org.tensorflow.op

Class Ops

java.lang.Object

org.tensorflow.op.Ops

public final class Ops
extends Object

An API for building a Graph with operation wrappers

Any operation wrapper found in the classpath properly annotated as an@Operator is exposed by this API or one of its subgroup.

Example usage:

 try (Graph g = new Graph()) {
   Ops ops = new Ops(g);
   // Operations are typed classes with convenience
   // builders in Ops.
   Constant three = ops.constant(3);
   // Single-result operations implement the Operand
   // interface, so this works too.
   Operand four = ops.constant(4);
   // Most builders are found within a group, and accept
   // Operand types as operands
   Operand nine = ops.math().add(four, ops.constant(5));
   // Multi-result operations however offer methods to
   // select a particular result for use.
   Operand result = 
       ops.math().add(ops.array().unique(s, a).y(), b);
   // Optional attributes
   ops.math().matMul(a, b, MatMul.transposeA(true));
   // Naming operators
   ops.withName(“foo”).constant(5); // name “foo”
   // Names can exist in a hierarchy
   Ops sub = ops.withSubScope(“sub”);
   sub.withName(“bar”).constant(4); // “sub/bar”
 }
 

Method Summary

Modifier and Type	Method and Description
Abort	abort(Abort.Options... options) Adds an Abort operation to the graph
<T extends Number> Abs<T>	abs(Operand<T> x) Adds an Abs operation to the graph
<T> AccumulateNV2<T>	accumulateNV2(Operand<T> inputs, Shape shape) Adds an AccumulateNV2 operation to the graph
<T> AccumulatorApplyGradient	accumulatorApplyGradient(Operand<String> handle, Operand<Long> localStep, Operand<T> gradient) Adds an AccumulatorApplyGradient operation to the graph
AccumulatorNumAccumulated	accumulatorNumAccumulated(Operand<String> handle) Adds an AccumulatorNumAccumulated operation to the graph
AccumulatorSetGlobalStep	accumulatorSetGlobalStep(Operand<String> handle, Operand<Long> newGlobalStep) Adds an AccumulatorSetGlobalStep operation to the graph
<T> AccumulatorTakeGradient<T>	accumulatorTakeGradient(Operand<String> handle, Operand<Integer> numRequired, Class<T> dtype) Adds an AccumulatorTakeGradient operation to the graph
<T> Acos<T>	acos(Operand<T> x) Adds an Acos operation to the graph
<T> Acosh<T>	acosh(Operand<T> x) Adds an Acosh operation to the graph
<T> Add<T>	add(Operand<T> x, Operand<T> y) Adds an Add operation to the graph
<T> AddManySparseToTensorsMap	addManySparseToTensorsMap(Operand<Long> sparseIndices, Operand<T> sparseValues, Operand<Long> sparseShape, AddManySparseToTensorsMap.Options... options) Adds an AddManySparseToTensorsMap operation to the graph
<T> AddN<T>	addN(Operand<T> inputs) Adds an AddN operation to the graph
<T> AddSparseToTensorsMap	addSparseToTensorsMap(Operand<Long> sparseIndices, Operand<T> sparseValues, Operand<Long> sparseShape, AddSparseToTensorsMap.Options... options) Adds an AddSparseToTensorsMap operation to the graph
<T> AddV2<T>	addV2(Operand<T> x, Operand<T> y) Adds an AddV2 operation to the graph
AdjustContrast	adjustContrast(Operand<Float> images, Operand<Float> contrastFactor) Adds an AdjustContrast operation to the graph
AdjustHue	adjustHue(Operand<Float> images, Operand<Float> delta) Adds an AdjustHue operation to the graph
AdjustSaturation	adjustSaturation(Operand<Float> images, Operand<Float> scale) Adds an AdjustSaturation operation to the graph
<T extends Number> All	all(Operand<Boolean> input, Operand<T> axis, All.Options... options) Adds an All operation to the graph
AllCandidateSampler	allCandidateSampler(Operand<Long> trueClasses, Long numTrue, Long numSampled, Boolean unique, AllCandidateSampler.Options... options) Adds an AllCandidateSampler operation to the graph
<T> Angle<Float>	angle(Operand<T> input) Adds an Angle operation to the graph
<U extends Number,T> Angle<U>	angle(Operand<T> input, Class<U> Tout) Adds an Angle operation to the graph
AnonymousIterator	anonymousIterator(List<Class<?>> outputTypes, List<Shape> outputShapes) Adds an AnonymousIterator operation to the graph
<T extends Number> Any	any(Operand<Boolean> input, Operand<T> axis, Any.Options... options) Adds an Any operation to the graph
<T> ApplyAdadelta<T>	applyAdadelta(Operand<T> var, Operand<T> accum, Operand<T> accumUpdate, Operand<T> lr, Operand<T> rho, Operand<T> epsilon, Operand<T> grad, ApplyAdadelta.Options... options) Adds an ApplyAdadelta operation to the graph
<T> ApplyAdagrad<T>	applyAdagrad(Operand<T> var, Operand<T> accum, Operand<T> lr, Operand<T> grad, ApplyAdagrad.Options... options) Adds an ApplyAdagrad operation to the graph
<T> ApplyAdagradDA<T>	applyAdagradDA(Operand<T> var, Operand<T> gradientAccumulator, Operand<T> gradientSquaredAccumulator, Operand<T> grad, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<Long> globalStep, ApplyAdagradDA.Options... options) Adds an ApplyAdagradDA operation to the graph
<T> ApplyAdam<T>	applyAdam(Operand<T> var, Operand<T> m, Operand<T> v, Operand<T> beta1Power, Operand<T> beta2Power, Operand<T> lr, Operand<T> beta1, Operand<T> beta2, Operand<T> epsilon, Operand<T> grad, ApplyAdam.Options... options) Adds an ApplyAdam operation to the graph
<T> ApplyAddSign<T>	applyAddSign(Operand<T> var, Operand<T> m, Operand<T> lr, Operand<T> alpha, Operand<T> signDecay, Operand<T> beta, Operand<T> grad, ApplyAddSign.Options... options) Adds an ApplyAddSign operation to the graph
<T> ApplyCenteredRMSProp<T>	applyCenteredRMSProp(Operand<T> var, Operand<T> mg, Operand<T> ms, Operand<T> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, ApplyCenteredRMSProp.Options... options) Adds an ApplyCenteredRMSProp operation to the graph
<T> ApplyFtrl<T>	applyFtrl(Operand<T> var, Operand<T> accum, Operand<T> linear, Operand<T> grad, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> lrPower, ApplyFtrl.Options... options) Adds an ApplyFtrl operation to the graph
<T> ApplyFtrlV2<T>	applyFtrlV2(Operand<T> var, Operand<T> accum, Operand<T> linear, Operand<T> grad, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> l2Shrinkage, Operand<T> lrPower, ApplyFtrlV2.Options... options) Adds an ApplyFtrlV2 operation to the graph
<T> ApplyGradientDescent<T>	applyGradientDescent(Operand<T> var, Operand<T> alpha, Operand<T> delta, ApplyGradientDescent.Options... options) Adds an ApplyGradientDescent operation to the graph
<T> ApplyMomentum<T>	applyMomentum(Operand<T> var, Operand<T> accum, Operand<T> lr, Operand<T> grad, Operand<T> momentum, ApplyMomentum.Options... options) Adds an ApplyMomentum operation to the graph
<T> ApplyPowerSign<T>	applyPowerSign(Operand<T> var, Operand<T> m, Operand<T> lr, Operand<T> logbase, Operand<T> signDecay, Operand<T> beta, Operand<T> grad, ApplyPowerSign.Options... options) Adds an ApplyPowerSign operation to the graph
<T> ApplyProximalAdagrad<T>	applyProximalAdagrad(Operand<T> var, Operand<T> accum, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> grad, ApplyProximalAdagrad.Options... options) Adds an ApplyProximalAdagrad operation to the graph
<T> ApplyProximalGradientDescent<T>	applyProximalGradientDescent(Operand<T> var, Operand<T> alpha, Operand<T> l1, Operand<T> l2, Operand<T> delta, ApplyProximalGradientDescent.Options... options) Adds an ApplyProximalGradientDescent operation to the graph
<T> ApplyRMSProp<T>	applyRMSProp(Operand<T> var, Operand<T> ms, Operand<T> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, ApplyRMSProp.Options... options) Adds an ApplyRMSProp operation to the graph
<T> ApproximateEqual	approximateEqual(Operand<T> x, Operand<T> y, ApproximateEqual.Options... options) Adds an ApproximateEqual operation to the graph
<T,U extends Number> ArgMax<Long>	argMax(Operand<T> input, Operand<U> dimension) Adds an ArgMax operation to the graph
<V extends Number,T,U extends Number> ArgMax<V>	argMax(Operand<T> input, Operand<U> dimension, Class<V> outputType) Adds an ArgMax operation to the graph
<T,U extends Number> ArgMin<Long>	argMin(Operand<T> input, Operand<U> dimension) Adds an ArgMin operation to the graph
<V extends Number,T,U extends Number> ArgMin<V>	argMin(Operand<T> input, Operand<U> dimension, Class<V> outputType) Adds an ArgMin operation to the graph
<T> Asin<T>	asin(Operand<T> x) Adds an Asin operation to the graph
<T> Asinh<T>	asinh(Operand<T> x) Adds an Asinh operation to the graph
<T> Assign<T>	assign(Operand<T> ref, Operand<T> value, Assign.Options... options) Adds an Assign operation to the graph
<T> AssignAdd<T>	assignAdd(Operand<T> ref, Operand<T> value, AssignAdd.Options... options) Adds an AssignAdd operation to the graph
<T> AssignAddVariableOp	assignAddVariableOp(Operand<?> resource, Operand<T> value) Adds an AssignAddVariableOp operation to the graph
<T> AssignSub<T>	assignSub(Operand<T> ref, Operand<T> value, AssignSub.Options... options) Adds an AssignSub operation to the graph
<T> AssignSubVariableOp	assignSubVariableOp(Operand<?> resource, Operand<T> value) Adds an AssignSubVariableOp operation to the graph
<T> AssignVariableOp	assignVariableOp(Operand<?> resource, Operand<T> value) Adds an AssignVariableOp operation to the graph
<T> AsString	asString(Operand<T> input, AsString.Options... options) Adds an AsString operation to the graph
<T> Atan<T>	atan(Operand<T> x) Adds an Atan operation to the graph
<T extends Number> Atan2<T>	atan2(Operand<T> y, Operand<T> x) Adds an Atan2 operation to the graph
<T> Atanh<T>	atanh(Operand<T> x) Adds an Atanh operation to the graph
AudioSpectrogram	audioSpectrogram(Operand<Float> input, Long windowSize, Long stride, AudioSpectrogram.Options... options) Adds an AudioSpectrogram operation to the graph
AudioSummary	audioSummary(Operand<String> tag, Operand<Float> tensor, Operand<Float> sampleRate, AudioSummary.Options... options) Adds an AudioSummary operation to the graph
<T extends Number> AvgPool<T>	avgPool(Operand<T> value, List<Long> ksize, List<Long> strides, String padding, AvgPool.Options... options) Adds an AvgPool operation to the graph
<T extends Number> AvgPool3D<T>	avgPool3D(Operand<T> input, List<Long> ksize, List<Long> strides, String padding, AvgPool3D.Options... options) Adds an AvgPool3D operation to the graph
<T extends Number> AvgPool3DGrad<T>	avgPool3DGrad(Operand<Integer> origInputShape, Operand<T> grad, List<Long> ksize, List<Long> strides, String padding, AvgPool3DGrad.Options... options) Adds an AvgPool3DGrad operation to the graph
Barrier	barrier(List<Class<?>> componentTypes, Barrier.Options... options) Adds an Barrier operation to the graph
BarrierClose	barrierClose(Operand<String> handle, BarrierClose.Options... options) Adds an BarrierClose operation to the graph
BarrierIncompleteSize	barrierIncompleteSize(Operand<String> handle) Adds an BarrierIncompleteSize operation to the graph
<T> BarrierInsertMany	barrierInsertMany(Operand<String> handle, Operand<String> keys, Operand<T> values, Long componentIndex) Adds an BarrierInsertMany operation to the graph
BarrierReadySize	barrierReadySize(Operand<String> handle) Adds an BarrierReadySize operation to the graph
BarrierTakeMany	barrierTakeMany(Operand<String> handle, Operand<Integer> numElements, List<Class<?>> componentTypes, BarrierTakeMany.Options... options) Adds an BarrierTakeMany operation to the graph
Batch	batch(Iterable<Operand<?>> inTensors, Long numBatchThreads, Long maxBatchSize, Long batchTimeoutMicros, Long gradTimeoutMicros, Batch.Options... options) Adds an Batch operation to the graph
<T extends Number> BatchCholesky<T>	batchCholesky(Operand<T> input) Adds an BatchCholesky operation to the graph
<T extends Number> BatchCholeskyGrad<T>	batchCholeskyGrad(Operand<T> l, Operand<T> grad) Adds an BatchCholeskyGrad operation to the graph
BatchFFT	batchFFT(Operand<?> input) Adds an BatchFFT operation to the graph
BatchFFT2D	batchFFT2D(Operand<?> input) Adds an BatchFFT2D operation to the graph
BatchFFT3D	batchFFT3D(Operand<?> input) Adds an BatchFFT3D operation to the graph
BatchIFFT	batchIFFT(Operand<?> input) Adds an BatchIFFT operation to the graph
BatchIFFT2D	batchIFFT2D(Operand<?> input) Adds an BatchIFFT2D operation to the graph
BatchIFFT3D	batchIFFT3D(Operand<?> input) Adds an BatchIFFT3D operation to the graph
<T> BatchMatMul<T>	batchMatMul(Operand<T> x, Operand<T> y, BatchMatMul.Options... options) Adds an BatchMatMul operation to the graph
<T> BatchMatrixBandPart<T>	batchMatrixBandPart(Operand<T> input, Operand<Long> numLower, Operand<Long> numUpper) Adds an BatchMatrixBandPart operation to the graph
