A<h;++A)y+=e[g*h+A]*p[A];for(let A=0;A<i;++A)m[g*i+A]=Math.trunc(y/f[A]),y%=f[A]}return[m,[o,i],l]}function I5(e,t,n,r,s,a=!1,o=0){let i=r.length;if(i!==s.length)throw new Error("segmentIds and indices should have same size.");let l=[t[0],e.length/t[0]],u=l[1],d=i>0?s[i-1]+1:0;if(d<0)throw new Error("segment ids must be >= 0");let h=t.slice();h[0]=d;let p=h.reduce((x,b)=>x*b,1),f=k.getArrayFromDType(n,p);if(i===0)return d>0&&f.fill(o),[f,h];if(d<=0)throw new Error("segment ids must be >= 0");let m=0,g=1,y=0,A=s[m];for(;;){let x=0;if(g<i){if(x=s[g],A===x){++g;continue}if(A>=x)throw new Error("segment ids are not increasing")}if(A<0||A>=d)throw new Error(`Segment id ${A} out of range [0, ${d}), possibly because segmentIds input is not sorted.`);A>y&&f.fill(o,y*u,A*u);for(let b=m;b<g;++b){let v=r[b];if(v<0||v>=l[0])throw new Error(`Bad: indices[${b}] == ${r[b]} out of range [0, ${l[0]})`);for(let I=0;I<u;I++)f[A*u+I]+=e[v*u+I]}if(a)for(let b=0;b<u;b++)f[A*u+b]/=g-m;if(m=g,++g,y=A+1,A=x,g>i)break}return y<d&&f.fill(o,y*u,d*u),[f,h]}var oN=en((e,t)=>{let n=e-t;return n*n}),Coe=xn(Po,oN),Eoe={kernelName:Po,backendName:"cpu",kernelFunc:Coe};function iN(e,t,n,r){let s=ze(e,t.dtype);for(let a=0;a<s.size;a++){let o=s.indexToLoc(a),i=new Array(o.length);for(let l=0;l<i.length;l++)i[l]=o[l]*n[l]+r[l];s.set(t.get(...i),...o)}return s}var $oe=class{constructor(e,t,n,r,s,a){this.separator=k.encodeString(e),this.nGramWidths=t,this.leftPad=k.encodeString(n),this.rightPad=k.encodeString(r),this.padWidth=s,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,r,s,a){for(let o=0;o<s;++o){let i=this.getPadWidth(a),l=Math.max(0,i-o),u=Math.max(0,i-(s-(o+1))),c=a-(l+u),d=t+(l>0?0:o-i),h=0;h+=l*this.leftPad.length;for(let y=0;y<c;++y)h+=e[d+y].length;h+=u*this.rightPad.length,h+=(l+u+c-1)*this.separator.length,n[r+o]=new Uint8Array(h);let f=n[r+o],m=0,g=y=>y.forEach(A=>f[m++]=A);for(let y=0;y<l;++y)g(this.leftPad),g(this.separator);for(let y=0;y<c-1;++y)g(e[d+y]),g(this.separator);if(c>0){g(e[d+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,r=t.length;if(r>0){let i=t[0];if(i!==0)throw new Error(`First split value must be 0, got ${i}`);for(let l=1;l<r;++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 s=r-1,a=k.getArrayFromDType("int32",r);if(n===0||r===0){let i=new Array(n);for(let l=0;l<=s;++l)a[l]=0;return[i,a]}a[0]=0;for(let i=1;i<=s;++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[s]);for(let i=0;i<s;++i){let l=t[i],u=a[i];if(this.nGramWidths.forEach(c=>{let d=t[i+1]-t[i],h=this.getNumNGrams(d,c);this.createNGrams(e,l,o,u,h,c),u+=h}),this.preserveShort&&u===a[i]){let c=t[i+1]-t[i];if(c===0)continue;let d=c+2*this.padWidth,h=1;this.createNGrams(e,l,o,u,h,d)}}return[o,a]}};function lN(e,t,n,r,s,a,o,i){return new $oe(n,r,s,a,o,i).compute(e,t)}function Roe(e,t,n){if(!e.length)return[];if(t.length===0){let a=new Array(e.length);for(let o=0;o<e.length;++o)a[o]=e.subarray(o,o+1);return a}if(t.length===1){let a=t[0],o=[],i=e.indexOf(a);for(;i!==-1;){let l=e.subarray(0,i);(!n||l.length!==0)&&o.push(l),e=e.subarray(i+1),i=e.indexOf(a)}return(!n||e.length!==0)&&o.push(e),o}let r=[],s=0;for(let a=0;a<e.length+1;a++)if(a===e.length||t.indexOf(e[a])!==-1){let o=e.subarray(s,a);(!n||o.length!==0)&&r.push(o),s=a+1}return r}function uN(e,t,n){let r=e.length,s=[],a=0,o=0,i=new Array(r);for(let h=0;h<r;++h){let p=Roe(e[h],t,n),f=p.length;i[h]=f,a+=f,o=Math.max(o,f),s.push(...p)}let l=k.getArrayFromDType("int32",a*2),u=new Array(a),c=[r,o],d=0;for(let h=0;h<r;++h)for(let p=0;p<i[h];++p)l[d*2]=h,l[d*2+1]=p,u[d]=s[d],++d;return[l,u,c]}function cN(e,t){let n=k.getArrayFromDType("int32",e.length);for(let r=0;r<e.length;++r)n[r]=k.fingerPrint64(e[r]).modulo(t).getLowBitsUnsigned();return n}var dN=en((e,t)=>e-t),_oe=b5((e,t,n,r)=>({real:e-n,imag:t-r})),S5=xn(zo,dN,_oe),Doe={kernelName:zo,backendName:"cpu",kernelFunc:S5};function hN(e,t){let n=new Array(e.rank);for(let s=0;s<n.length;s++)n[s]=e.shape[s]*t[s];let r=ze(n,e.dtype);for(let s=0;s<r.values.length;++s){let a=r.indexToLoc(s),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);r.values[s]=e.values[i]}return r}var Jd=(e,t)=>{let n=t.value-e.value;return n===0?e.index-t.index:n};function pN(e,t,n=0,r=e.length-1){for(;r>n;){if(r-n>600){let i=r-n+1,l=t-n+1,u=Math.log(i),c=.5*Math.exp(2*u/3),d=.5*Math.sqrt(u*c*(i-c)/i)*Math.sign(l-i/2),h=Math.max(n,Math.floor(t-l*c/i+d)),p=Math.min(r,Math.floor(t+(i-l)*c/i+d));pN(e,t,h,p)}let s=e[t],a=n,o=r;for(k.swap(e,n,t),Jd(e[r],s)>0&&k.swap(e,n,r);a<o;){for(k.swap(e,a,o),a++,o--;Jd(e[a],s)<0;)a=a+1;for(;Jd(e[o],s)>0;)o=o-1}Jd(e[n],s)===0?k.swap(e,n,o):(o=o+1,k.swap(e,o,r)),o<=t&&(n=o+1),t<=o&&(r=o-1)}}function fN(e,t,n,r,s){let a=t[t.length-1],[o,i]=[e.length/a,a],l=k.getTypedArrayFromDType(n,o*r),u=k.getTypedArrayFromDType("int32",o*r);for(let d=0;d<o;d++){let h=d*i,p=e.subarray(h,h+i),f=new Array(p.length);p.forEach((A,x)=>f[x]={value:A,index:x}),r<f.length&&(pN(f,r),f=f.slice(0,r)),s&&f.sort(Jd);let m=d*r,g=l.subarray(m,m+r),y=u.subarray(m,m+r);for(let A=0;A<r;A++)g[A]=f[A].value,y[A]=f[A].index}let c=t.slice();return c[c.length-1]=r,[ze(c,n,l),ze(c,"int32",u)]}function mN(e,t,n,r){let s=k.parseAxisParam(t,n)[0],a=[1,n[0],1];for(let f=0;f<s;f++)a[0]*=n[f];a[1]=n[s];for(let f=s+1;f<n.length;f++)a[2]*=n[f];let o={},i=new Int32Array(n[s]),l=new Jt(a,r,e),u=[],c=a[0]===1&&a[2]===1;for(let f=0;f<n[s];f++){let m;if(c)m=e[f].toString();else{let g=[];for(let y=0;y<a[0];y++)for(let A=0;A<a[2];A++)g.push(l.get(y,f,A));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 d=a.slice();d[1]=Object.keys(o).length;let h=new Jt(d,r);u.forEach((f,m)=>{for(let g=0;g<a[0];g++)for(let y=0;y<a[2];y++)h.set(l.get(g,f,y),g,m,y)});let p=n.slice();return p[s]=d[1],{outputValues:h.values,outputShape:p,indices:i}}var Foe="3.8.0";Cy("cpu",()=>new x5,1);var gN=At(_c,e=>e>=0?e:Math.exp(e)-1),Moe={kernelName:_c,backendName:"cpu",kernelFunc:gN};function yN(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{alpha:a}=r;Ne([s],"leakyRelu");let o=k.sizeFromShape(s.shape),i=n.data.get(s.dataId).values,l=k.getTypedArrayFromDType("float32",o);for(let u=0;u<i.length;u++)l[u]=i[u]<0?a*i[u]:i[u];return n.makeTensorInfo(s.shape,"float32",l)}var Ooe={kernelName:pl,backendName:"cpu",kernelFunc:yN},Poe=en((e,t)=>e<0?t*e:e);function AN(e){let{inputs:t,backend:n}=e,{x:r,alpha:s}=t;Ne([r,s],"prelu");let a=n.data.get(r.dataId).values,o=n.data.get(s.dataId).values,[i,l]=Poe(r.shape,s.shape,a,o,r.dtype);return n.makeTensorInfo(l,r.dtype,i)}var zoe={kernelName:Sl,backendName:"cpu",kernelFunc:AN},xN=At(Tl,e=>Math.max(0,e)),Loe={kernelName:Tl,backendName:"cpu",kernelFunc:xN},bN=At(Cl,e=>Math.min(Math.max(0,e),6)),Boe={kernelName:Cl,backendName:"cpu",kernelFunc:bN},vN=At(_l,e=>1/(1+Math.exp(-e))),Woe={kernelName:_l,backendName:"cpu",kernelFunc:vN};function T5(e,t,n,r,s){if(n==="linear")return Ls({inputs:{x:t},backend:e});if(n==="relu")return xN({inputs:{x:t},backend:e});if(n==="elu")return gN({inputs:{x:t},backend:e});if(n==="relu6")return bN({inputs:{x:t},backend:e});if(n==="prelu")return AN({inputs:{x:t,alpha:r},backend:e});if(n==="leakyrelu")return yN({inputs:{x:t},backend:e,attrs:{alpha:s}});if(n==="sigmoid")return vN({inputs:{x:t},backend:e});throw new Error(`Activation ${n} has not been implemented for the CPU backend.`)}function _t(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{shape:a}=r,o=k.sizeFromShape(s.shape),i=k.inferFromImplicitShape(a,o),l=k.sizeFromShape(i);k.assert(o===l,()=>`The new shape (${i}) has ${l} elements and the old shape (${s.shape}) has ${o} elements. The new shape and old shape must have the same number of elements.`),n.incRef(s.dataId);let u=n.data.get(s.dataId);if(u.complexTensorInfos!=null){let c=u.complexTensorInfos.real,d=u.complexTensorInfos.imag;c.shape=i,d.shape=i}return{dataId:s.dataId,shape:i,dtype:s.dtype}}var Voe={kernelName:Jc,backendName:"cpu",kernelFunc:_t};function wN(e){let{inputs:t,backend:n,attrs:r}=e,{a:s,b:a}=t,{transposeA:o,transposeB:i}=r;Ne([s,a],"matMul");let l=s.shape.length,u=a.shape.length,c=o?s.shape[l-2]:s.shape[l-1],d=i?a.shape[u-1]:a.shape[u-2],h=o?s.shape[l-1]:s.shape[l-2],p=i?a.shape[u-2]:a.shape[u-1],f=s.shape.slice(0,-2),m=a.shape.slice(0,-2),g=k.sizeFromShape(f),y=k.sizeFromShape(m),A=g===y||g===1||y===1;k.assert(l>=2&&u>=2&&A,()=>`Error in matMul: the input batch dimensions must either be the same or at least one input batch dimension must be 1. Got input batch dimensions of (${f}) and (${m}).`);let b=(g>y?s.shape.slice(0,-2):a.shape.slice(0,-2)).concat([h,p]);k.assert(c===d,()=>`Error in matMul: inner shapes (${c}) and (${d}) of Tensors with shapes ${s.shape} and ${a.shape} and transposeA=${o} and transposeB=${i} must match.`);let v=o?[g,c,h]:[g,h,c],I=i?[y,p,d]:[y,d,p],w=_t({inputs:{x:s},backend:n,attrs:{shape:v}}),S=_t({inputs:{x:a},backend:n,attrs:{shape:I}}),E=o?w.shape[1]:w.shape[2],D=o?w.shape[2]:w.shape[1],$=i?S.shape[1]:S.shape[2],R=Math.max(g,y),N=n.data.get(w.dataId).values,M=n.data.get(S.dataId).values,B=k.computeStrides(w.shape),q=k.computeStrides(S.shape),[X,J,ee]=o?[B[0],1,B[1]]:[B[0],B[1],1],[ae,se,oe]=i?[1,q[1],q[0]]:[q[1],1,q[0]],ne=D*$,ce=ze([R,D,$],w.dtype),he=ce.values,me=n.blockSize;for(let be=0;be<R;be++)for(let Ee=0;Ee<D;Ee+=me)for(let $e=0;$e<$;$e+=me)for(let Pe=0;Pe<E;Pe+=me){let je=Math.min(Ee+me,D),Be=Math.min($e+me,$),bt=Math.min(Pe+me,E);for(let pt=Ee;pt<je;pt++)for(let ft=$e;ft<Be;ft++){let dt=0;for(let xt=Pe;xt<bt;xt++){let Ye=Math.min(be,g-1)*X,Gn=Math.min(be,y-1)*oe,Bt=N[Ye+pt*J+xt*ee],or=M[xt*ae+ft*se+Gn];dt+=Bt*or}he[be*ne+(pt*$+ft)]+=dt}}return n.disposeIntermediateTensorInfo(w),n.disposeIntermediateTensorInfo(S),n.makeTensorInfo(b,ce.dtype,ce.values)}var Uoe={kernelName:Ji,backendName:"cpu",kernelFunc:wN};function Hoe(e){let{inputs:t,backend:n,attrs:r}=e,{a:s,b:a,bias:o,preluActivationWeights:i}=t,{transposeA:l,transposeB:u,activation:c,leakyreluAlpha:d}=r,h,p,f,m=[];h=wN({inputs:{a:s,b:a},attrs:{transposeA:l,transposeB:u},backend:n}),o&&(p=Yd({inputs:{a:h,b:o},backend:n}),m.push(h),h=p),c&&(f=T5(n,h,c,i,d),m.push(h),h=f);for(let y of m)n.disposeIntermediateTensorInfo(y);return h}var Goe={kernelName:Ll,backendName:"cpu",kernelFunc:Hoe},joe=At(xc,e=>Math.acos(e)),qoe={kernelName:xc,backendName:"cpu",kernelFunc:joe},Koe=At(bc,e=>Math.acosh(e)),Xoe={kernelName:bc,backendName:"cpu",kernelFunc:Koe};function Zoe(e){let{inputs:t,backend:n}=e,r=t;Ne(t,"addN");let s=r.map(i=>n.data.get(i.dataId).values),a=ze(r[0].shape,r[0].dtype),o=a.values;for(let i=0;i<r.length;i++){let l=s[i];for(let u=0;u<o.length;u++)o[u]+=l[u]}return n.makeTensorInfo(a.shape,a.dtype,a.values)}var Yoe={kernelName:Xi,backendName:"cpu",kernelFunc:Zoe};function Joe(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r;Ne(s,"all");let i=k.parseAxisParam(a,s.shape),l=i,u=_.getAxesPermutation(l,s.shape.length),c=s;u!=null&&(c=Dr({inputs:{x:s},backend:n,attrs:{perm:u}}),l=_.getInnerMostAxes(l.length,s.shape.length)),_.assertAxesAreInnerMostDims("all",l,c.shape.length);let[d,h]=_.computeOutAndReduceShapes(c.shape,l),p=k.sizeFromShape(h),f=k.makeZerosTypedArray(k.sizeFromShape(d),c.dtype),m=n.data.get(c.dataId).values;for(let y=0;y<f.length;++y){let A=y*p,x=m[A];for(let b=0;b<p;++b){let v=m[A+b];x=x&&v}f[y]=x}u!=null&&n.disposeIntermediateTensorInfo(c);let g=n.makeTensorInfo(d,c.dtype,f);if(o){let y=_.expandShapeToKeepDim(d,i),A=_t({inputs:{x:g},backend:n,attrs:{shape:y}});return n.disposeIntermediateTensorInfo(g),A}return g}var Qoe={kernelName:vc,backendName:"cpu",kernelFunc:Joe};function eie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r;Ne(s,"any");let i=k.parseAxisParam(a,s.shape),l=i,u=_.getAxesPermutation(l,s.shape.length),c=s;u!=null&&(c=Dr({inputs:{x:s},backend:n,attrs:{perm:u}}),l=_.getInnerMostAxes(l.length,s.shape.length)),_.assertAxesAreInnerMostDims("any",l,c.shape.length);let[d,h]=_.computeOutAndReduceShapes(c.shape,l),p=k.sizeFromShape(h),f=k.makeZerosTypedArray(k.sizeFromShape(d),c.dtype),m=n.data.get(c.dataId).values;for(let y=0;y<f.length;++y){let A=y*p,x=m[A];for(let b=0;b<p;++b){let v=m[A+b];x=x||v}f[y]=x}u!=null&&n.disposeIntermediateTensorInfo(c);let g=n.makeTensorInfo(d,c.dtype,f);if(o){let y=_.expandShapeToKeepDim(d,i),A=_t({inputs:{x:g},backend:n,attrs:{shape:y}});return n.disposeIntermediateTensorInfo(g),A}return g}var tie={kernelName:wc,backendName:"cpu",kernelFunc:eie};function nie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a}=r;Ne(s,"argMax");let o=k.parseAxisParam(a,s.shape),i=_.getAxesPermutation(o,s.shape.length),l=s,u=[];i!=null&&(l=Dr({inputs:{x:s},backend:n,attrs:{perm:i}}),u.push(l),o=_.getInnerMostAxes(o.length,l.shape.length)),o=[o[0]],_.assertAxesAreInnerMostDims("argMax",o,l.shape.length);let[c,d]=_.computeOutAndReduceShapes(l.shape,o),h=k.sizeFromShape(c),p=k.makeZerosTypedArray(h,"int32"),f=k.sizeFromShape(d),m=n.data.get(l.dataId).values;for(let g=0;g<p.length;++g){let y=g*f,A=m[y],x=0;for(let b=0;b<f;++b){let v=m[y+b];v>A&&(A=v,x=b)}p[g]=x}return u.forEach(g=>n.disposeIntermediateTensorInfo(g)),n.makeTensorInfo(c,"int32",p)}var rie={kernelName:Zi,backendName:"cpu",kernelFunc:nie};function sie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a}=r;Ne(s,"argMin");let o=k.parseAxisParam(a,s.shape),i=_.getAxesPermutation(o,s.shape.length),l=s,u=[];i!=null&&(l=Dr({inputs:{x:s},backend:n,attrs:{perm:i}}),u.push(l),o=_.getInnerMostAxes(o.length,l.shape.length)),o=[o[0]],_.assertAxesAreInnerMostDims("argMin",o,l.shape.length);let[c,d]=_.computeOutAndReduceShapes(l.shape,o),h=k.sizeFromShape(c),p=k.makeZerosTypedArray(h,"int32"),f=k.sizeFromShape(d),m=n.data.get(l.dataId).values;for(let g=0;g<p.length;++g){let y=g*f,A=m[y],x=0;for(let b=0;b<f;++b){let v=m[y+b];v<A&&(A=v,x=b)}p[g]=x}return u.forEach(g=>n.disposeIntermediateTensorInfo(g)),n.makeTensorInfo(c,"int32",p)}var aie={kernelName:qp,backendName:"cpu",kernelFunc:sie},oie=At(kc,e=>Math.asin(e)),iie={kernelName:kc,backendName:"cpu",kernelFunc:oie},lie=At(Ic,e=>Math.asinh(e)),uie={kernelName:Ic,backendName:"cpu",kernelFunc:lie},cie=At(Sc,e=>Math.atan(e)),die={kernelName:Sc,backendName:"cpu",kernelFunc:cie},hie=en((e,t)=>Math.atan2(e,t)),pie=xn(Nc,hie),fie={kernelName:Nc,backendName:"cpu",kernelFunc:pie},mie=At(Tc,e=>Math.atanh(e)),gie={kernelName:Tc,backendName:"cpu",kernelFunc:mie};function N5(e,t,n,r,s,a){let o=s.strideHeight,i=s.strideWidth,l=s.dilationHeight,u=s.dilationWidth,c=s.effectiveFilterHeight,d=s.effectiveFilterWidth,h=s.padInfo.top,p=s.padInfo.left,f=a==="max"?Number.NEGATIVE_INFINITY:Number.POSITIVE_INFINITY,m=ze(s.outShape,n),g=m.values,y=s.outShape[1]*s.outShape[2]*s.outShape[3],A=s.outShape[2]*s.outShape[3],x=s.outShape[3];for(let b=0;b<s.batchSize;++b){let v=b*y,I=b*r[0];for(let w=0;w<s.inChannels;++w)for(let S=0;S<s.outHeight;++S){let E=S*o-h,D=Math.max(0,E),$=Math.min(s.inHeight,c+E),R=v+S*A;for(let N=0;N<s.outWidth;++N){let M=N*i-p,B=Math.max(0,M),q=Math.min(s.inWidth,d+M),X=f,J=0,ee=0;for(let se=D;se<$;se+=l){let oe=I+se*r[1];for(let ne=B;ne<q;ne+=u){let ce=oe+ne*r[2],he=e[ce+w];a==="max"&&he>X?X=he:a==="avg"&&(J+=he,ee++)}if(isNaN(X))break}let ae=R+N*x+w;g[ae]=a==="avg"?J/ee:X}}}return m}function kN(e,t,n,r,s=!1,a=!1){let o=ze(r.outShape,"int32"),i=r.strideHeight,l=r.strideWidth,u=r.dilationHeight,c=r.dilationWidth,d=r.effectiveFilterHeight,h=r.effectiveFilterWidth,p=r.padInfo.top,f=r.padInfo.left,m=ze(t,n,e);for(let g=0;g<r.batchSize;++g)for(let y=0;y<r.inChannels;++y)for(let A=0;A<r.outHeight;++A){let x=A*i-p,b=x;for(;b<0;)b+=u;let v=Math.min(r.inHeight,d+x);for(let I=0;I<r.outWidth;++I){let w=I*l-f,S=w;for(;S<0;)S+=c;let E=Math.min(r.inWidth,h+w),D=Number.NEGATIVE_INFINITY,$=-1;for(let R=b;R<v;R+=u){let N=R-x;for(let M=S;M<E;M+=c){let B=M-w,q=m.get(g,R,M,y);q>D&&(D=q,s?$=a?((g*r.inHeight+R)*r.inWidth+M)*r.inChannels+y:(R*r.inWidth+M)*r.inChannels+y:$=N*h+B)}}o.set($,g,A,I,y)}}return o}function IN(e,t,n,r,s,a){let o=s.strideDepth,i=s.strideHeight,l=s.strideWidth,u=s.dilationDepth,c=s.dilationHeight,d=s.dilationWidth,h=s.effectiveFilterDepth,p=s.effectiveFilterHeight,f=s.effectiveFilterWidth,m=s.padInfo.front,g=s.padInfo.top,y=s.padInfo.left,A=a==="max"?Number.NEGATIVE_INFINITY:Number.POSITIVE_INFINITY,x=ze(s.outShape,n),b=x.values,v=s.outShape[1]*s.outShape[2]*s.outShape[3]*s.outShape[4],I=s.outShape[2]*s.outShape[3]*s.outShape[4],w=s.outShape[3]*s.outShape[4],S=s.outShape[4];for(let E=0;E<s.batchSize;++E){let D=E*v,$=E*r[0];for(let R=0;R<s.inChannels;++R)for(let N=0;N<s.outDepth;++N){let M=N*o-m,B=M;for(;B<0;)B+=u;let q=Math.min(s.inDepth,h+M),X=D+N*I;for(let J=0;J<s.outHeight;++J){let ee=J*i-g,ae=ee;for(;ae<0;)ae+=c;let se=Math.min(s.inHeight,p+ee),oe=X+J*w;for(let ne=0;ne<s.outWidth;++ne){let ce=ne*l-y,he=ce;for(;he<0;)he+=d;let me=Math.min(s.inWidth,f+ce),be=oe+ne*S,Ee=A,$e=0,Pe=0;for(let Be=B;Be<q;Be+=u){let bt=$+Be*r[1];for(let pt=ae;pt<se;pt+=c){let ft=bt+pt*r[2];for(let dt=he;dt<me;dt+=d){let xt=ft+dt*r[3],Ye=e[xt+R];if(a==="max"&&Ye>Ee?Ee=Ye:a==="avg"&&($e+=Ye,Pe++),isNaN(Ee))break}if(isNaN(Ee))break}if(isNaN(Ee))break}let je=be+R;b[je]=a==="avg"?$e/Pe:Ee}}}}return x}function yie(e,t){let n=ze(t.outShape,"int32"),r=t.strideDepth,s=t.strideHeight,a=t.strideWidth,o=t.dilationDepth,i=t.dilationHeight,l=t.dilationWidth,u=t.effectiveFilterDepth,c=t.effectiveFilterHeight,d=t.effectiveFilterWidth,h=t.padInfo.front,p=t.padInfo.top,f=t.padInfo.left;for(let m=0;m<t.batchSize;++m)for(let g=0;g<t.inChannels;++g)for(let y=0;y<t.outDepth;++y){let A=y*r-h,x=A;for(;x<0;)x+=o;let b=Math.min(t.inDepth,u+A);for(let v=0;v<t.outHeight;++v){let I=v*s-p,w=I;for(;w<0;)w+=i;let S=Math.min(t.inHeight,c+I);for(let E=0;E<t.outWidth;++E){let D=E*a-f,$=D;for(;$<0;)$+=l;let R=Math.min(t.inWidth,d+D),N=Number.NEGATIVE_INFINITY,M=-1;for(let B=x;B<b;B+=o){let q=B-A;for(let X=w;X<S;X+=i){let J=X-I;for(let ee=$;ee<R;ee+=l){let ae=ee-D,se=e.get(m,B,X,ee,g);se>=N&&(N=se,M=q*c*d+J*c+ae)}}}n.set(M,m,y,v,E,g)}}}return n}function Aie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t;Ne(s,"avgPool");let{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=r,u=1;k.assert(_.eitherStridesOrDilationsAreOne(o,u),()=>`Error in avgPool: Either strides or dilations must be 1. Got strides ${o} and dilations '${u}'`);let c=_.computePool2DInfo(s.shape,a,o,u,i,l),d;if(c.filterWidth===1&&c.filterHeight===1&&k.arraysEqual(c.inShape,c.outShape))d=Ls({inputs:{x:s},backend:n});else{let h=n.data.get(s.dataId).values,p=k.computeStrides(s.shape),f=N5(h,s.shape,s.dtype,p,c,"avg");d=n.makeTensorInfo(c.outShape,s.dtype,f.values)}return d}var xie={kernelName:Yi,backendName:"cpu",kernelFunc:Aie};function bie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l,dataFormat:u}=r;Ne(s,"avgPool3d");let c=_.computePool3DInfo(s.shape,a,o,1,i,l,u),d=n.data.get(s.dataId).values,h=IN(d,s.shape,s.dtype,k.computeStrides(s.shape),c,"avg");return n.makeTensorInfo(h.shape,"float32",h.values)}var vie={kernelName:Kp,backendName:"cpu",kernelFunc:bie};function wie(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,input:a}=t,{filterSize:o,strides:i,pad:l,dimRoundingMode:u}=r;Ne([s,a],"avgPool3DGrad");let c=_.computePool3DInfo(a.shape,o,i,1,l,u),d=c.strideDepth,h=c.strideHeight,p=c.strideWidth,f=c.filterDepth,m=c.filterHeight,g=c.filterWidth,y=c.dilationDepth,A=c.dilationHeight,x=c.dilationWidth,b=c.effectiveFilterDepth,v=c.effectiveFilterHeight,I=c.effectiveFilterWidth,w=b-1-c.padInfo.front,S=I-1-c.padInfo.left,E=v-1-c.padInfo.top,D=ze(a.shape,"float32"),$=1/(f*m*g),R=n.bufferSync(s);for(let N=0;N<c.batchSize;++N)for(let M=0;M<c.inChannels;++M)for(let B=0;B<c.inDepth;++B)for(let q=0;q<c.inHeight;++q)for(let X=0;X<c.inWidth;++X){let J=B-w,ee=q-E,ae=X-S,se=0;for(let oe=0;oe<b;oe+=y){let ne=(J+oe)/d;if(!(ne<0||ne>=c.outDepth||Math.floor(ne)!==ne))for(let ce=0;ce<v;ce+=A){let he=(ee+ce)/h;if(!(he<0||he>=c.outHeight||Math.floor(he)!==he))for(let me=0;me<I;me+=x){let be=(ae+me)/p;if(be<0||be>=c.outWidth||Math.floor(be)!==be)continue;se+=R.get(N,ne,he,be,M)}}}D.set(se*$,N,B,q,X,M)}return n.makeTensorInfo(D.shape,D.dtype,D.values)}var kie={kernelName:x1,backendName:"cpu",kernelFunc:wie};function Iie(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,input:a}=t,o=a;Ne([s,a],"avgPoolGrad");let{filterSize:i,strides:l,pad:u}=r,c=_.computePool2DInfo(o.shape,i,l,1,u),d=c.strideHeight,h=c.strideWidth,p=c.filterHeight,f=c.filterWidth,m=c.dilationHeight,g=c.dilationWidth,y=c.effectiveFilterHeight,A=c.effectiveFilterWidth,x=A-1-c.padInfo.left,b=y-1-c.padInfo.top,v=ze(o.shape,"float32"),I=1/(p*f),w=n.data.get(s.dataId).values,S=ze(s.shape,"float32",w);for(let E=0;E<c.batchSize;++E)for(let D=0;D<c.inChannels;++D)for(let $=0;$<c.inHeight;++$)for(let R=0;R<c.inWidth;++R){let N=$-b,M=R-x,B=0;for(let q=0;q<y;q+=m){let X=(N+q)/d;if(!(X<0||X>=c.outHeight||Math.floor(X)!==X))for(let J=0;J<A;J+=g){let ee=(M+J)/h;if(ee<0||ee>=c.outWidth||Math.floor(ee)!==ee)continue;B+=S.get(E,X,ee,D)}}v.set(B*I,E,$,R,D)}return n.makeTensorInfo(v.shape,v.dtype,v.values)}var Sie={kernelName:A1,backendName:"cpu",kernelFunc:Iie};function Tie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,scale:a,offset:o,mean:i,variance:l}=t;k.assert(i.shape.length===l.shape.length,()=>"Batch normalization gradient requires mean and variance to have equal ranks."),k.assert(o==null||i.shape.length===o.shape.length,()=>"Batch normalization gradient requires mean and offset to have equal ranks."),k.assert(a==null||i.shape.length===a.shape.length,()=>"Batch normalization gradient requires mean and scale to have equal ranks."),Ne([s,i,l,a,o],"batchNorm");let{varianceEpsilon:u}=r;u==null&&(u=.001);let c=n.data.get(s.dataId).values,d=n.data.get(i.dataId).values,h=n.data.get(l.dataId).values,p=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=p.length,A=h.length,x=d.length,b=0,v=0,I=0,w=0;for(let S=0;S<c.length;++S)m[S]=f[b++]+(c[S]-d[v++])*p[I++]/Math.sqrt(h[w++]+u),b>=g&&(b=0),v>=x&&(v=0),I>=y&&(I=0),w>=A&&(w=0);return n.makeTensorInfo(s.shape,s.dtype,m)}var Nie={kernelName:cl,backendName:"cpu",kernelFunc:Tie};function Cie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{blockShape:a,crops:o}=r;Ne([s],"batchToSpaceND");let i=a.reduce((y,A)=>y*A),l=_.getReshaped(s.shape,a,i),u=_.getPermuted(l.length,a.length),c=_.getReshapedPermuted(s.shape,a,i),d=_.getSliceBeginCoords(o,a.length),h=_.getSliceSize(c,o,a.length),p=_t({inputs:{x:s},backend:n,attrs:{shape:l}}),f=Dr({inputs:{x:p},backend:n,attrs:{perm:u}}),m=_t({inputs:{x:f},backend:n,attrs:{shape:c}}),g=hi({inputs:{x:m},backend:n,attrs:{begin:d,size:h}});return n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(f),n.disposeIntermediateTensorInfo(m),g}var Eie={kernelName:Cc,backendName:"cpu",kernelFunc:Cie};function $ie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,weights:a}=t,{size:o}=r,i=n.data.get(s.dataId).values,l=n.data.get(a.dataId).values,u=v5(i,l,a.dtype,a.shape,o);return n.makeTensorInfo([o],a.dtype,u)}var Rie={kernelName:b1,backendName:"cpu",kernelFunc:$ie},_ie=At(Co,(e,t)=>{let n=t;return e>n.clipValueMax?n.clipValueMax:e<n.clipValueMin?n.clipValueMin:e}),Die={kernelName:Co,backendName:"cpu",kernelFunc:_ie},Fie=e=>{let{x:t}=e.inputs,n=e.backend,r=new Float32Array(k.sizeFromShape(t.shape)),s=n.data.get(t.dataId),a=s.complexTensorInfos.real,o=s.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],d=l[u];r[u]=Math.hypot(c,d)}return n.makeOutput(r,t.shape,"float32")},Mie={kernelName:Xp,backendName:"cpu",kernelFunc:Fie};function uu(e){let{inputs:t,backend:n}=e,{input:r}=t,s=n.data.get(r.dataId).complexTensorInfos.imag,a=n.data.get(s.dataId).values;return n.makeTensorInfo(s.shape,s.dtype,a)}var Oie={kernelName:M1,backendName:"cpu",kernelFunc:uu};function cu(e){let{inputs:t,backend:n,attrs:r}=e,{axis:s}=r,a=k.parseAxisParam(s,t[0].shape)[0],o=_.computeOutShape(t.map(m=>m.shape),a);if(k.sizeFromShape(o)===0)return n.makeTensorInfo(o,t[0].dtype,[]);let i=t.filter(m=>k.sizeFromShape(m.shape)>0);if(i.length===1)return Ls({inputs:{x:i[0]},backend:n});let l=i.map(m=>m.shape);if(_.assertParamsConsistent(l,a),i[0].dtype==="complex64"){let m=i.map(b=>di({inputs:{input:b},backend:n})),g=i.map(b=>uu({inputs:{input:b},backend:n})),y=cu({inputs:m,backend:n,attrs:{axis:a}}),A=cu({inputs:g,backend:n,attrs:{axis:a}}),x=pr({inputs:{real:y,imag:A},backend:n});return m.forEach(b=>n.disposeIntermediateTensorInfo(b)),g.forEach(b=>n.disposeIntermediateTensorInfo(b)),n.disposeIntermediateTensorInfo(y),n.disposeIntermediateTensorInfo(A),x}let u=i.map(m=>{let g=k.sizeFromShape(m.shape.slice(a));return _t({inputs:{x:m},backend:n,attrs:{shape:[-1,g]}})}),c=u.map(m=>({vals:n.data.get(m.dataId).values,shape:m.shape}));o=_.computeOutShape(u.map(m=>m.shape),1);let d=u[0].shape[0]===1,h=FT(c,o,t[0].dtype,d),p=_.computeOutShape(i.map(m=>m.shape),a),f=n.makeTensorInfo(p,t[0].dtype,h);return u.forEach(m=>n.disposeIntermediateTensorInfo(m)),f}var Pie={kernelName:Ec,backendName:"cpu",kernelFunc:cu};function SN(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a}=t,{strides:o,pad:i,dataFormat:l,dilations:u,dimRoundingMode:c}=r;Ne([s,a],"conv2d");let d=_.convertConv2DDataFormat(l),h=_.computeConv2DInfo(s.shape,a.shape,o,u,i,c,!1,d),p=h.filterHeight,f=h.filterWidth,m=h.dilationHeight,g=h.dilationWidth,y=h.padInfo.left,A=h.padInfo.top,x=h.dataFormat==="channelsLast",b=new Jt(h.outShape,s.dtype),v=k.computeStrides(s.shape),I=k.computeStrides(a.shape),w=v[0],S=x?v[1]:v[2],E=x?v[2]:1,D=x?1:v[1],$=b.strides[0],R=x?b.strides[1]:b.strides[2],N=x?b.strides[2]:1,M=x?1:b.strides[1],B=n.data.get(s.dataId).values,q=n.data.get(a.dataId).values,X=b.values;for(let J=0;J<h.batchSize;++J){let ee=J*w,ae=J*$;for(let se=0;se<h.outHeight;++se){let oe=ae+se*R,ne=se*h.strideHeight-A;for(let ce=0;ce<p;++ce){let he=ne+ce*m;if(he<0||he>=h.inHeight)continue;let me=ce*I[0],be=ee+he*S;for(let Ee=0;Ee<h.outWidth;++Ee){let $e=oe+Ee*N,Pe=Ee*h.strideWidth-y;for(let je=0;je<f;++je){let Be=Pe+je*g;if(Be<0||Be>=h.inWidth)continue;let bt=me+je*I[1],pt=be+Be*E,ft=bt;for(let dt=0;dt<h.inChannels;++dt){let xt=B[pt+dt*D];for(let Ye=0;Ye<h.outChannels;++Ye)X[$e+Ye*M]+=xt*q[ft+Ye];ft+=h.outChannels}}}}}}return n.makeTensorInfo(b.shape,b.dtype,X)}var zie={kernelName:el,backendName:"cpu",kernelFunc:SN};function Lie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,dy:a}=t,{strides:o,pad:i,dataFormat:l,dimRoundingMode:u,filterShape:c}=r;Ne([s,a],"conv2dBackpropFilter");let d=_.convertConv2DDataFormat(l),h=_.computeConv2DInfo(s.shape,c,o,1,i,u,!1,d),{strideHeight:p,strideWidth:f,filterHeight:m,filterWidth:g}=h,y=h.dataFormat==="channelsLast",A=new Jt(h.filterShape,"float32"),x=h.padInfo.left,b=h.padInfo.top,v=n.data.get(s.dataId).values,I=n.data.get(a.dataId).values,w=new Jt(s.shape,s.dtype,v),S=new Jt(a.shape,a.dtype,I);for(let E=0;E<m;++E){let D=Math.max(0,Math.ceil((b-E)/p)),$=Math.min(h.outHeight,(h.inHeight+b-E)/p);for(let R=0;R<g;++R){let N=Math.max(0,Math.ceil((x-R)/f)),M=Math.min(h.outWidth,(h.inWidth+x-R)/f);for(let B=0;B<h.inChannels;++B)for(let q=0;q<h.outChannels;++q){let X=0;for(let J=0;J<h.batchSize;++J)for(let ee=D;ee<$;++ee){let ae=E+ee*p-b;for(let se=N;se<M;++se){let oe=R+se*f-x;y?X+=w.get(J,ae,oe,B)*S.get(J,ee,se,q):X+=w.get(J,B,ae,oe)*S.get(J,q,ee,se)}}A.set(X,E,R,B,q)}}}return n.makeTensorInfo(A.shape,A.dtype,A.values)}var Bie={kernelName:w1,backendName:"cpu",kernelFunc:Lie};function Wie(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,filter:a}=t,{inputShape:o,strides:i,pad:l,dataFormat:u,dimRoundingMode:c}=r;Ne([s,a],"conv2dBackpropInput");let d=k.computeStrides(a.shape),h=k.computeStrides(s.shape),p=_.convertConv2DDataFormat(u),f=_.computeConv2DInfo(o,a.shape,i,1,l,c,!1,p),m=new Jt(f.inShape,"float32"),g=m.values,y=n.data.get(s.dataId).values,A=n.data.get(a.dataId).values,[x,b,v]=d,{batchSize:I,filterHeight:w,filterWidth:S,inChannels:E,inHeight:D,inWidth:$,outChannels:R,outHeight:N,outWidth:M,strideHeight:B,strideWidth:q}=f;p=f.dataFormat;let X=w-1-f.padInfo.top,J=S-1-f.padInfo.left,ee=p==="channelsLast",ae=m.strides[0],se=ee?m.strides[1]:m.strides[2],oe=ee?m.strides[2]:1,ne=ee?1:m.strides[1],ce=h[0],he=ee?h[1]:h[2],me=ee?h[2]:1,be=ee?1:h[1];for(let Ee=0;Ee<I;++Ee)for(let $e=0;$e<E;++$e)for(let Pe=0;Pe<D;++Pe){let je=Pe-X,Be=Math.max(0,Math.ceil(je/B)),bt=Math.min(N,(w+je)/B);for(let pt=0;pt<$;++pt){let ft=pt-J,dt=Math.max(0,Math.ceil(ft/q)),xt=Math.min(M,(S+ft)/q),Ye=0;for(let Bt=Be;Bt<bt;++Bt){let or=Bt*B-je;for(let bn=dt;bn<xt;++bn){let zr=bn*q-ft,Rn=ce*Ee+he*Bt+me*bn,Ar=x*(w-1-or)+b*(S-1-zr)+v*$e;for(let xr=0;xr<R;++xr){let vn=y[Rn+be*xr],br=A[Ar+xr];Ye+=vn*br}}}let Gn=ae*Ee+se*Pe+oe*pt+ne*$e;g[Gn]=Ye}}return n.makeTensorInfo(m.shape,m.dtype,m.values)}var Vie={kernelName:tl,backendName:"cpu",kernelFunc:Wie};function Uie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a}=t,{strides:o,pad:i,dilations:l}=r;Ne([s,a],"conv3d");let u=_.computeConv3DInfo(s.shape,a.shape,o,l,i),{filterDepth:c,filterHeight:d,filterWidth:h,dilationDepth:p,dilationHeight:f,dilationWidth:m,padInfo:g}=u,y=g.front,A=g.left,x=g.top,b=new Jt(u.outShape,s.dtype),v=n.data.get(s.dataId).values,I=n.data.get(a.dataId).values,w=b.values,S=k.computeStrides(s.shape),E=k.computeStrides(a.shape);for(let D=0;D<u.batchSize;++D){let $=D*S[0],R=D*b.strides[0];for(let N=0;N<u.outDepth;++N){let M=R+N*b.strides[1],B=N*u.strideDepth-y;for(let q=0;q<c;++q){let X=B+q*p;if(X<0||X>=u.inDepth)continue;let J=q*E[0],ee=$+X*S[1];for(let ae=0;ae<u.outHeight;++ae){let se=M+ae*b.strides[2],oe=ae*u.strideHeight-x;for(let ne=0;ne<d;++ne){let ce=oe+ne*f;if(ce<0||ce>=u.inHeight)continue;let he=J+ne*E[1],me=ee+ce*S[2];for(let be=0;be<u.outWidth;++be){let Ee=se+be*u.outChannels,$e=be*u.strideWidth-A;for(let Pe=0;Pe<h;++Pe){let je=$e+Pe*m;if(je<0||je>=u.inWidth)continue;let Be=he+Pe*E[2],bt=me+je*u.inChannels,pt=Be;for(let ft=0;ft<u.inChannels;++ft){let dt=v[bt+ft];for(let xt=0;xt<u.outChannels;++xt)w[Ee+xt]+=dt*I[pt+xt];pt+=u.outChannels}}}}}}}}return n.makeTensorInfo(b.shape,b.dtype,b.values)}var Hie={kernelName:Zp,backendName:"cpu",kernelFunc:Uie};function Gie(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,dy:a}=t,{strides:o,pad:i,filterShape:l}=r;Ne([s,a],"conv3dBackpropFilterV2");let u=k.computeStrides(s.shape),c=k.computeStrides(a.shape),d=_.computeConv3DInfo(s.shape,l,o,1,i),h=d.strideDepth,p=d.strideHeight,f=d.strideWidth,m=d.filterDepth,g=d.filterHeight,y=d.filterWidth,A=new Jt(d.filterShape,"float32"),x=A.values,[b,v,I,w]=A.strides,S=n.data.get(a.dataId).values,[E,D,$,R]=c,N=n.data.get(s.dataId).values,[M,B,q,X]=u,J=d.padInfo.front,ee=d.padInfo.left,ae=d.padInfo.top;for(let se=0;se<m;++se){let oe=Math.max(0,Math.ceil((J-se)/h)),ne=Math.min(d.outDepth,(d.inDepth+J-se)/h),ce=se*b;for(let he=0;he<g;++he){let me=Math.max(0,Math.ceil((ae-he)/p)),be=Math.min(d.outHeight,(d.inHeight+ae-he)/p),Ee=he*v+ce;for(let $e=0;$e<y;++$e){let Pe=Math.max(0,Math.ceil((ee-$e)/f)),je=Math.min(d.outWidth,(d.inWidth+ee-$e)/f),Be=$e*I+Ee;for(let bt=0;bt<d.inChannels;++bt){let pt=bt*w+Be;for(let ft=0;ft<d.outChannels;++ft){let dt=0;for(let xt=0;xt<d.batchSize;++xt){let Ye=xt*M,Gn=xt*E;for(let Bt=oe;Bt<ne;++Bt){let bn=(se+Bt*h-J)*B+Ye,zr=Bt*D+Gn;for(let Rn=me;Rn<be;++Rn){let xr=(he+Rn*p-ae)*q+bn,vn=Rn*$+zr;for(let br=Pe;br<je;++br){let ir=($e+br*f-ee)*X+xr,bs=br*R+vn;dt+=N[ir+bt]*S[bs+ft]}}}}x[pt+ft]=dt}}}}}return n.makeTensorInfo(A.shape,A.dtype,A.values)}var jie={kernelName:k1,backendName:"cpu",kernelFunc:Gie};function qie(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,filter:a}=t,{pad:o,strides:i,inputShape:l}=r;Ne([s],"conv3dBackpropInputV2");let u=k.computeStrides(s.shape),c=k.computeStrides(a.shape),d=_.computeConv3DInfo(l,a.shape,i,1,o),h=new Jt(d.inShape,"float32"),p=h.values,[f,m,g,y]=h.strides,A=n.data.get(s.dataId).values,[x,b,v,I]=u,w=n.data.get(a.dataId).values,[S,E,D,$]=c,{batchSize:R,filterDepth:N,filterHeight:M,filterWidth:B,inChannels:q,inDepth:X,inHeight:J,inWidth:ee,outChannels:ae,outDepth:se,outHeight:oe,outWidth:ne,strideDepth:ce,strideHeight:he,strideWidth:me}=d,be=N-1-d.padInfo.front,Ee=M-1-d.padInfo.top,$e=B-1-d.padInfo.left;for(let Pe=0;Pe<R;++Pe)for(let je=0;je<q;++je)for(let Be=0;Be<X;++Be){let bt=Be-be,pt=Math.max(0,Math.ceil(bt/ce)),ft=Math.min(se,(N+bt)/ce);for(let dt=0;dt<J;++dt){let xt=dt-Ee,Ye=Math.max(0,Math.ceil(xt/he)),Gn=Math.min(oe,(M+xt)/he);for(let Bt=0;Bt<ee;++Bt){let or=Bt-$e,bn=Math.max(0,Math.ceil(or/me)),zr=Math.min(ne,(B+or)/me),Rn=0;for(let Ar=pt;Ar<ft;++Ar){let xr=Ar*ce-bt;for(let vn=Ye;vn<Gn;++vn){let br=vn*he-xt;for(let vr=bn;vr<zr;++vr){let ir=vr*me-or,bs=x*Pe+b*Ar+v*vn+I*vr,Hs=S*(N-1-xr)+E*(M-1-br)+D*(B-1-ir)+$*je;for(let xa=0;xa<ae;++xa){let Ii=A[bs+xa],vs=w[Hs+xa];Rn+=Ii*vs}}}}p[f*Pe+m*Be+g*dt+y*Bt+je]=Rn}}}return n.makeTensorInfo(h.shape,h.dtype,h.values)}var Kie={kernelName:I1,backendName:"cpu",kernelFunc:qie},Xie=At(nl,e=>Math.cos(e)),Zie={kernelName:nl,backendName:"cpu",kernelFunc:Xie},Yie=At(rl,e=>Math.cosh(e)),Jie={kernelName:rl,backendName:"cpu",kernelFunc:Yie};function Qie(e){let{inputs:t,backend:n,attrs:r}=e,{image:s,boxes:a,boxInd:o}=t,{cropSize:i,method:l,extrapolationValue:u}=r,[c,d,h,p]=s.shape,f=a.shape[0],[m,g]=i,y=ze([f,m,g,p],"float32"),A=n.data.get(a.dataId).values,x=n.data.get(o.dataId).values,b=n.data.get(s.dataId).values,v=k.computeStrides(s.shape),I=k.computeStrides(y.shape);for(let w=0;w<f;w++){let S=w*4,E=A[S],D=A[S+1],$=A[S+2],R=A[S+3],N=x[w];if(N>=c)continue;let M=m>1?($-E)*(d-1)/(m-1):0,B=g>1?(R-D)*(h-1)/(g-1):0;for(let q=0;q<m;q++){let X=m>1?E*(d-1)+q*M:.5*(E+$)*(d-1);if(X<0||X>d-1){for(let J=0;J<g;J++)for(let ee=0;ee<p;ee++){let ae=ee+J*I[2]+q*I[1]+w*I[0];y.values[ae]=u}continue}if(l==="bilinear"){let J=Math.floor(X),ee=Math.ceil(X),ae=X-J;for(let se=0;se<g;se++){let oe=g>1?D*(h-1)+se*B:.5*(D+R)*(h-1);if(oe<0||oe>h-1){for(let me=0;me<p;me++){let be=me+se*I[2]+q*I[1]+w*I[0];y.values[be]=u}continue}let ne=Math.floor(oe),ce=Math.ceil(oe),he=oe-ne;for(let me=0;me<p;me++){let be=me+ne*v[2]+J*v[1]+N*v[0],Ee=b[be];be=me+ce*v[2]+J*v[1]+N*v[0];let $e=b[be];be=me+ne*v[2]+ee*v[1]+N*v[0];let Pe=b[be];be=me+ce*v[2]+ee*v[1]+N*v[0];let je=b[be],Be=Ee+($e-Ee)*he,bt=Pe+(je-Pe)*he;be=me+se*I[2]+q*I[1]+w*I[0],y.values[be]=Be+(bt-Be)*ae}}}else for(let J=0;J<g;++J){let ee=g>1?D*(h-1)+J*B:.5*(D+R)*(h-1);if(ee<0||ee>h-1){for(let oe=0;oe<p;oe++){let ne=oe+J*I[2]+q*I[1]+w*I[0];y.values[ne]=u}continue}let ae=Math.round(ee),se=Math.round(X);for(let oe=0;oe<p;oe++){let ne=oe+ae*v[2]+se*v[1]+N*v[0],ce=oe+J*I[2]+q*I[1]+w*I[0];y.values[ce]=b[ne]}}}}return n.makeTensorInfo(y.shape,y.dtype,y.values)}var ele={kernelName:$c,backendName:"cpu",kernelFunc:Qie};function tle(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,exclusive:o,reverse:i}=r;Ne(s,"cumsum");let l=_.getAxesPermutation([a],s.shape.length),u=s;l!=null&&(u=Dr({inputs:{x:s},backend:n,attrs:{perm:l}}));let c=_.getInnerMostAxes(1,s.shape.length)[0];if(c!==u.shape.length-1)throw new Error(`backend.cumsum in CPU expects an inner-most axis=${u.shape.length-1} but got axis=${c}`);let d=Ur(u.dtype,"int32"),h=k.makeZerosTypedArray(k.sizeFromShape(u.shape),d),p=n.data.get(u.dataId).values,f=u.shape[u.shape.length-1],m=i?(y,A)=>y+f-A-1:(y,A)=>y+A;for(let y=0;y<p.length;y+=f)for(let A=0;A<f;A++){let x=m(y,A);if(A===0)h[x]=o?0:p[x];else{let b=m(y,A-1);h[x]=o?p[b]+h[b]:p[x]+h[b]}}let g=n.makeTensorInfo(u.shape,d,h);if(l!=null){let y=_.getUndoAxesPermutation(l),A=Dr({inputs:{x:g},backend:n,attrs:{perm:y}});return n.disposeIntermediateTensorInfo(g),n.disposeIntermediateTensorInfo(u),A}return g}var nle={kernelName:sl,backendName:"cpu",kernelFunc:tle};function rle(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,weights:a}=t,{size:o,binaryOutput:i}=r;if(s.shape.length===1){let l=n.data.get(s.dataId).values,u=n.data.get(a.dataId).values,c=v5(l,u,a.dtype,a.shape,o);return n.makeTensorInfo([o],a.dtype,c)}else if(s.shape.length===2){let l=n.bufferSync(s),u=n.bufferSync(a),c=_T(l,u,o,i);return n.makeTensorInfo(c.shape,a.dtype,c.values)}throw new Error(`Error in denseBincount: input must be at most rank 2, but got rank${s.shape.length}.`)}var sle={kernelName:S1,backendName:"cpu",kernelFunc:rle};function ale(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{blockSize:a,dataFormat:o}=r;k.assert(o==="NHWC",()=>`Only NHWC dataFormat supported on CPU for depthToSpace. Got ${o}`),k.assert(a>1,()=>`blockSize should be > 1 for depthToSpace, but was: ${a}`);let i=s.shape[0],l=s.shape[1],u=s.shape[2],c=s.shape[3],d=l*a,h=u*a,p=c/(a*a),f=n.data.get(s.dataId).values,m=new Float32Array(i*d*h*p),g=0;for(let y=0;y<i;++y)for(let A=0;A<d;++A){let x=Math.floor(A/a),b=A%a;for(let v=0;v<h;++v){let I=Math.floor(v/a),w=v%a,S=(b*a+w)*p;for(let E=0;E<p;++E){let $=E+S+c*(I+u*(x+l*y));m[g++]=f[$]}}}return n.makeTensorInfo([i,d,h,p],s.dtype,m)}var ole={kernelName:Rc,backendName:"cpu",kernelFunc:ale};function TN(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a}=t,{strides:o,pad:i,dilations:l,dimRoundingMode:u}=r;Ne([s,a],"depthwiseConv2DNative");let c=k.computeStrides(s.shape),d=k.computeStrides(a.shape),h=l;h==null&&(h=[1,1]),k.assert(_.eitherStridesOrDilationsAreOne(o,h),()=>`Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${o} and dilations '${h}'`);let p=_.computeConv2DInfo(s.shape,a.shape,o,h,i,u,!0),{filterHeight:f,filterWidth:m,dilationHeight:g,dilationWidth:y,padInfo:A}=p,x=A.left,b=A.top,v=p.outChannels/p.inChannels,I=new Jt(p.outShape,s.dtype),w=n.data.get(s.dataId).values,S=n.data.get(a.dataId).values,E=I.values;for(let D=0;D<p.batchSize;++D){let $=D*c[0],R=D*I.strides[0];for(let N=0;N<p.outHeight;++N){let M=R+N*I.strides[1],B=N*p.strideHeight-b;for(let q=0;q<f;++q){let X=B+q*g;if(X<0||X>=p.inHeight)continue;let J=q*d[0],ee=$+X*c[1];for(let ae=0;ae<p.outWidth;++ae){let se=M+ae*I.strides[2],oe=ae*p.strideWidth-x;for(let ne=0;ne<m;++ne){let ce=oe+ne*y;if(ce<0||ce>=p.inWidth)continue;let he=J+ne*d[1],me=ee+ce*p.inChannels,be=se,Ee=he;for(let $e=0;$e<p.inChannels;++$e){let Pe=w[me+$e];for(let je=0;je<v;++je)E[be+je]+=Pe*S[Ee+je];be+=v,Ee+=v}}}}}}return n.makeTensorInfo(I.shape,I.dtype,I.values)}var ile={kernelName:al,backendName:"cpu",kernelFunc:TN};function lle(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,dy:a}=t,{strides:o,dilations:i,pad:l,dimRoundingMode:u,filterShape:c}=r;Ne([s,a],"depthwiseConv2dNativeBackpropFilter");let d=_.computeConv2DInfo(s.shape,c,o,i,l,u,!0),{strideHeight:h,strideWidth:p,filterHeight:f,filterWidth:m}=d,g=new Jt(d.filterShape,"float32"),y=d.padInfo.left,A=d.padInfo.top,x=d.outChannels/d.inChannels,b=n.data.get(s.dataId).values,v=new Jt(s.shape,s.dtype,b),I=n.data.get(a.dataId).values,w=new Jt(a.shape,a.dtype,I);for(let S=0;S<f;++S){let E=Math.max(0,Math.ceil((A-S)/h)),D=Math.min(d.outHeight,(d.inHeight+A-S)/h);for(let $=0;$<m;++$){let R=Math.max(0,Math.ceil((y-$)/p)),N=Math.min(d.outWidth,(d.inWidth+y-$)/p);for(let M=0;M<d.outChannels;++M){let B=Math.trunc(M/x),q=M%x,X=0;for(let J=0;J<d.batchSize;++J)for(let ee=E;ee<D;++ee){let ae=S+ee*h-A;for(let se=R;se<N;++se){let oe=$+se*p-y;X+=v.get(J,ae,oe,B)*w.get(J,ee,se,M)}}g.set(X,S,$,B,q)}}}return n.makeTensorInfo(g.shape,g.dtype,g.values)}var ule={kernelName:T1,backendName:"cpu",kernelFunc:lle};function cle(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,filter:a}=t,{strides:o,dilations:i,pad:l,dimRoundingMode:u,inputShape:c}=r;Ne([s,a],"depthwiseConv2DNativeBackpropInput");let d=k.computeStrides(s.shape),h=k.computeStrides(a.shape),p=_.computeConv2DInfo(c,a.shape,o,i,l,u,!0),f=new Jt(p.inShape,"float32"),m=f.values,[g,y,A]=f.strides,x=n.data.get(s.dataId).values,[b,v,I]=d,w=n.data.get(a.dataId).values,[S,E,D]=h,{batchSize:$,filterHeight:R,filterWidth:N,inChannels:M,inHeight:B,inWidth:q,outChannels:X,outHeight:J,outWidth:ee,strideHeight:ae,strideWidth:se}=p,oe=R-1-p.padInfo.top,ne=N-1-p.padInfo.left,ce=X/M;for(let he=0;he<$;++he)for(let me=0;me<M;++me)for(let be=0;be<B;++be){let Ee=be-oe,$e=Math.max(0,Math.ceil(Ee/ae)),Pe=Math.min(J,(R+Ee)/ae);for(let je=0;je<q;++je){let Be=je-ne,bt=Math.max(0,Math.ceil(Be/se)),pt=Math.min(ee,(N+Be)/se),ft=0;for(let dt=$e;dt<Pe;++dt){let xt=dt*ae-Ee;for(let Ye=bt;Ye<pt;++Ye){let Gn=Ye*se-Be,Bt=b*he+v*dt+I*Ye,or=S*(R-1-xt)+E*(N-1-Gn)+D*me;for(let bn=0;bn<ce;++bn){let zr=me*ce+bn,Rn=x[Bt+zr],Ar=w[or+bn];ft+=Rn*Ar}}}m[g*he+y*be+A*je+me]=ft}}return n.makeTensorInfo(f.shape,f.dtype,f.values)}var dle={kernelName:N1,backendName:"cpu",kernelFunc:cle};function hle(e){let{inputs:t,backend:n}=e,{x:r}=t,s=k.sizeFromShape(r.shape),a=n.data.get(r.dataId).values,o=ze([s,s],r.dtype),i=o.values;for(let u=0;u<a.length;u++)i[u*s+u]=a[u];let l=[...r.shape,...r.shape];return n.makeTensorInfo(l,o.dtype,o.values)}var ple={kernelName:C1,backendName:"cpu",kernelFunc:hle},fle={kernelName:Yp,backendName:"cpu",kernelFunc:({inputs:e,backend:t,attrs:n})=>{let{x:r,filter:s}=e,{strides:a,pad:o,dilations:i}=n,l=t,u=l.data.get(r.dataId).values,c=r.shape.length,d=l.data.get(s.dataId).values,h=s.shape.length,{batchSize:p,inHeight:f,inWidth:m,inChannels:g,outHeight:y,outWidth:A,padInfo:x,strideHeight:b,strideWidth:v,filterHeight:I,filterWidth:w,dilationHeight:S,dilationWidth:E,outShape:D}=_.computeDilation2DInfo(r.shape,s.shape,a,o,"NHWC",i),$=k.sizeFromShape(D),R=D.length,N=k.getArrayFromDType(r.dtype,$);for(let B=0;B<p;++B)for(let q=0;q<y;++q){let X=q*b-x.top;for(let J=0;J<A;++J){let ee=J*v-x.left;for(let ae=0;ae<g;++ae){let se=Number.MIN_SAFE_INTEGER;for(let ne=0;ne<I;++ne){let ce=X+ne*S;if(ce>=0&&ce<f)for(let he=0;he<w;++he){let me=ee+he*E;if(me>=0&&me<m){let be=k.locToIndex([B,ce,me,ae],c,k.computeStrides(r.shape)),Ee=k.locToIndex([ne,he,ae],h,k.computeStrides(s.shape)),$e=u[be]+d[Ee];$e>se&&(se=$e)}}}let oe=k.locToIndex([B,q,J,ae],R,k.computeStrides(D));N[oe]=se}}}return{dataId:l.write(k.toTypedArray(N,r.dtype),D,r.dtype),shape:D,dtype:r.dtype}}},mle={kernelName:$1,backendName:"cpu",kernelFunc:({inputs:e,backend:t,attrs:n})=>{let{x:r,filter:s,dy:a}=e,{strides:o,pad:i,dilations:l}=n,u=t,c=k.toNestedArray(r.shape,u.data.get(r.dataId).values),d=k.toNestedArray(s.shape,u.data.get(s.dataId).values),{batchSize:h,inHeight:p,inWidth:f,inChannels:m,outHeight:g,outWidth:y,padInfo:A,strideHeight:x,strideWidth:b,filterHeight:v,filterWidth:I,dilationHeight:w,dilationWidth:S,outShape:E}=_.computeDilation2DInfo(r.shape,s.shape,o,i,"NHWC",l);k.assert(a.rank===E.length,()=>`Error in ${$1}, dy must have the same rank as output ${E.length}, but got ${a.rank}`);let D=k.toNestedArray(E,u.data.get(a.dataId).values),$=k.makeZerosNestedTypedArray(s.shape,s.dtype);for(let N=0;N<h;++N)for(let M=0;M<g;++M){let B=M*x-A.top;for(let q=0;q<y;++q){let X=q*b-A.left;for(let J=0;J<m;++J){let ee=Number.MIN_SAFE_INTEGER,ae=0,se=0;for(let oe=0;oe<v;++oe){let ne=B+oe*w;if(ne>=0&&ne<p)for(let ce=0;ce<I;++ce){let he=X+ce*S;if(he>=0&&he<f){let me=c[N][ne][he][J]+d[oe][ce][J];me>ee&&(ee=me,ae=oe,se=ce)}}}$[ae][se][J]+=D[N][M][q][J]}}}return{dataId:u.write(k.toTypedArray($,r.dtype),s.shape,s.dtype),shape:s.shape,dtype:s.dtype}}},gle={kernelName:E1,backendName:"cpu",kernelFunc:({inputs:e,backend:t,attrs:n})=>{let{x:r,filter:s,dy:a}=e,{strides:o,pad:i,dilations:l}=n,u=t,c=k.toNestedArray(r.shape,u.data.get(r.dataId).values),d=k.toNestedArray(s.shape,u.data.get(s.dataId).values),{batchSize:h,inHeight:p,inWidth:f,inChannels:m,outHeight:g,outWidth:y,padInfo:A,strideHeight:x,strideWidth:b,filterHeight:v,filterWidth:I,dilationHeight:w,dilationWidth:S,outShape:E}=_.computeDilation2DInfo(r.shape,s.shape,o,i,"NHWC",l);k.assert(a.rank===E.length,()=>`Error in ${E1}, dy must have the same rank as output ${E.length}, but got ${a.rank}`);let D=k.toNestedArray(E,u.data.get(a.dataId).values),$=k.makeZerosNestedTypedArray(r.shape,r.dtype);for(let N=0;N<h;++N)for(let M=0;M<g;++M){let B=M*x-A.top;for(let q=0;q<y;++q){let X=q*b-A.left;for(let J=0;J<m;++J){let ee=Number.MIN_SAFE_INTEGER,ae=B<0?0:B,se=X<0?0:X;for(let oe=0;oe<v;++oe){let ne=B+oe*w;if(ne>=0&&ne<p)for(let ce=0;ce<I;++ce){let he=X+ce*S;if(he>=0&&he<f){let me=c[N][ne][he][J]+d[oe][ce][J];me>ee&&(ee=me,ae=ne,se=he)}}}$[N][ae][se][J]+=D[N][M][q][J]}}}return{dataId:u.write(k.toTypedArray($,r.dtype),r.shape,r.dtype),shape:r.shape,dtype:r.dtype}}};function Qd(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r;Ne(s,"sum");let i;s.dtype==="bool"?i=Ja({inputs:{x:s},backend:n,attrs:{dtype:"int32"}}):i=Ls({inputs:{x:s},backend:n});let l=i.shape.length,u=k.parseAxisParam(a,i.shape),c=_.getAxesPermutation(u,l),d=u,h=i;c!=null&&(h=Dr({inputs:{x:i},backend:n,attrs:{perm:c}}),d=_.getInnerMostAxes(d.length,l)),_.assertAxesAreInnerMostDims("sum",d,h.shape.length);let[p,f]=_.computeOutAndReduceShapes(h.shape,d),m=_.upcastType(h.dtype,"int32"),g=km(n,p,m),y=k.sizeFromShape(f),A=n.data.get(g.dataId).values,x=n.data.get(h.dataId).values;for(let b=0;b<A.length;++b){let v=b*y,I=0;for(let w=0;w<y;++w)I+=x[v+w];A[b]=I}if(o){let b=_.expandShapeToKeepDim(g.shape,u),v=g;g=_t({inputs:{x:g},backend:n,attrs:{shape:b}}),n.disposeIntermediateTensorInfo(v)}return n.disposeIntermediateTensorInfo(i),c!=null&&n.disposeIntermediateTensorInfo(h),g}var yle={kernelName:Fl,backendName:"cpu",kernelFunc:Qd};function Ale(e){let{inputs:t,backend:n,attrs:r}=e,{equation:s}=r,a=t,{allDims:o,summedDims:i,idDims:l}=_.decodeEinsumEquation(s,a.length);_.checkEinsumDimSizes(o.length,l,a);let{path:u,steps:c}=_.getEinsumComputePath(i,l),d=c.length,h=null,p=o.length,f=[];for(let m=0;m<d;++m){for(let g of c[m]){let{permutationIndices:y,expandDims:A}=_.getEinsumPermutation(p,l[g]),x;_.isIdentityPermutation(y)?x=a[g]:(x=Dr({inputs:{x:a[g]},backend:n,attrs:{perm:y}}),f.push(x));let b=x.shape.slice();for(let v=0;v<A.length;++v)b.splice(A[v],0,1);k.arraysEqual(x.shape,b)||(x=_t({inputs:{x},backend:n,attrs:{shape:b}}),f.push(x)),h===null?h=x:(h=Im({inputs:{a:x,b:h},backend:n}),f.push(h))}m<d-1&&(u[m]>=0&&(h=Qd({inputs:{x:h},backend:n,attrs:{axis:u[m]-(o.length-p),keepDims:!1}}),f.push(h)),p--)}for(let m of f)m!==h&&n.disposeIntermediateTensorInfo(m);return h}var xle={kernelName:R1,backendName:"cpu",kernelFunc:Ale};function ble(e){let{inputs:t,backend:n}=e,{dy:r,y:s}=t;Ne([r,s],"eluGrad");let a=new Float32Array(k.sizeFromShape(s.shape)),o=n.data.get(s.dataId).values,i=n.data.get(r.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(s.shape,"float32",a)}var vle={kernelName:_1,backendName:"cpu",kernelFunc:ble},wle=_.ERF_P,kle=_.ERF_A1,Ile=_.ERF_A2,Sle=_.ERF_A3,Tle=_.ERF_A4,Nle=_.ERF_A5,Cle=At(Dc,e=>{let t=Math.sign(e),n=Math.abs(e),r=1/(1+wle*n);return t*(1-((((Nle*r+Tle)*r+Sle)*r+Ile)*r+kle)*r*Math.exp(-n*n))}),Ele={kernelName:Dc,backendName:"cpu",kernelFunc:Cle};function Sm(e){let{inputs:t,backend:n,attrs:r}=e,{input:s}=t,{dim:a}=r,o=s.shape.length,i=s.shape.slice(),l=a;return a<0&&(k.assert(-(o+1)<=a,()=>`Axis must be in the interval [${-(o+1)}, ${o}]`),l=o+a+1),i.splice(l,0,1),_t({inputs:{x:s},backend:n,attrs:{shape:i}})}var $le={kernelName:Fc,backendName:"cpu",kernelFunc:Sm},Rle=en((e,t)=>e/t),C5=xn(ol,Rle),E5={kernelName:ol,backendName:"cpu",kernelFunc:C5};function NN(e,t,n){let r=e.shape,s=r[0],a=r[1],o=n.data.get(e.dataId),i=o.complexTensorInfos.real,l=o.complexTensorInfos.imag,u=[s,a],c=k.sizeFromShape(u),d=k.getTypedArrayFromDType("float32",c),h=k.getTypedArrayFromDType("float32",c);for(let g=0;g<s;g++){let y=hi({inputs:{x:i},backend:n,attrs:{begin:[g,0],size:[1,a]}}),A=hi({inputs:{x:l},backend:n,attrs:{begin:[g,0],size:[1,a]}}),x=pr({inputs:{real:y,imag:A},backend:n}),{real:b,imag:v}=_le(x,t,n),I=_.mergeRealAndImagArrays(b,v);for(let w=0;w<a;w++){let S=_.getComplexWithIndex(I,w);d[g*a+w]=S.real,h[g*a+w]=S.imag}n.disposeIntermediateTensorInfo(y),n.disposeIntermediateTensorInfo(A),n.disposeIntermediateTensorInfo(x)}let p=n.makeTensorInfo(u,"float32",d),f=n.makeTensorInfo(u,"float32",h),m=pr({inputs:{real:p,imag:f},backend:n});return n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(f),m}function _le(e,t,n){let r=k.sizeFromShape(e.shape),s=n.data.get(e.dataId),a=n.data.get(s.complexTensorInfos.real.dataId).values,o=n.data.get(s.complexTensorInfos.imag.dataId).values;if(Dle(r)){let i=$5(a,o,r,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),d=n.makeTensorInfo([],"float32",k.createScalarValue(r,"float32")),h=Ls({inputs:{x:d},backend:n}),p=E5.kernelFunc({inputs:{a:u,b:d},backend:n}),f=E5.kernelFunc({inputs:{a:c,b:h},backend:n}),m=n.data.get(p.dataId).values,g=n.data.get(f.dataId).values;return n.disposeIntermediateTensorInfo(u),n.disposeIntermediateTensorInfo(c),n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(f),{real:m,imag:g}}return i}else{let i=_.mergeRealAndImagArrays(a,o),l=Fle(i,r,t);return _.splitRealAndImagArrays(l)}}function Dle(e){return(e&e-1)==0}function $5(e,t,n,r,s){if(n===1)return{real:e,imag:t};let a=_.mergeRealAndImagArrays(e,t),o=n/2,i=_.complexWithEvenIndex(a),l=i.real,u=i.imag,c=[l.length],d=s.makeTensorInfo(c,"float32",l),h=s.makeTensorInfo(c,"float32",u),p=pr({inputs:{real:d,imag:h},backend:s}),f=_.complexWithOddIndex(a),m=f.real,g=f.imag,y=[m.length],A=s.makeTensorInfo(y,"float32",m),x=s.makeTensorInfo(y,"float32",g),b=pr({inputs:{real:A,imag:x},backend:s}),v=$5(l,u,o,r,s),I=v.real,w=v.imag,S=[I.length],E=s.makeTensorInfo(S,"float32",I),D=s.makeTensorInfo(S,"float32",w),$=pr({inputs:{real:E,imag:D},backend:s}),R=$5(m,g,o,r,s),N=R.real,M=R.imag,B=[N.length],q=s.makeTensorInfo(B,"float32",N),X=s.makeTensorInfo(B,"float32",M),J=pr({inputs:{real:q,imag:X},backend:s}),ee=_.exponents(n,r),ae=[ee.real.length],se=s.makeTensorInfo(ae,"float32",ee.real),oe=s.makeTensorInfo(ae,"float32",ee.imag),ne=pr({inputs:{real:se,imag:oe},backend:s}),ce=Im({inputs:{a:ne,b:J},backend:s}),he=Yd({inputs:{a:$,b:ce},backend:s}),me=S5({inputs:{a:$,b:ce},backend:s}),be=di({inputs:{input:he},backend:s}),Ee=di({inputs:{input:me},backend:s}),$e=uu({inputs:{input:he},backend:s}),Pe=uu({inputs:{input:me},backend:s}),je=cu({inputs:[be,Ee],backend:s,attrs:{axis:0}}),Be=cu({inputs:[$e,Pe],backend:s,attrs:{axis:0}}),bt=s.data.get(je.dataId).values,pt=s.data.get(Be.dataId).values;return s.disposeIntermediateTensorInfo(d),s.disposeIntermediateTensorInfo(h),s.disposeIntermediateTensorInfo(p),s.disposeIntermediateTensorInfo(A),s.disposeIntermediateTensorInfo(x),s.disposeIntermediateTensorInfo(b),s.disposeIntermediateTensorInfo(E),s.disposeIntermediateTensorInfo(D),s.disposeIntermediateTensorInfo($),s.disposeIntermediateTensorInfo(q),s.disposeIntermediateTensorInfo(X),s.disposeIntermediateTensorInfo(J),s.disposeIntermediateTensorInfo(se),s.disposeIntermediateTensorInfo(oe),s.disposeIntermediateTensorInfo(ne),s.disposeIntermediateTensorInfo(ce),s.disposeIntermediateTensorInfo(he),s.disposeIntermediateTensorInfo(me),s.disposeIntermediateTensorInfo(be),s.disposeIntermediateTensorInfo($e),s.disposeIntermediateTensorInfo(Ee),s.disposeIntermediateTensorInfo(Pe),s.disposeIntermediateTensorInfo(je),s.disposeIntermediateTensorInfo(Be),{real:bt,imag:pt}}function Fle(e,t,n){let r=new Float32Array(t*2);for(let s=0;s<t;s++){let a=0,o=0;for(let i=0;i<t;i++){let l=_.exponent(s*i,t,n),u=_.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),_.assignToTypedArray(r,a,o,s)}return r}function Mle(e){let{inputs:t,backend:n}=e,{input:r}=t,s=k.sizeFromShape(r.shape),a=r.shape[r.shape.length-1],o=s/a,i=_t({inputs:{x:r},backend:n,attrs:{shape:[o,a]}}),l=NN(i,!1,n),u=_t({inputs:{x:l},backend:n,attrs:{shape:r.shape}});return n.disposeIntermediateTensorInfo(i),n.disposeIntermediateTensorInfo(l),u}var Ole={kernelName:D1,backendName:"cpu",kernelFunc:Mle};function R5(e){let{backend:t,attrs:n}=e,{shape:r,value:s,dtype:a}=n,o=a||k.inferDtype(s),i=k.getArrayFromDType(o,k.sizeFromShape(r));return zle(i,s,o),t.makeTensorInfo(r,o,i)}var Ple={kernelName:Jp,backendName:"cpu",kernelFunc:R5};function zle(e,t,n){e.fill(t)}var Lle={kernelName:Mc,backendName:"cpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{image:r}=e,s=n,a=k.getTypedArrayFromDType(r.dtype,k.sizeFromShape(r.shape)),[o,i,l,u]=r.shape,c=s.data.get(r.dataId).values;for(let h=0;h<o;h++){let p=h*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 A=0;A<u;A++){let x=Math.round(l-g-1),b=p+m+y+A,v=c[b];if(x>=0&&x<l){let I=x*u,w=p+m+I+A;v=c[w]}a[b]=v}}}}return{dataId:s.write(a,r.shape,r.dtype),shape:r.shape,dtype:r.dtype}}},Ble=en((e,t)=>Math.floor(e/t)),Wle=xn(ul,Ble,null,"int32"),Vle={kernelName:ul,backendName:"cpu",kernelFunc:Wle};function Ule(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dataFormat:c,dilations:d,dimRoundingMode:h,activation:p,leakyreluAlpha:f}=r,m=SN({inputs:{x:s,filter:a},backend:n,attrs:{strides:l,pad:u,dataFormat:c,dilations:d,dimRoundingMode:h}});if(o){let g=m;m=Yd({inputs:{a:m,b:o},backend:n}),n.disposeIntermediateTensorInfo(g)}if(p){let g=m;m=T5(n,m,p,i,f),n.disposeIntermediateTensorInfo(g)}return m}var Hle={kernelName:Bl,backendName:"cpu",kernelFunc:Ule};function Gle(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dataFormat:c,dilations:d,dimRoundingMode:h,activation:p,leakyreluAlpha:f}=r,m=TN({inputs:{x:s,filter:a},backend:n,attrs:{strides:l,pad:u,dataFormat:c,dilations:d,dimRoundingMode:h}});if(o){let g=m;m=Yd({inputs:{a:m,b:o},backend:n}),n.disposeIntermediateTensorInfo(g)}if(p){let g=m;m=T5(n,m,p,i,f),n.disposeIntermediateTensorInfo(g)}return m}var jle={kernelName:Wl,backendName:"cpu",kernelFunc:Gle};function qle(e){let{inputs:t,backend:n}=e,{params:r,indices:s}=t,a=k.sizeFromShape(r.shape),o=s.shape,i=o[o.length-1],[l,u,c,d]=_.prepareAndValidate(r,s);if(u===0)return n.makeTensorInfo(l,r.dtype,[]);let h=n.data.get(s.dataId).values,p=n.bufferSync(r),f=WT(h,p,r.dtype,u,i,c,d,r.shape,a);return n.makeTensorInfo(l,r.dtype,f.values)}var Kle={kernelName:Pc,backendName:"cpu",kernelFunc:qle};function Xle(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,indices:a}=t,{axis:o,batchDims:i}=r;Ne([s,a],"gatherV2");let l=i;i==null&&(l=0);let u=k.sizeFromShape(a.shape),c=k.parseAxisParam(o,s.shape)[0],d=_.segment_util.collectGatherOpShapeInfo(s,a,c,l),h=_t({inputs:{x:s},backend:n,attrs:{shape:[d.batchSize,d.outerSize,d.dimSize,d.sliceSize]}}),p=_t({inputs:{x:a},backend:n,attrs:{shape:[d.batchSize,u/d.batchSize]}}),f=[d.batchSize,d.outerSize,u/d.batchSize,d.sliceSize],m=n.bufferSync(p),g=n.bufferSync(h),y=VT(g,m,f);return n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(p),n.makeTensorInfo(d.outputShape,y.dtype,y.values)}var Zle={kernelName:Oc,backendName:"cpu",kernelFunc:Xle};function Yle(e){let{inputs:t,backend:n}=e,{input:r}=t,s=k.sizeFromShape(r.shape),a=r.shape[r.shape.length-1],o=s/a,i=_t({inputs:{x:r},backend:n,attrs:{shape:[o,a]}}),l=NN(i,!0,n),u=_t({inputs:{x:l},backend:n,attrs:{shape:r.shape}});return n.disposeIntermediateTensorInfo(i),n.disposeIntermediateTensorInfo(l),u}var Jle={kernelName:F1,backendName:"cpu",kernelFunc:Yle},Qle=At(zc,e=>Number.isFinite(e)?1:0,"bool"),eue={kernelName:zc,backendName:"cpu",kernelFunc:Qle},tue=At(Lc,e=>Math.abs(e)===1/0?1:0,"bool"),nue={kernelName:Lc,backendName:"cpu",kernelFunc:tue},rue=At(Bc,e=>Number.isNaN(e)?1:0,"bool"),sue={kernelName:Bc,backendName:"cpu",kernelFunc:rue};function aue(e){let{backend:t,attrs:n}=e,{start:r,stop:s,num:a}=n,o=qT(r,s,a);return t.makeTensorInfo([o.length],"float32",o)}var oue={kernelName:O1,backendName:"cpu",kernelFunc:aue},iue=At(Wc,e=>Math.log1p(e)),lue={kernelName:Wc,backendName:"cpu",kernelFunc:iue},uue=en((e,t)=>e&&t),cue=xn(Vc,uue,null,"bool"),due={kernelName:Vc,backendName:"cpu",kernelFunc:cue},hue=At(Qp,e=>e?0:1,"bool"),pue={kernelName:Qp,backendName:"cpu",kernelFunc:hue},fue=en((e,t)=>e||t),mue=xn(ef,fue,null,"bool"),gue={kernelName:ef,backendName:"cpu",kernelFunc:mue};function yue(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{depthRadius:a,bias:o,alpha:i,beta:l}=r;Ne(s,"LRN");let u=s.shape[3],c=u-1,d=n.data.get(s.dataId).values,h=k.sizeFromShape(s.shape),p=new Float32Array(h);function f(m){let g=m%u,y=m-g+Math.max(0,g-a),A=m-g+Math.min(g+a,c),x=0;for(;y<=A;y++){let b=d[y];x+=b*b}return x}for(let m=0;m<h;m++){let g=f(m),y=d[m]*Math.pow(o+i*g,-l);p[m]=y}return n.makeTensorInfo(s.shape,s.dtype,p)}var Aue={kernelName:tf,backendName:"cpu",kernelFunc:yue};function xue(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,y:a,dy:o}=t,{depthRadius:i,bias:l,alpha:u,beta:c}=r;Ne(o,"LRNGrad");let d=k.sizeFromShape(o.shape),h=o.shape[3],p=n.data.get(o.dataId).values,f=n.data.get(s.dataId).values,m=n.data.get(a.dataId).values,g=new Float32Array(d),y=d;for(let A=0;A<y;A++){let x=A%h,b=A-x+Math.max(0,x-i),v=A-x+Math.min(h,x+i+1),I=0;for(let w=b;w<v;w++)I+=Math.pow(f[w],2);I=u*I+l;for(let w=b;w<v;w++){let S=-2*u*c*f[w]*m[A]/I;A===w&&(S+=Math.pow(I,-c)),S*=p[A],g[w]+=S}}return n.makeTensorInfo(o.shape,s.dtype,g)}var bue={kernelName:P1,backendName:"cpu",kernelFunc:xue};function CN(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{reductionIndices:a,keepDims:o}=r,i=n,l=s.shape,u=l.length,c=k.parseAxisParam(a,l),d=c,h=_.getAxesPermutation(d,u),p=i.data.get(s.dataId).values;if(h!=null){let b=new Array(u);for(let v=0;v<b.length;v++)b[v]=l[h[v]];p=k5(p,l,s.dtype,h,b),d=_.getInnerMostAxes(d.length,u),l=b}Ne(s,"max"),_.assertAxesAreInnerMostDims("max",d,u);let[f,m]=_.computeOutAndReduceShapes(l,d),g=k.sizeFromShape(m),y=XT(p,g,f,s.dtype),A=i.write(y,f,s.dtype),x=f;return o&&(x=_.expandShapeToKeepDim(f,c)),{dataId:A,shape:x,dtype:s.dtype}}var vue={kernelName:gl,backendName:"cpu",kernelFunc:CN};function wue(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t;Ne(s,"maxPool");let{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=r,u=1;k.assert(_.eitherStridesOrDilationsAreOne(o,u),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${o} and dilations '${u}'`);let c=_.computePool2DInfo(s.shape,a,o,u,i,l),d;if(c.filterWidth===1&&c.filterHeight===1&&k.arraysEqual(c.inShape,c.outShape))d=Ls({inputs:{x:s},backend:n});else{let h=n.data.get(s.dataId).values,p=k.computeStrides(s.shape),f=N5(h,s.shape,s.dtype,p,c,"max");d=n.makeTensorInfo(c.outShape,s.dtype,f.values)}return d}var kue={kernelName:yl,backendName:"cpu",kernelFunc:wue};function Iue(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l,dataFormat:u}=r;Ne(s,"maxPool3d");let c=_.computePool3DInfo(s.shape,a,o,1,i,l,u),d=n.data.get(s.dataId).values,h=IN(d,s.shape,s.dtype,k.computeStrides(s.shape),c,"max");return n.makeTensorInfo(h.shape,"float32",h.values)}var Sue={kernelName:nf,backendName:"cpu",kernelFunc:Iue};function Tue(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,input:a}=t,{filterSize:o,strides:i,pad:l,dimRoundingMode:u}=r;Ne([s,a],"maxPool3DGrad");let c=_.computePool3DInfo(a.shape,o,i,1,l,u),d=n.bufferSync(a),h=yie(d,c),p=c.strideDepth,f=c.strideHeight,m=c.strideWidth,g=c.dilationDepth,y=c.dilationHeight,A=c.dilationWidth,x=c.effectiveFilterDepth,b=c.effectiveFilterHeight,v=c.effectiveFilterWidth,I=x-1-c.padInfo.front,w=v-1-c.padInfo.left,S=b-1-c.padInfo.top,E=ze(a.shape,"float32"),D=n.bufferSync(s);for(let $=0;$<c.batchSize;++$)for(let R=0;R<c.inChannels;++R)for(let N=0;N<c.inDepth;++N)for(let M=0;M<c.inHeight;++M)for(let B=0;B<c.inWidth;++B){let q=N-I,X=M-S,J=B-w,ee=0;for(let ae=0;ae<x;ae+=g){let se=(q+ae)/p;if(!(se<0||se>=c.outDepth||Math.floor(se)!==se))for(let oe=0;oe<b;oe+=y){let ne=(X+oe)/f;if(!(ne<0||ne>=c.outHeight||Math.floor(ne)!==ne))for(let ce=0;ce<v;ce+=A){let he=(J+ce)/m;if(he<0||he>=c.outWidth||Math.floor(he)!