<T> BatchMatrixDeterminant<T>	batchMatrixDeterminant(Operand<T> input) Adds an BatchMatrixDeterminant operation to the graph
<T> BatchMatrixDiag<T>	batchMatrixDiag(Operand<T> diagonal) Adds an BatchMatrixDiag operation to the graph
<T> BatchMatrixDiagPart<T>	batchMatrixDiagPart(Operand<T> input) Adds an BatchMatrixDiagPart operation to the graph
<T extends Number> BatchMatrixInverse<T>	batchMatrixInverse(Operand<T> input, BatchMatrixInverse.Options... options) Adds an BatchMatrixInverse operation to the graph
<T> BatchMatrixSetDiag<T>	batchMatrixSetDiag(Operand<T> input, Operand<T> diagonal) Adds an BatchMatrixSetDiag operation to the graph
<T extends Number> BatchMatrixSolve<T>	batchMatrixSolve(Operand<T> matrix, Operand<T> rhs, BatchMatrixSolve.Options... options) Adds an BatchMatrixSolve operation to the graph
<T extends Number> BatchMatrixSolveLs<T>	batchMatrixSolveLs(Operand<T> matrix, Operand<T> rhs, Operand<Double> l2Regularizer, BatchMatrixSolveLs.Options... options) Adds an BatchMatrixSolveLs operation to the graph
<T extends Number> BatchMatrixTriangularSolve<T>	batchMatrixTriangularSolve(Operand<T> matrix, Operand<T> rhs, BatchMatrixTriangularSolve.Options... options) Adds an BatchMatrixTriangularSolve operation to the graph
<T> BatchNormWithGlobalNormalization<T>	batchNormWithGlobalNormalization(Operand<T> t, Operand<T> m, Operand<T> v, Operand<T> beta, Operand<T> gamma, Float varianceEpsilon, Boolean scaleAfterNormalization) Adds an BatchNormWithGlobalNormalization operation to the graph
<T> BatchNormWithGlobalNormalizationGrad<T>	batchNormWithGlobalNormalizationGrad(Operand<T> t, Operand<T> m, Operand<T> v, Operand<T> gamma, Operand<T> backprop, Float varianceEpsilon, Boolean scaleAfterNormalization) Adds an BatchNormWithGlobalNormalizationGrad operation to the graph
<T extends Number> BatchSelfAdjointEig<T>	batchSelfAdjointEig(Operand<T> input) Adds an BatchSelfAdjointEig operation to the graph
<T extends Number> BatchSelfAdjointEigV2<T>	batchSelfAdjointEigV2(Operand<T> input, BatchSelfAdjointEigV2.Options... options) Adds an BatchSelfAdjointEigV2 operation to the graph
<T> BatchSvd<T>	batchSvd(Operand<T> input, BatchSvd.Options... options) Adds an BatchSvd operation to the graph
<T,U extends Number> BatchToSpace<T>	batchToSpace(Operand<T> input, Operand<U> crops, Long blockSize) Adds an BatchToSpace operation to the graph
<T,U extends Number,V extends Number> BatchToSpaceND<T>	batchToSpaceND(Operand<T> input, Operand<U> blockShape, Operand<V> crops) Adds an BatchToSpaceND operation to the graph
<T extends Number> BesselI0e<T>	besselI0e(Operand<T> x) Adds an BesselI0e operation to the graph
<T extends Number> BesselI1e<T>	besselI1e(Operand<T> x) Adds an BesselI1e operation to the graph
<T extends Number> Betainc<T>	betainc(Operand<T> a, Operand<T> b, Operand<T> x) Adds an Betainc operation to the graph
<T> BiasAdd<T>	biasAdd(Operand<T> value, Operand<T> bias, BiasAdd.Options... options) Adds an BiasAdd operation to the graph
<T> BiasAddGrad<T>	biasAddGrad(Operand<T> outBackprop, BiasAddGrad.Options... options) Adds an BiasAddGrad operation to the graph
BigQueryReader	bigQueryReader(String projectId, String datasetId, String tableId, List<String> columns, Long timestampMillis, BigQueryReader.Options... options) Adds an BigQueryReader operation to the graph
<T extends Number> Bincount<T>	bincount(Operand<Integer> arr, Operand<Integer> size, Operand<T> weights) Adds an Bincount operation to the graph
<U,T> Bitcast<U>	bitcast(Operand<T> input, Class<U> type) Adds an Bitcast operation to the graph
<T extends Number> BitwiseAnd<T>	bitwiseAnd(Operand<T> x, Operand<T> y) Adds an BitwiseAnd operation to the graph
<T extends Number> BitwiseOr<T>	bitwiseOr(Operand<T> x, Operand<T> y) Adds an BitwiseOr operation to the graph
<T extends Number> BitwiseXor<T>	bitwiseXor(Operand<T> x, Operand<T> y) Adds an BitwiseXor operation to the graph
<T extends Number> BroadcastDynamicShape<T>	broadcastDynamicShape(Operand<T> s0, Operand<T> s1) Adds an BroadcastDynamicShape operation to the graph
<T,U extends Number> BroadcastTo<T>	broadcastTo(Operand<T> input, Operand<U> shape) Adds an BroadcastTo operation to the graph
<T extends Number> Bucketize	bucketize(Operand<T> input, List<Float> boundaries) Adds an Bucketize operation to the graph
<U,T> Cast<U>	cast(Operand<T> x, Class<U> DstT, Cast.Options... options) Adds an Cast operation to the graph
<T extends Number> Ceil<T>	ceil(Operand<T> x) Adds an Ceil operation to the graph
<T extends Number> CheckNumerics<T>	checkNumerics(Operand<T> tensor, String message) Adds an CheckNumerics operation to the graph
<T> Cholesky<T>	cholesky(Operand<T> input) Adds an Cholesky operation to the graph
<T extends Number> CholeskyGrad<T>	choleskyGrad(Operand<T> l, Operand<T> grad) Adds an CholeskyGrad operation to the graph
<T> ClipByValue<T>	clipByValue(Operand<T> t, Operand<T> clipValueMin, Operand<T> clipValueMax) Adds an ClipByValue operation to the graph
<T> CompareAndBitpack	compareAndBitpack(Operand<T> input, Operand<T> threshold) Adds an CompareAndBitpack operation to the graph
<U,T extends Number> Complex<U>	complex(Operand<T> real, Operand<T> imag, Class<U> Tout) Adds an Complex operation to the graph
<T> ComplexAbs<Float>	complexAbs(Operand<T> x) Adds an ComplexAbs operation to the graph
<U extends Number,T> ComplexAbs<U>	complexAbs(Operand<T> x, Class<U> Tout) Adds an ComplexAbs operation to the graph
ComputeAccidentalHits	computeAccidentalHits(Operand<Long> trueClasses, Operand<Long> sampledCandidates, Long numTrue, ComputeAccidentalHits.Options... options) Adds an ComputeAccidentalHits operation to the graph
<T,U extends Number> Concat<T>	concat(Operand<T> values, Operand<U> axis) Adds an Concat operation to the graph
<T> ConditionalAccumulator	conditionalAccumulator(Class<T> dtype, Shape shape, ConditionalAccumulator.Options... options) Adds an ConditionalAccumulator operation to the graph
<T> Conj<T>	conj(Operand<T> input) Adds an Conj operation to the graph
<T,U extends Number> ConjugateTranspose<T>	conjugateTranspose(Operand<T> x, Operand<U> perm) Adds an ConjugateTranspose operation to the graph
Constant<Boolean>	constant(boolean data) Adds an Constant operation to the graph
Constant<Boolean>	constant(boolean[] data) Adds an Constant operation to the graph
Constant<Boolean>	constant(boolean[][] data) Adds an Constant operation to the graph
Constant<Boolean>	constant(boolean[][][] data) Adds an Constant operation to the graph
Constant<Boolean>	constant(boolean[][][][] data) Adds an Constant operation to the graph
Constant<Boolean>	constant(boolean[][][][][] data) Adds an Constant operation to the graph
Constant<Boolean>	constant(boolean[][][][][][] data) Adds an Constant operation to the graph
Constant<String>	constant(byte[] data) Adds an Constant operation to the graph
Constant<String>	constant(byte[][] data) Adds an Constant operation to the graph
Constant<String>	constant(byte[][][] data) Adds an Constant operation to the graph
Constant<String>	constant(byte[][][][] data) Adds an Constant operation to the graph
Constant<String>	constant(byte[][][][][] data) Adds an Constant operation to the graph
Constant<String>	constant(byte[][][][][][] data) Adds an Constant operation to the graph
<T> Constant<T>	constant(Class<T> type, long[] shape, ByteBuffer data) Adds an Constant operation to the graph
Constant<Double>	constant(double data) Adds an Constant operation to the graph
Constant<Double>	constant(double[] data) Adds an Constant operation to the graph
Constant<Double>	constant(double[][] data) Adds an Constant operation to the graph
Constant<Double>	constant(double[][][] data) Adds an Constant operation to the graph
Constant<Double>	constant(double[][][][] data) Adds an Constant operation to the graph
Constant<Double>	constant(double[][][][][] data) Adds an Constant operation to the graph
Constant<Double>	constant(double[][][][][][] data) Adds an Constant operation to the graph
Constant<Float>	constant(float data) Adds an Constant operation to the graph
Constant<Float>	constant(float[] data) Adds an Constant operation to the graph
Constant<Float>	constant(float[][] data) Adds an Constant operation to the graph
Constant<Float>	constant(float[][][] data) Adds an Constant operation to the graph
Constant<Float>	constant(float[][][][] data) Adds an Constant operation to the graph
Constant<Float>	constant(float[][][][][] data) Adds an Constant operation to the graph
Constant<Float>	constant(float[][][][][][] data) Adds an Constant operation to the graph
Constant<Integer>	constant(int data) Adds an Constant operation to the graph
Constant<Integer>	constant(int[] data) Adds an Constant operation to the graph
Constant<Integer>	constant(int[][] data) Adds an Constant operation to the graph
Constant<Integer>	constant(int[][][] data) Adds an Constant operation to the graph
Constant<Integer>	constant(int[][][][] data) Adds an Constant operation to the graph
Constant<Integer>	constant(int[][][][][] data) Adds an Constant operation to the graph
Constant<Integer>	constant(int[][][][][][] data) Adds an Constant operation to the graph
Constant<Long>	constant(long data) Adds an Constant operation to the graph
Constant<Long>	constant(long[] data) Adds an Constant operation to the graph
Constant<Long>	constant(long[][] data) Adds an Constant operation to the graph
Constant<Long>	constant(long[][][] data) Adds an Constant operation to the graph
Constant<Long>	constant(long[][][][] data) Adds an Constant operation to the graph
Constant<Long>	constant(long[][][][][] data) Adds an Constant operation to the graph
Constant<Long>	constant(long[][][][][][] data) Adds an Constant operation to the graph
Constant<Double>	constant(long[] shape, DoubleBuffer data) Adds an Constant operation to the graph
Constant<Float>	constant(long[] shape, FloatBuffer data) Adds an Constant operation to the graph
Constant<Integer>	constant(long[] shape, IntBuffer data) Adds an Constant operation to the graph
Constant<Long>	constant(long[] shape, LongBuffer data) Adds an Constant operation to the graph
<T> Constant<T>	constant(Object object, Class<T> type) Adds an Constant operation to the graph
Constant<String>	constant(String data) Adds an Constant operation to the graph
Constant<String>	constant(String data, Charset charset) Adds an Constant operation to the graph
ConsumeMutexLock	consumeMutexLock(Operand<?> mutexLock) Adds an ConsumeMutexLock operation to the graph
ControlTrigger	controlTrigger() Adds an ControlTrigger operation to the graph
<T extends Number> Conv2D<T>	conv2D(Operand<T> input, Operand<T> filter, List<Long> strides, String padding, Conv2D.Options... options) Adds an Conv2D operation to the graph
<T extends Number> Conv2DBackpropFilter<T>	conv2DBackpropFilter(Operand<T> input, Operand<Integer> filterSizes, Operand<T> outBackprop, List<Long> strides, String padding, Conv2DBackpropFilter.Options... options) Adds an Conv2DBackpropFilter operation to the graph
<T extends Number> Conv2DBackpropInput<T>	conv2DBackpropInput(Operand<Integer> inputSizes, Operand<T> filter, Operand<T> outBackprop, List<Long> strides, String padding, Conv2DBackpropInput.Options... options) Adds an Conv2DBackpropInput operation to the graph
<T extends Number> Conv3D<T>	conv3D(Operand<T> input, Operand<T> filter, List<Long> strides, String padding, Conv3D.Options... options) Adds an Conv3D operation to the graph
<T extends Number> Conv3DBackpropFilter<T>	conv3DBackpropFilter(Operand<T> input, Operand<T> filter, Operand<T> outBackprop, List<Long> strides, String padding, Conv3DBackpropFilter.Options... options) Adds an Conv3DBackpropFilter operation to the graph
<T extends Number> Conv3DBackpropFilterV2<T>	conv3DBackpropFilterV2(Operand<T> input, Operand<Integer> filterSizes, Operand<T> outBackprop, List<Long> strides, String padding, Conv3DBackpropFilterV2.Options... options) Adds an Conv3DBackpropFilterV2 operation to the graph
<T extends Number> Conv3DBackpropInput<T>	conv3DBackpropInput(Operand<T> input, Operand<T> filter, Operand<T> outBackprop, List<Long> strides, String padding, Conv3DBackpropInput.Options... options) Adds an Conv3DBackpropInput operation to the graph