==he)continue;let me=x*b*v-1-h.get($,se,ne,he,R),be=ae*b*v+oe*v+ce,Ee=me===be?1:0;if(Ee===0)continue;ee+=D.get($,se,ne,he,R)*Ee}}}E.set(ee,$,N,M,B,R)}return n.makeTensorInfo(E.shape,E.dtype,E.values)}var Nue={kernelName:L1,backendName:"cpu",kernelFunc:Tue};function Cue(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,input:a,output:o}=t,i=a;Ne([a,o],"maxPoolGrad");let{filterSize:l,strides:u,pad:c,dimRoundingMode:d}=r,h=_.computePool2DInfo(i.shape,l,u,1,c,d),p=n.data.get(i.dataId).values,f=ze(h.outShape,i.dtype,kN(p,i.shape,i.dtype,h).values),m=h.strideHeight,g=h.strideWidth,y=h.dilationHeight,A=h.dilationWidth,x=h.effectiveFilterHeight,b=h.effectiveFilterWidth,v=b-1-h.padInfo.left,I=x-1-h.padInfo.top,w=ze(i.shape,"float32"),S=n.data.get(s.dataId).values,E=ze(s.shape,"float32",S);for(let D=0;D<h.batchSize;++D)for(let $=0;$<h.inChannels;++$)for(let R=0;R<h.inHeight;++R)for(let N=0;N<h.inWidth;++N){let M=R-I,B=N-v,q=0;for(let X=0;X<x;X+=y){let J=(M+X)/m;if(!(J<0||J>=h.outHeight||Math.floor(J)!==J))for(let ee=0;ee<b;ee+=A){let ae=(B+ee)/g;if(ae<0||ae>=h.outWidth||Math.floor(ae)!==ae)continue;let se=x*b-1-f.get(D,J,ae,$),oe=X*b+ee,ne=se===oe?1:0;if(ne===0)continue;q+=E.get(D,J,ae,$)*ne}}w.set(q,D,R,N,$)}return n.makeTensorInfo(w.shape,w.dtype,w.values)}var Eue={kernelName:z1,backendName:"cpu",kernelFunc:Cue};function $ue(e,t,n,r,s){let a=k.computeStrides(t),o=N5(e,t,n,a,s,"max"),i=kN(e,t,n,s,!0,r);return[o.values,i.values]}var Rue={kernelName:B1,backendName:"cpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{x:r}=e,{filterSize:s,strides:a,pad:o,includeBatchInIndex:i}=t,l=n;Ne(r,"MaxPoolWithArgmax");let u=l.data.get(r.dataId).values,c=_.computePool2DInfo(r.shape,s,a,[1,1],o),[d,h]=$ue(u,r.shape,r.dtype,i,c),p=l.write(d,c.outShape,r.dtype),f=l.write(h,c.outShape,r.dtype);return[{dataId:p,shape:c.outShape,dtype:r.dtype},{dataId:f,shape:c.outShape,dtype:"int32"}]}};function _ue(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r,i=k.parseAxisParam(a,s.shape),u=_.computeOutAndReduceShapes(s.shape,i)[1],c=k.sizeFromShape(u),d=[],h=n.makeTensorInfo([],"float32",new Float32Array([c]));d.push(h);let p=Ja({inputs:{x:s},backend:n,attrs:{dtype:"float32"}});d.push(p);let f=C5({inputs:{a:p,b:h},backend:n});d.push(f);let m=Qd({inputs:{x:f},backend:n,attrs:{axis:a,keepDims:o}});return d.forEach(g=>n.disposeIntermediateTensorInfo(g)),m}var Due={kernelName:Al,backendName:"cpu",kernelFunc:_ue};function Fue(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r;Ne(s,"min");let i=k.parseAxisParam(a,s.shape),l=i,u=_.getAxesPermutation(l,s.shape.length),c=s;u!=null&&(c=Dr({inputs:{x:s},backend:n,attrs:{perm:u}}),l=_.getInnerMostAxes(l.length,s.shape.length)),_.assertAxesAreInnerMostDims("min",l,c.shape.length);let[d,h]=_.computeOutAndReduceShapes(c.shape,l),p=k.sizeFromShape(h),f=k.makeZerosTypedArray(k.sizeFromShape(d),c.dtype),m=n.data.get(c.dataId).values;for(let y=0;y<f.length;++y){let A=y*p,x=m[A];for(let b=0;b<p;++b){let v=m[A+b];(Number.isNaN(v)||v<x)&&(x=v)}f[y]=x}u!=null&&n.disposeIntermediateTensorInfo(c);let g=n.makeTensorInfo(d,c.dtype,f);if(o){let y=_.expandShapeToKeepDim(d,i),A=_t({inputs:{x:g},backend:n,attrs:{shape:y}});return n.disposeIntermediateTensorInfo(g),A}return g}var Mue={kernelName:xl,backendName:"cpu",kernelFunc:Fue};function Oue(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{paddings:a,mode:o}=r;Ne(s,"mirrorPad");let i=a.map((x,b)=>x[0]+s.shape[b]+x[1]),l=a.map(x=>x[0]),u=a.map((x,b)=>x[0]+s.shape[b]),c=o==="reflect"?0:1,d=n.data.get(s.dataId).values,h=s.shape.length,p=k.computeStrides(s.shape),f=k.sizeFromShape(i),m=i.length,g=k.computeStrides(i),y=k.getTypedArrayFromDType(s.dtype,f);for(let x=0;x<f;x++){let b=k.indexToLoc(x,m,g);for(let I=0;I<m;I++)b[I]<l[I]?b[I]=l[I]*2-b[I]-c:b[I]>=u[I]&&(b[I]=(u[I]-1)*2-b[I]+c);b=b.map((I,w)=>I-l[w]);let v=k.locToIndex(b,h,p);y[x]=d[v]}return{dataId:n.write(y,i,s.dtype),shape:i,dtype:s.dtype}}var Pue={kernelName:bl,backendName:"cpu",kernelFunc:Oue},zue=en((e,t)=>{let n=e%t;return e<0&&t<0||e>=0&&t>=0?n:(n+t)%t}),Lue=xn(Uc,zue),Bue={kernelName:Uc,backendName:"cpu",kernelFunc:Lue},Wue=Xs(Zg());function EN(e){let{inputs:t,backend:n,attrs:r}=e,{logits:s}=t,{dim:a}=r,o=s.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=k.parseAxisParam([i],s.shape),u=CN({inputs:{x:s},backend:n,attrs:{reductionIndices:l,keepDims:!1}}),c=_.expandShapeToKeepDim(u.shape,l),d=_t({inputs:{x:u},backend:n,attrs:{shape:c}}),h=S5({inputs:{a:s,b:d},backend:n}),p=zT({inputs:{x:h},backend:n}),f=Qd({inputs:{x:p},backend:n,attrs:{axis:l,keepDims:!1}}),m=_t({inputs:{x:f},backend:n,attrs:{shape:c}}),g=C5({inputs:{a:p,b:m},backend:n});return n.disposeIntermediateTensorInfo(u),n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(f),n.disposeIntermediateTensorInfo(m),g}var Vue={kernelName:Ml,backendName:"cpu",kernelFunc:EN};function Uue(e){let{inputs:t,backend:n,attrs:r}=e,{logits:s}=t,{numSamples:a,seed:o,normalized:i}=r;Ne(s,"multinomial");let l=i?s:EN({inputs:{logits:s},backend:n,attrs:{dim:-1}}),u=l.shape[0],c=l.shape[1],d=n.data.get(l.dataId).values,h=[u,a],p=k.makeZerosTypedArray(k.sizeFromShape(h),"int32");for(let f=0;f<u;++f){let m=f*c,g=new Float32Array(c-1);g[0]=d[m];for(let x=1;x<g.length;++x)g[x]=g[x-1]+d[m+x];let y=Wue.alea(o.toString()),A=f*a;for(let x=0;x<a;++x){let b=y();p[A+x]=g.length;for(let v=0;v<g.length;v++)if(b<g[v]){p[A+x]=v;break}}}return i||n.disposeIntermediateTensorInfo(l),n.makeTensorInfo(h,"int32",p)}var Hue={kernelName:W1,backendName:"cpu",kernelFunc:Uue},Gue=da.nonMaxSuppressionV3Impl;function jue(e){let{inputs:t,backend:n,attrs:r}=e,{boxes:s,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l}=r;Ne(s,"NonMaxSuppression");let u=n.data.get(s.dataId).values,c=n.data.get(a.dataId).values,{selectedIndices:d}=Gue(u,c,o,i,l);return n.makeTensorInfo([d.length],"int32",new Int32Array(d))}var que={kernelName:Gc,backendName:"cpu",kernelFunc:jue},Kue=da.nonMaxSuppressionV4Impl;function Xue(e){let{inputs:t,backend:n,attrs:r}=e,{boxes:s,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,padToMaxOutputSize:u}=r;Ne(s,"NonMaxSuppressionPadded");let c=n.data.get(s.dataId).values,d=n.data.get(a.dataId).values,{selectedIndices:h,validOutputs:p}=Kue(c,d,o,i,l,u);return[n.makeTensorInfo([h.length],"int32",new Int32Array(h)),n.makeTensorInfo([],"int32",new Int32Array([p]))]}var Zue={kernelName:jc,backendName:"cpu",kernelFunc:Xue},Yue=da.nonMaxSuppressionV5Impl;function Jue(e){let{inputs:t,backend:n,attrs:r}=e,{boxes:s,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,softNmsSigma:u}=r;Ne(s,"NonMaxSuppressionWithScore");let c=n.data.get(s.dataId).values,d=n.data.get(a.dataId).values,h=o,p=i,f=l,m=u,{selectedIndices:g,selectedScores:y}=Yue(c,d,h,p,f,m);return[n.makeTensorInfo([g.length],"int32",new Int32Array(g)),n.makeTensorInfo([y.length],"float32",new Float32Array(y))]}var Que={kernelName:qc,backendName:"cpu",kernelFunc:Jue};function ece(e){let{inputs:t,backend:n,attrs:r}=e,{indices:s}=t,{depth:a,onValue:o,offValue:i}=r;Ne(s,"oneHot");let l=k.sizeFromShape(s.shape),u=new Float32Array(l*a);u.fill(i);let c=n.data.get(s.dataId).values;for(let d=0;d<l;++d)c[d]>=0&&c[d]<a&&(u[d*a+c[d]]=o);return n.makeTensorInfo([...s.shape,a],"int32",u)}var tce={kernelName:wl,backendName:"cpu",kernelFunc:ece};function Tm(e){let{inputs:t,backend:n}=e,{x:r}=t;if(r.dtype==="string")throw new Error("zerosLike is not supported for string tensors");if(r.dtype==="complex64"){let s=di({inputs:{input:r},backend:n}),a=Tm({inputs:{x:s},backend:n}),o=uu({inputs:{input:r},backend:n}),i=Tm({inputs:{x:o},backend:n}),l=pr({inputs:{real:a,imag:i},backend:n});return n.disposeIntermediateTensorInfo(s),n.disposeIntermediateTensorInfo(a),n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}else return R5({backend:n,attrs:{shape:r.shape,value:0,dtype:r.dtype}})}var nce={kernelName:hd,backendName:"cpu",kernelFunc:Tm};function $N(e){let{inputs:t,backend:n}=e,{x:r}=t;if(r.dtype==="string")throw new Error("onesLike is not supported for string tensors");if(r.dtype==="complex64"){let s=di({inputs:{input:r},backend:n}),a=$N({inputs:{x:s},backend:n}),o=uu({inputs:{input:r},backend:n}),i=Tm({inputs:{x:o},backend:n}),l=pr({inputs:{real:a,imag:i},backend:n});return n.disposeIntermediateTensorInfo(s),n.disposeIntermediateTensorInfo(a),n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}else return R5({backend:n,attrs:{shape:r.shape,value:1,dtype:r.dtype}})}var rce={kernelName:Kc,backendName:"cpu",kernelFunc:$N};function RN(e){let{inputs:t,backend:n,attrs:r}=e,{axis:s}=r;if(t.length===1)return Sm({inputs:{input:t[0]},backend:n,attrs:{dim:s}});let a=t[0].shape,o=t[0].dtype;t.forEach(c=>{k.assertShapesMatch(a,c.shape,"All tensors passed to stack must have matching shapes"),k.assert(o===c.dtype,()=>"All tensors passed to stack must have matching dtypes")});let i=[],l=t.map(c=>{let d=Sm({inputs:{input:c},backend:n,attrs:{dim:s}});return i.push(d),d}),u=cu({inputs:l,backend:n,attrs:{axis:s}});return i.forEach(c=>n.disposeIntermediateTensorInfo(c)),u}var sce={kernelName:Xc,backendName:"cpu",kernelFunc:RN};function ace(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{paddings:a,constantValue:o}=r;Ne(s,"pad");let i=a.map((A,x)=>A[0]+s.shape[x]+A[1]),l=a.map(A=>A[0]),u=n.data.get(s.dataId).values,c=k.sizeFromShape(s.shape),d=s.shape.length,h=k.computeStrides(s.shape),p=k.sizeFromShape(i),f=i.length,m=k.computeStrides(i),g=k.getTypedArrayFromDType(s.dtype,p);o!==0&&g.fill(o);for(let A=0;A<c;A++){let b=k.indexToLoc(A,d,h).map((I,w)=>I+l[w]),v=k.locToIndex(b,f,m);g[v]=u[A]}return{dataId:n.write(g,i,s.dtype),shape:i,dtype:s.dtype}}var _N={kernelName:kl,backendName:"cpu",kernelFunc:ace},oce=en((e,t)=>Math.pow(e,t)),ice=xn(Il,oce),lce={kernelName:Il,backendName:"cpu",kernelFunc:ice};function uce(e){let{backend:t,attrs:n}=e,{start:r,stop:s,dtype:a,step:o}=n,i=tN(r,s,o,a);return t.makeTensorInfo([i.length],a,i)}var cce={kernelName:rf,backendName:"cpu",kernelFunc:uce},dce=At(Yc,e=>1/e),hce={kernelName:Yc,backendName:"cpu",kernelFunc:dce};function pce(e){let{inputs:t,backend:n,attrs:r}=e,{images:s}=t,{alignCorners:a,halfPixelCenters:o,size:i}=r;Ne(s,"resizeBilinear");let l=k.computeStrides(s.shape),[u,c]=i,[d,h,p,f]=s.shape,m=n.data.get(s.dataId).values,g=new Float32Array(k.sizeFromShape([d,u,c,f])),y=[a&&u>1?h-1:h,a&&c>1?p-1:p],A=[a&&u>1?u-1:u,a&&c>1?c-1:c],x=0,b=y[0]/A[0],v=y[1]/A[1];for(let I=0;I<d;I++)for(let w=0;w<u;w++){let S;o?S=b*(w+.5)-.5:S=b*w;let E=Math.max(0,Math.floor(S)),D=S-E,$=Math.min(h-1,Math.ceil(S)),R=I*l[0]+E*l[1],N=I*l[0]+$*l[1];for(let M=0;M<c;M++){let B;o?B=v*(M+.5)-.5:B=v*M;let q=Math.max(0,Math.floor(B)),X=B-q,J=Math.min(p-1,Math.ceil(B)),ee=R+q*l[2],ae=N+q*l[2],se=R+J*l[2],oe=N+J*l[2];for(let ne=0;ne<f;ne++){let ce=m[ee+ne],he=m[ae+ne],me=m[se+ne],be=m[oe+ne],Ee=ce+(me-ce)*X,$e=he+(be-he)*X,Pe=Ee+($e-Ee)*D;g[x++]=Pe}}}return n.makeTensorInfo([d,u,c,f],"float32",g)}var fce={kernelName:Nl,backendName:"cpu",kernelFunc:pce};function mce(e){let{inputs:t,backend:n,attrs:r}=e,{images:s,dy:a}=t,{alignCorners:o}=r;Ne([a,s],"resizeBilinearGrad");let i=k.computeStrides(s.shape),[l,u,c,d]=s.shape,[,h,p]=a.shape,f=new Float32Array(l*u*c*d),m=[o&&h>1?u-1:u,o&&p>1?c-1:c],g=[o&&h>1?h-1:h,o&&p>1?p-1:p],y=m[0]/g[0],A=m[1]/g[1],x=n.data.get(a.dataId).values,b=0;for(let v=0;v<l;v++){let I=v*i[0];for(let w=0;w<h;w++){let S=w*y,E=Math.floor(S),D=Math.min(Math.ceil(S),u-1),$=I+E*i[1],R=I+D*i[1],N=S-E,M=1-N;for(let B=0;B<p;B++){let q=B*A,X=Math.floor(q),J=Math.min(Math.ceil(q),c-1),ee=q-X,ae=1-ee,se=$+X*i[2],oe=$+J*i[2],ne=R+X*i[2],ce=R+J*i[2],he=M*ae,me=M*ee,be=N*ae,Ee=N*ee;for(let $e=0;$e<d;$e++){let Pe=x[b++];f[se+$e]+=Pe*he,f[oe+$e]+=Pe*me,f[ne+$e]+=Pe*be,f[ce+$e]+=Pe*Ee}}}}return n.makeTensorInfo([l,c,u,d],"float32",f)}var gce={kernelName:H1,backendName:"cpu",kernelFunc:mce};function yce(e){let{inputs:t,backend:n,attrs:r}=e,{images:s}=t,{alignCorners:a,halfPixelCenters:o,size:i}=r;Ne(s,"resizeNearestNeighbor");let l=k.computeStrides(s.shape),[u,c]=i,[d,h,p,f]=s.shape,m=n.data.get(s.dataId).values,g=new Float32Array(d*u*c*f),y=[a&&u>1?h-1:h,a&&c>1?p-1:p],A=[a&&u>1?u-1:u,a&&c>1?c-1:c],x=y[0]/A[0],b=y[1]/A[1],v=0;for(let I=0;I<d;I++){let w=I*l[0];for(let S=0;S<u;S++){let E=o?x*(S+.5):x*S,D=Math.min(h-1,a?Math.round(E):Math.floor(E));o&&(D=Math.max(0,D));let $=w+D*l[1];for(let R=0;R<c;R++){let N=o?b*(R+.5):b*R,M=Math.min(p-1,a?Math.round(N):Math.floor(N));o&&(M=Math.max(0,M));let B=$+M*l[2];for(let q=0;q<f;q++){let X=m[B+q];g[v++]=X}}}}return n.makeTensorInfo([d,u,c,f],s.dtype,g)}var Ace={kernelName:sf,backendName:"cpu",kernelFunc:yce};function xce(e){let{inputs:t,backend:n,attrs:r}=e,{images:s,dy:a}=t,{alignCorners:o}=r;Ne([a,s],"resizeNearestNeighborGrad");let i=k.computeStrides(s.shape),l=k.computeStrides(a.shape),[u,c,d,h]=s.shape,[,p,f]=a.shape,m=new Float32Array(u*c*d*h),g=n.data.get(a.dataId).values,y=[o&&p>1?c-1:c,o&&f>1?d-1:d],A=[o&&p>1?p-1:p,o&&f>1?f-1:f],x=y[0]/A[0],b=y[1]/A[1],v=1/x,I=1/b,w=Math.ceil(v)*2+2,S=Math.ceil(I)*2+2;for(let E=0;E<u;E++){let D=E*i[0];for(let $=0;$<c;$++){let R=D+$*i[1],N=Math.floor($*v),M=Math.floor(N-w/2);for(let B=0;B<d;B++){let q=R+B*i[2],X=Math.floor(B*I),J=Math.floor(X-S/2);for(let ee=0;ee<h;ee++){let ae=0;for(let se=0;se<w;se++){let oe=se+M;if(oe<0||oe>=p)continue;let ne=D+oe*l[1],ce=oe*x,he=Math.min(c-1,o?Math.round(ce):Math.floor(ce));if($===he)for(let me=0;me<S;me++){let be=me+J;if(be<0||be>=f)continue;let Ee=ne+be*l[2],$e=be*b,Pe=Math.min(d-1,o?Math.round($e):Math.floor($e));B===Pe&&(ae+=g[Ee+ee])}}m[q+ee]=ae}}}}return n.makeTensorInfo(s.shape,s.dtype,m)}var bce={kernelName:U1,backendName:"cpu",kernelFunc:xce};function vce(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{dims:a}=r;Ne(s,"reverse");let o=s.shape.length,i=k.parseAxisParam(a,s.shape);if(o===0)return Ls({inputs:{x:s},backend:n});let l=new Jt(s.shape,s.dtype),u=n.bufferSync(s);for(let c=0;c<l.size;c++){let d=l.indexToLoc(c),h=d.slice();i.forEach(p=>h[p]=s.shape[p]-1-h[p]),l.set(u.get(...h),...d)}return n.makeTensorInfo(l.shape,l.dtype,l.values)}var wce={kernelName:El,backendName:"cpu",kernelFunc:vce},kce={kernelName:pd,backendName:"cpu",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{image:r}=e,{radians:s,fillValue:a,center:o}=t,i=n,l=k.getTypedArrayFromDType(r.dtype,k.sizeFromShape(r.shape)),[u,c,d,h]=r.shape,[p,f]=_.getImageCenter(o,c,d),m=255,g=Math.sin(s),y=Math.cos(s),A=i.data.get(r.dataId).values;for(let b=0;b<u;b++){let v=b*d*c*h;for(let I=0;I<c;I++){let w=I*(d*h);for(let S=0;S<d;S++){let E=S*h;for(let D=0;D<h;D++){let $=[u,I,S,D],R=$[2],N=$[1],M=(R-p)*y-(N-f)*g,B=(R-p)*g+(N-f)*y;M=Math.round(M+p),B=Math.round(B+f);let q=a;if(typeof a!="number"&&(D===3?q=m:q=a[D]),M>=0&&M<d&&B>=0&&B<c){let J=B*(d*h),ee=M*h,ae=v+J+ee+D;q=A[ae]}let X=v+w+E+D;l[X]=q}}}}return{dataId:i.write(l,r.shape,r.dtype),shape:r.shape,dtype:r.dtype}}},Ice=At($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}),Sce={kernelName:$l,backendName:"cpu",kernelFunc:Ice};function DN(e,t,n,r,s,a,o,i,l,u){let c=[r/s,s],d=e.values,h=t.values;if(r===0)return ze(n,t.dtype);let p=ze(c,t.dtype);p.values.fill(l);for(let f=0;f<a;f++){let m=[],g=0;for(let y=0;y<o;y++){let A=d[f*o+y];m.push(A),g+=A*i[y]}if(g<0||g>=r/s)throw new Error(`Invalid indices: ${m} does not index into ${n}`);for(let y=0;y<s;y++)u?p.values[g*s+y]+=h[f*s+y]:p.values[g*s+y]=t.rank===0?h[0]:h[f*s+y]}return p}function Tce(e){let{inputs:t,backend:n,attrs:r}=e,{indices:s,updates:a}=t,{shape:o}=r,{sliceRank:i,numUpdates:l,sliceSize:u,strides:c,outputSize:d}=_.calculateShapes(a,s,o),h=!0,p=n.bufferSync(s),f=n.bufferSync(a),m=DN(p,f,o,d,u,l,i,c,0,h);return n.makeTensorInfo(o,m.dtype,m.values)}var Nce={kernelName:Qc,backendName:"cpu",kernelFunc:Tce};function Cce(e){let{inputs:t,backend:n}=e,{condition:r,t:s,e:a}=t;Ne([r,s,a],"select");let o=r.shape.length,i=n.data.get(r.dataId).values,l=n.data.get(s.dataId).values,u=n.data.get(a.dataId).values,c=Ur(s.dtype,a.dtype),d=k.makeZerosTypedArray(k.sizeFromShape(s.shape),c),h=0,p=o===0||o>1||s.shape.length===1?1:k.sizeFromShape(s.shape.slice(1));for(let f=0;f<i.length;f++)for(let m=0;m<p;m++)i[f]===1?d[h++]=l[f]:d[h++]=u[f];return n.makeTensorInfo(s.shape,c,d)}var Ece={kernelName:ed,backendName:"cpu",kernelFunc:Cce},$ce=_.SELU_SCALEALPHA,Rce=_.SELU_SCALE,_ce=At(td,e=>e>=0?Rce*e:$ce*(Math.exp(e)-1)),Dce={kernelName:td,backendName:"cpu",kernelFunc:_ce},Fce=At(sd,e=>e<0?-1:e>0?1:0),Mce={kernelName:sd,backendName:"cpu",kernelFunc:Fce},Oce=At(Rl,e=>Math.sin(e)),Pce={kernelName:Rl,backendName:"cpu",kernelFunc:Oce},zce=At(rd,e=>Math.sinh(e)),Lce={kernelName:rd,backendName:"cpu",kernelFunc:zce},Bce=11920928955078125e-23,FN=Math.log(Bce)+2,Wce=At(ad,e=>{let t=e>-FN,n=e<FN,r=Math.exp(e),s;return n?s=r:t?s=e:s=Math.log(1+r),s}),Vce={kernelName:ad,backendName:"cpu",kernelFunc:Wce};function Uce(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{blockShape:a,paddings:o}=r;Ne([s],"spaceToBatchND");let i=k.sizeFromShape(a),l=[[0,0]];l.push(...o);for(let I=1+a.length;I<s.shape.length;++I)l.push([0,0]);let u=_N.kernelFunc({inputs:{x:s},backend:n,attrs:{paddings:l,constantValue:0}}),c=_.getReshaped(u.shape,a,i,!1),d=_.getPermuted(c.length,a.length,!1),h=_.getReshapedPermuted(u.shape,a,i,!1),m=_t({inputs:{x:u},backend:n,attrs:{shape:c}}),A=Dr({inputs:{x:m},backend:n,attrs:{perm:d}}),v=_t({inputs:{x:A},backend:n,attrs:{shape:h}});return n.disposeIntermediateTensorInfo(u),n.disposeIntermediateTensorInfo(m),n.disposeIntermediateTensorInfo(A),v}var Hce={kernelName:od,backendName:"cpu",kernelFunc:Uce};function Gce(e){let{inputs:t,backend:n}=e,{indices:r,values:s,denseShape:a,defaultValue:o}=t;if(a.shape.length!==1)throw new Error(`Dense shape must be a vector, saw:
${a.shape}`);if(r.shape.length!==2)throw new Error(`Indices must be a matrix, saw:
${r.shape}`);if(s.shape.length!==1)throw new Error(`Values must be a vector, saw:
${s.shape}`);if(o.shape.length!==0)throw new Error(`Default value must be a scalar, saw:
${o.shape}`);let i=n.data.get(r.dataId).values,l=n.data.get(s.dataId).values,u=n.data.get(a.dataId).values,c=n.data.get(o.dataId).values[0],[d,h,p,f,m]=sN(i,r.shape,r.dtype,l,s.dtype,u,c);return[n.makeTensorInfo(h,r.dtype,d),n.makeTensorInfo([h[0]],s.dtype,p),n.makeTensorInfo([f.length],"bool",new Uint8Array(f.map(g=>Number(g)))),n.makeTensorInfo([m.length],r.dtype,new Int32Array(m))]}var jce={kernelName:G1,backendName:"cpu",kernelFunc:Gce};function qce(e){let{inputs:t,backend:n}=e,{inputIndices:r,inputShape:s,newShape:a}=t;if(r.shape.length!==2)throw new Error(`Input indices should be a matrix but received shape
${r.shape}`);if(s.shape.length!==1)throw new Error(`Input shape should be a vector but received shape
${s.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(s.dataId).values),i=n.data.get(r.dataId).values,l=Array.from(n.data.get(a.dataId).values),[u,c,d]=aN(i,r.shape,r.dtype,o,l);return[n.makeTensorInfo(c,r.dtype,u),n.makeTensorInfo([d.length],a.dtype,new Int32Array(d))]}var Kce={kernelName:j1,backendName:"cpu",kernelFunc:qce};function Xce(e){let{inputs:t,backend:n}=e,{data:r,indices:s,segmentIds:a}=t;if(r.shape.length<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(s.shape.length!==1)throw new Error(`Indices should be a vector but received shape
${s.shape}`);if(a.shape.length!==1)throw new Error(`Segment ids should be a vector but received shape
${a.shape}`);let o=n.data.get(r.dataId).values,i=n.data.get(s.dataId).values,l=n.data.get(a.dataId).values,[u,c]=I5(o,r.shape,r.dtype,i,l,!0);return n.makeTensorInfo(c,r.dtype,u)}var Zce={kernelName:q1,backendName:"cpu",kernelFunc:Xce};function Yce(e){let{inputs:t,backend:n}=e,{data:r,indices:s,segmentIds:a}=t;if(r.shape.length<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(s.shape.length!==1)throw new Error(`Indices should be a vector but received shape
${s.shape}`);if(a.shape.length!==1)throw new Error(`Segment ids should be a vector but received shape
${a.shape}`);let o=n.data.get(r.dataId).values,i=n.data.get(s.dataId).values,l=n.data.get(a.dataId).values,[u,c]=I5(o,r.shape,r.dtype,i,l);return n.makeTensorInfo(c,r.dtype,u)}var Jce={kernelName:K1,backendName:"cpu",kernelFunc:Yce};function Qce(e){let{inputs:t,backend:n,attrs:r}=e,{sparseIndices:s,sparseValues:a,defaultValue:o}=t,{outputShape:i}=r,{sliceRank:l,numUpdates:u,sliceSize:c,strides:d,outputSize:h}=_.calculateShapes(a,s,i),p=!1,f=n.bufferSync(s),m=n.bufferSync(a),g=n.data.get(o.dataId).values[0],y=DN(f,m,i,h,c,u,l,d,g,p);return n.makeTensorInfo(i,y.dtype,y.values)}var ede={kernelName:X1,backendName:"cpu",kernelFunc:Qce};function tde(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{numOrSizeSplits:a,axis:o}=r,i=k.parseAxisParam(o,s.shape)[0],l=_.prepareSplitSize(s,a,i),u=new Array(s.shape.length).fill(0),c=s.shape.slice();return l.map(d=>{let h=[...c];h[i]=d;let p=hi({inputs:{x:s},backend:n,attrs:{begin:u,size:h}});return u[i]+=d,p})}var nde={kernelName:id,backendName:"cpu",kernelFunc:tde},rde=At(Dl,e=>Math.sqrt(e)),sde={kernelName:Dl,backendName:"cpu",kernelFunc:rde},ade={kernelName:af,backendName:"cpu",kernelFunc:({inputs:e,backend:t})=>{let{x:n}=e,r=t;Ne(n,"square");let s=r.data.get(n.dataId).values,a=new Float32Array(s.length);for(let i=0;i<s.length;++i){let l=s[i];a[i]=l*l}return{dataId:r.write(a,n.shape,n.dtype),shape:n.shape,dtype:n.dtype}}},ode=At(Bo,(e,t)=>{let n=t;return isNaN(e)?NaN:e>0?1:n.alpha}),ide={kernelName:Bo,backendName:"cpu",kernelFunc:ode};function lde(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{begin:a,end:o,strides:i,beginMask:l,endMask:u,ellipsisMask:c,newAxisMask:d,shrinkAxisMask:h}=r;Ne(s,"stridedSlice");let{nonStrided:p,$begin:f,$strides:m,size:g,newShape:y,outShape:A}=Cn.sliceInfo(s.shape,a,o,i,l,u,c,d,h),x=_t({inputs:{x:s},backend:n,attrs:{shape:y}}),b;if(p){let I=hi({inputs:{x},backend:n,attrs:{begin:f,size:g}});b=_t({inputs:{x:I},backend:n,attrs:{shape:A}}),n.disposeIntermediateTensorInfo(I)}else if(A.some(I=>I===0))b=n.makeTensorInfo(A,s.dtype,[]);else{let I=n.bufferSync(x),w=iN(A,I,m,f);b=n.makeTensorInfo(w.shape,w.dtype,w.values)}let v=_t({inputs:{x:b},backend:n,attrs:{shape:A}});return n.disposeIntermediateTensorInfo(x),n.disposeIntermediateTensorInfo(b),v}var ude={kernelName:ld,backendName:"cpu",kernelFunc:lde};function cde(e){let{inputs:t,backend:n,attrs:r}=e,{separator:s,nGramWidths:a,leftPad:o,rightPad:i,padWidth:l,preserveShortSequences:u}=r,{data:c,dataSplits:d}=t,h=n.data.get(c.dataId).values,p=n.data.get(d.dataId).values,[f,m]=lN(h,p,s,a,o,i,l,u);return[n.makeTensorInfo([f.length],"string",f),n.makeTensorInfo(d.shape,"int32",m)]}var dde={kernelName:Z1,backendName:"cpu",kernelFunc:cde};function hde(e){let{inputs:t,backend:n,attrs:r}=e,{skipEmpty:s}=r,{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,d]=uN(i,l,s),h=c.length;return[n.makeTensorInfo([h,2],"int32",u),n.makeTensorInfo([h],"string",c),n.makeTensorInfo([2],"int32",new Int32Array(d))]}var pde={kernelName:Y1,backendName:"cpu",kernelFunc:hde};function fde(e){let{inputs:t,backend:n,attrs:r}=e,{numBuckets:s}=r,{input:a}=t;if(a.dtype!=="string")throw new Error("Input must be of datatype string");if(s<=0)throw new Error("Number of buckets must be at least 1");let o=n.data.get(a.dataId).values,i=cN(o,s);return n.makeTensorInfo(a.shape,"int32",i)}var mde={kernelName:J1,backendName:"cpu",kernelFunc:fde},gde=At(Ol,e=>Math.tan(e)),yde={kernelName:Ol,backendName:"cpu",kernelFunc:gde},Ade=At(Pl,e=>Math.tanh(e)),xde={kernelName:Pl,backendName:"cpu",kernelFunc:Ade};function bde(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{reps:a}=r;Ne(s,"tile");let o=hN(n.bufferSync(s),a);return n.makeTensorInfo(o.shape,o.dtype,o.values)}var vde={kernelName:Lo,backendName:"cpu",kernelFunc:bde};function wde(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{k:a,sorted:o}=r;Ne(s,"topk");let i=n.data.get(s.dataId).values,[l,u]=fN(i,s.shape,s.dtype,a,o);return[n.makeTensorInfo(l.shape,l.dtype,l.values),n.makeTensorInfo(u.shape,u.dtype,u.values)]}var kde={kernelName:ud,backendName:"cpu",kernelFunc:wde};function Ide(e){let{inputs:t,attrs:n,backend:r}=e,{image:s,transforms:a}=t,{interpolation:o,fillMode:i,fillValue:l,outputShape:u}=n,[c,d,h,p]=s.shape,[f,m]=u!=null?u:[d,h],g=[c,f,m,p],y=k.computeStrides(s.shape),A=y[0],x=y[1],b=y[2],v=k.getTypedArrayFromDType(s.dtype,k.sizeFromShape(g));v.fill(l);let I=r.data.get(s.dataId).values,w=r.data.get(a.dataId).values;for(let E=0;E<c;++E){let D=a.shape[0]===1?w:w.subarray(E*8,E*8+8);for(let $=0;$<f;++$)for(let R=0;R<m;++R)for(let N=0;N<p;++N){let M,B=D[6]*R+D[7]*$+1;if(B===0)continue;let q=(D[0]*R+D[1]*$+D[2])/B,X=(D[3]*R+D[4]*$+D[5])/B,J=MN(q,h,i),ee=MN(X,d,i);switch(o){case"nearest":M=$de(I,d,h,A,x,b,E,ee,J,N,l);break;case"bilinear":M=Rde(I,d,h,A,x,b,E,ee,J,N,l);break;default:throw new Error(`Error in Transform: Expect 'nearest' or 'bilinear', but got ${o}`)}let ae=E*A+$*x+R*b+N;v[ae]=M}return r.makeTensorInfo(g,s.dtype,v)}return{dataId:r.write(v,g,s.dtype),shape:s.shape,dtype:s.dtype}}var Sde={kernelName:cd,backendName:"cpu",kernelFunc:Ide};function MN(e,t,n){switch(n){case"reflect":return Tde(e,t);case"wrap":return Nde(e,t);case"nearest":return Ede(e,t);case"constant":default:return Cde(e,t)}}function Tde(e,t){let n=e;if(n<0)if(t<=1)n=0;else{let r=2*t;n<r&&(n=r*Math.trunc(-n/r)+n),n=n<-t?n+r:-n-1}else if(n>t-1)if(t<=1)n=0;else{let r=2*t;n-=r*Math.trunc(n/r),n>=t&&(n=r-n-1)}return k.clamp(0,n,t-1)}function Nde(e,t){let n=e;if(n<0)if(t<=1)n=0;else{let r=t-1;n+=t*(Math.trunc(-n/r)+1)}else if(n>t-1)if(t<=1)n=0;else{let r=t-1;n-=t*Math.trunc(n/r)}return k.clamp(0,n,t-1)}function Cde(e,t){return e}function Ede(e,t){return k.clamp(0,e,t-1)}function eh(e,t,n,r,s,a,o,i,l,u,c){let d=o*r+i*s+l*a+u;return 0<=i&&i<t&&0<=l&&l<n?e[d]:c}function $de(e,t,n,r,s,a,o,i,l,u,c){let d=Math.round(i),h=Math.round(l);return eh(e,t,n,r,s,a,o,d,h,u,c)}function Rde(e,t,n,r,s,a,o,i,l,u,c){let d=Math.floor(i),h=Math.floor(l),p=d+1,f=h+1,m=(f-l)*eh(e,t,n,r,s,a,o,d,h,u,c)+(l-h)*eh(e,t,n,r,s,a,o,d,f,u,c),g=(f-l)*eh(e,t,n,r,s,a,o,p,h,u,c)+(l-h)*eh(e,t,n,r,s,a,o,p,f,u,c);return(p-i)*m+(i-d)*g}function _de(e){let{inputs:t,attrs:n,backend:r}=e,{axis:s}=n,{x:a}=t;Ne(a,"unique");let o=r.data.get(a.dataId).values,{outputValues:i,outputShape:l,indices:u}=mN(o,s,a.shape,a.dtype);return[r.makeTensorInfo(l,a.dtype,i),r.makeTensorInfo([u.length],"int32",u)]}var Dde={kernelName:Q1,backendName:"cpu",kernelFunc:_de};function Fde(e){let{inputs:t,backend:n,attrs:r}=e,{value:s}=t,{axis:a}=r;a<0&&(a+=s.shape.length);let o=s.shape.length,i=s.shape[a],l=new Array(o-1),u=0;for(let p=0;p<o;p++)p!==a&&(l[u++]=s.shape[p]);let c=new Array(o).fill(0),d=s.shape.slice();d[a]=1;let h=new Array(i);for(let p=0;p<h.length;p++){c[a]=p;let f=hi({inputs:{x:s},backend:n,attrs:{begin:c,size:d}});h[p]=_t({inputs:{x:f},backend:n,attrs:{shape:l}}),n.disposeIntermediateTensorInfo(f)}return h}var Mde={kernelName:dd,backendName:"cpu",kernelFunc:Fde};function Ode(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,segmentIds:a}=t,{numSegments:o}=r;Ne(s,"unsortedSegmentSum");let i=s.shape.length,l=a.shape.length,u=[],c=[],d=i-l,h=a;for(let f=0;f<d;++f){let m=Sm({inputs:{input:h},backend:n,attrs:{dim:f+1}});h=m,c.push(m)}for(let f=0;f<o;++f){let m=k.createScalarValue(f,"int32"),g=n.makeTensorInfo([],"int32",m),y=OT({inputs:{a:g,b:h},backend:n}),A=Ja({inputs:{x:y},backend:n,attrs:{dtype:"float32"}}),x=Im({inputs:{a:A,b:s},backend:n}),b=Qd({inputs:{x},backend:n,attrs:{axis:0,keepDims:!1}});u.push(b),c.push(g),c.push(y),c.push(A),c.push(x),c.push(b)}let p=RN({inputs:u,backend:n,attrs:{axis:0}});return c.forEach(f=>n.disposeIntermediateTensorInfo(f)),p}var Pde={kernelName:of,backendName:"cpu",kernelFunc:Ode},zde=[Goe,Wae,qoe,Xoe,qae,Yoe,Qoe,tie,rie,aie,iie,uie,die,fie,gie,xie,vie,kie,Sie,Uoe,Nie,Eie,Rie,Gae,Xae,Die,Vae,Mie,Pie,Bie,Vie,zie,jie,Kie,Hie,Zie,Jie,ele,nle,sle,ole,ile,ule,dle,ple,fle,gle,mle,E5,xle,Moe,vle,Zae,Ele,Yae,$le,Qae,Ole,Ple,Lle,toe,Vle,Hle,jle,Kle,Zle,roe,aoe,Uae,Jle,Oie,eue,nue,sue,Ooe,ioe,uoe,oue,doe,lue,due,pue,gue,Aue,bue,poe,kue,Sue,Nue,Eue,Rue,vue,Due,Mue,moe,Pue,Bue,Hue,yoe,xoe,que,Zue,Que,voe,tce,rce,sce,_N,lce,zoe,Ioe,cce,Hae,hce,Loe,Boe,Voe,fce,gce,Ace,bce,wce,kce,Sce,Toe,Nce,Ece,Dce,Woe,Mce,Pce,Lce,Noe,Vue,Vce,Hce,jce,Kce,Zce,Jce,ede,nde,sde,ade,Eoe,ide,ude,dde,pde,mde,Doe,yle,yde,xde,vde,kde,woe,Sde,Dde,Mde,Pde,nce];for(let e of zde)ry(e);var ON={};_e(ON,{assertNotComplex:()=>hu,bindCanvasToFramebuffer:()=>Zde,bindColorTextureToFramebuffer:()=>Em,bindTextureToProgramUniformSampler:()=>YN,bindTextureUnit:()=>KN,bindVertexBufferToProgramAttribute:()=>F5,callAndCheck:()=>Ie,canBeRepresented:()=>PN,createFragmentShader:()=>BN,createFramebuffer:()=>qN,createProgram:()=>WN,createStaticIndexBuffer:()=>HN,createStaticVertexBuffer:()=>UN,createTexture:()=>GN,createVertexShader:()=>LN,getBatchDim:()=>fi,getExtensionOrThrow:()=>sh,getFramebufferErrorMessage:()=>JN,getMaxTexturesInShader:()=>nC,getNumChannels:()=>Kde,getProgramUniformLocation:()=>ZN,getProgramUniformLocationOrThrow:()=>XN,getRowsCols:()=>mi,getShapeAs3D:()=>$m,getTextureShapeFromLogicalShape:()=>eC,getWebGLDisjointQueryTimerVersion:()=>rC,getWebGLErrorMessage:()=>zN,getWebGLMaxTextureSize:()=>tC,hasExtension:()=>Mr,isCapableOfRenderingToFloatTexture:()=>sC,isDownloadFloatTextureEnabled:()=>aC,isReshapeFree:()=>oh,isWebGLFenceEnabled:()=>oC,isWebGLVersionEnabled:()=>O5,linkProgram:()=>VN,resetMaxTextureSize:()=>Yde,resetMaxTexturesInShader:()=>Jde,unbindColorTextureFromFramebuffer:()=>M5,unbindTextureUnit:()=>Xde,validateFramebuffer:()=>ah,validateProgram:()=>Cm,validateTextureSize:()=>jN});var pi={},_5={alpha:!1,antialias:!1,premultipliedAlpha:!1,preserveDrawingBuffer:!1,depth:!1,stencil:!1,failIfMajorPerformanceCaveat:!0};function Nm(e,t){pi[e]=t}function Bs(e){if(!(e in pi)){let n=Bde(e);if(n!==null)pi[e]=n;else return console.log("Could not get context for WebGL version",e),null}let t=pi[e];return t.isContextLost()?(delete pi[e],Bs(e)):(t.disable(t.DEPTH_TEST),t.disable(t.STENCIL_TEST),t.disable(t.BLEND),t.disable(t.DITHER),t.disable(t.POLYGON_OFFSET_FILL),t.disable(t.SAMPLE_COVERAGE),t.enable(t.SCISSOR_TEST),t.enable(t.CULL_FACE),t.cullFace(t.BACK),pi[e])}function Lde(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 Bde(e){if(e!==1&&e!==2)throw new Error("Cannot get WebGL rendering context, WebGL is disabled.");let t=Lde(e);return t.addEventListener("webglcontextlost",n=>{n.preventDefault(),delete pi[e]},!1),e===1?t.getContext("webgl",_5)||t.getContext("experimental-webgl",_5):t.getContext("webgl2",_5)}var th;(function(e){e[e.DENSE=0]="DENSE",e[e.SHARED_BATCH=1]="SHARED_BATCH"})(th||(th={}));var Fr;(function(e){e[e.RENDER=0]="RENDER",e[e.UPLOAD=1]="UPLOAD",e[e.PIXELS=2]="PIXELS",e[e.DOWNLOAD=3]="DOWNLOAD"})(Fr||(Fr={}));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 nh(e,t){return[t,e]}function Wde(e,t){return e*t}function rh(e){let t=k.sizeFromShape(e),n=Math.ceil(t/4);return k.sizeToSquarishShape(n)}function du(e,t){return[Math.max(1,Math.ceil(t/2)),Math.max(1,Math.ceil(e/2))]}function Vde(e,t){let[n,r]=du(e,t);return n*r*4}function D5(e,t){let n=e,r,s,a,o,i,l,u,c,d,h;return re().getNumber("WEBGL_VERSION")===2?(r=n.R32F,s=n.R16F,a=n.RGBA16F,o=n.RGBA32F,i=n.RED,u=4,c=1,d=n.HALF_FLOAT,h=n.FLOAT):(r=e.RGBA,s=e.RGBA,a=e.RGBA,o=n.RGBA,i=e.RGBA,u=4,c=4,d=t!=null?t.HALF_FLOAT_OES:null,h=e.FLOAT),l=e.RGBA,{internalFormatFloat:r,internalFormatHalfFloat:s,internalFormatPackedHalfFloat:a,internalFormatPackedFloat:o,textureFormatFloat:i,downloadTextureFormat:l,downloadUnpackNumChannels:u,defaultNumChannels:c,textureTypeHalfFloat:d,textureTypeFloat:h}}function Ie(e,t){let n=t();return re().getBool("DEBUG")&&Ude(e),n}function Ude(e){let t=e.getError();if(t!==e.NO_ERROR)throw new Error("WebGL Error: "+zN(e,t))}var Hde=596e-10,Gde=65504;function PN(e){return!!(re().getBool("WEBGL_RENDER_FLOAT32_ENABLED")||e===0||Hde<Math.abs(e)&&Math.abs(e)<Gde)}function zN(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 sh(e,t){return Aa(e,()=>e.getExtension(t),'Extension "'+t+'" not supported on this browser.')}function LN(e,t){let n=Aa(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 BN(e,t){let n=Aa(e,()=>e.createShader(e.FRAGMENT_SHADER),"Unable to create fragment WebGLShader.");if(Ie(e,()=>e.shaderSource(n,t)),Ie(e,()=>e.compileShader(n)),e.getShaderParameter(n,e.COMPILE_STATUS)===!1)throw qde(t,e.getShaderInfoLog(n)),new Error("Failed to compile fragment shader.");return n}var jde=/ERROR: [0-9]+:([0-9]+):/g;function qde(e,t){let n=jde.exec(t);if(n==null){console.log(`Couldn't parse line number in error: ${t}`),console.log(e);return}let r=+n[1],s=e.split(`
`),a=s.length.toString().length+2,o=s.map((d,h)=>k.rightPad((h+1).toString(),a)+d),i=0;for(let d=0;d<o.length;d++)i=Math.max(o[d].length,i);let l=o.slice(0,r-1),u=o.slice(r-1,r),c=o.slice(r);console.log(l.join(`
`)),console.log(t.split(`
`)[0]),console.log(`%c ${k.rightPad(u[0],i)}`,"border:1px solid red; background-color:#e3d2d2; color:#a61717"),console.log(c.join(`
`))}function WN(e){return Aa(e,()=>e.createProgram(),"Unable to create WebGLProgram.")}function VN(e,t){if(Ie(e,()=>e.linkProgram(t)),e.getProgramParameter(t,e.LINK_STATUS)===!1)throw console.log(e.getProgramInfoLog(t)),new Error("Failed to link vertex and fragment shaders.")}function Cm(e,t){if(Ie(e,()=>e.validateProgram(t)),e.getProgramParameter(t,e.VALIDATE_STATUS)===!1)throw console.log(e.getProgramInfoLog(t)),new Error("Shader program validation failed.")}function UN(e,t){let n=Aa(e,()=>e.createBuffer(),"Unable to create WebGLBuffer");return Ie(e,()=>e.bindBuffer(e.ARRAY_BUFFER,n)),Ie(e,()=>e.bufferData(e.ARRAY_BUFFER,t,e.STATIC_DRAW)),n}function HN(e,t){let n=Aa(e,()=>e.createBuffer(),"Unable to create WebGLBuffer");return Ie(e,()=>e.bindBuffer(e.ELEMENT_ARRAY_BUFFER,n)),Ie(e,()=>e.bufferData(e.ELEMENT_ARRAY_BUFFER,t,e.STATIC_DRAW)),n}function Kde(){return re().getNumber("WEBGL_VERSION")===2?1:4}function GN(e){return Aa(e,()=>e.createTexture(),"Unable to create WebGLTexture.")}function jN(e,t){let n=re().getNumber("WEBGL_MAX_TEXTURE_SIZE");if(e<=0||t<=0){let r=`[${e}x${t}]`;throw new Error("Requested texture size "+r+" is invalid.")}if(e>n||t>n){let r=`[${e}x${t}]`,s=`[${n}x${n}]`;throw new Error("Requested texture size "+r+" greater than WebGL maximum on this browser / GPU "+s+".")}}function qN(e){return Aa(e,()=>e.createFramebuffer(),"Unable to create WebGLFramebuffer.")}function F5(e,t,n,r,s,a,o){let i=e.getAttribLocation(t,n);return i===-1?!1:(Ie(e,()=>e.bindBuffer(e.ARRAY_BUFFER,r)),Ie(e,()=>e.vertexAttribPointer(i,s,e.FLOAT,!1,a,o)),Ie(e,()=>e.enableVertexAttribArray(i)),!0)}function KN(e,t,n){QN(e,n),Ie(e,()=>e.activeTexture(e.TEXTURE0+n)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,t))}function Xde(e,t){QN(e,t),Ie(e,()=>e.activeTexture(e.TEXTURE0+t)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,null))}function XN(e,t,n){return Aa(e,()=>e.getUniformLocation(t,n),'uniform "'+n+'" not present in program.')}function ZN(e,t,n){return e.getUniformLocation(t,n)}function YN(e,t,n,r){Ie(e,()=>KN(e,t,r)),Ie(e,()=>e.uniform1i(n,r))}function Zde(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 Em(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 M5(e,t){Ie(e,()=>e.bindFramebuffer(e.FRAMEBUFFER,t)),Ie(e,()=>e.framebufferTexture2D(e.FRAMEBUFFER,e.COLOR_ATTACHMENT0,e.TEXTURE_2D,null,0))}function ah(e){let t=e.checkFramebufferStatus(e.FRAMEBUFFER);if(t!==e.FRAMEBUFFER_COMPLETE)throw new Error("Error binding framebuffer: "+JN(e,t))}function JN(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 ${t}`}}function Aa(e,t,n){let r=Ie(e,()=>t());if(r==null)throw new Error(n);return r}function QN(e,t){let n=e.MAX_COMBINED_TEXTURE_IMAGE_UNITS-1,r=t+e.TEXTURE0;if(r<e.TEXTURE0||r>n){let s=`[gl.TEXTURE0, gl.TEXTURE${n}]`;throw new Error(`textureUnit must be in ${s}.`)}}function fi(e,t=2){return k.sizeFromShape(e.slice(0,e.length-t))}function mi(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 $m(e){let t=[1,1,1];return e.length===0||e.length===1&&e[0]===1||(t=[fi(e),...mi(e)]),t}function eC(e,t=!1){let n=re().getNumber("WEBGL_MAX_TEXTURE_SIZE");t&&(n=n*2,e=e.map((s,a)=>a>=e.length-2?k.nearestLargerEven(e[a]):e[a]),e.length===1&&(e=[2,e[0]])),e.length!==2&&(e=k.squeezeShape(e).newShape);let r=k.sizeFromShape(e);if(e.length<=1&&r<=n)return[1,r];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 s=fi(e),a=2,o=2;return e.length&&([a,o]=mi(e)),r=s*(a/2)*(o/2),k.sizeToSquarishShape(r).map(i=>i*2)}return k.sizeToSquarishShape(r)}function Rm(e){return e%2==0}function oh(e,t){if(e=e.slice(-2),t=t.slice(-2),k.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],r=t.slice(-1)[0];if(n===r||Rm(n)&&Rm(r)&&(e[0]===1||t[0]===1))return!0}return e[1]===t[1]&&Rm(e[0])&&Rm(t[0])}var _m,Dm;function tC(e){if(_m==null){let t=Bs(e);_m=t.getParameter(t.MAX_TEXTURE_SIZE)}return _m}function Yde(){_m=null}function Jde(){Dm=null}function nC(e){if(Dm==null){let t=Bs(e);Dm=t.getParameter(t.MAX_TEXTURE_IMAGE_UNITS)}return Math.min(16,Dm)}function rC(e){if(e===0)return 0;let t,n=Bs(e);return Mr(n,"EXT_disjoint_timer_query_webgl2")&&e===2?t=2:Mr(n,"EXT_disjoint_timer_query")?t=1:t=0,t}function Mr(e,t){return e.getExtension(t)!=null}function O5(e){try{if(Bs(e)!=null)return!0}catch(t){return console.log("Error when getting WebGL context: ",t),!1}return!1}function sC(e){if(e===0)return!1;let t=Bs(e);if(e===1){if(!Mr(t,"OES_texture_float"))return!1}else if(!Mr(t,"EXT_color_buffer_float"))return!1;return P5(t)}function aC(e){if(e===0)return!1;let t=Bs(e);if(e===1){if(!Mr(t,"OES_texture_float")||!Mr(t,"WEBGL_color_buffer_float"))return!1}else{if(Mr(t,"EXT_color_buffer_float"))return P5(t);let r="EXT_color_buffer_half_float";if(Mr(t,r)){let s=t.getExtension(r);return Qde(t,s)}return!1}return P5(t)}function P5(e){let t=D5(e),n=e.createTexture();e.bindTexture(e.TEXTURE_2D,n);let r=1,s=1;e.texImage2D(e.TEXTURE_2D,0,t.internalFormatFloat,r,s,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 Qde(e,t){let n=D5(e,t),r=e.createTexture();e.bindTexture(e.TEXTURE_2D,r);let s=1,a=1;e.texImage2D(e.TEXTURE_2D,0,n.internalFormatHalfFloat,s,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,r,0);let i=e.checkFramebufferStatus(e.FRAMEBUFFER)===e.FRAMEBUFFER_COMPLETE;return e.bindTexture(e.TEXTURE_2D,null),e.bindFramebuffer(e.FRAMEBUFFER,null),e.deleteTexture(r),e.deleteFramebuffer(o),i}function oC(e){return e!==2?!1:Bs(e).fenceSync!=null}function hu(e,t){Array.isArray(e)||(e=[e]),e.forEach(n=>{n!=null&&k.assert(n.dtype!=="complex64",()=>`${t} does not support complex64 tensors in the WebGL backend.`)})}var Fe=re();Fe.registerFlag("HAS_WEBGL",()=>Fe.getNumber("WEBGL_VERSION")>0);Fe.registerFlag("WEBGL_VERSION",()=>O5(2)?2:O5(1)?1:0);Fe.registerFlag("WEBGL_CHECK_NUMERICAL_PROBLEMS",()=>!1);Fe.registerFlag("WEBGL_BUFFER_SUPPORTED",()=>Fe.get("WEBGL_VERSION")===2);Fe.registerFlag("WEBGL_CPU_FORWARD",()=>!0);Fe.registerFlag("WEBGL_FORCE_F16_TEXTURES",()=>!1);Fe.registerFlag("WEBGL_PACK",()=>Fe.getBool("HAS_WEBGL"));Fe.registerFlag("WEBGL_PACK_NORMALIZATION",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_PACK_CLIP",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_PACK_DEPTHWISECONV",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_PACK_BINARY_OPERATIONS",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_PACK_UNARY_OPERATIONS",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_PACK_ARRAY_OPERATIONS",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_PACK_IMAGE_OPERATIONS",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_PACK_REDUCE",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_LAZILY_UNPACK",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_CONV_IM2COL",()=>Fe.getBool("WEBGL_PACK"));Fe.registerFlag("WEBGL_MAX_TEXTURE_SIZE",()=>tC(Fe.getNumber("WEBGL_VERSION")));Fe.registerFlag("WEBGL_MAX_TEXTURES_IN_SHADER",()=>nC(Fe.getNumber("WEBGL_VERSION")));Fe.registerFlag("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION",()=>{let e=Fe.getNumber("WEBGL_VERSION");return e===0?0:rC(e)});Fe.registerFlag("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE",()=>Fe.getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION")>0&&!mf.isMobile());Fe.registerFlag("WEBGL_RENDER_FLOAT32_CAPABLE",()=>sC(Fe.getNumber("WEBGL_VERSION")));Fe.registerFlag("WEBGL_RENDER_FLOAT32_ENABLED",()=>Fe.getBool("WEBGL_FORCE_F16_TEXTURES")?!1:Fe.getBool("WEBGL_RENDER_FLOAT32_CAPABLE"));Fe.registerFlag("WEBGL_DOWNLOAD_FLOAT_ENABLED",()=>aC(Fe.getNumber("WEBGL_VERSION")));Fe.registerFlag("WEBGL_FENCE_API_ENABLED",()=>oC(Fe.getNumber("WEBGL_VERSION")));Fe.registerFlag("WEBGL_SIZE_UPLOAD_UNIFORM",()=>Fe.getBool("WEBGL_RENDER_FLOAT32_ENABLED")?4:0);Fe.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}.`)});Fe.registerFlag("WEBGL_FLUSH_THRESHOLD",()=>mf.isMobile()&&Fe.getBool("IS_CHROME")?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}.`)});Fe.registerFlag("CPU_HANDOFF_SIZE_THRESHOLD",()=>128);Fe.registerFlag("WEBGL_USE_SHAPES_UNIFORMS",()=>!1);Fe.registerFlag("TOPK_LAST_DIM_CPU_HANDOFF_SIZE_THRESHOLD",()=>1e5);Fe.registerFlag("TOPK_K_CPU_HANDOFF_THRESHOLD",()=>128);function Wn(){let e,t,n,r,s,a,o,i,l,u;return re().getNumber("WEBGL_VERSION")===2?(e="#version 300 es",t="in",n="out",r="in",s="texture",a="outputColor",o="out vec4 outputColor;",i=`
bool isnan_custom(float val) {
return (val > 0.0 || val < 0.0) ? false : val != 0.0;
}
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",r="varying",s="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:r,texture2D:s,output:a,defineOutput:o,defineSpecialNaN:i,defineSpecialInf:l,defineRound:u}}function gi(e,t,n="index"){let r=k.computeStrides(t);return r.map((s,a)=>{let o=`int ${e[a]} = ${n} / ${s}`,i=a===r.length-1?`int ${e[a+1]} = ${n} - ${e[a]} * ${s}`:`index -= ${e[a]} * ${s}`;return`${o}; ${i};`}).join("")}function iC(e,t,n="index"){let r=k.computeStrides(t);return r.map((s,a)=>{let o=`int ${e[a]} = ${n} / outShapeStrides[${a}]`,i=a===r.length-1?`int ${e[a+1]} = ${n} - ${e[a]} * outShapeStrides[${a}]`:`index -= ${e[a]} * outShapeStrides[${a}]`;return`${o}; ${i};`}).join("")}function z5(e){let t=k.computeStrides(e).map(n=>n.toString());return`
int getFlatIndex(ivec3 coords) {
return coords.x * ${t[0]} + coords.y * ${t[1]} + coords.z;
}
`}var lC=`
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;
}
`,ehe=class{constructor(e){this.variableNames=["A"],this.packedInputs=!1,this.packedOutput=!0,this.outPackingScheme=th.DENSE;let t=rh(e),n=Wn();this.outputShape=e,this.userCode=`
ivec3 outCoordsFromFlatIndex(int index) {
${gi(["r","c","d"],e)}
return ivec3(r, c, d);
}
void main() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${t[0]}, ${t[1]}));
int index = 4 * (resTexRC.x * ${t[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);
}
${n.output} = result;
}
`}},the=class{constructor(e){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.outPackingScheme=th.DENSE;let t=rh(e),n=Wn();this.outputShape=e,this.userCode=`
ivec3 outCoordsFromFlatIndex(int index) {
${gi(["r","c","d"],e)}
return ivec3(r, c, d);
}
void main() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${t[0]}, ${t[1]}));
int index = 4 * (resTexRC.x * ${t[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));
}
${n.output} = result;
}
`}},nhe=class{constructor(e){this.variableNames=["A"],this.outTexUsage=Fr.DOWNLOAD;let t=Wn();this.outputShape=e,this.userCode=`
${lC}
void main() {
float x = getAAtOutCoords();
${t.output} = encode_float(x);
}
`}},rhe=class{constructor(e){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!1,this.outTexUsage=Fr.DOWNLOAD;let t=Wn();this.outputShape=e,this.userCode=`
${lC}
void main() {
ivec3 coords = getOutputCoords();
float x = getChannel(getAAtOutCoords(), vec2(coords.y, coords.z));
${t.output} = encode_float(x);
}
`}},she=class{constructor(e,t,n=!1){this.variableNames=["A"];let r=Wn(),[s,a]=t;this.outputShape=e;let o="result";n&&(o="floor(result * 255. + 0.5)"),this.userCode=`
${z5(e)}
void main() {
ivec3 coords = getOutputCoords();
int flatIndex = getFlatIndex(coords);
int offset = imod(flatIndex, 4);
flatIndex = idiv(flatIndex, 4, 1.);
int r = flatIndex / ${a};
int c = imod(flatIndex, ${a});
vec2 uv = (vec2(c, r) + halfCR) / vec2(${a}.0, ${s}.0);
vec4 values = ${r.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];
}
${r.output} = vec4(${o}, 0., 0., 0.);
}
`}},ahe=class{constructor(e,t,n=!1){this.variableNames=["A"],this.packedInputs=!1,this.packedOutput=!0;let r=Wn(),[s,a]=t;this.outputShape=e;let o="",i="result";n&&(i="floor(result * 255. + 0.5)");for(let l=0;l<=1;l++)for(let u=0;u<=1;u++){let c=l*2+u;o+=`
localCoords = coords;
if(localCoords[2] + ${u} < ${e[2]}) {
localCoords[2] += ${u};
if(localCoords[1] + ${l} < ${e[1]}) {
localCoords[1] += ${l};
flatIndex = getFlatIndex(localCoords);
offset = imod(flatIndex, 4);
flatIndex = idiv(flatIndex, 4, 1.);
r = flatIndex / ${a};
c = imod(flatIndex, ${a});
uv = (vec2(c, r) + halfCR) / vec2(${a}.0, ${s}.0);
values = ${r.texture2D}(A, uv);
if(offset == 0) {
result[${c}] = values[0];
} else if(offset == 1) {
result[${c}] = values[1];
} else if(offset == 2) {
result[${c}] = values[2];
} else {
result[${c}] = values[3];
}
}
}
`}this.userCode=`
${z5(e)}
void main() {
ivec3 coords = getOutputCoords();
vec4 result = vec4(0.);
int flatIndex, r, c, offset;
ivec3 localCoords;
vec2 uv;
vec4 values;
${o}
${r.output} = ${i};
}
`}},uC={};_e(uC,{bindVertexProgramAttributeStreams:()=>AC,createBufferFromOutputTexture:()=>vC,createFloat16MatrixTexture:()=>fC,createFloat16PackedMatrixTexture:()=>yC,createFloat32MatrixTexture:()=>pC,createIndexBuffer:()=>hC,createPackedMatrixTexture:()=>gC,createUnsignedBytesMatrixTexture:()=>mC,createVertexBuffer:()=>dC,createVertexShader:()=>cC,downloadByteEncodedFloatMatrixFromOutputTexture:()=>kC,downloadFloat32MatrixFromBuffer:()=>wC,downloadMatrixFromPackedOutputTexture:()=>SC,downloadPackedMatrixFromBuffer:()=>IC,getInternalFormatForFloat16MatrixTexture:()=>B5,getInternalFormatForFloat16PackedMatrixTexture:()=>U5,getInternalFormatForFloat32MatrixTexture:()=>L5,getInternalFormatForPackedMatrixTexture:()=>V5,getInternalFormatForUnsignedBytesMatrixTexture:()=>W5,uploadDenseMatrixToTexture:()=>xC,uploadPixelDataToTexture:()=>bC});function cC(e){let t=Wn(),n=`${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 LN(e,n)}function dC(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 UN(e,t)}function hC(e){let t=new Uint16Array([0,1,2,2,1,3]);return HN(e,t)}function ih(e,t,n,r,s,a){jN(t,n);let o=GN(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)),Ie(e,()=>e.texImage2D(i,0,r,t,n,0,s,a,null)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,null)),o}function L5(e){return e.internalFormatFloat}function pC(e,t,n,r){let[s,a]=nh(t,n);return ih(e,s,a,L5(r),r.textureFormatFloat,e.FLOAT)}function B5(e){return e.internalFormatHalfFloat}function fC(e,t,n,r){let[s,a]=nh(t,n);return ih(e,s,a,B5(r),r.textureFormatFloat,r.textureTypeHalfFloat)}function W5(e){return e.downloadTextureFormat}function mC(e,t,n,r){let[s,a]=nh(t,n);return ih(e,s,a,W5(r),e.RGBA,e.UNSIGNED_BYTE)}function V5(e){return e.internalFormatPackedFloat}function gC(e,t,n,r){let[s,a]=du(t,n);return ih(e,s,a,V5(r),e.RGBA,e.FLOAT)}function U5(e){return e.internalFormatPackedHalfFloat}function yC(e,t,n,r){let[s,a]=du(t,n);return ih(e,s,a,U5(r),e.RGBA,r.textureTypeHalfFloat)}function AC(e,t,n){let r=0,s=3*4,a=3*4+2*4;return Ie(e,()=>e.bindBuffer(e.ARRAY_BUFFER,n)),F5(e,t,"clipSpacePos",n,3,a,r)&&F5(e,t,"uv",n,2,a,s)}function xC(e,t,n,r,s,a){Ie(e,()=>e.bindTexture(e.TEXTURE_2D,t));let o,i,l;s instanceof Uint8Array?(o=new Uint8Array(n*r*4),i=e.UNSIGNED_BYTE,l=e.RGBA):(o=new Float32Array(n*r*4),i=e.FLOAT,l=a.internalFormatPackedFloat),o.set(s),Ie(e,()=>e.texImage2D(e.TEXTURE_2D,0,l,n,r,0,e.RGBA,i,o)),Ie(e,()=>e.bindTexture(e.TEXTURE_2D,null))}function bC(e,t,n){Ie(e,()=>e.bindTexture(e.TEXTURE_2D,t)),n.data instanceof Uint8Array?Ie(e,()=>e.texImage2D(e.TEXTURE_2D,0,e.RGBA,n.width,n.height,0,e.RGBA,e.UNSIGNED_BYTE,n.data)):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 vC(e,t,n,r){let s=e.createBuffer();Ie(e,()=>e.bindBuffer(e.PIXEL_PACK_BUFFER,s));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)),s}function wC(e,t,n){let r=e,s=new Float32Array(n);return r.bindBuffer(r.PIXEL_PACK_BUFFER,t),r.getBufferSubData(r.PIXEL_PACK_BUFFER,0,s),r.bindBuffer(r.PIXEL_PACK_BUFFER,null),s}function kC(e,t,n,r){let[s,a]=nh(t,n),o=4,i=new Uint8Array(Wde(t*n,o));return Ie(e,()=>e.readPixels(0,0,s,a,r.downloadTextureFormat,e.UNSIGNED_BYTE,i)),new Float32Array(i.buffer)}function IC(e,t,n,r,s,a,o,i){let l=e,u=new Float32Array(Vde(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 SC(e,t,n){let r=new Float32Array(t*n*4);return Ie(e,()=>e.readPixels(0,0,n,t,e.RGBA,e.FLOAT,r)),r}var Fm=class{constructor(e){this.outputTexture=null,this.program=null,this.disposed=!1,this.vertexAttrsAreBound=!1,this.itemsToPoll=[];let t=re().getNumber("WEBGL_VERSION");e!=null?(this.gl=e,Nm(t,e)):this.gl=Bs(t);let n="WEBGL_color_buffer_float",r="EXT_color_buffer_half_float";if(re().getNumber("WEBGL_VERSION")===1){let s="OES_texture_float",a="OES_texture_half_float";if(this.textureFloatExtension=sh(this.gl,s),Mr(this.gl,a))this.textureHalfFloatExtension=sh(this.gl,a);else if(re().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),Mr(this.gl,r))this.colorBufferHalfFloatExtension=sh(this.gl,r);else if(re().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",Mr(this.gl,n))this.colorBufferFloatExtension=this.gl.getExtension(n);else if(Mr(this.gl,r))this.colorBufferHalfFloatExtension=this.gl.getExtension(r);else throw new Error("GL context does not support color renderable floats");this.vertexBuffer=dC(this.gl),this.indexBuffer=hC(this.gl),this.framebuffer=qN(this.gl),this.textureConfig=D5(this.gl,this.textureHalfFloatExtension)}get debug(){return re().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(),pC(this.gl,e,t,this.textureConfig)}createFloat16MatrixTexture(e,t){return this.throwIfDisposed(),fC(this.gl,e,t,this.textureConfig)}createUnsignedBytesMatrixTexture(e,t){return this.throwIfDisposed(),mC(this.gl,e,t,this.textureConfig)}uploadPixelDataToTexture(e,t){this.throwIfDisposed(),bC(this.gl,e,t)}uploadDenseMatrixToTexture(e,t,n,r){this.throwIfDisposed(),xC(this.gl,e,t,n,r,this.textureConfig)}createFloat16PackedMatrixTexture(e,t){return this.throwIfDisposed(),yC(this.gl,e,t,this.textureConfig)}createPackedMatrixTexture(e,t){return this.throwIfDisposed(),gC(this.gl,e,t,this.textureConfig)}deleteMatrixTexture(e){this.throwIfDisposed(),this.outputTexture===e&&(M5(this.gl,this.framebuffer),this.outputTexture=null),Ie(this.gl,()=>this.gl.deleteTexture(e))}downloadByteEncodedFloatMatrixFromOutputTexture(e,t,n){return this.downloadMatrixDriver(e,()=>kC(this.gl,t,n,this.textureConfig))}downloadPackedMatrixFromBuffer(e,t,n,r,s,a){return IC(this.gl,e,t,n,r,s,a,this.textureConfig)}downloadFloat32MatrixFromBuffer(e,t){return wC(this.gl,e,t)}createBufferFromTexture(e,t,n){this.bindTextureToFrameBuffer(e);let r=vC(this.gl,t,n,this.textureConfig);return this.unbindTextureToFrameBuffer(),r}createAndWaitForFence(){let e=this.createFence(this.gl);return this.pollFence(e)}createFence(e){let t,n;if(re().getBool("WEBGL_FENCE_API_ENABLED")){let r=e,s=r.fenceSync(r.SYNC_GPU_COMMANDS_COMPLETE,0);e.flush(),n=()=>{let a=r.clientWaitSync(s,0,0);return a===r.ALREADY_SIGNALED||a===r.CONDITION_SATISFIED},t=s}else re().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION")>0?(t=this.beginQuery(),this.endQuery(),n=()=>this.isQueryAvailable(t,re().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION"))):n=()=>!0;return{query:t,isFencePassed:n}}downloadMatrixFromPackedTexture(e,t,n){return this.downloadMatrixDriver(e,()=>SC(this.gl,t,n))}createProgram(e){this.throwIfDisposed();let t=this.gl,n=BN(t,e);this.vertexShader==null&&(this.vertexShader=cC(t));let r=WN(t);return Ie(t,()=>t.attachShader(r,this.vertexShader)),Ie(t,()=>t.attachShader(r,n)),VN(t,r),this.debug&&Cm(t,r),this.vertexAttrsAreBound||(this.setProgram(r),this.vertexAttrsAreBound=AC(t,this.program,this.vertexBuffer)),r}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&&Cm(this.gl,this.program),Ie(this.gl,()=>this.gl.useProgram(e))}getUniformLocation(e,t,n=!0){return this.throwIfDisposed(),n?XN(this.gl,e,t):ZN(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(),YN(this.gl,e,t,n)}setOutputMatrixTexture(e,t,n){this.setOutputMatrixTextureDriver(e,n,t)}setOutputPackedMatrixTexture(e,t,n){this.throwIfDisposed();let[r,s]=du(t,n);this.setOutputMatrixTextureDriver(e,r,s)}setOutputMatrixWriteRegion(e,t,n,r){this.setOutputMatrixWriteRegionDriver(n,e,r,t)}setOutputPackedMatrixWriteRegion(e,t,n,r){throw new Error("setOutputPackedMatrixWriteRegion not implemented.")}debugValidate(){this.program!=null&&Cm(this.gl,this.program),ah(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=sh(this.gl,re().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(re().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION")===2){let n=this.gl,r=this.getQueryTimerExtensionWebGL2(),s=n.createQuery();return n.beginQuery(r.TIME_ELAPSED_EXT,s),s}let e=this.getQueryTimerExtensionWebGL1(),t=e.createQueryEXT();return e.beginQueryEXT(e.TIME_ELAPSED_EXT,t),t}endQuery(){if(re().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 k.repeatedTry(()=>this.disposed||this.isQueryAvailable(e,re().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION"))),this.getQueryTime(e,re().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,r=this.getQueryTimerExtensionWebGL2(),s=n.getQueryParameter(e,n.QUERY_RESULT_AVAILABLE);return this.disjoint==null&&(this.disjoint=this.gl.getParameter(r.GPU_DISJOINT_EXT)),s&&!this.disjoint}else{let n=this.getQueryTimerExtensionWebGL1(),r=n.getQueryObjectEXT(e,n.QUERY_RESULT_AVAILABLE_EXT);return this.disjoint==null&&(this.disjoint=this.gl.getParameter(n.GPU_DISJOINT_EXT)),r&&!this.disjoint}}pollFence(e){return new Promise(t=>{this.addItemToPoll(()=>e.isFencePassed(),()=>t())})}pollItems(){let e=ohe(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)&&k.repeatedTry(()=>(this.pollItems(),this.itemsToPoll.length===0))}bindTextureToFrameBuffer(e){this.throwIfDisposed(),Em(this.gl,e,this.framebuffer),this.debug&&ah(this.gl)}unbindTextureToFrameBuffer(){this.outputTexture!=null?(Em(this.gl,this.outputTexture,this.framebuffer),this.debug&&ah(this.gl)):M5(this.gl,this.framebuffer)}downloadMatrixDriver(e,t){this.bindTextureToFrameBuffer(e);let n=t();return this.unbindTextureToFrameBuffer(),n}setOutputMatrixTextureDriver(e,t,n){this.throwIfDisposed();let r=this.gl;Em(r,e,this.framebuffer),this.debug&&ah(r),this.outputTexture=e,Ie(r,()=>r.viewport(0,0,t,n)),Ie(r,()=>r.scissor(0,0,t,n))}setOutputMatrixWriteRegionDriver(e,t,n,r){this.throwIfDisposed(),Ie(this.gl,()=>this.gl.scissor(e,t,n,r))}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 ohe(e){let t=0;for(;t<e.length&&e[t]();++t);return t-1}var{getBroadcastDims:TC}=_;function ihe(e,t,n){let r=[];if(e.forEach(p=>{let f=k.sizeFromShape(p.shapeInfo.logicalShape);if(p.shapeInfo.isUniform?r.push(`uniform float ${p.name}${f>1?`[${f}]`:""};`):(r.push(`uniform sampler2D ${p.name};`),r.push(`uniform int offset${p.name};`)),n.enableShapeUniforms){let{uniformShape:m}=H5(n.packedInputs,p.shapeInfo.logicalShape,p.shapeInfo.texShape);switch(m.length){case 1:r.push(`uniform int ${p.name}Shape;`);break;case 2:r.push(`uniform ivec2 ${p.name}Shape;`);break;case 3:r.push(`uniform ivec3 ${p.name}Shape;`);break;case 4:r.push(`uniform ivec4 ${p.name}Shape;`);break;default:break}r.push(`uniform ivec2 ${p.name}TexShape;`)}}),n.enableShapeUniforms){switch(t.logicalShape.length){case 1:r.push("uniform int outShape;");break;case 2:r.push("uniform ivec2 outShape;"),r.push("uniform int outShapeStrides;");break;case 3:r.push("uniform ivec3 outShape;"),r.push("uniform ivec2 outShapeStrides;");break;case 4:r.push("uniform ivec4 outShape;"),r.push("uniform ivec3 outShapeStrides;");break;default:break}r.push("uniform ivec2 outTexShape;")}n.customUniforms&&n.customUniforms.forEach(p=>{r.push(`uniform ${p.type} ${p.name}${p.arrayIndex?`[${p.arrayIndex}]`:""};`)});let s=r.join(`
`),a=e.map(p=>lhe(p,t,n.packedInputs,n.enableShapeUniforms)).join(`
`),o=t.texShape,i=Wn(),l=dhe(i),u,c,d=fhe(i);return t.isPacked?(u=uhe(t.logicalShape,o,n.enableShapeUniforms),c=phe(i)):(u=che(t.logicalShape,o,n.enableShapeUniforms),c=hhe(i)),n.packedInputs&&(d+=Ahe),[d,l,c,s,u,a,n.userCode].join(`
`)}function pu(e,t=!1){let n=e.shapeInfo.logicalShape;switch(n.length){case 0:return $he(e,t);case 1:return _he(e,t);case 2:return Fhe(e,t);case 3:return Ohe(e,t);case 4:return zhe(e,t);case 5:return Lhe(e);case 6:return Bhe(e);default:throw new Error(`${n.length}-D input sampling is not yet supported`)}}function NC(e,t){switch(e.shapeInfo.logicalShape.length){case 0:return Ehe(e);case 1:return Rhe(e,t);case 2:return Dhe(e,t);case 3:return Mhe(e,t);default:return Phe(e,t)}}function lhe(e,t,n=!1,r){let s="";n?s+=NC(e,r):s+=pu(e,r);let a=e.shapeInfo.logicalShape,o=t.logicalShape;return a.length<=o.length&&(n?s+=Whe(e,t):s+=Vhe(e,t)),s}function uhe(e,t,n){switch(e.length){case 0:return CC();case 1:return xhe(e,t,n);case 2:return Nhe(e,t,n);case 3:return vhe(e,t,n);default:return khe(e,t,n)}}function che(e,t,n){switch(e.length){case 0:return CC();case 1:return bhe(e,t,n);case 2:return Che(e,t,n);case 3:return whe(e,t,n);case 4:return Ihe(e,t,n);case 5:return She(e,t);case 6:return The(e,t);default:throw new Error(`${e.length}-D output sampling is not yet supported`)}}function dhe(e){return`
float sampleTexture(sampler2D textureSampler, vec2 uv) {
return ${e.texture2D}(textureSampler, uv).r;
}
`}function hhe(e){return`
void setOutput(float val) {
${e.output} = vec4(val, 0, 0, 0);
}
`}function phe(e){return`
void setOutput(vec4 val) {
${e.output} = val;
}
`}function fhe(e){return`${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);
}
${mhe}
${ghe}
${yhe}
`}var mhe=`
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);
}
`,ghe=`
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);
}
`,yhe=`
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);
}
`,Ahe=`
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 CC(){return`
int getOutputCoords() {
return 0;
}
`}function xhe(e,t,n){let r=[Math.ceil(t[0]/2),Math.ceil(t[1]/2)];return r[0]===1?n?`
int getOutputCoords() {
return 2 * int(resultUV.x * ceil(float(outTexShape[1]) / 2.0));
}
`:`
int getOutputCoords() {
return 2 * int(resultUV.x * ${r[1]}.0);
}
`:r[1]===1?n?`
int getOutputCoords() {
return 2 * int(resultUV.y * ceil(float(outTexShape[0]) / 2.0));
}
`:`
int getOutputCoords() {
return 2 * int(resultUV.y * ${r[0]}.0);
}
`:n?`
int getOutputCoords() {
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(packedTexShape[0], packedTexShape[1]));
return 2 * (resTexRC.x * packedTexShape[1] + resTexRC.y);
}
`:`
int getOutputCoords() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${r[0]}, ${r[1]}));
return 2 * (resTexRC.x * ${r[1]} + resTexRC.y);
}
`}function bhe(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 * ${t[1]} + resTexRC.y;
}
`}function vhe(e,t,n){if(n)return`
ivec3 getOutputCoords() {
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
int texelsInLogicalRow = int(ceil(float(outShape[2]) / 2.0));
int texelsInBatch = texelsInLogicalRow * int(ceil(float(outShape[1]) / 2.0));
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(packedTexShape[0], packedTexShape[1]));