<U extends Number,T extends Number> Conv3DBackpropInputV2<U>	conv3DBackpropInputV2(Operand<T> inputSizes, Operand<U> filter, Operand<U> outBackprop, List<Long> strides, String padding, Conv3DBackpropInputV2.Options... options) Adds an Conv3DBackpropInputV2 operation to the graph
<T> Cos<T>	cos(Operand<T> x) Adds an Cos operation to the graph
<T> Cosh<T>	cosh(Operand<T> x) Adds an Cosh operation to the graph
<T extends Number> CountUpTo<T>	countUpTo(Operand<T> ref, Long limit) Adds an CountUpTo operation to the graph
static Ops	create(Graph graph) Creates an API for adding operations to the provided graph
<T extends Number> CropAndResize	cropAndResize(Operand<T> image, Operand<Float> boxes, Operand<Integer> boxInd, Operand<Integer> cropSize, CropAndResize.Options... options) Adds an CropAndResize operation to the graph
<T extends Number> CropAndResizeGradBoxes	cropAndResizeGradBoxes(Operand<Float> grads, Operand<T> image, Operand<Float> boxes, Operand<Integer> boxInd, CropAndResizeGradBoxes.Options... options) Adds an CropAndResizeGradBoxes operation to the graph
<T extends Number> CropAndResizeGradImage<T>	cropAndResizeGradImage(Operand<Float> grads, Operand<Float> boxes, Operand<Integer> boxInd, Operand<Integer> imageSize, Class<T> T, CropAndResizeGradImage.Options... options) Adds an CropAndResizeGradImage operation to the graph
<T extends Number> Cross<T>	cross(Operand<T> a, Operand<T> b) Adds an Cross operation to the graph
CTCBeamSearchDecoder	cTCBeamSearchDecoder(Operand<Float> inputs, Operand<Integer> sequenceLength, Long beamWidth, Long topPaths, CTCBeamSearchDecoder.Options... options) Adds an CTCBeamSearchDecoder operation to the graph
CTCGreedyDecoder	cTCGreedyDecoder(Operand<Float> inputs, Operand<Integer> sequenceLength, CTCGreedyDecoder.Options... options) Adds an CTCGreedyDecoder operation to the graph
CTCLoss	cTCLoss(Operand<Float> inputs, Operand<Long> labelsIndices, Operand<Integer> labelsValues, Operand<Integer> sequenceLength, CTCLoss.Options... options) Adds an CTCLoss operation to the graph
<T extends Number> CudnnRNN<T>	cudnnRNN(Operand<T> input, Operand<T> inputH, Operand<T> inputC, Operand<T> params, CudnnRNN.Options... options) Adds an CudnnRNN operation to the graph
<T extends Number> CudnnRNNBackprop<T>	cudnnRNNBackprop(Operand<T> input, Operand<T> inputH, Operand<T> inputC, Operand<T> params, Operand<T> output, Operand<T> outputH, Operand<T> outputC, Operand<T> outputBackprop, Operand<T> outputHBackprop, Operand<T> outputCBackprop, Operand<T> reserveSpace, CudnnRNNBackprop.Options... options) Adds an CudnnRNNBackprop operation to the graph
<T extends Number> CudnnRNNCanonicalToParams<T>	cudnnRNNCanonicalToParams(Operand<Integer> numLayers, Operand<Integer> numUnits, Operand<Integer> inputSize, Operand<T> weights, Iterable<Operand<T>> biases, CudnnRNNCanonicalToParams.Options... options) Adds an CudnnRNNCanonicalToParams operation to the graph
<U extends Number,T extends Number> CudnnRNNParamsSize<U>	cudnnRNNParamsSize(Operand<Integer> numLayers, Operand<Integer> numUnits, Operand<Integer> inputSize, Class<T> T, Class<U> S, CudnnRNNParamsSize.Options... options) Adds an CudnnRNNParamsSize operation to the graph
<T extends Number> CudnnRNNParamsToCanonical<T>	cudnnRNNParamsToCanonical(Operand<Integer> numLayers, Operand<Integer> numUnits, Operand<Integer> inputSize, Operand<T> params, Long numParams, CudnnRNNParamsToCanonical.Options... options) Adds an CudnnRNNParamsToCanonical operation to the graph
<T,U extends Number> Cumprod<T>	cumprod(Operand<T> x, Operand<U> axis, Cumprod.Options... options) Adds an Cumprod operation to the graph
<T,U extends Number> Cumsum<T>	cumsum(Operand<T> x, Operand<U> axis, Cumsum.Options... options) Adds an Cumsum operation to the graph
<T extends Number> DataFormatDimMap<T>	dataFormatDimMap(Operand<T> x, DataFormatDimMap.Options... options) Adds an DataFormatDimMap operation to the graph
<T extends Number> DataFormatVecPermute<T>	dataFormatVecPermute(Operand<T> x, DataFormatVecPermute.Options... options) Adds an DataFormatVecPermute operation to the graph
<T> DebugGradientIdentity<T>	debugGradientIdentity(Operand<T> input) Adds an DebugGradientIdentity operation to the graph
<T> DebugGradientRefIdentity<T>	debugGradientRefIdentity(Operand<T> input) Adds an DebugGradientRefIdentity operation to the graph
DecodeAndCropJpeg	decodeAndCropJpeg(Operand<String> contents, Operand<Integer> cropWindow, DecodeAndCropJpeg.Options... options) Adds an DecodeAndCropJpeg operation to the graph
DecodeBase64	decodeBase64(Operand<String> input) Adds an DecodeBase64 operation to the graph
DecodeBmp	decodeBmp(Operand<String> contents, DecodeBmp.Options... options) Adds an DecodeBmp operation to the graph
DecodeCompressed	decodeCompressed(Operand<String> bytes, DecodeCompressed.Options... options) Adds an DecodeCompressed operation to the graph
DecodeCSV	decodeCSV(Operand<String> records, Iterable<Operand<?>> recordDefaults, DecodeCSV.Options... options) Adds an DecodeCSV operation to the graph
DecodeGif	decodeGif(Operand<String> contents) Adds an DecodeGif operation to the graph
DecodeJpeg	decodeJpeg(Operand<String> contents, DecodeJpeg.Options... options) Adds an DecodeJpeg operation to the graph
DecodeJSONExample	decodeJSONExample(Operand<String> jsonExamples) Adds an DecodeJSONExample operation to the graph
<T extends Number> DecodePng<T>	decodePng(Operand<String> contents, Class<T> dtype, DecodePng.Options... options) Adds an DecodePng operation to the graph
DecodePng<UInt8>	decodePng(Operand<String> contents, DecodePng.Options... options) Adds an DecodePng operation to the graph
DecodeProtoV2	decodeProtoV2(Operand<String> bytes, String messageType, List<String> fieldNames, List<Class<?>> outputTypes, DecodeProtoV2.Options... options) Adds an DecodeProtoV2 operation to the graph
<T extends Number> DecodeRaw<T>	decodeRaw(Operand<String> bytes, Class<T> outType, DecodeRaw.Options... options) Adds an DecodeRaw operation to the graph
DecodeWav	decodeWav(Operand<String> contents, DecodeWav.Options... options) Adds an DecodeWav operation to the graph
<T> DeepCopy<T>	deepCopy(Operand<T> x) Adds an DeepCopy operation to the graph
DeleteSessionTensor	deleteSessionTensor(Operand<String> handle) Adds an DeleteSessionTensor operation to the graph
<T> DenseToDenseSetOperation<T>	denseToDenseSetOperation(Operand<T> set1, Operand<T> set2, String setOperation, DenseToDenseSetOperation.Options... options) Adds an DenseToDenseSetOperation operation to the graph
<T> DenseToSparseSetOperation<T>	denseToSparseSetOperation(Operand<T> set1, Operand<Long> set2Indices, Operand<T> set2Values, Operand<Long> set2Shape, String setOperation, DenseToSparseSetOperation.Options... options) Adds an DenseToSparseSetOperation operation to the graph
<T> DepthToSpace<T>	depthToSpace(Operand<T> input, Long blockSize, DepthToSpace.Options... options) Adds an DepthToSpace operation to the graph
<T extends Number> DepthwiseConv2dNative<T>	depthwiseConv2dNative(Operand<T> input, Operand<T> filter, List<Long> strides, String padding, DepthwiseConv2dNative.Options... options) Adds an DepthwiseConv2dNative operation to the graph
<T extends Number> DepthwiseConv2dNativeBackpropFilter<T>	depthwiseConv2dNativeBackpropFilter(Operand<T> input, Operand<Integer> filterSizes, Operand<T> outBackprop, List<Long> strides, String padding, DepthwiseConv2dNativeBackpropFilter.Options... options) Adds an DepthwiseConv2dNativeBackpropFilter operation to the graph
<T extends Number> DepthwiseConv2dNativeBackpropInput<T>	depthwiseConv2dNativeBackpropInput(Operand<Integer> inputSizes, Operand<T> filter, Operand<T> outBackprop, List<Long> strides, String padding, DepthwiseConv2dNativeBackpropInput.Options... options) Adds an DepthwiseConv2dNativeBackpropInput operation to the graph
<T> Dequantize	dequantize(Operand<T> input, Operand<Float> minRange, Operand<Float> maxRange, Dequantize.Options... options) Adds an Dequantize operation to the graph
DeserializeIterator	deserializeIterator(Operand<?> resourceHandle, Operand<?> serialized) Adds an DeserializeIterator operation to the graph
<T> DeserializeManySparse<T>	deserializeManySparse(Operand<String> serializedSparse, Class<T> dtype) Adds an DeserializeManySparse operation to the graph
<U,T> DeserializeSparse<U>	deserializeSparse(Operand<T> serializedSparse, Class<U> dtype) Adds an DeserializeSparse operation to the graph
DestroyResourceOp	destroyResourceOp(Operand<?> resource, DestroyResourceOp.Options... options) Adds an DestroyResourceOp operation to the graph
<T> DestroyTemporaryVariable<T>	destroyTemporaryVariable(Operand<T> ref, String varName) Adds an DestroyTemporaryVariable operation to the graph
<T> Diag<T>	diag(Operand<T> diagonal) Adds an Diag operation to the graph
<T> DiagPart<T>	diagPart(Operand<T> input) Adds an DiagPart operation to the graph
<T extends Number> Digamma<T>	digamma(Operand<T> x) Adds an Digamma operation to the graph
<T extends Number> Dilation2D<T>	dilation2D(Operand<T> input, Operand<T> filter, List<Long> strides, List<Long> rates, String padding) Adds an Dilation2D operation to the graph
<T extends Number> Dilation2DBackpropFilter<T>	dilation2DBackpropFilter(Operand<T> input, Operand<T> filter, Operand<T> outBackprop, List<Long> strides, List<Long> rates, String padding) Adds an Dilation2DBackpropFilter operation to the graph
<T extends Number> Dilation2DBackpropInput<T>	dilation2DBackpropInput(Operand<T> input, Operand<T> filter, Operand<T> outBackprop, List<Long> strides, List<Long> rates, String padding) Adds an Dilation2DBackpropInput operation to the graph
<T> Div<T>	div(Operand<T> x, Operand<T> y) Adds an Div operation to the graph
<T extends Number> DivNoNan<T>	divNoNan(Operand<T> x, Operand<T> y) Adds an DivNoNan operation to the graph
<T extends Number> DrawBoundingBoxes<T>	drawBoundingBoxes(Operand<T> images, Operand<Float> boxes) Adds an DrawBoundingBoxes operation to the graph
<T> DynamicPartition<T>	dynamicPartition(Operand<T> data, Operand<Integer> partitions, Long numPartitions) Adds an DynamicPartition operation to the graph
<T> DynamicStitch<T>	dynamicStitch(Iterable<Operand<Integer>> indices, Operand<T> data) Adds an DynamicStitch operation to the graph
EagerPyFunc	eagerPyFunc(Iterable<Operand<?>> input, String token, List<Class<?>> Tout) Adds an EagerPyFunc operation to the graph
<T> EditDistance	editDistance(Operand<Long> hypothesisIndices, Operand<T> hypothesisValues, Operand<Long> hypothesisShape, Operand<Long> truthIndices, Operand<T> truthValues, Operand<Long> truthShape, EditDistance.Options... options) Adds an EditDistance operation to the graph
<T extends Number> Elu<T>	elu(Operand<T> features) Adds an Elu operation to the graph
<T> Empty<T>	empty(Operand<Integer> shape, Class<T> dtype, Empty.Options... options) Adds an Empty operation to the graph
<T extends Number,U> EmptyTensorList	emptyTensorList(Operand<T> elementShape, Operand<Integer> maxNumElements, Class<U> elementDtype) Adds an EmptyTensorList operation to the graph
EncodeBase64	encodeBase64(Operand<String> input, EncodeBase64.Options... options) Adds an EncodeBase64 operation to the graph
EncodeJpeg	encodeJpeg(Operand<UInt8> image, EncodeJpeg.Options... options) Adds an EncodeJpeg operation to the graph
<T extends Number> EncodePng	encodePng(Operand<T> image, EncodePng.Options... options) Adds an EncodePng operation to the graph
EncodeProto	encodeProto(Operand<Integer> sizes, Iterable<Operand<?>> values, List<String> fieldNames, String messageType, EncodeProto.Options... options) Adds an EncodeProto operation to the graph
EncodeWav	encodeWav(Operand<Float> audio, Operand<Integer> sampleRate) Adds an EncodeWav operation to the graph
<T> EnsureShape<T>	ensureShape(Operand<T> input, Shape shape) Adds an EnsureShape operation to the graph
<T> Equal	equal(Operand<T> x, Operand<T> y) Adds an Equal operation to the graph
<T extends Number> Erf<T>	erf(Operand<T> x) Adds an Erf operation to the graph
<T extends Number> Erfc<T>	erfc(Operand<T> x) Adds an Erfc operation to the graph
<T> Exp<T>	exp(Operand<T> x) Adds an Exp operation to the graph
<T,U extends Number> ExpandDims<T>	expandDims(Operand<T> input, Operand<U> axis) Adds an ExpandDims operation to the graph
<T> Expm1<T>	expm1(Operand<T> x) Adds an Expm1 operation to the graph
ExtractGlimpse	extractGlimpse(Operand<Float> input, Operand<Integer> size, Operand<Float> offsets, ExtractGlimpse.Options... options) Adds an ExtractGlimpse operation to the graph