int index = resTexRC.x * packedTexShape[1] + resTexRC.y;
int b = index / texelsInBatch;
index -= b * texelsInBatch;
int r = 2 * (index / texelsInLogicalRow);
int c = imod(index, texelsInLogicalRow) * 2;
return ivec3(b, r, c);
}
`;let r=[Math.ceil(t[0]/2),Math.ceil(t[1]/2)],s=Math.ceil(e[2]/2),a=s*Math.ceil(e[1]/2);return`
ivec3 getOutputCoords() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${r[0]}, ${r[1]}));
int index = resTexRC.x * ${r[1]} + resTexRC.y;
int b = index / ${a};
index -= b * ${a};
int r = 2 * (index / ${s});
int c = imod(index, ${s}) * 2;
return ivec3(b, r, c);
}
`}function whe(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;
${iC(["r","c","d"],e)}
return ivec3(r, c, d);
}
`;let r=gi(["r","c","d"],e);return`
ivec3 getOutputCoords() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${t[0]}, ${t[1]}));
int index = resTexRC.x * ${t[1]} + resTexRC.y;
${r}
return ivec3(r, c, d);
}
`}function khe(e,t,n){if(n)return`
ivec4 getOutputCoords() {
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(packedTexShape[0], packedTexShape[1]));
int index = resTexRC.x * packedTexShape[1] + resTexRC.y;
int texelsInLogicalRow = int(ceil(float(outShape[3]) / 2.0));
int texelsInBatch = texelsInLogicalRow * int(ceil(float(outShape[2]) / 2.0));
int texelsInBatchN = texelsInBatch * outShape[1];
int b2 = index / texelsInBatchN;
index -= b2 * texelsInBatchN;
int b = index / texelsInBatch;
index -= b * texelsInBatch;
int r = 2 * (index / texelsInLogicalRow);
int c = imod(index, texelsInLogicalRow) * 2;
return ivec4(b2, b, r, c);
}
`;let r=[Math.ceil(t[0]/2),Math.ceil(t[1]/2)],s=Math.ceil(e[e.length-1]/2),a=s*Math.ceil(e[e.length-2]/2),o=a,i="",l="b, r, c";for(let u=2;u<e.length-1;u++)o*=e[e.length-u-1],i=`
int b${u} = index / ${o};
index -= b${u} * ${o};
`+i,l=`b${u}, `+l;return`
ivec${e.length} getOutputCoords() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${r[0]}, ${r[1]}));
int index = resTexRC.x * ${r[1]} + resTexRC.y;
${i}
int b = index / ${a};
index -= b * ${a};
int r = 2 * (index / ${s});
int c = imod(index, ${s}) * 2;
return ivec${e.length}(${l});
}
`}function Ihe(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;
${iC(["r","c","d","d2"],e)}
return ivec4(r, c, d, d2);
}
`;let r=gi(["r","c","d","d2"],e);return`
ivec4 getOutputCoords() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${t[0]}, ${t[1]}));
int index = resTexRC.x * ${t[1]} + resTexRC.y;
${r}
return ivec4(r, c, d, d2);
}
`}function She(e,t){let n=gi(["r","c","d","d2","d3"],e);return`
ivec5 getOutputCoords() {
ivec2 resTexRC = ivec2(resultUV.yx * vec2(${t[0]},
${t[1]}));
int index = resTexRC.x * ${t[1]} + resTexRC.y;
${n}
ivec5 outShape = ivec5(r, c, d, d2, d3);
return outShape;
}
`}function The(e,t){let n=gi(["r","c","d","d2","d3","d4"],e);return`
ivec6 getOutputCoords() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${t[0]}, ${t[1]}));
int index = resTexRC.x * ${t[1]} + resTexRC.y;
${n}
ivec6 result = ivec6(r, c, d, d2, d3, d4);
return result;
}
`}function Nhe(e,t,n){let r=[Math.ceil(t[0]/2),Math.ceil(t[1]/2)];if(k.arraysEqual(e,t))return n?`
ivec2 getOutputCoords() {
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
return 2 * ivec2(resultUV.yx * vec2(packedTexShape[0], packedTexShape[1]));
}
`:`
ivec2 getOutputCoords() {
return 2 * ivec2(resultUV.yx * vec2(${r[0]}, ${r[1]}));
}
`;let s=Math.ceil(e[1]/2);return n?`
ivec2 getOutputCoords() {
ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
int texelsInLogicalRow = int(ceil(float(outShape[1]) / 2.0));
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(packedTexShape[0], packedTexShape[1]));
int index = resTexRC.x * packedTexShape[1] + resTexRC.y;
int r = 2 * (index / texelsInLogicalRow);
int c = imod(index, texelsInLogicalRow) * 2;
return ivec2(r, c);
}
`:`
ivec2 getOutputCoords() {
ivec2 resTexRC = ivec2(resultUV.yx *
vec2(${r[0]}, ${r[1]}));
int index = resTexRC.x * ${r[1]} + resTexRC.y;
int r = 2 * (index / ${s});
int c = imod(index, ${s}) * 2;
return ivec2(r, c);
}
`}function Che(e,t,n){return k.arraysEqual(e,t)?n?`
ivec2 getOutputCoords() {
return ivec2(resultUV.yx * vec2(outTexShape[0], outTexShape[1]));
}
`:`
ivec2 getOutputCoords() {
return ivec2(resultUV.yx * vec2(${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 * ${e[1]};
return ivec2(r, c);
}
`}function yi(e){return`offset${e}`}function Ehe(e){let t=e.name,n="get"+t.charAt(0).toUpperCase()+t.slice(1),r=Wn();return`
vec4 ${n}() {
return ${r.texture2D}(${t}, halfCR);
}
`}function $he(e,t){let n=e.name,r="get"+n.charAt(0).toUpperCase()+n.slice(1);if(e.shapeInfo.isUniform)return`float ${r}() {return ${n};}`;let[s,a]=e.shapeInfo.texShape;if(s===1&&a===1)return`
float ${r}() {
return sampleTexture(${n}, halfCR);
}
`;let o=yi(n);if(t)return`
float ${r}() {
vec2 uv = uvFromFlat(${n}TexShape[0], ${n}TexShape[1], ${o});
return sampleTexture(${n}, uv);
}
`;let[i,l]=e.shapeInfo.texShape;return`
float ${r}() {
vec2 uv = uvFromFlat(${i}, ${l}, ${o});
return sampleTexture(${n}, uv);
}
`}function Rhe(e,t){let n=e.name,r="get"+n.charAt(0).toUpperCase()+n.slice(1),s=e.shapeInfo.texShape,a=Wn();if(t)return`
vec4 ${r}(int index) {
ivec2 packedTexShape = ivec2(ceil(float(${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(s[0]/2),Math.ceil(s[1]/2)];return`
vec4 ${r}(int index) {
vec2 uv = packedUVfrom1D(
${o[0]}, ${o[1]}, index);
return ${a.texture2D}(${n}, uv);
}
`}function _he(e,t){let n=e.name,r="get"+n.charAt(0).toUpperCase()+n.slice(1);if(e.shapeInfo.isUniform)return`
float ${r}(int index) {
${fu(e)}
}
`;let s=e.shapeInfo.texShape,a=s[0],o=s[1];if(o===1&&a===1)return`
float ${r}(int index) {
return sampleTexture(${n}, halfCR);
}
`;let i=yi(n);return o===1?t?`
float ${r}(int index) {
vec2 uv = vec2(0.5, (float(index + ${i}) + 0.5) / float(${n}TexShape[0]));
return sampleTexture(${n}, uv);
}
`:`
float ${r}(int index) {
vec2 uv = vec2(0.5, (float(index + ${i}) + 0.5) / ${a}.0);
return sampleTexture(${n}, uv);
}
`:a===1?t?`
float ${r}(int index) {
vec2 uv = vec2((float(index + ${i}) + 0.5) / float(${n}TexShape[1]), 0.5);
return sampleTexture(${n}, uv);
}
`:`
float ${r}(int index) {
vec2 uv = vec2((float(index + ${i}) + 0.5) / ${o}.0, 0.5);
return sampleTexture(${n}, uv);
}
`:t?`
float ${r}(int index) {
vec2 uv = uvFromFlat(${n}TexShape[0], ${n}TexShape[1], index + ${i});
return sampleTexture(${n}, uv);
}
`:`
float ${r}(int index) {
vec2 uv = uvFromFlat(${a}, ${o}, index + ${i});
return sampleTexture(${n}, uv);
}
`}function Dhe(e,t){let n=e.shapeInfo.logicalShape,r=e.name,s="get"+r.charAt(0).toUpperCase()+r.slice(1),a=e.shapeInfo.texShape,o=a[0],i=a[1],l=Wn();if(a!=null&&k.arraysEqual(n,a))return t?`
vec4 ${s}(int row, int col) {
vec2 uv = (vec2(col, row) + halfCR) / vec2(${r}TexShape[1], ${r}TexShape[0]);
return ${l.texture2D}(${r}, uv);
}
`:`
vec4 ${s}(int row, int col) {
vec2 uv = (vec2(col, row) + halfCR) / vec2(${i}.0, ${o}.0);
return ${l.texture2D}(${r}, uv);
}
`;if(t)return`
vec4 ${s}(int row, int col) {
ivec2 packedTexShape = ivec2(ceil(float(${r}TexShape[0]) / 2.0), ceil(float(${r}TexShape[1]) / 2.0));
int valuesPerRow = int(ceil(float(${r}Shape[1]) / 2.0));
vec2 uv = packedUVfrom2D(valuesPerRow, packedTexShape[0], packedTexShape[1], row, col);
return ${l.texture2D}(${r}, uv);
}
`;let u=[Math.ceil(a[0]/2),Math.ceil(a[1]/2)],c=Math.ceil(n[1]/2);return`
vec4 ${s}(int row, int col) {
vec2 uv = packedUVfrom2D(${c}, ${u[0]}, ${u[1]}, row, col);
return ${l.texture2D}(${r}, uv);
}
`}function Fhe(e,t){let n=e.shapeInfo.logicalShape,r=e.name,s="get"+r.charAt(0).toUpperCase()+r.slice(1),a=e.shapeInfo.texShape;if(a!=null&&k.arraysEqual(n,a)){if(t)return`
float ${s}(int row, int col) {
vec2 uv = (vec2(col, row) + halfCR) / vec2(${r}TexShape[1], ${r}TexShape[0]);
return sampleTexture(${r}, uv);
}
`;let h=a[0],p=a[1];return`
float ${s}(int row, int col) {
vec2 uv = (vec2(col, row) + halfCR) / vec2(${p}.0, ${h}.0);
return sampleTexture(${r}, uv);
}
`}let{newShape:o,keptDims:i}=k.squeezeShape(n),l=o;if(l.length<n.length){let h=mu(e,l),p=["row","col"];return`
${pu(h,t)}
float ${s}(int row, int col) {
return ${s}(${gu(p,i)});
}
`}if(e.shapeInfo.isUniform)return`
float ${s}(int row, int col) {
int index = round(dot(vec2(row, col), vec2(${n[1]}, 1)));
${fu(e)}
}
`;let u=a[0],c=a[1],d=yi(r);return c===1?t?`
float ${s}(int row, int col) {
float index = dot(vec3(row, col, ${d}), vec3(${r}Shape[1], 1, 1));
vec2 uv = vec2(0.5, (index + 0.5) / float(${r}TexShape[0]));
return sampleTexture(${r}, uv);
}
`:`
float ${s}(int row, int col) {
float index = dot(vec3(row, col, ${d}), vec3(${n[1]}, 1, 1));
vec2 uv = vec2(0.5, (index + 0.5) / ${u}.0);
return sampleTexture(${r}, uv);
}
`:u===1?t?`
float ${s}(int row, int col) {
float index = dot(vec3(row, col, ${d}), vec3(${r}Shape[1], 1, 1));
vec2 uv = vec2((index + 0.5) / float(${r}TexShape[1]), 0.5);
return sampleTexture(${r}, uv);
}
`:`
float ${s}(int row, int col) {
float index = dot(vec3(row, col, ${d}), vec3(${n[1]}, 1, 1));
vec2 uv = vec2((index + 0.5) / ${c}.0, 0.5);
return sampleTexture(${r}, uv);
}
`:t?`
float ${s}(int row, int col) {
// Explicitly use integer operations as dot() only works on floats.
int index = row * ${r}Shape[1] + col + ${d};
vec2 uv = uvFromFlat(${r}TexShape[0], ${r}TexShape[1], index);
return sampleTexture(${r}, uv);
}
`:`
float ${s}(int row, int col) {
// Explicitly use integer operations as dot() only works on floats.
int index = row * ${n[1]} + col + ${d};
vec2 uv = uvFromFlat(${u}, ${c}, index);
return sampleTexture(${r}, uv);
}
`}function Mhe(e,t){let n=e.shapeInfo.logicalShape,r=e.name,s="get"+r.charAt(0).toUpperCase()+r.slice(1),a=e.shapeInfo.texShape,o=[Math.ceil(a[0]/2),Math.ceil(a[1]/2)];if(n[0]===1){let h=n.slice(1),p=[1,2],f=mu(e,h),m=["b","row","col"];return`
${NC(f,t)}
vec4 ${s}(int b, int row, int col) {
return ${s}(${gu(m,p)});
}
`}let i=Wn();if(t)return`
vec4 ${s}(int b, int row, int col) {
ivec2 packedTexShape = ivec2(ceil(float(${r}TexShape[0]) / 2.0), ceil(float(${r}TexShape[1]) / 2.0));
int valuesPerRow = int(ceil(float(${r}Shape[2]) / 2.0));
int texelsInBatch = valuesPerRow * int(ceil(float(${r}Shape[1]) / 2.0));
vec2 uv = packedUVfrom3D(
packedTexShape[0], packedTexShape[1], texelsInBatch, valuesPerRow, b, row, col);
return ${i.texture2D}(${r}, uv);
}
`;let l=o[0],u=o[1],c=Math.ceil(n[2]/2),d=c*Math.ceil(n[1]/2);return`
vec4 ${s}(int b, int row, int col) {
vec2 uv = packedUVfrom3D(
${l}, ${u}, ${d}, ${c}, b, row, col);
return ${i.texture2D}(${r}, uv);
}
`}function Ohe(e,t){let n=e.shapeInfo.logicalShape,r=e.name,s="get"+r.charAt(0).toUpperCase()+r.slice(1),a=n[1]*n[2],o=n[2],{newShape:i,keptDims:l}=k.squeezeShape(n),u=i;if(u.length<n.length){let m=mu(e,u),g=["row","col","depth"];return`
${pu(m,t)}
float ${s}(int row, int col, int depth) {
return ${s}(${gu(g,l)});
}
`}if(e.shapeInfo.isUniform)return`
float ${s}(int row, int col, int depth) {
int index = round(dot(vec3(row, col, depth),
vec3(${a}, ${o}, 1)));
${fu(e)}
}
`;let c=e.shapeInfo.texShape,d=c[0],h=c[1],p=e.shapeInfo.flatOffset;if(h===a&&p==null)return t?`
float ${s}(int row, int col, int depth) {
int stride1 = ${r}Shape[2];
float texR = float(row);
float texC = dot(vec2(col, depth), vec2(stride1, 1));
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${r}TexShape[1], ${r}TexShape[0]);
return sampleTexture(${r}, uv);
}
`:`
float ${s}(int row, int col, int depth) {
float texR = float(row);
float texC = dot(vec2(col, depth), vec2(${o}, 1));
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${h}.0, ${d}.0);
return sampleTexture(${r}, uv);
}
`;if(h===o&&p==null)return t?`
float ${s}(int row, int col, int depth) {
float texR = dot(vec2(row, col), vec2(${r}Shape[1], 1));
float texC = float(depth);
vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${r}TexShape[1], ${r}TexShape[0]);
return sampleTexture(${r}, uv);
}
`:`
float ${s}(int row, int col, int depth) {
float texR = dot(vec2(row, col), vec2(${n[1]}, 1));
float texC = float(depth);
vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${h}.0, ${d}.0);
return sampleTexture(${r}, uv);
}
`;let f=yi(r);return t?`
float ${s}(int row, int col, int depth) {
// Explicitly use integer operations as dot() only works on floats.
int stride0 = ${r}Shape[1] * ${r}Shape[2];
int stride1 = ${r}Shape[2];
int index = row * ${a} + col * ${o} + depth + ${f};
vec2 uv = uvFromFlat(${r}TexShape[0], ${r}TexShape[1], index);
return sampleTexture(${r}, uv);
}
`:`
float ${s}(int row, int col, int depth) {
// Explicitly use integer operations as dot() only works on floats.
int index = row * ${a} + col * ${o} + depth + ${f};
vec2 uv = uvFromFlat(${d}, ${h}, index);
return sampleTexture(${r}, uv);
}
`}function Phe(e,t){let n=e.name,r="get"+n.charAt(0).toUpperCase()+n.slice(1),s=Wn();if(t)return`
vec4 ${r}(int b2, int b, int row, int col) {
int valuesPerRow = int(ceil(float(${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 ${s.texture2D}(${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],d=Math.ceil(a[o-1]/2),h=d*Math.ceil(a[o-2]/2),p="int b, int row, int col",f=`b * ${h} + (row / 2) * ${d} + (col / 2)`;for(let m=2;m<o-1;m++)p=`int b${m}, `+p,h*=a[o-m-1],f=`b${m} * ${h} + `+f;return`
vec4 ${r}(${p}) {
int index = ${f};
int texR = index / ${c};
int texC = index - texR * ${c};
vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${c}, ${u});
return ${s.texture2D}(${n}, uv);
}
`}function zhe(e,t){let n=e.shapeInfo.logicalShape,r=e.name,s="get"+r.charAt(0).toUpperCase()+r.slice(1),a=n[3],o=n[2]*a,i=n[1]*o,{newShape:l,keptDims:u}=k.squeezeShape(n);if(l.length<n.length){let A=mu(e,l),x=["row","col","depth","depth2"];return`
${pu(A,t)}
float ${s}(int row, int col, int depth, int depth2) {
return ${s}(${gu(x,u)});
}
`}if(e.shapeInfo.isUniform)return`
float ${s}(int row, int col, int depth, int depth2) {
int index = round(dot(vec4(row, col, depth, depth2),
vec4(${i}, ${o}, ${a}, 1)));
${fu(e)}
}
`;let c=e.shapeInfo.flatOffset,d=e.shapeInfo.texShape,h=d[0],p=d[1],f=`int stride2 = ${r}Shape[3];`,m=`int stride1 = ${r}Shape[2] * stride2;`,g=`int stride0 = ${r}Shape[1] * stride1;`;if(p===i&&c==null)return t?`
float ${s}(int row, int col, int depth, int depth2) {
${f}
${m}
float texR = float(row);
float texC =
dot(vec3(col, depth, depth2),
vec3(stride1, stride2, 1));
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${r}TexShape[1], ${r}TexShape[0]);
return sampleTexture(${r}, uv);
}
`:`
float ${s}(int row, int col, int depth, int depth2) {
float texR = float(row);
float texC =
dot(vec3(col, depth, depth2),
vec3(${o}, ${a}, 1));
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${p}.0, ${h}.0);
return sampleTexture(${r}, uv);
}
`;if(p===a&&c==null)return t?`
float ${s}(int row, int col, int depth, int depth2) {
float texR = dot(vec3(row, col, depth),
vec3(${r}Shape[1] * ${r}Shape[2], ${r}Shape[2], 1));
float texC = float(depth2);
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${r}TexShape[1], ${r}TexShape[0]);
return sampleTexture(${r}, uv);
}
`:`
float ${s}(int row, int col, int depth, int depth2) {
float texR = dot(vec3(row, col, depth),
vec3(${n[1]*n[2]}, ${n[2]}, 1));
float texC = float(depth2);
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${p}.0, ${h}.0);
return sampleTexture(${r}, uv);
}
`;let y=yi(r);return t?`
float ${s}(int row, int col, int depth, int depth2) {
// Explicitly use integer operations as dot() only works on floats.
${f}
${m}
${g}
int index = row * stride0 + col * stride1 +
depth * stride2 + depth2;
vec2 uv = uvFromFlat(${r}TexShape[0], ${r}TexShape[1], index + ${y});
return sampleTexture(${r}, uv);
}
`:`
float ${s}(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 * ${a} + depth2;
vec2 uv = uvFromFlat(${h}, ${p}, index + ${y});
return sampleTexture(${r}, uv);
}
`}function Lhe(e){let t=e.shapeInfo.logicalShape,n=e.name,r="get"+n.charAt(0).toUpperCase()+n.slice(1),s=t[4],a=t[3]*s,o=t[2]*a,i=t[1]*o,{newShape:l,keptDims:u}=k.squeezeShape(t);if(l.length<t.length){let m=mu(e,l),g=["row","col","depth","depth2","depth3"];return`
${pu(m)}
float ${r}(int row, int col, int depth, int depth2, int depth3) {
return ${r}(${gu(g,u)});
}
`}if(e.shapeInfo.isUniform)return`
float ${r}(int row, int col, int depth, int depth2, int depth3) {
float index = dot(
vec4(row, col, depth, depth2),
vec4(${i}, ${o}, ${a}, ${s})) +
depth3;
${fu(e)}
}
`;let c=e.shapeInfo.flatOffset,d=e.shapeInfo.texShape,h=d[0],p=d[1];if(p===i&&c==null)return`
float ${r}(int row, int col, int depth, int depth2, int depth3) {
int texR = row;
float texC = dot(vec4(col, depth, depth2, depth3),
vec4(${o}, ${a}, ${s}, 1));
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${p}.0, ${h}.0);
return sampleTexture(${n}, uv);
}
`;if(p===s&&c==null)return`
float ${r}(int row, int col, int depth, int depth2, int depth3) {
float texR = dot(
vec4(row, col, depth, depth2),
vec4(${t[1]*t[2]*t[3]},
${t[2]*t[3]}, ${t[3]}, 1));
int texC = depth3;
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${p}.0, ${h}.0);
return sampleTexture(${n}, uv);
}
`;let f=yi(n);return`
float ${r}(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 * ${s} + depth3 + ${f};
vec2 uv = uvFromFlat(${h}, ${p}, index);
return sampleTexture(${n}, uv);
}
`}function Bhe(e){let t=e.shapeInfo.logicalShape,n=e.name,r="get"+n.charAt(0).toUpperCase()+n.slice(1),{newShape:s,keptDims:a}=k.squeezeShape(t);if(s.length<t.length){let g=mu(e,s),y=["row","col","depth","depth2","depth3","depth4"];return`
${pu(g)}
float ${r}(int row, int col, int depth,
int depth2, int depth3, int depth4) {
return ${r}(${gu(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 ${r}(int row, int col, int depth,
int depth2, int depth3, int depth4) {
int index = round(dot(
vec4(row, col, depth, depth2),
vec4(${c}, ${u}, ${l}, ${i})) +
dot(
vec2(depth3, depth4),
vec2(${o}, 1)));
${fu(e)}
}
`;let d=e.shapeInfo.flatOffset,h=e.shapeInfo.texShape,p=h[0],f=h[1];if(f===c&&d==null)return`
float ${r}(int row, int col, int depth,
int depth2, int depth3, int depth4) {
int texR = row;
float texC = dot(vec4(col, depth, depth2, depth3),
vec4(${u}, ${l}, ${i}, ${o})) +
float(depth4);
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${f}.0, ${p}.0);
return sampleTexture(${n}, uv);
}
`;if(f===o&&d==null)return`
float ${r}(int row, int col, int depth,
int depth2, int depth3, int depth4) {
float texR = dot(vec4(row, col, depth, depth2),
vec4(${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, ${p}.0);
return sampleTexture(${n}, uv);
}
`;let m=yi(n);return`
float ${r}(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 + ${m};
vec2 uv = uvFromFlat(${p}, ${f}, index);
return sampleTexture(${n}, uv);
}
`}function fu(e){let t=e.name,n=k.sizeFromShape(e.shapeInfo.logicalShape);return n<2?`return ${t};`:`
for (int i = 0; i < ${n}; i++) {
if (i == index) {
return ${t}[i];
}
}
`}function Whe(e,t){let n=e.name,r=n.charAt(0).toUpperCase()+n.slice(1),s="get"+r+"AtOutCoords",a=e.shapeInfo.logicalShape.length,o=t.logicalShape.length,i=TC(e.shapeInfo.logicalShape,t.logicalShape),l=wt(o),u=o-a,c,d=["x","y","z","w","u","v"];a===0?c="":o<2&&i.length>=1?c="coords = 0;":c=i.map(A=>`coords.${d[A+u]} = 0;`).join(`
`);let h="";o<2&&a>0?h="coords":h=e.shapeInfo.logicalShape.map((A,x)=>`coords.${d[x+u]}`).join(", ");let p="return outputValue;",m=k.sizeFromShape(e.shapeInfo.logicalShape)===1,y=k.sizeFromShape(t.logicalShape)===1;if(a===1&&!m&&!y)p=`
return vec4(outputValue.xy, outputValue.xy);
`;else if(m&&!y)o===1?p=`
return vec4(outputValue.x, outputValue.x, 0., 0.);
`:p=`
return vec4(outputValue.x);
`;else if(i.length){let A=a-2,x=a-1;i.indexOf(A)>-1&&i.indexOf(x)>-1?p="return vec4(outputValue.x);":i.indexOf(A)>-1?p="return vec4(outputValue.x, outputValue.y, outputValue.x, outputValue.y);":i.indexOf(x)>-1&&(p="return vec4(outputValue.xx, outputValue.zz);")}return`
vec4 ${s}() {
${l} coords = getOutputCoords();
${c}
vec4 outputValue = get${r}(${h});
${p}
}
`}function Vhe(e,t){let n=e.name,r=n.charAt(0).toUpperCase()+n.slice(1),s="get"+r+"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&&k.arraysEqual(o,a))return`
float ${s}() {
return sampleTexture(${n}, resultUV);
}
`;let u=wt(l),c=TC(e.shapeInfo.logicalShape,t.logicalShape),d=l-i,h,p=["x","y","z","w","u","v"];i===0?h="":l<2&&c.length>=1?h="coords = 0;":h=c.map(m=>`coords.${p[m+d]} = 0;`).join(`
`);let f="";return l<2&&i>0?f="coords":f=e.shapeInfo.logicalShape.map((m,g)=>`coords.${p[g+d]}`).join(", "),`
float ${s}() {
${u} coords = getOutputCoords();
${h}
return get${r}(${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 ${e} is not yet supported`)}function H5(e,t,n){let{newShape:r}=k.squeezeShape(t),s=t.length,a=e&&s===3&&t[0]===1,o=a?t.slice(1):r,i=!e&&s>1&&!k.arraysEqual(t,n)&&r.length<s||a;return{useSqueezeShape:i,uniformShape:i?o:t}}function mu(e,t){let n=JSON.parse(JSON.stringify(e));return n.shapeInfo.logicalShape=t,n}function gu(e,t){return t.map(n=>e[n]).join(", ")}function Uhe(e,t,n,r){let s=n.map((x,b)=>{let v={logicalShape:x.shape,texShape:x.isUniform?null:x.texData.texShape,isUniform:x.isUniform,isPacked:x.isUniform?!1:x.texData.isPacked,flatOffset:null};return x.texData!=null&&x.texData.slice!=null&&x.texData.slice.flatOffset>0&&(v.flatOffset=x.texData.slice.flatOffset),{name:t.variableNames[b],shapeInfo:v}}),a=s.map(x=>x.shapeInfo),o={logicalShape:r.shape,texShape:r.texData.texShape,isUniform:!1,isPacked:r.texData.isPacked,flatOffset:null},i=ihe(s,o,t),l=e.createProgram(i),u=null,c=e.getUniformLocation(l,"NAN",!1);re().getNumber("WEBGL_VERSION")===1&&(u=e.getUniformLocation(l,"INFINITY",!1));let d=!1,h={},p={},f={};for(let x=0;x<t.variableNames.length;x++){let b=t.variableNames[x];h[b]=e.getUniformLocation(l,b,d),h[`offset${b}`]=e.getUniformLocation(l,`offset${b}`,d),t.enableShapeUniforms&&(p[`${b}Shape`]=e.getUniformLocation(l,`${b}Shape`,d),f[`${b}TexShape`]=e.getUniformLocation(l,`${b}TexShape`,d))}let m,g,y;t.enableShapeUniforms&&(m=e.getUniformLocation(l,"outShape",d),y=e.getUniformLocation(l,"outShapeStrides",d),g=e.getUniformLocation(l,"outTexShape",d));let A=[];return t.customUniforms&&t.customUniforms.forEach((x,b)=>{A[b]=e.getUniformLocation(l,x.name,d)}),{program:t,source:i,webGLProgram:l,uniformLocations:h,customUniformLocations:A,inShapeInfos:a,outShapeInfo:o,infLoc:u,nanLoc:c,inShapesLocations:p,inTexShapesLocations:f,outShapeLocation:m,outShapeStridesLocation:y,outTexShapeLocation:g}}function EC(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,r)=>{let s=n.logicalShape,a=t[r],o=a.shape;if(!k.arraysEqual(s,o))throw Error(`Binary was compiled with different shapes than the current args. Shapes ${s} and ${o} must match`);if(n.isUniform&&a.isUniform)return;let i=n.texShape,l=a.isUniform?null:a.texData.texShape;if(!k.arraysEqual(i,l))throw Error(`Binary was compiled with different texture shapes than the current args. Shape ${i} and ${l} must match`)})}function Hhe(e,t,n,r,s){t.program.enableShapeUniforms||(EC(t.inShapeInfos,n),EC([t.outShapeInfo],[r]));let a=r.texData.texture,o=r.texData.texShape;r.texData.isPacked?e.setOutputPackedMatrixTexture(a,o[0],o[1]):e.setOutputMatrixTexture(a,o[0],o[1]),e.setProgram(t.webGLProgram),re().getNumber("WEBGL_VERSION")===1&&t.infLoc!==null&&e.gl.uniform1f(t.infLoc,1/0),t.nanLoc!==null&&e.gl.uniform1f(t.nanLoc,NaN),n.forEach((l,u)=>{let c=t.program.variableNames[u],d=t.uniformLocations[c],h=t.uniformLocations[`offset${c}`],p=t.inShapesLocations[`${c}Shape`],f=t.inTexShapesLocations[`${c}TexShape`];if(p){let{uniformShape:m}=H5(t.program.packedInputs,l.shape,l.texData.texShape);switch(m.length){case 1:e.gl.uniform1iv(p,new Int32Array(m));break;case 2:e.gl.uniform2iv(p,new Int32Array(m));break;case 3:e.gl.uniform3iv(p,new Int32Array(m));break;case 4:e.gl.uniform4iv(p,new Int32Array(m));break;default:break}}if(f&&e.gl.uniform2i(f,l.texData.texShape[0],l.texData.texShape[1]),d!=null){if(l.isUniform){if(k.sizeFromShape(l.shape)<2)e.gl.uniform1f(d,l.uniformValues[0]);else{let m=l.uniformValues;m instanceof Float32Array||(m=new Float32Array(m)),e.gl.uniform1fv(d,m)}return}l.texData.slice!=null&&h!=null&&e.gl.uniform1i(h,l.texData.slice.flatOffset),e.setInputMatrixTexture(l.texData.texture,d,u)}});let i=t.outShapeLocation;if(i)switch(r.shape.length){case 1:e.gl.uniform1iv(i,new Int32Array(r.shape));break;case 2:e.gl.uniform2iv(i,new Int32Array(r.shape));break;case 3:e.gl.uniform3iv(i,new Int32Array(r.shape));break;case 4:e.gl.uniform4iv(i,new Int32Array(r.shape));break;default:break}if(t.outShapeStridesLocation){let l=k.computeStrides(r.shape);switch(r.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,r.texData.texShape[0],r.texData.texShape[1]),t.program.customUniforms&&s&&t.program.customUniforms.forEach((l,u)=>{let c=t.customUniformLocations[u],d=s[u];if(l.type==="float")e.gl.uniform1fv(c,d);else if(l.type==="vec2")e.gl.uniform2fv(c,d);else if(l.type==="vec3")e.gl.uniform3fv(c,d);else if(l.type==="vec4")e.gl.uniform4fv(c,d);else if(l.type==="int")e.gl.uniform1iv(c,d);else if(l.type==="ivec2")e.gl.uniform2iv(c,d);else if(l.type==="ivec3")e.gl.uniform3iv(c,d);else if(l.type==="ivec4")e.gl.uniform4iv(c,d);else throw Error(`uniform type ${l.type} is not supported yet.`)}),e.executeProgram()}function Ghe(e,t,n){let r="";t.concat(n).forEach(o=>{let i=o.texData!=null&&o.texData.slice!=null&&o.texData.slice.flatOffset>0;if(e.enableShapeUniforms&&!o.isUniform){let l=o.texData.texShape,{useSqueezeShape:u,uniformShape:c}=H5(e.packedInputs,o.shape,l),d="",h="",p="";if(c.length===1&&e.packedInputs){let b=[Math.ceil(l[0]/2),Math.ceil(l[1]/2)];d=`${b[0]>1}_${b[1]>1}`}else if(c.length===2&&!e.packedInputs)h=`${c[0]>1}_${c[1]>1}`;else if(c.length>2&&!e.packedInputs){let b=k.computeStrides(c);p=`${b[0]===l[1]}_${b[b.length-1]===l[1]}`}let f=o.shape.length,m=f===2&&k.arraysEqual(o.shape,l),g=k.sizeFromShape(o.shape)===1,y=_.getBroadcastDims(o.shape,n.shape),A=!e.packedInputs&&f===n.shape.length&&k.arraysEqual(l,n.texData.texShape),x=e.packedInputs||f>2?"":`${l[0]>1}_${l[1]>1}`;r+=`${f}_${A}_${u}_${c.length}_${g}_${y}_${m}_${d}_${h}_${p}_${x}_${i}`}else{let l=o.isUniform?"uniform":o.texData.texShape;r+=`${o.shape}_${l}_${i}`}});let s=e.userCode,a=e.constructor.name;return a+="_"+r+"_"+s+`${re().getNumber("WEBGL_VERSION")}`,a}function Mm(e){return re().getBool("WEBGL_USE_SHAPES_UNIFORMS")&&e<=4}var $C={};_e($C,{addImpl:()=>DC,bincountImpl:()=>Xhe,bincountReduceImpl:()=>Zhe,ceilImpl:()=>FC,concatImpl:()=>MC,equalImpl:()=>OC,expImpl:()=>PC,expm1Impl:()=>zC,floorImpl:()=>LC,gatherNdImpl:()=>Yhe,gatherV2Impl:()=>Jhe,greaterEqualImpl:()=>WC,greaterImpl:()=>BC,lessEqualImpl:()=>UC,lessImpl:()=>VC,linSpaceImpl:()=>Qhe,logImpl:()=>HC,maxImpl:()=>epe,maximumImpl:()=>GC,minimumImpl:()=>jC,multiplyImpl:()=>K5,negImpl:()=>npe,notEqualImpl:()=>qC,prodImpl:()=>spe,rangeImpl:()=>KC,rsqrtImpl:()=>XC,simpleAbsImpl:()=>jhe,sliceImpl:()=>X5,sparseFillEmptyRowsImpl:()=>ape,sparseReshapeImpl:()=>ope,sparseSegmentReductionImpl:()=>ipe,squaredDifferenceImpl:()=>ZC,stridedSliceImpl:()=>lpe,stringNGramsImpl:()=>cpe,stringSplitImpl:()=>hpe,stringToHashBucketFastImpl:()=>ppe,subImpl:()=>YC,tileImpl:()=>mpe,topKImpl:()=>gpe,transposeImpl:()=>rpe,uniqueImpl:()=>ype});function RC(e,t){Array.isArray(e)||(e=[e]),e.forEach(n=>{n!=null&&k.assert(n.dtype!=="complex64",()=>`${t} does not support complex64 tensors in the CPU backend.`)})}function jhe(e){let t=new Float32Array(e.length);for(let n=0;n<e.length;++n)t[n]=Math.abs(e[n]);return t}function Or(e){return(t,n,r,s,a)=>{let o=_.assertAndGetBroadcastShape(t,n),i=o.length,l=k.computeStrides(o),u=k.sizeFromShape(o),c=k.getTypedArrayFromDType(a,u),d=t.length,h=n.length,p=k.computeStrides(t),f=k.computeStrides(n),m=_.getBroadcastDims(t,o),g=_.getBroadcastDims(n,o);if(m.length+g.length===0)for(let y=0;y<c.length;++y)c[y]=e(r[y%r.length],s[y%s.length]);else for(let y=0;y<c.length;++y){let A=k.indexToLoc(y,i,l),x=A.slice(-d);m.forEach(w=>x[w]=0);let b=k.locToIndex(x,d,p),v=A.slice(-h);g.forEach(w=>v[w]=0);let I=k.locToIndex(v,h,f);c[y]=e(r[b],s[I])}return[c,o]}}function G5(e){let{inputs:t,backend:n}=e,{real:r,imag:s}=t,a=n.data.get(r.dataId).values,o=n.data.get(s.dataId).values,i=n.makeTensorInfo(r.shape,"complex64"),l=n.data.get(i.dataId);return l.complexTensorInfos={real:n.makeTensorInfo(r.shape,"float32",a),imag:n.makeTensorInfo(s.shape,"float32",o)},i}function j5(e,t,n="float32"){if(n==="complex64"){let s=j5(e,t,"float32"),a=j5(e,t,"float32");return G5({inputs:{real:s,imag:a},backend:e})}let r=k.makeZerosTypedArray(k.sizeFromShape(t),n);return e.makeTensorInfo(t,n,r)}function _C(e){let{inputs:t,backend:n}=e,{x:r}=t;return n.incRef(r.dataId),{dataId:r.dataId,shape:r.shape,dtype:r.dtype}}function qhe(e){let{inputs:t,backend:n}=e,{input:r}=t,s=n.data.get(r.dataId).complexTensorInfos.real,a=n.data.get(s.dataId).values;return n.makeTensorInfo(s.shape,s.dtype,a)}function Om(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{dtype:a}=r;if(a==="complex64"){if(s.dtype==="complex64")return _C({inputs:{x:s},backend:n});let o=j5(n,s.shape,s.dtype),i=Om({inputs:{x:s},backend:n,attrs:{dtype:"float32"}}),l=G5({inputs:{real:i,imag:o},backend:n});return n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}if(s.dtype==="complex64"){let o=qhe({inputs:{input:s},backend:n}),i=Om({inputs:{x:o},backend:n,attrs:{dtype:a}});return n.disposeIntermediateTensorInfo(o),i}if(!k.hasEncodingLoss(s.dtype,a)){let o=_C({inputs:{x:s},backend:n});return{dataId:o.dataId,shape:o.shape,dtype:a}}if(a==="int32"){let o=n.data.get(s.dataId).values,i=Int32Array.from(o);return n.makeTensorInfo(s.shape,"int32",i)}if(a==="bool"){let o=n.data.get(s.dataId).values,i=k.toTypedArray([0],s.dtype),[l,u]=Or((c,d)=>c!==d?1:0)(s.shape,[],o,i,"bool");return n.makeTensorInfo(u,"bool",l)}throw new Error(`Error in Cast: failed to cast ${s.dtype} to ${a}`)}function Qr(e,t,n,r){return n==null?({inputs:s,backend:a})=>{let{a:o,b:i}=s,l=a;RC([o,i],e);let u=l.data.get(o.dataId).values,c=l.data.get(i.dataId).values,d=o.dtype==="string"?_.fromUint8ToStringArray(u):u,h=o.dtype==="string"?_.fromUint8ToStringArray(c):c,p=r||o.dtype,[f,m]=t(o.shape,i.shape,d,h,p);return l.makeTensorInfo(m,p,f)}:({inputs:s,backend:a})=>{let{a:o,b:i}=s,l=a;if(o.dtype==="complex64"||i.dtype==="complex64"){let u=Om({inputs:{x:o},backend:l,attrs:{dtype:"complex64"}}),c=l.data.get(u.dataId),d=c.complexTensorInfos.real,h=c.complexTensorInfos.imag,p=l.data.get(d.dataId).values,f=l.data.get(h.dataId).values,m=Om({inputs:{x:i},backend:l,attrs:{dtype:"complex64"}}),g=l.data.get(m.dataId),y=g.complexTensorInfos.real,A=g.complexTensorInfos.imag,x=l.data.get(y.dataId).values,b=l.data.get(A.dataId).values,[v,I,w]=n(o.shape,i.shape,p,f,x,b),S=l.makeTensorInfo(w,"float32",v),E=l.makeTensorInfo(w,"float32",I),D=G5({inputs:{real:S,imag:E},backend:l});return l.disposeIntermediateTensorInfo(u),l.disposeIntermediateTensorInfo(m),l.disposeIntermediateTensorInfo(S),l.disposeIntermediateTensorInfo(E),D}else{let u=l.data.get(o.dataId).values,c=l.data.get(i.dataId).values,d=r||o.dtype,[h,p]=t(o.shape,i.shape,u,c,d);return l.makeTensorInfo(p,d,h)}}}function q5(e){return(t,n,r,s,a,o)=>{let i=_.assertAndGetBroadcastShape(t,n),l=k.sizeFromShape(i),u=i.length,c=k.computeStrides(i),d=k.getTypedArrayFromDType("float32",l),h=k.getTypedArrayFromDType("float32",l),p=_.getBroadcastDims(t,i),f=_.getBroadcastDims(n,i),m=_.mergeRealAndImagArrays(r,s),g=_.mergeRealAndImagArrays(a,o),y=t.length,A=k.computeStrides(t),x=n.length,b=k.computeStrides(n);if(p.length+f.length===0)for(let v=0;v<d.length;v++){let I=v%m.length,w=v%g.length,S=e(m[I*2],m[I*2+1],g[w*2],g[w*2+1]);d[v]=S.real,h[v]=S.imag}else for(let v=0;v<d.length;v++){let I=k.indexToLoc(v,u,c),w=I.slice(-y);p.forEach(R=>w[R]=0);let S=k.locToIndex(w,y,A),E=I.slice(-x);f.forEach(R=>E[R]=0);let D=k.locToIndex(E,x,b),$=e(m[S*2],m[S*2+1],g[D*2],g[D*2+1]);d[v]=$.real,h[v]=$.imag}return[d,h,i]}}var DC=Or((e,t)=>e+t),Khe=q5((e,t,n,r)=>({real:e+n,imag:t+r})),g7e=Qr(Ma,DC,Khe);function Xhe(e,t,n,r,s){let a=k.sizeFromShape(r),o=k.makeZerosTypedArray(s,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>=s||(a>0?o[l]+=t[i]:o[l]+=1)}return o}function Zhe(e,t,n,r=!1){let s=e.shape[0],a=e.shape[1],o=ze([s,n],t.dtype);for(let i=0;i<s;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||(r?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 yu(e){return(t,n,r)=>{let s=k.getTypedArrayFromDType(n,t.length);for(let a=0;a<t.length;++a)s[a]=e(t[a],r);return s}}function Au(e,t,n){return({inputs:r,attrs:s,backend:a})=>{let{x:o}=r;if(RC(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,s);return i.makeTensorInfo(o.shape,u,c)}}var FC=yu(e=>Math.ceil(e)),y7e=Au(No,FC);function MC(e,t,n,r){let s=k.getArrayFromDType(n,k.sizeFromShape(t));if(r&&n!=="string"){let a=0;e.forEach(o=>{let i=k.sizeFromShape(o.shape);s.set(o.vals,a),a+=i})}else{let a=0;e.forEach(o=>{let i=n==="string"?_.fromUint8ToStringArray(o.vals):o.vals,l=0;for(let u=0;u<o.shape[0];++u){let c=u*t[1]+a;for(let d=0;d<o.shape[1];++d)s[c+d]=i[l++]}a+=o.shape[1]})}return s}var OC=Or((e,t)=>e===t?1:0),A7e=Qr(il,OC,null,"bool"),PC=yu(e=>Math.exp(e)),x7e=Au(Eo,PC),zC=yu(e=>Math.expm1(e)),b7e=Au(ll,zC),LC=yu(e=>Math.floor(e)),v7e=Au($o,LC);function Yhe(e,t,n,r,s,a,o,i,l){let u=ze([r,a],n);for(let c=0;c<r;c++){let d=[],h=0;for(let p=0;p<s;p++){let f=e[c*s+p];h+=f*o[p],d.push(f)}if(h<0||h>=l/a)throw new Error(`Invalid indices: ${d} does not index into ${i}`);for(let p=0;p<a;p++)u.values[c*a+p]=t.get(...t.indexToLoc(h*a+p))}return u}function Jhe(e,t,n){let r=ze(n,e.dtype);for(let s=0;s<r.size;++s){let o=r.indexToLoc(s).slice(),i=o[0],l=o[2],u=t.locToIndex([i,l]);o[2]=t.values[u];let c=e.locToIndex(o);r.values[s]=e.values[c]}return r}var BC=Or((e,t)=>e>t?1:0),w7e=Qr(dl,BC,null,"bool"),WC=Or((e,t)=>e>=t?1:0),k7e=Qr(Ro,WC,null,"bool"),VC=Or((e,t)=>e<t?1:0),I7e=Qr(fl,VC,null,"bool"),UC=Or((e,t)=>e<=t?1:0),S7e=Qr(ml,UC,null,"bool");function Qhe(e,t,n){let r=(t-e)/(n-1),s=k.makeZerosTypedArray(n,"float32");s[0]=e;for(let a=1;a<s.length;a++)s[a]=s[a-1]+r;return s}var HC=yu(e=>Math.log(e)),T7e=Au(_o,HC);function epe(e,t,n,r){let s=k.getTypedArrayFromDType(r,k.sizeFromShape(n));for(let a=0;a<s.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)}s[a]=i}return s}var GC=Or((e,t)=>Math.max(e,t)),N7e=Qr(Do,GC),jC=Or((e,t)=>Math.min(e,t)),C7e=Qr(Fo,jC),K5=Or((e,t)=>e*t),tpe=q5((e,t,n,r)=>({real:e*n-t*r,imag:e*r+t*n})),E7e=Qr(Mo,K5,tpe);function npe(e,t,n){let r=k.createScalarValue(-1,n);return K5([],t,r,e,n)}var qC=Or((e,t)=>e!==t?1:0),$7e=Qr(vl,qC,null,"bool");function rpe(e,t,n,r,s){let a=t.length,o=k.sizeFromShape(t),i=k.computeStrides(t),l=k.computeStrides(s),u=k.getTypedArrayFromDType(n,k.sizeFromShape(s));for(let c=0;c<o;++c){let d=k.indexToLoc(c,a,i),h=new Array(d.length);for(let f=0;f<h.length;f++)h[f]=d[r[f]];let p=k.locToIndex(h,a,l);u[p]=e[c]}return u}function spe(e,t,n,r){let[s,a]=_.computeOutAndReduceShapes(e,r),o=Ur(t,"int32"),i=k.makeZerosTypedArray(k.sizeFromShape(s),o),l=k.sizeFromShape(a);for(let u=0;u<i.length;++u){let c=u*l,d=1;for(let h=0;h<l;++h)d*=n[c+h];i[u]=d}return{outVals:i,outShape:s,outDtype:o}}function KC(e,t,n,r){let s=e===t,a=e<t&&n<0,o=t<e&&n>1;if(s||a||o)return k.makeZerosTypedArray(0,r);let i=Math.abs(Math.ceil((t-e)/n)),l=k.makeZerosTypedArray(i,r);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 XC=yu(e=>1/Math.sqrt(e)),R7e=Au(Oo,XC);function X5(e,t,n,r,s){let a=Cn.isSliceContinous(r,t,n),o=k.sizeFromShape(n),i=k.computeStrides(r);if(a){let d=Cn.computeFlatOffset(t,i);return s==="string"?e.slice(d,d+o):e.subarray(d,d+o)}let l=s==="string"?_.fromUint8ToStringArray(e):e,u=ze(r,s,l),c=ze(n,s);for(let d=0;d<c.size;++d){let h=c.indexToLoc(d),p=h.map((f,m)=>f+t[m]);c.set(u.get(...p),...h)}return s==="string"?_.fromStringArrayToUint8(c.values):c.values}function ape(e,t,n,r,s,a,o){let i=t[0],l=a[0],u=new Array(l),c=new Array(i),d=t[1];if(l===0){if(i!==0)throw new Error(`Received SparseTensor with denseShape[0] = 0 but
indices.shape[0] = ${i}`);let g=k.getArrayFromDType(n,0),y=k.getArrayFromDType(s,0);return[g,[0,d],y,u,c]}let h=!0,p=0,f=new Array(l).fill(0);for(let g=0;g<i;++g){let y=e[g*d];if(y<0)throw new Error(`indices(${g}, 0) is invalid: ${y} < 0`);if(y>=l)throw new Error(`indices(${g}, 0) is invalid: ${y} >= ${l}`);++f[y],h=h&&y>=p,p=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&&h){let g=e,y=r;for(let A=0;A<i;++A)c[A]=A;return[g,[i,d],y,u,c]}else{let g=f[l-1],y=k.getArrayFromDType(n,g*d),A=k.getArrayFromDType(s,g),x=new Array(l).fill(0);for(let b=0;b<i;++b){let v=e[b*d],I=x[v],w=(v===0?0:f[v-1])+I;x[v]++;for(let S=0;S<d;++S)y[w*d+S]=e[b*d+S];A[w]=r[b],c[b]=w}for(let b=0;b<l;++b)if(x[b]===0){let I=b===0?0:f[b-1];y[I*d+0]=b;for(let w=1;w<d;++w)y[I*d+w]=0;A[I]=o}return[y,[g,d],A,u,c]}}function ope(e,t,n,r,s){let a=k.sizeFromShape(r),o=t[0],i=s.length,l=[],u=1,c=-1;for(let g=0;g<i;++g){let y=s[g];if(y===-1){if(c!==-1)throw new Error(`only one output dimension may be -1, not both ${c} and ${g}`);c=g,l.push(1)}else{if(y<0)throw new Error(`size ${g} must be non-negative, not ${y}`);u*=y,l.push(y)}}if(c!==-1){if(u<=0)throw new Error("reshape cannot infer the missing input size for an empty tensor unless all specified input sizes are non-zero");let g=Math.trunc(a/u);if(u*g!==a)throw new Error(`Input to reshape is a SparseTensor with ${a}
dense values, but the requested shape requires a multiple of ${u}. inputShape=${r} outputShape= ${l}`);l[c]=g}let d=k.sizeFromShape(l);if(d!==a)throw new Error(`Input to reshape is a tensor with ${a} dense values, but the requested shape has ${d}. inputShape=${r} outputShape=${l}`);let h=r.length,p=[];if(h>0){p[h-1]=1;for(let g=h-2;g>=0;--g)p[g]=p[g+1]*r[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=k.getArrayFromDType(n,o*i);for(let g=0;g<o;++g){let y=0;for(let A=0;A<h;++A)y+=e[g*h+A]*p[A];for(let A=0;A<i;++A)m[g*i+A]=Math.trunc(y/f[A]),y%=f[A]}return[m,[o,i],l]}function ipe(e,t,n,r,s,a=!1,o=0){let i=r.length;if(i!==s.length)throw new Error("segmentIds and indices should have same size.");let l=[t[0],e.length/t[0]],u=l[1],d=i>0?s[i-1]+1:0;if(d<0)throw new Error("segment ids must be >= 0");let h=t.slice();h[0]=d;let p=h.reduce((x,b)=>x*b,1),f=k.getArrayFromDType(n,p);if(i===0)return d>0&&f.fill(o),[f,h];if(d<=0)throw new Error("segment ids must be >= 0");let m=0,g=1,y=0,A=s[m];for(;;){let x=0;if(g<i){if(x=s[g],A===x){++g;continue}if(A>=x)throw new Error("segment ids are not increasing")}if(A<0||A>=d)throw new Error(`Segment id ${A} out of range [0, ${d}), possibly because segmentIds input is not sorted.`);A>y&&f.fill(o,y*u,A*u);for(let b=m;b<g;++b){let v=r[b];if(v<0||v>=l[0])throw new Error(`Bad: indices[${b}] == ${r[b]} out of range [0, ${l[0]})`);for(let I=0;I<u;I++)f[A*u+I]+=e[v*u+I]}if(a)for(let b=0;b<u;b++)f[A*u+b]/=g-m;if(m=g,++g,y=A+1,A=x,g>i)break}return y<d&&f.fill(o,y*u,d*u),[f,h]}var ZC=Or((e,t)=>{let n=e-t;return n*n}),_7e=Qr(Po,ZC);function lpe(e,t,n,r){let s=ze(e,t.dtype);for(let a=0;a<s.size;a++){let o=s.indexToLoc(a),i=new Array(o.length);for(let l=0;l<i.length;l++)i[l]=o[l]*n[l]+r[l];s.set(t.get(...i),...o)}return s}var upe=class{constructor(e,t,n,r,s,a){this.separator=k.encodeString(e),this.nGramWidths=t,this.leftPad=k.encodeString(n),this.rightPad=k.encodeString(r),this.padWidth=s,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,r,s,a){for(let o=0;o<s;++o){let i=this.getPadWidth(a),l=Math.max(0,i-o),u=Math.max(0,i-(s-(o+1))),c=a-(l+u),d=t+(l>0?0:o-i),h=0;h+=l*this.leftPad.length;for(let y=0;y<c;++y)h+=e[d+y].length;h+=u*this.rightPad.length,h+=(l+u+c-1)*this.separator.length,n[r+o]=new Uint8Array(h);let f=n[r+o],m=0,g=y=>y.forEach(A=>f[m++]=A);for(let y=0;y<l;++y)g(this.leftPad),g(this.separator);for(let y=0;y<c-1;++y)g(e[d+y]),g(this.separator);if(c>0){g(e[d+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,r=t.length;if(r>0){let i=t[0];if(i!==0)throw new Error(`First split value must be 0, got ${i}`);for(let l=1;l<r;++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 s=r-1,a=k.getArrayFromDType("int32",r);if(n===0||r===0){let i=new Array(n);for(let l=0;l<=s;++l)a[l]=0;return[i,a]}a[0]=0;for(let i=1;i<=s;++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[s]);for(let i=0;i<s;++i){let l=t[i],u=a[i];if(this.nGramWidths.forEach(c=>{let d=t[i+1]-t[i],h=this.getNumNGrams(d,c);this.createNGrams(e,l,o,u,h,c),u+=h}),this.preserveShort&&u===a[i]){let c=t[i+1]-t[i];if(c===0)continue;let d=c+2*this.padWidth,h=1;this.createNGrams(e,l,o,u,h,d)}}return[o,a]}};function cpe(e,t,n,r,s,a,o,i){return new upe(n,r,s,a,o,i).compute(e,t)}function dpe(e,t,n){if(!e.length)return[];if(t.length===0){let a=new Array(e.length);for(let o=0;o<e.length;++o)a[o]=e.subarray(o,o+1);return a}if(t.length===1){let a=t[0],o=[],i=e.indexOf(a);for(;i!==-1;){let l=e.subarray(0,i);(!n||l.length!==0)&&o.push(l),e=e.subarray(i+1),i=e.indexOf(a)}return(!n||e.length!==0)&&o.push(e),o}let r=[],s=0;for(let a=0;a<e.length+1;a++)if(a===e.length||t.indexOf(e[a])!==-1){let o=e.subarray(s,a);(!n||o.length!==0)&&r.push(o),s=a+1}return r}function hpe(e,t,n){let r=e.length,s=[],a=0,o=0,i=new Array(r);for(let h=0;h<r;++h){let p=dpe(e[h],t,n),f=p.length;i[h]=f,a+=f,o=Math.max(o,f),s.push(...p)}let l=k.getArrayFromDType("int32",a*2),u=new Array(a),c=[r,o],d=0;for(let h=0;h<r;++h)for(let p=0;p<i[h];++p)l[d*2]=h,l[d*2+1]=p,u[d]=s[d],++d;return[l,u,c]}function ppe(e,t){let n=k.getArrayFromDType("int32",e.length);for(let r=0;r<e.length;++r)n[r]=k.fingerPrint64(e[r]).modulo(t).getLowBitsUnsigned();return n}var YC=Or((e,t)=>e-t),fpe=q5((e,t,n,r)=>({real:e-n,imag:t-r})),D7e=Qr(zo,YC,fpe);function mpe(e,t){let n=new Array(e.rank);for(let s=0;s<n.length;s++)n[s]=e.shape[s]*t[s];let r=ze(n,e.dtype);for(let s=0;s<r.values.length;++s){let a=r.indexToLoc(s),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);r.values[s]=e.values[i]}return r}var lh=(e,t)=>{let n=t.value-e.value;return n===0?e.index-t.index:n};function JC(e,t,n=0,r=e.length-1){for(;r>n;){if(r-n>600){let i=r-n+1,l=t-n+1,u=Math.log(i),c=.5*Math.exp(2*u/3),d=.5*Math.sqrt(u*c*(i-c)/i)*Math.sign(l-i/2),h=Math.max(n,Math.floor(t-l*c/i+d)),p=Math.min(r,Math.floor(t+(i-l)*c/i+d));JC(e,t,h,p)}let s=e[t],a=n,o=r;for(k.swap(e,n,t),lh(e[r],s)>0&&k.swap(e,n,r);a<o;){for(k.swap(e,a,o),a++,o--;lh(e[a],s)<0;)a=a+1;for(;lh(e[o],s)>0;)o=o-1}lh(e[n],s)===0?k.swap(e,n,o):(o=o+1,k.swap(e,o,r)),o<=t&&(n=o+1),t<=o&&(r=o-1)}}function gpe(e,t,n,r,s){let a=t[t.length-1],[o,i]=[e.length/a,a],l=k.getTypedArrayFromDType(n,o*r),u=k.getTypedArrayFromDType("int32",o*r);for(let d=0;d<o;d++){let h=d*i,p=e.subarray(h,h+i),f=new Array(p.length);p.forEach((A,x)=>f[x]={value:A,index:x}),r<f.length&&(JC(f,r),f=f.slice(0,r)),s&&f.sort(lh);let m=d*r,g=l.subarray(m,m+r),y=u.subarray(m,m+r);for(let A=0;A<r;A++)g[A]=f[A].value,y[A]=f[A].index}let c=t.slice();return c[c.length-1]=r,[ze(c,n,l),ze(c,"int32",u)]}function ype(e,t,n,r){let s=k.parseAxisParam(t,n)[0],a=[1,n[0],1];for(let f=0;f<s;f++)a[0]*=n[f];a[1]=n[s];for(let f=s+1;f<n.length;f++)a[2]*=n[f];let o={},i=new Int32Array(n[s]),l=new Jt(a,r,e),u=[],c=a[0]===1&&a[2]===1;for(let f=0;f<n[s];f++){let m;if(c)m=e[f].toString();else{let g=[];for(let y=0;y<a[0];y++)for(let A=0;A<a[2];A++)g.push(l.get(y,f,A));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 d=a.slice();d[1]=Object.keys(o).length;let h=new Jt(d,r);u.forEach((f,m)=>{for(let g=0;g<a[0];g++)for(let y=0;y<a[2];y++)h.set(l.get(g,f,y),g,m,y)});let p=n.slice();return p[s]=d[1],{outputValues:h.values,outputShape:p,indices:i}}var{addImpl:Ape,bincountImpl:QC,bincountReduceImpl:xpe,ceilImpl:bpe,concatImpl:vpe,equalImpl:wpe,expImpl:kpe,expm1Impl:Ipe,floorImpl:Spe,gatherNdImpl:Tpe,gatherV2Impl:Npe,greaterImpl:Cpe,greaterEqualImpl:Epe,lessImpl:$pe,lessEqualImpl:Rpe,linSpaceImpl:_pe,logImpl:Dpe,maxImpl:Fpe,maximumImpl:Mpe,minimumImpl:Ope,multiplyImpl:Ppe,negImpl:zpe,notEqualImpl:Lpe,prodImpl:Bpe,rangeImpl:Wpe,rsqrtImpl:Vpe,simpleAbsImpl:eE,sliceImpl:Upe,sparseFillEmptyRowsImpl:Hpe,sparseReshapeImpl:Gpe,sparseSegmentReductionImpl:tE,stridedSliceImpl:jpe,stringNGramsImpl:qpe,stringSplitImpl:Kpe,stringToHashBucketFastImpl:Xpe,subImpl:Zpe,tileImpl:Ype,topKImpl:Jpe,transposeImpl:Z5,uniqueImpl:Qpe}=$C;function nE(e,t){return["x","y","z","w","u","v"].slice(0,t).map(n=>`${e}.${n}`)}function Vn(e,t){return t===1?[e]:nE(e,t)}function efe(e,t){if(e===1)return"rc";let n="";for(let r=0;r<e;r++)n+=t[r],r<e-1&&(n+=",");return n}var tfe=class{constructor(e){this.variableNames=["A"],this.packedInputs=!1,this.packedOutput=!0,this.outputShape=e;let t=e.length;if(t===0)this.userCode=`
void main() {
setOutput(vec4(getA(), 0., 0., 0.));
}
`;else{let n=Vn("rc",t),r=wt(t),s=rfe(t,e,n),a=sfe(t,e[e.length-1],e[e.length-2],n),o=afe(e,n);this.userCode=`
void main() {
${r} rc = getOutputCoords();
if(${s}) {
setOutput(vec4(0));
} else {
${a}
setOutput(vec4(${o}));
}
}
`}}};function nfe(e,t){let n=[];for(let r=0;r<=1;r++)for(let s=0;s<=1;s++){let a=`${r===0?"r":"rp1"}, ${s===0?"c":"cp1"}`;for(let o=2;o<e;o++)a=`${t[t.length-1-o]},`+a;n.push(a)}return n}function rfe(e,t,n){if(e===1)return`rc > ${t[0]}`;let r="";for(let s=e-2;s<e;s++)r+=`${n[s]} >= ${t[s]}`,s<e-1&&(r+="||");return r}function sfe(e,t,n,r){if(e===1)return"";let s=r.slice(-2);return`
int r = ${s[0]};
int c = ${s[1]};
int rp1 = r + 1;
int cp1 = c + 1;
bool cEdge = cp1 >= ${t};
bool rEdge = rp1 >= ${n};
`}function afe(e,t){let n=e.length,r=nfe(n,t);return n===1?`getA(rc),
rc + 1 >= ${e[0]} ? 0. : getA(rc + 1),
0, 0`:`getA(${r[0]}),
cEdge ? 0. : getA(${r[1]}),
rEdge ? 0. : getA(${r[2]}),
rEdge || cEdge ? 0. : getA(${r[3]})`}var rE=class{constructor(e,t){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e;let n="";for(let r=0;r<4;r++){let s="thisRC = rc;";r%2==1&&(s+="thisRC.z += 1;"),r>1&&(s+="thisRC.y += 1;"),n+=`
${s}
${r>0?"if(thisRC.y < rows && thisRC.z < cols){":""}
int flatIndex = getFlatIndex(thisRC);
ivec3 inputRC = inputCoordsFromReshapedOutCoords(flatIndex);
vec2 inputRCInnerDims = vec2(float(inputRC.y),float(inputRC.z));
result[${r}] =
getChannel(getA(inputRC.x, inputRC.y, inputRC.z), inputRCInnerDims);
${r>0?"}":""}
`}this.userCode=`
${ofe(t)}
${z5(e)}
void main() {
ivec3 rc = getOutputCoords();
vec4 result = vec4(0.);
ivec3 thisRC;
int rows = ${e[1]};
int cols = ${e[2]};
${n}
setOutput(result);
}
`}};function ofe(e){return`
ivec3 inputCoordsFromReshapedOutCoords(int index) {
${gi(["r","c","d"],e)}
return ivec3(r, c, d);
}
`}var ife=class{constructor(e){this.gpgpu=e,this.numUsedTextures=0,this.numFreeTextures=0,this._numBytesAllocated=0,this._numBytesFree=0,this.freeTextures={},this.logEnabled=!1,this.usedTextures={}}acquireTexture(e,t,n){let r=aE(t,n),s=oE(e,r,n);s in this.freeTextures||(this.freeTextures[s]=[]),s in this.usedTextures||(this.usedTextures[s]=[]);let a=sE(e,r,this.gpgpu.gl,this.gpgpu.textureConfig,n);if(this.freeTextures[s].length>0){this.numFreeTextures--,this.numUsedTextures++,this._numBytesFree-=a,this.log();let i=this.freeTextures[s].shift();return this.usedTextures[s].push(i),i}let o;return r===Sn.PACKED_2X2_FLOAT32?o=this.gpgpu.createPackedMatrixTexture(e[0],e[1]):r===Sn.PACKED_2X2_FLOAT16?o=this.gpgpu.createFloat16PackedMatrixTexture(e[0],e[1]):r===Sn.UNPACKED_FLOAT32?o=this.gpgpu.createFloat32MatrixTexture(e[0],e[1]):r===Sn.UNPACKED_FLOAT16?o=this.gpgpu.createFloat16MatrixTexture(e[0],e[1]):r===Sn.PACKED_4X1_UNSIGNED_BYTE&&(o=this.gpgpu.createUnsignedBytesMatrixTexture(e[0],e[1])),this.usedTextures[s].push(o),this.numUsedTextures++,this._numBytesAllocated+=a,this.log(),o}releaseTexture(e,t,n,r){if(this.freeTextures==null)return;let s=aE(n,r),a=oE(t,s,r);a in this.freeTextures||(this.freeTextures[a]=[]);let o=sE(t,s,this.gpgpu.gl,this.gpgpu.textureConfig,r),i=re().get("WEBGL_DELETE_TEXTURE_THRESHOLD");i!==-1&&this._numBytesAllocated>i?(this.gpgpu.deleteMatrixTexture(e),this._numBytesAllocated-=o):(this.freeTextures[a].push(e),this.numFreeTextures++,this._numBytesFree+=o),this.numUsedTextures--;let l=this.usedTextures[a],u=l.indexOf(e);if(u<0)throw new Error("Cannot release a texture that was never provided by this texture manager");l.splice(u,1),this.log()}log(){if(!this.logEnabled)return;let e=this.numFreeTextures+this.numUsedTextures;console.log("Free/Used",`${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)});for(let e in this.usedTextures)this.usedTextures[e].forEach(t=>{this.gpgpu.deleteMatrixTexture(t)});this.freeTextures=null,this.usedTextures=null,this.numUsedTextures=0,this.numFreeTextures=0,this._numBytesAllocated=0,this._numBytesFree=0}}};function lfe(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;throw new Error(`Unknown internal format ${t}`)}function sE(e,t,n,r,s){let a=ufe(t,r),o;if(s){let[l,u]=du(e[0],e[1]);o=l*u}else{let[l,u]=nh(e[0],e[1]);o=l*u}let i=lfe(n,a);return o*i}function ufe(e,t){switch(e){case Sn.PACKED_2X2_FLOAT32:return V5(t);case Sn.PACKED_2X2_FLOAT16:return U5(t);case Sn.UNPACKED_FLOAT32:return L5(t);case Sn.UNPACKED_FLOAT16:return B5(t);case Sn.PACKED_4X1_UNSIGNED_BYTE:return W5(t);default:throw new Error(`Unknown physical texture type ${e}`)}}function cfe(e){return re().getBool("WEBGL_RENDER_FLOAT32_ENABLED")?e?Sn.PACKED_2X2_FLOAT32:Sn.UNPACKED_FLOAT32:e?Sn.PACKED_2X2_FLOAT16:Sn.UNPACKED_FLOAT16}function aE(e,t){if(e===Fr.UPLOAD)return Sn.PACKED_2X2_FLOAT32;if(e===Fr.RENDER||e==null)return cfe(t);if(e===Fr.DOWNLOAD||e===Fr.PIXELS)return Sn.PACKED_4X1_UNSIGNED_BYTE;throw new Error(`Unknown logical texture type ${e}`)}function oE(e,t,n){return`${e[0]}_${e[1]}_${t}_${n}`}var Qa=class{constructor(e,t){this.variableNames=["A"],this.outputShape=e,this.enableShapeUniforms=Mm(this.outputShape.length),this.userCode=`
float unaryOperation(float x) {
${t}
}
void main() {
float x = getAAtOutCoords();
float y = unaryOperation(x);
setOutput(y);
}
`}},gs="if (isnan(x)) return x;",dfe="return x;",iE="return abs(x);",hfe="return (x >= 0.0) ? x : (exp(x) - 1.0);",pfe=gs+`
return (x < 0.0) ? 0.0 : x;
`,ffe=gs+`
return (x < 0.0) ? 0.0 : min(6.0, x);
`,Pm="return x;",mfe="return 1.0 / (1.0 + exp(-1.0 * x));",gfe="return x;",yfe=`
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;
`,Afe=`
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;
`,xfe=`
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;
`,bfe="return 1.0 / (1.0 + exp(-1.0 * x));",xu=class{constructor(e,t){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e,this.enableShapeUniforms=Mm(this.outputShape.length),this.userCode=`
vec4 unaryOperation(vec4 x) {
${t}
}
void main() {
vec4 x = getAAtOutCoords();
vec4 y = unaryOperation(x);
setOutput(y);
}
`}},vfe=class{constructor(e){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!1,this.outputShape=e;let t=e.length,n=Vn("rc",t),r=wt(t),s=efe(t,n),a=n.slice(-2),o=t<=1?"rc":`vec2(${a.join(",")})`;this.userCode=`
void main() {
${r} rc = getOutputCoords();
vec4 packedInput = getA(${s});
setOutput(getChannel(packedInput, ${o}));
}
`}},wfe=da.whereImpl,kfe=1e-7,Ife=1e-4,zm={};function Sfe(e){return e in zm||(zm[e]={}),zm[e]}var Tfe=re().getNumber("CPU_HANDOFF_SIZE_THRESHOLD"),Nfe=600;function Cfe(){return re().global.screen==null?1024:re().global.screen.height*re().global.screen.width*window.devicePixelRatio*Nfe/1024/1024}var lE=class extends Lp{constructor(e){super();if(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,!re().getBool("HAS_WEBGL"))throw new Error("WebGL is not supported on this device");if(e==null){let t=Bs(re().getNumber("WEBGL_VERSION"));this.binaryCache=Sfe(re().getNumber("WEBGL_VERSION")),this.gpgpu=new Fm(t),this.canvas=t.canvas,this.gpgpuCreatedLocally=!0}else this.gpgpu=e,this.binaryCache={},this.gpgpuCreatedLocally=!1,this.canvas=e.gl.canvas;this.textureManager=new ife(this.gpgpu),this.numMBBeforeWarning=Cfe(),this.texData=new d1(this,Ba())}nextDataId(){return lE.nextDataId++}numDataIds(){return this.texData.numDataIds()-this.pendingDeletes}write(e,t,n){if((re().getBool("WEBGL_CHECK_NUMERICAL_PROBLEMS")||re().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 r={id:this.nextDataId()};return this.texData.set(r,{shape:t,dtype:n,values:e,usage:Fr.UPLOAD,refCount:1}),r}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,r,s){if(re().getBool("DEBUG")&&this.checkNumericalProblems(t),r==="complex64")throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");this.texData.set(e,{shape:n,dtype:r,values:t,usage:Fr.UPLOAD,refCount:s})}disposeIntermediateTensorInfo(e){this.disposeData(e.dataId)}readSync(e){let t=this.texData.get(e),{values:n,dtype:r,complexTensorInfos:s,slice:a,shape:o,isPacked:i}=t;if(a!=null){let d;i?d=new xu(o,Pm):d=new Qa(o,Pm);let h=this.runWebGLProgram(d,[{dataId:e,shape:o,dtype:r}],r),p=this.readSync(h.dataId);return this.disposeIntermediateTensorInfo(h),p}if(n!=null)return this.convertAndCacheOnCPU(e);if(r==="string")return n;let l=this.activeTimers!=null,u;l&&(u=k.now());let c;if(r==="complex64"){let d=this.readSync(s.real.dataId),h=this.readSync(s.imag.dataId);c=_.mergeRealAndImagArrays(d,h)}else c=this.getValuesFromTexture(e);return l&&(this.downloadWaitMs+=k.now()-u),this.convertAndCacheOnCPU(e,c)}async read(e){if(this.pendingRead.has(e)){let p=this.pendingRead.get(e);return new Promise(f=>p.push(f))}let t=this.texData.get(e),{values:n,shape:r,slice:s,dtype:a,complexTensorInfos:o,isPacked:i}=t;if(s!=null){let p;i?p=new xu(r,Pm):p=new Qa(r,Pm);let f=this.runWebGLProgram(p,[{dataId:e,shape:r,dtype:a}],a),m=this.read(f.dataId);return this.disposeIntermediateTensorInfo(f),m}if(n!=null)return this.convertAndCacheOnCPU(e);if(!re().getBool("WEBGL_DOWNLOAD_FLOAT_ENABLED")&&re().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"&&re().get("WEBGL_BUFFER_SUPPORTED")){u=this.decode(e);let p=this.texData.get(u.dataId);l=this.gpgpu.createBufferFromTexture(p.texture,...rh(r))}this.pendingRead.set(e,[]),a!=="complex64"&&await this.gpgpu.createAndWaitForFence();let c;if(a==="complex64"){let p=await Promise.all([this.read(o.real.dataId),this.read(o.imag.dataId)]),f=p[0],m=p[1];c=_.mergeRealAndImagArrays(f,m)}else if(l==null)c=this.getValuesFromTexture(e);else{let p=k.sizeFromShape(r);c=this.gpgpu.downloadFloat32MatrixFromBuffer(l,p)}if(u!=null&&this.disposeIntermediateTensorInfo(u),l!=null){let p=this.gpgpu.gl;Ie(p,()=>p.deleteBuffer(l))}let d=this.convertAndCacheOnCPU(e,c),h=this.pendingRead.get(e);return this.pendingRead.delete(e),h.forEach(p=>p(d)),this.pendingDisposal.has(e)&&(this.pendingDisposal.delete(e),this.disposeData(e)&&Ba().removeDataId(e,this),this.pendingDeletes--),d}bufferSync(e){let t=this.readSync(e.dataId),n=t;if(e.dtype==="string")try{n=t.map(r=>k.decodeString(r))}catch(r){throw new Error("Failed to decode encoded string bytes into utf-8")}return ze(e.shape,e.dtype,n)}checkNumericalProblems(e){if(e!=null)for(let t=0;t<e.length;t++){let n=e[t];if(!PN(n))throw re().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 ${n} cannot be represented on this device.`)}}getValuesFromTexture(e){let{shape:t,dtype:n,isPacked:r}=this.texData.get(e),s=k.sizeFromShape(t);if(re().getBool("WEBGL_DOWNLOAD_FLOAT_ENABLED")){let d=this.decode(e),h=this.texData.get(d.dataId),p=this.gpgpu.downloadMatrixFromPackedTexture(h.texture,...rh(t)).subarray(0,s);return this.disposeIntermediateTensorInfo(d),p}let a=re().getBool("WEBGL_PACK")&&r===!0,o=a?$m(t):t,i=a?new rhe(o):new nhe(o),l=this.runWebGLProgram(i,[{shape:o,dtype:n,dataId:e}],"float32"),u=this.texData.get(l.dataId),c=this.gpgpu.downloadByteEncodedFloatMatrixFromOutputTexture(u.texture,u.texShape[0],u.texShape[1]).subarray(0,s);return this.disposeIntermediateTensorInfo(l),c}timerAvailable(){return re().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0}async time(e){let t=this.activeTimers,n=[],r=!1;this.programTimersStack==null?(this.programTimersStack=n,r=!0):this.activeTimers.push(n),this.activeTimers=n,e();let s=k.flatten(this.activeTimers.map(i=>i.query)).filter(i=>i!=null),a=k.flatten(this.activeTimers.map(i=>i.name)).filter(i=>i!=null);this.activeTimers=t,r&&(this.programTimersStack=null);let o={uploadWaitMs:this.uploadWaitMs,downloadWaitMs:this.downloadWaitMs,kernelMs:null,wallMs:null};if(re().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0){let i=await Promise.all(s);o.kernelMs=k.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 re().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0?this.gpgpu.beginQuery():{startMs:k.now(),endMs:null}}endTimer(e){return re().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE")>0?(this.gpgpu.endQuery(),e):(e.endMs=k.now(),e)}async getQueryTime(e){if(re().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:r,usage:s,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(r,n),this.textureManager.releaseTexture(t,r,s,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}getDataInfo(e){return this.texData.get(e)}shouldExecuteOnCPU(e,t=Tfe){return re().getBool("WEBGL_CPU_FORWARD")&&e.every(n=>this.texData.get(n.dataId).texture==null&&k.sizeFromShape(n.shape)<t)}getGPGPUContext(){return this.gpgpu}where(e){_.warn("tf.where() in webgl locks the UI thread. Call tf.whereAsync() instead");let t=e.dataSync();return wfe(e.shape,t)}packedUnaryOp(e,t,n){let r=new xu(e.shape,t),s=this.compileAndRun(r,[e],n);return Ba().makeTensorFromDataId(s.dataId,s.shape,s.dtype)}abs(e){if(this.shouldExecuteOnCPU([e])&&e.dtype!=="complex64"){let r=eE(this.texData.get(e.dataId).values);return this.makeOutput(e.shape,e.dtype,r)}if(re().getBool("WEBGL_PACK_UNARY_OPERATIONS"))return this.packedUnaryOp(e,iE,e.dtype);let t=new Qa(e.shape,iE),n=this.compileAndRun(t,[e]);return Ba().makeTensorFromDataId(n.dataId,n.shape,n.dtype)}makeTensorInfo(e,t,n){let r;if(t==="string"&&n!=null&&n.length>0&&k.isString(n[0])){let s=n.map(a=>k.encodeString(a));r=this.write(s,e,t)}else r=this.write(n,e,t);return this.texData.get(r).usage=null,{dataId:r,shape:e,dtype:t}}makeOutput(e,t,n){let{dataId:r}=this.makeTensorInfo(e,t,n);return Ba().makeTensorFromDataId(r,e,t,this)}unpackTensor(e){let t=new vfe(e.shape);return this.runWebGLProgram(t,[e],e.dtype)}packTensor(e){let t=new tfe(e.shape),n=!0;return this.runWebGLProgram(t,[e],e.dtype,null,n)}packedReshape(e,t){let n=[fi(e.shape),...mi(e.shape)],r={dtype:e.dtype,shape:n,dataId:e.dataId},s=[fi(t),...mi(t)],a=new rE(s,n),o=!0,i=this.runWebGLProgram(a,[r],e.dtype,null,o);return{dataId:i.dataId,shape:t,dtype:i.dtype}}decode(e){let t=this.texData.get(e),{isPacked:n,shape:r,dtype:s}=t,a=$m(r),o;n?o=new the(a):o=new ehe(a);let i=!0,l=this.runWebGLProgram(o,[{shape:a,dtype:s,dataId:e}],s,null,i);return{dtype:s,shape:r,dataId:l.dataId}}runWebGLProgram(e,t,n,r,s=!1){let a=this.makeTensorInfo(e.outputShape,n),o=this.texData.get(a.dataId);if(e.packedOutput&&(o.isPacked=!0),e.outPackingScheme===th.DENSE){let m=rh(e.outputShape);o.texShape=m.map(g=>g*2)}if(e.outTexUsage!=null&&(o.usage=e.outTexUsage),k.sizeFromShape(a.shape)===0)return o.values=k.getTypedArrayFromDType(a.dtype,0),a;let i=[],l=t.map(m=>{if(m.dtype==="complex64")throw new Error("GPGPUProgram does not support complex64 input. For complex64 dtypes, please separate the program into real and imaginary parts.");let g=this.texData.get(m.dataId);if(g.texture==null){if(!e.packedInputs&&k.sizeFromShape(m.shape)<=re().getNumber("WEBGL_SIZE_UPLOAD_UNIFORM"))return{shape:m.shape,texData:null,isUniform:!0,uniformValues:g.values};e.packedInputs&&(g.isPacked=!0,g.shape=m.shape)}else if(!!g.isPacked!=!!e.packedInputs)m=g.isPacked?this.unpackTensor(m):this.packTensor(m),i.push(m),g=this.texData.get(m.dataId);else if(g.isPacked&&!oh(g.shape,m.shape)){let y=m,A=m.shape;m.shape=g.shape,m=this.packedReshape(m,A),i.push(m),g=this.texData.get(m.dataId),y.shape=A}return this.uploadToGPU(m.dataId),{shape:m.shape,texData:g,isUniform:!1}});this.uploadToGPU(a.dataId);let u={shape:a.shape,texData:o,isUniform:!1},c=Ghe(e,l,u),d=this.getAndSaveBinary(c,()=>Uhe(this.gpgpu,e,l,u)),h=this.activeTimers!=null,p;h&&(p=this.startTimer()),Hhe(this.gpgpu,d,l,u,r),i.forEach(m=>this.disposeIntermediateTensorInfo(m)),h&&(p=this.endTimer(p),this.activeTimers.push({name:e.constructor.name,query:this.getQueryTime(p)}));let f=re().get("WEBGL_FLUSH_THRESHOLD");if(f>0){let m=k.now();m-this.lastGlFlushTime>f&&(this.gpgpu.gl.flush(),this.lastGlFlushTime=m)}if(!re().getBool("WEBGL_LAZILY_UNPACK")&&o.isPacked&&s===!1){let m=this.unpackTensor(a);return this.disposeIntermediateTensorInfo(a),m}return a}compileAndRun(e,t,n,r,s=!1){return n=n||t[0].dtype,this.runWebGLProgram(e,t,n,r,s)}getAndSaveBinary(e,t){return e in this.binaryCache||(this.binaryCache[e]=t()),this.binaryCache[e]}getTextureManager(){return this.textureManager}dispose(){this.disposed||(re().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(!re().get("WEBGL_RENDER_FLOAT32_ENABLED")){let e=re().getBool("DEBUG");re().set("DEBUG",!1);let t=this.abs(De(1e-8)).dataSync()[0];if(re().set("DEBUG",e),t>0)return 32}return 16})),this.floatPrecisionValue}epsilon(){return this.floatPrecision()===32?kfe:Ife}uploadToGPU(e){let t=this.texData.get(e),{shape:n,dtype:r,values:s,texture:a,usage:o,isPacked:i}=t;if(a!=null)return;let l=this.activeTimers!=null,u;l&&(u=k.now());let c=t.texShape;if(c==null&&(c=eC(n,i),t.texShape=c),s!=null){let d=$m(n),h,p=c[1],f=c[0],m=s instanceof Uint8Array;i?([p,f]=du(c[0],c[1]),h=new ahe(d,[f,p],m)):h=new she(d,[f,p],m);let g=this.makeTensorInfo([f,p],r);m?this.texData.get(g.dataId).usage=Fr.PIXELS:this.texData.get(g.dataId).usage=Fr.UPLOAD,this.gpgpu.uploadDenseMatrixToTexture(this.getTexture(g.dataId),p,f,s);let y=!0,A=this.runWebGLProgram(h,[g],r,null,y),x=this.texData.get(A.dataId);t.texture=x.texture,t.texShape=x.texShape,t.isPacked=x.isPacked,t.usage=x.usage,this.disposeIntermediateTensorInfo(g),this.texData.delete(A.dataId),t.values=null,l&&(this.uploadWaitMs+=k.now()-u)}else{let d=this.acquireTexture(c,o,r,i);t.texture=d}}convertAndCacheOnCPU(e,t){let n=this.texData.get(e),{dtype:r}=n;return this.releaseGPUData(e),t!=null&&(n.values=Efe(t,r)),n.values}acquireTexture(e,t,n,r){if(this.numBytesInGPU+=this.computeBytes(e,n),!this.warnedAboutMemory&&this.numBytesInGPU>this.numMBBeforeWarning*1024*1024){let s=(this.numBytesInGPU/1024/1024).toFixed(2);this.warnedAboutMemory=!0,console.warn(`High memory usage in GPU: ${s} MB, most likely due to a memory leak`)}return this.textureManager.acquireTexture(e,t,r)}computeBytes(e,t){return e[0]*e[1]*k.bytesPerElement(t)}},uh=lE;uh.nextDataId=0;function Efe(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 r=0;r<n.length;++r)n[r]=Math.round(e[r]);return n}else throw new Error(`Unknown dtype ${t}`)}var $fe="3.8.0";function uE(){re().set("WEBGL_FORCE_F16_TEXTURES",!0)}mf.isBrowser()&&Cy("webgl",()=>new uh,2);var Rfe={forceHalfFloat:uE},cE=`