<T extends Number> ExtractImagePatches<T>	extractImagePatches(Operand<T> images, List<Long> ksizes, List<Long> strides, List<Long> rates, String padding) Adds an ExtractImagePatches operation to the graph
ExtractJpegShape<Integer>	extractJpegShape(Operand<String> contents) Adds an ExtractJpegShape operation to the graph
<T extends Number> ExtractJpegShape<T>	extractJpegShape(Operand<String> contents, Class<T> outputType) Adds an ExtractJpegShape operation to the graph
<T extends Number> ExtractVolumePatches<T>	extractVolumePatches(Operand<T> input, List<Long> ksizes, List<Long> strides, String padding) Adds an ExtractVolumePatches operation to the graph
Fact	fact() Adds an Fact operation to the graph
FakeQuantWithMinMaxArgs	fakeQuantWithMinMaxArgs(Operand<Float> inputs, FakeQuantWithMinMaxArgs.Options... options) Adds an FakeQuantWithMinMaxArgs operation to the graph
FakeQuantWithMinMaxArgsGradient	fakeQuantWithMinMaxArgsGradient(Operand<Float> gradients, Operand<Float> inputs, FakeQuantWithMinMaxArgsGradient.Options... options) Adds an FakeQuantWithMinMaxArgsGradient operation to the graph
FakeQuantWithMinMaxVars	fakeQuantWithMinMaxVars(Operand<Float> inputs, Operand<Float> min, Operand<Float> max, FakeQuantWithMinMaxVars.Options... options) Adds an FakeQuantWithMinMaxVars operation to the graph
FakeQuantWithMinMaxVarsGradient	fakeQuantWithMinMaxVarsGradient(Operand<Float> gradients, Operand<Float> inputs, Operand<Float> min, Operand<Float> max, FakeQuantWithMinMaxVarsGradient.Options... options) Adds an FakeQuantWithMinMaxVarsGradient operation to the graph
FakeQuantWithMinMaxVarsPerChannel	fakeQuantWithMinMaxVarsPerChannel(Operand<Float> inputs, Operand<Float> min, Operand<Float> max, FakeQuantWithMinMaxVarsPerChannel.Options... options) Adds an FakeQuantWithMinMaxVarsPerChannel operation to the graph
FakeQuantWithMinMaxVarsPerChannelGradient	fakeQuantWithMinMaxVarsPerChannelGradient(Operand<Float> gradients, Operand<Float> inputs, Operand<Float> min, Operand<Float> max, FakeQuantWithMinMaxVarsPerChannelGradient.Options... options) Adds an FakeQuantWithMinMaxVarsPerChannelGradient operation to the graph
<T> FFT<T>	fFT(Operand<T> input) Adds an FFT operation to the graph
<T> FFT2D<T>	fFT2D(Operand<T> input) Adds an FFT2D operation to the graph
<T> FFT3D<T>	fFT3D(Operand<T> input) Adds an FFT3D operation to the graph
FIFOQueue	fIFOQueue(List<Class<?>> componentTypes, FIFOQueue.Options... options) Adds an FIFOQueue operation to the graph
<U,T extends Number> Fill<U>	fill(Operand<T> dims, Operand<U> value) Adds an Fill operation to the graph
FixedLengthRecordReader	fixedLengthRecordReader(Long recordBytes, FixedLengthRecordReader.Options... options) Adds an FixedLengthRecordReader operation to the graph
FixedUnigramCandidateSampler	fixedUnigramCandidateSampler(Operand<Long> trueClasses, Long numTrue, Long numSampled, Boolean unique, Long rangeMax, FixedUnigramCandidateSampler.Options... options) Adds an FixedUnigramCandidateSampler operation to the graph
<T extends Number> Floor<T>	floor(Operand<T> x) Adds an Floor operation to the graph
<T> FloorDiv<T>	floorDiv(Operand<T> x, Operand<T> y) Adds an FloorDiv operation to the graph
<T extends Number> FloorMod<T>	floorMod(Operand<T> x, Operand<T> y) Adds an FloorMod operation to the graph
<T extends Number> FractionalAvgPool<T>	fractionalAvgPool(Operand<T> value, List<Float> poolingRatio, FractionalAvgPool.Options... options) Adds an FractionalAvgPool operation to the graph
<T extends Number> FractionalMaxPool<T>	fractionalMaxPool(Operand<T> value, List<Float> poolingRatio, FractionalMaxPool.Options... options) Adds an FractionalMaxPool operation to the graph
<T extends Number> FusedBatchNorm<T>	fusedBatchNorm(Operand<T> x, Operand<T> scale, Operand<T> offset, Operand<T> mean, Operand<T> variance, FusedBatchNorm.Options... options) Adds an FusedBatchNorm operation to the graph
<T extends Number> FusedBatchNormGrad<T>	fusedBatchNormGrad(Operand<T> yBackprop, Operand<T> x, Operand<T> scale, Operand<T> reserveSpace1, Operand<T> reserveSpace2, FusedBatchNormGrad.Options... options) Adds an FusedBatchNormGrad operation to the graph
<T extends Number,U extends Number> FusedBatchNormGradV2<T,U>	fusedBatchNormGradV2(Operand<T> yBackprop, Operand<T> x, Operand<Float> scale, Operand<U> reserveSpace1, Operand<U> reserveSpace2, FusedBatchNormGradV2.Options... options) Adds an FusedBatchNormGradV2 operation to the graph
<T extends Number,U extends Number> FusedBatchNormV2<T,U>	fusedBatchNormV2(Operand<T> x, Operand<U> scale, Operand<U> offset, Operand<U> mean, Operand<U> variance, FusedBatchNormV2.Options... options) Adds an FusedBatchNormV2 operation to the graph
<T extends Number> FusedPadConv2D<T>	fusedPadConv2D(Operand<T> input, Operand<Integer> paddings, Operand<T> filter, String mode, List<Long> strides, String padding) Adds an FusedPadConv2D operation to the graph
<T extends Number> FusedResizeAndPadConv2D<T>	fusedResizeAndPadConv2D(Operand<T> input, Operand<Integer> size, Operand<Integer> paddings, Operand<T> filter, String mode, List<Long> strides, String padding, FusedResizeAndPadConv2D.Options... options) Adds an FusedResizeAndPadConv2D operation to the graph
<T,U extends Number> Gather<T>	gather(Operand<T> params, Operand<U> indices, Gather.Options... options) Adds an Gather operation to the graph
<T,U extends Number> GatherNd<T>	gatherNd(Operand<T> params, Operand<U> indices) Adds an GatherNd operation to the graph
<T,U extends Number,V extends Number> GatherV2<T>	gatherV2(Operand<T> params, Operand<U> indices, Operand<V> axis) Adds an GatherV2 operation to the graph
GcsConfigureBlockCache	gcsConfigureBlockCache(Operand<?> maxCacheSize, Operand<?> blockSize, Operand<?> maxStaleness) Adds an GcsConfigureBlockCache operation to the graph
GcsConfigureCredentials	gcsConfigureCredentials(Operand<String> json) Adds an GcsConfigureCredentials operation to the graph
GenerateBigQueryReaderPartitions	generateBigQueryReaderPartitions(String projectId, String datasetId, String tableId, List<String> columns, Long timestampMillis, Long numPartitions, GenerateBigQueryReaderPartitions.Options... options) Adds an GenerateBigQueryReaderPartitions operation to the graph
GenerateVocabRemapping	generateVocabRemapping(Operand<String> newVocabFile, Operand<String> oldVocabFile, Long newVocabOffset, Long numNewVocab, GenerateVocabRemapping.Options... options) Adds an GenerateVocabRemapping operation to the graph
<T> GetSessionHandle	getSessionHandle(Operand<T> value) Adds an GetSessionHandle operation to the graph
<T> GetSessionHandleV2	getSessionHandleV2(Operand<T> value) Adds an GetSessionHandleV2 operation to the graph
<T> GetSessionTensor<T>	getSessionTensor(Operand<String> handle, Class<T> dtype) Adds an GetSessionTensor operation to the graph
Gradients	gradients(Iterable<? extends Operand<?>> y, Iterable<? extends Operand<?>> x, Gradients.Options... options) Adds an Gradients operation to the graph
Gradients	gradients(Operand<?> y, Iterable<? extends Operand<?>> x, Gradients.Options... options) Adds an Gradients operation to the graph
<T extends Number> Greater	greater(Operand<T> x, Operand<T> y) Adds an Greater operation to the graph
<T extends Number> GreaterEqual	greaterEqual(Operand<T> x, Operand<T> y) Adds an GreaterEqual operation to the graph
<T> GuaranteeConst<T>	guaranteeConst(Operand<T> input) Adds an GuaranteeConst operation to the graph
<T,U> HashTable	hashTable(Class<T> keyDtype, Class<U> valueDtype, HashTable.Options... options) Adds an HashTable operation to the graph
<T extends Number> HistogramFixedWidth<Integer>	histogramFixedWidth(Operand<T> values, Operand<T> valueRange, Operand<Integer> nbins) Adds an HistogramFixedWidth operation to the graph
<U extends Number,T extends Number> HistogramFixedWidth<U>	histogramFixedWidth(Operand<T> values, Operand<T> valueRange, Operand<Integer> nbins, Class<U> dtype) Adds an HistogramFixedWidth operation to the graph
<T extends Number> HistogramSummary	histogramSummary(Operand<String> tag, Operand<T> values) Adds an HistogramSummary operation to the graph
<T extends Number> HSVToRGB<T>	hSVToRGB(Operand<T> images) Adds an HSVToRGB operation to the graph
<T> Identity<T>	identity(Operand<T> input) Adds an Identity operation to the graph
IdentityN	identityN(Iterable<Operand<?>> input) Adds an IdentityN operation to the graph
IdentityReader	identityReader(IdentityReader.Options... options) Adds an IdentityReader operation to the graph
<T> IFFT<T>	iFFT(Operand<T> input) Adds an IFFT operation to the graph
<T> IFFT2D<T>	iFFT2D(Operand<T> input) Adds an IFFT2D operation to the graph
<T> IFFT3D<T>	iFFT3D(Operand<T> input) Adds an IFFT3D operation to the graph
<T extends Number> Igamma<T>	igamma(Operand<T> a, Operand<T> x) Adds an Igamma operation to the graph
<T extends Number> Igammac<T>	igammac(Operand<T> a, Operand<T> x) Adds an Igammac operation to the graph
<T> Imag<Float>	imag(Operand<T> input) Adds an Imag operation to the graph
<U extends Number,T> Imag<U>	imag(Operand<T> input, Class<U> Tout) Adds an Imag operation to the graph
<T extends Number> ImageSummary	imageSummary(Operand<String> tag, Operand<T> tensor, ImageSummary.Options... options) Adds an ImageSummary operation to the graph
<T> ImmutableConst<T>	immutableConst(Class<T> dtype, Shape shape, String memoryRegionName) Adds an ImmutableConst operation to the graph
<T,U> InitializeTable	initializeTable(Operand<?> tableHandle, Operand<T> keys, Operand<U> values) Adds an InitializeTable operation to the graph
InitializeTableFromTextFile	initializeTableFromTextFile(Operand<?> tableHandle, Operand<String> filename, Long keyIndex, Long valueIndex, InitializeTableFromTextFile.Options... options) Adds an InitializeTableFromTextFile operation to the graph
<T> InplaceAdd<T>	inplaceAdd(Operand<T> x, Operand<Integer> i, Operand<T> v) Adds an InplaceAdd operation to the graph
<T> InplaceSub<T>	inplaceSub(Operand<T> x, Operand<Integer> i, Operand<T> v) Adds an InplaceSub operation to the graph
<T> InplaceUpdate<T>	inplaceUpdate(Operand<T> x, Operand<Integer> i, Operand<T> v) Adds an InplaceUpdate operation to the graph
<T extends Number> InTopK	inTopK(Operand<Float> predictions, Operand<T> targets, Long k) Adds an InTopK operation to the graph
<T extends Number> InTopKV2	inTopKV2(Operand<Float> predictions, Operand<T> targets, Operand<T> k) Adds an InTopKV2 operation to the graph
<T> Inv<T>	inv(Operand<T> x) Adds an Inv operation to the graph
<T extends Number> Invert<T>	invert(Operand<T> x) Adds an Invert operation to the graph
<T extends Number> InvertPermutation<T>	invertPermutation(Operand<T> x) Adds an InvertPermutation operation to the graph
IRFFT	iRFFT(Operand<?> input, Operand<Integer> fftLength) Adds an IRFFT operation to the graph
IRFFT2D	iRFFT2D(Operand<?> input, Operand<Integer> fftLength) Adds an IRFFT2D operation to the graph
IRFFT3D	iRFFT3D(Operand<?> input, Operand<Integer> fftLength) Adds an IRFFT3D operation to the graph
<T extends Number> IsFinite	isFinite(Operand<T> x) Adds an IsFinite operation to the graph
<T extends Number> IsInf	isInf(Operand<T> x) Adds an IsInf operation to the graph
<T extends Number> IsNan	isNan(Operand<T> x) Adds an IsNan operation to the graph
<T> IsVariableInitialized	isVariableInitialized(Operand<T> ref) Adds an IsVariableInitialized operation to the graph
Iterator	iterator(String sharedName, String container, List<Class<?>> outputTypes, List<Shape> outputShapes) Adds an Iterator operation to the graph
IteratorFromStringHandle	iteratorFromStringHandle(Operand<String> stringHandle, List<Class<?>> outputTypes, IteratorFromStringHandle.Options... options) Adds an IteratorFromStringHandle operation to the graph
IteratorGetNext	iteratorGetNext(Operand<?> iterator, List<Class<?>> outputTypes, List<Shape> outputShapes) Adds an IteratorGetNext operation to the graph
IteratorGetNextAsOptional	iteratorGetNextAsOptional(Operand<?> iterator, List<Class<?>> outputTypes, List<Shape> outputShapes) Adds an IteratorGetNextAsOptional operation to the graph
IteratorGetNextSync	iteratorGetNextSync(Operand<?> iterator, List<Class<?>> outputTypes, List<Shape> outputShapes) Adds an IteratorGetNextSync operation to the graph
IteratorToStringHandle	iteratorToStringHandle(Operand<?> resourceHandle) Adds an IteratorToStringHandle operation to the graph
<T extends Number> L2Loss<T>	l2Loss(Operand<T> t) Adds an L2Loss operation to the graph
LearnedUnigramCandidateSampler	learnedUnigramCandidateSampler(Operand<Long> trueClasses, Long numTrue, Long numSampled, Boolean unique, Long rangeMax, LearnedUnigramCandidateSampler.Options... options) Adds an LearnedUnigramCandidateSampler operation to the graph
<T extends Number> LeftShift<T>	leftShift(Operand<T> x, Operand<T> y) Adds an LeftShift operation to the graph
<T extends Number> Less	less(Operand<T> x, Operand<T> y) Adds an Less operation to the graph
<T extends Number> LessEqual	lessEqual(Operand<T> x, Operand<T> y) Adds an LessEqual operation to the graph
<T extends Number> Lgamma<T>	lgamma(Operand<T> x) Adds an Lgamma operation to the graph
<T extends Number,U extends Number> LinSpace<T>	linSpace(Operand<T> start, Operand<T> stop, Operand<U> num) Adds an LinSpace operation to the graph
LMDBReader	lMDBReader(LMDBReader.Options... options) Adds an LMDBReader operation to the graph
LoadAndRemapMatrix	loadAndRemapMatrix(Operand<String> ckptPath, Operand<String> oldTensorName, Operand<Long> rowRemapping, Operand<Long> colRemapping, Operand<Float> initializingValues, Long numRows, Long numCols, LoadAndRemapMatrix.Options... options) Adds an LoadAndRemapMatrix operation to the graph
<T> Log<T>	log(Operand<T> x) Adds an Log operation to the graph
<T> Log1p<T>	log1p(Operand<T> x) Adds an Log1p operation to the graph
LogicalAnd	logicalAnd(Operand<Boolean> x, Operand<Boolean> y) Adds an LogicalAnd operation to the graph
LogicalNot	logicalNot(Operand<Boolean> x) Adds an LogicalNot operation to the graph
LogicalOr	logicalOr(Operand<Boolean> x, Operand<Boolean> y) Adds an LogicalOr operation to the graph
<T> LogMatrixDeterminant<T>	logMatrixDeterminant(Operand<T> input) Adds an LogMatrixDeterminant operation to the graph
<T extends Number> LogSoftmax<T>	logSoftmax(Operand<T> logits) Adds an LogSoftmax operation to the graph
LogUniformCandidateSampler	logUniformCandidateSampler(Operand<Long> trueClasses, Long numTrue, Long numSampled, Boolean unique, Long rangeMax, LogUniformCandidateSampler.Options... options) Adds an LogUniformCandidateSampler operation to the graph
<T,U> LookupTableExport<T,U>	lookupTableExport(Operand<?> tableHandle, Class<T> Tkeys, Class<U> Tvalues) Adds an LookupTableExport operation to the graph
<U,T> LookupTableFind<U>	lookupTableFind(Operand<?> tableHandle, Operand<T> keys, Operand<U> defaultValue) Adds an LookupTableFind operation to the graph
<T,U> LookupTableImport	lookupTableImport(Operand<?> tableHandle, Operand<T> keys, Operand<U> values) Adds an LookupTableImport operation to the graph
<T,U> LookupTableInsert	lookupTableInsert(Operand<?> tableHandle, Operand<T> keys, Operand<U> values) Adds an LookupTableInsert operation to the graph
LookupTableSize	lookupTableSize(Operand<?> tableHandle) Adds an LookupTableSize operation to the graph
LoopCond	loopCond(Operand<Boolean> input) Adds an LoopCond operation to the graph
<T extends Number> LRN<T>	lRN(Operand<T> input, LRN.Options... options) Adds an LRN operation to the graph
<T> Lu<T,Integer>	lu(Operand<T> input) Adds an Lu operation to the graph
<T,U extends Number> Lu<T,U>	lu(Operand<T> input, Class<U> outputIdxType) Adds an Lu operation to the graph
MakeIterator	makeIterator(Operand<?> dataset, Operand<?> iterator) Adds an MakeIterator operation to the graph
MapClear	mapClear(List<Class<?>> dtypes, MapClear.Options... options) Adds an MapClear operation to the graph
MapIncompleteSize	mapIncompleteSize(List<Class<?>> dtypes, MapIncompleteSize.Options... options) Adds an MapIncompleteSize operation to the graph
MapPeek	mapPeek(Operand<Long> key, Operand<Integer> indices, List<Class<?>> dtypes, MapPeek.Options... options) Adds an MapPeek operation to the graph
MapSize	mapSize(List<Class<?>> dtypes, MapSize.Options... options) Adds an MapSize operation to the graph
MapStage	mapStage(Operand<Long> key, Operand<Integer> indices, Iterable<Operand<?>> values, List<Class<?>> dtypes, MapStage.Options... options) Adds an MapStage operation to the graph
MapUnstage	mapUnstage(Operand<Long> key, Operand<Integer> indices, List<Class<?>> dtypes, MapUnstage.Options... options) Adds an MapUnstage operation to the graph
MapUnstageNoKey	mapUnstageNoKey(Operand<Integer> indices, List<Class<?>> dtypes, MapUnstageNoKey.Options... options) Adds an MapUnstageNoKey operation to the graph
MatchingFiles	matchingFiles(Operand<String> pattern) Adds an MatchingFiles operation to the graph
<T> MatMul<T>	matMul(Operand<T> a, Operand<T> b, MatMul.Options... options) Adds an MatMul operation to the graph
<T,U extends Number> MatrixBandPart<T>	matrixBandPart(Operand<T> input, Operand<U> numLower, Operand<U> numUpper) Adds an MatrixBandPart operation to the graph
<T> MatrixDeterminant<T>	matrixDeterminant(Operand<T> input) Adds an MatrixDeterminant operation to the graph
<T> MatrixDiag<T>	matrixDiag(Operand<T> diagonal) Adds an MatrixDiag operation to the graph
<T> MatrixDiagPart<T>	matrixDiagPart(Operand<T> input) Adds an MatrixDiagPart operation to the graph
<T> MatrixInverse<T>	matrixInverse(Operand<T> input, MatrixInverse.Options... options) Adds an MatrixInverse operation to the graph
<T> MatrixSetDiag<T>	matrixSetDiag(Operand<T> input, Operand<T> diagonal) Adds an MatrixSetDiag operation to the graph
<T> MatrixSolve<T>	matrixSolve(Operand<T> matrix, Operand<T> rhs, MatrixSolve.Options... options) Adds an MatrixSolve operation to the graph
<T> MatrixSolveLs<T>	matrixSolveLs(Operand<T> matrix, Operand<T> rhs, Operand<Double> l2Regularizer, MatrixSolveLs.Options... options) Adds an MatrixSolveLs operation to the graph
<T> MatrixSquareRoot<T>	matrixSquareRoot(Operand<T> input) Adds an MatrixSquareRoot operation to the graph
<T> MatrixTriangularSolve<T>	matrixTriangularSolve(Operand<T> matrix, Operand<T> rhs, MatrixTriangularSolve.Options... options) Adds an MatrixTriangularSolve operation to the graph
<T,U extends Number> Max<T>	max(Operand<T> input, Operand<U> axis, Max.Options... options) Adds an Max operation to the graph
<T extends Number> Maximum<T>	maximum(Operand<T> x, Operand<T> y) Adds an Maximum operation to the graph
<T> MaxPool<T>	maxPool(Operand<T> input, List<Long> ksize, List<Long> strides, String padding, MaxPool.Options... options) Adds an MaxPool operation to the graph
<T extends Number> MaxPool3D<T>	maxPool3D(Operand<T> input, List<Long> ksize, List<Long> strides, String padding, MaxPool3D.Options... options) Adds an MaxPool3D operation to the graph
<U extends Number,T extends Number> MaxPool3DGrad<U>	maxPool3DGrad(Operand<T> origInput, Operand<T> origOutput, Operand<U> grad, List<Long> ksize, List<Long> strides, String padding, MaxPool3DGrad.Options... options) Adds an MaxPool3DGrad operation to the graph
<T extends Number> MaxPool3DGradGrad<T>	maxPool3DGradGrad(Operand<T> origInput, Operand<T> origOutput, Operand<T> grad, List<Long> ksize, List<Long> strides, String padding, MaxPool3DGradGrad.Options... options) Adds an MaxPool3DGradGrad operation to the graph
<T extends Number> MaxPoolGradGrad<T>	maxPoolGradGrad(Operand<T> origInput, Operand<T> origOutput, Operand<T> grad, List<Long> ksize, List<Long> strides, String padding, MaxPoolGradGrad.Options... options) Adds an MaxPoolGradGrad operation to the graph
<T extends Number> MaxPoolGradGradV2<T>	maxPoolGradGradV2(Operand<T> origInput, Operand<T> origOutput, Operand<T> grad, Operand<Integer> ksize, Operand<Integer> strides, String padding, MaxPoolGradGradV2.Options... options) Adds an MaxPoolGradGradV2 operation to the graph
<T extends Number,U extends Number> MaxPoolGradGradWithArgmax<T>	maxPoolGradGradWithArgmax(Operand<T> input, Operand<T> grad, Operand<U> argmax, List<Long> ksize, List<Long> strides, String padding) Adds an MaxPoolGradGradWithArgmax operation to the graph
<T extends Number> MaxPoolGradV2<T>	maxPoolGradV2(Operand<T> origInput, Operand<T> origOutput, Operand<T> grad, Operand<Integer> ksize, Operand<Integer> strides, String padding, MaxPoolGradV2.Options... options) Adds an MaxPoolGradV2 operation to the graph
<T> MaxPoolV2<T>	maxPoolV2(Operand<T> input, Operand<Integer> ksize, Operand<Integer> strides, String padding, MaxPoolV2.Options... options) Adds an MaxPoolV2 operation to the graph
<T extends Number,U extends Number> MaxPoolWithArgmax<T,U>	maxPoolWithArgmax(Operand<T> input, List<Long> ksize, List<Long> strides, Class<U> Targmax, String padding) Adds an MaxPoolWithArgmax operation to the graph
<T extends Number> MaxPoolWithArgmax<T,Long>	maxPoolWithArgmax(Operand<T> input, List<Long> ksize, List<Long> strides, String padding) Adds an MaxPoolWithArgmax operation to the graph
<T,U extends Number> Mean<T>	mean(Operand<T> input, Operand<U> axis, Mean.Options... options) Adds an Mean operation to the graph
<T> Merge<T>	merge(Operand<T> inputs) Adds an Merge operation to the graph
MergeSummary	mergeSummary(Iterable<Operand<String>> inputs) Adds an MergeSummary operation to the graph
MergeV2Checkpoints	mergeV2Checkpoints(Operand<String> checkpointPrefixes, Operand<String> destinationPrefix, MergeV2Checkpoints.Options... options) Adds an MergeV2Checkpoints operation to the graph
Mfcc	mfcc(Operand<Float> spectrogram, Operand<Integer> sampleRate, Mfcc.Options... options) Adds an Mfcc operation to the graph
<T,U extends Number> Min<T>	min(Operand<T> input, Operand<U> axis, Min.Options... options) Adds an Min operation to the graph
<T extends Number> Minimum<T>	minimum(Operand<T> x, Operand<T> y) Adds an Minimum operation to the graph
<T,U extends Number> MirrorPad<T>	mirrorPad(Operand<T> input, Operand<U> paddings, String mode) Adds an MirrorPad operation to the graph
<T extends Number> Mod<T>	mod(Operand<T> x, Operand<T> y) Adds an Mod operation to the graph
<T> Mul<T>	mul(Operand<T> x, Operand<T> y) Adds an Mul operation to the graph
<U extends Number,T extends Number> Multinomial<U>	multinomial(Operand<T> logits, Operand<Integer> numSamples, Class<U> outputDtype, Multinomial.Options... options) Adds an Multinomial operation to the graph
<T extends Number> Multinomial<Long>	multinomial(Operand<T> logits, Operand<Integer> numSamples, Multinomial.Options... options) Adds an Multinomial operation to the graph
<T> Multiply<T>	multiply(Operand<T> x, Operand<T> y) Adds an Multiply operation to the graph
<T,U> MutableDenseHashTable	mutableDenseHashTable(Operand<T> emptyKey, Operand<T> deletedKey, Class<U> valueDtype, MutableDenseHashTable.Options... options) Adds an MutableDenseHashTable operation to the graph
<T,U> MutableHashTable	mutableHashTable(Class<T> keyDtype, Class<U> valueDtype, MutableHashTable.Options... options) Adds an MutableHashTable operation to the graph
<T,U> MutableHashTableOfTensors	mutableHashTableOfTensors(Class<T> keyDtype, Class<U> valueDtype, MutableHashTableOfTensors.Options... options) Adds an MutableHashTableOfTensors operation to the graph
MutexLock	mutexLock(Operand<?> mutex) Adds an MutexLock operation to the graph
MutexV2	mutexV2(MutexV2.Options... options) Adds an MutexV2 operation to the graph
<T> Neg<T>	neg(Operand<T> x) Adds an Neg operation to the graph
<T> Negate<T>	negate(Operand<T> x) Adds an Negate operation to the graph
NegTrain	negTrain(Operand<Float> wIn, Operand<Float> wOut, Operand<Integer> examples, Operand<Integer> labels, Operand<Float> lr, List<Long> vocabCount, Long numNegativeSamples) Adds an NegTrain operation to the graph
<T> NextIteration<T>	nextIteration(Operand<T> data) Adds an NextIteration operation to the graph
NonMaxSuppression	nonMaxSuppression(Operand<Float> boxes, Operand<Float> scores, Operand<Integer> maxOutputSize, NonMaxSuppression.Options... options) Adds an NonMaxSuppression operation to the graph
<T extends Number> NonMaxSuppressionV2	nonMaxSuppressionV2(Operand<T> boxes, Operand<T> scores, Operand<Integer> maxOutputSize, Operand<Float> iouThreshold) Adds an NonMaxSuppressionV2 operation to the graph
<T extends Number> NonMaxSuppressionV3	nonMaxSuppressionV3(Operand<T> boxes, Operand<T> scores, Operand<Integer> maxOutputSize, Operand<Float> iouThreshold, Operand<Float> scoreThreshold) Adds an NonMaxSuppressionV3 operation to the graph
<T extends Number> NonMaxSuppressionV4	nonMaxSuppressionV4(Operand<T> boxes, Operand<T> scores, Operand<Integer> maxOutputSize, Operand<Float> iouThreshold, Operand<Float> scoreThreshold, NonMaxSuppressionV4.Options... options) Adds an NonMaxSuppressionV4 operation to the graph