if (isnan(a)) return a;
if (isnan(b)) return b;
`,bu=class{constructor(e,t,n){this.variableNames=["A","B"],this.outputShape=_.assertAndGetBroadcastShape(t,n),this.enableShapeUniforms=Mm(this.outputShape.length),this.userCode=`
float binaryOperation(float a, float b) {
${e}
}
void main() {
float a = getAAtOutCoords();
float b = getBAtOutCoords();
setOutput(binaryOperation(a, b));
}
`}},Lm=`
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;
`,ch=class{constructor(e,t,n,r=!1){this.variableNames=["A","B"],this.supportsBroadcasting=!0,this.packedInputs=!0,this.packedOutput=!0,this.outputShape=_.assertAndGetBroadcastShape(t,n);let s=this.outputShape.length;this.enableShapeUniforms=Mm(s);let a="";if(r)if(s===0||k.sizeFromShape(this.outputShape)===1)a=`
result.y = 0.;
result.z = 0.;
result.w = 0.;
`;else if(a=`
${wt(s)} coords = getOutputCoords();
`,s===1)this.enableShapeUniforms?a+=`
result.y = (coords + 1) >= outShape ? 0. : result.y;
result.z = 0.;
result.w = 0.;
`:a+=`
result.y = (coords + 1) >= ${this.outputShape[0]} ? 0. : result.y;
result.z = 0.;
result.w = 0.;
`;else{let i=Vn("coords",s);this.enableShapeUniforms?a+=`
bool nextRowOutOfBounds =
(${i[s-2]} + 1) >= outShape[${s} - 2];
bool nextColOutOfBounds =
(${i[s-1]} + 1) >= outShape[${s} - 1];
result.y = nextColOutOfBounds ? 0. : result.y;
result.z = nextRowOutOfBounds ? 0. : result.z;
result.w = nextColOutOfBounds || nextRowOutOfBounds ? 0. : result.w;
`:a+=`
bool nextRowOutOfBounds =
(${i[s-2]} + 1) >= ${this.outputShape[s-2]};
bool nextColOutOfBounds =
(${i[s-1]} + 1) >= ${this.outputShape[s-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 fr(e){let{inputs:t,backend:n}=e,{x:r}=t;return n.incRef(r.dataId),{dataId:r.dataId,shape:r.shape,dtype:r.dtype}}var _fe={kernelName:hl,backendName:"webgl",kernelFunc:fr};function eo(e){let{inputs:t,backend:n}=e,{real:r,imag:s}=t,a=n.makeTensorInfo(r.shape,"complex64"),o=n.texData.get(a.dataId),i=fr({inputs:{x:r},backend:n}),l=fr({inputs:{x:s},backend:n});return o.complexTensorInfos={real:i,imag:l},a}var Dfe={kernelName:v1,backendName:"webgl",kernelFunc:eo},dE="return (a < 0.) ? b * a : a;",hE=`
vec4 aLessThanZero = vec4(lessThan(a, vec4(0.)));
return (aLessThanZero * (b * a)) + ((vec4(1.0) - aLessThanZero) * a);
`;function Ffe(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{alpha:a}=r,o=n.makeTensorInfo([],"float32",k.createScalarValue(a,"float32")),i=re().getBool("WEBGL_PACK_BINARY_OPERATIONS")?new ch(hE,s.shape,o.shape):new bu(dE,s.shape,o.shape),l=n.runWebGLProgram(i,[s,o],s.dtype);return n.disposeIntermediateTensorInfo(o),l}var Mfe={kernelName:pl,backendName:"webgl",kernelFunc:Ffe},pE="return (a < 0.) ? b * a : a;",fE=`
vec4 aLessThanZero = vec4(lessThan(a, vec4(0.)));
return (aLessThanZero * (b * a)) + ((vec4(1.0) - aLessThanZero) * a);
`;function Ofe(e){let{inputs:t,backend:n}=e,{x:r,alpha:s}=t,a=re().getBool("WEBGL_PACK_BINARY_OPERATIONS")?new ch(fE,r.shape,s.shape):new bu(pE,r.shape,s.shape);return n.runWebGLProgram(a,[r,s],r.dtype)}var Pfe={kernelName:Sl,backendName:"webgl",kernelFunc:Ofe},mE="if (isnan(x)) return x;",zfe=`
if (isnan(a)) return a;
if (isnan(b)) return b;
`,Lfe=`
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 it({opSnippet:e,packedOpSnippet:t,cpuKernelImpl:n,dtype:r}){return({inputs:s,backend:a})=>{let{x:o}=s,i=a,l=r||o.dtype;if(i.shouldExecuteOnCPU([o])&&n!=null){let d=i.texData.get(o.dataId),h=n(d.values,l);return i.makeTensorInfo(o.shape,l,h)}let u=re().getBool("WEBGL_PACK_UNARY_OPERATIONS")&&t!=null,c;return u?c=new xu(o.shape,t):c=new Qa(o.shape,e),i.runWebGLProgram(c,[o],l)}}function Tn({opSnippet:e,packedOpSnippet:t,checkOutOfBounds:n=!1,supportsComplex:r=!1,cpuKernelImpl:s,dtype:a}){return({inputs:o,backend:i})=>{let{a:l,b:u}=o,c=i;if(r&&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(x=>{let[b,v]=x,I={dataId:b.dataId,dtype:b.dtype,shape:l.shape},w={dataId:v.dataId,dtype:v.dtype,shape:u.shape},S=new bu(e,l.shape,u.shape);return c.runWebGLProgram(S,[I,w],Ur(b.dtype,v.dtype))}),A=eo({inputs:{real:g,imag:y},backend:c});return c.disposeIntermediateTensorInfo(g),c.disposeIntermediateTensorInfo(y),A}let d=a||Ur(l.dtype,u.dtype);if((l.dtype==="string"||u.dtype==="string"||c.shouldExecuteOnCPU([l,u]))&&s!=null){let f=c.texData.get(l.dataId).values,m=c.texData.get(u.dataId).values,g=l.dtype==="string"?_.fromUint8ToStringArray(f):f,y=l.dtype==="string"?_.fromUint8ToStringArray(m):m,[A,x]=s(l.shape,u.shape,g,y,d),b=c.makeTensorInfo(x,d),v=c.texData.get(b.dataId);return v.values=A,b}let h=re().getBool("WEBGL_PACK_BINARY_OPERATIONS")&&t!=null,p;return h?p=new ch(t,l.shape,u.shape,n):p=new bu(e,l.shape,u.shape),c.runWebGLProgram(p,[l,u],d)}}function Bm(e,t=!1){if(e==="linear")return t?gfe:dfe;if(e==="relu")return t?Afe:pfe;if(e==="elu")return t?yfe:hfe;if(e==="relu6")return t?xfe:ffe;if(e==="prelu")return t?fE:pE;if(e==="leakyrelu")return t?hE:dE;if(e==="sigmoid")return t?bfe:mfe;throw new Error(`Activation ${e} has not been implemented for the WebGL backend.`)}var gE=class{constructor(e,t,n,r=!1,s=!1,a=!1,o=null,i=!1,l=!1){this.variableNames=["matrixA","matrixB"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=n;let u=r?e[1]:e[2],c=Math.ceil(u/2),d=r?"i * 2, rc.y":"rc.y, i * 2",h=s?"rc.z, i * 2":"i * 2, rc.z",p=r?["a.xxyy","a.zzww"]:["a.xxzz","a.yyww"],f=s?["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 A="rc.x",x="rc.x";e[0]<t[0]?A=`int(min(float(rc.x), ${e[0]-1}.))`:t[0]<e[0]&&(x=`int(min(float(rc.x), ${t[0]-1}.))`),this.userCode=`
${m}
const float sharedDimension = ${c}.0;
vec4 dot2x2ARowBCol(ivec3 rc) {
vec4 result = vec4(0);
for (int i = 0; i < ${c}; i++) {
int batchA = ${A};
int batchB = ${x};
vec4 a = getMatrixA(batchA, ${d});
vec4 b = getMatrixB(batchB, ${h});
// These swizzled products need to be separately added.
// See: https://github.com/tensorflow/tfjs/issues/1735
result += (${p[0]} * ${f[0]});
result += (${p[1]} * ${f[1]});
}
return result;
}
void main() {
ivec3 rc = getOutputCoords();
vec4 result = dot2x2ARowBCol(rc);
${y}
${g}
setOutput(result);
}
`}},yE={REAL:"return areal * breal - aimag * bimag;",IMAG:"return areal * bimag + aimag * breal;"},AE=class{constructor(e,t,n){this.variableNames=["AReal","AImag","BReal","BImag"],this.outputShape=_.assertAndGetBroadcastShape(t,n),this.userCode=`
float binaryOpComplex(
float areal, float aimag, float breal, float bimag) {
${e}
}
void main() {
float areal = getARealAtOutCoords();
float aimag = getAImagAtOutCoords();
float breal = getBRealAtOutCoords();
float bimag = getBImagAtOutCoords();
setOutput(binaryOpComplex(areal, aimag, breal, bimag));
}
`}},xE="return a * b;";function Y5(e){let{inputs:t,backend:n}=e,{a:r,b:s}=t,a=_.upcastType(r.dtype,s.dtype);if(r.dtype==="complex64"){let i=n.texData.get(r.dataId),l=n.texData.get(s.dataId),u=new AE(yE.REAL,r.shape,s.shape),c=new AE(yE.IMAG,r.shape,s.shape),d=[{dataId:i.complexTensorInfos.real.dataId,dtype:i.complexTensorInfos.real.dtype,shape:r.shape},{dataId:i.complexTensorInfos.imag.dataId,dtype:i.complexTensorInfos.imag.dtype,shape:r.shape},{dataId:l.complexTensorInfos.real.dataId,dtype:l.complexTensorInfos.real.dtype,shape:s.shape},{dataId:l.complexTensorInfos.imag.dataId,dtype:l.complexTensorInfos.imag.dtype,shape:s.shape}],h=n.runWebGLProgram(u,d,"float32"),p=n.runWebGLProgram(c,d,"float32"),f=eo({inputs:{real:h,imag:p},backend:n});return n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(p),f}if(n.shouldExecuteOnCPU([r,s])){let i=n.texData.get(r.dataId),l=n.texData.get(s.dataId),[u,c]=Ppe(r.shape,s.shape,i.values,l.values,a),d=n.makeTensorInfo(c,a),h=n.texData.get(d.dataId);return h.values=u,d}let o;return re().getBool("WEBGL_PACK_BINARY_OPERATIONS")?o=new ch(xE,r.shape,s.shape):o=new bu(xE,r.shape,s.shape),n.runWebGLProgram(o,[r,s],a)}var Bfe={kernelName:Mo,backendName:"webgl",kernelFunc:Y5};function Wfe(e,t,n){let r=[fi(e.shape),...mi(e.shape)],s={dtype:e.dtype,shape:r,dataId:e.dataId},a=[fi(t),...mi(t)],o=new rE(a,r),i=!0,l=n.runWebGLProgram(o,[s],e.dtype,null,i);return{dataId:l.dataId,shape:t,dtype:l.dtype}}function ve(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{shape:a}=r,o=n,i=k.sizeFromShape(s.shape),l=k.inferFromImplicitShape(a,i),u=k.sizeFromShape(l);k.assert(i===u,()=>`The new shape (${l}) has ${u} elements and the old shape (${s.shape}) has ${i} elements. The new shape and old shape must have the same number of elements.`);let c=o.texData.get(s.dataId);return c.isPacked&&!oh(s.shape,l)&&!(c.texture!==null&&oh(c.shape,l))?Wfe(s,l,o):(o.incRef(s.dataId),{dataId:s.dataId,shape:l,dtype:s.dtype})}var Vfe={kernelName:Jc,backendName:"webgl",kernelFunc:ve},bE=class{constructor(e,t){this.variableNames=["x"];let{windowSize:n,batchSize:r,inSize:s,outSize:a}=e;this.outputShape=[r,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 * ${k.isInt(c)?c.toPrecision(2):c}, ones);`}let u="";s%n>0&&(u=`
if (inIdx < 0 || inIdx >= ${s}) {
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);
${l}
}
setOutput(sumValue);
}
`}},Ufe=class{constructor(e,t){this.variableNames=["x"];let{windowSize:n,batchSize:r,inSize:s,outSize:a}=e;this.outputShape=[r,a];let o="0.0",i="";t==="prod"?o="1.0":t==="min"?(o="1.0 / 1e-20",i="min"):t==="max"&&(o="-1.0 / 1e-20",i="max");let l=`${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,d=`
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);
}
}
}
`,h="vec4";t==="all"?(o="1.0",d=`
bool reducedAllValue = all(values);
float floatedReducedAllValue = float(reducedAllValue);
allValue = float(allValue >= 1.0 && floatedReducedAllValue >= 1.0);
`,h="bvec4"):t==="any"&&(o="0.0",d=`
bool reducedAnyValue = any(values);
float floatedReducedAnyValue = float(reducedAnyValue);
anyValue = float(anyValue >= 1.0 || floatedReducedAnyValue >= 1.0);
`,h="bvec4");let p="";s%n>0&&(p=`
if (inIdx < 0 || inIdx >= ${s}) {
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) {
${p}
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;
${h} values = ${h}(
getValue(batch, inIdx),
getValue(batch, inIdx + 1),
getValue(batch, inIdx + 2),
getValue(batch, inIdx + 3)
);
${d}
}
int inIdx = inOffset + ${u};
if (${c===1}) {
${h} values = ${h}(
getValue(batch, inIdx),
initializationValue,
initializationValue,
initializationValue
);
${d}
} else if (${c===2}) {
${h} values = ${h}(
getValue(batch, inIdx),
getValue(batch, inIdx + 1),
initializationValue,
initializationValue
);
${d}
} else if (${c===3}) {
${h} values = ${h}(
getValue(batch, inIdx),
getValue(batch, inIdx + 1),
getValue(batch, inIdx + 2),
initializationValue
);
${d}
}
setOutput(${l});
}
`}};function Hfe(e){let t=[];for(;t.length===0||t[t.length-1].outSize!==1;){let n=t.length?t[t.length-1].outSize:e[1],r=_.computeOptimalWindowSize(n);t.push({inSize:n,windowSize:r,outSize:Math.ceil(n/r)})}return t}function Ai(e,t,n,r){let s=Hfe(e.shape),a=e;for(let o=0;o<s.length;o++){let{inSize:i,windowSize:l,outSize:u}=s[o],c,d;n==="mean"?c=o===0?new bE({windowSize:l,inSize:i,batchSize:e.shape[0],outSize:u},i):new bE({windowSize:l,inSize:i,batchSize:e.shape[0],outSize:u}):c=new Ufe({windowSize:l,inSize:i,batchSize:e.shape[0],outSize:u},n),d=a,a=r.runWebGLProgram(c,[a],t),d.dataId!==e.dataId&&r.disposeIntermediateTensorInfo(d)}return a}var Gfe=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 r=wt(this.rank),s=jfe(t);this.userCode=`
void main() {
${r} resRC = getOutputCoords();
setOutput(getA(${s}));
}
`}};function jfe(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"],r=new Array(t);for(let s=0;s<e.length;s++)r[e[s]]=n[s];return r.join()}var qfe=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 ${this.rank} is not yet supported.`);let r=wt(this.rank),s=nE("rc",this.rank),a=new Array(this.rank);for(let u=0;u<t.length;u++)a[t[u]]=s[u];let o=`vec2(${a.slice(-2).join()})`,i=`++${s[this.rank-1]} < ${n[this.rank-1]}`,l=`getChannel(getA(${a.join()}), ${o})`;this.userCode=`
void main() {
${r} rc = getOutputCoords();
vec4 result = vec4(0.);
result[0] = ${l};
if(${i}) {
result[1] = ${l};
}
--${s[this.rank-1]};
if(++${s[this.rank-2]} < ${n[this.rank-2]}) {
result[2] = ${l};
if(${i}) {
result[3] = ${l};
}
}
setOutput(result);
}
`}};function Wm(e,t,n){let r=re().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new qfe(e.shape,t):new Gfe(e.shape,t);return n.runWebGLProgram(r,[e],e.dtype)}function Kfe(e,t,n,r){let s=t,a=e.shape.length,o=k.parseAxisParam(s,e.shape),i=o,l=_.getAxesPermutation(i,a),u=l!=null,c=e;u&&(c=Wm(e,l,r),i=_.getInnerMostAxes(i.length,a)),_.assertAxesAreInnerMostDims("sum",i,a);let[d,h]=_.computeOutAndReduceShapes(c.shape,i),p=d;n&&(p=_.expandShapeToKeepDim(d,o));let f=k.sizeFromShape(h),g=k.sizeFromShape(e.shape)/f,y=ve({inputs:{x:c},attrs:{shape:[g,f]},backend:r}),A=cy(e.dtype),x=Ai(y,A,"sum",r),b=ve({inputs:{x},attrs:{shape:p},backend:r});return r.disposeIntermediateTensorInfo(y),r.disposeIntermediateTensorInfo(x),u&&r.disposeIntermediateTensorInfo(c),b}function Vm(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r;return Kfe(s,a,o,n)}var Xfe={kernelName:Fl,backendName:"webgl",kernelFunc:Vm};function Un(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{perm:a}=r,o=n,i=s.shape.length,l=new Array(i);for(let c=0;c<l.length;c++)l[c]=s.shape[a[c]];let u;if(o.shouldExecuteOnCPU([s])){let d=o.texData.get(s.dataId).values,h=Z5(d,s.shape,s.dtype,a,l);u=o.makeTensorInfo(l,s.dtype);let p=o.texData.get(u.dataId);p.values=h}else u=Wm(s,a,o);return u}var Zfe={kernelName:zl,backendName:"webgl",kernelFunc:Un},vE=1e3;function Um({a:e,b:t,transposeA:n,transposeB:r,backend:s,bias:a=null,preluActivationWeights:o=null,leakyreluAlpha:i=0,activation:l=null}){let u=e.shape.length,c=t.shape.length,d=n?e.shape[u-2]:e.shape[u-1],h=r?t.shape[c-1]:t.shape[c-2],p=n?e.shape[u-1]:e.shape[u-2],f=r?t.shape[c-2]:t.shape[c-1],m=e.shape.slice(0,-2),g=t.shape.slice(0,-2),y=k.sizeFromShape(m),A=k.sizeFromShape(g),x=y===A||y===1||A===1;k.assert(u>=2&&c>=2&&x,()=>`Error in matMul: the input batch dimensions must either be the same or at least one input batch dimension must be 1. Got input batch dimensions of (${m}) and (${g}).`);let v=(y>A?e.shape.slice(0,-2):t.shape.slice(0,-2)).concat([p,f]);k.assert(d===h,()=>`Error in matMul: inner shapes (${d}) and (${h}) of Tensors with shapes ${e.shape} and ${t.shape} and transposeA=${n} and transposeB=${r} must match.`);let I=n?[y,d,p]:[y,p,d],w=r?[A,f,h]:[A,h,f],S=ve({inputs:{x:e},backend:s,attrs:{shape:I}}),E=ve({inputs:{x:t},backend:s,attrs:{shape:w}}),D=[S,E],$=Math.max(y,A),R=n?S.shape[1]:S.shape[2],N=a!=null,M=o!=null,B=l==="leakyrelu",q=l!=null?Bm(l,!0):null,X=N||M||B||q!=null,J;if((p===1||f===1)&&R>vE&&X===!1){let ae=S,se=E;n&&(ae=Un({inputs:{x:S},backend:s,attrs:{perm:[0,2,1]}}),D.push(ae)),r&&(se=Un({inputs:{x:E},backend:s,attrs:{perm:[0,2,1]}}),D.push(se));let oe=f!==1,ne=f===1,ce=ae;oe&&(ce=ve({inputs:{x:ae},backend:s,attrs:{shape:[$,R,1]}}),D.push(ce));let he=f===1?2:1,me=se;ne&&(me=ve({inputs:{x:se},backend:s,attrs:{shape:[$,1,R]}}),D.push(me));let be=Y5({inputs:{a:ce,b:me},backend:s});J=Vm({inputs:{x:be},backend:s,attrs:{axis:he,keepDims:!0}}),D.push(be)}else{let ae=Ur(e.dtype,t.dtype),se=new gE(I,w,[$,p,f],n,r,N,q,M,B),oe=[S,E];if(a!=null&&oe.push(a),M&&oe.push(o),B){let ne=s.makeTensorInfo([],"float32",k.createScalarValue(i,"float32"));oe.push(ne),D.push(ne)}J=s.runWebGLProgram(se,oe,ae)}let ee=ve({inputs:{x:J},backend:s,attrs:{shape:v}});D.push(J);for(let ae of D)s.disposeIntermediateTensorInfo(ae);return ee}function Yfe(e){let{inputs:t,backend:n,attrs:r}=e,{a:s,b:a,bias:o,preluActivationWeights:i}=t,{transposeA:l,transposeB:u,activation:c,leakyreluAlpha:d}=r;return Um({a:s,b:a,transposeA:l,transposeB:u,backend:n,bias:o,preluActivationWeights:i,leakyreluAlpha:d,activation:c})}var Jfe={kernelName:Ll,backendName:"webgl",kernelFunc:Yfe},wE="return abs(x);";function Qfe(e){let{inputs:t,backend:n}=e,{x:r}=t;if(n.shouldExecuteOnCPU([r])&&r.dtype!=="complex64"){let a=n.texData.get(r.dataId),o=eE(a.values);return n.makeTensorInfo(r.shape,r.dtype,o)}let s;return re().getBool("WEBGL_PACK_UNARY_OPERATIONS")?s=new xu(r.shape,wE):s=new Qa(r.shape,wE),n.runWebGLProgram(s,[r],r.dtype)}var eme={kernelName:Ac,backendName:"webgl",kernelFunc:Qfe},tme=gs+`
if (abs(x) > 1.) {
return NAN;
}
return acos(x);
`,nme=it({opSnippet:tme}),rme={kernelName:xc,backendName:"webgl",kernelFunc:nme},sme=gs+`
if (x < 1.0) return NAN;
return log(x + sqrt(x * x - 1.0));`,ame=it({opSnippet:sme}),ome={kernelName:bc,backendName:"webgl",kernelFunc:ame},kE="return a + b;",ime=Tn({opSnippet:kE,packedOpSnippet:kE,supportsComplex:!0,cpuKernelImpl:Ape}),lme={kernelName:Ma,backendName:"webgl",kernelFunc:ime},ume=class{constructor(e,t){this.outputShape=[],this.outputShape=e,this.variableNames=t.map((s,a)=>`T${a}`);let n=[];this.variableNames.forEach(s=>{n.push(`float v${s} = get${s}AtOutCoords();`)});let r=this.variableNames.map(s=>`v${s}`).join(" + ");this.userCode=`
void main() {
${n.join(`
`)}
float result = ${r};
setOutput(result);
}
`}},cme=class{constructor(e,t){this.outputShape=[],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e,this.variableNames=t.map((s,a)=>`T${a}`);let n=[];this.variableNames.forEach(s=>{n.push(`vec4 v${s} = get${s}AtOutCoords();`)});let r=this.variableNames.map(s=>`v${s}`).join(" + ");this.userCode=`
void main() {
${n.join(`
`)}
vec4 result = ${r};
setOutput(result);
}
`}};function Hm(e){let{inputs:t,backend:n}=e,r=t;if(r.length===1)return fr({inputs:{x:r[0]},backend:n});if(r.length>re().get("WEBGL_MAX_TEXTURES_IN_SHADER")){let l=Math.floor(r.length/2),u=Hm({inputs:r.slice(0,l),backend:n}),c=Hm({inputs:r.slice(l),backend:n});return Hm({inputs:[u,c],backend:n})}let s=r.map(l=>l.dtype).reduce((l,u)=>Ur(l,u)),a=r.map(l=>l.shape),i=re().getBool("WEBGL_PACK")?new cme(r[0].shape,a):new ume(r[0].shape,a);return n.runWebGLProgram(i,r,s)}var dme={kernelName:Xi,backendName:"webgl",kernelFunc:Hm};function hme(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r,i=s.shape.length,l=k.parseAxisParam(a,s.shape),u=l,c=_.getAxesPermutation(u,i),d=s;c!=null&&(d=Un({inputs:{x:s},backend:n,attrs:{perm:c}}),u=_.getInnerMostAxes(u.length,i)),_.assertAxesAreInnerMostDims("all",u,i);let[h,p]=_.computeOutAndReduceShapes(d.shape,u),f=k.sizeFromShape(p),m=ve({inputs:{x:d},backend:n,attrs:{shape:[-1,f]}}),g=Ai(m,m.dtype,"all",n),y;if(o){let A=_.expandShapeToKeepDim(h,l);y=ve({inputs:{x:g},backend:n,attrs:{shape:A}})}else y=ve({inputs:{x:g},backend:n,attrs:{shape:h}});return n.disposeIntermediateTensorInfo(m),n.disposeIntermediateTensorInfo(g),c!=null&&n.disposeIntermediateTensorInfo(d),y}var pme={kernelName:vc,backendName:"webgl",kernelFunc:hme};function fme(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r,i=s.shape.length,l=k.parseAxisParam(a,s.shape),u=l,c=_.getAxesPermutation(u,i),d=s;c!=null&&(d=Un({inputs:{x:s},backend:n,attrs:{perm:c}}),u=_.getInnerMostAxes(u.length,i)),_.assertAxesAreInnerMostDims("any",u,i);let[h,p]=_.computeOutAndReduceShapes(d.shape,u),f=k.sizeFromShape(p),m=ve({inputs:{x:d},backend:n,attrs:{shape:[-1,f]}}),g=Ai(m,m.dtype,"any",n),y;if(o){let A=_.expandShapeToKeepDim(h,l);y=ve({inputs:{x:g},backend:n,attrs:{shape:A}})}else y=ve({inputs:{x:g},backend:n,attrs:{shape:h}});return n.disposeIntermediateTensorInfo(m),n.disposeIntermediateTensorInfo(g),c!=null&&n.disposeIntermediateTensorInfo(d),y}var mme={kernelName:wc,backendName:"webgl",kernelFunc:fme},gme=class{constructor(e,t,n){this.variableNames=["A"];let{windowSize:r,batchSize:s,outSize:a}=e;n||this.variableNames.push("bestIndicesA"),this.outputShape=[s,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 * ${r};
int bestIndex = inOffset;
float bestValue = getA(batch, bestIndex);
for (int i = 0; i < ${r}; i++) {
int inIdx = ${i};
float candidate = getA(batch, inIdx);
if (candidate ${o} bestValue) {
bestValue = candidate;
bestIndex = inIdx;
}
}
setOutput(float(bestIndex));
}
`}},yme=class{constructor(e,t,n,r){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,k.assert(e.length>2,()=>`Packed arg${n.charAt(0).toUpperCase()+n.slice(1)} supports only inputs with rank above 2.`);let s=e[e.length-1],a=Math.ceil(s/t);this.outputShape=e.slice(0,-1),a>1&&this.outputShape.push(a),r||this.variableNames.push("bestIndicesA");let o=this.outputShape,i=o.length,l=wt(i),u=Vn("coords",i),c,d;if(a===1){d=i+1;let w=wt(d);c=`
${w} sourceLocR = ${w}(${u.join()}, 0);
++${u[i-1]};
${w} sourceLocG = ${w}(${u.join()}, 0);
++${u[i-2]};
${w} sourceLocA = ${w}(${u.join()}, 0);
--${u[i-1]};
${w} sourceLocB = ${w}(${u.join()}, 0);
--${u[i-2]};`}else d=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 h=["x","y","z","w","u","v"].slice(0,d),p="."+h[d-1],f=h.map(w=>"int "+w),m=Vn("sourceLocR",d-1).concat("inIdx.r"),g=Vn("sourceLocG",d-1).concat("inIdx.g"),y=Vn("sourceLocB",d-1).concat("inIdx.b"),A=Vn("sourceLocA",d-1).concat("inIdx.a"),x=n==="max"?"greaterThan":"lessThan",b=r?"":`
inIdx = round(vec4(getBestIndicesAChannel(${m.join()}),
getBestIndicesAChannel(${g.join()}),
getBestIndicesAChannel(${y.join()}),
getBestIndicesAChannel(${A.join()})));`,v=`vec4(
getAChannel(${m.join()}),
hasNextCol ? getAChannel(${g.join()}) : 0.,
hasNextRow ? getAChannel(${y.join()}) : 0.,
hasNextRow && hasNextCol ? getAChannel(${A.join()}) : 0.)`,I=r?"":`
float getBestIndicesAChannel(${f.join()}) {
return getChannel(getBestIndicesA(${h.join()}),
vec2(${h.slice(-2).join()}));
}`;this.userCode=`
float getAChannel(${f.join()}) {
return getChannel(getA(${h.join()}),
vec2(${h.slice(-2).join()}));
}
${I}
void main() {
${l} coords = getOutputCoords();
bool hasNextCol = ${u[i-1]} < ${o[i-1]-1};
bool hasNextRow = ${u[i-2]} < ${o[i-2]-1};
${c}
ivec4 srcIdx = ivec4(sourceLocR${p}, sourceLocG${p},
sourceLocB${p}, sourceLocA${p}) * ${t};
ivec4 inIdx = srcIdx;
vec4 bestIndex = vec4(inIdx);
vec4 bestValue = ${v};
for (int i = 0; i < ${t}; i++) {
inIdx = srcIdx;
${b}
vec4 candidate = ${v};
bvec4 nan = isnan(candidate);
bvec4 replace = bvec4(
vec4(${x}(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 IE(e,t,n,r=null){let s=t.shape[0],a=t.shape[1];r!=null&&(s=r.shape[0],a=r.shape[1]);let o=_.computeOptimalWindowSize(a),i={windowSize:o,inSize:a,batchSize:s,outSize:Math.ceil(a/o)},l=new gme(i,n,r==null),u=[t];r!=null&&u.push(r);let c=e.runWebGLProgram(l,u,"int32");if(c.shape[1]===1)return c;let d=IE(e,t,n,c);return e.disposeIntermediateTensorInfo(c),d}function SE(e,t,n,r=null){let s=r!=null?r.shape:t.shape,a=s[s.length-1],o=_.computeOptimalWindowSize(a),i=new yme(s,o,n,r==null),l=r==null?[t]:[t,r],u=e.runWebGLProgram(i,l,"int32");if(u.shape.length===t.shape.length){let c=SE(e,t,n,u);return e.disposeIntermediateTensorInfo(u),c}return u}function TE(e,t,n,r){let s=[n];if(_.assertAxesAreInnerMostDims("arg"+r.charAt(0).toUpperCase()+r.slice(1),s,t.shape.length),!re().getBool("WEBGL_PACK_REDUCE")||t.shape.length<=2){let a=[],[o,i]=_.computeOutAndReduceShapes(t.shape,s),l=k.sizeFromShape(i),u=ve({inputs:{x:t},backend:e,attrs:{shape:[-1,l]}});a.push(u);let c=IE(e,u,r);a.push(c);let d=ve({inputs:{x:c},backend:e,attrs:{shape:o}});return a.forEach(h=>e.disposeIntermediateTensorInfo(h)),d}return SE(e,t,r)}function Ame(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a}=r,o=k.parseAxisParam(a,s.shape),i=_.getAxesPermutation(o,s.shape.length),l=s,u=[];i!=null&&(l=Un({inputs:{x:s},backend:n,attrs:{perm:i}}),u.push(l),o=_.getInnerMostAxes(o.length,l.shape.length)),_.assertAxesAreInnerMostDims("argMax",[o[0]],l.shape.length);let c=TE(n,l,o[0],"max");return u.forEach(d=>n.disposeIntermediateTensorInfo(d)),c}var xme={kernelName:Zi,backendName:"webgl",kernelFunc:Ame};function bme(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a}=r,o=k.parseAxisParam(a,s.shape),i=_.getAxesPermutation(o,s.shape.length),l=s,u=[];i!=null&&(l=Un({inputs:{x:s},backend:n,attrs:{perm:i}}),u.push(l),o=_.getInnerMostAxes(o.length,l.shape.length)),_.assertAxesAreInnerMostDims("argMin",[o[0]],l.shape.length);let c=TE(n,l,o[0],"min");return u.forEach(d=>n.disposeIntermediateTensorInfo(d)),c}var vme={kernelName:qp,backendName:"webgl",kernelFunc:bme},wme=gs+`
if (abs(x) > 1.) {
return NAN;
}
return asin(x);
`,kme=it({opSnippet:wme}),Ime={kernelName:kc,backendName:"webgl",kernelFunc:kme},Sme=gs+"return log(x + sqrt(x * x + 1.0));",Tme=it({opSnippet:Sme}),Nme={kernelName:Ic,backendName:"webgl",kernelFunc:Tme},Cme=gs+`
return atan(x);
`,Eme=it({opSnippet:Cme}),$me={kernelName:Sc,backendName:"webgl",kernelFunc:Eme},Rme=zfe+`
return atan(a, b);
`,_me=`
vec4 result = atan(a, b);
vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
`+Lfe+`
return result;
`,Dme=Tn({opSnippet:Rme,packedOpSnippet:_me}),Fme={kernelName:Nc,backendName:"webgl",kernelFunc:Dme},Mme=gs+`
if ((x < -1.0) || (x > 1.0)) return NAN;
return (log(1.0 + x) - log(1.0 - x)) / 2.0;`,Ome=it({opSnippet:Mme}),Pme={kernelName:Tc,backendName:"webgl",kernelFunc:Ome},dh=class{constructor(e,t,n,r=!1,s=!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,d=e.effectiveFilterWidth,h=e.padInfo.top,p=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) * ${e.inChannels} + d`,y="0.0";if(f||(y="-1.0 / 1e-20"),n){let w=">=";this.userCode=`
const ivec2 strides = ivec2(${o}, ${i});
const ivec2 pads = ivec2(${h}, ${p});
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 < ${d};
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 ${w} currMinMaxValue) {
minMaxValue = value;
minMaxValueFound = 1.0;
minMaxPosition = ${r?s?m:g:`wR * ${d} + wC`};
}
}
}
setOutput(float(minMaxPosition));
}
`;return}let A="max",x=`${t}(${t}(${t}(minMaxValue[0], minMaxValue[1]), minMaxValue[2]), minMaxValue[3])`;t==="avg"&&(x="avgValue / count");let b=Math.floor(a/4)*4,v=a%4,I=`
if (${f}) {
avgValue += dot(values, ones);
} else {
minMaxValue = ${A}(values, minMaxValue);
}
`;this.userCode=`
const ivec2 strides = ivec2(${o}, ${i});
const ivec2 pads = ivec2(${h}, ${p});
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)
);
${I}
}
int xC = xCCorner + ${b};
if (${v===1}) {
vec4 values = vec4(
getValue(batch, xR, xC, d),
initializationValue,
initializationValue,
initializationValue
);
${I}
} else if (${v===2}) {
vec4 values = vec4(
getValue(batch, xR, xC, d),
getValue(batch, xR, xC + ${u}, d),
initializationValue,
initializationValue
);
${I}
} else if (${v===3}) {
vec4 values = vec4(
getValue(batch, xR, xC, d),
getValue(batch, xR, xC + ${u}, d),
getValue(batch, xR, xC + 2 * ${u}, d),
initializationValue
);
${I}
}
}
setOutput(${x});
}
`}},J5=class{constructor(e,t,n,r=!1,s=!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,d=e.dilationWidth,h=e.effectiveFilterDepth,p=e.effectiveFilterHeight,f=e.effectiveFilterWidth,m=e.padInfo.front,g=e.padInfo.top,y=e.padInfo.left;this.outputShape=e.outShape;let A=t==="avg",x="0.0";if(A||(x="-1.0 / 1e-20"),n){let E=">=";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 < ${h};
wD += ${u}) {
int xD = xDCorner + wD;
if (xD < 0 || xD >= ${e.inDepth}) {
continue;
}
for (int wR = 0; wR < ${p};
wR += ${c}) {
int xR = xRCorner + wR;
if (xR < 0 || xR >= ${e.inHeight}) {
continue;
}
for (int wC = 0; wC < ${f};
wC += ${d}) {
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 ${E} currMinMaxValue) {
minMaxValue = value;
minMaxValueFound = 1.0;
minMaxPosition = ${r?s?`(((batch * ${e.inDepth} + xD) * ${e.inHeight} + xR) * ${e.inWidth} + xC) * ${e.inChannels} + ch`:`((xD * ${e.inHeight} + xR) * ${e.inWidth} + xC) * ${e.inChannels} + ch`:`wD * ${p} * ${f} +
wR * ${f} + wC`};
}
}
}
}
setOutput(float(minMaxPosition));
}
`;return}let b="max",v=`${t}(${t}(${t}(minMaxValue[0], minMaxValue[1]), minMaxValue[2]), minMaxValue[3])`;t==="avg"&&(v="avgValue / count");let I=Math.floor(a/4)*4,w=a%4,S=`
if (${A}) {
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 = ${x};
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 >= ${e.inWidth}) {
return initializationValue;
}
count += 1.0;
return getX(batch, xD, xR, xC, ch);
}
void main() {
ivec5 coords = getOutputCoords();
int batch = coords.x;
int ch = coords.u;
ivec3 xCorner = ivec3(coords.y, coords.z, coords.w) * strides - pads;
int xDCorner = xCorner.x;
int xRCorner = xCorner.y;
int xCCorner = xCorner.z;
// max/min x(?, ?, ?, d) to get y(yD, yR, yC, ch).
// ? = to be determined
vec4 minMaxValue = vec4(${x});
float avgValue = 0.0;
count = 0.0;
for (int wD = 0; wD < ${h};
wD += ${u}) {
int xD = xDCorner + wD;
if (xD < 0 || xD >= ${e.inDepth}) {
continue;
}
for (int wR = 0; wR < ${p};
wR += ${c}) {
int xR = xRCorner + wR;
if (xR < 0 || xR >= ${e.inHeight}) {
continue;
}
for (int wC = 0; wC < ${I}; wC += 4) {
int xC = xCCorner + wC * ${d};
vec4 values = vec4(
getValue(batch, xD, xR, xC, ch),
getValue(batch, xD, xR, xC + ${d}, ch),
getValue(batch, xD, xR, xC + 2 * ${d}, ch),
getValue(batch, xD, xR, xC + 3 * ${d}, ch)
);
${S}
}
int xC = xCCorner + ${I};
if (${w===1}) {
vec4 values = vec4(
getValue(batch, xD, xR, xC, ch),
initializationValue,
initializationValue,
initializationValue
);
${S}
} else if (${w===2}) {
vec4 values = vec4(
getValue(batch, xD, xR, xC, ch),
getValue(batch, xD, xR, xC + ${d}, ch),
initializationValue,
initializationValue
);
${S}
} else if (${w===3}) {
vec4 values = vec4(
getValue(batch, xD, xR, xC, ch),
getValue(batch, xD, xR, xC + ${d}, ch),
getValue(batch, xD, xR, xC + 2 * ${d}, ch),
initializationValue
);
${S}
}
}
setOutput(${v});
}
}
`}};function zme(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t;hu(s,"avgPool");let{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=r,u=1;k.assert(_.eitherStridesOrDilationsAreOne(o,u),()=>`Error in avgPool: Either strides or dilations must be 1. Got strides ${o} and dilations '${u}'`);let c=_.computePool2DInfo(s.shape,a,o,u,i,l);if(c.filterWidth===1&&c.filterHeight===1&&k.arraysEqual(c.inShape,c.outShape))return fr({inputs:{x:s},backend:n});let d=new dh(c,"avg",!1);return n.runWebGLProgram(d,[s],"float32")}var Lme={kernelName:Yi,backendName:"webgl",kernelFunc:zme};function Bme(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{filterSize:a,strides:o,pad:i,dimRoundingMode:l,dataFormat:u}=r,c=[1,1,1],d=_.computePool3DInfo(s.shape,a,o,c,i,l,u),h=new J5(d,"avg",!1);return n.runWebGLProgram(h,[s],"float32")}var Wme={kernelName:Kp,backendName:"webgl",kernelFunc:Bme},Vme=class{constructor(e){this.variableNames=["dy"],this.outputShape=e.inShape;let t=e.filterHeight,n=e.filterWidth,r=e.strideHeight,s=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,d=1/(t*n);this.userCode=`
const ivec2 pads = ivec2(${u}, ${c});
const float avgMultiplier = float(${d});
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) / ${r}.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) / ${s}.0;
if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
fract(dyC) > 0.0) {
continue;
}
int idyC = int(dyC);
float dyValue = getDy(b, idyR, idyC, d);
dotProd += dyValue * avgMultiplier;
}
}
setOutput(dotProd);
}
`}},Ume=class{constructor(e){this.variableNames=["dy"],this.outputShape=e.inShape;let t=e.filterDepth,n=e.filterHeight,r=e.filterWidth,s=e.strideDepth,a=e.strideHeight,o=e.strideWidth,i=e.dilationDepth,l=e.dilationHeight,u=e.dilationWidth,c=e.effectiveFilterDepth,d=e.effectiveFilterHeight,h=e.effectiveFilterWidth,p=c-1-e.padInfo.front,f=d-1-e.padInfo.top,m=h-1-e.padInfo.left,g=1/(t*n*r);this.userCode=`
const ivec3 pads = ivec3(${p}, ${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) / ${s}.0;
if (dyD < 0.0 || dyD >= ${e.outDepth}.0 || fract(dyD) > 0.0) {
continue;
}
int idyD = int(dyD);
for (int wR = 0; wR < ${d};
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 < ${h};
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 Hme(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,input:a}=t,o=a,{filterSize:i,strides:l,pad:u,dimRoundingMode:c}=r,d=[1,1,1],h=_.computePool3DInfo(o.shape,i,l,d,u,c),p=new Ume(h);return n.runWebGLProgram(p,[s],o.dtype)}var Gme={kernelName:x1,backendName:"webgl",kernelFunc:Hme};function jme(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,input:a}=t,o=a;hu([s,a],"avgPoolGrad");let{filterSize:i,strides:l,pad:u}=r,c=_.computePool2DInfo(o.shape,i,l,1,u),d=new Vme(c);return n.runWebGLProgram(d,[s],o.dtype)}var qme={kernelName:A1,backendName:"webgl",kernelFunc:jme};function Kme(e){let{inputs:t,backend:n,attrs:r}=e,{a:s,b:a}=t,{transposeA:o,transposeB:i}=r;return Um({a:s,b:a,transposeA:o,transposeB:i,backend:n})}var Xme={kernelName:Ji,backendName:"webgl",kernelFunc:Kme},Zme=class{constructor(e,t,n,r,s,a){this.outputShape=[],this.variableNames=["x","mean","variance"],_.assertAndGetBroadcastShape(e,t),_.assertAndGetBroadcastShape(e,n);let o="0.0";r!=null&&(_.assertAndGetBroadcastShape(e,r),this.variableNames.push("offset"),o="getOffsetAtOutCoords()");let i="1.0";s!=null&&(_.assertAndGetBroadcastShape(e,s),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)));
}
`}},Yme=class{constructor(e,t,n,r,s,a){this.packedInputs=!0,this.packedOutput=!0,this.variableNames=["x","mean","variance"],_.assertAndGetBroadcastShape(e,t),_.assertAndGetBroadcastShape(e,n);let o="vec4(0.0)";r!=null&&(_.assertAndGetBroadcastShape(e,r),this.variableNames.push("offset"),o="getOffsetAtOutCoords()");let i="vec4(1.0)";s!=null&&(_.assertAndGetBroadcastShape(e,s),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);
}
`}},Jme=({inputs:e,backend:t,attrs:n})=>{let{x:r,mean:s,variance:a,offset:o,scale:i}=e;k.assert(s.shape.length===a.shape.length,()=>"Batch normalization gradient requires mean and variance to have equal ranks."),k.assert(o==null||s.shape.length===o.shape.length,()=>"Batch normalization gradient requires mean and offset to have equal ranks."),k.assert(i==null||s.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=[r,s,a],c=null;o!=null&&(c=o.shape,u.push(o));let d=null;i!=null&&(d=i.shape,u.push(i));let h=re().getBool("WEBGL_PACK_NORMALIZATION")?new Yme(r.shape,s.shape,a.shape,c,d,l):new Zme(r.shape,s.shape,a.shape,c,d,l);return t.runWebGLProgram(h,u,u[0].dtype)},Qme={kernelName:cl,backendName:"webgl",kernelFunc:Jme},e0e=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=t0e(this.rank),r,s=e.map((a,o)=>`sourceLoc.${Q5[o]} = start[${o}] + coords.${Q5[o]};`);r=`
${t} sourceLoc;
${t} coords = getOutputCoords();
${s.join(`
`)}
`,this.userCode=`
void main() {
${r}
setOutput(getSource(${n}));
}
`}},Q5=["x","y","z","w","u","v"];function t0e(e){if(e===1)return"sourceLoc";if(e<=6)return Q5.slice(0,e).map(t=>"sourceLoc."+t).join(",");throw Error(`Slicing for rank ${e} is not yet supported`)}var n0e=class{constructor(e){this.variableNames=["source"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e,this.rank=e.length,this.customUniforms=[{name:"start",arrayIndex:this.rank,type:"int"}];let t=wt(this.rank),n=Vn("coords",this.rank),r=Vn("sourceLoc",this.rank),s=this.rank===1?"sourceLoc":`vec2(${r.slice(-2).join()})`,a=`getChannel(getSource(${r.join()}), ${s})`,o=`
result.x = ${a};
if (++${n[this.rank-1]} < ${e[this.rank-1]}) {
++${r[this.rank-1]};
result.y = ${a};
--${r[this.rank-1]};
}
`,i=this.rank===1?"":`
--${n[this.rank-1]};
if (++${n[this.rank-2]} < ${e[this.rank-2]}) {
++${r[this.rank-2]};
result.z = ${a};
if (++${n[this.rank-1]} < ${e[this.rank-1]}) {
++${r[this.rank-1]};
result.w = ${a};
}
}
`,l=this.rank<=4?`sourceLoc = coords +
${t}(${e.map((u,c)=>`start[${c}]`).join()});`:e.map((u,c)=>`${r[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 r0e(e,t,n,r){let s=r.texData.get(e.dataId),a=r.makeTensorInfo(n,e.dtype),o=r.texData.get(a.dataId);Object.assign(o,s),o.refCount=1,o.shape=n,o.dtype=e.dtype;let i=Cn.computeFlatOffset(t,k.computeStrides(e.shape));s.slice&&(i+=s.slice.flatOffset),o.slice={flatOffset:i,origDataId:s.slice&&s.slice.origDataId||e.dataId};let l=r.dataRefCount.get(o.slice.origDataId)||1;return r.dataRefCount.set(o.slice.origDataId,l+1),a}function vu(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{begin:a,size:o}=r,[i,l]=Cn.parseSliceParams(s,a,o);if(Cn.assertParamsValid(s,i,l),k.sizeFromShape(l)===0)return n.makeTensorInfo(l,s.dtype,[]);if(n.shouldExecuteOnCPU([s])||s.dtype==="string"){let d=n.texData.get(s.dataId),h=Upe(d.values,i,l,s.shape,s.dtype);return n.makeTensorInfo(l,s.dtype,h)}let{isPacked:u}=n.texData.get(s.dataId),c=Cn.isSliceContinous(s.shape,i,l);if(u||!c){let d=re().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new n0e(l):new e0e(l),h=[i];return n.runWebGLProgram(d,[s],s.dtype,h)}return n.uploadToGPU(s.dataId),r0e(s,i,l,n)}var s0e={kernelName:nd,backendName:"webgl",kernelFunc:vu},a0e=e=>{let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{blockShape:a,crops:o}=r;k.assert(s.shape.length<=4,()=>"batchToSpaceND for rank > 4 with a WebGL backend not implemented yet");let i=a.reduce((A,x)=>A*x),l=_.getReshaped(s.shape,a,i),u=_.getPermuted(l.length,a.length),c=_.getReshapedPermuted(s.shape,a,i),d=_.getSliceBeginCoords(o,a.length),h=_.getSliceSize(c,o,a.length),p=[],f=ve({inputs:{x:s},backend:n,attrs:{shape:l}}),m=Un({inputs:{x:f},backend:n,attrs:{perm:u}}),g=ve({inputs:{x:m},backend:n,attrs:{shape:c}}),y=vu({inputs:{x:g},backend:n,attrs:{begin:d,size:h}});return p.push(f),p.push(m),p.push(g),p.forEach(A=>n.disposeIntermediateTensorInfo(A)),y},o0e={kernelName:Cc,backendName:"webgl",kernelFunc:a0e};function i0e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,weights:a}=t,{size:o}=r,i=n.readSync(s.dataId),l=n.readSync(a.dataId),u=QC(i,l,a.dtype,a.shape,o);return n.makeTensorInfo([o],a.dtype,u)}var l0e={kernelName:b1,backendName:"webgl",kernelFunc:i0e},u0e="return float(a != b);",NE=Tn({opSnippet:u0e,cpuKernelImpl:Lpe,dtype:"bool"}),c0e={kernelName:vl,backendName:"webgl",kernelFunc:NE};function hh(e){let{inputs:t,backend:n}=e,{input:r}=t,s=n.texData.get(r.dataId);return fr({inputs:{x:s.complexTensorInfos.real},backend:n})}var d0e={kernelName:V1,backendName:"webgl",kernelFunc:hh},h0e="return float(int(x));";function p0e(e,t){let n=new Qa(e.shape,h0e),r=t.runWebGLProgram(n,[e],"int32");return{dataId:r.dataId,shape:r.shape,dtype:r.dtype}}function eb(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{dtype:a}=r;if(a==="complex64"){if(s.dtype==="complex64")return fr({inputs:{x:s},backend:n});let o=ln(s.shape),i=eb({inputs:{x:s},backend:n,attrs:{dtype:"float32"}}),l=eo({inputs:{real:i,imag:o},backend:n});return o.dispose(),n.disposeIntermediateTensorInfo(i),l}if(s.dtype==="complex64"){let o=hh({inputs:{input:s},backend:n}),i=eb({inputs:{x:o},backend:n,attrs:{dtype:a}});return n.disposeIntermediateTensorInfo(o),i}if(!k.hasEncodingLoss(s.dtype,a)){let o=fr({inputs:{x:s},backend:n});return{dataId:o.dataId,shape:o.shape,dtype:a}}if(a==="int32")return p0e(s,n);if(a==="bool"){let o=n.makeTensorInfo([],"bool",k.getTypedArrayFromDType("bool",1)),l=NE({inputs:{a:s,b:o},backend:n});return n.disposeIntermediateTensorInfo(o),l}throw new Error(`Error in Cast: failed to cast ${s.dtype} to ${a}`)}var f0e={kernelName:Qi,backendName:"webgl",kernelFunc:eb},CE="return ceil(x);",m0e=it({opSnippet:CE,packedOpSnippet:CE,cpuKernelImpl:bpe}),g0e={kernelName:No,backendName:"webgl",kernelFunc:m0e},y0e=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));
}
`}},A0e=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 x0e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{clipValueMin:a,clipValueMax:o}=r,i;re().getBool("WEBGL_PACK_CLIP")?i=new A0e(s.shape):i=new y0e(s.shape);let l=[[a],[o]];return n.runWebGLProgram(i,[s],s.dtype,l)}var b0e={kernelName:Co,backendName:"webgl",kernelFunc:x0e},v0e=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 EE(e,t){return{dataId:t.dataId,dtype:t.dtype,shape:e.shape}}function w0e(e){let{inputs:t,backend:n}=e,{x:r}=t,s=n.texData.get(r.dataId),a=new v0e(r.shape),o=[EE(r,s.complexTensorInfos.real),EE(r,s.complexTensorInfos.imag)];return n.runWebGLProgram(a,o,o[0].dtype)}var k0e={kernelName:Xp,backendName:"webgl",kernelFunc:w0e},I0e=class{constructor(e){this.outputShape=[],this.outputShape=_.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-${o}));`)}let r=t.length,s=t[t.length-1];n.push(`else setOutput(getT${r}(yR, yC-${s}));`),this.userCode=`
void main() {
ivec2 coords = getOutputCoords();
int yR = coords.x;
int yC = coords.y;
${n.join(`
`)}
}
`}},S0e=class{constructor(e,t){this.packedInputs=!0,this.packedOutput=!0,this.outputShape=[],this.outputShape=_.computeOutShape(e,t);let n=this.outputShape,r=n.length,s=wt(r),a=Vn("coords",r),o=["x","y","z","w","u","v"].slice(0,r);this.variableNames=e.map((f,m)=>`T${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(),d=`if (${l} < ${i[0]}) {
return getChannel(
getT0(${c}), vec2(${u.join()}));
}`;for(let f=1;f<i.length;f++){let m=i[f-1];d+=`
if (${l} < ${i[f]} && ${l} >= ${i[f-1]}) {
return getChannel(
getT${f}(${Gm(o,l,m)}),
vec2(${Gm(u,l,m)}));
}`}let h=i.length,p=i[i.length-1];d+=`
return getChannel(
getT${h}(${Gm(o,l,p)}),
vec2(${Gm(u,l,p)}));`,this.userCode=`
float getValue(${o.map(f=>"int "+f)}) {
${d}
}
void main() {
${s} coords = getOutputCoords();
vec4 result = vec4(getValue(${a}), 0., 0., 0.);
${a[r-1]} = ${a[r-1]} + 1;
if (${a[r-1]} < ${n[r-1]}) {
result.g = getValue(${a});
}
${a[r-2]} = ${a[r-2]} + 1;
if (${a[r-2]} < ${n[r-2]}) {
result.a = getValue(${a});
}
${a[r-1]} = ${a[r-1]} - 1;
if (${a[r-2]} < ${n[r-2]} &&
${a[r-1]} < ${n[r-1]}) {
result.b = getValue(${a});
}
setOutput(result);
}
`}};function Gm(e,t,n){let r=e.indexOf(t);return e.map((a,o)=>o===r?`${a} - ${n}`:a).join()}function jm(e){let{inputs:t,backend:n}=e,{input:r}=t,s=n.texData.get(r.dataId);return fr({inputs:{x:s.complexTensorInfos.imag},backend:n})}var T0e={kernelName:M1,backendName:"webgl",kernelFunc:jm};function wu(e,t,n){let r=e[0].dtype;if(r==="complex64"){let c=e.map(m=>hh({inputs:{input:m},backend:n})),d=e.map(m=>jm({inputs:{input:m},backend:n})),h=wu(c,t,n),p=wu(d,t,n),f=eo({inputs:{real:h,imag:p},backend:n});return c.forEach(m=>n.disposeIntermediateTensorInfo(m)),d.forEach(m=>n.disposeIntermediateTensorInfo(m)),n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(p),f}let s=n.shouldExecuteOnCPU(e);if(r==="string"&&(s=!0),s){let c=e.map(y=>{let A=k.sizeFromShape(y.shape.slice(t));return ve({inputs:{x:y},backend:n,attrs:{shape:[-1,A]}})}),d=c.map(y=>({vals:n.readSync(y.dataId),shape:y.shape})),h=_.computeOutShape(c.map(y=>y.shape),1),p=c[0].shape[0]===1,f=vpe(d,h,r,p),m=_.computeOutShape(e.map(y=>y.shape),t),g=n.makeTensorInfo(m,r,f);return c.forEach(y=>n.disposeIntermediateTensorInfo(y)),g}if(e.length>re().getNumber("WEBGL_MAX_TEXTURES_IN_SHADER")){let c=Math.floor(e.length/2),d=wu(e.slice(0,c),t,n),h=wu(e.slice(c),t,n),p=wu([d,h],t,n);return n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),p}if(re().getBool("WEBGL_PACK_ARRAY_OPERATIONS")&&e[0].shape.length>1){let c=new S0e(e.map(d=>d.shape),t);return n.runWebGLProgram(c,e,r)}let{tensors2D:a,outShape:o}=N0e(e,t,n),i=new I0e(a.map(c=>c.shape)),l=n.runWebGLProgram(i,a,r);a.forEach(c=>n.disposeIntermediateTensorInfo(c));let u=ve({inputs:{x:l},attrs:{shape:o},backend:n});return n.disposeIntermediateTensorInfo(l),u}function N0e(e,t,n){let r=_.computeOutShape(e.map(a=>a.shape),t);return{tensors2D:e.map(a=>ve({inputs:{x:a},attrs:{shape:[-1,k.sizeFromShape(a.shape.slice(t))]},backend:n})),outShape:r}}function $E(e){let{inputs:t,backend:n,attrs:r}=e,{axis:s}=r,a=k.parseAxisParam(s,t[0].shape)[0],o=_.computeOutShape(t.map(u=>u.shape),a);if(k.sizeFromShape(o)===0)return n.makeTensorInfo(o,t[0].dtype,[]);let i=t.filter(u=>k.sizeFromShape(u.shape)>0);if(i.length===1)return fr({inputs:{x:i[0]},backend:n});let l=i.map(u=>u.shape);return _.assertParamsConsistent(l,a),wu(i,a,n)}var C0e={kernelName:Ec,backendName:"webgl",kernelFunc:$E},RE=class{constructor(e,t=!1,n=null,r=!1,s=!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,d=e.filterHeight,h=e.filterWidth,p=Math.floor(e.inChannels/4)*4,f=e.inChannels%4,m=e.dataFormat==="channelsLast",g=m?1:2,y=m?2:3,A=m?3:1,x="",b="";n&&(r?x=`float activation(float a) {
float b = getPreluActivationWeightsAtOutCoords();
${n}
}`:s?x=`float activation(float a) {
float b = getLeakyreluAlphaAtOutCoords();
${n}
}`:x=`
float activation(float x) {
${n}
}
`,b="result = activation(result);");let v=t?"result += getBiasAtOutCoords();":"";t&&this.variableNames.push("bias"),r&&this.variableNames.push("preluActivationWeights"),s&&this.variableNames.push("leakyreluAlpha"),this.userCode=`
${x}
const ivec2 strides = ivec2(${i}, ${l});
const ivec2 pads = ivec2(${a}, ${o});
void main() {
ivec4 coords = getOutputCoords();
int batch = coords[0];
int d2 = coords[${A}];
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 < ${d}; wR++) {
int xR = xRCorner + wR * ${u};
if (xR < 0 || xR >= ${e.inHeight}) {
continue;
}
for (int wC = 0; wC < ${h}; wC++) {
int xC = xCCorner + wC * ${c};
if (xC < 0 || xC >= ${e.inWidth}) {
continue;
}
for (int d1 = 0; d1 < ${p}; d1 += 4) {
vec4 wValues = vec4(
getW(wR, wC, d1, d2),
getW(wR, wC, d1 + 1, d2),
getW(wR, wC, d1 + 2, d2),
getW(wR, wC, d1 + 3, d2)
);
if (${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, ${p}) *
getW(wR, wC, ${p}, d2);
} else {
dotProd +=
getX(batch, ${p}, xR, xC) *
getW(wR, wC, ${p}, d2);
}
} else if (${f===2}) {
vec2 wValues = vec2(
getW(wR, wC, ${p}, d2),
getW(wR, wC, ${p} + 1, d2)
);
if (${m}) {
vec2 xValues = vec2(
getX(batch, xR, xC, ${p}),
getX(batch, xR, xC, ${p} + 1)
);
dotProd += dot(xValues, wValues);
} else {
vec2 xValues = vec2(
getX(batch, ${p}, xR, xC),
getX(batch, ${p} + 1, xR, xC)
);
dotProd += dot(xValues, wValues);
}
} else if (${f===3}) {
vec3 wValues = vec3(
getW(wR, wC, ${p}, d2),
getW(wR, wC, ${p} + 1, d2),
getW(wR, wC, ${p} + 2, d2)
);
if (${m}) {
vec3 xValues = vec3(
getX(batch, xR, xC, ${p}),
getX(batch, xR, xC, ${p} + 1),
getX(batch, xR, xC, ${p} + 2)
);
dotProd += dot(xValues, wValues);
} else {
vec3 xValues = vec3(
getX(batch, ${p}, xR, xC),
getX(batch, ${p} + 1, xR, xC),
getX(batch, ${p} + 2, xR, xC)
);
dotProd += dot(xValues, wValues);
}
}
}
}
float result = dotProd;
${v}
${b}
setOutput(result);
}
`}},E0e=class{constructor(e){this.variableNames=["x","W"],this.outputShape=e.outShape;let t=e.padInfo.front,n=e.padInfo.top,r=e.padInfo.left,s=e.strideDepth,a=e.strideHeight,o=e.strideWidth,i=e.dilationDepth,l=e.dilationHeight,u=e.dilationWidth,c=e.filterDepth,d=e.filterHeight,h=e.filterWidth,p=Math.floor(e.inChannels/4)*4,f=e.inChannels%4;this.userCode=`
const ivec3 strides = ivec3(${s}, ${a}, ${o});
const ivec3 pads = ivec3(${t}, ${n}, ${r});
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 < ${d}; wR++) {
int xR = xRCorner + wR * ${l};
if (xR < 0 || xR >= ${e.inHeight}) {
continue;
}
for (int wC = 0; wC < ${h}; wC++) {
int xC = xCCorner + wC * ${u};
if (xC < 0 || xC >= ${e.inWidth}) {
continue;
}
for (int d1 = 0; d1 < ${p}; d1 += 4) {
vec4 xValues = vec4(
getX(batch, xF, xR, xC, d1),
getX(batch, xF, xR, xC, d1 + 1),
getX(batch, xF, xR, xC, d1 + 2),
getX(batch, xF, xR, xC, d1 + 3)
);
vec4 wValues = vec4(
getW(wF, wR, wC, d1, d2),
getW(wF, wR, wC, d1 + 1, d2),
getW(wF, wR, wC, d1 + 2, d2),
getW(wF, wR, wC, d1 + 3, d2)
);
dotProd += dot(xValues, wValues);
}
if (${f===1}) {
dotProd +=
getX(batch, xF, xR, xC, ${p}) *
getW(wF, wR, wC, ${p}, d2);
} else if (${f===2}) {
vec2 xValues = vec2(
getX(batch, xF, xR, xC, ${p}),
getX(batch, xF, xR, xC, ${p} + 1)
);
vec2 wValues = vec2(
getW(wF, wR, wC, ${p}, d2),
getW(wF, wR, wC, ${p} + 1, d2)
);
dotProd += dot(xValues, wValues);
} else if (${f===3}) {
vec3 xValues = vec3(
getX(batch, xF, xR, xC, ${p}),
getX(batch, xF, xR, xC, ${p} + 1),
getX(batch, xF, xR, xC, ${p} + 2)
);
vec3 wValues = vec3(
getW(wF, wR, wC, ${p}, d2),
getW(wF, wR, wC, ${p} + 1, d2),
getW(wF, wR, wC, ${p} + 2, d2)
);
dotProd += dot(xValues, wValues);
}
}
}
}
setOutput(dotProd);
}
`}},$0e=class{constructor(e,t,n){this.variableNames=["A"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e;let{filterWidth:r,inChannels:s,strideWidth:a,strideHeight:o,padInfo:i,outWidth:l,dilationWidth:u,dilationHeight:c,dataFormat:d}=n,{left:h,top:p}=i,f=s*r,m=Wn(),g=d==="channelsLast",y=g?0:1,A=g?1:2,x="";for(let b=0;b<=1;b++)for(let v=0;v<=1;v++)x+=`
blockIndex = rc.y + ${v};
pos = rc.x + ${b};
if(blockIndex < ${e[1]} && pos < ${e[0]}) {
offsetY = int(blockIndex / (${l})) * ${o} - ${p};
d0 = offsetY + ${c} * (pos / ${f});
if(d0 < ${t[y]} && d0 >= 0) {
offsetX = int(mod(float(blockIndex), ${l}.) * ${a}. - ${h}.);
d1 = offsetX + ${u} * (int(mod(float(pos), ${f}.) / ${s}.));
if(d1 < ${t[A]} && d1 >= 0) {
ch = int(mod(float(pos), ${s}.));
if (${g}) {
innerDims = vec2(d1, ch);
result[${b*2+v}] = getChannel(
getA(d0, int(innerDims.x),
int(innerDims.y)), innerDims);
} else {
innerDims = vec2(d0, d1);
result[${b*2+v}] = getChannel(
getA(ch, int(innerDims.x),
int(innerDims.y)), innerDims);
}
}
}
}
`;this.userCode=`
void main() {
ivec2 rc = getOutputCoords();
vec4 result = vec4(0);
int blockIndex, pos, offsetY, d0, offsetX, d1, ch;
vec2 innerDims;
${x}
${m.output} = result;
}
`}};function _E({x:e,filter:t,convInfo:n,backend:r,bias:s=null,preluActivationWeights:a=null,leakyreluAlpha:o=0,activation:i=null}){let l=e.shape,u=r.texData.get(e.dataId),c=n.inChannels,d=l[0]*l[1]*l[2],h=n.outChannels,p=n.dataFormat==="channelsLast",f=!1,m=!1,g,y=[],A=(d===1||h===1)&&c>vE,x=l[2]%2!=0&&!!u.isPacked;if(A||!re().getBool("WEBGL_LAZILY_UNPACK")||!re().getBool("WEBGL_PACK_BINARY_OPERATIONS")||!x){let b=p?l[0]*l[1]*l[2]:l[0]*l[2]*l[3],v=ve({inputs:{x:e},backend:r,attrs:{shape:[1,b,n.inChannels]}}),I=ve({inputs:{x:t},backend:r,attrs:{shape:[1,n.inChannels,n.outChannels]}}),w=Um({a:v,b:I,transposeA:f,transposeB:m,backend:r,bias:s,activation:i,preluActivationWeights:a,leakyreluAlpha:o});g=ve({inputs:{x:w},backend:r,attrs:{shape:n.outShape}}),y.push(v),y.push(I),y.push(w)}else{let b=p?l[0]*l[1]*(l[2]+1):l[0]*l[2]*(l[3]+1),v={dataId:e.dataId,shape:[1,b,n.inChannels],dtype:e.dtype},I=u.shape;u.shape=u.shape.slice(),u.shape[u.shape.length-2]++,k.assert(oh(u.shape,v.shape),()=>`packed reshape ${u.shape} to ${v.shape} isn't free`);let w=ve({inputs:{x:t},backend:r,attrs:{shape:[1,n.inChannels,n.outChannels]}});y.push(w);let S=Um({a:v,b:w,backend:r,transposeA:f,transposeB:m,bias:s,activation:i,preluActivationWeights:a,leakyreluAlpha:o}),E=r.texData.get(S.dataId);k.assert(E.isPacked,()=>"batchMatMul result is expected to be packed"),u.shape=I,E.shape=n.outShape,g=fr({inputs:{x:S},backend:r}),g.shape=n.outShape,y.push(S)}for(let b of y)r.disposeIntermediateTensorInfo(b);return g}function DE({x:e,filter:t,convInfo:n,backend:r,bias:s=null,preluActivationWeights:a=null,leakyreluAlpha:o=0,activation:i=null}){let{filterWidth:l,filterHeight:u,inChannels:c,outWidth:d,outHeight:h,dataFormat:p}=n,f=p==="channelsLast",m=l*u*c,g=h*d,y=[m,g],A=!0,x=!1,b=[],v=ve({inputs:{x:e},backend:r,attrs:{shape:e.shape.slice(1)}}),I=ve({inputs:{x:t},backend:r,attrs:{shape:[1,m,k.sizeFromShape(t.shape)/m]}});b.push(v),b.push(I);let w=new $0e(y,v.shape,n),S=r.runWebGLProgram(w,[v],"float32"),E=ve({inputs:{x:S},backend:r,attrs:{shape:[1,y[0],y[1]]}});b.push(S),b.push(E);let D=s!=null,$=a!=null,R=i==="leakyrelu",N=i?Bm(i,!0):null,M=new gE(E.shape,I.shape,[1,g,n.outChannels],A,x,D,N,$,R),B=[E,I];if(s&&B.push(s),$&&B.push(a),R){let ee=r.makeTensorInfo([],"float32",k.createScalarValue(o,"float32"));B.push(ee),b.push(ee)}let q=r.runWebGLProgram(M,B,"float32"),X=f?[1,h,d,n.outChannels]:[1,n.outChannels,h,d],J=ve({inputs:{x:q},backend:r,attrs:{shape:X}});b.push(q);for(let ee of b)r.disposeIntermediateTensorInfo(ee);return J}function R0e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a}=t,{strides:o,pad:i,dataFormat:l,dilations:u,dimRoundingMode:c}=r,d=_.convertConv2DDataFormat(l),h=_.computeConv2DInfo(s.shape,a.shape,o,u,i,c,!1,d),p;if(h.filterHeight===1&&h.filterWidth===1&&h.dilationHeight===1&&h.dilationWidth===1&&h.strideHeight===1&&h.strideWidth===1&&(h.padInfo.type==="SAME"||h.padInfo.type==="VALID"))p=_E({x:s,filter:a,convInfo:h,backend:n});else if(re().getBool("WEBGL_CONV_IM2COL")&&s.shape[0]===1)p=DE({x:s,filter:a,convInfo:h,backend:n});else{let m=new RE(h);p=n.runWebGLProgram(m,[s,a],"float32")}let f=ve({inputs:{x:p},backend:n,attrs:{shape:h.outShape}});return n.disposeIntermediateTensorInfo(p),f}var _0e={kernelName:el,backendName:"webgl",kernelFunc:R0e},D0e=class{constructor(e){this.variableNames=["x","dy"],this.outputShape=e.filterShape;let t=e.strideHeight,n=e.strideWidth,r=e.padInfo.top,s=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} - ${r};
if (xR < 0 || xR >= ${e.inHeight}) {
continue;
}
for (int yC = 0; yC < ${e.outWidth}; yC++) {
int xC = wC + yC * ${n} - ${s};
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);
}
`}},F0e=class{constructor(e){this.variableNames=["dy","W"],this.outputShape=e.inShape;let t=e.filterHeight,n=e.filterWidth,r=e.strideHeight,s=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) / ${r}.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) / ${s}.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);
}
`}},M0e=class{constructor(e){this.variableNames=["x","dy"],this.outputShape=e.filterShape;let t=e.strideDepth,n=e.strideHeight,r=e.strideWidth,s=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} - ${s};
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 * ${r} - ${o};
if (xC < 0 || xC >= ${e.inWidth}) {
continue;
}
float dyValue = getDy(b, yF, yR, yC, d2);
float xValue = getX(b, xF, xR, xC, d1);
dotProd += (xValue * dyValue);
}
}
}
}
setOutput(dotProd);
}
`}},O0e=class{constructor(e){this.variableNames=["dy","W"],this.outputShape=e.inShape;let t=e.filterDepth,n=e.filterHeight,r=e.filterWidth,s=e.strideDepth,a=e.strideHeight,o=e.strideWidth,i=t-1-e.padInfo.front,l=n-1-e.padInfo.top,u=r-1-e.padInfo.left;this.userCode=`
const ivec3 pads = ivec3(${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) / ${s}.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 < ${r}; 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 = ${r} - 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 P0e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,dy:a}=t,{strides:o,pad:i,dataFormat:l,dimRoundingMode:u,filterShape:c}=r,d=_.convertConv2DDataFormat(l),h=_.computeConv2DInfo(s.shape,c,o,1,i,u,!1,d),p=new D0e(h);return n.runWebGLProgram(p,[s,a],"float32")}var z0e={kernelName:w1,backendName:"webgl",kernelFunc:P0e};function L0e(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,filter:a}=t,{inputShape:o,strides:i,pad:l,dataFormat:u,dimRoundingMode:c}=r,d=_.convertConv2DDataFormat(u),h=_.computeConv2DInfo(o,a.shape,i,1,l,c,!1,d),p=new F0e(h);return n.runWebGLProgram(p,[s,a],"float32")}var B0e={kernelName:tl,backendName:"webgl",kernelFunc:L0e};function W0e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a}=t,{strides:o,pad:i,dilations:l}=r,u=_.computeConv3DInfo(s.shape,a.shape,o,l,i),c=new E0e(u);return n.runWebGLProgram(c,[s,a],"float32")}var V0e={kernelName:Zp,backendName:"webgl",kernelFunc:W0e};function U0e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,dy:a}=t,{strides:o,pad:i,filterShape:l}=r,u=_.computeConv3DInfo(s.shape,l,o,1,i),c=new M0e(u);return n.runWebGLProgram(c,[s,a],"float32")}var H0e={kernelName:k1,backendName:"webgl",kernelFunc:U0e};function G0e(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,filter:a}=t,{pad:o,strides:i,inputShape:l}=r,u=_.computeConv3DInfo(l,a.shape,i,1,o),c=new O0e(u);return n.runWebGLProgram(c,[s,a],"float32")}var j0e={kernelName:I1,backendName:"webgl",kernelFunc:G0e},q0e=mE+`
return cos(x);
`,K0e=it({opSnippet:q0e}),X0e={kernelName:nl,backendName:"webgl",kernelFunc:K0e},Z0e=`
float e2x = exp(-x);
return (e2x + 1.0 / e2x) / 2.0;
`,Y0e=it({opSnippet:Z0e}),J0e={kernelName:rl,backendName:"webgl",kernelFunc:Y0e},Q0e=class{constructor(e,t,n,r,s){this.variableNames=["Image","Boxes","BoxInd"],this.outputShape=[];let[a,o,i,l]=e,[u]=t,[c,d]=n;this.outputShape=[u,c,d,l];let h=r==="bilinear"?1:0,[p,f]=[`${o-1}.0`,`${i-1}.0`],[m,g,y]=c>1?[`${(o-1)/(c-1)}`,"(y2-y1) * height_ratio",`y1*${p} + float(y)*(height_scale)`]:["0.0","0.0",`0.5 * (y1+y2) * ${p}`],[A,x,b]=d>1?[`${(i-1)/(d-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(${A});
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 = ${x};
float in_y = ${y};
if( in_y < 0.0 || in_y > ${p} ) {
setOutput(float(${s}));
return;
}
float in_x = ${b};
if( in_x < 0.0 || in_x > ${f} ) {
setOutput(float(${s}));
return;
}
vec2 sourceFracIndexCR = vec2(in_x,in_y);
if(${h} == 1) {
// Compute the four integer indices.