NonMaxSuppressionWithOverlaps	nonMaxSuppressionWithOverlaps(Operand<Float> overlaps, Operand<Float> scores, Operand<Integer> maxOutputSize, Operand<Float> overlapThreshold, Operand<Float> scoreThreshold) Adds an NonMaxSuppressionWithOverlaps operation to the graph
NoOp	noOp() Adds an NoOp operation to the graph
<T> NotEqual	notEqual(Operand<T> x, Operand<T> y) Adds an NotEqual operation to the graph
<T extends Number> NthElement<T>	nthElement(Operand<T> input, Operand<Integer> n, NthElement.Options... options) Adds an NthElement operation to the graph
<U,T extends Number> OneHot<U>	oneHot(Operand<T> indices, Operand<Integer> depth, Operand<U> onValue, Operand<U> offValue, OneHot.Options... options) Adds an OneHot operation to the graph
<T> OnesLike<T>	onesLike(Operand<T> x) Adds an OnesLike operation to the graph
OptionalFromValue	optionalFromValue(Iterable<Operand<?>> components) Adds an OptionalFromValue operation to the graph
OptionalGetValue	optionalGetValue(Operand<?> optional, List<Class<?>> outputTypes, List<Shape> outputShapes) Adds an OptionalGetValue operation to the graph
OptionalHasValue	optionalHasValue(Operand<?> optional) Adds an OptionalHasValue operation to the graph
OptionalNone	optionalNone() Adds an OptionalNone operation to the graph
OrderedMapClear	orderedMapClear(List<Class<?>> dtypes, OrderedMapClear.Options... options) Adds an OrderedMapClear operation to the graph
OrderedMapIncompleteSize	orderedMapIncompleteSize(List<Class<?>> dtypes, OrderedMapIncompleteSize.Options... options) Adds an OrderedMapIncompleteSize operation to the graph
OrderedMapPeek	orderedMapPeek(Operand<Long> key, Operand<Integer> indices, List<Class<?>> dtypes, OrderedMapPeek.Options... options) Adds an OrderedMapPeek operation to the graph
OrderedMapSize	orderedMapSize(List<Class<?>> dtypes, OrderedMapSize.Options... options) Adds an OrderedMapSize operation to the graph
OrderedMapStage	orderedMapStage(Operand<Long> key, Operand<Integer> indices, Iterable<Operand<?>> values, List<Class<?>> dtypes, OrderedMapStage.Options... options) Adds an OrderedMapStage operation to the graph
OrderedMapUnstage	orderedMapUnstage(Operand<Long> key, Operand<Integer> indices, List<Class<?>> dtypes, OrderedMapUnstage.Options... options) Adds an OrderedMapUnstage operation to the graph
OrderedMapUnstageNoKey	orderedMapUnstageNoKey(Operand<Integer> indices, List<Class<?>> dtypes, OrderedMapUnstageNoKey.Options... options) Adds an OrderedMapUnstageNoKey operation to the graph
<T,U extends Number> Pad<T>	pad(Operand<T> input, Operand<U> paddings) Adds an Pad operation to the graph
PaddingFIFOQueue	paddingFIFOQueue(List<Class<?>> componentTypes, PaddingFIFOQueue.Options... options) Adds an PaddingFIFOQueue operation to the graph
<T,U extends Number> PadV2<T>	padV2(Operand<T> input, Operand<U> paddings, Operand<T> constantValues) Adds an PadV2 operation to the graph
<T> ParallelConcat<T>	parallelConcat(Operand<T> values, Shape shape) Adds an ParallelConcat operation to the graph
<T> ParallelDynamicStitch<T>	parallelDynamicStitch(Iterable<Operand<Integer>> indices, Operand<T> data) Adds an ParallelDynamicStitch operation to the graph
<U extends Number,T extends Number> ParameterizedTruncatedNormal<U>	parameterizedTruncatedNormal(Operand<T> shape, Operand<U> means, Operand<U> stdevs, Operand<U> minvals, Operand<U> maxvals, ParameterizedTruncatedNormal.Options... options) Adds an ParameterizedTruncatedNormal operation to the graph
ParseExample	parseExample(Operand<String> serialized, Operand<String> names, Iterable<Operand<String>> sparseKeys, Iterable<Operand<String>> denseKeys, Iterable<Operand<?>> denseDefaults, List<Class<?>> sparseTypes, List<Shape> denseShapes) Adds an ParseExample operation to the graph
ParseSequenceExample	parseSequenceExample(Operand<String> serialized, Operand<String> debugName, Iterable<Operand<?>> contextDenseDefaults, List<String> featureListDenseMissingAssumedEmpty, List<String> contextSparseKeys, List<String> contextDenseKeys, List<String> featureListSparseKeys, List<String> featureListDenseKeys, List<Class<?>> contextSparseTypes, List<Class<?>> featureListDenseTypes, List<Class<?>> featureListSparseTypes, ParseSequenceExample.Options... options) Adds an ParseSequenceExample operation to the graph
ParseSingleExample	parseSingleExample(Operand<String> serialized, Iterable<Operand<?>> denseDefaults, Long numSparse, List<String> sparseKeys, List<String> denseKeys, List<Class<?>> sparseTypes, List<Shape> denseShapes) Adds an ParseSingleExample operation to the graph
ParseSingleSequenceExample	parseSingleSequenceExample(Operand<String> serialized, Operand<String> featureListDenseMissingAssumedEmpty, Iterable<Operand<String>> contextSparseKeys, Iterable<Operand<String>> contextDenseKeys, Iterable<Operand<String>> featureListSparseKeys, Iterable<Operand<String>> featureListDenseKeys, Iterable<Operand<?>> contextDenseDefaults, Operand<String> debugName, List<Class<?>> contextSparseTypes, List<Class<?>> featureListDenseTypes, List<Class<?>> featureListSparseTypes, ParseSingleSequenceExample.Options... options) Adds an ParseSingleSequenceExample operation to the graph
<T> ParseTensor<T>	parseTensor(Operand<String> serialized, Class<T> outType) Adds an ParseTensor operation to the graph
<T> Placeholder<T>	placeholder(Class<T> dtype, Placeholder.Options... options) Adds an Placeholder operation to the graph
<T> PlaceholderV2<T>	placeholderV2(Class<T> dtype, Shape shape) Adds an PlaceholderV2 operation to the graph
<T> PlaceholderWithDefault<T>	placeholderWithDefault(Operand<T> input, Shape shape) Adds an PlaceholderWithDefault operation to the graph
<T extends Number> Polygamma<T>	polygamma(Operand<T> a, Operand<T> x) Adds an Polygamma operation to the graph
<T extends Number> PopulationCount	populationCount(Operand<T> x) Adds an PopulationCount operation to the graph
<T> Pow<T>	pow(Operand<T> x, Operand<T> y) Adds an Pow operation to the graph
<T> PreventGradient<T>	preventGradient(Operand<T> input, PreventGradient.Options... options) Adds an PreventGradient operation to the graph
<T> Print<T>	print(Operand<T> input, Iterable<Operand<?>> data, Print.Options... options) Adds an Print operation to the graph
PrintV2	printV2(Operand<String> input, PrintV2.Options... options) Adds an PrintV2 operation to the graph
PriorityQueue	priorityQueue(List<Class<?>> componentTypes, List<Shape> shapes, PriorityQueue.Options... options) Adds an PriorityQueue operation to the graph
<T,U extends Number> Prod<T>	prod(Operand<T> input, Operand<U> axis, Prod.Options... options) Adds an Prod operation to the graph
<T> Qr<T>	qr(Operand<T> input, Qr.Options... options) Adds an Qr operation to the graph
<T extends Number> QuantizeAndDequantize<T>	quantizeAndDequantize(Operand<T> input, QuantizeAndDequantize.Options... options) Adds an QuantizeAndDequantize operation to the graph
<T extends Number> QuantizeAndDequantizeV2<T>	quantizeAndDequantizeV2(Operand<T> input, Operand<T> inputMin, Operand<T> inputMax, QuantizeAndDequantizeV2.Options... options) Adds an QuantizeAndDequantizeV2 operation to the graph
<T extends Number> QuantizeAndDequantizeV3<T>	quantizeAndDequantizeV3(Operand<T> input, Operand<T> inputMin, Operand<T> inputMax, Operand<Integer> numBits, QuantizeAndDequantizeV3.Options... options) Adds an QuantizeAndDequantizeV3 operation to the graph
<V,T,U> QuantizedAdd<V>	quantizedAdd(Operand<T> x, Operand<U> y, Operand<Float> minX, Operand<Float> maxX, Operand<Float> minY, Operand<Float> maxY, Class<V> Toutput) Adds an QuantizedAdd operation to the graph
<T> QuantizedAvgPool<T>	quantizedAvgPool(Operand<T> input, Operand<Float> minInput, Operand<Float> maxInput, List<Long> ksize, List<Long> strides, String padding) Adds an QuantizedAvgPool operation to the graph
<U,T> QuantizedBatchNormWithGlobalNormalization<U>	quantizedBatchNormWithGlobalNormalization(Operand<T> t, Operand<Float> tMin, Operand<Float> tMax, Operand<T> m, Operand<Float> mMin, Operand<Float> mMax, Operand<T> v, Operand<Float> vMin, Operand<Float> vMax, Operand<T> beta, Operand<Float> betaMin, Operand<Float> betaMax, Operand<T> gamma, Operand<Float> gammaMin, Operand<Float> gammaMax, Class<U> outType, Float varianceEpsilon, Boolean scaleAfterNormalization) Adds an QuantizedBatchNormWithGlobalNormalization operation to the graph
<V,T,U> QuantizedBiasAdd<V>	quantizedBiasAdd(Operand<T> input, Operand<U> bias, Operand<Float> minInput, Operand<Float> maxInput, Operand<Float> minBias, Operand<Float> maxBias, Class<V> outType) Adds an QuantizedBiasAdd operation to the graph
<T> QuantizedConcat<T>	quantizedConcat(Operand<Integer> concatDim, Operand<T> values, Iterable<Operand<Float>> inputMins, Iterable<Operand<Float>> inputMaxes) Adds an QuantizedConcat operation to the graph
<V,T,U> QuantizedConv2D<V>	quantizedConv2D(Operand<T> input, Operand<U> filter, Operand<Float> minInput, Operand<Float> maxInput, Operand<Float> minFilter, Operand<Float> maxFilter, Class<V> outType, List<Long> strides, String padding, QuantizedConv2D.Options... options) Adds an QuantizedConv2D operation to the graph
<T> QuantizedInstanceNorm<T>	quantizedInstanceNorm(Operand<T> x, Operand<Float> xMin, Operand<Float> xMax, QuantizedInstanceNorm.Options... options) Adds an QuantizedInstanceNorm operation to the graph
<V,T,U,W> QuantizedMatMul<V>	quantizedMatMul(Operand<T> a, Operand<U> b, Operand<Float> minA, Operand<Float> maxA, Operand<Float> minB, Operand<Float> maxB, Class<V> Toutput, Class<W> Tactivation, QuantizedMatMul.Options... options) Adds an QuantizedMatMul operation to the graph
<T> QuantizedMaxPool<T>	quantizedMaxPool(Operand<T> input, Operand<Float> minInput, Operand<Float> maxInput, List<Long> ksize, List<Long> strides, String padding) Adds an QuantizedMaxPool operation to the graph
<V,T,U> QuantizedMul<V>	quantizedMul(Operand<T> x, Operand<U> y, Operand<Float> minX, Operand<Float> maxX, Operand<Float> minY, Operand<Float> maxY, Class<V> Toutput) Adds an QuantizedMul operation to the graph
<U,T> QuantizeDownAndShrinkRange<U>	quantizeDownAndShrinkRange(Operand<T> input, Operand<Float> inputMin, Operand<Float> inputMax, Class<U> outType) Adds an QuantizeDownAndShrinkRange operation to the graph
<U,T> QuantizedRelu<U>	quantizedRelu(Operand<T> features, Operand<Float> minFeatures, Operand<Float> maxFeatures, Class<U> outType) Adds an QuantizedRelu operation to the graph
<U,T> QuantizedRelu6<U>	quantizedRelu6(Operand<T> features, Operand<Float> minFeatures, Operand<Float> maxFeatures, Class<U> outType) Adds an QuantizedRelu6 operation to the graph
<U,T> QuantizedReluX<U>	quantizedReluX(Operand<T> features, Operand<Float> maxValue, Operand<Float> minFeatures, Operand<Float> maxFeatures, Class<U> outType) Adds an QuantizedReluX operation to the graph
<T,U extends Number> QuantizedReshape<T>	quantizedReshape(Operand<T> tensor, Operand<U> shape, Operand<Float> inputMin, Operand<Float> inputMax) Adds an QuantizedReshape operation to the graph
<T> QuantizedResizeBilinear<T>	quantizedResizeBilinear(Operand<T> images, Operand<Integer> size, Operand<Float> min, Operand<Float> max, QuantizedResizeBilinear.Options... options) Adds an QuantizedResizeBilinear operation to the graph
<T> QuantizeV2<T>	quantizeV2(Operand<Float> input, Operand<Float> minRange, Operand<Float> maxRange, Class<T> T, QuantizeV2.Options... options) Adds an QuantizeV2 operation to the graph
QueueClose	queueClose(Operand<?> handle, QueueClose.Options... options) Adds an QueueClose operation to the graph
QueueDequeue	queueDequeue(Operand<?> handle, List<Class<?>> componentTypes, QueueDequeue.Options... options) Adds an QueueDequeue operation to the graph
QueueDequeueMany	queueDequeueMany(Operand<?> handle, Operand<Integer> n, List<Class<?>> componentTypes, QueueDequeueMany.Options... options) Adds an QueueDequeueMany operation to the graph
QueueDequeueUpTo	queueDequeueUpTo(Operand<?> handle, Operand<Integer> n, List<Class<?>> componentTypes, QueueDequeueUpTo.Options... options) Adds an QueueDequeueUpTo operation to the graph
QueueEnqueue	queueEnqueue(Operand<?> handle, Iterable<Operand<?>> components, QueueEnqueue.Options... options) Adds an QueueEnqueue operation to the graph
QueueEnqueueMany	queueEnqueueMany(Operand<?> handle, Iterable<Operand<?>> components, QueueEnqueueMany.Options... options) Adds an QueueEnqueueMany operation to the graph