ivec2 sourceFloorCR = ivec2(sourceFracIndexCR);
ivec2 sourceCeilCR = ivec2(ceil(sourceFracIndexCR));
float topLeft = getImage(b, sourceFloorCR.y, sourceFloorCR.x, d);
float bottomLeft = getImage(b, sourceCeilCR.y, sourceFloorCR.x, d);
float topRight = getImage(b, sourceFloorCR.y, sourceCeilCR.x, d);
float bottomRight = getImage(b, sourceCeilCR.y, sourceCeilCR.x, d);
vec2 fracCR = sourceFracIndexCR - vec2(sourceFloorCR);
float top = topLeft + (topRight - topLeft) * fracCR.x;
float bottom = bottomLeft + (bottomRight - bottomLeft) * fracCR.x;
float newValue = top + (bottom - top) * fracCR.y;
setOutput(newValue);
} else {
// Compute the coordinators of nearest neighbor point.
ivec2 sourceNearestCR = ivec2(floor(
sourceFracIndexCR + vec2(0.5,0.5)));
float newValue = getImage(b, sourceNearestCR.y, sourceNearestCR.x, d);
setOutput(newValue);
}
}
`}},ege=e=>{let{inputs:t,backend:n,attrs:r}=e,{image:s,boxes:a,boxInd:o}=t,{cropSize:i,method:l,extrapolationValue:u}=r,c=new Q0e(s.shape,a.shape,i,l,u);return n.runWebGLProgram(c,[s,a,o],"float32")},tge={kernelName:$c,backendName:"webgl",kernelFunc:ege},FE=class{constructor(e,t,n){this.variableNames=["x"],this.customUniforms=[{name:"index",type:"float"}],this.outputShape=e;let r=e.length,s=t?"0.0":`getX(${ME(r,"coords")})`,a=e[e.length-1],o="",i="";t?(o=n?`end != ${a-1}`:"end != 0",i=n?"end + 1":"end - 1"):(o=n?`end + pow2 < ${a}`:"end >= pow2",i=n?"end + pow2":"end - pow2"),this.userCode=`
void main() {
${wt(r)} coords = getOutputCoords();
int end = ${OE(r,"coords")};
float val = ${s};
int pow2 = int(pow(2.0, index));
if (${o}) {
int idx = ${i};
${OE(r,"coords")} = idx;
val += getX(${ME(r,"coords")});
}
setOutput(val);
}
`}};function ME(e,t){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 sum for rank ${e} is not yet supported`)}function OE(e,t){if(e===1)return`${t}`;if(e===2)return`${t}.y`;if(e===3)return`${t}.z`;if(e===4)return`${t}.w`;throw Error(`Cumulative sum for rank ${e} is not yet supported`)}function nge(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,exclusive:o,reverse:i}=r,l=s.shape.length,u=_.getAxesPermutation([a],l),c=s;u!=null&&(c=Un({inputs:{x:s},backend:n,attrs:{perm:u}}));let d=_.getInnerMostAxes(1,l)[0];if(d!==l-1)throw new Error(`WebGL cumsum shader expects an inner-most axis=${s.shape.length-1} but got axis=${a}`);let h=c.shape[d],p=fr({inputs:{x:c},backend:n});for(let f=0;f<=Math.ceil(Math.log2(h))-1;f++){let m=new FE(c.shape,!1,i),g=[[f]],y=p;p=n.runWebGLProgram(m,[p],p.dtype,g),n.disposeIntermediateTensorInfo(y)}if(o){let f=new FE(c.shape,o,i),m=p;p=n.runWebGLProgram(f,[p],p.dtype),n.disposeIntermediateTensorInfo(m)}if(u!=null){let f=_.getUndoAxesPermutation(u),m=Un({inputs:{x:p},backend:n,attrs:{perm:f}});return n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(c),m}return p}var rge={kernelName:sl,backendName:"webgl",kernelFunc:nge};function sge(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,weights:a}=t,{size:o,binaryOutput:i}=r;if(s.shape.length===1){let l=n.readSync(s.dataId),u=n.readSync(a.dataId),c=QC(l,u,a.dtype,a.shape,o);return n.makeTensorInfo([o],a.dtype,c)}else if(s.shape.length===2){let l=n.bufferSync(s),u=n.bufferSync(a),c=xpe(l,u,o,i);return n.makeTensorInfo(c.shape,a.dtype,c.values)}throw new Error(`Error in denseBincount: input must be at most rank 2, but got rank${s.shape.length}.`)}var age={kernelName:S1,backendName:"webgl",kernelFunc:sge},oge=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 ige(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{blockSize:a,dataFormat:o}=r;k.assert(a>1,()=>`blockSize should be > 1 for depthToSpace, but was: ${a}`);let i=s.shape[0],l=o==="NHWC"?s.shape[1]:s.shape[2],u=o==="NHWC"?s.shape[2]:s.shape[3],c=o==="NHWC"?s.shape[3]:s.shape[1],d=l*a,h=u*a,p=c/(a*a),f=o==="NHWC"?[i,d,h,p]:[i,p,d,h],m=new oge(f,a,o);return n.runWebGLProgram(m,[s],s.dtype)}var lge={kernelName:Rc,backendName:"webgl",kernelFunc:ige},PE=class{constructor(e,t=!1,n=null,r=!1,s=!1){this.variableNames=["x","W"],this.outputShape=e.outShape;let a=e.inHeight,o=e.inWidth,i=e.padInfo.top,l=e.padInfo.left,u=e.strideHeight,c=e.strideWidth,d=e.dilationHeight,h=e.dilationWidth,p=e.filterHeight,f=e.filterWidth,m=e.outChannels/e.inChannels,g="",y="";n&&(r?g=`float activation(float a) {
float b = getPreluActivationWeightsAtOutCoords();
${n}
}`:s?g=`float activation(float a) {
float b = getLeakyreluAlphaAtOutCoords();
${n}
}`:g=`
float activation(float x) {
${n}
}
`,y="result = activation(result);");let A=t?"result += getBiasAtOutCoords();":"";t&&this.variableNames.push("bias"),r&&this.variableNames.push("preluActivationWeights"),s&&this.variableNames.push("leakyreluAlpha"),this.userCode=`
${g}
const ivec2 strides = ivec2(${u}, ${c});
const ivec2 pads = ivec2(${i}, ${l});
void main() {
ivec4 coords = getOutputCoords();
int batch = coords.x;
ivec2 xRCCorner = coords.yz * strides - pads;
int d2 = coords.w;
int d1 = d2 / ${m};
int q = d2 - d1 * ${m};
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 < ${p}; wR++) {
int xR = xRCorner + wR * ${d};
if (xR < 0 || xR >= ${a}) {
continue;
}
for (int wC = 0; wC < ${f}; wC++) {
int xC = xCCorner + wC * ${h};
if (xC < 0 || xC >= ${o}) {
continue;
}
float xVal = getX(batch, xR, xC, d1);
float wVal = getW(wR, wC, d1, q);
dotProd += xVal * wVal;
}
}
float result = dotProd;
${A}
${y}
setOutput(result);
}
`}},zE=class{constructor(e,t=!1,n=null,r=!1,s=!1){this.variableNames=["x","W"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=e.outShape;let a=e.outChannels/e.inChannels,o=e.inHeight,i=e.inWidth,l=e.padInfo.top,u=e.padInfo.left,c=e.strideHeight,d=e.strideWidth,h=e.dilationHeight,p=e.dilationWidth,f=e.filterHeight,m=e.filterWidth,g=m,y=`
int xR; int xC; int xCOffset;
vec4 wTexel; vec4 previous; vec4 final;`;for(let v=0;v<m;v++)y+=`
vec4 xTexelC${v*2};
int xTexelC${v*2}Ready;
vec4 xTexelC${v*2+1};
int xTexelC${v*2+1}Ready;
vec4 xC${v};`;for(let v=0;v<f;v++){for(let I=0;I<m;I++)y+=`
xTexelC${I*2} = vec4(0.0);
xTexelC${I*2}Ready = 0;
xTexelC${I*2+1} = vec4(0.0);
xTexelC${I*2+1}Ready = 0;
xC${I} = vec4(0.0);`;y+=`
xR = xRCorner + ${v*h};
if (xR >=0 && xR < ${o}) {
`;for(let I=0;I<(g+1)/2;I++){let w=I*2,S=w*p;if(y+=`
xC = xCCorner + ${S};
`,d===1){if(w<m&&(u%2==1?(y+=`
xCOffset = xC + 1;
if (xCOffset >= 0 && xCOffset < ${i} && xTexelC${w}Ready == 0) {
xTexelC${w} = getX(batch, xR, xCOffset, d1);
// Need to manually clear unused channels in case
// we're reading from recycled texture.
if (xCOffset + 1 >= ${i}) {
xTexelC${w}.zw = vec2(0.0);
}
xTexelC${w}Ready = 1;
}
`,p===1&&S>0?y+=`
xC${w} = vec4(xTexelC${w-2}.zw, xTexelC${w}.xy);
`:y+=`
xCOffset = xC + 1 - 2;
if (xCOffset >= 0 && xCOffset < ${i}) {
previous = getX(batch, xR, xCOffset, d1);
// Need to manually clear unused channels in case
// we're reading from recycled texture.
if (xCOffset + 1 >= ${i}) {
previous.zw = vec2(0.0);
}
xC${w} = vec4(previous.zw, xTexelC${w}.xy);
} else {
xC${w} = vec4(0.0, 0.0, xTexelC${w}.xy);
}
`):y+=`
if (xC >= 0 && xC < ${i} && xTexelC${w}Ready == 0) {
xTexelC${w} = getX(batch, xR, xC, d1);
if (xC + 1 >= ${i}) {
xTexelC${w}.zw = vec2(0.0);
}
xTexelC${w}Ready = 1;
}
xC${w} = xTexelC${w};
`,S+1<m)){let E=u%2==0?k.nearestLargerEven(p):p;p%2==0&&u%2==1||p%2!=0&&u%2!=1?(y+=`
xCOffset = xC + ${u%2} + ${E};
if (xCOffset >= 0 && xCOffset < ${i} && xTexelC${w+1}Ready == 0) {
xTexelC${w+1} = getX(batch, xR, xCOffset, d1);
// Need to manually clear unused channels in case
// we're reading from recycled texture.
if (xCOffset + 1 >= ${i}) {
xTexelC${w+1}.zw = vec2(0.0);
}
xTexelC${w+1}Ready = 1;
}
`,p>1&&(y+=`
xCOffset -= 2;
if (xCOffset >= 0 && xCOffset < ${i} && xTexelC${w}Ready == 0) {
xTexelC${w} = getX(batch, xR, xCOffset, d1);
xTexelC${w}Ready = 1;
}
`),y+=`
xC${w+1} = vec4(xTexelC${w}.zw, xTexelC${w+1}.xy);
`):E===1?y+=`
xC${w+1} = xTexelC${w};
`:y+=`
xCOffset = xC + ${E};
if (xCOffset >= 0 && xCOffset < ${i} && xTexelC${w+1}Ready == 0) {
xTexelC${w+1} = getX(batch, xR, xCOffset, d1);
if (xCOffset + 1 >= ${i}) {
xTexelC${w+1}.zw = vec2(0.0);
}
xTexelC${w+1}Ready = 1;
}
xC${w+1} = xTexelC${w+1};
`}}else S<m&&(u%2==1?(y+=`
xCOffset = xC + 1 - ${d};
if(xCOffset >= 0 && xCOffset < ${i} && xTexelC${w}Ready == 0) {
xTexelC${w} = getX(batch, xR, xCOffset, d1);
// Need to manually clear unused channels in case
// we're reading from recycled texture.
if (xCOffset + 1 >= ${i}) {
xTexelC${w}.zw = vec2(0.0);
}
xTexelC${w}Ready = 1;
}
if(xC + 1 >= 0 && xC + 1 < ${i} && xTexelC${w+1}Ready == 0) {
xTexelC${w+1} = getX(batch, xR, xC + 1, d1);
// Need to manually clear unused channels in case
// we're reading from recycled texture.
if (xC + 2 >= ${i}) {
xTexelC${w+1}.zw = vec2(0.0);
}
xTexelC${w+1}Ready = 1;
}
xC${w} = vec4(xTexelC${w}.zw, xTexelC${w+1}.zw);
`,S+1<m&&(y+=`
final = vec4(0.0);
xCOffset = xC + 1 + ${d};
if(xCOffset >= 0 && xCOffset < ${i}) {
final = getX(batch, xR, xCOffset, d1);
}
xC${w+1} = vec4(xTexelC${w+1}.xy, final.xy);
`)):(y+=`
if(xC >= 0 && xC < ${i} && xTexelC${w}Ready == 0) {
xTexelC${w} = getX(batch, xR, xC, d1);
if (xC + 1 >= ${i}) {
xTexelC${w}.zw = vec2(0.0);
}
xTexelC${w}Ready = 1;
}
xCOffset = xC + ${d};
if(xCOffset >= 0 && xCOffset < ${i} && xTexelC${w+1}Ready == 0) {
xTexelC${w+1} = getX(batch, xR, xCOffset, d1);
if (xCOffset + 1 >= ${i}) {
xTexelC${w+1}.zw = vec2(0.);
}
xTexelC${w+1}Ready = 1;
}
xC${w} = vec4(
xTexelC${w}.xy, xTexelC${w+1}.xy);
`,S+1<m&&(y+=`
xC${w+1} = vec4(xTexelC${w}.zw, xTexelC${w+1}.zw);
`)));w<m&&(y+=`
wTexel = getW(${v}, ${S}, d1, q);
dotProd += xC${w} * vec4(wTexel.xz, wTexel.xz);
`,S+1<m&&(y+=`
wTexel = getW(${v}, ${S+1}, d1, q);
dotProd += xC${w+1} * vec4(wTexel.xz, wTexel.xz);
`))}y+=`
}
`}let A="",x="";n&&(r?A=`vec4 activation(vec4 a) {
vec4 b = getPreluActivationWeightsAtOutCoords();
${n}
}`:s?A=`vec4 activation(vec4 a) {
vec4 b = getLeakyreluAlphaAtOutCoords();
${n}
}`:A=`vec4 activation(vec4 x) {
${n}
}`,x="result = activation(result);");let b=t?"result += getBiasAtOutCoords();":"";t&&this.variableNames.push("bias"),r&&this.variableNames.push("preluActivationWeights"),s&&this.variableNames.push("leakyreluAlpha"),this.userCode=`
${A}
const ivec2 strides = ivec2(${c}, ${d});
const ivec2 pads = ivec2(${l}, ${u});
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);
${y}
vec4 result = dotProd - vec4(0.000000000000001);
${b}
${x}
setOutput(result);
}
`}};function uge(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a}=t,{strides:o,pad:i,dilations:l,dimRoundingMode:u}=r,c=l;c==null&&(c=[1,1]),k.assert(_.eitherStridesOrDilationsAreOne(o,c),()=>`Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${o} and dilations '${c}'`);let d=_.computeConv2DInfo(s.shape,a.shape,o,c,i,u,!0),h;return re().getBool("WEBGL_PACK_DEPTHWISECONV")&&d.strideWidth<=2&&d.outChannels/d.inChannels==1?h=new zE(d):h=new PE(d),n.runWebGLProgram(h,[s,a],"float32")}var cge={kernelName:al,backendName:"webgl",kernelFunc:uge},dge=class{constructor(e){this.variableNames=["x","dy"],this.outputShape=e.filterShape;let t=e.strideHeight,n=e.strideWidth,r=e.padInfo.top,s=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} - ${r};
if (xR < 0 || xR >= ${e.inHeight}) {
continue;
}
for (int yC = 0; yC < ${e.outWidth}; yC++) {
int xC = wC + yC * ${n} - ${s};
if (xC < 0 || xC >= ${e.inWidth}) {
continue;
}
float dyValue = getDy(b, yR, yC, d2);
float xValue = getX(b, xR, xC, d1);
dotProd += (xValue * dyValue);
}
}
}
setOutput(dotProd);
}
`}},hge=class{constructor(e){this.variableNames=["dy","W"],this.outputShape=e.inShape;let t=e.filterHeight,n=e.filterWidth,r=e.strideHeight,s=e.strideWidth,a=t-1-e.padInfo.top,o=n-1-e.padInfo.left,i=e.outChannels/e.inChannels;this.userCode=`
const ivec2 pads = ivec2(${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) / ${r}.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) / ${s}.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 * ${i} + dm;
float xValue = getDy(batch, idyR, idyC, d2);
float wValue = getW(wRPerm, wCPerm, d1, dm);
dotProd += xValue * wValue;
}
}
}
setOutput(dotProd);
}
`}};function pge(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,dy:a}=t,{strides:o,dilations:i,pad:l,dimRoundingMode:u,filterShape:c}=r,d=_.computeConv2DInfo(s.shape,c,o,i,l,u,!0),h=new dge(d);return n.runWebGLProgram(h,[s,a],"float32")}var fge={kernelName:T1,backendName:"webgl",kernelFunc:pge};function mge(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,filter:a}=t,{strides:o,dilations:i,pad:l,dimRoundingMode:u,inputShape:c}=r,d=_.computeConv2DInfo(c,a.shape,o,i,l,u,!0),h=new hge(d);return n.runWebGLProgram(h,[s,a],"float32")}var gge={kernelName:N1,backendName:"webgl",kernelFunc:mge},yge=class{constructor(e){this.variableNames=["X"],this.outputShape=[e,e],this.userCode=`
void main() {
ivec2 coords = getOutputCoords();
float val = coords[0] == coords[1] ? getX(coords[0]) : 0.0;
setOutput(val);
}
`}};function Age(e){let{inputs:t,backend:n}=e,{x:r}=t,s=[...r.shape,...r.shape],a=k.sizeFromShape(r.shape),o=ve({inputs:{x:r},backend:n,attrs:{shape:[a]}}),i=new yge(a),l=n.runWebGLProgram(i,[o],o.dtype),u=ve({inputs:{x:l},backend:n,attrs:{shape:s}});return n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(l),u}var xge={kernelName:C1,backendName:"webgl",kernelFunc:Age},bge=class{constructor(e){this.variableNames=["x","W"],this.outputShape=e.outShape;let{inHeight:t,inWidth:n,padInfo:r,strideHeight:s,strideWidth:a,filterHeight:o,filterWidth:i,dilationHeight:l,dilationWidth:u}=e,{top:c,left:d}=r;this.userCode=`
const ivec2 strides = ivec2(${s}, ${a});
const ivec2 pads = ivec2(${c}, ${d});
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 < ${n}) {
float xVal = getX(batch, hIn, wIn, d1);
float wVal = getW(h, w, d1);
float val = xVal + wVal;
if (val > curVal) {
curVal = val;
}
}
}
}
}
float result = curVal;
setOutput(result);
}
`}};function vge(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a}=t,{strides:o,pad:i,dilations:l}=r,u=_.computeDilation2DInfo(s.shape,a.shape,o,i,"NHWC",l),c,d=new bge(u);c=n.runWebGLProgram(d,[s,a],"float32");let h=ve({inputs:{x:c},backend:n,attrs:{shape:u.outShape}});return n.disposeIntermediateTensorInfo(c),h}var wge={kernelName:Yp,backendName:"webgl",kernelFunc:vge};function kge(e){let{inputs:t,backend:n,attrs:r}=e,{equation:s}=r,a=t,{allDims:o,summedDims:i,idDims:l}=_.decodeEinsumEquation(s,a.length);_.checkEinsumDimSizes(o.length,l,a);let{path:u,steps:c}=_.getEinsumComputePath(i,l),d=c.length,h=null,p=o.length,f=[];for(let m=0;m<d;++m){for(let g of c[m]){let{permutationIndices:y,expandDims:A}=_.getEinsumPermutation(p,l[g]),x;_.isIdentityPermutation(y)?x=a[g]:(x=Un({inputs:{x:a[g]},backend:n,attrs:{perm:y}}),f.push(x));let b=x.shape.slice();for(let v=0;v<A.length;++v)b.splice(A[v],0,1);k.arraysEqual(x.shape,b)||(x=ve({inputs:{x},backend:n,attrs:{shape:b}}),f.push(x)),h===null?h=x:(h=Y5({inputs:{a:x,b:h},backend:n}),f.push(h))}m<d-1&&(u[m]>=0&&(h=Vm({inputs:{x:h},backend:n,attrs:{axis:u[m]-(o.length-p),keepDims:!1}}),f.push(h)),p--)}for(let m of f)m!==h&&n.disposeIntermediateTensorInfo(m);return h}var Ige={kernelName:R1,backendName:"webgl",kernelFunc:kge},Sge="return (x >= 0.0) ? x : (exp(x) - 1.0);",Tge=`
vec4 result;
result.r = (x.r >= 0.0) ? x.r : (exp(x.r) - 1.0);
result.g = (x.g >= 0.0) ? x.g : (exp(x.g) - 1.0);
result.b = (x.b >= 0.0) ? x.b : (exp(x.b) - 1.0);
result.a = (x.a >= 0.0) ? x.a : (exp(x.a) - 1.0);
return result;
`,Nge=it({opSnippet:Sge,packedOpSnippet:Tge}),Cge={kernelName:_c,backendName:"webgl",kernelFunc:Nge},Ege="return (b >= 1.0) ? a : a * (b + 1.0);",$ge=`
vec4 bGTEZero = vec4(greaterThanEqual(b, vec4(0.)));
return (bGTEZero * a) + ((vec4(1.0) - bGTEZero) * (a * (b + vec4(1.0))));
`,Rge=e=>{let{inputs:t,backend:n}=e,{dy:r,y:s}=t,a=re().getBool("WEBGL_PACK_BINARY_OPERATIONS")?new ch($ge,r.shape,s.shape):new bu(Ege,r.shape,s.shape);return n.runWebGLProgram(a,[r,s],r.dtype)},_ge={kernelName:_1,backendName:"webgl",kernelFunc:Rge},Dge=`
return vec4(equal(a, b));
`,Fge="return float(a == b);",Mge=Tn({opSnippet:Fge,packedOpSnippet:Dge,dtype:"bool",cpuKernelImpl:wpe}),Oge={kernelName:il,backendName:"webgl",kernelFunc:Mge},Pge=`
// Error function is calculated approximately with elementary function.
// See "Handbook of Mathematical Functions with Formulas,
// Graphs, and Mathematical Tables", Abramowitz and Stegun.
float p = ${_.ERF_P};
float a1 = ${_.ERF_A1};
float a2 = ${_.ERF_A2};
float a3 = ${_.ERF_A3};
float a4 = ${_.ERF_A4};
float a5 = ${_.ERF_A5};
float sign = sign(x);
x = abs(x);
float t = 1.0 / (1.0 + p * x);
return sign * (1.0 - (((((a5*t + a4)*t) + a3)*t + a2)*t + a1)*t*exp(-x*x));
`,zge=it({opSnippet:Pge}),Lge={kernelName:Dc,backendName:"webgl",kernelFunc:zge},LE="return exp(x);",BE=it({opSnippet:LE,packedOpSnippet:LE,cpuKernelImpl:kpe}),Bge={kernelName:Eo,backendName:"webgl",kernelFunc:BE};function tb(e){let{inputs:t,attrs:n,backend:r}=e,{dim:s}=n,{input:a}=t,o=a.shape.length,i=a.shape.slice(),l=s;return s<0&&(k.assert(-(o+1)<=s,()=>`Axis must be in the interval [${-(o+1)}, ${o}]`),l=o+s+1),i.splice(l,0,1),ve({inputs:{x:a},backend:r,attrs:{shape:i}})}var Wge={kernelName:Fc,backendName:"webgl",kernelFunc:tb},WE="return exp(x) - 1.0;",Vge=it({opSnippet:WE,packedOpSnippet:WE,cpuKernelImpl:Ipe}),Uge={kernelName:ll,backendName:"webgl",kernelFunc:Vge},VE=class{constructor(e,t,n){this.variableNames=["real","imag"];let r=t[1];this.outputShape=t;let s=n?`2.0 * ${Math.PI}`:`-2.0 * ${Math.PI}`,a=n?`${r}.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 = ${s};
float unaryOpComplex(float real, float expR, float imag, float expI) {
${o}
}
float mulMatDFT(int batch, int index) {
float indexRatio = float(index) / float(${r});
float exponentMultiplierTimesIndexRatio =
exponentMultiplier * indexRatio;
float result = 0.0;
for (int i = 0; i < ${r}; 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 UE(e,t,n){let r=n.texData.get(e.dataId),s=k.sizeFromShape(e.shape),a=e.shape[e.shape.length-1],o=s/a,i=ve({inputs:{x:e},backend:n,attrs:{shape:[o,a]}}),l=i.shape,u=new VE("real",l,t),c=new VE("imag",l,t),d=[{dataId:r.complexTensorInfos.real.dataId,dtype:r.complexTensorInfos.real.dtype,shape:l},{dataId:r.complexTensorInfos.imag.dataId,dtype:r.complexTensorInfos.imag.dtype,shape:l}],h=n.runWebGLProgram(u,d,"float32"),p=n.runWebGLProgram(c,d,"float32"),f=eo({inputs:{real:h,imag:p},backend:n});n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(p);let m=ve({inputs:{x:f},backend:n,attrs:{shape:e.shape}});return n.disposeIntermediateTensorInfo(i),n.disposeIntermediateTensorInfo(f),m}function Hge(e){let{inputs:t,backend:n}=e,{input:r}=t;return UE(r,!1,n)}var Gge={kernelName:D1,backendName:"webgl",kernelFunc:Hge},jge=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 qm(e){let{backend:t,attrs:n}=e,{shape:r,value:s}=n,{dtype:a}=n;if(a=a||k.inferDtype(s),a==="string"){let o=k.getArrayFromDType(a,k.sizeFromShape(r));return o.fill(s),t.makeTensorInfo(r,a,o)}else{let o=new jge(r,s),i=[[s]];return t.runWebGLProgram(o,[],a,i)}}var qge={kernelName:Jp,backendName:"webgl",kernelFunc:qm},Kge=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 < ${t}) {
outputValue = getImage(coords[0], coords[1], coordX, coords[3]);
} else {
outputValue = getImage(coords[0], coords[1], coords[2], coords[3]);
}
setOutput(outputValue);
}
`}},Xge={kernelName:Mc,backendName:"webgl",kernelFunc:({inputs:e,backend:t})=>{let{image:n}=e,r=t,s=new Kge(n.shape);return r.runWebGLProgram(s,[n],n.dtype)}},HE="return floor(x);",Zge=it({opSnippet:HE,packedOpSnippet:HE,cpuKernelImpl:Spe}),Yge={kernelName:$o,backendName:"webgl",kernelFunc:Zge},Jge=`
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;
}
`,Qge=`
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);
`,e2e=Tn({opSnippet:Jge,packedOpSnippet:Qge,dtype:"int32"}),t2e={kernelName:ul,backendName:"webgl",kernelFunc:e2e},n2e=class{constructor(e){this.variableNames=["A"];let t=Wn(),[n,r]=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(${r}.0, ${n}.0);
vec4 values = ${t.texture2D}(A, uv);
float value;
if (depth == 0) {
value = values.r;
} else if (depth == 1) {
value = values.g;
} else if (depth == 2) {
value = values.b;
} else if (depth == 3) {
value = values.a;
}
setOutput(floor(value * 255.0 + 0.5));
}
`}},r2e=class{constructor(e){this.variableNames=["A"],this.packedInputs=!1,this.packedOutput=!0;let t=Wn(),[n,r]=e;this.outputShape=e,this.userCode=`
void main() {
ivec3 coords = getOutputCoords();
int texR = coords[0];
int texC = coords[1];
int depth = coords[2];
vec4 result = vec4(0.);
for(int row=0; row<=1; row++) {
for(int col=0; col<=1; col++) {
texC = coords[1] + row;
depth = coords[2] + col;
vec2 uv = (vec2(texC, texR) + halfCR) /
vec2(${r}.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;
}
`}},s2e={kernelName:ey,backendName:"webgl",kernelFunc:a2e},ku;function a2e(e){let{inputs:t,backend:n,attrs:r}=e,{pixels:s}=t,{numChannels:a}=r,o=typeof HTMLVideoElement!="undefined"&&s instanceof HTMLVideoElement,i=typeof HTMLImageElement!="undefined"&&s instanceof HTMLImageElement,[l,u]=o?[s.videoWidth,s.videoHeight]:[s.width,s.height],c=[u,l],d=[u,l,a];(i||o)&&(ku==null&&(ku=document.createElement("canvas").getContext("2d")),ku.canvas.width=l,ku.canvas.height=u,ku.drawImage(s,0,0,l,u),s=ku.canvas);let h=n.makeTensorInfo(c,"int32");n.texData.get(h.dataId).usage=Fr.PIXELS,n.gpgpu.uploadPixelDataToTexture(n.getTexture(h.dataId),s);let p=re().getBool("WEBGL_PACK")?new r2e(d):new n2e(d),f=n.runWebGLProgram(p,[h],"int32");return n.disposeData(h.dataId),f}function o2e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dataFormat:c,dilations:d,dimRoundingMode:h,activation:p,leakyreluAlpha:f}=r,m=_.convertConv2DDataFormat(c),g=_.computeConv2DInfo(s.shape,a.shape,l,d,u,h,!1,m),y,A=[];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=_E({x:s,filter:a,convInfo:g,backend:n,bias:o,activation:p,preluActivationWeights:i,leakyreluAlpha:f});else if(re().getBool("WEBGL_CONV_IM2COL")&&s.shape[0]===1)y=DE({x:s,filter:a,convInfo:g,backend:n,bias:o,activation:p,preluActivationWeights:i,leakyreluAlpha:f});else{let b=o!=null,v=i!=null,I=p==="leakyrelu",w=p?Bm(p,!1):null,S=new RE(g,b,w,v,I),E=[s,a];if(o&&E.push(o),i&&E.push(i),I){let D=n.makeTensorInfo([],"float32",k.createScalarValue(f,"float32"));E.push(D),A.push(D)}y=n.runWebGLProgram(S,E,"float32")}let x=ve({inputs:{x:y},backend:n,attrs:{shape:g.outShape}});return A.push(y),A.forEach(b=>n.disposeIntermediateTensorInfo(b)),x}var i2e={kernelName:Bl,backendName:"webgl",kernelFunc:o2e};function l2e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dilations:c,dimRoundingMode:d,activation:h,leakyreluAlpha:p}=r,f=[],m=c;m==null&&(m=[1,1]),k.assert(_.eitherStridesOrDilationsAreOne(l,m),()=>`Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${l} and dilations '${m}'`);let g=_.computeConv2DInfo(s.shape,a.shape,l,m,u,d,!0),y=re().getBool("WEBGL_PACK_DEPTHWISECONV")&&g.strideWidth<=2&&g.outChannels/g.inChannels==1,A=h?Bm(h,y):null,x=[s,a],b=o!=null,v=i!=null,I=h==="leakyrelu";if(b&&x.push(o),v&&x.push(i),I){let E=n.makeTensorInfo([],"float32",k.createScalarValue(p,"float32"));x.push(E),f.push(E)}let w;y?w=new zE(g,b,A,v,I):w=new PE(g,b,A,v,I);let S=n.runWebGLProgram(w,x,"float32");return f.forEach(E=>n.disposeIntermediateTensorInfo(E)),S}var u2e={kernelName:Wl,backendName:"webgl",kernelFunc:l2e},c2e=class{constructor(e,t,n){this.sliceDim=e,this.strides=t,this.variableNames=["x","indices"],this.outputShape=n;let r=wt(t.length),s=wt(n.length),a=this.sliceDim>1?"strides[j]":"strides";this.userCode=`
${r} strides = ${r}(${this.strides});
void main() {
${s} coords = getOutputCoords();
int flattenIndex = 0;
for (int j = 0; j < ${this.sliceDim}; j++) {
int index = round(getIndices(coords[0], j));
flattenIndex += index * ${a};
}
setOutput(getX(flattenIndex, coords[1]));
}
`}};function d2e(e){let{inputs:t,backend:n}=e,{params:r,indices:s}=t,a=s.shape,o=a[a.length-1],i=k.sizeFromShape(r.shape),[l,u,c,d]=_.prepareAndValidate(r,s),h=ve({inputs:{x:s},backend:n,attrs:{shape:[u,o]}}),p=ve({inputs:{x:r},backend:n,attrs:{shape:[k.sizeFromShape(r.shape)/c,c]}});if(n.shouldExecuteOnCPU([r,s])||r.dtype==="string"){let y=n.readSync(s.dataId),A=n.bufferSync(r),x=Tpe(y,A,r.dtype,u,o,c,d,r.shape,i);return n.makeTensorInfo(l,r.dtype,x.values)}let f=new c2e(o,d,[u,c]),m=n.runWebGLProgram(f,[p,h],p.dtype),g=ve({inputs:{x:m},backend:n,attrs:{shape:l}});return n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(m),g}var h2e={kernelName:Pc,backendName:"webgl",kernelFunc:d2e},p2e=class{constructor(e,t){this.variableNames=["A","indices"],this.outputShape=t,this.rank=t.length;let n=wt(this.rank),r=f2e(e,2);this.userCode=`
void main() {
${n} resRC = getOutputCoords();
setOutput(getA(${r}));
}
`}};function f2e(e,t){let n=["resRC.x","resRC.y","resRC.z","resRC.w"],r=[];for(let s=0;s<e.length;s++)s===2?r.push("int(getIndices(resRC.x, resRC.z))"):r.push(`${n[s]}`);return r.join()}function GE(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,indices:a}=t,{axis:o,batchDims:i}=r,l=k.parseAxisParam(o,s.shape)[0],u=_.segment_util.collectGatherOpShapeInfo(s,a,l,i),c=k.sizeFromShape(a.shape),d=[],h=ve({inputs:{x:s},backend:n,attrs:{shape:[u.batchSize,u.outerSize,u.dimSize,u.sliceSize]}}),p=ve({inputs:{x:a},backend:n,attrs:{shape:[u.batchSize,c/u.batchSize]}});d.push(h),d.push(p);let f=[u.batchSize,u.outerSize,c/u.batchSize,u.sliceSize];if(n.shouldExecuteOnCPU([s,a])||s.dtype==="string"){let A=n.bufferSync(p),x=n.bufferSync(h),b=Npe(x,A,f);return d.forEach(v=>n.disposeIntermediateTensorInfo(v)),n.makeTensorInfo(u.outputShape,b.dtype,b.values)}let m=new p2e(h.shape,f),g=n.runWebGLProgram(m,[h,p],h.dtype);d.push(g);let y=ve({inputs:{x:g},backend:n,attrs:{shape:u.outputShape}});return d.forEach(A=>n.disposeIntermediateTensorInfo(A)),y}var m2e={kernelName:Oc,backendName:"webgl",kernelFunc:GE},g2e="return float(a > b);",y2e=`
return vec4(greaterThan(a, b));
`,A2e=Tn({opSnippet:g2e,packedOpSnippet:y2e,cpuKernelImpl:Cpe,dtype:"bool"}),x2e={kernelName:dl,backendName:"webgl",kernelFunc:A2e},b2e="return float(a >= b);",v2e=`
return vec4(greaterThanEqual(a, b));
`,w2e=Tn({opSnippet:b2e,packedOpSnippet:v2e,dtype:"bool",cpuKernelImpl:Epe}),k2e={kernelName:Ro,backendName:"webgl",kernelFunc:w2e};function I2e(e){let{inputs:t,backend:n}=e,{input:r}=t;return UE(r,!0,n)}var S2e={kernelName:F1,backendName:"webgl",kernelFunc:I2e},T2e="return float(!isnan(x) && !isinf(x));",N2e=it({opSnippet:T2e,dtype:"bool"}),C2e={kernelName:zc,backendName:"webgl",kernelFunc:N2e},E2e="return float(isinf(x));",$2e=it({opSnippet:E2e,dtype:"bool"}),R2e={kernelName:Lc,backendName:"webgl",kernelFunc:$2e},_2e="return float(isnan(x));",D2e=it({opSnippet:_2e,dtype:"bool"}),F2e={kernelName:Bc,backendName:"webgl",kernelFunc:D2e},M2e="return float(a < b);",O2e=`
return vec4(lessThan(a, b));
`,P2e=Tn({opSnippet:M2e,packedOpSnippet:O2e,cpuKernelImpl:$pe,dtype:"bool"}),z2e={kernelName:fl,backendName:"webgl",kernelFunc:P2e},L2e="return float(a <= b);",B2e=`
return vec4(lessThanEqual(a, b));
`,W2e=Tn({opSnippet:L2e,packedOpSnippet:B2e,cpuKernelImpl:Rpe,dtype:"bool"}),V2e={kernelName:ml,backendName:"webgl",kernelFunc:W2e};function U2e(e){let{backend:t,attrs:n}=e,{start:r,stop:s,num:a}=n,o=_pe(r,s,a);return t.makeTensorInfo([o.length],"float32",o)}var H2e={kernelName:O1,backendName:"webgl",kernelFunc:U2e},G2e=`if (x < 0.0) return NAN;
return log(x);`,j2e=`
vec4 result = log(x);
vec4 isNaN = vec4(lessThan(x, vec4(0.0)));
result.r = isNaN.r == 1.0 ? NAN : result.r;
result.g = isNaN.g == 1.0 ? NAN : result.g;
result.b = isNaN.b == 1.0 ? NAN : result.b;
result.a = isNaN.a == 1.0 ? NAN : result.a;
return result;
`,q2e=it({opSnippet:G2e,packedOpSnippet:j2e,cpuKernelImpl:Dpe}),K2e={kernelName:_o,backendName:"webgl",kernelFunc:q2e},X2e="return log(1.0 + x);",Z2e=it({opSnippet:X2e}),Y2e={kernelName:Wc,backendName:"webgl",kernelFunc:Z2e},J2e="return float(a >= 1.0 && b >= 1.0);",Q2e=`
return vec4(
vec4(greaterThanEqual(a, vec4(1.0))) *
vec4(greaterThanEqual(b, vec4(1.0))));
`,e1e=Tn({opSnippet:J2e,packedOpSnippet:Q2e,dtype:"bool"}),t1e={kernelName:Vc,backendName:"webgl",kernelFunc:e1e},n1e="return float(!(x >= 1.0));",r1e=it({opSnippet:n1e}),s1e={kernelName:Qp,backendName:"webgl",kernelFunc:r1e},a1e="return float(a >= 1.0 || b >= 1.0);",o1e=`
return min(
vec4(greaterThanEqual(a, vec4(1.0))) +
vec4(greaterThanEqual(b, vec4(1.0))),
vec4(1.0));
`,i1e=Tn({opSnippet:a1e,packedOpSnippet:o1e,dtype:"bool"}),l1e={kernelName:ef,backendName:"webgl",kernelFunc:i1e},u1e=class{constructor(e,t,n,r,s){this.variableNames=["x"],this.outputShape=[];let a=t,o=e[3]-1;this.outputShape=e;let i,l=`float(${n}) + float(${r}) * sum`;s===.5?i=`inversesqrt(${l})`:s===1?i=`1.0/(${l})`:i=`exp(log(${l}) * float(-${s}));`,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 * ${i};
setOutput(val);
}
`}},c1e=class{constructor(e,t,n,r,s){this.variableNames=["x"],this.outputShape=[],this.packedInputs=!0,this.packedOutput=!0;let a=t,o=e[3]-1;this.outputShape=e;let i,l=`float(${n}) + float(${r}) * sum`;s===.5?i=`inversesqrt(${l})`:s===1?i=`1.0/(${l})`:i=`exp(log(${l}) * float(-${s}));`,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);
}
`}},d1e=e=>{let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{depthRadius:a,bias:o,alpha:i,beta:l}=r,u=re().getBool("WEBGL_PACK_NORMALIZATION")?new c1e(s.shape,a,o,i,l):new u1e(s.shape,a,o,i,l);return n.runWebGLProgram(u,[s],s.dtype)},h1e={kernelName:tf,backendName:"webgl",kernelFunc:d1e},p1e=class{constructor(e,t,n,r,s){this.variableNames=["inputImage","outputImage","dy"],this.outputShape=[],this.outputShape=e,this.depth=e[3],this.depthRadius=t,this.bias=n,this.alpha=r,this.beta=s,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 = ${this.depth};
float norm = 0.0;
for (int k = MIN_DEPTH_BEGIN; k < MAX_DEPTH_END; ++k) {
if (k < depthBegin){
continue;
}
else if (k >= depthBegin && k < depthEnd) {
norm += getInputImage(b, r, c, k) * getInputImage(b, r, c, k);
}
else {
break;
}
}
norm = float(${r}) * norm + float(${n});
for(int k = MIN_DEPTH_BEGIN; k < MAX_DEPTH_END; ++k){
if (k < depthBegin){
continue;
}
else if (k >= depthBegin && k < depthEnd){
float dyi = -2.0 * float(${r})
* float(${s})
* getInputImage(b ,r ,c, k) * getOutputImage(b, r, c, d)
/ norm;
if (k == d) {
dyi += pow(norm, -1.0 * ${s});
}
if (k == coords[3]) {
dyi *= getDy(b, r, c, d);
result += dyi;
}
}
else {
break;
}
}
}
setOutput(result);
}
`}},f1e=e=>{let{inputs:t,backend:n,attrs:r}=e,{x:s,y:a,dy:o}=t,{depthRadius:i,bias:l,alpha:u,beta:c}=r,d=new p1e(s.shape,i,l,u,c);return n.runWebGLProgram(d,[s,a,o],s.dtype)},m1e={kernelName:P1,backendName:"webgl",kernelFunc:f1e};function g1e(e,t,n,r){let s=k.sizeFromShape(t),o=k.sizeFromShape(e.shape)/s,i=ve({inputs:{x:e},attrs:{shape:[o,s]},backend:r}),l=Ai(i,e.dtype,"max",r),u=ve({inputs:{x:l},attrs:{shape:n},backend:r});return r.disposeIntermediateTensorInfo(i),r.disposeIntermediateTensorInfo(l),u}function jE(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{reductionIndices:a,keepDims:o}=r,i=s.shape.length,l=k.parseAxisParam(a,s.shape),u=l,c=_.getAxesPermutation(u,i),d=c!=null,h=n.shouldExecuteOnCPU([s]),p=s;if(d){if(h){let x=n.texData.get(p.dataId).values,b=new Array(i);for(let w=0;w<b.length;w++)b[w]=s.shape[c[w]];let v=Z5(x,s.shape,s.dtype,c,b);p=n.makeTensorInfo(b,s.dtype);let I=n.texData.get(p.dataId);I.values=v}else p=Wm(s,c,n);u=_.getInnerMostAxes(u.length,i)}_.assertAxesAreInnerMostDims("max",u,i);let[f,m]=_.computeOutAndReduceShapes(p.shape,u),g=f;o&&(g=_.expandShapeToKeepDim(f,l));let y;if(h){let x=n.texData.get(p.dataId).values,b=Fpe(x,k.sizeFromShape(m),g,s.dtype);y=n.makeTensorInfo(g,s.dtype);let v=n.texData.get(y.dataId);v.values=b}else y=g1e(p,m,g,n);return d&&n.disposeIntermediateTensorInfo(p),y}var y1e={kernelName:gl,backendName:"webgl",kernelFunc:jE},A1e=cE+`
return max(a, b);
`,x1e=`
vec4 result = vec4(max(a, b));
vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
`+Lm+`
return result;
`,b1e=Tn({opSnippet:A1e,packedOpSnippet:x1e,cpuKernelImpl:Mpe}),v1e={kernelName:Do,backendName:"webgl",kernelFunc:b1e};function w1e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t;hu(s,"maxPool");let{filterSize:a,strides:o,pad:i,dimRoundingMode:l}=r,u=1;k.assert(_.eitherStridesOrDilationsAreOne(o,u),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${o} and dilations '${u}'`);let c=_.computePool2DInfo(s.shape,a,o,u,i,l);if(c.filterWidth===1&&c.filterHeight===1&&k.arraysEqual(c.inShape,c.outShape))return fr({inputs:{x:s},backend:n});let d=new dh(c,"max",!1);return n.runWebGLProgram(d,[s],s.dtype)}var k1e={kernelName:yl,backendName:"webgl",kernelFunc:w1e};function I1e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{filterSize:a,strides:o,pad:i,dataFormat:l,dimRoundingMode:u}=r,c=[1,1,1],d=_.computePool3DInfo(s.shape,a,o,c,i,u,l),h=new J5(d,"max",!1);return n.runWebGLProgram(h,[s],s.dtype)}var S1e={kernelName:nf,backendName:"webgl",kernelFunc:I1e},T1e=class{constructor(e){this.variableNames=["dy","maxPos"],this.outputShape=e.inShape;let t=e.strideHeight,n=e.strideWidth,r=e.dilationHeight,s=e.effectiveFilterHeight,a=e.effectiveFilterWidth,o=s-1-e.padInfo.top,i=a-1-e.padInfo.left,l=s*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 < ${s};
wR += ${r}) {
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 * ${a} + wC;
float mask = float(maxPosValue == curPosValue ? 1.0 : 0.0);
dotProd += dyValue * mask;
}
}
setOutput(dotProd);
}
`}},N1e=class{constructor(e){this.variableNames=["dy","maxPos"],this.outputShape=e.inShape;let t=e.strideDepth,n=e.strideHeight,r=e.strideWidth,s=e.dilationDepth,a=e.dilationHeight,o=e.dilationWidth,i=e.effectiveFilterDepth,l=e.effectiveFilterHeight,u=e.effectiveFilterWidth,c=i-1-e.padInfo.front,d=l-1-e.padInfo.top,h=u-1-e.padInfo.left,p=i*l*u-1;this.userCode=`
const ivec3 pads = ivec3(${c}, ${d}, ${h});
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 += ${s}) {
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) / ${r}.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 = ${p} -
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 C1e(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,input:a}=t,o=a,{filterSize:i,strides:l,pad:u,dimRoundingMode:c}=r,d=[1,1,1],h=_.computePool3DInfo(o.shape,i,l,d,u,c),p=new J5(h,"max",!0),f=n.runWebGLProgram(p,[o],o.dtype),m=new N1e(h),g=n.runWebGLProgram(m,[s,f],o.dtype);return n.disposeIntermediateTensorInfo(f),g}var E1e={kernelName:L1,backendName:"webgl",kernelFunc:C1e};function $1e(e){let{inputs:t,backend:n,attrs:r}=e,{dy:s,input:a,output:o}=t,i=a;hu([a,o],"maxPoolGrad");let{filterSize:l,strides:u,pad:c,dimRoundingMode:d}=r,h=_.computePool2DInfo(i.shape,l,u,1,c,d),p=!0,f=new dh(h,"max",p),m=n.runWebGLProgram(f,[i],i.dtype),g=new T1e(h),y=n.runWebGLProgram(g,[s,m],i.dtype);return n.disposeIntermediateTensorInfo(m),y}var R1e={kernelName:z1,backendName:"webgl",kernelFunc:$1e};function _1e(e,t,n,r){let s=new dh(n,"max",!1),a=r.runWebGLProgram(s,[e],"float32");s=new dh(n,"max",!0,!0,t);let o=r.runWebGLProgram(s,[e],"float32");return[a,o]}var D1e={kernelName:B1,backendName:"webgl",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{x:r}=e,{filterSize:s,strides:a,pad:o,includeBatchInIndex:i}=t,l=n;k.assert(r.shape.length===4,()=>`Error in maxPool: input must be rank 4 but got rank ${r.shape.length}.`);let u=[1,1];k.assert(_.eitherStridesOrDilationsAreOne(a,u),()=>`Error in maxPool: Either strides or dilations must be 1. Got strides ${a} and dilations '${u}'`);let c=_.computePool2DInfo(r.shape,s,a,u,o),[d,h]=_1e(r,i,c,l);return[d,h]}};function F1e(e,t,n,r){let s=k.sizeFromShape(t),o=k.sizeFromShape(e.shape)/s,i=ve({inputs:{x:e},attrs:{shape:[o,s]},backend:r}),l=Ai(i,"float32","mean",r),u=ve({inputs:{x:l},attrs:{shape:n},backend:r});return r.disposeIntermediateTensorInfo(i),r.disposeIntermediateTensorInfo(l),u}var M1e={kernelName:Al,backendName:"webgl",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{x:r}=e,{keepDims:s,axis:a}=t,o=n,i=r.shape.length,l=k.parseAxisParam(a,r.shape),u=l,c=_.getAxesPermutation(u,i),d=c!=null,h=o.shouldExecuteOnCPU([r]),p=[],f=r;if(d){if(h){let b=o.texData.get(f.dataId).values,v=new Array(i);for(let S=0;S<v.length;S++)v[S]=r.shape[c[S]];let I=Z5(b,r.shape,r.dtype,c,v);f=o.makeTensorInfo(v,r.dtype);let w=o.texData.get(f.dataId);w.values=I}else f=Wm(r,c,o);p.push(f),u=_.getInnerMostAxes(u.length,i)}_.assertAxesAreInnerMostDims("sum",u,i);let[m,g]=_.computeOutAndReduceShapes(f.shape,u),y=m;s&&(y=_.expandShapeToKeepDim(m,l));let A=F1e(f,g,y,o);for(let x of p)o.disposeIntermediateTensorInfo(x);return A}};function O1e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r,i=s.shape.length,l=k.parseAxisParam(a,s.shape),u=l,c=_.getAxesPermutation(u,i),d=s;c!=null&&(d=Un({inputs:{x:s},backend:n,attrs:{perm:c}}),u=_.getInnerMostAxes(u.length,s.shape.length)),_.assertAxesAreInnerMostDims("min",u,i);let[h,p]=_.computeOutAndReduceShapes(d.shape,u),f=k.sizeFromShape(p),m=ve({inputs:{x:d},backend:n,attrs:{shape:[-1,f]}}),g=Ai(m,m.dtype,"min",n),y;if(o){let A=_.expandShapeToKeepDim(h,l);y=ve({inputs:{x:g},backend:n,attrs:{shape:A}})}else y=ve({inputs:{x:g},backend:n,attrs:{shape:h}});return n.disposeIntermediateTensorInfo(m),n.disposeIntermediateTensorInfo(g),c!=null&&n.disposeIntermediateTensorInfo(d),y}var P1e={kernelName:xl,backendName:"webgl",kernelFunc:O1e},z1e=cE+`
return min(a, b);
`,L1e=`
vec4 result = vec4(min(a, b));
vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
`+Lm+`
return result;
`,B1e=Tn({opSnippet:z1e,packedOpSnippet:L1e,cpuKernelImpl:Ope}),W1e={kernelName:Fo,backendName:"webgl",kernelFunc:B1e},V1e=class{constructor(e,t,n){this.variableNames=["x"],this.outputShape=t.map((u,c)=>u[0]+e[c]+u[1]);let r=e.length,s=wt(r),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,r),l=n==="reflect"?0:1;if(r===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=`
${s} start = ${s}(${a});
${s} end = ${s}(${o});
void main() {
${s} outC = getOutputCoords();
for (int i = 0; i < ${r}; 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};
}
}
${s} coords = outC - start;
setOutput(getX(${i}));
}
`}},U1e=class{constructor(e,t,n){this.variableNames=["x"],this.packedInputs=!0,this.packedOutput=!0,this.outputShape=t.map((p,f)=>p[0]+e[f]+p[1]);let r=e.length,s=wt(r),a=t.map(p=>p[0]).join(","),o=t.map((p,f)=>p[0]+e[f]).join(","),i=Vn("rc",r),l=Vn("source",r),u=`${i[r-1]} < ${this.outputShape[r-1]}`,c=r===1?"source":`vec2(${l.slice(-2).join()})`,d=n==="reflect"?0:1,h="";if(r===1){let p=`
${s} source = rc;
if (source < start) {
source = start * 2 - source - ${d};
} else if (source >= end) {
source = (end - 1) * 2 - source + ${d};
}
source -= start;
`;h=`
${s} rc = outputLoc;
${p}
result[0] = getChannel(getX(${l.join()}), ${c});
${i[r-1]} += 1;
if(${u}) {
${p}
result[1] = getChannel(getX(${l.join()}), ${c});
}
`}else{let p=`
${s} source = rc;
${s} lt = ${s}(lessThan(source, start));
${s} gte = ${s}(greaterThanEqual(source, end));
${s} orig = 1 - (lt + gte);
source = orig * source +
lt * (start * 2 - source - ${d}) +
gte * ((end - 1) * 2 - source + ${d});
source -= start;
`;h=`
${s} rc = outputLoc;
${p}
result[0] = getChannel(getX(${l.join()}), ${c});
${i[r-1]} += 1;
if(${u}) {
${p}
result[1] = getChannel(getX(${l.join()}), ${c});
}
rc = outputLoc;
${i[r-2]} += 1;
if(${i[r-2]} < ${this.outputShape[r-2]}) {
${p}
result[2] = getChannel(getX(${l.join()}), ${c});
${i[r-1]} += 1;
if(${u}) {
${p}
result[3] = getChannel(getX(${l.join()}), ${c});
}
}
`}this.userCode=`
const ${s} start = ${s}(${a});
const ${s} end = ${s}(${o});
void main() {
${s} outputLoc = getOutputCoords();
vec4 result = vec4(0.);
${h}
setOutput(result);
}
`}},H1e=({inputs:e,backend:t,attrs:n})=>{let{x:r}=e,{paddings:s,mode:a}=n,o=re().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new U1e(r.shape,s,a):new V1e(r.shape,s,a);return t.runWebGLProgram(o,[r],r.dtype)},G1e={kernelName:bl,backendName:"webgl",kernelFunc:H1e},j1e=`if (b == 0.0) return NAN;
return mod(a, b);`,q1e=`
vec4 result = mod(a, b);
vec4 isNaN = vec4(equal(b, vec4(0.0)));
`+Lm+`
return result;
`,K1e=Tn({opSnippet:j1e,packedOpSnippet:q1e}),X1e={kernelName:Uc,backendName:"webgl",kernelFunc:K1e},Z1e=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}));
}
`}},Y1e=`
if (a == b) {
return 1.0;
};
return a / b;`,J1e=`
// 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;
`,qE=Tn({opSnippet:Y1e,packedOpSnippet:J1e,checkOutOfBounds:!0}),Q1e={kernelName:ol,backendName:"webgl",kernelFunc:qE},KE="return a - b;",XE=Tn({opSnippet:KE,packedOpSnippet:KE,supportsComplex:!0,cpuKernelImpl:Zpe}),eye={kernelName:zo,backendName:"webgl",kernelFunc:XE};function ZE(e){let{inputs:t,backend:n,attrs:r}=e,{logits:s}=t,{dim:a}=r,o=k.parseAxisParam([a],s.shape),i=jE({inputs:{x:s},backend:n,attrs:{reductionIndices:o,keepDims:!1}}),l=_.expandShapeToKeepDim(i.shape,o),u=ve({inputs:{x:i},backend:n,attrs:{shape:l}}),c=XE({inputs:{a:s,b:u},backend:n}),d=BE({inputs:{x:c},backend:n}),h=Vm({inputs:{x:d},backend:n,attrs:{axis:o,keepDims:!1}}),p=ve({inputs:{x:h},backend:n,attrs:{shape:l}}),f=qE({inputs:{a:d,b:p},backend:n});return n.disposeIntermediateTensorInfo(i),n.disposeIntermediateTensorInfo(u),n.disposeIntermediateTensorInfo(c),n.disposeIntermediateTensorInfo(d),n.disposeIntermediateTensorInfo(h),n.disposeIntermediateTensorInfo(p),f}var tye={kernelName:Ml,backendName:"webgl",kernelFunc:ZE};function nye(e){let{inputs:t,backend:n,attrs:r}=e,{logits:s}=t,{numSamples:a,seed:o,normalized:i}=r,l=i?s:ZE({inputs:{logits:s},backend:n,attrs:{dim:s.shape.length-1}}),u=l.shape[0],c=l.shape[1],d=new Z1e(u,c,a),h=[[o]],p=n.runWebGLProgram(d,[l],"int32",h);return i||n.disposeIntermediateTensorInfo(l),p}var rye={kernelName:W1,backendName:"webgl",kernelFunc:nye},YE="return -x;";function sye(e){let{inputs:t,backend:n}=e,{x:r}=t;if(n.shouldExecuteOnCPU([r])){let a=n.texData.get(r.dataId),[o,i]=zpe(a.values,r.shape,r.dtype);return n.makeTensorInfo(i,r.dtype,o)}let s;return re().getBool("WEBGL_PACK_UNARY_OPERATIONS")?s=new xu(r.shape,YE):s=new Qa(r.shape,YE),n.runWebGLProgram(s,[r],r.dtype)}var aye={kernelName:Hc,backendName:"webgl",kernelFunc:sye},oye=da.nonMaxSuppressionV3Impl;function iye(e){_.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");let{inputs:t,backend:n,attrs:r}=e,{boxes:s,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l}=r,u=n.readSync(s.dataId),c=n.readSync(a.dataId),{selectedIndices:d}=oye(u,c,o,i,l);return n.makeTensorInfo([d.length],"int32",new Int32Array(d))}var lye={kernelName:Gc,backendName:"webgl",kernelFunc:iye},uye=da.nonMaxSuppressionV4Impl;function cye(e){_.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");let{inputs:t,backend:n,attrs:r}=e,{boxes:s,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,padToMaxOutputSize:u}=r,c=n.readSync(s.dataId),d=n.readSync(a.dataId),{selectedIndices:h,validOutputs:p}=uye(c,d,o,i,l,u);return[n.makeTensorInfo([h.length],"int32",new Int32Array(h)),n.makeTensorInfo([],"int32",new Int32Array([p]))]}var dye={kernelName:jc,backendName:"webgl",kernelFunc:cye},hye=da.nonMaxSuppressionV5Impl;function pye(e){_.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");let{inputs:t,backend:n,attrs:r}=e,{boxes:s,scores:a}=t,{maxOutputSize:o,iouThreshold:i,scoreThreshold:l,softNmsSigma:u}=r,c=n.readSync(s.dataId),d=n.readSync(a.dataId),h=o,p=i,f=l,m=u,{selectedIndices:g,selectedScores:y}=hye(c,d,h,p,f,m);return[n.makeTensorInfo([g.length],"int32",new Int32Array(g)),n.makeTensorInfo([y.length],"float32",new Float32Array(y))]}var fye={kernelName:qc,backendName:"webgl",kernelFunc:pye},mye=class{constructor(e,t,n,r){this.variableNames=["indices"],this.outputShape=[e,t],this.userCode=`
void main() {
ivec2 coords = getOutputCoords();
int index = round(getIndices(coords.x));
setOutput(mix(float(${r}), float(${n}),
float(index == coords.y)));
}
`}},gye=e=>{let{inputs:t,backend:n,attrs:r}=e,{indices:s}=t,{depth:a,onValue:o,offValue:i}=r,l=k.sizeFromShape(s.shape),u=new mye(l,a,o,i),c=ve({inputs:{x:s},backend:n,attrs:{shape:[l]}}),d=n.runWebGLProgram(u,[c],s.dtype);n.disposeIntermediateTensorInfo(c);let h=[...s.shape,a],p=ve({inputs:{x:d},backend:n,attrs:{shape:h}});return n.disposeIntermediateTensorInfo(d),p},yye={kernelName:wl,backendName:"webgl",kernelFunc:gye};function Km(e){let{inputs:t,backend:n}=e,{x:r}=t;if(r.dtype==="complex64"){let s=hh({inputs:{input:r},backend:n}),a=Km({inputs:{x:s},backend:n}),o=jm({inputs:{input:r},backend:n}),i=Km({inputs:{x:o},backend:n}),l=eo({inputs:{real:a,imag:i},backend:n});return n.disposeIntermediateTensorInfo(s),n.disposeIntermediateTensorInfo(a),n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}else return qm({attrs:{shape:r.shape,dtype:r.dtype,value:r.dtype==="string"?"":0},backend:n})}var Aye={kernelName:hd,backendName:"webgl",kernelFunc:Km};function JE(e){let{inputs:t,backend:n}=e,{x:r}=t;if(r.dtype==="string")throw new Error("onesLike is not supported under string dtype");if(r.dtype==="complex64"){let s=hh({inputs:{input:r},backend:n}),a=JE({inputs:{x:s},backend:n}),o=jm({inputs:{input:r},backend:n}),i=Km({inputs:{x:o},backend:n}),l=eo({inputs:{real:a,imag:i},backend:n});return n.disposeIntermediateTensorInfo(s),n.disposeIntermediateTensorInfo(a),n.disposeIntermediateTensorInfo(o),n.disposeIntermediateTensorInfo(i),l}else return qm({attrs:{shape:r.shape,dtype:r.dtype,value:1},backend:n})}var xye={kernelName:Kc,backendName:"webgl",kernelFunc:JE};function bye(e){let{inputs:t,backend:n,attrs:r}=e,{axis:s}=r;if(t.length===1)return tb({inputs:{input:t[0]},backend:n,attrs:{dim:s}});let a=t[0].shape,o=t[0].dtype;t.forEach(c=>{k.assertShapesMatch(a,c.shape,"All tensors passed to stack must have matching shapes"),k.assert(o===c.dtype,()=>"All tensors passed to stack must have matching dtypes")});let i=[],l=t.map(c=>{let d=tb({inputs:{input:c},backend:n,attrs:{dim:s}});return i.push(d),d}),u=$E({inputs:l,backend:n,attrs:{axis:s}});return i.forEach(c=>n.disposeIntermediateTensorInfo(c)),u}var vye={kernelName:Xc,backendName:"webgl",kernelFunc:bye},wye=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 r=e.length,s=wt(r),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,r);if(r===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=`
${s} start = ${s}(${a});
${s} end = ${s}(${o});
void main() {
${s} outC = getOutputCoords();
if (any(lessThan(outC, start)) || any(greaterThanEqual(outC, end))) {
setOutput(value);
} else {
${s} coords = outC - start;
setOutput(getX(${i}));
}
}
`}},kye=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 r=e.length,s=wt(r),a=t.map(f=>f[0]).join(","),o=t.map((f,m)=>f[0]+e[m]).join(","),i=Vn("rc",r),l=Vn("source",r),u=`${i[r-1]} < ${this.outputShape[r-1]}`,c=r===1?"source":`vec2(${l.slice(-2).join()})`,d=[`${s} rc = outputLoc;`,`${i[r-1]} += 1;
if(${u}) {
`,r===1?"":`}
rc = outputLoc;
${i[r-2]} += 1;
if(${i[r-2]} < ${this.outputShape[r-2]}) {`,r===1?"":` ${i[r-1]} += 1;
if(${u}) {`],h=r===1?"rc < start || rc >= end":"any(lessThan(rc, start)) || any(greaterThanEqual(rc, end))",p="";for(let f=0,m=r===1?2:4;f<m;f++)p+=`
${d[f]}
if (${h}) {
result[${f}] = float(value);
} else {
${s} source = rc - start;
result[${f}] = getChannel(getX(${l.join()}), ${c});
}
`;p+=r===1?"} ":"}}",this.userCode=`
const ${s} start = ${s}(${a});
const ${s} end = ${s}(${o});
void main() {
${s} outputLoc = getOutputCoords();
vec4 result = vec4(0.);
${p}
setOutput(result);
}
`}},QE=e=>{let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{paddings:a,constantValue:o}=r,i=re().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new kye(s.shape,a,o):new wye(s.shape,a,o),l=[[o]];return n.runWebGLProgram(i,[s],s.dtype,l)},Iye={kernelName:kl,backendName:"webgl",kernelFunc:QE},Sye=`
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);
`,Tye=`
// 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));
`+Lm+`
return result;
`,Nye=Tn({opSnippet:Sye,packedOpSnippet:Tye}),Cye={kernelName:Il,backendName:"webgl",kernelFunc:Nye};function Eye(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,keepDims:o}=r,i=s.shape.length,l=[],u=k.parseAxisParam(a,s.shape),c=u,d=_.getAxesPermutation(c,i),h=s;d!=null&&(h=Un({inputs:{x:s},backend:n,attrs:{perm:d}}),c=_.getInnerMostAxes(c.length,i),l.push(h)),_.assertAxesAreInnerMostDims("prod",c,i);let p;if(n.shouldExecuteOnCPU([h])){let f=n.texData.get(h.dataId).values,{outVals:m,outShape:g,outDtype:y}=Bpe(h.shape,h.dtype,f,c);p=n.makeTensorInfo(g,y,m)}else{let[f,m]=_.computeOutAndReduceShapes(h.shape,c),g=k.sizeFromShape(m),y=ve({inputs:{x:h},backend:n,attrs:{shape:[-1,g]}}),A=cy(s.dtype),x=Ai(y,A,"prod",n);p=ve({inputs:{x},backend:n,attrs:{shape:f}}),l.push(y),l.push(x)}if(o){l.push(p);let f=_.expandShapeToKeepDim(p.shape,u);p=ve({inputs:{x:p},backend:n,attrs:{shape:f}})}return l.forEach(f=>n.disposeIntermediateTensorInfo(f)),p}var $ye={kernelName:Zc,backendName:"webgl",kernelFunc:Eye},e9=e=>{let{backend:t,attrs:n}=e,{start:r,stop:s,step:a,dtype:o}=n,i=Wpe(r,s,a,o);return t.makeTensorInfo([i.length],o,i)},Rye={kernelName:rf,backendName:"webgl",kernelFunc:e9},_ye="return 1.0 / x;",Dye=it({opSnippet:_ye}),Fye={kernelName:Yc,backendName:"webgl",kernelFunc:Dye},Mye=gs+`
return (x < 0.0) ? 0.0 : x;
`,Oye=`
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;
`,Pye=it({opSnippet:Mye,packedOpSnippet:Oye}),zye={kernelName:Tl,backendName:"webgl",kernelFunc:Pye},Lye=gs+`
return (x < 0.0) ? 0.0 : min(6.0, x);
`,Bye=`
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;
`,Wye=it({opSnippet:Lye,packedOpSnippet:Bye}),Vye={kernelName:Cl,backendName:"webgl",kernelFunc:Wye},Uye=class{constructor(e,t,n,r,s){this.variableNames=["A"],this.outputShape=[];let[a,o,i,l]=e;this.outputShape=[a,t,n,l];let u=[r&&t>1?o-1:o,r&&n>1?i-1:i],c=[r&&t>1?t-1:t,r&&n>1?n-1:n],d;s?d="(vec2(yRC) + vec2(0.5)) * effectiveInputOverOutputRatioRC - vec2(0.5)":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 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);
}
`}},Hye=class{constructor(e,t,n,r,s){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=[r&&t>1?o-1:o,r&&n>1?i-1:i],c=[r&&t>1?t-1:t,r&&n>1?n-1:n],d;s?d="(vec3(yRC) + vec3(0.5)) * effectiveInputOverOutputRatioRC - vec3(0.5)":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 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 Gye(e){let{inputs:t,backend:n,attrs:r}=e,{images:s}=t,{alignCorners:a,halfPixelCenters:o,size:i}=r,[l,u]=i,c=re().getBool("WEBGL_PACK_IMAGE_OPERATIONS")?new Hye(s.shape,l,u,a,o):new Uye(s.shape,l,u,a,o);return n.runWebGLProgram(c,[s],"float32")}var jye={kernelName:Nl,backendName:"webgl",kernelFunc:Gye},qye=class{constructor(e,t,n){this.variableNames=["dy"],this.outputShape=[],this.outputShape=t;let[,r,s]=t,[,a,o]=e,i=[n&&a>1?r-1:r,n&&o>1?s-1:s],l=[n&&a>1?a-1:a,n&&o>1?o-1:o],u=i[0]/l[0],c=i[1]/l[1],d=1/u,h=1/c,p=Math.ceil(d)*2+2,f=Math.ceil(h)*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(${c});
const float invHeightScale = float(${d});
const float invWidthScale = float(${h});
const int winHeight = int(${p});
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), ${r-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), ${s-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 Kye(e){let{inputs:t,backend:n,attrs:r}=e,{images:s,dy:a}=t,{alignCorners:o}=r,i=new qye(a.shape,s.shape,o);return n.runWebGLProgram(i,[a],a.dtype)}var Xye={kernelName:H1,backendName:"webgl",kernelFunc:Kye},Zye=class{constructor(e,t,n,r,s){this.variableNames=["A"],this.outputShape=[];let[a,o,i,l]=e;this.outputShape=[a,t,n,l];let u=[r&&t>1?o-1:o,r&&n>1?i-1:i],c=[r&&t>1?t-1:t,r&&n>1?n-1:n],d=r?"0.5":"0.0",h;s?h="max((vec2(yRC) + vec2(0.5)) * effectiveInputOverOutputRatioRC, vec2(0.0))":h="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 = ${h};
// Compute the coordinators of nearest neighbor point.
ivec2 sourceNearestRC = ivec2(
min(inputShapeRC - 1.0, floor(sourceFracIndexRC + ${d})));
float newValue = getA(b, sourceNearestRC.x, sourceNearestRC.y, d);
setOutput(newValue);
}
`}},Yye=class{constructor(e,t,n,r,s){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=[r&&t>1?o-1:o,r&&n>1?i-1:i],c=[r&&t>1?t-1:t,r&&n>1?n-1:n],d=r?"0.5":"0.0",h;s?h="max((vec3(yRC) + vec3(0.5)) * effectiveInputOverOutputRatioRC, vec3(0.0))":h="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 = ${h};
// Compute the coordinators of nearest neighbor point.
ivec3 sourceNearestRC = ivec3(
min(inputShapeRC - 1.0, floor(sourceFracIndexRC + ${d})));
// Should we calculate next column and row elements in 2x2 packed cell.
bool hasNextCol = d < ${l-1};
bool hasNextRow = coords.z < ${n-1};
vec4 newValue = vec4(
getAValue(b, sourceNearestRC.x, sourceNearestRC.y, d),
hasNextCol ? getAValue(b, sourceNearestRC.x, sourceNearestRC.y, d + 1)
: 0.0,
hasNextRow ? getAValue(b, sourceNearestRC.x, sourceNearestRC.z, d)
: 0.0,
(hasNextRow && hasNextCol) ?
getAValue(b, sourceNearestRC.x, sourceNearestRC.z, d + 1) : 0.0);
setOutput(newValue);
}
`}};function Jye(e){let{inputs:t,backend:n,attrs:r}=e,{images:s}=t,{alignCorners:a,halfPixelCenters:o,size:i}=r,[l,u]=i,c=re().getBool("WEBGL_PACK_IMAGE_OPERATIONS")?new Yye(s.shape,l,u,a,o):new Zye(s.shape,l,u,a,o);return n.runWebGLProgram(c,[s],s.dtype)}var Qye={kernelName:sf,backendName:"webgl",kernelFunc:Jye},eAe=class{constructor(e,t,n){this.variableNames=["dy"],this.outputShape=[],this.outputShape=t;let[,r,s]=t,[,a,o]=e,i=[n&&a>1?r-1:r,n&&o>1?s-1:s],l=[n&&a>1?a-1:a,n&&o>1?o-1:o],u=i[0]/l[0],c=i[1]/l[1],d=1/u,h=1/c,p=Math.ceil(d)*2+2,f=Math.ceil(h)*2+2;this.userCode=`
void main() {
ivec4 coords = getOutputCoords();
int b = coords[0];
int d = coords[3];
int r = coords[1];
int c = coords[2];
float accumulator = 0.0;
const float heightScale = float(${u});
const float widthScale = float(${c});
const float invHeightScale = float(${d});
const float invWidthScale = float(${h});
const int winHeight = int(${p});
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(${l[1]}));
int sourceNearestRow = int(min(
float(int(${r}) - 1),
${n} ? float(round(sourceFracRow)) :
float(floor(sourceFracRow))));
int sourceNearestCol = int(min(
float(int(${s}) - 1),
${n} ? float(round(sourceFracCol)) :
float(floor(sourceFracCol))));
if (r == sourceNearestRow && c == sourceNearestCol) {
accumulator += getDy(b, dyR, dyC, d);
}
}
}
// End loop over dy
setOutput(accumulator);
}
`}};function tAe(e){let{inputs:t,backend:n,attrs:r}=e,{images:s,dy:a}=t,{alignCorners:o}=r,i=new eAe(a.shape,s.shape,o);return n.runWebGLProgram(i,[a],a.dtype)}var nAe={kernelName:U1,backendName:"webgl",kernelFunc:tAe},rAe=class{constructor(e,t){this.variableNames=["x"];let n=e.length;if(n>4)throw new Error(`WebGL backend: Reverse of rank-${n} tensor is not yet supported`);if(this.outputShape=e,n===1){this.userCode=`
void main() {
int coord = getOutputCoords();
setOutput(getX(${e[0]} - coord - 1));
}
`;return}let r=o=>t.indexOf(o)!==-1&&e[o]!==1?`${e[o]} - coords[${o}] - 1`:`coords[${o}]`,s=e.map((o,i)=>r(i)).join(","),a=wt(n);this.userCode=`
void main() {
${a} coords = getOutputCoords();
setOutput(getX(${s}));
}
`}},sAe=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-${n} tensor is not yet supported`);this.outputShape=e;let r=Vn("rc",n),s=`${r[n-1]} + 1 < ${this.outputShape[n-1]}`,a=`${r[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),
${e[0]} - rc - 1);
if(${s}){
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 = ${i(r.slice())};
if(${s}){
result.g = ${l(r.slice())};
}
if(${a}) {
result.b = ${u(r.slice())};
if(${s}) {
result.a = ${c(r.slice())};
}
}
setOutput(result);
}
`;function i(p){return d(p)}function l(p){return p[n-1]="("+p[n-1]+" + 1)",d(p)}function u(p){return p[n-2]="("+p[n-2]+" + 1)",d(p)}function c(p){return p[n-1]="("+p[n-1]+" + 1)",p[n-2]="("+p[n-2]+" + 1)",d(p)}function d(p){let f=e.map((y,A)=>h(A,p)),m=f.join(","),g=f.slice(-2).join(",");return`getChannel(getX(${m}), vec2(${g}))`}function h(p,f){return t.indexOf(p)!==-1&&e[p]!==1?`${e[p]} - ${f[p]} - 1`:`${f[p]}`}}};function aAe(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{dims:a}=r,o=s.shape.length,i=k.parseAxisParam(a,s.shape);if(o===0)return fr({inputs:{x:s},backend:n});let l=re().getBool("WEBGL_PACK_ARRAY_OPERATIONS")?new sAe(s.shape,i):new rAe(s.shape,i);return n.runWebGLProgram(l,[s],s.dtype)}var oAe={kernelName:El,backendName:"webgl",kernelFunc:aAe},iAe=class{constructor(e,t){this.variableNames=["Image"],this.outputShape=[],this.customUniforms=[{name:"params",type:"vec4"}];let n=e[1],r=e[2];this.outputShape=e;let s="";typeof t=="number"?s=`float outputValue = ${t.toFixed(2)};`:s=`
vec3 fill = vec3(${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]));
${s}
if(coordX >= 0 && coordX < ${r} && coordY >= 0 && coordY < ${n}) {
outputValue = getImage(coords[0], coordY, coordX, coords[3]);
}
setOutput(outputValue);
}
`}},lAe={kernelName:pd,backendName:"webgl",kernelFunc:({inputs:e,attrs:t,backend:n})=>{let{image:r}=e,{radians:s,fillValue:a,center:o}=t,i=n,l=new iAe(r.shape,a),[u,c]=_.getImageCenter(o,r.shape[1],r.shape[2]),d=[[u,c,Math.sin(s),Math.cos(s)]];return i.runWebGLProgram(l,[r],r.dtype,d)}},uAe=`
// OpenGL ES does not support round function.
// The algorithm is based on banker's rounding.
float base = floor(x);
if ((x - base) < 0.5) {
return floor(x);
} else if ((x - base) > 0.5) {
return ceil(x);
} else {
if (mod(base, 2.0) == 0.0) {
return base;
} else {
return base + 1.0;
}
}
`,cAe=it({opSnippet:uAe}),dAe={kernelName:$l,backendName:"webgl",kernelFunc:cAe},hAe="return inversesqrt(x);",pAe=it({opSnippet:hAe,cpuKernelImpl:Vpe}),fAe={kernelName:Oo,backendName:"webgl",kernelFunc:pAe},t9=class{constructor(e,t,n,r,s,a,o=!0){this.variableNames=["updates","indices","defaultValue"],this.outputShape=a;let i=wt(s.length),l=wt(a.length),u="";n===1?u="i":n===2&&(u="i, j");let c=`getIndices(${u})`,d="";r===1?d="i":r===2&&(d="i, coords[1]");let h=`getUpdates(${d})`,p=t>1?"strides[j]":"strides";this.userCode=`
${i} strides = ${i}(${s});
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 * ${p};
}
if (flattenedIndex == coords[0]) {
sum += ${h};
found = true;
}
}
setOutput(mix(getDefaultValue(), sum, float(found)));
}
`}};function mAe(e){let{inputs:t,backend:n,attrs:r}=e,{indices:s,updates:a}=t,{shape:o}=r,{sliceRank:i,numUpdates:l,sliceSize:u,strides:c,outputSize:d}=_.calculateShapes(a,s,o),h=[d/u,u];if(d===0)return n.makeTensorInfo(o,s.dtype);let p=ve({inputs:{x:s},backend:n,attrs:{shape:[l,i]}}),f=ve({inputs:{x:a},backend:n,attrs:{shape:[l,u]}}),m=n.makeTensorInfo([],"float32",new Float32Array([0])),g=new t9(l,i,p.shape.length,f.shape.length,c,h),y=n.runWebGLProgram(g,[f,p,m],f.dtype),A=ve({inputs:{x:y},backend:n,attrs:{shape:o}});return n.disposeIntermediateTensorInfo(p),n.disposeIntermediateTensorInfo(f),n.disposeIntermediateTensorInfo(y),n.disposeIntermediateTensorInfo(m),A}var gAe={kernelName:Qc,backendName:"webgl",kernelFunc:mAe},yAe=class{constructor(e,t,n){this.variableNames=["c","a","b"],this.outputShape=t;let r,s;if(n>4)throw Error(`Where for rank ${n} is not yet supported`);if(n===1)s="resRC",r="resRC";else{let o=["resRC.x","resRC.y","resRC.z","resRC.w"],i=[],l=[];for(let u=0;u<t.length;u++)l.push(`${o[u]}`),u<e&&i.push(`${o[u]}`);r=i.join(),s=l.join()}let a=wt(n);this.userCode=`
void main() {
${a} resRC = getOutputCoords();
float cVal = getC(${r});
if (cVal >= 1.0) {
setOutput(getA(${s}));
} else {
setOutput(getB(${s}));
}
}
`}};function AAe(e){let{inputs:t,backend:n}=e,{condition:r,t:s,e:a}=t,o=new yAe(r.shape.length,s.shape,s.shape.length);return n.runWebGLProgram(o,[r,s,a],Ur(s.dtype,a.dtype))}var xAe={kernelName:ed,backendName:"webgl",kernelFunc:AAe},bAe=`
// Stable and Attracting Fixed Point (0, 1) for Normalized Weights.
// see: https://arxiv.org/abs/1706.02515
float scaleAlpha = ${_.SELU_SCALEALPHA};
float scale = ${_.SELU_SCALE};
return (x >= 0.0) ? scale * x : scaleAlpha * (exp(x) - 1.0);
`,vAe=it({opSnippet:bAe}),wAe={kernelName:td,backendName:"webgl",kernelFunc:vAe},kAe="return 1.0 / (1.0 + exp(-1.0 * x));",IAe=it({opSnippet:kAe}),SAe={kernelName:_l,backendName:"webgl",kernelFunc:IAe},TAe=`
if (isnan(x)) { return 0.0; }
return sign(x);
`,NAe=it({opSnippet:TAe}),CAe={kernelName:sd,backendName:"webgl",kernelFunc:NAe},EAe=mE+`
return sin(x);
`,$Ae=it({opSnippet:EAe}),RAe={kernelName:Rl,backendName:"webgl",kernelFunc:$Ae},_Ae=`
float e2x = exp(x);
return (e2x - 1.0 / e2x) / 2.0;
`,DAe=it({opSnippet:_Ae}),FAe={kernelName:rd,backendName:"webgl",kernelFunc:DAe},MAe=`
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;
`,OAe=it({opSnippet:MAe}),PAe={kernelName:ad,backendName:"webgl",kernelFunc:OAe},zAe=e=>{let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{blockShape:a,paddings:o}=r;k.assert(s.shape.length<=4,()=>"spaceToBatchND for rank > 4 with a WebGL backend not implemented yet");let i=a.reduce((y,A)=>y*A),l=[[0,0]];l.push(...o);for(let y=1+a.length;y<s.shape.length;++y)l.push([0,0]);let u=[],c=QE({inputs:{x:s},backend:n,attrs:{paddings:l,constantValue:0}}),d=_.getReshaped(c.shape,a,i,!1),h=_.getPermuted(d.length,a.length,!1),p=_.getReshapedPermuted(c.shape,a,i,!1),f=ve({inputs:{x:c},backend:n,attrs:{shape:d}}),m=Un({inputs:{x:f},backend:n,attrs:{perm:h}}),g=ve({inputs:{x:m},backend:n,attrs:{shape:p}});return u.push(c),u.push(f),u.push(m),u.forEach(y=>n.disposeIntermediateTensorInfo(y)),g},LAe={kernelName:od,backendName:"webgl",kernelFunc:zAe};function BAe(e){let{inputs:t,backend:n}=e,{indices:r,values:s,denseShape:a,defaultValue:o}=t;if(a.shape.length!==1)throw new Error(`Dense shape must be a vector, saw:
${a.shape}`);if(r.shape.length!==2)throw new Error(`Indices must be a matrix, saw:
${r.shape}`);if(s.shape.length!==1)throw new Error(`Values must be a vector, saw:
${s.shape}`);if(o.shape.length!==0)throw new Error(`Default value must be a scalar, saw:
${o.shape}`);let i=n.readSync(r.dataId),l=n.readSync(s.dataId),u=n.readSync(a.dataId),c=n.readSync(o.dataId)[0],[d,h,p,f,m]=Hpe(i,r.shape,r.dtype,l,s.dtype,u,c);return[n.makeTensorInfo(h,r.dtype,d),n.makeTensorInfo([h[0]],s.dtype,p),n.makeTensorInfo([f.length],"bool",new Uint8Array(f.map(g=>Number(g)))),n.makeTensorInfo([m.length],r.dtype,new Int32Array(m))]}var WAe={kernelName:G1,backendName:"webgl",kernelFunc:BAe};function VAe(e){let{inputs:t,backend:n}=e,{inputIndices:r,inputShape:s,newShape:a}=t;if(r.shape.length!==2)throw new Error(`Input indices should be a matrix but received shape ${r.shape}`);if(s.shape.length!==1)throw new Error(`Input shape should be a vector but received shape ${s.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(s.dataId)),i=n.readSync(r.dataId),l=Array.from(n.readSync(a.dataId)),[u,c,d]=Gpe(i,r.shape,r.dtype,o,l);return[n.makeTensorInfo(c,r.dtype,u),n.makeTensorInfo([d.length],a.dtype,new Int32Array(d))]}var UAe={kernelName:j1,backendName:"webgl",kernelFunc:VAe};function HAe(e){let{inputs:t,backend:n}=e,{data:r,indices:s,segmentIds:a}=t;if(r.shape.length<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(s.shape.length!==1)throw new Error(`Indices should be a vector but received shape
${s.shape}`);if(a.shape.length!==1)throw new Error(`Segment ids should be a vector but received shape
${a.shape}`);let o=n.readSync(r.dataId),i=n.readSync(s.dataId),l=n.readSync(a.dataId),[u,c]=tE(o,r.shape,r.dtype,i,l,!0);return n.makeTensorInfo(c,r.dtype,u)}var GAe={kernelName:q1,backendName:"webgl",kernelFunc:HAe};function jAe(e){let{inputs:t,backend:n}=e,{data:r,indices:s,segmentIds:a}=t;if(r.shape.length<1)throw new Error("Data should be at least 1 dimensional but received scalar");if(s.shape.length!==1)throw new Error(`Indices should be a vector but received shape
${s.shape}`);if(a.shape.length!==1)throw new Error(`Segment ids should be a vector but received shape
${a.shape}`);let o=n.readSync(r.dataId),i=n.readSync(s.dataId),l=n.readSync(a.dataId),[u,c]=tE(o,r.shape,r.dtype,i,l);return n.makeTensorInfo(c,r.dtype,u)}var qAe={kernelName:K1,backendName:"webgl",kernelFunc:jAe};function KAe(e){let{inputs:t,backend:n,attrs:r}=e,{sparseIndices:s,sparseValues:a,defaultValue:o}=t,{outputShape:i}=r,{sliceRank:l,numUpdates:u,strides:c,outputSize:d}=_.calculateShapes(a,s,i),h=!1,p=new t9(u,l,s.shape.length,a.shape.length,c,[d,1],h),f=n.runWebGLProgram(p,[a,s,o],a.dtype),m=ve({inputs:{x:f},backend:n,attrs:{shape:i}});return n.disposeIntermediateTensorInfo(f),m}var XAe={kernelName:X1,backendName:"webgl",kernelFunc:KAe};function ZAe(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{numOrSizeSplits:a,axis:o}=r,i=k.parseAxisParam(o,s.shape)[0],l=_.prepareSplitSize(s,a,i),u=s.shape.length,c=new Array(u).fill(0),d=s.shape.slice();return l.map(h=>{let p=[...d];p[i]=h;let f=vu({inputs:{x:s},backend:n,attrs:{begin:c,size:p}});return c[i]+=h,f})}var YAe={kernelName:id,backendName:"webgl",kernelFunc:ZAe},JAe="return sqrt(x);",QAe=it({opSnippet:JAe}),exe={kernelName:Dl,backendName:"webgl",kernelFunc:QAe},txe="return x * x;",nxe=it({opSnippet:txe}),rxe={kernelName:af,backendName:"webgl",kernelFunc:nxe},n9="return (a - b) * (a - b);",sxe=Tn({opSnippet:n9,packedOpSnippet:n9}),axe={kernelName:Po,backendName:"webgl",kernelFunc:sxe};function oxe({inputs:e,attrs:t,backend:n}){let{x:r}=e,s=gs+`
return x > 0.0 ? 1.0 : float(${t.alpha});
`,a=new Qa(r.shape,s);return n.runWebGLProgram(a,[r],r.dtype)}var ixe={kernelName:Bo,backendName:"webgl",kernelFunc:oxe},lxe=class{constructor(e,t,n){this.variableNames=["x"],this.outputShape=n;let r=n.length,s=wt(n.length),a=wt(n.length),o="";if(r===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=`
${s} begin = ${s}(${e});
${s} strides = ${s}(${t});
void main() {
${a} coords = getOutputCoords();
setOutput(getX(${o}));
}
`}};function uxe(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{begin:a,end:o,strides:i,beginMask:l,endMask:u,ellipsisMask:c,newAxisMask:d,shrinkAxisMask:h}=r,{nonStrided:p,$begin:f,$strides:m,size:g,newShape:y,outShape:A}=Cn.sliceInfo(s.shape,a,o,i,l,u,c,d,h),x=ve({inputs:{x:s},backend:n,attrs:{shape:y}}),b;if(p){let I=vu({inputs:{x},backend:n,attrs:{begin:f,size:g}});b=ve({inputs:{x:I},backend:n,attrs:{shape:A}}),n.disposeIntermediateTensorInfo(I)}else if(A.some(I=>I===0))b=n.makeTensorInfo(A,s.dtype,[]);else if(n.shouldExecuteOnCPU([x])){let S=n.texData.get(x.dataId).values,E=ze(x.shape,x.dtype,S),D=jpe(A,E,m,f);b=n.makeTensorInfo(A,x.dtype,D.values)}else{let w=new lxe(f,m,A);b=n.runWebGLProgram(w,[x],x.dtype)}let v=ve({inputs:{x:b},backend:n,attrs:{shape:A}});return n.disposeIntermediateTensorInfo(x),n.disposeIntermediateTensorInfo(b),v}var cxe={kernelName:ld,backendName:"webgl",kernelFunc:uxe};function dxe(e){let{inputs:t,backend:n,attrs:r}=e,{separator:s,nGramWidths:a,leftPad:o,rightPad:i,padWidth:l,preserveShortSequences:u}=r,{data:c,dataSplits:d}=t,h=n.readSync(c.dataId),p=n.readSync(d.dataId),[f,m]=qpe(h,p,s,a,o,i,l,u);return[n.makeTensorInfo([f.length],"string",f),n.makeTensorInfo(d.shape,"int32",m)]}var hxe={kernelName:Z1,backendName:"webgl",kernelFunc:dxe};function pxe(e){let{inputs:t,backend:n,attrs:r}=e,{skipEmpty:s}=r,{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,d]=Kpe(i,l,s),h=c.length;return[n.makeTensorInfo([h,2],"int32",u),n.makeTensorInfo([h],"string",c),n.makeTensorInfo([2],"int32",new Int32Array(d))]}var fxe={kernelName:Y1,backendName:"webgl",kernelFunc:pxe};function mxe(e){let{inputs:t,backend:n,attrs:r}=e,{numBuckets:s}=r,{input:a}=t;if(a.dtype!=="string")throw new Error("Input must be of datatype string");if(s<=0)throw new Error("Number of buckets must be at least 1");let o=n.readSync(a.dataId),i=Xpe(o,s);return n.makeTensorInfo(a.shape,"int32",i)}var gxe={kernelName:J1,backendName:"webgl",kernelFunc:mxe},yxe="return tan(x);",Axe=it({opSnippet:yxe}),xxe={kernelName:Ol,backendName:"webgl",kernelFunc:Axe},bxe=`
float e2x = exp(-2.0 * abs(x));
return sign(x) * (1.0 - e2x) / (1.0 + e2x);
`,vxe=it({opSnippet:bxe}),wxe={kernelName:Pl,backendName:"webgl",kernelFunc:vxe},kxe=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 r=wt(this.rank),s=Ixe(e);this.userCode=`
void main() {
${r} resRC = getOutputCoords();
setOutput(getA(${s}));
}
`}};function Ixe(e){let t=e.length;if(t>5)throw Error(`Tile for rank ${t} is not yet supported`);if(t===1)return`imod(resRC, ${e[0]})`;let n=["resRC.x","resRC.y","resRC.z","resRC.w","resRC.u"],r=[];for(let s=0;s<e.length;s++)r.push(`imod(${n[s]}, ${e[s]})`);return r.join()}function r9(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{reps:a}=r;if(s.dtype==="string"||s.shape.length>5){let l=n.readSync(s.dataId),u=s.dtype==="string"?l.map(h=>k.decodeString(h)):l,c=ze(s.shape,s.dtype,u),d=Ype(c,a);return n.makeTensorInfo(d.shape,d.dtype,d.values)}let o=new kxe(s.shape,a);return n.runWebGLProgram(o,[s],s.dtype)}var Sxe={kernelName:Lo,backendName:"webgl",kernelFunc:r9},Txe=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));
}
}
`}},Nxe=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 xi(e,t){t!==null&&e.disposeIntermediateTensorInfo(t)}function s9(e){let t=1;for(;t<e;)t*=2;return t}function Cxe(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{k:a,sorted:o}=r,i=re().getNumber("TOPK_LAST_DIM_CPU_HANDOFF_SIZE_THRESHOLD"),l=re().getNumber("TOPK_K_CPU_HANDOFF_THRESHOLD"),u=s.shape,c=u[u.length-1];if(n.shouldExecuteOnCPU([s])||c<i||a>l){let D=n.readSync(s.dataId),[$,R]=Jpe(D,u,s.dtype,a,o);return[n.makeTensorInfo($.shape,$.dtype,$.values),n.makeTensorInfo(R.shape,R.dtype,R.values)]}if(a===0)return u[u.length-1]=0,[n.makeTensorInfo(u,s.dtype,[]),n.makeTensorInfo(u,"int32",[])];if(c===1)return[s,qm({attrs:{shape:u,dtype:"int32",value:0},backend:n})];let d=n.texData.get(s.dataId),h=d!==null&&d.isPacked,p=h?n.unpackTensor(s):s,m=k.sizeFromShape(u)/c,g=ve({inputs:{x:p},attrs:{shape:[m,c]},backend:n});h&&xi(n,p);let y=s9(a),A=s9(c),x=null,b=()=>x===null?[g,g]:[g,x],v=(D,$,R)=>{let N=b(),M=new Txe(R),q=[[c],[x===null?1:0],[Number.NEGATIVE_INFINITY],[D],[$]],X=x;x=n.runWebGLProgram(M,N,"int32",q),xi(n,X)};for(let D=1;D<y;D*=2){let $=D*2;for(let R=D;R>=1;R/=2)v($,R,[m,A])}for(let D=A;D>y;D/=2){let $=b(),R=new Nxe([m,D/2]),M=[[c],[x===null?1:0],[y]],B=x;x=n.runWebGLProgram(R,$,"int32",M),xi(n,B);let q=y/2,X=q*2;for(let J=q;J>=1;J/=2)v(X,J,x.shape)}let I=x;x=vu({inputs:{x},backend:n,attrs:{begin:0,size:[m,a]}}),xi(n,I);let w=GE({inputs:{x:g,indices:x},backend:n,attrs:{axis:1,batchDims:1}});xi(n,g);let S=u.slice(0,-1);S.push(a),I=x,x=ve({inputs:{x},attrs:{shape:S},backend:n}),xi(n,I);let E=w;return w=ve({inputs:{x:w},attrs:{shape:S},backend:n}),xi(n,E),[w,x]}var Exe={kernelName:ud,backendName:"webgl",kernelFunc:Cxe},$xe=class{constructor(e,t,n,r,s,a){this.variableNames=["Image","Transforms"],this.outputShape=a;let o=n==="nearest"?1:2,i;switch(r){case"constant":i=1;break;case"reflect":i=2;break;case"wrap":i=3;break;case"nearest":i=4;break;default:i=1;break}this.userCode=`
float mapCoord(float outCoord, float len) {
float inCoord = outCoord;
if(${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 < ${t}) {
outputValue = getImage(batch, coordY, coordX, channel);
} else {
outputValue = float(${s});
}
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(${s});
} else {
float inX = (a1 * xf + a2 * yf + a3) / projection;
float inY = (b1 * xf + b2 * yf + b3) / projection;
float mapX = mapCoord(inX, float(${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 Rxe(e){let{inputs:t,backend:n,attrs:r}=e,{image:s,transforms:a}=t,{interpolation:o,fillMode:i,fillValue:l,outputShape:u}=r,[c,d,h,p]=s.shape,[f,m]=u!=null?u:[d,h],g=[c,f,m,p],y=new $xe(d,h,o,i,l,g);return n.runWebGLProgram(y,[s,a],"float32")}var _xe={kernelName:cd,backendName:"webgl",kernelFunc:Rxe};function Dxe(e){let{inputs:t,attrs:n,backend:r}=e,{axis:s}=n,{x:a}=t;hu(a,"unique"),console.warn("WARNING: ","UI might be locked temporarily as data is being downloaded");let o=r.readSync(a.dataId),{outputValues:i,outputShape:l,indices:u}=Qpe(o,s,a.shape,a.dtype);return[r.makeTensorInfo(l,a.dtype,i),r.makeTensorInfo([u.length],"int32",u)]}var Fxe={kernelName:Q1,backendName:"webgl",kernelFunc:Dxe};function Mxe(e){let{inputs:t,backend:n,attrs:r}=e,{value:s}=t,{axis:a}=r;a<0&&(a+=s.shape.length);let o=s,i=o.shape.length,l=s.shape[a],u=new Array(i-1),c=0;for(let m=0;m<i;m++)m!==a&&(u[c++]=o.shape[m]);let d=[],h=new Array(i).fill(0),p=o.shape.slice();p[a]=1;let f=new Array(l);for(let m=0;m<f.length;m++){h[a]=m;let g=vu({inputs:{x:o},backend:n,attrs:{begin:h,size:p}}),y=ve({inputs:{x:g},backend:n,attrs:{shape:u}});f[m]=y,d.push(g)}return d.forEach(m=>n.disposeIntermediateTensorInfo(m)),f}var Oxe={kernelName:dd,backendName:"webgl",kernelFunc:Mxe},Pxe=class{constructor(e,t){this.variableNames=["x","segmentIds"];let n=e.windowSize,r=e.batchSize,s=e.inSize,a=e.numSegments,o=a*Math.ceil(s/n);this.outputShape=[r,o];let i="0.0",l="sumValue",u=Math.floor(n/4)*4,c=n%4,d=`
sumValue += dot(values, segFilter);
`,h="";s%n>0&&(h=`
if (inIdx < 0 || inIdx >= ${s}) {
return initializationValue;
}
`);let p="";s%n>0&&(p=`
if (inIdx < 0 || inIdx >= ${s}) {
return -1.0;
}
`),this.userCode=`
const float initializationValue = ${i};
float getValue(int batch, int inIdx) {
${h}
return getX(batch, inIdx);
}
float getSegmentIdAtIndex(int inIdx) {
${p}
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
);
${d}
}
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
);
${d}
} else if (${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
);
${d}
} else if (${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
);
${d}
}
setOutput(${l});
}
`}};function zxe(e){let{inputs:t,backend:n,attrs:r}=e,{x:s,segmentIds:a}=t,{numSegments:o}=r,i=s.shape.length,l=[],u=0,c=_.getAxesPermutation([u],i),d=s;c!=null&&(d=Un({inputs:{x:s},backend:n,attrs:{perm:c}}),l.push(d),u=_.getInnerMostAxes(1,i)[0]);let h=_.segment_util.computeOutShape(d.shape,u,o),p=k.sizeFromShape([d.shape[u]]),f=ve({inputs:{x:d},backend:n,attrs:{shape:[-1,p]}});l.push(f);let m=cy(s.dtype),g=(b,v,I,w,S)=>{let E=b.shape[0],D=b.shape[1],$=_.segment_util.segOpComputeOptimalWindowSize(D,S),R={windowSize:$,inSize:D,batchSize:E,numSegments:S},N=new Pxe(R,v),M=n.compileAndRun(N,[b,I],w);if(l.push(M),M.shape[1]===S)return M;let B=e9({backend:n,attrs:{start:0,stop:S,step:1,dtype:"float32"}}),q=r9({inputs:{x:B},backend:n,attrs:{reps:[D/$]}});return l.push(B),l.push(q),g(M,v,q,w,S)},y=g(f,"unsortedSegmentSum",a,m,o),A=ve({inputs:{x:y},backend:n,attrs:{shape:h}}),x=A;if(c!=null){l.push(A);let b=_.getUndoAxesPermutation(c);x=Un({inputs:{x},backend:n,attrs:{perm:b}})}return l.forEach(b=>n.disposeIntermediateTensorInfo(b)),x}var Lxe={kernelName:of,backendName:"webgl",kernelFunc:zxe},Bxe=[h1e,m1e,Jfe,eme,rme,ome,lme,dme,pme,mme,xme,vme,Ime,Nme,Fme,$me,Pme,Wme,Lme,Gme,qme,Xme,Qme,o0e,l0e,f0e,g0e,b0e,k0e,Dfe,C0e,z0e,B0e,_0e,H0e,j0e,V0e,X0e,J0e,tge,rge,age,lge,fge,gge,cge,xge,wge,Ige,Cge,_ge,Oge,Lge,Bge,Wge,Uge,Gge,qge,Xge,Yge,t2e,s2e,i2e,u2e,h2e,m2e,x2e,k2e,_fe,S2e,T0e,C2e,R2e,F2e,Mfe,z2e,V2e,H2e,Y2e,K2e,t1e,s1e,l1e,y1e,S1e,k1e,E1e,R1e,D1e,v1e,M1e,P1e,W1e,G1e,X1e,rye,Bfe,aye,lye,dye,fye,c0e,yye,xye,vye,Iye,Cye,Pfe,$ye,Rye,d0e,Q1e,Fye,Vye,zye,Vfe,jye,Xye,Qye,nAe,oAe,lAe,dAe,fAe,gAe,xAe,wAe,SAe,CAe,RAe,FAe,s0e,tye,PAe,LAe,WAe,UAe,GAe,qAe,XAe,YAe,exe,rxe,axe,ixe,cxe,hxe,fxe,gxe,eye,Xfe,xxe,wxe,Sxe,Exe,_xe,Zfe,Fxe,Oxe,Lxe,Aye];for(let e of Bxe)ry(e);var tr;(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"})(tr||(tr={}));var ph;(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"})(ph||(ph={}));var a9;function Wxe(e){a9=e.wasm.cwrap(Ll,null,["number","array","number","number","array","number","number","number","number","number","number","number","number"])}function Vxe(e){let{inputs:t,backend:n,attrs:r}=e,{a:s,b:a,bias:o,preluActivationWeights:i}=t;if(s.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:d}=r,h=n.dataIdMap.get(s.dataId).id,p=n.dataIdMap.get(a.dataId).id,f=0;if(o!=null){let S=n.dataIdMap.get(o.dataId);if(S.shape.length!==1)throw new Error(`_FusedMatMul only supports rank-1 bias but got rank ${S.shape.length}.`);f=S.id}let m=i==null?0:n.dataIdMap.get(i.dataId).id,g=ph[c];if(g==null)throw new Error(`${c} activation not yet supported for FusedConv2D in the wasm backend.`);let y=l?s.shape[2]:s.shape[1],A=u?a.shape[1]:a.shape[2],x=s.shape[0],b=n.makeOutput([x,y,A],s.dtype),v=n.dataIdMap.get(b.dataId).id,I=new Uint8Array(new Int32Array(s.shape).buffer),w=new Uint8Array(new Int32Array(a.shape).buffer);return a9(h,I,s.shape.length,p,w,a.shape.length,l,u,g,f,m,d||0,v),b}var Uxe={kernelName:Ll,backendName:"wasm",setupFunc:Wxe,kernelFunc:Vxe};function $n(e){let t;function n(s){t=s.wasm.cwrap(e,null,["number","number"])}function r(s){let{backend:a,inputs:{x:o}}=s,i=a.dataIdMap.get(o.dataId).id,l=a.makeOutput(o.shape,o.dtype),u=a.dataIdMap.get(l.dataId).id;return k.sizeFromShape(l.shape)===0||t(i,u),l}return{kernelName:e,backendName:"wasm",setupFunc:n,kernelFunc:r}}var Hxe=$n(Ac);function Hn(e,t,n){let r;function s(o){r=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,d=i.dataIdMap.get(u.dataId).id,h=i.dataIdMap.get(c.dataId).id,p=n!=null?n:u.dtype,f=_.assertAndGetBroadcastShape(u.shape,c.shape),m=i.makeOutput(f,p);if(k.sizeFromShape(f)===0)return m;let g=new Uint8Array(new Int32Array(u.shape).buffer),y=new Uint8Array(new Int32Array(c.shape).buffer),A=i.dataIdMap.get(m.dataId).id,x=()=>r(d,g,u.shape.length,h,y,c.shape.length,tr[u.dtype],A);if(t&&u.dtype==="float32")return x(),m;let b=_.getBroadcastDims(u.shape,f),v=_.getBroadcastDims(c.shape,f),I=b.every((S,E)=>S===E),w=v.every((S,E)=>S===E);if(I&&w)return x(),m;throw new Error(`Broadcasting along outer dims is not yet supported for ${u.dtype} ${e}.`)}return{kernelName:e,backendName:"wasm",setupFunc:s,kernelFunc:a}}var Gxe=!0,jxe=Hn(Ma,Gxe),o9;function qxe(e){o9=e.wasm.cwrap(Xi,null,["array","number","number","number"])}function Kxe(e){let{inputs:t,backend:n}=e,r=n.makeOutput(t[0].shape,t[0].dtype);if(k.sizeFromShape(r.shape)===0)return r;let s=t.map(i=>n.dataIdMap.get(i.dataId).id),a=new Uint8Array(new Int32Array(s).buffer),o=n.dataIdMap.get(r.dataId).id;return o9(a,s.length,tr[r.dtype],o),r}var Xxe={kernelName:Xi,backendName:"wasm",setupFunc:qxe,kernelFunc:Kxe};function Xm(e){let{inputs:{x:t},backend:n}=e,r=n.makeOutput(t.shape,t.dtype),s=n.typedArrayFromHeap(t);return n.typedArrayFromHeap(r).set(s),r}var Zxe={kernelName:hl,backendName:"wasm",kernelFunc:Xm},i9;function Yxe(e){i9=e.wasm.cwrap(zl,null,["number","array","number","number","number","array","number"])}function Iu(e){let{inputs:t,backend:n,attrs:r}=e,[s,a]=Qxe(t.x.shape,r.perm),o=!0;for(let f=0;f<a.length;f++)a[f]!==f&&(o=!1);let i=Jxe(t.x.shape,r.perm),l={dataId:t.x.dataId,shape:s,dtype:t.x.dtype};if(o){let f=Xm({inputs:t,backend:n});return f.shape=i,f}let u=n.makeOutput(i,l.dtype),c=n.dataIdMap.get(l.dataId).id,d=n.dataIdMap.get(u.dataId).id,h=new Uint8Array(new Int32Array(a).buffer),p=new Uint8Array(new Int32Array(l.shape).buffer);return i9(c,p,l.shape.length,tr[l.dtype],d,h,a.length),u}function Jxe(e,t){let n=new Array(e.length);for(let r=0;r<n.length;r++)n[r]=e[t[r]];return n}function Qxe(e,t){let n=[],r=[];for(let s=0;s<e.length;++s)e[s]!==1&&n.push(e[s]),e[t[s]]!==1&&r.push(t[s]);for(let s=0;s<r.length;++s){let a=-1;for(let o=0;o<r.length;++o)r[o]>=s&&(a===-1||r[a]>r[o])&&(a=o);r[a]=s}return[n,r]}var e5e={kernelName:zl,backendName:"wasm",kernelFunc:Iu,setupFunc:Yxe};function to(e,t,n){let r=e.shape,s=e.shape.length,a=k.parseAxisParam(t,r),o=a,i=_.getAxesPermutation(o,s),l=null,u=!1;if(i!=null){let c=new Array(s);for(let p=0;p<c.length;p++)c[p]=r[i[p]];o=_.getInnerMostAxes(o.length,s),l=Iu({inputs:{x:e},attrs:{perm:i},backend:n});let d=n.dataIdMap.get(e.dataId).id;n.dataIdMap.get(l.dataId).id!==d&&(u=!0)}return{transposed:l,originalAxes:a,axes:o,inputWasTransposed:u}}var l9;function t5e(e){l9=e.wasm.cwrap(vc,null,["number, number, number"])}function n5e(e){let{backend:t,inputs:n,attrs:r}=e,{axis:s,keepDims:a}=r,{x:o}=n,l=t.dataIdMap.get(o.dataId).id,u=o,{transposed:c,axes:d,originalAxes:h,inputWasTransposed:p}=to(o,s,t);if(p){let x=t.dataIdMap.get(c.dataId).id;u=c,l=x}let f=u.shape.length;_.assertAxesAreInnerMostDims("all",d,f);let[m,g]=_.computeOutAndReduceShapes(u.shape,d),y=k.sizeFromShape(g),A=t.makeOutput(m,o.dtype);if(k.sizeFromShape(u.shape)!==0){let x=t.dataIdMap.get(A.dataId).id;l9(l,y,x)}if(p&&t.disposeData(c.dataId),a){let x=_.expandShapeToKeepDim(A.shape,h);A.shape=x}return A}var r5e={kernelName:vc,backendName:"wasm",setupFunc:t5e,kernelFunc:n5e},u9;function s5e(e){u9=e.wasm.cwrap(wc,null,["number, number, number"])}function a5e(e){let{backend:t,inputs:n,attrs:r}=e,{axis:s,keepDims:a}=r,{x:o}=n,l=t.dataIdMap.get(o.dataId).id,u=o,{transposed:c,axes:d,originalAxes:h,inputWasTransposed:p}=to(o,s,t);if(p){let x=t.dataIdMap.get(c.dataId).id;u=c,l=x}let f=u.shape.length;_.assertAxesAreInnerMostDims("any",d,f);let[m,g]=_.computeOutAndReduceShapes(u.shape,d),y=k.sizeFromShape(g),A=t.makeOutput(m,o.dtype);if(k.sizeFromShape(u.shape)!==0){let x=t.dataIdMap.get(A.dataId).id;u9(l,y,x)}if(p&&t.disposeData(c.dataId),a){let x=_.expandShapeToKeepDim(A.shape,h);A.shape=x}return A}var o5e={kernelName:wc,backendName:"wasm",setupFunc:s5e,kernelFunc:a5e},c9;function i5e(e){c9=e.wasm.cwrap(Zi,null,["number","number","number","number","number"])}function l5e(e){let{backend:t,inputs:n,attrs:r}=e,{axis:s}=r,{x:a}=n,o=t.dataIdMap.get(a.dataId).id,i=o,l=a,{transposed:u,axes:c,inputWasTransposed:d}=to(a,s,t);if(d){let y=t.dataIdMap.get(u.dataId).id;y!==o&&(l=u,i=y)}let h=l.shape.slice(0,-1),p=t.makeOutput(h,"int32"),f=t.dataIdMap.get(p.dataId).id,m=k.sizeFromShape(p.shape),g=l.shape[c[0]];return c9(i,tr[l.dtype],m,g,f),d&&t.disposeData(u.dataId),p}var u5e={kernelName:Zi,backendName:"wasm",kernelFunc:l5e,setupFunc:i5e},d9;function c5e(e){d9=e.wasm.cwrap(Yi,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function d5e(e){let{inputs:t,attrs:n,backend:r}=e,s=t.x,a=r.dataIdMap.get(s.dataId).id,{filterSize:o,strides:i,pad:l,dimRoundingMode:u}=n,c=_.computePool2DInfo(s.shape,o,i,1,l,u),d=c.filterHeight,h=c.filterWidth,p=c.padInfo.top,f=c.padInfo.right,m=c.padInfo.bottom,g=c.padInfo.left,y=c.strideHeight,A=c.strideWidth,x=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=r.makeOutput(c.outShape,"float32"),v=r.dataIdMap.get(b.dataId).id;return d9(a,s.shape[0],s.shape[1],s.shape[2],d,h,p,f,m,g,y,A,x,v),b}var h5e={kernelName:Yi,backendName:"wasm",setupFunc:c5e,kernelFunc:d5e};function nr(e){let{inputs:t,attrs:n}=e,{x:r}=t,{shape:s}=n,a=k.sizeFromShape(r.shape),o=k.inferFromImplicitShape(s,a);return k.assert(a===k.sizeFromShape(o),()=>`new shape: ${o}, old shape: ${r.shape}. New shape and old shape must have the same number of elements.`),e.backend.incRef(r.dataId),{dataId:r.dataId,shape:o,dtype:r.dtype}}var p5e={kernelName:Jc,backendName:"wasm",kernelFunc:nr},h9;function f5e(e){h9=e.wasm.cwrap(Ji,null,["number","array","number","number","array","number","number","number","number"])}function m5e(e){let{inputs:t,backend:n,attrs:r}=e,{a:s,b:a}=t,{transposeA:o,transposeB:i}=r;if(s.dtype!=="float32"||a.dtype!=="float32")throw new Error("BatchMatMul for non non-float32 tensors not yet supported.");let l=s.shape.length,u=a.shape.length,c=o?s.shape[l-2]:s.shape[l-1],d=i?a.shape[u-1]:a.shape[u-2],h=o?s.shape[l-1]:s.shape[l-2],p=i?a.shape[u-2]:a.shape[u-1],f=s.shape.slice(0,-2),m=a.shape.slice(0,-2),g=k.sizeFromShape(f),y=k.sizeFromShape(m),A=g===y||g===1||y===1;k.assert(l>=2&&u>=2&&A,()=>`Error in matMul: the input batch dimensions must either be the same or at least one input batch dimension must be 1. Got input batch dimensions of (${f}) and (${m}).`);let b=(g>y?s.shape.slice(0,-2):a.shape.slice(0,-2)).concat([h,p]);k.assert(c===d,()=>`Error in matMul: inner shapes (${c}) and (${d}) of Tensors with shapes ${s.shape} and ${a.shape} and transposeA=${o} and transposeB=${i} must match.`);let v=o?[g,c,h]:[g,h,c],I=i?[y,p,d]:[y,d,p],w=nr({inputs:{x:s},backend:n,attrs:{shape:v}}),S=nr({inputs:{x:a},backend:n,attrs:{shape:I}}),E=n.dataIdMap.get(w.dataId).id,D=n.dataIdMap.get(S.dataId).id,$=o?w.shape[2]:w.shape[1],R=i?S.shape[1]:S.shape[2],N=Math.max(g,y),M=n.makeOutput([N,$,R],w.dtype),B=n.dataIdMap.get(M.dataId).id,q=new Uint8Array(new Int32Array(w.shape).buffer),X=new Uint8Array(new Int32Array(S.shape).buffer);return h9(E,q,w.shape.length,D,X,S.shape.length,o,i,B),n.disposeData(w.dataId),n.disposeData(S.dataId),M.shape=b,M}var g5e={kernelName:Ji,backendName:"wasm",setupFunc:f5e,kernelFunc:m5e};function fh(e){let{inputs:{x:t},attrs:{begin:n,size:r},backend:s}=e,[a,o]=Cn.parseSliceParams(t,n,r),i=Cn.isSliceContinous(t.shape,a,o),l=s.readSync(t.dataId),u=s.makeOutput(o,t.dtype),c=k.computeStrides(t.shape),d=s.dataIdMap.get(u.dataId);if(i){let f=Cn.computeFlatOffset(a,c);return t.dtype==="string"?d.stringBytes=l.slice(f,f+k.sizeFromShape(o)):s.typedArrayFromHeap(u).set(l.subarray(f,f+k.sizeFromShape(o))),u}if(t.dtype==="string"){let f=X5(l,a,o,t.shape,t.dtype);return d.stringBytes=f,u}let h=s.typedArrayFromHeap(u),p=t.shape.length;if(p===2)y5e(l,c[0],h,a,o);else if(p===3)A5e(l,c[0],c[1],h,a,o);else if(p===4)x5e(l,c[0],c[1],c[2],h,a,o);else{let f=X5(l,a,o,t.shape,t.dtype);h.set(f)}return u}function y5e(e,t,n,r,s){let a=0,o=r[0],i=r[1],l=o+s[0];for(let u=o;u<l;u++){let c=u*t+i;n.set(e.subarray(c,c+s[1]),a),a+=s[1]}}function A5e(e,t,n,r,s,a){let o=0,i=s[0],l=s[1],u=s[2],c=i+a[0],d=l+a[1];for(let h=i;h<c;h++)for(let p=l;p<d;p++){let f=h*t+p*n+u;r.set(e.subarray(f,f+a[2]),o),o+=a[2]}}function x5e(e,t,n,r,s,a,o){let i=0,l=a[0],u=a[1],c=a[2],d=l+o[0],h=u+o[1],p=c+o[2],f=a[3];for(let m=l;m<d;m++)for(let g=u;g<h;g++)for(let y=c;y<p;y++){let A=m*t+g*n+y*r+f;s.set(e.subarray(A,A+o[3]),i),i+=o[3]}}var b5e={kernelName:nd,backendName:"wasm",kernelFunc:fh};function v5e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{blockShape:a,crops:o}=r,i=a.reduce((y,A)=>y*A),l=_.getReshaped(s.shape,a,i),u=_.getPermuted(l.length,a.length),c=_.getReshapedPermuted(s.shape,a,i),d=_.getSliceBeginCoords(o,a.length),h=_.getSliceSize(c,o,a.length),p=nr({inputs:{x:s},backend:n,attrs:{shape:l}}),f=Iu({inputs:{x:p},backend:n,attrs:{perm:u}}),m=nr({inputs:{x:f},backend:n,attrs:{shape:c}}),g=fh({inputs:{x:m},backend:n,attrs:{begin:d,size:h}});return n.disposeData(p.dataId),n.disposeData(f.dataId),n.disposeData(p.dataId),g}var w5e={kernelName:Cc,backendName:"wasm",kernelFunc:v5e};function Zm(e){let{inputs:{x:t},attrs:{dtype:n},backend:r}=e,s=r.makeOutput(t.shape,n),a=r.typedArrayFromHeap(t);return r.typedArrayFromHeap(s).set(a),s}var k5e={kernelName:Qi,backendName:"wasm",kernelFunc:Zm},I5e=$n(No),p9;function S5e(e){p9=e.wasm.cwrap(Co,null,["number","number","number","number"])}function T5e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{clipValueMin:a,clipValueMax:o}=r,i=n.dataIdMap.get(s.dataId).id,l=n.makeOutput(s.shape,s.dtype),u=n.dataIdMap.get(l.dataId).id;return p9(i,a,o,u),l}var N5e={kernelName:Co,backendName:"wasm",setupFunc:S5e,kernelFunc:T5e};function f9(e){let{inputs:t,backend:n}=e,r=k.parseAxisParam(e.attrs.axis,t[0].shape)[0],s=_.computeOutShape(t.map(p=>p.shape),r),a=t.filter(p=>k.sizeFromShape(p.shape)>0);if(a.length===1)return Xm({inputs:{x:a[0]},backend:n});let o=n.makeOutput(s,t[0].dtype);if(k.sizeFromShape(s)===0)return o;let i=a.map(p=>p.shape);if(_.assertParamsConsistent(i,r),a[0].dtype==="string"){let p=a.map(x=>{let b=k.sizeFromShape(x.shape.slice(r));return nr({inputs:{x},backend:n,attrs:{shape:[-1,b]}})}),f=p.map(x=>({vals:n.readSync(x.dataId),shape:x.shape}));s=_.computeOutShape(p.map(x=>x.shape),1);let m=p[0].shape[0]===1,g=MC(f,s,t[0].dtype,m),y=_.computeOutShape(a.map(x=>x.shape),r);o.shape=y;let A=n.dataIdMap.get(o.dataId);return A.stringBytes=_.fromStringArrayToUint8(g),p.forEach(x=>n.disposeData(x.dataId)),o}let l=k.sizeFromShape(a[0].shape.slice(0,r)),u=0,c=a.map(p=>{let f=k.sizeFromShape(p.shape.slice(r));return u+=f,f}),d=a.map(p=>n.typedArrayFromHeap(p)),h=n.typedArrayFromHeap(o);for(let p=0;p<l;p++){let f=p*u;for(let m=0;m<d.length;m++){let g=c[m],y=p*g,A=d[m].subarray(y,y+g);h.set(A,f),f+=g}}return o}var C5e={kernelName:Ec,backendName:"wasm",kernelFunc:f9},m9;function E5e(e){m9=e.wasm.cwrap(el,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function $5e(e){let{inputs:t,attrs:n,backend:r}=e,{x:s,filter:a}=t,o=r.dataIdMap.get(s.dataId).id,i=r.dataIdMap.get(a.dataId).id,{strides:l,dilations:u,pad:c,dimRoundingMode:d,dataFormat:h}=n,p=_.convertConv2DDataFormat(h),f=_.computeConv2DInfo(s.shape,a.shape,l,u,c,d,!1,p),m=f.filterHeight,g=f.filterWidth,y=f.padInfo.top,A=f.padInfo.right,x=f.padInfo.bottom,b=f.padInfo.left,v=f.dilationHeight,I=f.dilationWidth,w=f.strideHeight,S=f.strideWidth,E=f.inChannels,D=f.outChannels,$=f.padInfo.type==="SAME"?1:0;if(f.dataFormat!=="channelsLast")throw new Error(`wasm backend Conv2D does not support dataFormat:'${f.dataFormat}'. Please use 'channelsLast'.`);let R=r.makeOutput(f.outShape,"float32"),N=r.dataIdMap.get(R.dataId).id;return m9(o,s.shape[0],s.shape[1],s.shape[2],i,m,g,y,A,x,b,$,v,I,w,S,E,D,N),R}var R5e={kernelName:el,backendName:"wasm",setupFunc:E5e,kernelFunc:$5e},g9;function _5e(e){g9=e.wasm.cwrap(tl,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function D5e(e){let{backend:t,inputs:n,attrs:r}=e,{dy:s,filter:a}=n,{strides:o,pad:i,dataFormat:l,dimRoundingMode:u,inputShape:c}=r,d=1,h=_.convertConv2DDataFormat(l),p=_.computeConv2DInfo(c,a.shape,o,d,i,u,!1,h),{batchSize:f,filterHeight:m,filterWidth:g,inChannels:y,inHeight:A,inWidth:x,outChannels:b,outHeight:v,outWidth:I,strideHeight:w,strideWidth:S}=p,E=m-1-p.padInfo.top,D=g-1-p.padInfo.left,$=p.dataFormat==="channelsLast",R=k.computeStrides(p.inShape),N=k.computeStrides(s.shape),[M,B,q]=k.computeStrides(a.shape),X=R[0],J=$?R[1]:R[2],ee=$?R[2]:1,ae=$?1:R[1],se=N[0],oe=$?N[1]:N[2],ne=$?N[2]:1,ce=$?1:N[1],he=t.makeOutput(p.inShape,"float32"),me=t.dataIdMap.get(he.dataId).id,be=t.dataIdMap.get(s.dataId).id,Ee=t.dataIdMap.get(a.dataId).id;return g9(be,Ee,f,m,g,A,x,y,v,I,b,w,S,E,D,M,B,q,X,J,ee,ae,se,oe,ne,ce,me),he}var F5e={kernelName:tl,backendName:"wasm",setupFunc:_5e,kernelFunc:D5e},M5e=$n(nl),O5e=$n(rl),nb;(function(e){e[e.bilinear=0]="bilinear",e[e.nearest=1]="nearest"})(nb||(nb={}));var y9;function P5e(e){y9=e.wasm.cwrap($c,null,["number","number","number","number","array","number","number","number","number","number"])}function z5e(e){let{backend:t,inputs:n,attrs:r}=e,{method:s,extrapolationValue:a,cropSize:o}=r,{image:i,boxes:l,boxInd:u}=n,c=l.shape[0],[d,h]=o,p=[c,d,h,i.shape[3]],f=t.dataIdMap.get(i.dataId),m;i.dtype!=="float32"&&(m=Zm({backend:t,inputs:{x:i},attrs:{dtype:"float32"}}),f=t.dataIdMap.get(m.dataId));let g=f.id,y=t.dataIdMap.get(l.dataId).id,A=t.dataIdMap.get(u.dataId).id,x=t.makeOutput(p,"float32"),b=t.dataIdMap.get(x.dataId).id,v=new Uint8Array(new Int32Array(i.shape).buffer);return y9(g,y,A,c,v,d,h,nb[s],a,b),m!=null&&t.disposeData(m.dataId),x}var L5e={kernelName:$c,backendName:"wasm",setupFunc:P5e,kernelFunc:z5e},A9;function B5e(e){A9=e.wasm.cwrap(sl,null,["number","number","number","number","number","number"])}function W5e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{axis:a,exclusive:o,reverse:i}=r,l=s.shape.length;k.assert(s.dtype==="float32"||s.dtype==="int32",()=>`cumsum does not support ${s.dtype} tensors in the WASM backend`);let u=_.getAxesPermutation([a],l),c=s;u!==null&&(c=Iu({inputs:{x:s},attrs:{perm:u},backend:n}));let d=_.getInnerMostAxes(1,l)[0];_.assertAxesAreInnerMostDims("cumsum",[d],l);let h=n.makeOutput(c.shape,c.dtype),p=c.shape[d],f=n.dataIdMap.get(c.dataId).id,m=n.dataIdMap.get(h.dataId).id;A9(f,o?1:0,i?1:0,p,m,tr[s.dtype]);let g=h;if(u!==null){let y=_.getUndoAxesPermutation(u);g=Iu({inputs:{x:h},attrs:{perm:y},backend:n}),n.disposeData(c.dataId),n.disposeData(h.dataId)}return g}var V5e={kernelName:sl,backendName:"wasm",setupFunc:B5e,kernelFunc:W5e},x9;function U5e(e){x9=e.wasm.cwrap(Rc,null,["number","number","number","array","number","array","array","number","number"])}function H5e(e){let{backend:t,inputs:n,attrs:r}=e,{x:s}=n,{blockSize:a,dataFormat:o}=r;k.assert(a>1,()=>`blockSize should be > 1 for depthToSpace, but was: ${a}`);let i=s.shape[0],l=o==="NHWC"?s.shape[1]:s.shape[2],u=o==="NHWC"?s.shape[2]:s.shape[3],c=o==="NHWC"?s.shape[3]:s.shape[1],d=l*a,h=u*a,p=c/(a*a),f=o==="NHWC"?[i,d,h,p]:[i,p,d,h],m=t.makeOutput(f,"float32"),y=t.dataIdMap.get(s.dataId).id,A=new Uint8Array(new Int32Array(k.computeStrides(s.shape)).buffer),x=new Uint8Array(new Int32Array(f).buffer),b=new Uint8Array(new Int32Array(k.computeStrides(f)).buffer),v=t.dataIdMap.get(m.dataId).id;return x9(y,a,o==="NHWC"?1:0,A,s.shape.length-1,x,b,f.length,v),m}var G5e={kernelName:Rc,backendName:"wasm",setupFunc:U5e,kernelFunc:H5e},b9;function j5e(e){b9=e.wasm.cwrap(al,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function q5e(e){let{inputs:t,attrs:n,backend:r}=e,{x:s,filter:a}=t,o=r.dataIdMap.get(s.dataId).id,i=r.dataIdMap.get(a.dataId).id,{strides:l,dilations:u,pad:c,dimRoundingMode:d}=n,h=u==null?[1,1]:u,p=_.computeConv2DInfo(s.shape,a.shape,l,h,c,d,!0),f=p.filterHeight,m=p.filterWidth,g=p.padInfo.top,y=p.padInfo.right,A=p.padInfo.bottom,x=p.padInfo.left,b=p.dilationHeight,v=p.dilationWidth,I=p.strideHeight,w=p.strideWidth,S=p.inChannels,E=p.outChannels,D=p.padInfo.type==="SAME"?1:0;if(p.dataFormat!=="channelsLast")throw new Error(`wasm backend DepthwiseConv2dNative does not support dataFormat:'${p.dataFormat}'. Please use 'channelsLast'.`);let $=r.makeOutput(p.outShape,"float32"),R=r.dataIdMap.get($.dataId).id;return b9(o,s.shape[0],s.shape[1],s.shape[2],i,f,m,g,y,A,x,D,b,v,I,w,S,E,R),$}var K5e={kernelName:al,backendName:"wasm",setupFunc:j5e,kernelFunc:q5e},X5e=!1,Z5e=Hn(il,X5e,"bool"),Y5e=$n(Eo);function rb(e){let{inputs:t,attrs:n,backend:r}=e,{input:s}=t,{dim:a}=n,o=s.shape.length,i=s.shape.slice(),l=a;return a<0&&(k.assert(-(o+1)<=a,()=>`Axis must be in the interval [${-(o+1)}, ${o}]`),l=o+a+1),i.splice(l,0,1),nr({inputs:{x:s},backend:r,attrs:{shape:i}})}var J5e={kernelName:Fc,backendName:"wasm",kernelFunc:rb};function Q5e(e){let{attrs:{shape:t,value:n,dtype:r},backend:s}=e,a=s.makeOutput(t,r);return s.typedArrayFromHeap(a).fill(n),a}var ebe={kernelName:Jp,backendName:"wasm",kernelFunc:Q5e},v9;function tbe(e){v9=e.wasm.cwrap(Mc,null,["number","number","number","number","number","number"])}function nbe(e){let{inputs:t,backend:n}=e,{image:r}=t,s=n.makeOutput(r.shape,r.dtype),a=n.dataIdMap.get(r.dataId).id,o=n.dataIdMap.get(s.dataId).id,[i,l,u,c]=r.shape;return v9(a,i,l,u,c,o),s}var rbe={kernelName:Mc,backendName:"wasm",kernelFunc:nbe,setupFunc:tbe},sbe=$n($o),abe=!1,obe=Hn(ul,abe),w9;function ibe(e){w9=e.wasm.cwrap(cl,null,["number","number","number","number","number","number","number"])}function lbe(e){let{backend:t,inputs:n,attrs:r}=e,{varianceEpsilon:s}=r,{x:a,mean:o,variance:i,offset:l,scale:u}=n,c=t.dataIdMap.get(a.dataId).id,d=t.dataIdMap.get(o.dataId).id,h=t.dataIdMap.get(i.dataId).id,p=l!=null?t.dataIdMap.get(l.dataId).id:0,f=u!=null?t.dataIdMap.get(u.dataId).id:0,m=t.makeOutput(a.shape,a.dtype);if(k.sizeFromShape(a.shape)===0)return m;let g=t.dataIdMap.get(m.dataId).id;return w9(c,d,h,p,f,s,g),m}var ube={kernelName:cl,backendName:"wasm",setupFunc:ibe,kernelFunc:lbe},k9;function cbe(e){k9=e.wasm.cwrap(Bl,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function dbe(e){let{inputs:t,attrs:n,backend:r}=e,{x:s,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dilations:c,dataFormat:d,dimRoundingMode:h,activation:p,leakyreluAlpha:f}=n,m=_.computeConv2DInfo(s.shape,a.shape,l,c,u,h),g=ph[p];if(g==null)throw new Error(`${p} activation not yet supported for FusedConv2D in the wasm backend.`);let y=r.dataIdMap.get(s.dataId).id,A=r.dataIdMap.get(a.dataId).id,x=m.outChannels,b=0;if(o!=null){let ne=r.dataIdMap.get(o.dataId);if(ne.shape.length!==1)throw new Error(`FusedConv2D only supports rank-1 bias but got rank ${ne.shape.length}.`);if(ne.shape[0]!==x)throw new Error(`FusedConv2D bias shape (${ne.shape}) does not match the number of output channels (${x})`);b=ne.id}let v=m.filterHeight,I=m.filterWidth,w=m.padInfo.top,S=m.padInfo.right,E=m.padInfo.bottom,D=m.padInfo.left,$=m.dilationHeight,R=m.dilationWidth,N=m.strideHeight,M=m.strideWidth,B=m.inChannels,q=m.padInfo.type==="SAME"?1:0,X=m.batchSize,J=m.inHeight,ee=m.inWidth;if(d!=="NHWC")throw new Error(`wasm backend FusedConv2D does not support dataFormat:'${d}'. Please use 'NHWC'.`);let ae=r.makeOutput(m.outShape,"float32"),se=r.dataIdMap.get(ae.dataId).id,oe=i==null?0:r.dataIdMap.get(i.dataId).id;return k9(y,X,J,ee,A,v,I,b,w,S,E,D,q,$,R,N,M,B,x,g,oe,f||0,se),ae}var hbe={kernelName:Bl,backendName:"wasm",setupFunc:cbe,kernelFunc:dbe},I9;function pbe(e){I9=e.wasm.cwrap(Wl,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 fbe(e){let{inputs:t,attrs:n,backend:r}=e,{x:s,filter:a,bias:o,preluActivationWeights:i}=t,{strides:l,pad:u,dilations:c,dataFormat:d,dimRoundingMode:h,activation:p,leakyreluAlpha:f}=n,m=_.computeConv2DInfo(s.shape,a.shape,l,c,u,h,!0),g=ph[p];if(g==null)throw new Error(`${p} activation not yet supported for FusedDepthwiseConv2D in the wasm backend.`);let y=r.dataIdMap.get(s.dataId).id,A=r.dataIdMap.get(a.dataId).id,x=m.outChannels,b=0;if(o!=null){let ne=r.dataIdMap.get(o.dataId);if(ne.shape.length!==1)throw new Error(`FusedDepthwiseConv2D only supports rank-1 bias but got rank ${ne.shape.length}.`);if(ne.shape[0]!==x)throw new Error(`FusedDepthwiseConv2D bias shape (${ne.shape}) does not match the number of output channels (${x})`);b=ne.id}let v=m.filterHeight,I=m.filterWidth,w=m.padInfo.top,S=m.padInfo.right,E=m.padInfo.bottom,D=m.padInfo.left,$=m.dilationHeight,R=m.dilationWidth,N=m.strideHeight,M=m.strideWidth,B=m.inChannels,q=m.padInfo.type==="SAME"?1:0,X=m.batchSize,J=m.inHeight,ee=m.inWidth;if(d!=="NHWC")throw new Error(`wasm backend FusedDepthwiseConv2D does not support dataFormat:'${d}'. Please use 'NHWC'.`);let ae=r.makeOutput(m.outShape,"float32"),se=r.dataIdMap.get(ae.dataId).id,oe=i==null?0:r.dataIdMap.get(i.dataId).id;return I9(y,X,J,ee,A,v,I,b,w,S,E,D,q,$,R,N,M,B,x,g,oe,f||0,se),ae}var mbe={kernelName:Wl,backendName:"wasm",setupFunc:pbe,kernelFunc:fbe},S9;function gbe(e){S9=e.wasm.cwrap(Pc,null,["number","number","number","number","number","number","array","number"])}function ybe(e){let{backend:t,inputs:n}=e,{params:r,indices:s}=n,[a,o,i,l]=x6.prepareAndValidate(r,s),u=t.makeOutput(a,r.dtype);if(o===0)return u;let c=s.shape,d=c[c.length-1],p=t.dataIdMap.get(r.dataId).id,m=t.dataIdMap.get(s.dataId).id,g=new Uint8Array(new Int32Array(l).buffer),y=t.dataIdMap.get(u.dataId).id;return S9(p,tr[r.dtype],m,o,d,i,g,y),u}var Abe={kernelName:Pc,backendName:"wasm",setupFunc:gbe,kernelFunc:ybe},T9;function xbe(e){T9=e.wasm.cwrap("Gather",null,["number","number","array","number","number","number","array","number"])}function bbe(e){let{backend:t,inputs:n,attrs:r}=e,{x:s,indices:a}=n,{axis:o,batchDims:i}=r,l=k.parseAxisParam(o,s.shape)[0],u=_.segment_util.collectGatherOpShapeInfo(s,a,l,i),c=nr({inputs:{x:s},attrs:{shape:[u.batchSize,u.outerSize,u.dimSize,u.sliceSize]},backend:t}),d=k.sizeFromShape(a.shape),h=nr({inputs:{x:a},attrs:{shape:[u.batchSize,d/u.batchSize]},backend:t}),p=[u.batchSize,u.outerSize,d/u.batchSize,u.sliceSize],f=t.makeOutput(p,s.dtype);if(k.sizeFromShape(s.shape)===0)return f;let m=c.shape.length-1,y=t.dataIdMap.get(c.dataId).id,x=t.dataIdMap.get(h.dataId).id,b=t.dataIdMap.get(f.dataId).id,v=new Uint8Array(new Int32Array(k.computeStrides(c.shape)).buffer),I=new Uint8Array(new Int32Array(k.computeStrides(p)).buffer);return T9(y,tr[s.dtype],v,m,x,u.batchSize,I,b),t.disposeData(c.dataId),t.disposeData(h.dataId),f.shape=u.outputShape,f}var vbe={kernelName:Oc,backendName:"wasm",setupFunc:xbe,kernelFunc:bbe},wbe=!1,kbe=Hn(dl,wbe,"bool"),Ibe=!1,Sbe=Hn(Ro,Ibe,"bool"),N9;function Tbe(e){N9=e.wasm.cwrap(pl,null,["number","number","number"])}function Nbe(e){let{inputs:{x:t},attrs:{alpha:n},backend:r}=e,s=r.dataIdMap.get(t.dataId).id,a=r.makeOutput(t.shape,t.dtype);if(k.sizeFromShape(t.shape)!==0){let o=r.dataIdMap.get(a.dataId).id;N9(s,n,o)}return a}var Cbe={kernelName:pl,backendName:"wasm",setupFunc:Tbe,kernelFunc:Nbe},Ebe=!1,$be=Hn(fl,Ebe,"bool"),Rbe=!1,_be=Hn(ml,Rbe,"bool"),Dbe=$n(_o),Fbe=!1,Mbe=Hn(Vc,Fbe,"bool"),C9;function Obe(e){C9=e.wasm.cwrap(gl,null,["number, number, number"])}function Pbe(e){let{backend:t,inputs:n,attrs:r}=e,{reductionIndices:s,keepDims:a}=r,{x:o}=n,l=t.dataIdMap.get(o.dataId).id,u=o,{transposed:c,axes:d,originalAxes:h,inputWasTransposed:p}=to(o,s,t);if(p){let x=t.dataIdMap.get(c.dataId).id;u=c,l=x}let f=u.shape.length;_.assertAxesAreInnerMostDims("max",d,f);let[m,g]=_.computeOutAndReduceShapes(u.shape,d),y=k.sizeFromShape(g),A=t.makeOutput(m,o.dtype);if(k.sizeFromShape(u.shape)!==0){let x=t.dataIdMap.get(A.dataId).id;C9(l,y,x)}if(p&&t.disposeData(c.dataId),a){let x=_.expandShapeToKeepDim(A.shape,h);A.shape=x}return A}var zbe={kernelName:gl,backendName:"wasm",setupFunc:Obe,kernelFunc:Pbe},Lbe=!1,Bbe=Hn(Do,Lbe),E9;function Wbe(e){E9=e.wasm.cwrap(yl,null,["number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number","number"])}function Vbe(e){let{inputs:t,attrs:n,backend:r}=e,s=t.x,a=r.dataIdMap.get(s.dataId).id,{filterSize:o,strides:i,pad:l,dimRoundingMode:u}=n,c=_.computePool2DInfo(s.shape,o,i,1,l,u),d=c.filterHeight,h=c.filterWidth,p=c.padInfo.top,f=c.padInfo.right,m=c.padInfo.bottom,g=c.padInfo.left,y=c.dilationHeight,A=c.dilationWidth,x=c.strideHeight,b=c.strideWidth,v=c.inChannels,I=c.outChannels;if(c.dataFormat!=="channelsLast")throw new Error(`wasm backend does not support dataFormat:'${c.dataFormat}'. Please use 'channelsLast'.`);let w=r.makeOutput(c.outShape,"float32"),S=r.dataIdMap.get(w.dataId).id;return E9(a,s.shape[0],s.shape[1],s.shape[2],d,h,p,f,m,g,y,A,x,b,v,I,S),w}var Ube={kernelName:yl,backendName:"wasm",setupFunc:Wbe,kernelFunc:Vbe},$9;function Hbe(e){$9=e.wasm.cwrap(Al,null,["number, number, number"])}function Gbe(e){let{backend:t,inputs:n,attrs:r}=e,{axis:s,keepDims:a}=r,{x:o}=n,i=t.dataIdMap.get(o.dataId).id,l=i,u=o,{transposed:c,axes:d,originalAxes:h,inputWasTransposed:p}=to(o,s,t),f=d;if(p){let b=t.dataIdMap.get(c.dataId).id;b!==i&&(u=c,l=b,f=_.getInnerMostAxes(f.length,u.shape.length))}_.assertAxesAreInnerMostDims("mean",f,u.shape.length);let[m,g]=_.computeOutAndReduceShapes(u.shape,f),y=k.sizeFromShape(g),A=u;u.dtype!=="float32"&&(A=Zm({backend:t,inputs:{x:u},attrs:{dtype:"float32"}}),l=t.dataIdMap.get(A.dataId).id);let x=t.makeOutput(m,"float32");if(k.sizeFromShape(u.shape)!==0){let b=t.dataIdMap.get(x.dataId).id;$9(l,y,b)}if(p&&t.disposeData(c.dataId),a){let b=_.expandShapeToKeepDim(x.shape,h);x.shape=b}return u.dtype!=="float32"&&t.disposeData(A.dataId),x}var jbe={kernelName:Al,backendName:"wasm",setupFunc:Hbe,kernelFunc:Gbe},R9;function qbe(e){R9=e.wasm.cwrap(xl,null,["number, number, number"])}function Kbe(e){let{backend:t,inputs:n,attrs:r}=e,{axis:s,keepDims:a}=r,{x:o}=n,i=t.dataIdMap.get(o.dataId).id,l=i,u=o,{transposed:c,axes:d,originalAxes:h,inputWasTransposed:p}=to(o,s,t);if(p){let x=t.dataIdMap.get(c.dataId).id;x!==i&&(u=c,l=x)}let f=u.shape.length;_.assertAxesAreInnerMostDims("min",d,f);let[m,g]=_.computeOutAndReduceShapes(u.shape,d),y=k.sizeFromShape(g),A=t.makeOutput(m,u.dtype);if(k.sizeFromShape(u.shape)!==0){let x=t.dataIdMap.get(A.dataId).id;R9(l,y,x)}if(p&&t.disposeData(c.dataId),a){let x=_.expandShapeToKeepDim(A.shape,h);A.shape=x}return A}var Xbe={kernelName:xl,backendName:"wasm",setupFunc:qbe,kernelFunc:Kbe},Zbe=!1,Ybe=Hn(Fo,Zbe),sb;(function(e){e[e.reflect=0]="reflect",e[e.symmetric=1]="symmetric"})(sb||(sb={}));var _9;function Jbe(e){_9=e.wasm.cwrap(bl,null,["number","array","number","number","array","array","number","number"])}function Qbe(e){let{inputs:{x:t},backend:n,attrs:{paddings:r,mode:s}}=e,a=r.map((f,m)=>f[0]+t.shape[m]+f[1]),o=n.dataIdMap.get(t.dataId).id,i=n.makeOutput(a,t.dtype),l=n.dataIdMap.get(i.dataId).id,u=new Uint8Array(new Int32Array(t.shape).buffer),c=r.map(f=>f[0]),d=r.map(f=>f[1]),h=new Uint8Array(new Int32Array(c).buffer),p=new Uint8Array(new Int32Array(d).buffer);return _9(o,u,t.shape.length,tr[t.dtype],h,p,sb[s],l),i}var e3e={kernelName:bl,backendName:"wasm",kernelFunc:Qbe,setupFunc:Jbe},t3e=!0,n3e=Hn(Mo,t3e),r3e=$n(Hc);function ab(e,t){let n=new Int32Array(e.wasm.HEAPU8.buffer,t,4),r=n[0],s=n[1],a=n[2],o=n[3];return e.wasm._free(t),{pSelectedIndices:r,selectedSize:s,pSelectedScores:a,pValidOutputs:o}}var D9;function s3e(e){D9=e.wasm.cwrap(Gc,"number",["number","number","number","number","number"])}function a3e(e){let{backend:t,inputs:n,attrs:r}=e,{iouThreshold:s,maxOutputSize:a,scoreThreshold:o}=r,{boxes:i,scores:l}=n,u=t.dataIdMap.get(i.dataId).id,c=t.dataIdMap.get(l.dataId).id,d=D9(u,c,a,s,o),{pSelectedIndices:h,selectedSize:p,pSelectedScores:f,pValidOutputs:m}=ab(t,d);return t.wasm._free(f),t.wasm._free(m),t.makeOutput([p],"int32",h)}var o3e={kernelName:Gc,backendName:"wasm",setupFunc:s3e,kernelFunc:a3e},F9;function i3e(e){F9=e.wasm.cwrap(jc,"number",["number","number","number","number","number","bool"])}function l3e(e){let{backend:t,inputs:n,attrs:r}=e,{iouThreshold:s,maxOutputSize:a,scoreThreshold:o,padToMaxOutputSize:i}=r,{boxes:l,scores:u}=n,c=t.dataIdMap.get(l.dataId).id,d=t.dataIdMap.get(u.dataId).id,h=F9(c,d,a,s,o,i),{pSelectedIndices:p,selectedSize:f,pSelectedScores:m,pValidOutputs:g}=ab(t,h);t.wasm._free(m);let y=t.makeOutput([f],"int32",p),A=t.makeOutput([],"int32",g);return[y,A]}var u3e={kernelName:jc,backendName:"wasm",setupFunc:i3e,kernelFunc:l3e},M9;function c3e(e){M9=e.wasm.cwrap(qc,"number",["number","number","number","number","number","number"])}function d3e(e){let{backend:t,inputs:n,attrs:r}=e,{iouThreshold:s,maxOutputSize:a,scoreThreshold:o,softNmsSigma:i}=r,{boxes:l,scores:u}=n,c=t.dataIdMap.get(l.dataId).id,d=t.dataIdMap.get(u.dataId).id,h=M9(c,d,a,s,o,i),{pSelectedIndices:p,selectedSize:f,pSelectedScores:m,pValidOutputs:g}=ab(t,h);t.wasm._free(g);let y=t.makeOutput([f],"int32",p),A=t.makeOutput([f],"float32",m);return[y,A]}var h3e={kernelName:qc,backendName:"wasm",setupFunc:c3e,kernelFunc:d3e},p3e=!1,f3e=Hn(vl,p3e,"bool"),O9;function m3e(e){O9=e.wasm.cwrap(wl,null,["number","number","number","number","number"])}function g3e(e){let{inputs:t,backend:n,attrs:r}=e,{indices:s}=t,{depth:a,onValue:o,offValue:i}=r,l=n.makeOutput([...s.shape,a],"int32"),u=n.dataIdMap.get(l.dataId).id,d=n.dataIdMap.get(s.dataId).id;return O9(d,a,o,i,u),l}var y3e={kernelName:wl,backendName:"wasm",setupFunc:m3e,kernelFunc:g3e};function A3e(e){let{inputs:{x:t},backend:n}=e,r=n.makeOutput(t.shape,t.dtype);return n.typedArrayFromHeap(r).fill(1),r}var x3e={kernelName:Kc,backendName:"wasm",kernelFunc:A3e};function b3e(e){let{inputs:t,backend:n,attrs:r}=e,{axis:s}=r;if(t.length===1)return rb({inputs:{input:t[0]},backend:n,attrs:{dim:s}});let a=t[0].shape,o=t[0].dtype;t.forEach(c=>{k.assertShapesMatch(a,c.shape,"All tensors passed to stack must have matching shapes"),k.assert(o===c.dtype,()=>"All tensors passed to stack must have matching dtypes")});let i=[],l=t.map(c=>{let d=rb({inputs:{input:c},backend:n,attrs:{dim:s}});return i.push(d),d}),u=f9({inputs:l,backend:n,attrs:{axis:s}});return i.forEach(c=>n.disposeData(c.dataId)),u}var v3e={kernelName:Xc,backendName:"wasm",kernelFunc:b3e},P9;function w3e(e){P9=e.wasm.cwrap(kl,null,["number","array","number","number","array","array","number","number"])}function k3e(e){let{inputs:{x:t},backend:n,attrs:{paddings:r,constantValue:s}}=e,a=r.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=r.map(f=>f[0]),d=r.map(f=>f[1]),h=new Uint8Array(new Int32Array(c).buffer),p=new Uint8Array(new Int32Array(d).buffer);return P9(o,u,t.shape.length,tr[t.dtype],h,p,s,l),i}var z9={kernelName:kl,backendName:"wasm",kernelFunc:k3e,setupFunc:w3e},I3e=!1,S3e=Hn(Il,I3e),L9;function T3e(e){L9=e.wasm.cwrap(Sl,null,["number","number","number"])}function N3e(e){let{inputs:t,backend:n}=e,{x:r,alpha:s}=t,a=n.dataIdMap.get(r.dataId).id,o=n.dataIdMap.get(s.dataId).id,i=n.makeOutput(r.shape,"float32"),l=n.dataIdMap.get(i.dataId).id;return L9(a,o,l),i}var C3e={kernelName:Sl,backendName:"wasm",setupFunc:T3e,kernelFunc:N3e},B9;function E3e(e){B9=e.wasm.cwrap(Zc,null,["number","number","number","number"])}function $3e(e){let{backend:t,inputs:n,attrs:r}=e,{axis:s,keepDims:a}=r,{x:o}=n,i=t.dataIdMap.get(o.dataId).id,l=i,u=o,{transposed:c,axes:d,originalAxes:h,inputWasTransposed:p}=to(o,s,t),f=d;if(p){let x=t.dataIdMap.get(c.dataId).id;x!==i&&(u=c,l=x,f=_.getInnerMostAxes(f.length,u.shape.length))}_.assertAxesAreInnerMostDims("prod",f,u.shape.length);let[m,g]=_.computeOutAndReduceShapes(u.shape,f),y=k.sizeFromShape(g),A=t.makeOutput(m,u.dtype);if(k.sizeFromShape(u.shape)!==0){let x=t.dataIdMap.get(A.dataId).id;B9(l,y,tr[A.dtype],x)}if(p&&t.disposeData(c.dataId),a){let x=_.expandShapeToKeepDim(A.shape,h);A.shape=x}return A}var R3e={kernelName:Zc,backendName:"wasm",setupFunc:E3e,kernelFunc:$3e},_3e=e=>{let{backend:t,attrs:n}=e,{start:r,stop:s,step:a,dtype:o}=n,i=KC(r,s,a,o),l=t.makeOutput([i.length],o);return t.typedArrayFromHeap(l).set(i),l},D3e={kernelName:rf,backendName:"wasm",kernelFunc:_3e},F3e=!0,M3e=Hn(ol,F3e),O3e=$n(Tl),P3e=$n(Cl),W9;function z3e(e){W9=e.wasm.cwrap(Nl,null,["number","number","number","number","number","number","number","number","number","number"])}function L3e(e){let{backend:t,inputs:n,attrs:r}=e,{images:s}=n,{alignCorners:a,halfPixelCenters:o,size:i}=r,[l,u]=i,[c,d,h,p]=s.shape,f=[c,l,u,p],m=t.dataIdMap.get(s.dataId),g;m.dtype!=="float32"&&(g=Zm({backend:t,inputs:{x:s},attrs:{dtype:"float32"}}),m=t.dataIdMap.get(g.dataId));let y=m.id,A=t.makeOutput(f,"float32");if(k.sizeFromShape(s.shape)===0)return A;let x=t.dataIdMap.get(A.dataId).id;return W9(y,c,d,h,p,l,u,a?1:0,o?1:0,x),g!=null&&t.disposeData(g.dataId),A}var B3e={kernelName:Nl,backendName:"wasm",setupFunc:z3e,kernelFunc:L3e},V9;function W3e(e){V9=e.wasm.cwrap(El,null,["number","array","number","array","number","number"])}function V3e(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{dims:a}=r,o=k.parseAxisParam(a,s.shape);if(s.shape.length===0)return Xm({inputs:{x:s},backend:n});let i=n.makeOutput(s.shape,s.dtype),l=n.dataIdMap.get(s.dataId).id,u=n.dataIdMap.get(i.dataId).id,c=new Uint8Array(new Int32Array(o).buffer),d=new Uint8Array(new Int32Array(s.shape).buffer);V9(l,c,o.length,d,s.shape.length,u);let h=nr({inputs:{x:i},attrs:{shape:s.shape},backend:n});return n.disposeData(i.dataId),h}var U3e={kernelName:El,backendName:"wasm",kernelFunc:V3e,setupFunc:W3e},U9;function H3e(e){U9=e.wasm.cwrap(pd,null,["number","number","number","number","number","number","number","number","array","number","number"])}function G3e(e){let{inputs:t,backend:n,attrs:r}=e,{image:s}=t,{radians:a,fillValue:o,center:i}=r,l=n.makeOutput(s.shape,s.dtype),u=n.dataIdMap.get(s.dataId).id,c=n.dataIdMap.get(l.dataId).id,[d,h,p,f]=s.shape,[m,g]=_.getImageCenter(i,h,p),y=o===0,A=255,x=typeof o=="number"?