QueueIsClosed	queueIsClosed(Operand<String> handle) Adds an QueueIsClosed operation to the graph
QueueIsClosedV2	queueIsClosedV2(Operand<?> handle) Adds an QueueIsClosedV2 operation to the graph
QueueSize	queueSize(Operand<?> handle) Adds an QueueSize operation to the graph
<T extends Number> RandomCrop<T>	randomCrop(Operand<T> image, Operand<Long> size, RandomCrop.Options... options) Adds an RandomCrop operation to the graph
<U extends Number,T extends Number> RandomGamma<U>	randomGamma(Operand<T> shape, Operand<U> alpha, RandomGamma.Options... options) Adds an RandomGamma operation to the graph
<U extends Number,T extends Number> RandomNormal<U>	randomNormal(Operand<T> shape, Class<U> dtype, RandomNormal.Options... options) Adds an RandomNormal operation to the graph
<U extends Number,T extends Number> RandomPoisson<U>	randomPoisson(Operand<T> shape, Operand<U> rate, RandomPoisson.Options... options) Adds an RandomPoisson operation to the graph
<V extends Number,T extends Number,U extends Number> RandomPoissonV2<V>	randomPoissonV2(Operand<T> shape, Operand<U> rate, Class<V> dtype, RandomPoissonV2.Options... options) Adds an RandomPoissonV2 operation to the graph
<T extends Number,U extends Number> RandomPoissonV2<Long>	randomPoissonV2(Operand<T> shape, Operand<U> rate, RandomPoissonV2.Options... options) Adds an RandomPoissonV2 operation to the graph
<T> RandomShuffle<T>	randomShuffle(Operand<T> value, RandomShuffle.Options... options) Adds an RandomShuffle operation to the graph
RandomShuffleQueue	randomShuffleQueue(List<Class<?>> componentTypes, RandomShuffleQueue.Options... options) Adds an RandomShuffleQueue operation to the graph
<U extends Number,T extends Number> RandomUniform<U>	randomUniform(Operand<T> shape, Class<U> dtype, RandomUniform.Options... options) Adds an RandomUniform operation to the graph
<U extends Number,T extends Number> RandomUniformInt<U>	randomUniformInt(Operand<T> shape, Operand<U> minval, Operand<U> maxval, RandomUniformInt.Options... options) Adds an RandomUniformInt operation to the graph
<T extends Number> Range<T>	range(Operand<T> start, Operand<T> limit, Operand<T> delta) Adds an Range operation to the graph
<T> Rank	rank(Operand<T> input) Adds an Rank operation to the graph
ReaderNumRecordsProduced	readerNumRecordsProduced(Operand<?> readerHandle) Adds an ReaderNumRecordsProduced operation to the graph
ReaderNumWorkUnitsCompleted	readerNumWorkUnitsCompleted(Operand<?> readerHandle) Adds an ReaderNumWorkUnitsCompleted operation to the graph
ReaderRead	readerRead(Operand<?> readerHandle, Operand<?> queueHandle) Adds an ReaderRead operation to the graph
ReaderReadUpTo	readerReadUpTo(Operand<?> readerHandle, Operand<?> queueHandle, Operand<Long> numRecords) Adds an ReaderReadUpTo operation to the graph
ReaderReset	readerReset(Operand<?> readerHandle) Adds an ReaderReset operation to the graph
ReaderRestoreState	readerRestoreState(Operand<?> readerHandle, Operand<String> state) Adds an ReaderRestoreState operation to the graph
ReaderSerializeState	readerSerializeState(Operand<?> readerHandle) Adds an ReaderSerializeState operation to the graph
ReadFile	readFile(Operand<String> filename) Adds an ReadFile operation to the graph
<T> ReadVariableOp<T>	readVariableOp(Operand<?> resource, Class<T> dtype) Adds an ReadVariableOp operation to the graph
<T> Real<Float>	real(Operand<T> input) Adds an Real operation to the graph
<U extends Number,T> Real<U>	real(Operand<T> input, Class<U> Tout) Adds an Real operation to the graph
<T> RealDiv<T>	realDiv(Operand<T> x, Operand<T> y) Adds an RealDiv operation to the graph
<T> Reciprocal<T>	reciprocal(Operand<T> x) Adds an Reciprocal operation to the graph
RecordInput	recordInput(String filePattern, RecordInput.Options... options) Adds an RecordInput operation to the graph
<T extends Number> ReduceAll	reduceAll(Operand<Boolean> input, Operand<T> axis, ReduceAll.Options... options) Adds an ReduceAll operation to the graph
<T extends Number> ReduceAny	reduceAny(Operand<Boolean> input, Operand<T> axis, ReduceAny.Options... options) Adds an ReduceAny operation to the graph
ReduceJoin	reduceJoin(Operand<String> inputs, Operand<Integer> reductionIndices, ReduceJoin.Options... options) Adds an ReduceJoin operation to the graph
<T,U extends Number> ReduceMax<T>	reduceMax(Operand<T> input, Operand<U> axis, ReduceMax.Options... options) Adds an ReduceMax operation to the graph
<T,U extends Number> ReduceMean<T>	reduceMean(Operand<T> input, Operand<U> axis, ReduceMean.Options... options) Adds an ReduceMean operation to the graph
<T,U extends Number> ReduceMin<T>	reduceMin(Operand<T> input, Operand<U> axis, ReduceMin.Options... options) Adds an ReduceMin operation to the graph
<T,U extends Number> ReduceProd<T>	reduceProd(Operand<T> input, Operand<U> axis, ReduceProd.Options... options) Adds an ReduceProd operation to the graph
<T,U extends Number> ReduceSum<T>	reduceSum(Operand<T> input, Operand<U> axis, ReduceSum.Options... options) Adds an ReduceSum operation to the graph
<T> RefNextIteration<T>	refNextIteration(Operand<T> data) Adds an RefNextIteration operation to the graph
<T> RefSelect<T>	refSelect(Operand<Integer> index, Operand<T> inputs) Adds an RefSelect operation to the graph
<T> RefSwitch<T>	refSwitch(Operand<T> data, Operand<Boolean> pred) Adds an RefSwitch operation to the graph
RegexFullMatch	regexFullMatch(Operand<String> input, Operand<String> pattern) Adds an RegexFullMatch operation to the graph
RegexReplace	regexReplace(Operand<String> input, Operand<String> pattern, Operand<String> rewrite, RegexReplace.Options... options) Adds an RegexReplace operation to the graph
<T> Relu<T>	relu(Operand<T> features) Adds an Relu operation to the graph
<T extends Number> Relu6<T>	relu6(Operand<T> features) Adds an Relu6 operation to the graph
RemoteFusedGraphExecute	remoteFusedGraphExecute(Iterable<Operand<?>> inputs, List<Class<?>> Toutputs, String serializedRemoteFusedGraphExecuteInfo) Adds an RemoteFusedGraphExecute operation to the graph
<T> RequantizationRange	requantizationRange(Operand<T> input, Operand<Float> inputMin, Operand<Float> inputMax) Adds an RequantizationRange operation to the graph
<U,T> Requantize<U>	requantize(Operand<T> input, Operand<Float> inputMin, Operand<Float> inputMax, Operand<Float> requestedOutputMin, Operand<Float> requestedOutputMax, Class<U> outType) Adds an Requantize operation to the graph
<T,U extends Number> Reshape<T>	reshape(Operand<T> tensor, Operand<U> shape) Adds an Reshape operation to the graph
<T extends Number> ResizeArea	resizeArea(Operand<T> images, Operand<Integer> size, ResizeArea.Options... options) Adds an ResizeArea operation to the graph
<T extends Number> ResizeBicubic	resizeBicubic(Operand<T> images, Operand<Integer> size, ResizeBicubic.Options... options) Adds an ResizeBicubic operation to the graph
<T extends Number> ResizeBilinear	resizeBilinear(Operand<T> images, Operand<Integer> size, ResizeBilinear.Options... options) Adds an ResizeBilinear operation to the graph
<T extends Number> ResizeNearestNeighbor<T>	resizeNearestNeighbor(Operand<T> images, Operand<Integer> size, ResizeNearestNeighbor.Options... options) Adds an ResizeNearestNeighbor operation to the graph
<T> ResourceApplyAdadelta	resourceApplyAdadelta(Operand<?> var, Operand<?> accum, Operand<?> accumUpdate, Operand<T> lr, Operand<T> rho, Operand<T> epsilon, Operand<T> grad, ResourceApplyAdadelta.Options... options) Adds an ResourceApplyAdadelta operation to the graph
<T> ResourceApplyAdagrad	resourceApplyAdagrad(Operand<?> var, Operand<?> accum, Operand<T> lr, Operand<T> grad, ResourceApplyAdagrad.Options... options) Adds an ResourceApplyAdagrad operation to the graph
<T> ResourceApplyAdagradDA	resourceApplyAdagradDA(Operand<?> var, Operand<?> gradientAccumulator, Operand<?> gradientSquaredAccumulator, Operand<T> grad, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<Long> globalStep, ResourceApplyAdagradDA.Options... options) Adds an ResourceApplyAdagradDA operation to the graph
<T> ResourceApplyAdam	resourceApplyAdam(Operand<?> var, Operand<?> m, Operand<?> v, Operand<T> beta1Power, Operand<T> beta2Power, Operand<T> lr, Operand<T> beta1, Operand<T> beta2, Operand<T> epsilon, Operand<T> grad, ResourceApplyAdam.Options... options) Adds an ResourceApplyAdam operation to the graph
<T> ResourceApplyAdamWithAmsgrad	resourceApplyAdamWithAmsgrad(Operand<?> var, Operand<?> m, Operand<?> v, Operand<?> vhat, Operand<T> beta1Power, Operand<T> beta2Power, Operand<T> lr, Operand<T> beta1, Operand<T> beta2, Operand<T> epsilon, Operand<T> grad, ResourceApplyAdamWithAmsgrad.Options... options) Adds an ResourceApplyAdamWithAmsgrad operation to the graph
<T> ResourceApplyAddSign	resourceApplyAddSign(Operand<?> var, Operand<?> m, Operand<T> lr, Operand<T> alpha, Operand<T> signDecay, Operand<T> beta, Operand<T> grad, ResourceApplyAddSign.Options... options) Adds an ResourceApplyAddSign operation to the graph
<T> ResourceApplyCenteredRMSProp	resourceApplyCenteredRMSProp(Operand<?> var, Operand<?> mg, Operand<?> ms, Operand<?> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, ResourceApplyCenteredRMSProp.Options... options) Adds an ResourceApplyCenteredRMSProp operation to the graph
<T> ResourceApplyFtrl	resourceApplyFtrl(Operand<?> var, Operand<?> accum, Operand<?> linear, Operand<T> grad, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> lrPower, ResourceApplyFtrl.Options... options) Adds an ResourceApplyFtrl operation to the graph
<T> ResourceApplyFtrlV2	resourceApplyFtrlV2(Operand<?> var, Operand<?> accum, Operand<?> linear, Operand<T> grad, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> l2Shrinkage, Operand<T> lrPower, ResourceApplyFtrlV2.Options... options) Adds an ResourceApplyFtrlV2 operation to the graph
<T> ResourceApplyGradientDescent	resourceApplyGradientDescent(Operand<?> var, Operand<T> alpha, Operand<T> delta, ResourceApplyGradientDescent.Options... options) Adds an ResourceApplyGradientDescent operation to the graph
<T> ResourceApplyKerasMomentum	resourceApplyKerasMomentum(Operand<?> var, Operand<?> accum, Operand<T> lr, Operand<T> grad, Operand<T> momentum, ResourceApplyKerasMomentum.Options... options) Adds an ResourceApplyKerasMomentum operation to the graph
<T> ResourceApplyMomentum	resourceApplyMomentum(Operand<?> var, Operand<?> accum, Operand<T> lr, Operand<T> grad, Operand<T> momentum, ResourceApplyMomentum.Options... options) Adds an ResourceApplyMomentum operation to the graph
<T> ResourceApplyPowerSign	resourceApplyPowerSign(Operand<?> var, Operand<?> m, Operand<T> lr, Operand<T> logbase, Operand<T> signDecay, Operand<T> beta, Operand<T> grad, ResourceApplyPowerSign.Options... options) Adds an ResourceApplyPowerSign operation to the graph
<T> ResourceApplyProximalAdagrad	resourceApplyProximalAdagrad(Operand<?> var, Operand<?> accum, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> grad, ResourceApplyProximalAdagrad.Options... options) Adds an ResourceApplyProximalAdagrad operation to the graph
<T> ResourceApplyProximalGradientDescent	resourceApplyProximalGradientDescent(Operand<?> var, Operand<T> alpha, Operand<T> l1, Operand<T> l2, Operand<T> delta, ResourceApplyProximalGradientDescent.Options... options) Adds an ResourceApplyProximalGradientDescent operation to the graph
<T> ResourceApplyRMSProp	resourceApplyRMSProp(Operand<?> var, Operand<?> ms, Operand<?> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, ResourceApplyRMSProp.Options... options) Adds an ResourceApplyRMSProp operation to the graph
<T extends Number> ResourceCountUpTo<T>	resourceCountUpTo(Operand<?> resource, Long limit, Class<T> T) Adds an ResourceCountUpTo operation to the graph
<U,T extends Number> ResourceGather<U>	resourceGather(Operand<?> resource, Operand<T> indices, Class<U> dtype, ResourceGather.Options... options) Adds an ResourceGather operation to the graph
<T extends Number,U> ResourceScatterAdd	resourceScatterAdd(Operand<?> resource, Operand<T> indices, Operand<U> updates) Adds an ResourceScatterAdd operation to the graph
<T extends Number,U> ResourceScatterDiv	resourceScatterDiv(Operand<?> resource, Operand<T> indices, Operand<U> updates) Adds an ResourceScatterDiv operation to the graph