[o,o,o,y?0:A]:[...o,A],b=new Uint8Array(new Int32Array(x).buffer);return U9(u,d,h,p,f,a,m,g,b,x.length,c),l}var j3e={kernelName:pd,backendName:"wasm",kernelFunc:G3e,setupFunc:H3e},q3e=$n($l),K3e=$n(Oo),H9;function X3e(e){H9=e.wasm.cwrap(Qc,null,["number","number","number","number","number","number","array","number","number"])}function Z3e(e){let{backend:t,inputs:n,attrs:r}=e,{indices:s,updates:a}=n,{shape:o}=r,i=t.makeOutput(o,a.dtype);if(k.sizeFromShape(o)===0)return i;let{sliceRank:l,numUpdates:u,sliceSize:c,strides:d,outputSize:h}=v6.calculateShapes(a,s,o),f=t.dataIdMap.get(s.dataId).id,g=t.dataIdMap.get(a.dataId).id,y=new Uint8Array(new Int32Array(d).buffer),A=t.dataIdMap.get(i.dataId).id;return H9(f,g,tr[a.dtype],l,u,c,y,h,A),i}var Y3e={kernelName:Qc,backendName:"wasm",setupFunc:X3e,kernelFunc:Z3e},G9;function J3e(e){G9=e.wasm.cwrap("SelectV2",null,["number","number","number","number","number"])}function Q3e(e){let{inputs:t,backend:n}=e,{condition:r,t:s,e:a}=t,o=n.dataIdMap.get(r.dataId).id,i=n.dataIdMap.get(s.dataId).id,l=n.dataIdMap.get(a.dataId).id,u=n.makeOutput(s.shape,s.dtype),c=n.dataIdMap.get(u.dataId).id,d=r.shape.length,h=s.shape.length,p=d===0||d>1||h===1?1:k.sizeFromShape(s.shape.slice(1));return G9(o,i,l,p,c),u}var eve={kernelName:ed,backendName:"wasm",kernelFunc:Q3e,setupFunc:J3e},j9;function tve(e){j9=e.wasm.cwrap(_l,null,["number","number"])}function nve(e){let{backend:t,inputs:{x:n}}=e,r=t.dataIdMap.get(n.dataId).id,s=t.makeOutput(n.shape,n.dtype),a=t.dataIdMap.get(s.dataId).id;return k.sizeFromShape(s.shape)===0||j9(r,a),s}var rve={kernelName:"Sigmoid",backendName:"wasm",setupFunc:tve,kernelFunc:nve},sve=$n(Rl),q9;function ave(e){q9=e.wasm.cwrap(Ml,null,["number","number","number","number"])}function ove(e){let{backend:t,inputs:{logits:n},attrs:{dim:r}}=e,s=t.dataIdMap.get(n.dataId).id,a=t.makeOutput(n.shape,n.dtype),o=t.dataIdMap.get(a.dataId).id,i=n.shape[r],l=k.sizeFromShape(n.shape)/i;return k.sizeFromShape(a.shape)===0||q9(s,o,i,l),a}var ive={kernelName:Ml,backendName:"wasm",setupFunc:ave,kernelFunc:ove};function lve(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,{blockShape:a,paddings:o}=r,i=k.sizeFromShape(a),l=[[0,0]];l.push(...o);for(let I=1+a.length;I<s.shape.length;++I)l.push([0,0]);let u=z9.kernelFunc({inputs:{x:s},backend:n,attrs:{paddings:l,constantValue:0}}),c=_.getReshaped(u.shape,a,i,!1),d=_.getPermuted(c.length,a.length,!1),h=_.getReshapedPermuted(u.shape,a,i,!1),m=nr({inputs:{x:u},backend:n,attrs:{shape:c}}),A=Iu({inputs:{x:m},backend:n,attrs:{perm:d}}),v=nr({inputs:{x:A},backend:n,attrs:{shape:h}});return n.disposeData(u.dataId),n.disposeData(m.dataId),n.disposeData(A.dataId),v}var uve={kernelName:od,backendName:"wasm",kernelFunc:lve};function cve(e){let{inputs:t,attrs:n,backend:r}=e,{x:s}=t,{numOrSizeSplits:a,axis:o}=n,i=k.parseAxisParam(o,s.shape)[0],l=_.prepareSplitSize(s,a,i),u=new Array(s.shape.length).fill(0),c=s.shape.slice();return l.map(d=>{let h=[...c];h[i]=d;let p=fh({inputs:{x:s},attrs:{begin:u,size:h},backend:r});return u[i]+=d,p})}var dve={kernelName:id,backendName:"wasm",kernelFunc:cve},hve=$n(Dl),pve=$n(af),fve=!0,mve=Hn(Po,fve),K9;function gve(e){K9=e.wasm.cwrap(Bo,null,["number","number","number"])}function yve(e){let{backend:t,inputs:n,attrs:r}=e,{alpha:s}=r,{x:a}=n,o=t.dataIdMap.get(a.dataId).id,i=t.makeOutput(a.shape,a.dtype),l=t.dataIdMap.get(i.dataId).id;return K9(o,s,l),i}var Ave={kernelName:Bo,backendName:"wasm",setupFunc:gve,kernelFunc:yve},X9;function xve(e){X9=e.wasm.cwrap(ld,null,["number","array","number","array","array","array","array","array","number","number"])}function bve(e){let{backend:t,inputs:n,attrs:r}=e,{x:s}=n,{begin:a,end:o,strides:i}=r;i==null&&(i=new Array(a.length));let{beginMask:l,endMask:u,ellipsisMask:c,newAxisMask:d,shrinkAxisMask:h}=r,p=_.slice_util.maskToAxes(c);if(p.length>1)throw new Error("Multiple ellipses in slice is not allowed.");if(c!==0&&d!==0)throw new Error("Using both ellipsisMask and newAxisMask is not yet supported.");if(c!==0&&h!==0)throw new Error("Using both ellipsisMask and shrinkAxisMask is not yet supported.");let f=s.shape.length-a.length,m=_.slice_util.maskToAxes(d),g=s.shape.slice();m.forEach($=>{a[$]=0,o[$]=1,g.splice($,0,1)});let y=nr({inputs:{x:s},attrs:{shape:g},backend:t}),{begin:A,end:x,strides:b}=_.slice_util.getNormalizedAxes(y.shape,p,f,a,o,i,l,u,c);a=A,o=x,i=b;let v=_.slice_util.maskToAxes(h);v.forEach($=>{o[$]=a[$]+1,i[$]=1});let I=_.slice_util.computeOutShape(a,o,i),w=I.filter(($,R)=>v.indexOf(R)===-1);if(i.every($=>$===1)){let $=fh({inputs:{x:y},attrs:{begin:a,size:I},backend:t});t.disposeData(y.dataId);let R=nr({inputs:{x:$},attrs:{shape:w},backend:t});return t.disposeData($.dataId),R}let E=t.makeOutput(w,"float32");if(!w.some($=>$===0)){let $=t.dataIdMap.get(y.dataId).id,R=new Uint8Array(new Int32Array(k.computeStrides(y.shape)).buffer),N=new Uint8Array(new Int32Array(a).buffer),M=new Uint8Array(new Int32Array(o).buffer),B=new Uint8Array(new Int32Array(i).buffer),q=new Uint8Array(new Int32Array(w).buffer),X=new Uint8Array(new Int32Array(k.computeStrides(w)).buffer),J=t.dataIdMap.get(E.dataId).id;X9($,R,y.shape.length,N,M,B,q,X,w.length,J)}t.disposeData(y.dataId);let D=nr({inputs:{x:E},attrs:{shape:w},backend:t});return t.disposeData(E.dataId),D}var vve={kernelName:ld,backendName:"wasm",setupFunc:xve,kernelFunc:bve},wve=!0,kve=Hn(zo,wve),Z9;function Ive(e){Z9=e.wasm.cwrap(Fl,null,["number, number, number"])}function Sve(e){let{backend:t,inputs:n,attrs:r}=e,{axis:s,keepDims:a}=r,{x:o}=n,i=t.dataIdMap.get(o.dataId).id,l=i,u=o,{transposed:c,axes:d,originalAxes:h,inputWasTransposed:p}=to(o,s,t),f=d;if(p){let x=t.dataIdMap.get(c.dataId).id;x!==i&&(u=c,l=x,f=_.getInnerMostAxes(f.length,u.shape.length))}_.assertAxesAreInnerMostDims("sum",f,u.shape.length);let[m,g]=_.computeOutAndReduceShapes(u.shape,f),y=k.sizeFromShape(g),A=t.makeOutput(m,u.dtype);if(k.sizeFromShape(u.shape)!==0){let x=t.dataIdMap.get(A.dataId).id;Z9(l,y,x)}if(p&&t.disposeData(c.dataId),a){let x=_.expandShapeToKeepDim(A.shape,h);A.shape=x}return A}var Tve={kernelName:Fl,backendName:"wasm",setupFunc:Ive,kernelFunc:Sve},Nve=$n(Ol),Cve=$n(Pl),Y9;function Eve(e){Y9=e.wasm.cwrap(Lo,null,["number","array","number","array","number","number"])}function $ve(e){let{inputs:t,backend:n,attrs:r}=e,{x:s}=t,a=n.dataIdMap.get(s.dataId).id,{reps:o}=r,i=new Array(s.shape.length);for(let h=0;h<i.length;h++)i[h]=s.shape[h]*o[h];let l=new Uint8Array(new Int32Array(s.shape).buffer),u=new Uint8Array(new Int32Array(i).buffer),c=n.makeOutput(i,s.dtype),d=n.dataIdMap.get(c.dataId).id;return Y9(a,l,s.shape.length,u,i.length,tr[c.dtype],d),c}var Rve={kernelName:Lo,backendName:"wasm",setupFunc:Eve,kernelFunc:$ve},J9;function _ve(e){J9=e.wasm.cwrap(ud,null,["number","array","number","number","number","bool","number","number"])}var Dve=({inputs:e,backend:t,attrs:n})=>{let{x:r}=e,{k:s,sorted:a}=n,o=t.dataIdMap.get(r.dataId).id,i=new Uint8Array(new Int32Array(r.shape).buffer),l=r.shape.slice();l[l.length-1]=s;let u=t.makeOutput(l,r.dtype),c=t.dataIdMap.get(u.dataId).id,d=t.makeOutput(l,"int32"),h=t.dataIdMap.get(d.dataId).id;return J9(o,i,r.shape.length,tr[r.dtype],s,a,c,h),[u,d]},Fve={kernelName:ud,backendName:"wasm",setupFunc:_ve,kernelFunc:Dve},Q9;function Mve(e){Q9=e.wasm.cwrap(cd,null,["number","number","bool","number","number","number","number","number","number","array","number","number","number","number","number"])}function Ove(e){let{backend:t,inputs:n,attrs:r}=e,{image:s,transforms:a}=n,{interpolation:o,fillMode:i,fillValue:l,outputShape:u}=r,[c,d,h,p]=s.shape,[f,m]=u!=null?u:[d,h],g=[c,f,m,p],y=new Uint8Array(new Int32Array(k.computeStrides(s.shape)).buffer),A=t.makeOutput(g,s.dtype),x=t.dataIdMap.get(A.dataId).id,v=t.dataIdMap.get(s.dataId).id,w=t.dataIdMap.get(a.dataId).id,S=o==="nearest"?1:2,E;switch(i){case"constant":E=1;break;case"reflect":E=2;break;case"wrap":E=3;break;case"nearest":E=4;break;default:E=1;break}return Q9(v,w,a.shape[0]>1,c,f,m,p,h,d,y,s.shape.length-1,S,E,l,x),A}var Pve={kernelName:cd,backendName:"wasm",setupFunc:Mve,kernelFunc:Ove};function zve(e){let{inputs:t,backend:n,attrs:r}=e,{value:s}=t,{axis:a}=r;a<0&&(a+=s.shape.length);let o=s.shape[a],i=s.shape.length,l=new Array(i-1),u=0;for(let p=0;p<i;p++)p!==a&&(l[u++]=s.shape[p]);let c=new Array(o),d=new Array(i).fill(0),h=s.shape.slice();h[a]=1;for(let p=0;p<c.length;p++)d[a]=p,c[p]=fh({inputs:{x:s},attrs:{begin:d,size:h},backend:n});return c.map(({dataId:p,dtype:f})=>({dataId:p,dtype:f,shape:l}))}var Lve={kernelName:dd,backendName:"wasm",kernelFunc:zve};function Bve(e){let{inputs:{x:t},backend:n}=e,r=n.makeOutput(t.shape,t.dtype);return n.typedArrayFromHeap(r).fill(0),r}var Wve={kernelName:hd,backendName:"wasm",kernelFunc:Bve},Vve=[Hxe,jxe,Xxe,r5e,o5e,u5e,h5e,g5e,w5e,k5e,I5e,N5e,C5e,R5e,F5e,M5e,O5e,L5e,V5e,G5e,K5e,Z5e,Y5e,J5e,ebe,rbe,sbe,obe,Uxe,ube,hbe,mbe,Abe,vbe,kbe,Sbe,Zxe,Cbe,$be,_be,Dbe,Mbe,zbe,Bbe,Ube,jbe,Xbe,Ybe,e3e,n3e,r3e,o3e,u3e,h3e,f3e,y3e,x3e,v3e,z9,S3e,C3e,R3e,D3e,M3e,O3e,P3e,p5e,B3e,U3e,j3e,K3e,q3e,Y3e,eve,rve,sve,b5e,ive,uve,dve,hve,pve,mve,Ave,vve,kve,Tve,Nve,Cve,Rve,Fve,Pve,e5e,Lve,Wve];for(let e of Vve)ry(e);var ob=re();ob.registerFlag("WASM_HAS_SIMD_SUPPORT",async()=>WebAssembly.validate(new Uint8Array([0,97,115,109,1,0,0,0,1,4,1,96,0,0,3,2,1,0,10,9,1,7,0,65,0,253,15,26,11])));ob.registerFlag("WASM_HAS_MULTITHREAD_SUPPORT",async()=>{if(ob.get("IS_NODE"))return!1;try{return new MessageChannel().port1.postMessage(new SharedArrayBuffer(1)),WebAssembly.validate(new Uint8Array([0,97,115,109,1,0,0,0,1,4,1,96,0,0,3,2,1,0,5,4,1,3,1,1,10,11,1,9,0,65,0,254,16,2,0,26,11]))}catch(e){return!1}});var e$=Xs(y_()),Uve='var Module={};function threadPrintErr(){var text=Array.prototype.slice.call(arguments).join(" ");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;this.alert=threadAlert;Module["instantiateWasm"]=function(info,receiveInstance){var instance=new WebAssembly.Instance(Module["wasmModule"],info);Module["wasmModule"]=null;receiveInstance(instance);return instance.exports};function moduleLoaded(){}this.onmessage=function(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;moduleLoaded()})}else if(e.data.cmd==="objectTransfer"){Module["PThread"].receiveObjectTransfer(e.data)}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);var max=e.data.stackBase;var top=e.data.stackBase+e.data.stackSize;Module["establishStackSpace"](top,max);Module["_emscripten_tls_init"]();Module["PThread"].receiveObjectTransfer(e.data);Module["PThread"].setThreadStatus(Module["_pthread_self"](),1);try{var result=Module["invokeEntryPoint"](e.data.start_routine,e.data.arg);if(!Module["getNoExitRuntime"]())Module["PThread"].threadExit(result)}catch(ex){if(ex==="Canceled!"){Module["PThread"].threadCancel()}else if(ex!="unwind"){if(ex instanceof Module["ExitStatus"]){if(Module["getNoExitRuntime"]()){}else{Module["PThread"].threadExit(ex.status)}}else{Module["PThread"].threadExit(-2);throw ex}}}}else if(e.data.cmd==="cancel"){if(Module["_pthread_self"]()){Module["PThread"].threadCancel()}}else if(e.data.target==="setimmediate"){}else if(e.data.cmd==="processThreadQueue"){if(Module["_pthread_self"]()){Module["_emscripten_current_thread_process_queued_calls"]()}}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);throw ex}};if(typeof process==="object"&&typeof process.versions==="object"&&typeof process.versions.node==="string"){self={location:{href:__filename}};var onmessage=this.onmessage;var nodeWorkerThreads=require("worker_threads");global.Worker=nodeWorkerThreads.Worker;var parentPort=nodeWorkerThreads.parentPort;parentPort.on("message",function(data){onmessage({data:data})});var nodeFS=require("fs");var nodeRead=function(filename){return nodeFS.readFileSync(filename,"utf8")};function globalEval(x){global.require=require;global.Module=Module;eval.call(null,x)}importScripts=function(f){globalEval(nodeRead(f))};postMessage=function(msg){parentPort.postMessage(msg)};if(typeof performance==="undefined"){performance={now:function(){return Date.now()}}}}',Hve=Xs(A_()),t$=class extends Lp{constructor(e){super();this.wasm=e,this.dataIdNextNumber=1,this.wasm.tfjs.init(),this.dataIdMap=new d1(this,Ba())}write(e,t,n){let r={id:this.dataIdNextNumber++};return this.move(r,e,t,n,1),r}numDataIds(){return this.dataIdMap.numDataIds()}async time(e){let t=k.now();return e(),{kernelMs:k.now()-t}}move(e,t,n,r,s){let a=this.dataIdNextNumber++;if(r==="string"){let u=t;this.dataIdMap.set(e,{id:a,stringBytes:u,shape:n,dtype:r,memoryOffset:null,refCount:s});return}let o=k.sizeFromShape(n),i=o*k.bytesPerElement(r),l=this.wasm._malloc(i);this.dataIdMap.set(e,{id:a,memoryOffset:l,shape:n,dtype:r,refCount:s}),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){let{memoryOffset:t,dtype:n,shape:r,stringBytes:s}=this.dataIdMap.get(e);if(n==="string")return s;let a=this.wasm.HEAPU8.slice(t,t+k.sizeFromShape(r)*k.bytesPerElement(n));return qve(a.buffer,n)}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 r;if(n==null)r=this.write(null,e,t);else{let s=this.dataIdNextNumber++;r={id:s},this.dataIdMap.set(r,{id:s,memoryOffset:n,shape:e,dtype:t,refCount:1});let a=k.sizeFromShape(e);this.wasm.tfjs.registerTensor(s,a,n)}return{dataId:r,shape:e,dtype:t}}typedArrayFromHeap({shape:e,dtype:t,dataId:n}){let r=this.wasm.HEAPU8.buffer,{memoryOffset:s}=this.dataIdMap.get(n),a=k.sizeFromShape(e);switch(t){case"float32":return new Float32Array(r,s,a);case"int32":return new Int32Array(r,s,a);case"bool":return new Uint8Array(r,s,a);default:throw new Error(`Unknown dtype ${t}`)}}};function Gve(e){return(t,n)=>(k.fetch(e,{credentials:"same-origin"}).then(r=>{r.ok||t.env.a(`failed to load wasm binary file at '${e}'`),r.arrayBuffer().then(s=>{WebAssembly.instantiate(s,t).then(a=>{n(a.instance,a.module)})})}),{})}function n$(e,t,n){if(Ym!=null)return Ym;let r="tfjs-backend-wasm.wasm";return e&&t?r="tfjs-backend-wasm-threaded-simd.wasm":e&&(r="tfjs-backend-wasm-simd.wasm"),gh!=null&&gh[r]!=null?gh[r]:n+r}async function jve(){let[e,t]=await Promise.all([re().getAsync("WASM_HAS_SIMD_SUPPORT"),re().getAsync("WASM_HAS_MULTITHREAD_SUPPORT")]);return new Promise((n,r)=>{let s={};s.locateFile=(i,l)=>{if(i.endsWith(".worker.js")){let u=Uve,c=new Blob([u],{type:"application/javascript"});return URL.createObjectURL(c)}return i.endsWith(".wasm")?n$(e,t,mh!=null?mh:l):l+i},ib&&(s.instantiateWasm=Gve(n$(e,t,mh!=null?mh:"")));let a=!1;s.onAbort=()=>{if(a||yh)return;yh=!0,r({message:"Make sure the server can serve the `.wasm` file relative to the bundled js file. For more details see https://github.com/tensorflow/tfjs/blob/master/tfjs-backend-wasm/README.md#using-bundlers"})};let o;t&&e&&Ym==null?(s.mainScriptUrlOrBlob=new Blob(["var WasmBackendModuleThreadedSimd = "+e$.default.toString()],{type:"text/javascript"}),o=(0,e$.default)(s)):o=(0,Hve.default)(s),o.then(i=>{a=!0,yh=!1;let l=null;i.tfjs={init:i.cwrap("init",null,[]),registerTensor:i.cwrap("register_tensor",null,["number","number","number"]),disposeData:i.cwrap("dispose_data",l,["number"]),dispose:i.cwrap("dispose",l,[])},n({wasm:i})})})}function qve(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 Kve=["tfjs-backend-wasm.wasm","tfjs-backend-wasm-simd.wasm","tfjs-backend-wasm-threaded-simd.wasm"],Ym=null,mh=null,gh={},yh=!1,ib=!1;function Xve(e,t=!1){if(M6("setWasmPath has been deprecated in favor of setWasmPaths and will be removed in a future release."),yh)throw new Error("The WASM backend was already initialized. Make sure you call `setWasmPath()` before you call `tf.setBackend()` or `tf.ready()`");Ym=e,ib=t}function Zve(e,t=!1){if(yh)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")mh=e;else{gh=e;let n=Kve.filter(r=>gh[r]==null);if(n.length>0)throw new Error(`There were no entries found for the following binaries: ${n.join(",")}. Please either call setWasmPaths with a map providing a path for each binary, or with a string indicating the directory where all the binaries can be found.`)}ib=t}var Yve="3.8.0",Jve=2;Cy("wasm",async()=>{let{wasm:e}=await jve();return new t$(e)},Jve);var Qve={tfjs:x_,"tfjs-core":b_,"tfjs-data":v_,"tfjs-layers":w_,"tfjs-converter":k_,"tfjs-backend-cpu":I_,"tfjs-backend-webgl":S_,"tfjs-backend-wasm":T_};var rr={name:"humangl",priority:99,canvas:null,gl:null,width:1024,height:1024,extensions:[],webGLattr:{alpha:!1,antialias:!1,premultipliedAlpha:!1,preserveDrawingBuffer:!1,depth:!1,stencil:!1,failIfMajorPerformanceCaveat:!1,desynchronized:!0}};function ewe(){let e=rr.gl;!e||(rr.extensions=e.getSupportedExtensions())}function r$(){if(!H2(rr.name)){try{rr.canvas=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(rr.width,rr.height):document.createElement("canvas")}catch(e){fe("error: cannot create canvas:",e);return}try{rr.gl=rr.canvas.getContext("webgl2",rr.webGLattr)}catch(e){fe("error: cannot get WebGL2 context:",e);return}try{Nm(2,rr.gl)}catch(e){fe("error: cannot set WebGL2 context:",e);return}try{let e=new Fm(rr.gl);G2(rr.name,()=>new uh(e),rr.priority)}catch(e){fe("error: cannot register WebGL backend:",e);return}try{zi("webgl").forEach(t=>{let n={...t,backendName:rr.name};rp(n)})}catch(e){fe("error: cannot update WebGL backend registration:",e);return}try{kr.set("WEBGL_VERSION",2)}catch(e){fe("error: cannot set WebGL backend flags:",e);return}ewe(),fe("backend registered:",rr.name)}}function s$(e,t){let n=[e.startPoint[0]*t[0],e.startPoint[1]*t[1]],r=[e.endPoint[0]*t[0],e.endPoint[1]*t[1]];return{startPoint:n,endPoint:r}}function xh(e){return[Math.abs(e.endPoint[0]-e.startPoint[0]),Math.abs(e.endPoint[1]-e.startPoint[1])]}function Su(e){return[e.startPoint[0]+(e.endPoint[0]-e.startPoint[0])/2,e.startPoint[1]+(e.endPoint[1]-e.startPoint[1])/2]}function Tu(e,t,n){let r=t.shape[1],s=t.shape[2],a=[[e.startPoint[1]/r,e.startPoint[0]/s,e.endPoint[1]/r,e.endPoint[0]/s]];return Ze.cropAndResize(t,a,[0],n)}function Jm(e,t=1.5){let n=Su(e),r=xh(e),s=[t*r[0]/2,t*r[1]/2],a=[n[0]-s[0],n[1]-s[1]],o=[n[0]+s[0],n[1]+s[1]];return{startPoint:a,endPoint:o,landmarks:e.landmarks}}function Qm(e){let t=Su(e),n=xh(e),s=Math.max(...n)/2,a=[Math.round(t[0]-s),Math.round(t[1]-s)],o=[Math.round(t[0]+s),Math.round(t[1]+s)];return{startPoint:a,endPoint:o,landmarks:e.landmarks}}function lb(e){let t=e.map(a=>a[0]),n=e.map(a=>a[1]),r=[Math.min(...t),Math.min(...n)],s=[Math.max(...t),Math.max(...n)];return{startPoint:r,endPoint:s,landmarks:e}}var a$=e=>({startPoint:Xe(e,[0,0],[-1,2]),endPoint:Xe(e,[0,2],[-1,2])});var e0=[[1,0,0],[0,1,0],[0,0,1]];function twe(e){return e-2*Math.PI*Math.floor((e+Math.PI)/(2*Math.PI))}function ub(e,t){let n=Math.PI/2-Math.atan2(-(t[1]-e[1]),t[0]-e[0]);return twe(n)}function o$(e,t){return[[1,0,e],[0,1,t],[0,0,1]]}function no(e,t){let n=0;for(let r=0;r<e.length;r++)n+=e[r]*t[r];return n}function nwe(e,t){let n=[];for(let r=0;r<e.length;r++)n.push(e[r][t]);return n}function i$(e,t){let n=[],r=e.length;for(let s=0;s<r;s++){n.push([]);for(let a=0;a<r;a++)n[s].push(no(e[s],nwe(t,a)))}return n}function t0(e,t){let n=Math.cos(e),r=Math.sin(e),s=[[n,-r,0],[r,n,0],[0,0,1]],a=o$(t[0],t[1]),o=i$(a,s),i=o$(-t[0],-t[1]);return i$(o,i)}function l$(e){let t=[[e[0][0],e[1][0]],[e[0][1],e[1][1]]],n=[e[0][2],e[1][2]],r=[-no(t[0],n),-no(t[1],n)];return[t[0].concat(r[0]),t[1].concat(r[1]),[0,0,1]]}function u$(e,t){return[no(e,t[0]),no(e,t[1])]}function c$(e){let t={strides:[e/16,e/8],anchors:[2,6]},n=[];for(let r=0;r<t.strides.length;r++){let s=t.strides[r],a=Math.floor((e+s-1)/s),o=Math.floor((e+s-1)/s),i=t.anchors[r];for(let l=0;l<a;l++){let u=s*(l+.5);for(let c=0;c<o;c++){let d=s*(c+.5);for(let h=0;h<i;h++)n.push([d,u])}}}return n}var d$=6;function rwe(e,t,n){let r=Xe(e,[0,1],[-1,2]),s=Me(r,t),a=Xe(e,[0,3],[-1,2]),o=Je(a,n),i=Je(s,n),l=Je(o,2),u=Ue(i,l),c=Me(i,l),d=pe(u,n),h=pe(c,n);return ic([d,h],1)}var h$=class{constructor(t,n){this.model=t,this.anchorsData=c$(t.inputs[0].shape[1]),this.anchors=sa(this.anchorsData),this.inputSize=t.inputs[0].shape[2],this.config=n}async getBoundingBoxes(t,n){if(!t||t.isDisposedInternal||t.shape.length!==4||t.shape[1]<1||t.shape[2]<1)return null;let[r,s,a]=Ve(()=>{let c=Ze.resizeBilinear(t,[this.inputSize,this.inputSize]).div(127.5).sub(.5),d=this.model.execute(c),h;if(Array.isArray(d)){let g=d.sort((b,v)=>b.size-v.size),y=rn([g[0],g[2]],2),A=rn([g[1],g[3]],2);h=rn([A,y],1).squeeze(0)}else h=Xn(d);let p=rwe(h,this.anchors,[this.inputSize,this.inputSize]),f=Xe(h,[0,0],[-1,1]),m=Is(f).squeeze().dataSync();return[h,p,m]});this.config=Fn(this.config,n);let o=await Ze.nonMaxSuppressionAsync(s,a,this.config.face.detector.maxDetected,this.config.face.detector.iouThreshold,this.config.face.detector.minConfidence),i=o.arraySync();o.dispose();let l=[];for(let u=0;u<i.length;u++){let c=a[i[u]];if(c>this.config.face.detector.minConfidence){let d=Xe(s,[i[u],0],[1,-1]),h=a$(d);d.dispose();let p=this.anchorsData[i[u]],f=Ve(()=>Xe(r,[i[u],d$-1],[1,-1]).squeeze().reshape([d$,-1]));l.push({box:h,landmarks:f,anchor:p,confidence:c})}}return r.dispose(),s.dispose(),{boxes:l,scaleFactor:[t.shape[2]/this.inputSize,t.shape[1]/this.inputSize]}}};async function p$(e){let t=await Nt(Ct(e.modelBasePath,e.face.detector.modelPath),{fromTFHub:e.face.detector.modelPath.includes("tfhub.dev")}),n=new h$(t,e);return!t||!t.modelUrl?fe("load model failed:",e.face.detector.modelPath):e.debug&&fe("load model:",t.modelUrl),n}var Ws={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:[61,185,40,39,37,0,267,269,270,409,291],lipsLowerOuter:[146,91,181,84,17,314,405,321,375,291],lipsUpperInner:[78,191,80,81,82,13,312,311,310,415,308],lipsLowerInner:[78,95,88,178,87,14,317,402,318,324,308],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]},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]}],bh=[[.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]],bi=[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 swe=[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],awe=[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],owe=[33,133,362,263,1,78,308],G7e=swe.map(e=>bh[e]),j7e=awe.map(e=>bh[e]),q7e=owe.map(e=>bh[e]);var db=Ws.leftEyeLower0,hb=Ws.rightEyeLower0,Nu={leftBounds:[db[0],db[db.length-1]],rightBounds:[hb[0],hb[hb.length-1]]},n0={count:468,mouth:13,symmetryLine:[13,Ws.midwayBetweenEyes[0]]},f$={leftEye:0,rightEye:1,nose:2,mouth:3,leftEar:4,rightEar:5,symmetryLine:[3,2]},Cu={upperCenter:3,lowerCenter:4,index:71,numCoordinates:76};function r0(e,t,n,r){for(let s=0;s<cb.length;s++){let{key:a,indices:o}=cb[s],i=Ws[`${n}${a}`];if(!r||r.includes(a))for(let l=0;l<o.length;l++){let u=o[l];e[i[l]]=[t[u][0],t[u][1],(t[u][2]+e[i[l]][2])/2]}}}var pb=class{constructor(t,n,r){var s,a;this.storedBoxes=[],this.boundingBoxDetector=t,this.meshDetector=n,this.irisModel=r,this.boxSize=((s=t==null?void 0:t.model)==null?void 0:s.inputs[0].shape[2])||0,this.meshSize=(n==null?void 0:n.inputs[0].shape[2])||((a=t==null?void 0:t.model)==null?void 0:a.inputs[0].shape[2]),this.irisSize=(r==null?void 0:r.inputs[0].shape[1])||0,this.irisEnlarge=2.3,this.skipped=0,this.detectedFaces=0}transformRawCoords(t,n,r,s){let a=xh({startPoint:n.startPoint,endPoint:n.endPoint}),o=t.map(d=>[a[0]/this.meshSize*(d[0]-this.meshSize/2),a[1]/this.meshSize*(d[1]-this.meshSize/2),d[2]]),i=r!==0?t0(r,[0,0]):e0,l=r!==0?o.map(d=>[...u$(d,i),d[2]]):o,u=r!==0?l$(s):e0,c=[...Su({startPoint:n.startPoint,endPoint:n.endPoint}),1];return l.map(d=>[Math.round(d[0]+no(c,u[0])),Math.round(d[1]+no(c,u[1])),Math.round(d[2])])}getLeftToRightEyeDepthDifference(t){let n=t[Nu.leftBounds[0]][2],r=t[Nu.rightBounds[0]][2];return n-r}getEyeBox(t,n,r,s,a=!1){let o=Qm(Jm(lb([t[r],t[s]]),this.irisEnlarge)),i=xh(o),l=Ze.cropAndResize(n,[[o.startPoint[1]/this.meshSize,o.startPoint[0]/this.meshSize,o.endPoint[1]/this.meshSize,o.endPoint[0]/this.meshSize]],[0],[this.irisSize,this.irisSize]);return a&&kr.flags.IS_BROWSER&&(l=Ze.flipLeftRight(l)),{box:o,boxSize:i,crop:l}}getEyeCoords(t,n,r,s=!1){let a=[];for(let o=0;o<Cu.numCoordinates;o++){let i=t[o*3],l=t[o*3+1],u=t[o*3+2];a.push([(s?1-i/this.irisSize:i/this.irisSize)*r[0]+n.startPoint[0],l/this.irisSize*r[1]+n.startPoint[1],u])}return{rawCoords:a,iris:a.slice(Cu.index)}}getAdjustedIrisCoords(t,n,r){let s=t[Ws[`${r}EyeUpper0`][Cu.upperCenter]][2],a=t[Ws[`${r}EyeLower0`][Cu.lowerCenter]][2],o=(s+a)/2;return n.map((i,l)=>{let u=o;return l===2?u=s:l===4&&(u=a),[i[0],i[1],u]})}async predict(t,n){let r=!1,s;if((this.skipped===0||this.skipped>n.face.detector.skipFrames||!n.face.mesh.enabled||!n.skipFrame)&&(s=await this.boundingBoxDetector.getBoundingBoxes(t,n),this.skipped=0),n.skipFrame&&this.skipped++,!n.skipFrame||s&&s.boxes&&(!n.face.mesh.enabled||s.boxes.length!==this.detectedFaces&&this.detectedFaces!==n.face.detector.maxDetected)){this.storedBoxes=[],this.detectedFaces=0;for(let o of s.boxes)this.storedBoxes.push({startPoint:o.box.startPoint.dataSync(),endPoint:o.box.endPoint.dataSync(),landmarks:o.landmarks.arraySync(),confidence:o.confidence});this.storedBoxes.length>0&&(r=!0)}if(r){if(!s||!s.boxes||s.boxes.length===0)return this.storedBoxes=[],this.detectedFaces=0,null;for(let o=0;o<this.storedBoxes.length;o++){let i=s$({startPoint:this.storedBoxes[o].startPoint,endPoint:this.storedBoxes[o].endPoint},s.scaleFactor),l=Jm(i),u=Qm(l),c=this.storedBoxes[o].landmarks,d=this.storedBoxes[o].confidence;this.storedBoxes[o]={...u,confidence:d,landmarks:c}}}s&&s.boxes&&s.boxes.forEach(o=>{o.box.startPoint.dispose(),o.box.endPoint.dispose(),o.landmarks.dispose()});let a=Ve(()=>this.storedBoxes.map((o,i)=>{let l,u=0,c;if(n.face.detector.rotation&&n.face.mesh.enabled&&kr.flags.IS_BROWSER){let[x,b]=o.landmarks.length>=n0.count?n0.symmetryLine:f$.symmetryLine;u=ub(o.landmarks[x],o.landmarks[b]);let v=Su({startPoint:o.startPoint,endPoint:o.endPoint}),I=[v[0]/t.shape[2],v[1]/t.shape[1]],w=Ze.rotateWithOffset(t,u,0,I);c=t0(-u,v),n.face.mesh.enabled?l=Tu({startPoint:o.startPoint,endPoint:o.endPoint},w,[this.meshSize,this.meshSize]).div(255):l=Tu({startPoint:o.startPoint,endPoint:o.endPoint},w,[this.boxSize,this.boxSize]).div(255)}else{c=e0;let x=t.clone();n.face.mesh.enabled?l=Tu({startPoint:o.startPoint,endPoint:o.endPoint},x,[this.meshSize,this.meshSize]).div(255):l=Tu({startPoint:o.startPoint,endPoint:o.endPoint},x,[this.boxSize,this.boxSize]).div(255)}if(!n.face.mesh.enabled)return{mesh:[],box:o,faceConfidence:null,boxConfidence:o.confidence,confidence:o.confidence,image:l};let[,d,h]=this.meshDetector.execute(l),p=d.dataSync()[0];if(p<n.face.detector.minConfidence)return this.storedBoxes[i].confidence=p,null;let m=le(h,[-1,3]).arraySync();if(n.face.iris.enabled){let{box:x,boxSize:b,crop:v}=this.getEyeBox(m,l,Nu.leftBounds[0],Nu.leftBounds[1],!0),{box:I,boxSize:w,crop:S}=this.getEyeBox(m,l,Nu.rightBounds[0],Nu.rightBounds[1]),D=this.irisModel.predict(rn([v,S])).dataSync(),$=D.slice(0,Cu.numCoordinates*3),{rawCoords:R,iris:N}=this.getEyeCoords($,x,b,!0),M=D.slice(Cu.numCoordinates*3),{rawCoords:B,iris:q}=this.getEyeCoords(M,I,w),X=this.getLeftToRightEyeDepthDifference(m);Math.abs(X)<30?(r0(m,R,"left",null),r0(m,B,"right",null)):X<1?r0(m,R,"left",["EyeUpper0","EyeLower0"]):r0(m,B,"right",["EyeUpper0","EyeLower0"]);let J=this.getAdjustedIrisCoords(m,N,"left"),ee=this.getAdjustedIrisCoords(m,q,"right");m=m.concat(J).concat(ee)}let g=this.transformRawCoords(m,o,u,c),y=o.confidence;if(o=Jm(lb(g),1.5),o.confidence=y,n.face.detector.rotation&&n.face.mesh.enabled&&n.face.description.enabled&&kr.flags.IS_BROWSER){let[x,b]=o.landmarks.length>=n0.count?n0.symmetryLine:f$.symmetryLine;u=ub(o.landmarks[x],o.landmarks[b]);let v=Su({startPoint:o.startPoint,endPoint:o.endPoint}),I=[v[0]/t.shape[2],v[1]/t.shape[1]],w=Ze.rotateWithOffset(t.toFloat(),u,0,I);c=t0(-u,v),l=Tu({startPoint:o.startPoint,endPoint:o.endPoint},w,[this.meshSize,this.meshSize]).div(255)}let A={mesh:g,box:o,faceConfidence:p,boxConfidence:o.confidence,image:l};return this.storedBoxes[i]={...Qm(o),confidence:o.confidence,faceConfidence:p},A}));return n.face.mesh.enabled&&(this.storedBoxes=this.storedBoxes.filter(o=>o.confidence>n.face.detector.minConfidence)),this.detectedFaces=a.length,a}};var Kt=[null,null,null],fb;async function m$(e,t){let n=await fb.predict(e,t),r=[],s=0;for(let a of n||[]){if(!a||a.isDisposedInternal)continue;let o=a.mesh.map(c=>[c[0]/(e.shape[2]||0),c[1]/(e.shape[1]||0),c[2]/fb.meshSize]),i={};if(a.mesh&&a.mesh.length>0)for(let c of Object.keys(Ws))i[c]=Ws[c].map(d=>a.mesh[d]);let l=a.box?[Math.trunc(Math.max(0,a.box.startPoint[0])),Math.trunc(Math.max(0,a.box.startPoint[1])),Math.trunc(Math.min(e.shape[2]||0,a.box.endPoint[0])-Math.max(0,a.box.startPoint[0])),Math.trunc(Math.min(e.shape[1]||0,a.box.endPoint[1])-Math.max(0,a.box.startPoint[1]))]:[0,0,0,0],u=a.box?[a.box.startPoint[0]/(e.shape[2]||0),a.box.startPoint[1]/(e.shape[1]||0),(a.box.endPoint[0]-a.box.startPoint[0])/(e.shape[2]||0),(a.box.endPoint[1]-a.box.startPoint[1])/(e.shape[1]||0)]:[0,0,0,0];r.push({id:s++,score:Math.round(100*a.faceConfidence||100*a.boxConfidence||0)/100,boxScore:Math.round(100*a.boxConfidence)/100,faceScore:Math.round(100*a.faceConfidence)/100,box:l,boxRaw:u,mesh:a.mesh,meshRaw:o,annotations:i,image:a.image,tensor:a.image}),a.coords&&a.coords.dispose()}return r}async function mb(e){return!Kt[0]&&e.face.enabled||!Kt[1]&&e.face.mesh.enabled||!Kt[2]&&e.face.iris.enabled?(Kt=await Promise.all([!Kt[0]&&e.face.enabled?p$(e):null,!Kt[1]&&e.face.mesh.enabled?Nt(Ct(e.modelBasePath,e.face.mesh.modelPath),{fromTFHub:e.face.mesh.modelPath.includes("tfhub.dev")}):null,!Kt[2]&&e.face.iris.enabled?Nt(Ct(e.modelBasePath,e.face.iris.modelPath),{fromTFHub:e.face.iris.modelPath.includes("tfhub.dev")}):null]),e.face.mesh.enabled&&(!Kt[1]||!Kt[1].modelUrl?fe("load model failed:",e.face.mesh.modelPath):e.debug&&fe("load model:",Kt[1].modelUrl)),e.face.iris.enabled&&(!Kt[2]||!Kt[2].modelUrl?fe("load model failed:",e.face.iris.modelPath):e.debug&&fe("load model:",Kt[2].modelUrl))):e.debug&&(Kt[0]&&fe("cached model:",Kt[0].model.modelUrl),Kt[1]&&fe("cached model:",Kt[1].modelUrl),Kt[2]&&fe("cached model:",Kt[2].modelUrl)),fb=new pb(Kt[0],Kt[1],Kt[2]),Kt}var g$=bi,y$=bh;var ys,s0=[],A$=0,gb=Number.MAX_SAFE_INTEGER;async function yb(e){let t=Ct(e.modelBasePath,e.face.description.modelPath);return ys?e.debug&&fe("cached model:",t):(ys=await Nt(t),ys?e.debug&&fe("load model:",t):fe("load model failed:",e.face.description.modelPath)),ys}function Ab(e,t,n=2){if(!e||!t||(e==null?void 0:e.length)===0||(t==null?void 0:t.length)===0||(e==null?void 0:e.length)!==(t==null?void 0:t.length))return 0;let r=5*e.map((a,o)=>Math.abs(e[o]-t[o])**n).reduce((a,o)=>a+o,0)**(1/n);return Math.max(0,100-r)/100}function x$(e,t,n=0){let r={similarity:0,name:"",source:"",embedding:[]};if(!e||!t||!Array.isArray(e)||!Array.isArray(t))return r;for(let s of t)if(s.embedding&&s.name){let a=Ab(e,s.embedding);a>n&&a>r.similarity&&(r={...s,similarity:a})}return r}function xb(e){return Ve(()=>{let n=e.image||e.tensor||e;if(!(n instanceof It))return null;let r=[[.05,.15,.85,.85]];return ys.inputs[0].shape?(n.shape.length===3?Ze.cropAndResize(ta(n,0),r,[0],[ys.inputs[0].shape[2],ys.inputs[0].shape[1]]):Ze.cropAndResize(n,r,[0],[ys.inputs[0].shape[2],ys.inputs[0].shape[1]])).mul(255):null})}async function bb(e,t,n,r){var s,a;return ys?gb<t.face.description.skipFrames&&t.skipFrame&&A$===r&&((s=s0[n])==null?void 0:s.age)&&((a=s0[n])==null?void 0:a.age)>0?(gb++,s0[n]):(gb=0,new Promise(async o=>{let i=xb(e),l,u={age:0,gender:"unknown",genderScore:0,descriptor:[]};t.face.description.enabled&&(l=await ys.predict(i)),We(i),l&&(Ve(()=>{let c=l.find(m=>m.shape[1]===1).dataSync(),d=Math.trunc(200*Math.abs(c[0]-.5))/100;d>t.face.description.minConfidence&&(u.gender=c[0]<=.5?"female":"male",u.genderScore=Math.min(.99,d));let h=l.find(m=>m.shape[1]===100).argMax(1).dataSync()[0],p=l.find(m=>m.shape[1]===100).dataSync();u.age=Math.round(p[h-1]>p[h+1]?10*h-100*p[h-1]:10*h+100*p[h+1])/10;let f=l.find(m=>m.shape[1]===1024);u.descriptor=[...f.dataSync()]}),l.forEach(c=>We(c))),s0[n]=u,A$=r,o(u)})):null}var iwe=["angry","disgust","fear","happy","sad","surprise","neutral"],As,a0=[],b$=0,vb=Number.MAX_SAFE_INTEGER,wb=[.2989,.587,.114];async function kb(e){return As?e.debug&&fe("cached model:",As.modelUrl):(As=await Nt(Ct(e.modelBasePath,e.face.emotion.modelPath)),!As||!As.modelUrl?fe("load model failed:",e.face.emotion.modelPath):e.debug&&fe("load model:",As.modelUrl)),As}async function Ib(e,t,n,r){return As?vb<t.face.emotion.skipFrames&&t.skipFrame&&b$===r&&a0[n]&&a0[n].length>0?(vb++,a0[n]):(vb=0,new Promise(async s=>{let a=Ze.resizeBilinear(e,[As.inputs[0].shape[2],As.inputs[0].shape[1]],!1),[o,i,l]=na(a,3,3);a.dispose();let u=pe(o,wb[0]),c=pe(i,wb[1]),d=pe(l,wb[2]);o.dispose(),i.dispose(),l.dispose();let h=j2([u,c,d]);u.dispose(),c.dispose(),d.dispose();let p=Ve(()=>h.sub(.5).mul(2));h.dispose();let f=[];if(t.face.emotion.enabled){let m=await As.predict(p),g=m.dataSync();We(m);for(let y=0;y<g.length;y++)g[y]>t.face.emotion.minConfidence&&f.push({score:Math.min(.99,Math.trunc(100*g[y])/100),emotion:iwe[y]});f.sort((y,A)=>A.score-y.score)}p.dispose(),a0[n]=f,b$=r,s(f)})):null}var vh=["nose","leftEye","rightEye","leftEar","rightEar","leftShoulder","rightShoulder","leftElbow","rightElbow","leftWrist","rightWrist","leftHip","rightHip","leftKnee","rightKnee","leftAnkle","rightAnkle"],v$=vh.length,wh=vh.reduce((e,t,n)=>(e[t]=n,e),{}),lwe=[["leftHip","leftShoulder"],["leftElbow","leftShoulder"],["leftElbow","leftWrist"],["leftHip","leftKnee"],["leftKnee","leftAnkle"],["rightHip","rightShoulder"],["rightElbow","rightShoulder"],["rightElbow","rightWrist"],["rightHip","rightKnee"],["rightKnee","rightAnkle"],["leftShoulder","rightShoulder"],["leftHip","rightHip"]],uwe=lwe.map(([e,t])=>[wh[e],wh[t]]),w$=[["nose","leftEye"],["leftEye","leftEar"],["nose","rightEye"],["rightEye","rightEar"],["nose","leftShoulder"],["leftShoulder","leftElbow"],["leftElbow","leftWrist"],["leftShoulder","leftHip"],["leftHip","leftKnee"],["leftKnee","leftAnkle"],["nose","rightShoulder"],["rightShoulder","rightElbow"],["rightElbow","rightWrist"],["rightShoulder","rightHip"],["rightHip","rightKnee"],["rightKnee","rightAnkle"]];function k$(e){let t=e.reduce(({maxX:n,maxY:r,minX:s,minY:a},{position:{x:o,y:i}})=>({maxX:Math.max(n,o),maxY:Math.max(r,i),minX:Math.min(s,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 I$(e,[t,n],[r,s]){let a=t/r,o=n/s,i=(u,c)=>({id:c,score:u.score,boxRaw:[u.box[0]/s,u.box[1]/r,u.box[2]/s,u.box[3]/r],box:[Math.trunc(u.box[0]*o),Math.trunc(u.box[1]*a),Math.trunc(u.box[2]*o),Math.trunc(u.box[3]*a)],keypoints:u.keypoints.map(({score:d,part:h,position:p})=>({score:d,part:h,position:[Math.trunc(p.x*o),Math.trunc(p.y*a)],positionRaw:[p.x/r,p.y/r]}))});return e.map((u,c)=>i(u,c))}var Sb=class{constructor(t,n){this.priorityQueue=new Array(t),this.numberOfElements=-1,this.getElementValue=n}enqueue(t){this.priorityQueue[++this.numberOfElements]=t,this.swim(this.numberOfElements)}dequeue(){let t=this.priorityQueue[0];return this.exchange(0,this.numberOfElements--),this.sink(0),this.priorityQueue[this.numberOfElements+1]=null,t}empty(){return this.numberOfElements===-1}size(){return this.numberOfElements+1}all(){return this.priorityQueue.slice(0,this.numberOfElements+1)}max(){return this.priorityQueue[0]}swim(t){for(;t>0&&this.less(Math.floor(t/2),t);)this.exchange(t,Math.floor(t/2)),t=Math.floor(t/2)}sink(t){for(;2*t<=this.numberOfElements;){let n=2*t;if(n<this.numberOfElements&&this.less(n,n+1)&&n++,!this.less(t,n))break;this.exchange(t,n),t=n}}getValueAt(t){return this.getElementValue(this.priorityQueue[t])}less(t,n){return this.getValueAt(t)<this.getValueAt(n)}exchange(t,n){let r=this.priorityQueue[t];this.priorityQueue[t]=this.priorityQueue[n],this.priorityQueue[n]=r}};function Tb(e,t,n,r){return{y:r.get(e,t,n),x:r.get(e,t,n+v$)}}function Nb(e,t,n){let{heatmapY:r,heatmapX:s,id:a}=e,{y:o,x:i}=Tb(r,s,a,n);return{x:e.heatmapX*t+i,y:e.heatmapY*t+o}}function Cb(e,t,n){return e<t?t:e>n?n:e}function S$(e,t,n,r){let s=n-e,a=r-t;return s*s+a*a}function Eb(e,t){return{x:e.x+t.x,y:e.y+t.y}}var o0=1,Eu=16,cwe=50**2;function T$(e,t,n,r,s,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,A,x)=>({y:Cb(Math.round(y.y/Eu),0,A-1),x:Cb(Math.round(y.x/Eu),0,x-1)}),[u,c]=r.shape,d=l(t.position,u,c),h=i(d),f=Eb(t.position,h);for(let y=0;y<o;y++){let A=l(f,u,c),x=Tb(A.y,A.x,n,s);f=Eb({x:A.x*Eu,y:A.y*Eu},{x:x.x,y:x.y})}let m=l(f,u,c),g=r.get(m.y,m.x,n);return{position:f,part:vh[n],score:g}}function dwe(e,t,n,r,s){let a=w$.map(([h,p])=>[wh[h],wh[p]]),o=a.map(([,h])=>h),i=a.map(([h])=>h),l=t.shape[2],u=o.length,c=new Array(l),d=Nb(e.part,Eu,n);c[e.part.id]={score:e.score,part:vh[e.part.id],position:d};for(let h=u-1;h>=0;--h){let p=o[h],f=i[h];c[p]&&!c[f]&&(c[f]=T$(h,c[p],f,t,n,s))}for(let h=0;h<u;++h){let p=i[h],f=o[h];c[p]&&!c[f]&&(c[f]=T$(h,c[p],f,t,n,r))}return c}function hwe(e,t,n,r,s){let[a,o]=s.shape,i=!0,l=Math.max(n-o0,0),u=Math.min(n+o0+1,a);for(let c=l;c<u;++c){let d=Math.max(r-o0,0),h=Math.min(r+o0+1,o);for(let p=d;p<h;++p)if(s.get(c,p,e)>t){i=!1;break}if(!i)break}return i}function pwe(e,t){let[n,r,s]=t.shape,a=new Sb(n*r*s,({score:o})=>o);for(let o=0;o<n;++o)for(let i=0;i<r;++i)for(let l=0;l<s;++l){let u=t.get(o,i,l);u<e||hwe(l,u,o,i,t)&&a.enqueue({score:u,part:{heatmapY:o,heatmapX:i,id:l}})}return a}function N$(e,{x:t,y:n},r){return e.some(({keypoints:s})=>{var o;let a=(o=s[r])==null?void 0:o.position;return a?S$(n,t,a.y,a.x)<=cwe:!1})}function fwe(e,t){return t.reduce((r,{position:s,score:a},o)=>(N$(e,s,o)||(r+=a),r),0)/t.length}function C$(e,t,n,r,s,a){let o=[],i=pwe(a,t);for(;o.length<s&&!i.empty();){let l=i.dequeue(),u=Nb(l.part,Eu,e);if(N$(o,u,l.part.id))continue;let c=dwe(l,t,e,n,r);c=c.filter(p=>p.score>a);let d=fwe(o,c),h=k$(c);d>a&&o.push({keypoints:c,box:h,score:Math.round(100*d)/100})}return o}var mr,mwe=["MobilenetV1/offset_2/BiasAdd","MobilenetV1/heatmap_2/BiasAdd","MobilenetV1/displacement_fwd_2/BiasAdd","MobilenetV1/displacement_bwd_2/BiasAdd"];async function $b(e,t){let n=Ve(()=>{if(!mr.inputs[0].shape)return[];let i=Ze.resizeBilinear(e,[mr.inputs[0].shape[2],mr.inputs[0].shape[1]]).toFloat().div(127.5).sub(1),u=mr.execute(i,mwe).map(c=>Xn(c,[0]));return u[1]=u[1].sigmoid(),u}),r=await Promise.all(n.map(o=>o.buffer()));for(let o of n)o.dispose();let s=await C$(r[0],r[1],r[2],r[3],t.body.maxDetected,t.body.minConfidence);return mr.inputs[0].shape?I$(s,[e.shape[1],e.shape[2]],[mr.inputs[0].shape[2],mr.inputs[0].shape[1]]):[]}async function Rb(e){return mr?e.debug&&fe("cached model:",mr.modelUrl):(mr=await Nt(Ct(e.modelBasePath,e.body.modelPath)),!mr||!mr.modelUrl?fe("load model failed:",e.body.modelPath):e.debug&&fe("load model:",mr.modelUrl)),mr}function i0(e){return[Math.abs(e.endPoint[0]-e.startPoint[0]),Math.abs(e.endPoint[1]-e.startPoint[1])]}function kh(e){return[e.startPoint[0]+(e.endPoint[0]-e.startPoint[0])/2,e.startPoint[1]+(e.endPoint[1]-e.startPoint[1])/2]}function E$(e,t,n){let r=t.shape[1],s=t.shape[2],a=[[e.startPoint[1]/r,e.startPoint[0]/s,e.endPoint[1]/r,e.endPoint[0]/s]];return Ze.cropAndResize(t,a,[0],n)}function $$(e,t){let n=[e.startPoint[0]*t[0],e.startPoint[1]*t[1]],r=[e.endPoint[0]*t[0],e.endPoint[1]*t[1]],s=e.palmLandmarks.map(a=>[a[0]*t[0],a[1]*t[1]]);return{startPoint:n,endPoint:r,palmLandmarks:s,confidence:e.confidence}}function l0(e,t=1.5){let n=kh(e),r=i0(e),s=[t*r[0]/2,t*r[1]/2],a=[n[0]-s[0],n[1]-s[1]],o=[n[0]+s[0],n[1]+s[1]];return{startPoint:a,endPoint:o,palmLandmarks:e.palmLandmarks}}function u0(e){let t=kh(e),n=i0(e),s=Math.max(...n)/2,a=[t[0]-s,t[1]-s],o=[t[0]+s,t[1]+s];return{startPoint:a,endPoint:o,palmLandmarks:e.palmLandmarks}}var R$=[{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 _b=class{constructor(t){var n;this.model=t,this.anchors=R$.map(r=>[r.x,r.y]),this.anchorsTensor=sa(this.anchors),this.inputSize=(n=this.model)==null?void 0:n.inputs[0].shape[2],this.inputSizeTensor=ur([this.inputSize,this.inputSize]),this.doubleInputSizeTensor=ur([this.inputSize*2,this.inputSize*2])}normalizeBoxes(t){return Ve(()=>{let n=Xe(t,[0,0],[-1,2]),r=Xe(t,[0,2],[-1,2]),s=Me(Je(n,this.inputSizeTensor),this.anchorsTensor),a=Je(r,this.doubleInputSizeTensor),o=pe(Ue(s,a),this.inputSizeTensor),i=pe(Me(s,a),this.inputSizeTensor);return ic([o,i],1)})}normalizeLandmarks(t,n){return Ve(()=>{let r=Me(Je(t.reshape([-1,7,2]),this.inputSizeTensor),this.anchors[n]);return pe(r,this.inputSizeTensor)})}async getBoxes(t,n){let r=this.model.predict(t),s=Xn(r);r.dispose();let a=Ve(()=>Is(Xe(s,[0,0],[-1,1])).squeeze()),o=a.dataSync(),i=Xe(s,[0,1],[-1,4]),l=this.normalizeBoxes(i);i.dispose();let u=await Ze.nonMaxSuppressionAsync(l,o,n.hand.maxDetected,n.hand.iouThreshold,n.hand.minConfidence),c=u.arraySync();a.dispose(),u.dispose();let d=[];for(let h of c)if(o[h]>=n.hand.minConfidence){let p=Xe(l,[h,0],[1,-1]),f=Xe(s,[h,5],[1,14]),m=Ve(()=>this.normalizeLandmarks(f,h).reshape([-1,2]));f.dispose(),d.push({box:p,palmLandmarks:m,confidence:o[h]})}return s.dispose(),l.dispose(),d}async estimateHandBounds(t,n){let r=t.shape[1],s=t.shape[2],a=Ve(()=>t.resizeBilinear([this.inputSize,this.inputSize]).div(127.5).sub(1)),o=await this.getBoxes(a,n);a.dispose();let i=[];if(!o||o.length===0)return i;for(let l of o){let u=l.box.dataSync(),c=u.slice(0,2),d=u.slice(2,4),h=l.palmLandmarks.arraySync();l.box.dispose(),l.palmLandmarks.dispose(),i.push($$({startPoint:c,endPoint:d,palmLandmarks:h,confidence:l.confidence},[s/this.inputSize,r/this.inputSize]))}return i}};function gwe(e){return e-2*Math.PI*Math.floor((e+Math.PI)/(2*Math.PI))}function _$(e,t){let n=Math.PI/2-Math.atan2(-(t[1]-e[1]),t[0]-e[0]);return gwe(n)}var D$=(e,t)=>[[1,0,e],[0,1,t],[0,0,1]];function ro(e,t){let n=0;for(let r=0;r<e.length;r++)n+=e[r]*t[r];return n}function ywe(e,t){let n=[];for(let r=0;r<e.length;r++)n.push(e[r][t]);return n}function F$(e,t){let n=[],r=e.length;for(let s=0;s<r;s++){n.push([]);for(let a=0;a<r;a++)n[s].push(ro(e[s],ywe(t,a)))}return n}function Db(e,t){let n=Math.cos(e),r=Math.sin(e),s=[[n,-r,0],[r,n,0],[0,0,1]],a=D$(t[0],t[1]),o=F$(a,s),i=D$(-t[0],-t[1]);return F$(o,i)}function M$(e){let t=[[e[0][0],e[1][0]],[e[0][1],e[1][1]]],n=[e[0][2],e[1][2]],r=[-ro(t[0],n),-ro(t[1],n)];return[t[0].concat(r[0]),t[1].concat(r[1]),[0,0,1]]}function Fb(e,t){return[ro(e,t[0]),ro(e,t[1])]}var Awe=5,O$=1.65,P$=[0,5,9,13,17,1,2],xwe=0,bwe=2,Mb=class{constructor(t,n){var r;this.handDetector=t,this.handPoseModel=n,this.inputSize=(r=this.handPoseModel)==null?void 0:r.inputs[0].shape[2],this.storedBoxes=[],this.skipped=0,this.detectedHands=0}calculateLandmarksBoundingBox(t){let n=t.map(o=>o[0]),r=t.map(o=>o[1]),s=[Math.min(...n),Math.min(...r)],a=[Math.max(...n),Math.max(...r)];return{startPoint:s,endPoint:a}}getBoxForPalmLandmarks(t,n){let r=t.map(a=>Fb([...a,1],n)),s=this.calculateLandmarksBoundingBox(r);return l0(u0(s),Awe)}getBoxForHandLandmarks(t){let n=this.calculateLandmarksBoundingBox(t),r=l0(u0(n),O$);r.palmLandmarks=[];for(let s=0;s<P$.length;s++)r.palmLandmarks.push(t[P$[s]].slice(0,2));return r}transformRawCoords(t,n,r,s){let a=i0(n),o=[a[0]/this.inputSize,a[1]/this.inputSize,(a[0]+a[1])/this.inputSize/2],i=t.map(p=>[o[0]*(p[0]-this.inputSize/2),o[1]*(p[1]-this.inputSize/2),o[2]*p[2]]),l=Db(r,[0,0]),u=i.map(p=>[...Fb(p,l),p[2]]),c=M$(s),d=[...kh(n),1],h=[ro(d,c[0]),ro(d,c[1])];return u.map(p=>[Math.trunc(p[0]+h[0]),Math.trunc(p[1]+h[1]),Math.trunc(p[2])])}async estimateHands(t,n){let r=!1,s;(this.skipped===0||this.skipped>n.hand.skipFrames||!n.hand.landmarks||!n.skipFrame)&&(s=await this.handDetector.estimateHandBounds(t,n),this.skipped=0),n.skipFrame&&this.skipped++,s&&s.length>0&&(s.length!==this.detectedHands&&this.detectedHands!==n.hand.maxDetected||!n.hand.landmarks)&&(this.detectedHands=0,this.storedBoxes=[...s],this.storedBoxes.length>0&&(r=!0));let a=[];for(let o=0;o<this.storedBoxes.length;o++){let i=this.storedBoxes[o];if(!!i)if(n.hand.landmarks){let l=n.hand.rotation?_$(i.palmLandmarks[xwe],i.palmLandmarks[bwe]):0,u=kh(i),c=[u[0]/t.shape[2],u[1]/t.shape[1]],d=n.hand.rotation&&kr.flags.IS_BROWSER?Ze.rotateWithOffset(t,l,0,c):t.clone(),h=Db(-l,u),p=r?this.getBoxForPalmLandmarks(i.palmLandmarks,h):i,f=E$(p,d,[this.inputSize,this.inputSize]),m=f.div(255);f.dispose(),d.dispose();let[g,y]=await this.handPoseModel.predict(m);m.dispose();let A=g.dataSync()[0];if(g.dispose(),A>=n.hand.minConfidence){let x=le(y,[-1,3]),b=x.arraySync();y.dispose(),x.dispose();let v=this.transformRawCoords(b,p,l,h),I=this.getBoxForHandLandmarks(v);this.storedBoxes[o]={...I,confidence:A};let w={landmarks:v,confidence:A,box:{topLeft:I.startPoint,bottomRight:I.endPoint}};a.push(w)}else this.storedBoxes[o]=null;y.dispose()}else{let l=l0(u0(i),O$),u={confidence:i.confidence,box:{topLeft:l.startPoint,bottomRight:l.endPoint}};a.push(u)}}return this.storedBoxes=this.storedBoxes.filter(o=>o!==null),this.detectedHands=a.length,a}};var z$={thumb:[1,2,3,4],indexFinger:[5,6,7,8],middleFinger:[9,10,11,12],ringFinger:[13,14,15,16],pinky:[17,18,19,20],palmBase:[0]},so,ao,L$;async function Ob(e,t){let n=await L$.estimateHands(e,t);if(!n)return[];let r=[];for(let s=0;s<n.length;s++){let a={};if(n[s].landmarks)for(let u of Object.keys(z$))a[u]=z$[u].map(c=>n[s].landmarks[c]);let o=n[s].landmarks,i=[Number.MAX_SAFE_INTEGER,Number.MAX_SAFE_INTEGER,0,0],l=[0,0,0,0];if(o&&o.length>0){for(let u of o)u[0]<i[0]&&(i[0]=u[0]),u[1]<i[1]&&(i[1]=u[1]),u[0]>i[2]&&(i[2]=u[0]),u[1]>i[3]&&(i[3]=u[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[s].box?[Math.trunc(Math.max(0,n[s].box.topLeft[0])),Math.trunc(Math.max(0,n[s].box.topLeft[1])),Math.trunc(Math.min(e.shape[2]||0,n[s].box.bottomRight[0])-Math.max(0,n[s].box.topLeft[0])),Math.trunc(Math.min(e.shape[1]||0,n[s].box.bottomRight[1])-Math.max(0,n[s].box.topLeft[1]))]:[0,0,0,0],l=[n[s].box.topLeft[0]/(e.shape[2]||0),n[s].box.topLeft[1]/(e.shape[1]||0),(n[s].box.bottomRight[0]-n[s].box.topLeft[0])/(e.shape[2]||0),(n[s].box.bottomRight[1]-n[s].box.topLeft[1])/(e.shape[1]||0)];r.push({id:s,score:Math.round(100*n[s].confidence)/100,box:i,boxRaw:l,keypoints:o,annotations:a})}return r}async function Pb(e){!so||!ao?([so,ao]=await Promise.all([e.hand.enabled?Nt(Ct(e.modelBasePath,e.hand.detector.modelPath),{fromTFHub:e.hand.detector.modelPath.includes("tfhub.dev")}):null,e.hand.landmarks?Nt(Ct(e.modelBasePath,e.hand.skeleton.modelPath),{fromTFHub:e.hand.skeleton.modelPath.includes("tfhub.dev")}):null]),e.hand.enabled&&(!so||!so.modelUrl?fe("load model failed:",e.hand.detector.modelPath):e.debug&&fe("load model:",so.modelUrl),!ao||!ao.modelUrl?fe("load model failed:",e.hand.skeleton.modelPath):e.debug&&fe("load model:",ao.modelUrl))):(e.debug&&fe("cached model:",so.modelUrl),e.debug&&fe("cached model:",ao.modelUrl));let t=new _b(so);return L$=new Mb(t,ao),[so,ao]}var B$=["nose","leftEyeInside","leftEye","leftEyeOutside","rightEyeInside","rightEye","rightEyeOutside","leftEar","rightEar","leftMouth","rightMouth","leftShoulder","rightShoulder","leftElbow","rightElbow","leftWrist","rightWrist","leftPalm","rightPalm","leftIndex","rightIndex","leftPinky","rightPinky","leftHip","rightHip","leftKnee","rightKnee","leftAnkle","rightAnkle","leftHeel","rightHeel","leftFoot","rightFoot","midHip","forehead","leftThumb","leftHand","rightThumb","rightHand"],W$=["nose","leftEyeInside","leftEye","leftEyeOutside","rightEyeInside","rightEye","rightEyeOutside","leftEar","rightEar","leftMouth","rightMouth","leftShoulder","rightShoulder","leftElbow","rightElbow","left:15","right:16","left:17","right:18","left:19","right:20","left:21","right:22","leftChest","rightChest","neck","forehead","left:27","right:28","left:29","right:30"];var sr;async function c0(e){return sr?e.debug&&fe("cached model:",sr.modelUrl):(sr=await Nt(Ct(e.modelBasePath,e.body.modelPath)),sr.width=parseInt(sr.signature.inputs["input_1:0"].tensorShape.dim[2].size),sr.height=parseInt(sr.signature.inputs["input_1:0"].tensorShape.dim[1].size),!sr||!sr.modelUrl?fe("load model failed:",e.body.modelPath):e.debug&&fe("load model:",sr.modelUrl)),sr}async function zb(e,t){var m;if(!sr)return[];if(!t.body.enabled)return[];let n={width:e.shape[2]||0,height:e.shape[1]||0},r=Ze.resizeBilinear(e,[sr.width,sr.height],!1),s=Je(r,[255]);r.dispose();let a=await sr.predict(s),o=((m=a.find(g=>g.size===195||g.size===155))==null?void 0:m.dataSync())||[];a.forEach(g=>g.dispose()),s.dispose();let i=[],l=(o==null?void 0:o.length)===195?B$:W$,u=5;for(let g=0;g<o.length/u;g++)i.push({id:g,part:l[g],position:[Math.trunc(n.width*o[u*g+0]/255),Math.trunc(n.height*o[u*g+1]/255),Math.trunc(o[u*g+2])+0],positionRaw:[o[u*g+0]/255,o[u*g+1]/255,o[u*g+2]+0],score:(100-Math.trunc(100/(1+Math.exp(o[u*g+3]))))/100,presence:(100-Math.trunc(100/(1+Math.exp(o[u*g+4]))))/100});let c=i.map(g=>g.position[0]),d=i.map(g=>g.position[1]),h=[Math.min(...c),Math.min(...d),Math.max(...c)-Math.min(...c),Math.max(...d)-Math.min(...c)],p=[0,0,0,0],f=i.reduce((g,y)=>y.score>g?y.score:g,0);return[{id:0,score:f,box:h,boxRaw:p,keypoints:i}]}var ar,Vs=[],Lb=[0,0,0,0],Bb=[0,0,0,0],d0=0,Wb=Number.MAX_SAFE_INTEGER,vwe=["head","neck","rightShoulder","rightElbow","rightWrist","chest","leftShoulder","leftElbow","leftWrist","pelvis","rightHip","rightKnee","rightAnkle","leftHip","leftKnee","leftAnkle"];async function V$(e){return ar?e.debug&&fe("cached model:",ar.modelUrl):(ar=await Nt(Ct(e.modelBasePath,e.body.modelPath)),!ar||!ar.modelUrl?fe("load model failed:",e.body.modelPath):e.debug&&fe("load model:",ar.modelUrl)),ar}function wwe(e,t){let[n,r]=e.shape;return Ve(()=>{let s=(i,l)=>Ue(i,pe(Je(i,ut(l,"int32")),ut(l,"int32"))),a=le(e,[r*n]),o=_a(a,0).dataSync()[0];if(o>t){let i=q2(a,0),l=s(i,n).dataSync()[0],u=Je(i,ut(n,"int32")).dataSync()[0];return[l,u,o]}return[0,0,o]})}async function Vb(e,t){return Wb<t.body.skipFrames&&t.skipFrame&&Object.keys(Vs).length>0?(Wb++,[{id:0,score:d0,box:Lb,boxRaw:Bb,keypoints:Vs}]):(Wb=0,new Promise(async n=>{let r=Ve(()=>{if(!ar.inputs[0].shape)return null;let u=Ze.resizeBilinear(e,[ar.inputs[0].shape[2],ar.inputs[0].shape[1]],!1);return pe(u,2).sub(1)}),s;if(t.body.enabled&&(s=await ar.predict(r)),r.dispose(),s){Vs.length=0;let u=s.squeeze();We(s);let c=u.unstack(2);We(u);for(let d=0;d<c.length;d++){let[h,p,f]=wwe(c[d],t.body.minConfidence);d0>t.body.minConfidence&&Vs.push({score:Math.round(100*f)/100,part:vwe[d],positionRaw:[h/ar.inputs[0].shape[2],p/ar.inputs[0].shape[1]],position:[Math.round(e.shape[2]*h/ar.inputs[0].shape[2]),Math.round(e.shape[1]*p/ar.inputs[0].shape[1])]})}c.forEach(d=>We(d))}d0=Vs.reduce((u,c)=>c.score>u?c.score:u,0);let a=Vs.map(u=>u.position[0]),o=Vs.map(u=>u.position[1]);Lb=[Math.min(...a),Math.min(...o),Math.max(...a)-Math.min(...a),Math.max(...o)-Math.min(...o)];let i=Vs.map(u=>u.positionRaw[0]),l=Vs.map(u=>u.positionRaw[1]);Bb=[Math.min(...i),Math.min(...l),Math.max(...i)-Math.min(...i),Math.max(...l)-Math.min(...l)],n([{id:0,score:d0,box:Lb,boxRaw:Bb,keypoints:Vs}])}))}var xs,Us=[],Ub=[0,0,0,0],Hb=[0,0,0,0],$u=0,Gb=Number.MAX_SAFE_INTEGER,kwe=["nose","leftEye","rightEye","leftEar","rightEar","leftShoulder","rightShoulder","leftElbow","rightElbow","leftWrist","rightWrist","leftHip","rightHip","leftKnee","rightKnee","leftAnkle","rightAnkle"];async function jb(e){return xs?e.debug&&fe("cached model:",xs.modelUrl):(xs=await Nt(Ct(e.modelBasePath,e.body.modelPath)),!xs||!xs.modelUrl?fe("load model failed:",e.body.modelPath):e.debug&&fe("load model:",xs.modelUrl)),xs}async function qb(e,t){return Gb<t.body.skipFrames&&t.skipFrame&&Object.keys(Us).length>0?(Gb++,[{id:0,score:$u,box:Ub,boxRaw:Hb,keypoints:Us}]):(Gb=0,new Promise(async n=>{let r=Ve(()=>{if(!xs.inputs[0].shape)return null;let u=Ze.resizeBilinear(e,[xs.inputs[0].shape[2],xs.inputs[0].shape[1]],!1);return Mt(u,"int32")}),s;if(t.body.enabled&&(s=await xs.predict(r)),r.dispose(),s){Us.length=0;let u=s.arraySync();We(s);let c=u[0][0];for(let d=0;d<c.length;d++)$u=c[d][2],$u>t.body.minConfidence&&Us.push({score:Math.round(100*$u)/100,part:kwe[d],positionRaw:[c[d][1],c[d][0]],position:[Math.round((e.shape[2]||0)*c[d][1]),Math.round((e.shape[1]||0)*c[d][0])]})}$u=Us.reduce((u,c)=>c.score>u?c.score:u,0);let a=Us.map(u=>u.position[0]),o=Us.map(u=>u.position[1]);Ub=[Math.min(...a),Math.min(...o),Math.max(...a)-Math.min(...a),Math.max(...o)-Math.min(...o)];let i=Us.map(u=>u.positionRaw[0]),l=Us.map(u=>u.positionRaw[1]);Hb=[Math.min(...i),Math.min(...l),Math.max(...i)-Math.min(...i),Math.max(...l)-Math.min(...l)],n([{id:0,score:$u,box:Ub,boxRaw:Hb,keypoints:Us}])}))}var Ru=[{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 gr,Kb=[],Xb=Number.MAX_SAFE_INTEGER,h0=2.5;async function Zb(e){if(gr)e.debug&&fe("cached model:",gr.modelUrl);else{gr=await Nt(Ct(e.modelBasePath,e.object.modelPath));let t=Object.values(gr.modelSignature.inputs);if(gr.inputSize=Array.isArray(t)?parseInt(t[0].tensorShape.dim[2].size):null,!gr.inputSize)throw new Error(`Human: Cannot determine model inputSize: ${e.object.modelPath}`);!gr||!gr.modelUrl?fe("load model failed:",e.object.modelPath):e.debug&&fe("load model:",gr.modelUrl)}return gr}async function Iwe(e,t,n,r){let s=0,a=[];for(let u of[1,2,4])Ve(()=>{var g,y;let c=u*13,d=(g=e.find(A=>A.shape[1]===c**2&&A.shape[2]===Ru.length))==null?void 0:g.squeeze(),h=(y=e.find(A=>A.shape[1]===c**2&&A.shape[2]<Ru.length))==null?void 0:y.squeeze(),f=h.reshape([-1,4,h.shape[1]/4]).argMax(2).arraySync(),m=d.arraySync();for(let A=0;A<d.shape[0];A++)for(let x=0;x<d.shape[1];x++){let b=m[A][x];if(b>r.object.minConfidence&&x!==61){let v=(.5+Math.trunc(A%c))/c,I=(.5+Math.trunc(A/c))/c,w=f[A].map(B=>B*(c/u/t)),[S,E]=[v-h0/u*w[0],I-h0/u*w[1]],[D,$]=[v+h0/u*w[2]-S,I+h0/u*w[3]-E],R=[S,E,D,$];R=R.map(B=>Math.max(0,Math.min(B,1)));let N=[R[0]*n[0],R[1]*n[1],R[2]*n[0],R[3]*n[1]],M={id:s++,score:Math.round(100*b)/100,class:x+1,label:Ru[x].label,box:N.map(B=>Math.trunc(B)),boxRaw:R};a.push(M)}}});e.forEach(u=>We(u));let o=a.map(u=>[u.boxRaw[1],u.boxRaw[0],u.boxRaw[3],u.boxRaw[2]]),i=a.map(u=>u.score),l=[];if(o&&o.length>0){let u=await Ze.nonMaxSuppressionAsync(o,i,r.object.maxDetected,r.object.iouThreshold,r.object.minConfidence);l=u.dataSync(),We(u)}return a=a.filter((u,c)=>l.includes(c)).sort((u,c)=>c.score-u.score),a}async function Yb(e,t){return Xb<t.object.skipFrames&&t.skipFrame&&Kb.length>0?(Xb++,Kb):(Xb=0,new Promise(async n=>{let r=[e.shape[2],e.shape[1]],s=Ze.resizeBilinear(e,[gr.inputSize,gr.inputSize],!1),a=s.div(255),o=a.transpose([0,3,1,2]);a.dispose(),s.dispose();let i;t.object.enabled&&(i=await gr.predict(o)),o.dispose();let l=await Iwe(i,gr.inputSize,r,t);Kb=l,n(l)}))}var yr,Jb=[],Qb=Number.MAX_SAFE_INTEGER;async function e3(e){if(yr)e.debug&&fe("cached model:",yr.modelUrl);else{yr=await Nt(Ct(e.modelBasePath,e.object.modelPath));let t=Object.values(yr.modelSignature.inputs);if(yr.inputSize=Array.isArray(t)?parseInt(t[0].tensorShape.dim[2].size):null,!yr.inputSize)throw new Error(`Human: Cannot determine model inputSize: ${e.object.modelPath}`);!yr||!yr.modelUrl?fe("load model failed:",e.object.modelPath):e.debug&&fe("load model:",yr.modelUrl)}return yr}async function Swe(e,t,n,r){if(!e)return[];let s=[],a=e.arraySync(),o=Xn(e);e.dispose();let i=na(o,6,1);o.dispose();let u=So([i[1],i[0],i[3],i[2]],1).squeeze(),c=i[4].squeeze(),d=i[5].squeeze();i.forEach(m=>m.dispose());let h=await Ze.nonMaxSuppressionAsync(u,c,r.object.maxDetected,r.object.iouThreshold,r.object.minConfidence);u.dispose(),c.dispose(),d.dispose();let p=h.dataSync();h.dispose();let f=0;for(let m of p){let g=Math.trunc(100*a[0][m][4])/100,y=a[0][m][5],A=Ru[y].label,[x,b]=[a[0][m][0]/t,a[0][m][1]/t],v=[x,b,a[0][m][2]/t-x,a[0][m][3]/t-b],I=[Math.trunc(v[0]*n[0]),Math.trunc(v[1]*n[1]),Math.trunc(v[2]*n[0]),Math.trunc(v[3]*n[1])];s.push({id:f++,score:g,class:y,label:A,box:I,boxRaw:v})}return s}async function t3(e,t){return Qb<t.object.skipFrames&&t.skipFrame&&Jb.length>0?(Qb++,Jb):(Qb=0,new Promise(async n=>{let r=[e.shape[2],e.shape[1]],s=Ze.resizeBilinear(e,[yr.inputSize,yr.inputSize]),a=t.object.enabled?yr.execute(s,["tower_0/detections"]):null;s.dispose();let o=await Swe(a,yr.inputSize,r,t);Jb=o,n(o)}))}function Twe(e,t,n){let r=function(i,l,u){let c=new RegExp("\\b"+l+" \\w+ (\\w+)","ig");i.replace(c,(d,h)=>(u[h]=0,d))},s=function(i,l){let u=e.createShader(l);if(e.shaderSource(u,i),e.compileShader(u),!e.getShaderParameter(u,e.COMPILE_STATUS))throw new Error("Filter: GL compile failed",e.getShaderInfoLog(u));return u};this.uniform={},this.attribute={};let a=s(t,e.VERTEX_SHADER),o=s(n,e.FRAGMENT_SHADER);if(this.id=e.createProgram(),e.attachShader(this.id,a),e.attachShader(this.id,o),e.linkProgram(this.id),!e.getProgramParameter(this.id,e.LINK_STATUS))throw new Error("Filter: GL link failed",e.getProgramInfoLog(this.id));e.useProgram(this.id),r(t,"attribute",this.attribute);for(let i in this.attribute)this.attribute[i]=e.getAttribLocation(this.id,i);r(t,"uniform",this.uniform),r(n,"uniform",this.uniform);for(let i in this.uniform)this.uniform[i]=e.getUniformLocation(this.id,i)}function U$(e){e||(e={});let t=0,n=null,r=!1,s=-1,a=[null,null],o=[],i=-1,l=-1,u=null,c=null,d={},h=e.canvas||document.createElement("canvas"),p={},f={INTERMEDIATE:1},m=h.getContext("webgl");if(!m)throw new Error("Filter: getContext() failed");this.addFilter=function(v){let I=Array.prototype.slice.call(arguments,1),w=d[v];o.push({func:w,args:I})},this.reset=function(){o=[]};let g=function(v,I){if(!(v===i&&I===l)){if(h.width=v,i=v,h.height=I,l=I,!u){let w=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]);u=m.createBuffer(),m.bindBuffer(m.ARRAY_BUFFER,u),m.bufferData(m.ARRAY_BUFFER,w,m.STATIC_DRAW),m.pixelStorei(m.UNPACK_PREMULTIPLY_ALPHA_WEBGL,!0)}m.viewport(0,0,i,l),a=[null,null]}},y=function(v,I){let w=m.createFramebuffer();m.bindFramebuffer(m.FRAMEBUFFER,w);let S=m.createRenderbuffer();m.bindRenderbuffer(m.RENDERBUFFER,S);let E=m.createTexture();return m.bindTexture(m.TEXTURE_2D,E),m.texImage2D(m.TEXTURE_2D,0,m.RGBA,v,I,0,m.RGBA,m.UNSIGNED_BYTE,null),m.texParameteri(m.TEXTURE_2D,m.TEXTURE_MAG_FILTER,m.LINEAR),m.texParameteri(m.TEXTURE_2D,m.TEXTURE_MIN_FILTER,m.LINEAR),m.texParameteri(m.TEXTURE_2D,m.TEXTURE_WRAP_S,m.CLAMP_TO_EDGE),m.texParameteri(m.TEXTURE_2D,m.TEXTURE_WRAP_T,m.CLAMP_TO_EDGE),m.framebufferTexture2D(m.FRAMEBUFFER,m.COLOR_ATTACHMENT0,m.TEXTURE_2D,E,0),m.bindTexture(m.TEXTURE_2D,null),m.bindFramebuffer(m.FRAMEBUFFER,null),{fbo:w,texture:E}},A=function(v){return a[v]=a[v]||y(i,l),a[v]},x=function(v=null){var E,D;let I=null,w=null,S=!1;t===0?I=n:I=(E=A(s))==null?void 0:E.texture,t++,r&&!(v&f.INTERMEDIATE)?(w=null,S=t%2==0):(s=(s+1)%2,w=(D=A(s))==null?void 0:D.fbo),m.bindTexture(m.TEXTURE_2D,I),m.bindFramebuffer(m.FRAMEBUFFER,w),m.uniform1f(c.uniform.flipY,S?-1:1),m.drawArrays(m.TRIANGLES,0,6)};this.apply=function(v){if(g(v.width,v.height),t=0,n||(n=m.createTexture()),m.bindTexture(m.TEXTURE_2D,n),m.texParameteri(m.TEXTURE_2D,m.TEXTURE_WRAP_S,m.CLAMP_TO_EDGE),m.texParameteri(m.TEXTURE_2D,m.TEXTURE_WRAP_T,m.CLAMP_TO_EDGE),m.texParameteri(m.TEXTURE_2D,m.TEXTURE_MIN_FILTER,m.NEAREST),m.texParameteri(m.TEXTURE_2D,m.TEXTURE_MAG_FILTER,m.NEAREST),m.texImage2D(m.TEXTURE_2D,0,m.RGBA,m.RGBA,m.UNSIGNED_BYTE,v),o.length===0)return x(),h;for(let I=0;I<o.length;I++){r=I===o.length-1;let w=o[I];w.func.apply(this,w.args||[])}return h};let b=function(v){if(p[v])return c=p[v],m.useProgram(c.id),c;let I={};I.VERTEX_IDENTITY=["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.);","}"].join(`
`),I.FRAGMENT_IDENTITY=["precision highp float;","varying vec2 vUv;","uniform sampler2D texture;","void main(void) {","gl_FragColor = texture2D(texture, vUv);","}"].join(`
`),c=new Twe(m,I.VERTEX_IDENTITY,v);let w=Float32Array.BYTES_PER_ELEMENT,S=4*w;return m.enableVertexAttribArray(c.attribute.pos),m.vertexAttribPointer(c.attribute.pos,2,m.FLOAT,!1,S,0*w),m.enableVertexAttribArray(c.attribute.uv),m.vertexAttribPointer(c.attribute.uv,2,m.FLOAT,!1,S,2*w),p[v]=c,c};d.colorMatrix=function(v){let I=new Float32Array(v);I[4]/=255,I[9]/=255,I[14]/=255,I[19]/=255;let w=I[18]===1&&I[3]===0&&I[8]===0&&I[13]===0&&I[15]===0&&I[16]===0&&I[17]===0&&I[19]===0?d.colorMatrix.SHADER.WITHOUT_ALPHA:d.colorMatrix.SHADER.WITH_ALPHA,S=b(w);m.uniform1fv(S.uniform.m,I),x()},d.colorMatrix.SHADER={},d.colorMatrix.SHADER.WITH_ALPHA=["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];","}"].join(`
`),d.colorMatrix.SHADER.WITHOUT_ALPHA=["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;","}"].join(`
`),d.brightness=function(v){let I=(v||0)+1;d.colorMatrix([I,0,0,0,0,0,I,0,0,0,0,0,I,0,0,0,0,0,1,0])},d.saturation=function(v){let I=(v||0)*2/3+1,w=(I-1)*-.5;d.colorMatrix([I,w,w,0,0,w,I,w,0,0,w,w,I,0,0,0,0,0,1,0])},d.desaturate=function(){d.saturation(-1)},d.contrast=function(v){let I=(v||0)+1,w=-128*(I-1);d.colorMatrix([I,0,0,0,w,0,I,0,0,w,0,0,I,0,w,0,0,0,1,0])},d.negative=function(){d.contrast(-2)},d.hue=function(v){v=(v||0)/180*Math.PI;let I=Math.cos(v),w=Math.sin(v),S=.213,E=.715,D=.072;d.colorMatrix([S+I*(1-S)+w*-S,E+I*-E+w*-E,D+I*-D+w*(1-D),0,0,S+I*-S+w*.143,E+I*(1-E)+w*.14,D+I*-D+w*-.283,0,0,S+I*-S+w*-(1-S),E+I*-E+w*E,D+I*(1-D)+w*D,0,0,0,0,0,1,0])},d.desaturateLuminance=function(){d.colorMatrix([.2764723,.929708,.0938197,0,-37.1,.2764723,.929708,.0938197,0,-37.1,.2764723,.929708,.0938197,0,-37.1,0,0,0,1,0])},d.sepia=function(){d.colorMatrix([.393,.7689999,.18899999,0,0,.349,.6859999,.16799999,0,0,.272,.5339999,.13099999,0,0,0,0,0,1,0])},d.brownie=function(){d.colorMatrix([.5997023498159715,.34553243048391263,-.2708298674538042,0,47.43192855600873,-.037703249837783157,.8609577587992641,.15059552388459913,0,-36.96841498319127,.24113635128153335,-.07441037908422492,.44972182064877153,0,-7.562075277591283,0,0,0,1,0])},d.vintagePinhole=function(){d.colorMatrix([.6279345635605994,.3202183420819367,-.03965408211312453,0,9.651285835294123,.02578397704808868,.6441188644374771,.03259127616149294,0,7.462829176470591,.0466055556782719,-.0851232987247891,.5241648018700465,0,5.159190588235296,0,0,0,1,0])},d.kodachrome=function(){d.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])},d.technicolor=function(){d.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])},d.polaroid=function(){d.colorMatrix([1.438,-.062,-.062,0,0,-.122,1.378,-.122,0,0,-.016,-.016,1.483,0,0,0,0,0,1,0])},d.shiftToBGR=function(){d.colorMatrix([0,0,1,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,1,0])},d.convolution=function(v){let I=new Float32Array(v),w=1/i,S=1/l,E=b(d.convolution.SHADER);m.uniform1fv(E.uniform.m,I),m.uniform2f(E.uniform.px,w,S),x()},d.convolution.SHADER=["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);","vec4 c12 = texture2D(texture, vec2(vUv.x, vUv.y - px.y));","vec4 c13 = texture2D(texture, vec2(vUv.x + px.x, vUv.y - px.y));","vec4 c21 = texture2D(texture, vec2(vUv.x - px.x, vUv.y) );","vec4 c22 = texture2D(texture, vUv);","vec4 c23 = texture2D(texture, vec2(vUv.x + px.x, vUv.y) );","vec4 c31 = texture2D(texture, vec2(vUv.x - px.x, vUv.y + px.y) );","vec4 c32 = texture2D(texture, vec2(vUv.x, vUv.y + px.y) );","vec4 c33 = texture2D(texture, vUv + px );","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;","}"].join(`
`),d.detectEdges=function(){d.convolution.call(this,[0,1,0,1,-4,1,0,1,0])},d.sobelX=function(){d.convolution.call(this,[-1,0,1,-2,0,2,-1,0,1])},d.sobelY=function(){d.convolution.call(this,[-1,-2,-1,0,0,0,1,2,1])},d.sharpen=function(v){let I=v||1;d.convolution.call(this,[0,-1*I,0,-1*I,1+4*I,-1*I,0,-1*I,0])},d.emboss=function(v){let I=v||1;d.convolution.call(this,[-2*I,-1*I,0,-1*I,1,1*I,0,1*I,2*I])},d.blur=function(v){let I=v/7/i,w=v/7/l,S=b(d.blur.SHADER);m.uniform2f(S.uniform.px,0,w),x(f.INTERMEDIATE),m.uniform2f(S.uniform.px,I,0),x()},d.blur.SHADER=["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;","}"].join(`
`),d.pixelate=function(v){let I=v/i,w=v/l,S=b(d.pixelate.SHADER);m.uniform2f(S.uniform.size,I,w),x()},d.pixelate.SHADER=["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);","}"].join(`
`)}var p0=2048,Oe,Lt,tn;function vi(e,t){let n;if(!e)throw new Error("Human: Input is missing");if(!(e instanceof It)&&!(typeof Image!="undefined"&&e instanceof Image)&&!(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("Human: Input type is not recognized");if(e instanceof It)if(e.shape&&e.shape.length===4&&e.shape[0]===1&&e.shape[3]===3)n=Qs(e);else throw new Error(`Human: Input tensor shape must be [1, height, width, 3] and instead was ${e.shape}`);else{let s=e.naturalWidth||e.videoWidth||e.width||e.shape&&e.shape[1]>0,a=e.naturalHeight||e.videoHeight||e.height||e.shape&&e.shape[2]>0;if(!s||!a)return{tensor:null,canvas:Oe};let o=s,i=a;if(o>p0&&(o=p0,i=o*a/s),i>p0&&(i=p0,o=i*s/a),t.filter.width>0?o=t.filter.width:t.filter.height>0&&(o=s*(t.filter.height/a)),t.filter.height>0?i=t.filter.height:t.filter.width>0&&(i=a*(t.filter.width/s)),!o||!i)throw new Error("Human: Input cannot determine dimension");(!Oe||(Oe==null?void 0:Oe.width)!==o||(Oe==null?void 0:Oe.height)!==i)&&(Oe=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(o,i):document.createElement("canvas"),(Oe==null?void 0:Oe.width)!==o&&(Oe.width=o),(Oe==null?void 0:Oe.height)!==i&&(Oe.height=i));let l=Oe.getContext("2d");if(e instanceof ImageData?l.putImageData(e,0,0):t.filter.flip&&typeof l.translate!="undefined"?(l.translate(s,0),l.scale(-1,1),l.drawImage(e,0,0,s,a,0,0,Oe==null?void 0:Oe.width,Oe==null?void 0:Oe.height),l.setTransform(1,0,0,1,0,0)):l.drawImage(e,0,0,s,a,0,0,Oe==null?void 0:Oe.width,Oe==null?void 0:Oe.height),t.filter.enabled){if((!tn||!Lt||Oe.width!==Lt.width||(Oe==null?void 0:Oe.height)!==(Lt==null?void 0:Lt.height))&&(Lt=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(Oe==null?void 0:Oe.width,Oe==null?void 0:Oe.height):document.createElement("canvas"),(Lt==null?void 0:Lt.width)!==(Oe==null?void 0:Oe.width)&&(Lt.width=Oe==null?void 0:Oe.width),(Lt==null?void 0:Lt.height)!==(Oe==null?void 0:Oe.height)&&(Lt.height=Oe==null?void 0:Oe.height),tn=kr.flags.IS_BROWSER?new U$({canvas:Lt}):null),!tn)return{tensor:null,canvas:Oe};tn.reset(),tn.addFilter("brightness",t.filter.brightness),t.filter.contrast!==0&&tn.addFilter("contrast",t.filter.contrast),t.filter.sharpness!==0&&tn.addFilter("sharpen",t.filter.sharpness),t.filter.blur!==0&&tn.addFilter("blur",t.filter.blur),t.filter.saturation!==0&&tn.addFilter("saturation",t.filter.saturation),t.filter.hue!==0&&tn.addFilter("hue",t.filter.hue),t.filter.negative&&tn.addFilter("negative"),t.filter.sepia&&tn.addFilter("sepia"),t.filter.vintage&&tn.addFilter("brownie"),t.filter.sepia&&tn.addFilter("sepia"),t.filter.kodachrome&&tn.addFilter("kodachrome"),t.filter.technicolor&&tn.addFilter("technicolor"),t.filter.polaroid&&tn.addFilter("polaroid"),t.filter.pixelate!==0&&tn.addFilter("pixelate",t.filter.pixelate),tn.apply(Oe)}else Lt=Oe,tn&&(tn=null);let u;if(Lt.data){let c=[Lt.height,Lt.width,3];u=fp(Lt.data,c,"int32")}else if(Lt instanceof ImageData)u=Br?Br.fromPixels(Lt):null;else if(t.backend==="webgl"||t.backend==="humangl"){let c=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(o,i):document.createElement("canvas");c.width=o,c.height=i;let d=c.getContext("2d");d==null||d.drawImage(Lt,0,0),u=Br?Br.fromPixels(c):null}else{let c=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(o,i):document.createElement("canvas");c.width=o,c.height=i;let d=c.getContext("2d");d==null||d.drawImage(Lt,0,0);let h=d==null?void 0:d.getImageData(0,0,o,i);u=Br?Br.fromPixels(h):null}if(u){let c=u.toFloat();n=c.expandDims(0),u.dispose(),c.dispose()}}let r=t.filter.return?Lt:null;return{tensor:n,canvas:r}}var Pr,n3=!1;async function f0(e){return Pr?e.debug&&fe("cached model:",Pr.modelUrl):(Pr=await Nt(Ct(e.modelBasePath,e.segmentation.modelPath)),!Pr||!Pr.modelUrl?fe("load model failed:",e.segmentation.modelPath):e.debug&&fe("load model:",Pr.modelUrl)),Pr}async function r3(e){var f,m;let t=((f=e.tensor)==null?void 0:f.shape[1])||0,n=((m=e.tensor)==null?void 0:m.shape[2])||0;if(!e.tensor||!Pr||!Pr.inputs[0].shape)return null;let r=Ze.resizeBilinear(e.tensor,[Pr.inputs[0].shape[1],Pr.inputs[0].shape[2]],!1),s=r.div(255),a=Pr.predict(s);We(r),We(s);let o=Xn(a,0),i;if(o.shape[2]===2){let g=o.softmax(),[y,A]=pc(g,2),x=A.expandDims(2),b=x.expandDims(0);We(g),We(y),We(A);let v=Ze.cropAndResize(b,[[0,0,.5,.5]],[0],[t,n]);i=v.squeeze(0),We(v),We(x),We(b)}else i=Ze.resizeBilinear(o,[t,n]);if(typeof document=="undefined")return i.dataSync();let l=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(t,n):document.createElement("canvas");l.width=t,l.height=n,Br&&await Br.toPixels(i,l),We(i),We(o),We(a);let u=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(t,n):document.createElement("canvas");u.width=t,u.height=n;let c=u.getContext("2d");c.filter="blur(8px",await c.drawImage(l,0,0);let d=c.getImageData(0,0,t,n).data,h=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(t,n):document.createElement("canvas");h.width=t,h.height=n;let p=h.getContext("2d");return e.canvas&&await p.drawImage(e.canvas,0,0),p.globalCompositeOperation="darken",p.filter="blur(8px)",await p.drawImage(l,0,0),p.globalCompositeOperation="source-over",p.filter="none",e.canvas=h,d}async function H$(e,t,n){var a;if(n3)return null;n3=!0,Pr||await f0(n);let r=vi(e,n),s=await r3(r);if(We(r.tensor),t&&s){let o=vi(t,n),i=o.canvas;We(o.tensor);let l=r.canvas,u=(a=l.getContext("2d"))==null?void 0:a.getImageData(0,0,l.width,l.height).data,c=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(l.width,l.height):document.createElement("canvas");c.width=l.width,c.height=l.height;let d=c.getContext("2d");d.globalCompositeOperation="copy",d.drawImage(i,0,0,c.width,c.height);let h=d.getImageData(0,0,c.width,c.height);for(let p=0;p<c.width*c.height;p++)h.data[4*p+0]=(255-s[4*p+0])/255*h.data[4*p+0]+s[4*p+0]/255*u[4*p+0],h.data[4*p+1]=(255-s[4*p+1])/255*h.data[4*p+1]+s[4*p+1]/255*u[4*p+1],h.data[4*p+2]=(255-s[4*p+2])/255*h.data[4*p+2]+s[4*p+2]/255*u[4*p+2],h.data[4*p+3]=(255-s[4*p+3])/255*h.data[4*p+3]+s[4*p+3]/255*u[4*p+3];d.putImageData(h,0,0),r.canvas=c}return n3=!1,r.canvas}async function G$(e){e.config.async?[e.models.face,e.models.emotion,e.models.handpose,e.models.posenet,e.models.blazepose,e.models.efficientpose,e.models.movenet,e.models.nanodet,e.models.centernet,e.models.faceres,e.models.segmentation]=await Promise.all([e.models.face||(e.config.face.enabled?mb(e.config):null),e.models.emotion||(e.config.face.enabled&&e.config.face.emotion.enabled?kb(e.config):null),e.models.handpose||(e.config.hand.enabled?Pb(e.config):null),e.models.posenet||(e.config.body.enabled&&e.config.body.modelPath.includes("posenet")?Rb(e.config):null),e.models.blazepose||(e.config.body.enabled&&e.config.body.modelPath.includes("blazepose")?c0(e.config):null),e.models.efficientpose||(e.config.body.enabled&&e.config.body.modelPath.includes("efficientpose")?V$(e.config):null),e.models.movenet||(e.config.body.enabled&&e.config.body.modelPath.includes("movenet")?jb(e.config):null),e.models.nanodet||(e.config.object.enabled&&e.config.object.modelPath.includes("nanodet")?Zb(e.config):null),e.models.centernet||(e.config.object.enabled&&e.config.object.modelPath.includes("centernet")?e3(e.config):null),e.models.faceres||(e.config.face.enabled&&e.config.face.description.enabled?yb(e.config):null),e.models.segmentation||(e.config.segmentation.enabled?f0(e.config):null)]):(e.config.face.enabled&&!e.models.face&&(e.models.face=await mb(e.config)),e.config.face.enabled&&e.config.face.emotion.enabled&&!e.models.emotion&&(e.models.emotion=await kb(e.config)),e.config.hand.enabled&&!e.models.handpose&&(e.models.handpose=await Pb(e.config)),e.config.body.enabled&&!e.models.posenet&&e.config.body.modelPath.includes("posenet")&&(e.models.posenet=await Rb(e.config)),e.config.body.enabled&&!e.models.blazepose&&e.config.body.modelPath.includes("blazepose")&&(e.models.blazepose=await c0(e.config)),e.config.body.enabled&&!e.models.efficientpose&&e.config.body.modelPath.includes("efficientpose")&&(e.models.efficientpose=await c0(e.config)),e.config.body.enabled&&!e.models.movenet&&e.config.body.modelPath.includes("movenet")&&(e.models.movenet=await jb(e.config)),e.config.object.enabled&&!e.models.nanodet&&e.config.object.modelPath.includes("nanodet")&&(e.models.nanodet=await Zb(e.config)),e.config.object.enabled&&!e.models.centernet&&e.config.object.modelPath.includes("centernet")&&(e.models.centernet=await e3(e.config)),e.config.face.enabled&&e.config.face.description.enabled&&!e.models.faceres&&(e.models.faceres=await yb(e.config)),e.config.segmentation.enabled&&!e.models.segmentation&&(e.models.segmentation=await f0(e.config)))}var Nwe=e=>{let t=(d,h)=>Math.atan2(d[1]-h[1],d[0]-h[0]);if(!e.annotations.rightEyeIris||!e.annotations.leftEyeIris)return{bearing:0,strength:0};let n=[0,-.1],r=1,s=e.mesh[33][2]>e.mesh[263][2],a=s?e.mesh[473]:e.mesh[468],o=s?[(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=s?[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],r*(a[1]-o[1])/i[1]-n[1]],u=Math.sqrt(l[0]**2+l[1]**2);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}},Cwe=(e,t)=>{let n=g=>{let y=Math.sqrt(g[0]*g[0]+g[1]*g[1]+g[2]*g[2]);return g[0]/=y,g[1]/=y,g[2]/=y,g},r=(g,y)=>{let A=g[0]-y[0],x=g[1]-y[1],b=g[2]-y[2];return[A,x,b]},s=(g,y)=>{let A=g[1]*y[2]-g[2]*y[1],x=g[2]*y[0]-g[0]*y[2],b=g[0]*y[1]-g[1]*y[0];return[A,x,b]},a=g=>{let[y,A,x,b,v,I,w,S,E]=g,D,$,R;return b<1?b>-1?(R=Math.asin(b),$=Math.atan2(-w,y),D=Math.atan2(-I,v)):(R=-Math.PI/2,$=-Math.atan2(S,E),D=0):(R=Math.PI/2,$=Math.atan2(S,E),D=0),{pitch:2*-D,yaw:2*-$,roll:2*-R}},o=g=>{let y=(x,b,v,I)=>Math.atan2(I-b,v-x);return{pitch:y(g[10][1],g[10][2],g[152][1],g[152][2]),yaw:y(g[33][0],g[33][2],g[263][0],g[263][2]),roll:y(g[33][0],g[33][1],g[263][0],g[263][1])}},i=e.meshRaw;if(!i||i.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 l=Math.max(e.boxRaw[2]*t[0],e.boxRaw[3]*t[1])/1.5,u=[i[10],i[152],i[234],i[454]].map(g=>[g[0]*t[0]/l,g[1]*t[1]/l,g[2]]),c=n(r(u[1],u[0])),d=n(r(u[3],u[2])),h=n(s(d,c));d=s(c,h);let p=[d[0],d[1],d[2],c[0],c[1],c[2],h[0],h[1],h[2]],f=a(p),m=i.length===478?Nwe(e):{bearing:0,strength:0};return{angle:f,matrix:p,gaze:m}},s3=async(e,t)=>{var c,d,h,p,f,m;let n,r,s,a,o,i,l=[];e.state="run:face",n=nt();let u=await m$(t,e.config);if(e.performance.face=Math.trunc(nt()-n),!t.shape||t.shape.length!==4)return[];if(!u)return[];for(let g=0;g<u.length;g++){if(e.analyze("Get Face"),!u[g].image||u[g].image.isDisposedInternal){fe("Face object is disposed:",u[g].image);continue}let y=Cwe(u[g],[t.shape[2],t.shape[1]]);e.analyze("Start Emotion:"),e.config.async?a=e.config.face.emotion.enabled?Ib(u[g].image||ns([]),e.config,g,u.length):{}:(e.state="run:emotion",n=nt(),a=e.config.face.emotion.enabled?await Ib(u[g].image||ns([]),e.config,g,u.length):{},e.performance.emotion=Math.trunc(nt()-n)),e.analyze("End Emotion:"),e.analyze("Start Description:"),e.config.async?i=e.config.face.description.enabled?bb(u[g].image||ns([]),e.config,g,u.length):[]:(e.state="run:description",n=nt(),i=e.config.face.description.enabled?await bb(u[g].image||ns([]),e.config,g,u.length):[],e.performance.embedding=Math.trunc(nt()-n)),e.analyze("End Description:"),e.config.async&&([r,s,a,o,i]=await Promise.all([r,s,a,o,i])),e.analyze("Finish Face:"),!e.config.face.iris.enabled&&((d=(c=u[g])==null?void 0:c.annotations)==null?void 0:d.leftEyeIris)&&((p=(h=u[g])==null?void 0:h.annotations)==null?void 0:p.rightEyeIris)&&(delete u[g].annotations.leftEyeIris,delete u[g].annotations.rightEyeIris);let A=((f=u[g].annotations)==null?void 0:f.leftEyeIris)&&((m=u[g].annotations)==null?void 0:m.rightEyeIris)?Math.max(Math.abs(u[g].annotations.leftEyeIris[3][0]-u[g].annotations.leftEyeIris[1][0]),Math.abs(u[g].annotations.rightEyeIris[4][1]-u[g].annotations.rightEyeIris[2][1]))/t.shape[2]:0;l.push({...u[g],id:g,age:i.age,gender:i.gender,genderScore:i.genderScore,embedding:i.descriptor,emotion:a,iris:A!==0?Math.trunc(500/A/11.7)/100:0,rotation:y,tensor:e.config.face.detector.return?Xn(u[g].image):null}),We(u[g].image),u[g].image&&delete u[g].image,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),l};var j$=e=>{if(!e)return[];let t=[];for(let n=0;n<e.length;n++){let r=e[n].keypoints.find(l=>l.part==="leftWrist"),s=e[n].keypoints.find(l=>l.part==="rightWrist"),a=e[n].keypoints.find(l=>l.part==="nose");a&&r&&s&&r.position.y<a.position.y&&s.position.y<a.position.y?t.push({body:n,gesture:"i give up"}):a&&r&&r.position.y<a.position.y?t.push({body:n,gesture:"raise left hand"}):a&&s&&s.position.y<a.position.y&&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&&t.push({body:n,gesture:`leaning ${o.position.y>i.position.y?"left":"right"}`})}return t},q$=e=>{if(!e)return[];let t=[];for(let n=0;n<e.length;n++)if(e[n].mesh&&e[n].mesh.length>0){let r=e[n].mesh[33][2]-e[n].mesh[263][2];Math.abs(r)<10?t.push({face:n,gesture:"facing center"}):t.push({face:n,gesture:`facing ${r<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 o=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]));o>10&&t.push({face:n,gesture:`mouth ${Math.trunc(o)}% open`});let i=e[n].mesh[152][2];Math.abs(i)>10&&t.push({face:n,gesture:`head ${i<0?"up":"down"}`})}return t},K$=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.rightEyeIris)continue;let r=e[n].annotations.leftEyeIris[3][0]-e[n].annotations.leftEyeIris[1][0],s=e[n].annotations.leftEyeIris[4][1]-e[n].annotations.leftEyeIris[2][1],a=Math.abs(r*s),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 d=Math.abs(e[n].mesh[33][0]-e[n].annotations.rightEyeIris[0][0])/e[n].box[2],h=Math.abs(e[n].mesh[263][0]-e[n].annotations.leftEyeIris[0][0])/e[n].box[2];(h>.06||d>.06)&&(u=!1),h>.06&&t.push({iris:n,gesture:"looking right"}),d>.06&&t.push({iris:n,gesture:"looking left"});let p=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||p<.01||f>.022||p>.022)&&(u=!1),(f<.01||p<.01)&&t.push({iris:n,gesture:"looking down"}),(f>.022||p>.022)&&t.push({iris:n,gesture:"looking up"}),u&&t.push({iris:n,gesture:"looking center"})}return t},X$=e=>{if(!e)return[];let t=[];for(let n=0;n<e.length;n++){let r=[];for(let[s,a]of Object.entries(e[n].annotations))s!=="palmBase"&&Array.isArray(a)&&r.push({name:s.toLowerCase(),position:a[0]});if(r&&r.length>0){let s=r.reduce((o,i)=>o.position[2]<i.position[2]?o:i),a=r.reduce((o,i)=>o.position[1]<i.position[1]?o:i);t.push({hand:n,gesture:`${s.name} forward ${a.name} up`})}}return t};var i3={};Xg(i3,{all:()=>Rwe,body:()=>J$,canvas:()=>$we,face:()=>Y$,gesture:()=>Z$,hand:()=>Q$,object:()=>eR,options:()=>oo,person:()=>Ewe});var oo={color:"rgba(173, 216, 230, 0.6)",labelColor:"rgba(173, 216, 230, 1)",shadowColor:"black",font:'small-caps 14px "Segoe UI"',lineHeight:18,lineWidth:4,pointSize:2,roundRect:8,drawPoints:!1,drawLabels:!0,drawBoxes:!0,drawPolygons:!0,drawGaze:!0,fillPolygons:!1,useDepth:!0,useCurves:!1,bufferedOutput:!0},m0=e=>Math.round(e*180/Math.PI);function a3(e,t,n,r=0,s){e.fillStyle=s.useDepth&&r?`rgba(${127.5+2*r}, ${127.5-2*r}, 255, 0.3)`:s.color,e.beginPath(),e.arc(t,n,s.pointSize,0,2*Math.PI),e.fill()}function Ih(e,t,n,r,s,a){if(e.beginPath(),a.useCurves){let o=(t+t+r)/2,i=(n+n+s)/2;e.ellipse(o,i,r/2,s/2,0,0,2*Math.PI)}else e.lineWidth=a.lineWidth,e.moveTo(t+a.roundRect,n),e.lineTo(t+r-a.roundRect,n),e.quadraticCurveTo(t+r,n,t+r,n+a.roundRect),e.lineTo(t+r,n+s-a.roundRect),e.quadraticCurveTo(t+r,n+s,t+r-a.roundRect,n+s),e.lineTo(t+a.roundRect,n+s),e.quadraticCurveTo(t,n+s,t,n+s-a.roundRect),e.lineTo(t,n+a.roundRect),e.quadraticCurveTo(t,n,t+a.roundRect,n),e.closePath();e.stroke()}function o3(e,t=[],n){if(!(t===void 0||t.length===0)){e.beginPath(),e.moveTo(t[0][0],t[0][1]);for(let r of t){let s=r[2]||0;e.strokeStyle=n.useDepth&&s?`rgba(${127.5+2*s}, ${127.5-2*s}, 255, 0.3)`:n.color,e.fillStyle=n.useDepth&&s?`rgba(${127.5+2*s}, ${127.5-2*s}, 255, 0.3)`:n.color,e.lineTo(r[0],Math.round(r[1]))}e.stroke(),n.fillPolygons&&(e.closePath(),e.fill())}}function Sh(e,t=[],n){if(!(t===void 0||t.length===0)){if(!n.useCurves||t.length<=2){o3(e,t,n);return}e.moveTo(t[0][0],t[0][1]);for(let r=0;r<t.length-2;r++){let s=(t[r][0]+t[r+1][0])/2,a=(t[r][1]+t[r+1][1])/2;e.quadraticCurveTo(t[r][0],t[r][1],s,a)}e.quadraticCurveTo(t[t.length-2][0],t[t.length-2][1],t[t.length-1][0],t[t.length-1][1]),e.stroke(),n.fillPolygons&&(e.closePath(),e.fill())}}async function Z$(e,t,n){let r=Fn(oo,n);if(!t||!e||!(e instanceof HTMLCanvasElement))return;let s=e.getContext("2d");if(!s)return;s.font=r.font,s.fillStyle=r.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}: ${l[1]}`;r.shadowColor&&r.shadowColor!==""&&(s.fillStyle=r.shadowColor,s.fillText(c,8,2+a*r.lineHeight)),s.fillStyle=r.labelColor,s.fillText(c,6,0+a*r.lineHeight),a+=1}}}async function Y$(e,t,n){var a,o,i,l;let r=Fn(oo,n);if(!t||!e||!(e instanceof HTMLCanvasElement))return;let s=e.getContext("2d");if(!!s)for(let u of t){s.font=r.font,s.strokeStyle=r.color,s.fillStyle=r.color,r.drawBoxes&&Ih(s,u.box[0],u.box[1],u.box[2],u.box[3],r);let c=[];if(c.push(`face: ${Math.trunc(100*u.score)}%`),u.genderScore&&c.push(`${u.gender||""} ${Math.trunc(100*u.genderScore)}%`),u.age&&c.push(`age: ${u.age||""}`),u.iris&&c.push(`distance: ${u.iris}`),u.emotion&&u.emotion.length>0){let d=u.emotion.map(h=>`${Math.trunc(100*h.score)}% ${h.emotion}`);d.length>3&&(d.length=3),c.push(d.join(" "))}u.rotation&&u.rotation.angle&&u.rotation.gaze&&(u.rotation.angle.roll&&c.push(`roll: ${m0(u.rotation.angle.roll)}\xB0 yaw:${m0(u.rotation.angle.yaw)}\xB0 pitch:${m0(u.rotation.angle.pitch)}\xB0`),u.rotation.gaze.bearing&&c.push(`gaze: ${m0(u.rotation.gaze.bearing)}\xB0`)),c.length===0&&c.push("face"),s.fillStyle=r.color;for(let d=c.length-1;d>=0;d--){let h=Math.max(u.box[0],0),p=d*r.lineHeight+u.box[1];r.shadowColor&&r.shadowColor!==""&&(s.fillStyle=r.shadowColor,s.fillText(c[d],h+5,p+16)),s.fillStyle=r.labelColor,s.fillText(c[d],h+4,p+15)}if(s.lineWidth=1,u.mesh&&u.mesh.length>0){if(r.drawPoints)for(let d of u.mesh)a3(s,d[0],d[1],d[2],r);if(r.drawPolygons){s.lineWidth=1;for(let d=0;d<bi.length/3;d++){let h=[bi[d*3+0],bi[d*3+1],bi[d*3+2]].map(p=>u.mesh[p]);o3(s,h,r)}if(u.annotations&&u.annotations.leftEyeIris){s.strokeStyle=r.useDepth?"rgba(255, 200, 255, 0.3)":r.color,s.beginPath();let d=Math.abs(u.annotations.leftEyeIris[3][0]-u.annotations.leftEyeIris[1][0])/2,h=Math.abs(u.annotations.leftEyeIris[4][1]-u.annotations.leftEyeIris[2][1])/2;s.ellipse(u.annotations.leftEyeIris[0][0],u.annotations.leftEyeIris[0][1],d,h,0,0,2*Math.PI),s.stroke(),r.fillPolygons&&(s.fillStyle=r.useDepth?"rgba(255, 255, 200, 0.3)":r.color,s.fill())}if(u.annotations&&u.annotations.rightEyeIris){s.strokeStyle=r.useDepth?"rgba(255, 200, 255, 0.3)":r.color,s.beginPath();let d=Math.abs(u.annotations.rightEyeIris[3][0]-u.annotations.rightEyeIris[1][0])/2,h=Math.abs(u.annotations.rightEyeIris[4][1]-u.annotations.rightEyeIris[2][1])/2;s.ellipse(u.annotations.rightEyeIris[0][0],u.annotations.rightEyeIris[0][1],d,h,0,0,2*Math.PI),s.stroke(),r.fillPolygons&&(s.fillStyle=r.useDepth?"rgba(255, 255, 200, 0.3)":r.color,s.fill())}if(r.drawGaze&&((o=(a=u.rotation)==null?void 0:a.gaze)==null?void 0:o.strength)&&((l=(i=u.rotation)==null?void 0:i.gaze)==null?void 0:l.bearing)&&u.annotations.leftEyeIris&&u.annotations.rightEyeIris&&u.annotations.leftEyeIris[0]&&u.annotations.rightEyeIris[0]){s.strokeStyle="pink",s.beginPath();let d=[u.annotations.leftEyeIris[0][0]+Math.sin(u.rotation.gaze.bearing)*u.rotation.gaze.strength*u.box[3],u.annotations.leftEyeIris[0][1]+Math.cos(u.rotation.gaze.bearing)*u.rotation.gaze.strength*u.box[2]];s.moveTo(u.annotations.leftEyeIris[0][0],u.annotations.leftEyeIris[0][1]),s.lineTo(d[0],d[1]);let h=[u.annotations.rightEyeIris[0][0]+Math.sin(u.rotation.gaze.bearing)*u.rotation.gaze.strength*u.box[3],u.annotations.rightEyeIris[0][1]+Math.cos(u.rotation.gaze.bearing)*u.rotation.gaze.strength*u.box[2]];s.moveTo(u.annotations.rightEyeIris[0][0],u.annotations.rightEyeIris[0][1]),s.lineTo(h[0],h[1]),s.stroke()}}}}}async function J$(e,t,n){var a;let r=Fn(oo,n);if(!t||!e||!(e instanceof HTMLCanvasElement))return;let s=e.getContext("2d");if(!!s){s.lineJoin="round";for(let o=0;o<t.length;o++){if(s.strokeStyle=r.color,s.fillStyle=r.color,s.lineWidth=r.lineWidth,s.font=r.font,r.drawBoxes&&t[o].box&&((a=t[o].box)==null?void 0:a.length)===4&&(Ih(s,t[o].box[0],t[o].box[1],t[o].box[2],t[o].box[3],r),r.drawLabels&&(r.shadowColor&&r.shadowColor!==""&&(s.fillStyle=r.shadowColor,s.fillText(`body ${100*t[o].score}%`,t[o].box[0]+3,1+t[o].box[1]+r.lineHeight,t[o].box[2])),s.fillStyle=r.labelColor,s.fillText(`body ${100*t[o].score}%`,t[o].box[0]+2,0+t[o].box[1]+r.lineHeight,t[o].box[2]))),r.drawPoints)for(let i=0;i<t[o].keypoints.length;i++)s.fillStyle=r.useDepth&&t[o].keypoints[i].position[2]?`rgba(${127.5+2*(t[o].keypoints[i].position[2]||0)}, ${127.5-2*(t[o].keypoints[i].position[2]||0)}, 255, 0.5)`:r.color,a3(s,t[o].keypoints[i].position[0],t[o].keypoints[i].position[1],0,r);if(r.drawLabels&&(s.font=r.font,t[o].keypoints))for(let i of t[o].keypoints)s.fillStyle=r.useDepth&&i.position[2]?`rgba(${127.5+2*i.position[2]}, ${127.5-2*i.position[2]}, 255, 0.5)`:r.color,s.fillText(`${i.part} ${Math.trunc(100*i.score)}%`,i.position[0]+4,i.position[1]+4);if(r.drawPolygons&&t[o].keypoints){let i,l=[];l.length=0,i=t[o].keypoints.find(u=>u.part==="leftShoulder"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightShoulder"),i&&l.push([i.position[0],i.position[1]]),Sh(s,l,r),l.length=0,i=t[o].keypoints.find(u=>u.part==="rightShoulder"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightHip"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftHip"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftShoulder"),i&&l.push([i.position[0],i.position[1]]),l.length===4&&o3(s,l,r),l.length=0,i=t[o].keypoints.find(u=>u.part==="leftHip"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftKnee"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftAnkle"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftHeel"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftFoot"),i&&l.push([i.position[0],i.position[1]]),Sh(s,l,r),l.length=0,i=t[o].keypoints.find(u=>u.part==="rightHip"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightKnee"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightAnkle"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightHeel"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightFoot"),i&&l.push([i.position[0],i.position[1]]),Sh(s,l,r),l.length=0,i=t[o].keypoints.find(u=>u.part==="leftShoulder"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftElbow"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftWrist"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="leftPalm"),i&&l.push([i.position[0],i.position[1]]),Sh(s,l,r),l.length=0,i=t[o].keypoints.find(u=>u.part==="rightShoulder"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightElbow"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightWrist"),i&&l.push([i.position[0],i.position[1]]),i=t[o].keypoints.find(u=>u.part==="rightPalm"),i&&l.push([i.position[0],i.position[1]]),Sh(s,l,r)}}}}async function Q$(e,t,n){let r=Fn(oo,n);if(!t||!e||!(e instanceof HTMLCanvasElement))return;let s=e.getContext("2d");if(!!s){s.lineJoin="round",s.font=r.font;for(let a of t){if(r.drawBoxes&&(s.strokeStyle=r.color,s.fillStyle=r.color,Ih(s,a.box[0],a.box[1],a.box[2],a.box[3],r),r.drawLabels&&(r.shadowColor&&r.shadowColor!==""&&(s.fillStyle=r.shadowColor,s.fillText("hand",a.box[0]+3,1+a.box[1]+r.lineHeight,a.box[2])),s.fillStyle=r.labelColor,s.fillText("hand",a.box[0]+2,0+a.box[1]+r.lineHeight,a.box[2])),s.stroke()),r.drawPoints&&a.keypoints&&a.keypoints.length>0)for(let o of a.keypoints)s.fillStyle=r.useDepth?`rgba(${127.5+2*o[2]}, ${127.5-2*o[2]}, 255, 0.5)`:r.color,a3(s,o[0],o[1],0,r);if(r.drawLabels){let o=(i,l)=>{s.fillStyle=r.useDepth?`rgba(${127.5+2*i[i.length-1][2]}, ${127.5-2*i[i.length-1][2]}, 255, 0.5)`:r.color,s.fillText(l,i[i.length-1][0]+4,i[i.length-1][1]+4)};s.font=r.font,o(a.annotations.indexFinger,"index"),o(a.annotations.middleFinger,"middle"),o(a.annotations.ringFinger,"ring"),o(a.annotations.pinky,"pinky"),o(a.annotations.thumb,"thumb"),o(a.annotations.palmBase,"palm")}if(r.drawPolygons){let o=i=>{if(!!i)for(let l=0;l<i.length;l++)s.beginPath(),s.strokeStyle=r.useDepth?`rgba(${127.5+2*i[l][2]}, ${127.5-2*i[l][2]}, 255, 0.5)`:r.color,s.moveTo(i[l>0?l-1:0][0],i[l>0?l-1:0][1]),s.lineTo(i[l][0],i[l][1]),s.stroke()};s.lineWidth=r.lineWidth,o(a.annotations.indexFinger),o(a.annotations.middleFinger),o(a.annotations.ringFinger),o(a.annotations.pinky),o(a.annotations.thumb)}}}}async function eR(e,t,n){let r=Fn(oo,n);if(!t||!e||!(e instanceof HTMLCanvasElement))return;let s=e.getContext("2d");if(!!s){s.lineJoin="round",s.font=r.font;for(let a of t)if(r.drawBoxes){if(s.strokeStyle=r.color,s.fillStyle=r.color,Ih(s,a.box[0],a.box[1],a.box[2],a.box[3],r),r.drawLabels){let o=`${a.label} ${Math.round(100*a.score)}%`;r.shadowColor&&r.shadowColor!==""&&(s.fillStyle=r.shadowColor,s.fillText(o,a.box[0]+3,1+a.box[1]+r.lineHeight,a.box[2])),s.fillStyle=r.labelColor,s.fillText(o,a.box[0]+2,0+a.box[1]+r.lineHeight,a.box[2])}s.stroke()}}}async function Ewe(e,t,n){let r=Fn(oo,n);if(!t||!e||!(e instanceof HTMLCanvasElement))return;let s=e.getContext("2d");if(!!s){s.lineJoin="round",s.font=r.font;for(let a=0;a<t.length;a++)if(r.drawBoxes){if(s.strokeStyle=r.color,s.fillStyle=r.color,Ih(s,t[a].box[0],t[a].box[1],t[a].box[2],t[a].box[3],r),r.drawLabels){let o=`person #${a}`;r.shadowColor&&r.shadowColor!==""&&(s.fillStyle=r.shadowColor,s.fillText(o,t[a].box[0]+3,1+t[a].box[1]+r.lineHeight,t[a].box[2])),s.fillStyle=r.labelColor,s.fillText(o,t[a].box[0]+2,0+t[a].box[1]+r.lineHeight,t[a].box[2])}s.stroke()}}}async function $we(e,t){if(!e||!t||!(e instanceof HTMLCanvasElement)||!(t instanceof HTMLCanvasElement))return;let n=e.getContext("2d");n==null||n.drawImage(e,0,0)}async function Rwe(e,t,n){let r=nt(),s=Fn(oo,n);!t||!e||e instanceof HTMLCanvasElement&&(Y$(e,t.face,s),J$(e,t.body,s),Q$(e,t.hand,s),eR(e,t.object,s),Z$(e,t.gesture,s),t.performance.draw=Math.trunc(nt()-r))}function tR(e,t,n,r,s){var i,l,u,c,d,h,p,f,m,g,y,A,x,b,v,I;let a=0,o=[];for(let w of e){let S={id:a++,face:w,body:null,hands:{left:null,right:null},gestures:[],box:[0,0,0,0]};for(let M of t)w.box[0]>M.box[0]&&w.box[0]<M.box[0]+M.box[2]&&w.box[1]+w.box[3]>M.box[1]&&w.box[1]+w.box[3]<M.box[1]+M.box[3]&&(S.body=M);if(S.body)for(let M of n)M.box[0]+M.box[2]>S.body.box[0]&&M.box[0]+M.box[2]<S.body.box[0]+S.body.box[2]&&M.box[1]+M.box[3]>S.body.box[1]&&M.box[1]+M.box[3]<S.body.box[1]+S.body.box[3]&&S.hands&&(S.hands.left=M),M.box[0]<S.body.box[0]+S.body.box[2]&&M.box[0]>S.body.box[0]&&M.box[1]+M.box[3]>S.body.box[1]&&M.box[1]+M.box[3]<S.body.box[1]+S.body.box[3]&&S.hands&&(S.hands.right=M);for(let M of r)M.face!==void 0&&M.face===w.id?(i=S.gestures)==null||i.push(M):M.iris!==void 0&&M.iris===w.id?(l=S.gestures)==null||l.push(M):M.body!==void 0&&M.body===((u=S.body)==null?void 0:u.id)?(c=S.gestures)==null||c.push(M):M.hand!==void 0&&M.hand===((h=(d=S.hands)==null?void 0:d.left)==null?void 0:h.id)?(p=S.gestures)==null||p.push(M):M.hand!==void 0&&M.hand===((m=(f=S.hands)==null?void 0:f.right)==null?void 0:m.id)&&((g=S.gestures)==null||g.push(M));let E=[],D=[],$=M=>{M&&M.length===4&&(E.push(M[0],M[0]+M[2]),D.push(M[1],M[1]+M[3]))};$((y=S.face)==null?void 0:y.box),$((A=S.body)==null?void 0:A.box),$((b=(x=S.hands)==null?void 0:x.left)==null?void 0:b.box),$((I=(v=S.hands)==null?void 0:v.right)==null?void 0:I.box);let R=Math.min(...E),N=Math.min(...D);S.box=[R,N,Math.max(...E)-R,Math.max(...D)-N],s&&s.length===4&&(S.boxRaw=[S.box[0]/s[2],S.box[1]/s[1],S.box[2]/s[2],S.box[3]/s[1]]),o.push(S)}return o}var Le={face:[],body:[],hand:[],gesture:[],object:[],persons:[],performance:{},timestamp:0};function nR(e){var r,s,a,o,i,l,u,c,d,h,p,f,m,g,y,A,x,b,v,I,w;let t=Date.now()-e.timestamp,n=t<1e3?8-Math.log(t):1;if(Le.canvas=e.canvas,!Le.body||e.body.length!==Le.body.length)Le.body=JSON.parse(JSON.stringify(e.body));else for(let S=0;S<e.body.length;S++){let E=e.body[S].box.map((R,N)=>((n-1)*Le.body[S].box[N]+R)/n),D=e.body[S].boxRaw.map((R,N)=>((n-1)*Le.body[S].boxRaw[N]+R)/n),$=e.body[S].keypoints.map((R,N)=>({score:R.score,part:R.part,position:[Le.body[S].keypoints[N]?((n-1)*Le.body[S].keypoints[N].position[0]+R.position[0])/n:R.position[0],Le.body[S].keypoints[N]?((n-1)*Le.body[S].keypoints[N].position[1]+R.position[1])/n:R.position[1]],positionRaw:[Le.body[S].keypoints[N]?((n-1)*Le.body[S].keypoints[N].positionRaw[0]+R.positionRaw[0])/n:R.position[0],Le.body[S].keypoints[N]?((n-1)*Le.body[S].keypoints[N].positionRaw[1]+R.positionRaw[1])/n:R.position[1]]}));Le.body[S]={...e.body[S],box:E,boxRaw:D,keypoints:$}}if(!Le.hand||e.hand.length!==Le.hand.length)Le.hand=JSON.parse(JSON.stringify(e.hand));else for(let S=0;S<e.hand.length;S++){let E=e.hand[S].box.map((M,B)=>((n-1)*Le.hand[S].box[B]+M)/n),D=e.hand[S].boxRaw.map((M,B)=>((n-1)*Le.hand[S].boxRaw[B]+M)/n),$=e.hand[S].keypoints.map((M,B)=>M.map((q,X)=>((n-1)*Le.hand[S].keypoints[B][X]+q)/n)),R=Object.keys(e.hand[S].annotations),N={};for(let M of R)N[M]=e.hand[S].annotations[M].map((B,q)=>B.map((X,J)=>((n-1)*Le.hand[S].annotations[M][q][J]+X)/n));Le.hand[S]={...e.hand[S],box:E,boxRaw:D,keypoints:$,annotations:N}}if(!Le.face||e.face.length!==Le.face.length)Le.face=JSON.parse(JSON.stringify(e.face));else for(let S=0;S<e.face.length;S++){let E=e.face[S].box.map((R,N)=>((n-1)*Le.face[S].box[N]+R)/n),D=e.face[S].boxRaw.map((R,N)=>((n-1)*Le.face[S].boxRaw[N]+R)/n),$={matrix:[0,0,0,0,0,0,0,0,0],angle:{roll:0,yaw:0,pitch:0},gaze:{bearing:0,strength:0}};$.matrix=(r=e.face[S].rotation)==null?void 0:r.matrix,$.angle={roll:((n-1)*(((a=(s=Le.face[S].rotation)==null?void 0:s.angle)==null?void 0:a.roll)||0)+(((i=(o=e.face[S].rotation)==null?void 0:o.angle)==null?void 0:i.roll)||0))/n,yaw:((n-1)*(((u=(l=Le.face[S].rotation)==null?void 0:l.angle)==null?void 0:u.yaw)||0)+(((d=(c=e.face[S].rotation)==null?void 0:c.angle)==null?void 0:d.yaw)||0))/n,pitch:((n-1)*(((p=(h=Le.face[S].rotation)==null?void 0:h.angle)==null?void 0:p.pitch)||0)+(((m=(f=e.face[S].rotation)==null?void 0:f.angle)==null?void 0:m.pitch)||0))/n},$.gaze={bearing:((n-1)*(((y=(g=Le.face[S].rotation)==null?void 0:g.gaze)==null?void 0:y.bearing)||0)+(((x=(A=e.face[S].rotation)==null?void 0:A.gaze)==null?void 0:x.bearing)||0))/n,strength:((n-1)*(((v=(b=Le.face[S].rotation)==null?void 0:b.gaze)==null?void 0:v.strength)||0)+(((w=(I=e.face[S].rotation)==null?void 0:I.gaze)==null?void 0:w.strength)||0))/n},Le.face[S]={...e.face[S],rotation:$,box:E,boxRaw:D}}if(!Le.object||e.object.length!==Le.object.length)Le.object=JSON.parse(JSON.stringify(e.object));else for(let S=0;S<e.object.length;S++){let E=e.object[S].box.map(($,R)=>((n-1)*Le.object[S].box[R]+$)/n),D=e.object[S].boxRaw.map(($,R)=>((n-1)*Le.object[S].boxRaw[R]+$)/n);Le.object[S]={...e.object[S],box:E,boxRaw:D}}if(e.persons){let S=e.persons;if(!Le.persons||S.length!==Le.persons.length)Le.persons=JSON.parse(JSON.stringify(S));else for(let E=0;E<S.length;E++)Le.persons[E].box=S[E].box.map((D,$)=>((n-1)*Le.persons[E].box[$]+D)/n)}return e.gesture&&(Le.gesture=e.gesture),e.performance&&(Le.performance=e.performance),Le}var g0=`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==`,y0=`
/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==`;var rR="2.0.3";var _u,Th,Nh,wi,ki,Du,A0,Ch,x0,b0,v0,w0,sR=class{constructor(t){wr(this,_u,void 0);wr(this,Th,void 0);wr(this,Nh,void 0);wr(this,wi,void 0);wr(this,ki,void 0);wr(this,Du,void 0);this.analyze=(...t)=>{if(!Dn(this,Th))return;let n=this.tf.engine().state.numTensors,r=Dn(this,_u);es(this,_u,n);let s=n-r;s!==0&&fe(...t,s)};wr(this,A0,t=>{if(!Dn(this,Nh))return null;if(!t)return"input is not defined";if(this.tf.ENV.flags.IS_NODE&&!(t instanceof It))return"input must be a tensor";try{this.tf.getBackend()}catch(n){return"backend not loaded"}return null});wr(this,Ch,async(t=!1)=>{var n;if(this.config.backend&&this.config.backend.length>0&&t||this.tf.getBackend()!==this.config.backend){let r=nt();if(this.state="backend",this.config.backend&&this.config.backend.length>0){if(typeof window=="undefined"&&typeof WorkerGlobalScope!="undefined"&&this.config.debug&&fe("running inside web worker"),this.tf.ENV.flags.IS_BROWSER&&this.config.backend==="tensorflow"&&(this.config.backend="webgl"),this.tf.ENV.flags.IS_NODE&&(this.config.backend==="webgl"||this.config.backend==="humangl")&&(this.config.backend="tensorflow"),this.config.debug&&fe("setting backend:",this.config.backend),this.config.backend==="wasm"){if(this.config.debug&&fe("wasm path:",this.config.wasmPath),typeof((n=this.tf)==null?void 0:n.setWasmPaths)!="undefined")this.tf.setWasmPaths(this.config.wasmPath);else throw new Error("Human: WASM backend is not loaded");let s=await this.tf.env().getAsync("WASM_HAS_SIMD_SUPPORT"),a=await this.tf.env().getAsync("WASM_HAS_MULTITHREAD_SUPPORT");this.config.debug&&fe(`wasm execution: ${s?"SIMD":"no SIMD"} ${a?"multithreaded":"singlethreaded"}`),this.config.debug&&!s&&fe("warning: wasm simd support is not enabled")}this.config.backend==="humangl"&&r$();try{await this.tf.setBackend(this.config.backend)}catch(s){fe("error: cannot set backend:",this.config.backend,s)}}if(this.tf.enableProdMode(),this.tf.getBackend()==="webgl"||this.tf.getBackend()==="humangl"){this.tf.ENV.set("CHECK_COMPUTATION_FOR_ERRORS",!1),this.tf.ENV.set("WEBGL_CPU_FORWARD",!0),this.tf.ENV.set("WEBGL_PACK_DEPTHWISECONV",!0),typeof this.config.deallocate!="undefined"&&this.config.deallocate&&(fe("changing webgl: WEBGL_DELETE_TEXTURE_THRESHOLD:",!0),this.tf.ENV.set("WEBGL_DELETE_TEXTURE_THRESHOLD",0));let s=await this.tf.backend().getGPGPUContext().gl;this.config.debug&&fe(`gl version:${s.getParameter(s.VERSION)} renderer:${s.getParameter(s.RENDERER)}`)}await this.tf.ready(),this.performance.backend=Math.trunc(nt()-r)}});this.next=t=>nR(t||this.result);wr(this,x0,async t=>{if(this.config.cacheSensitivity===0)return!1;let n=32,r=t.resizeBilinear([Math.trunc(t.shape[1]/n),Math.trunc(t.shape[2]/n)]),s=r.dataSync(),a=0;for(let l=0;l<s.length/3;l++)a+=s[3*l+2];r.dispose();let o=100*(Math.max(a,Dn(this,ki))/Math.min(a,Dn(this,ki))-1);es(this,ki,a);let i=o<Math.max(this.config.cacheSensitivity,Dn(this,Du));return es(this,Du,o>10*this.config.cacheSensitivity?0:o),i});wr(this,b0,async()=>{let t=(s,a="application/octet-stream")=>fetch(`data:${a};base64,${s}`).then(o=>o.blob()),n,r;switch(this.config.warmup){case"face":n=await t(g0);break;case"full":n=await t(y0);break;default:n=null}if(n){let s=await createImageBitmap(n);r=await this.detect(s,this.config),s.close()}return r});wr(this,v0,async()=>new Promise(t=>{let n,r=0;switch(this.config.warmup){case"face":r=256,n="data:image/jpeg;base64,"+g0;break;case"full":case"body":r=1200,n="data:image/jpeg;base64,"+y0;break;default:n=null}let s=new Image;s.onload=async()=>{let a=typeof OffscreenCanvas!="undefined"?new OffscreenCanvas(r,r):document.createElement("canvas");a.width=s.naturalWidth,a.height=s.naturalHeight;let o=a.getContext("2d");o==null||o.drawImage(s,0,0);let i=await this.detect(a,this.config);t(i)},n?s.src=n:t(null)}));wr(this,w0,async()=>{let t=s=>Buffer.from(s,"base64"),n;if(this.config.warmup==="face"&&(n=t(g0)),(this.config.warmup==="body"||this.config.warmup==="full")&&(n=t(y0)),!n)return null;let r;if(typeof void 0!="undefined"){let s=(void 0).decodeJpeg(n),a=s.expandDims(0);this.tf.dispose(s),r=await this.detect(a,this.config),this.tf.dispose(a)}else this.config.debug&&fe("Warmup tfjs-node not loaded");return r});this.config=Fn(w3,t||{}),this.tf=Ah,this.draw=i3,this.version=rR,this.state="idle",es(this,_u,0),es(this,Th,!1),es(this,Nh,!1),es(this,wi,!0),es(this,Du,0),this.performance={backend:0,load:0,image:0,frames:0,cached:0,changed:0,total:0,draw:0},this.models={face:null,posenet:null,blazepose:null,efficientpose:null,movenet:null,handpose:null,age:null,gender:null,emotion:null,embedding:null,nanodet:null,centernet:null,faceres:null,segmentation:null},this.image=n=>vi(n,this.config),this.faceTriangulation=g$,this.faceUVMap=y$,this.sysinfo=k3(),es(this,ki,1)}similarity(t,n){return Ab(t,n)}segmentation(t,n){return H$(t,n,this.config)}enhance(t){return xb(t)}match(t,n,r=0){return x$(t,n,r)}async load(t){this.state="load";let n=nt();t&&(this.config=Fn(this.config,t)),Dn(this,wi)&&(this.config.debug&&fe(`version: ${this.version}`),this.config.debug&&fe(`tfjs version: ${this.tf.version_core}`),this.config.debug&&fe("platform:",this.sysinfo.platform),this.config.debug&&fe("agent:",this.sysinfo.agent),await Dn(this,Ch).call(this,!0),this.tf.ENV.flags.IS_BROWSER&&(this.config.debug&&fe("configuration:",this.config),this.config.debug&&fe("tf flags:",this.tf.ENV.flags))),await G$(this),Dn(this,wi)&&(this.config.debug&&fe("tf engine state:",this.tf.engine().state.numBytes,"bytes",this.tf.engine().state.numTensors,"tensors"),es(this,wi,!1));let r=Math.trunc(nt()-n);r>(this.performance.load||0)&&(this.performance.load=r)}async detect(t,n){return new Promise(async r=>{this.state="config";let s,a;this.config=Fn(this.config,n),this.state="check";let o=Dn(this,A0).call(this,t);o&&(fe(o,t),r({error:o}));let i=nt();await Dn(this,Ch).call(this),await this.load(),s=nt();let l=vi(t,this.config);if(this.performance.image=Math.trunc(nt()-s),this.analyze("Get Image:"),this.config.segmentation.enabled&&l&&l.tensor&&(this.analyze("Start Segmentation:"),this.state="run:segmentation",s=nt(),await r3(l),a=Math.trunc(nt()-s),a>0&&(this.performance.segmentation=a),l.canvas&&(l.tensor.dispose(),l=vi(l.canvas,this.config)),this.analyze("End Segmentation:")),!l||!l.tensor){fe("could not convert input to tensor"),r({error:"could not convert input to tensor"});return}s=nt(),this.config.skipFrame=await Dn(this,x0).call(this,l.tensor),this.performance.frames||(this.performance.frames=0),this.performance.cached||(this.performance.cached=0),this.performance.frames++,this.config.skipFrame&&this.performance.cached++,this.performance.changed=Math.trunc(nt()-s),this.analyze("Check Changed:");let u,c,d,h;this.config.async?(u=this.config.face.enabled?s3(this,l.tensor):[],this.performance.face&&delete this.performance.face):(this.state="run:face",s=nt(),u=this.config.face.enabled?await s3(this,l.tensor):[],a=Math.trunc(nt()-s),a>0&&(this.performance.face=a)),this.analyze("Start Body:"),this.config.async?(this.config.body.modelPath.includes("posenet")?c=this.config.body.enabled?$b(l.tensor,this.config):[]:this.config.body.modelPath.includes("blazepose")?c=this.config.body.enabled?zb(l.tensor,this.config):[]:this.config.body.modelPath.includes("efficientpose")?c=this.config.body.enabled?Vb(l.tensor,this.config):[]:this.config.body.modelPath.includes("movenet")&&(c=this.config.body.enabled?qb(l.tensor,this.config):[]),this.performance.body&&delete this.performance.body):(this.state="run:body",s=nt(),this.config.body.modelPath.includes("posenet")?c=this.config.body.enabled?await $b(l.tensor,this.config):[]:this.config.body.modelPath.includes("blazepose")?c=this.config.body.enabled?await zb(l.tensor,this.config):[]:this.config.body.modelPath.includes("efficientpose")?c=this.config.body.enabled?await Vb(l.tensor,this.config):[]:this.config.body.modelPath.includes("movenet")&&(c=this.config.body.enabled?await qb(l.tensor,this.config):[]),a=Math.trunc(nt()-s),a>0&&(this.performance.body=a)),this.analyze("End Body:"),this.analyze("Start Hand:"),this.config.async?(d=this.config.hand.enabled?Ob(l.tensor,this.config):[],this.performance.hand&&delete this.performance.hand):(this.state="run:hand",s=nt(),d=this.config.hand.enabled?await Ob(l.tensor,this.config):[],a=Math.trunc(nt()-s),a>0&&(this.performance.hand=a)),this.analyze("End Hand:"),this.analyze("Start Object:"),this.config.async?(this.config.object.modelPath.includes("nanodet")?h=this.config.object.enabled?Yb(l.tensor,this.config):[]:this.config.object.modelPath.includes("centernet")&&(h=this.config.object.enabled?t3(l.tensor,this.config):[]),this.performance.object&&delete this.performance.object):(this.state="run:object",s=nt(),this.config.object.modelPath.includes("nanodet")?h=this.config.object.enabled?await Yb(l.tensor,this.config):[]:this.config.object.modelPath.includes("centernet")&&(h=this.config.object.enabled?await t3(l.tensor,this.config):[]),a=Math.trunc(nt()-s),a>0&&(this.performance.object=a)),this.analyze("End Object:"),this.config.async&&([u,c,d,h]=await Promise.all([u,c,d,h]));let p=[];this.config.gesture.enabled&&(s=nt(),p=[...q$(u),...j$(c),...X$(d),...K$(u)],this.config.async?this.performance.gesture&&delete this.performance.gesture:this.performance.gesture=Math.trunc(nt()-s)),this.performance.total=Math.trunc(nt()-i),this.state="idle",this.result={face:u,body:c,hand:d,gesture:p,object:h,performance:this.performance,canvas:l.canvas,timestamp:Date.now(),get persons(){var f;return tR(u,c,d,p,(f=l==null?void 0:l.tensor)==null?void 0:f.shape)}},We(l.tensor),r(this.result)})}async warmup(t){let n=nt();if(t&&(this.config=Fn(this.config,t)),!this.config.warmup||this.config.warmup==="none")return{error:"null"};let r;typeof createImageBitmap=="function"?r=await Dn(this,b0).call(this):typeof Image!="undefined"?r=await Dn(this,v0).call(this):r=await Dn(this,w0).call(this);let s=nt();return this.config.debug&&fe("Warmup",this.config.warmup,Math.round(s-n),"ms",r),r}};_u=new WeakMap,Th=new WeakMap,Nh=new WeakMap,wi=new WeakMap,ki=new WeakMap,Du=new WeakMap,A0=new WeakMap,Ch=new WeakMap,x0=new WeakMap,b0=new WeakMap,v0=new WeakMap,w0=new WeakMap;return Dwe;})();
/**
* @license
* Copyright 2017 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/**
* @license
* Copyright 2018 Google LLC
*
* Use of this source code is governed by an MIT-style
* license that can be found in the LICENSE file or at
* https://opensource.org/licenses/MIT.
* =============================================================================
*/
/**
* @license
* Copyright 2018 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
* =============================================================================
*/
/**
* @license
* Copyright 2018 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/**
* @license
* Copyright 2019 Google LLC
*
* Use of this source code is governed by an MIT-style
* license that can be found in the LICENSE file or at
* https://opensource.org/licenses/MIT.
* =============================================================================
*/
/**
* @license
* Copyright 2019 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
* =============================================================================
*/
/**
* @license
* Copyright 2019 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/**
* @license
* Copyright 2020 Google Inc. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/**
* @license
* Copyright 2020 Google LLC
*
* Use of this source code is governed by an MIT-style
* license that can be found in the LICENSE file or at
* https://opensource.org/licenses/MIT.
* =============================================================================
*/
/**
* @license
* Copyright 2020 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/**
* @license
* Copyright 2020 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the License);
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an AS IS BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/**
* @license
* Copyright 2021 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/**
* @license
* Copyright 2021 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* https://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/**
* @license
* Copyright 2018 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/** @license See the LICENSE file. */
Reference in New Issue View Git Blame Copy Permalink