<T extends Number,U> ResourceScatterMax	resourceScatterMax(Operand<?> resource, Operand<T> indices, Operand<U> updates) Adds an ResourceScatterMax operation to the graph
<T extends Number,U> ResourceScatterMin	resourceScatterMin(Operand<?> resource, Operand<T> indices, Operand<U> updates) Adds an ResourceScatterMin operation to the graph
<T extends Number,U> ResourceScatterMul	resourceScatterMul(Operand<?> resource, Operand<T> indices, Operand<U> updates) Adds an ResourceScatterMul operation to the graph
<T extends Number,U> ResourceScatterNdAdd	resourceScatterNdAdd(Operand<?> ref, Operand<T> indices, Operand<U> updates, ResourceScatterNdAdd.Options... options) Adds an ResourceScatterNdAdd operation to the graph
<T extends Number,U> ResourceScatterNdUpdate	resourceScatterNdUpdate(Operand<?> ref, Operand<T> indices, Operand<U> updates, ResourceScatterNdUpdate.Options... options) Adds an ResourceScatterNdUpdate operation to the graph
<T extends Number,U> ResourceScatterSub	resourceScatterSub(Operand<?> resource, Operand<T> indices, Operand<U> updates) Adds an ResourceScatterSub operation to the graph
<T extends Number,U> ResourceScatterUpdate	resourceScatterUpdate(Operand<?> resource, Operand<T> indices, Operand<U> updates) Adds an ResourceScatterUpdate operation to the graph
<T,U extends Number> ResourceSparseApplyAdadelta	resourceSparseApplyAdadelta(Operand<?> var, Operand<?> accum, Operand<?> accumUpdate, Operand<T> lr, Operand<T> rho, Operand<T> epsilon, Operand<T> grad, Operand<U> indices, ResourceSparseApplyAdadelta.Options... options) Adds an ResourceSparseApplyAdadelta operation to the graph
<T,U extends Number> ResourceSparseApplyAdagrad	resourceSparseApplyAdagrad(Operand<?> var, Operand<?> accum, Operand<T> lr, Operand<T> grad, Operand<U> indices, ResourceSparseApplyAdagrad.Options... options) Adds an ResourceSparseApplyAdagrad operation to the graph
<T,U extends Number> ResourceSparseApplyAdagradDA	resourceSparseApplyAdagradDA(Operand<?> var, Operand<?> gradientAccumulator, Operand<?> gradientSquaredAccumulator, Operand<T> grad, Operand<U> indices, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<Long> globalStep, ResourceSparseApplyAdagradDA.Options... options) Adds an ResourceSparseApplyAdagradDA operation to the graph
<T,U extends Number> ResourceSparseApplyCenteredRMSProp	resourceSparseApplyCenteredRMSProp(Operand<?> var, Operand<?> mg, Operand<?> ms, Operand<?> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, Operand<U> indices, ResourceSparseApplyCenteredRMSProp.Options... options) Adds an ResourceSparseApplyCenteredRMSProp operation to the graph
<T,U extends Number> ResourceSparseApplyFtrl	resourceSparseApplyFtrl(Operand<?> var, Operand<?> accum, Operand<?> linear, Operand<T> grad, Operand<U> indices, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> lrPower, ResourceSparseApplyFtrl.Options... options) Adds an ResourceSparseApplyFtrl operation to the graph
<T,U extends Number> ResourceSparseApplyFtrlV2	resourceSparseApplyFtrlV2(Operand<?> var, Operand<?> accum, Operand<?> linear, Operand<T> grad, Operand<U> indices, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> l2Shrinkage, Operand<T> lrPower, ResourceSparseApplyFtrlV2.Options... options) Adds an ResourceSparseApplyFtrlV2 operation to the graph
<T,U extends Number> ResourceSparseApplyKerasMomentum	resourceSparseApplyKerasMomentum(Operand<?> var, Operand<?> accum, Operand<T> lr, Operand<T> grad, Operand<U> indices, Operand<T> momentum, ResourceSparseApplyKerasMomentum.Options... options) Adds an ResourceSparseApplyKerasMomentum operation to the graph
<T,U extends Number> ResourceSparseApplyMomentum	resourceSparseApplyMomentum(Operand<?> var, Operand<?> accum, Operand<T> lr, Operand<T> grad, Operand<U> indices, Operand<T> momentum, ResourceSparseApplyMomentum.Options... options) Adds an ResourceSparseApplyMomentum operation to the graph
<T,U extends Number> ResourceSparseApplyProximalAdagrad	resourceSparseApplyProximalAdagrad(Operand<?> var, Operand<?> accum, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> grad, Operand<U> indices, ResourceSparseApplyProximalAdagrad.Options... options) Adds an ResourceSparseApplyProximalAdagrad operation to the graph
<T,U extends Number> ResourceSparseApplyProximalGradientDescent	resourceSparseApplyProximalGradientDescent(Operand<?> var, Operand<T> alpha, Operand<T> l1, Operand<T> l2, Operand<T> grad, Operand<U> indices, ResourceSparseApplyProximalGradientDescent.Options... options) Adds an ResourceSparseApplyProximalGradientDescent operation to the graph
<T,U extends Number> ResourceSparseApplyRMSProp	resourceSparseApplyRMSProp(Operand<?> var, Operand<?> ms, Operand<?> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, Operand<U> indices, ResourceSparseApplyRMSProp.Options... options) Adds an ResourceSparseApplyRMSProp operation to the graph
<T extends Number,U> ResourceStridedSliceAssign	resourceStridedSliceAssign(Operand<?> ref, Operand<T> begin, Operand<T> end, Operand<T> strides, Operand<U> value, ResourceStridedSliceAssign.Options... options) Adds an ResourceStridedSliceAssign operation to the graph
<T> Restore<T>	restore(Operand<String> filePattern, Operand<String> tensorName, Class<T> dt, Restore.Options... options) Adds an Restore operation to the graph
<T> RestoreSlice<T>	restoreSlice(Operand<String> filePattern, Operand<String> tensorName, Operand<String> shapeAndSlice, Class<T> dt, RestoreSlice.Options... options) Adds an RestoreSlice operation to the graph
RestoreV2	restoreV2(Operand<String> prefix, Operand<String> tensorNames, Operand<String> shapeAndSlices, List<Class<?>> dtypes) Adds an RestoreV2 operation to the graph
<T,U extends Number> Reverse<T>	reverse(Operand<T> tensor, Operand<U> axis) Adds an Reverse operation to the graph
<T,U extends Number> ReverseSequence<T>	reverseSequence(Operand<T> input, Operand<U> seqLengths, Long seqDim, ReverseSequence.Options... options) Adds an ReverseSequence operation to the graph
RFFT	rFFT(Operand<Float> input, Operand<Integer> fftLength) Adds an RFFT operation to the graph
RFFT2D	rFFT2D(Operand<Float> input, Operand<Integer> fftLength) Adds an RFFT2D operation to the graph
RFFT3D	rFFT3D(Operand<Float> input, Operand<Integer> fftLength) Adds an RFFT3D operation to the graph
<T extends Number> RGBToHSV<T>	rGBToHSV(Operand<T> images) Adds an RGBToHSV operation to the graph
<T extends Number> RightShift<T>	rightShift(Operand<T> x, Operand<T> y) Adds an RightShift operation to the graph
<T extends Number> Rint<T>	rint(Operand<T> x) Adds an Rint operation to the graph
<T,U extends Number,V extends Number> Roll<T>	roll(Operand<T> input, Operand<U> shift, Operand<V> axis) Adds an Roll operation to the graph
<T> Round<T>	round(Operand<T> x) Adds an Round operation to the graph
Rpc	rpc(Operand<String> address, Operand<String> method, Operand<String> request, Rpc.Options... options) Adds an Rpc operation to the graph
<T> Rsqrt<T>	rsqrt(Operand<T> x) Adds an Rsqrt operation to the graph
<T extends Number> SampleDistortedBoundingBox<T>	sampleDistortedBoundingBox(Operand<T> imageSize, Operand<Float> boundingBoxes, SampleDistortedBoundingBox.Options... options) Adds an SampleDistortedBoundingBox operation to the graph
<T extends Number> SampleDistortedBoundingBoxV2<T>	sampleDistortedBoundingBoxV2(Operand<T> imageSize, Operand<Float> boundingBoxes, Operand<Float> minObjectCovered, SampleDistortedBoundingBoxV2.Options... options) Adds an SampleDistortedBoundingBoxV2 operation to the graph
Save	save(Operand<String> filename, Operand<String> tensorNames, Iterable<Operand<?>> data) Adds an Save operation to the graph
SaveSlices	saveSlices(Operand<String> filename, Operand<String> tensorNames, Operand<String> shapesAndSlices, Iterable<Operand<?>> data) Adds an SaveSlices operation to the graph
SaveV2	saveV2(Operand<String> prefix, Operand<String> tensorNames, Operand<String> shapeAndSlices, Iterable<Operand<?>> tensors) Adds an SaveV2 operation to the graph
<T extends Number> ScalarSummary	scalarSummary(Operand<String> tags, Operand<T> values) Adds an ScalarSummary operation to the graph
<T,U extends Number> ScatterAdd<T>	scatterAdd(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterAdd.Options... options) Adds an ScatterAdd operation to the graph
<T,U extends Number> ScatterDiv<T>	scatterDiv(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterDiv.Options... options) Adds an ScatterDiv operation to the graph
<T extends Number,U extends Number> ScatterMax<T>	scatterMax(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterMax.Options... options) Adds an ScatterMax operation to the graph
<T extends Number,U extends Number> ScatterMin<T>	scatterMin(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterMin.Options... options) Adds an ScatterMin operation to the graph
<T,U extends Number> ScatterMul<T>	scatterMul(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterMul.Options... options) Adds an ScatterMul operation to the graph
<U,T extends Number> ScatterNd<U>	scatterNd(Operand<T> indices, Operand<U> updates, Operand<T> shape) Adds an ScatterNd operation to the graph
<T,U extends Number> ScatterNdAdd<T>	scatterNdAdd(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterNdAdd.Options... options) Adds an ScatterNdAdd operation to the graph
<T,U extends Number> ScatterNdNonAliasingAdd<T>	scatterNdNonAliasingAdd(Operand<T> input, Operand<U> indices, Operand<T> updates) Adds an ScatterNdNonAliasingAdd operation to the graph
<T,U extends Number> ScatterNdSub<T>	scatterNdSub(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterNdSub.Options... options) Adds an ScatterNdSub operation to the graph
<T,U extends Number> ScatterNdUpdate<T>	scatterNdUpdate(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterNdUpdate.Options... options) Adds an ScatterNdUpdate operation to the graph
<T,U extends Number> ScatterSub<T>	scatterSub(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterSub.Options... options) Adds an ScatterSub operation to the graph
<T,U extends Number> ScatterUpdate<T>	scatterUpdate(Operand<T> ref, Operand<U> indices, Operand<T> updates, ScatterUpdate.Options... options) Adds an ScatterUpdate operation to the graph
Scope	scope() Returns the current scope of this API
SdcaFprint	sdcaFprint(Operand<String> input) Adds an SdcaFprint operation to the graph
SdcaOptimizer	sdcaOptimizer(Iterable<Operand<Long>> sparseExampleIndices, Iterable<Operand<Long>> sparseFeatureIndices, Iterable<Operand<Float>> sparseFeatureValues, Iterable<Operand<Float>> denseFeatures, Operand<Float> exampleWeights, Operand<Float> exampleLabels, Iterable<Operand<Long>> sparseIndices, Iterable<Operand<Float>> sparseWeights, Iterable<Operand<Float>> denseWeights, Operand<Float> exampleStateData, String lossType, Float l1, Float l2, Long numLossPartitions, Long numInnerIterations, SdcaOptimizer.Options... options) Adds an SdcaOptimizer operation to the graph
SdcaShrinkL1	sdcaShrinkL1(Iterable<Operand<Float>> weights, Float l1, Float l2) Adds an SdcaShrinkL1 operation to the graph
<T extends Number,U extends Number> SegmentMax<T>	segmentMax(Operand<T> data, Operand<U> segmentIds) Adds an SegmentMax operation to the graph
<T,U extends Number> SegmentMean<T>	segmentMean(Operand<T> data, Operand<U> segmentIds) Adds an SegmentMean operation to the graph
<T extends Number,U extends Number> SegmentMin<T>	segmentMin(Operand<T> data, Operand<U> segmentIds) Adds an SegmentMin operation to the graph
<T,U extends Number> SegmentProd<T>	segmentProd(Operand<T> data, Operand<U> segmentIds) Adds an SegmentProd operation to the graph
<T,U extends Number> SegmentSum<T>	segmentSum(Operand<T> data, Operand<U> segmentIds) Adds an SegmentSum operation to the graph
<T> SelfAdjointEig<T>	selfAdjointEig(Operand<T> input, SelfAdjointEig.Options... options) Adds an SelfAdjointEig operation to the graph
<T extends Number> Selu<T>	selu(Operand<T> features) Adds an Selu operation to the graph
SerializeIterator	serializeIterator(Operand<?> resourceHandle) Adds an SerializeIterator operation to the graph
<T> SerializeManySparse<String>	serializeManySparse(Operand<Long> sparseIndices, Operand<T> sparseValues, Operand<Long> sparseShape) Adds an SerializeManySparse operation to the graph
<U,T> SerializeManySparse<U>	serializeManySparse(Operand<Long> sparseIndices, Operand<T> sparseValues, Operand<Long> sparseShape, Class<U> outType) Adds an SerializeManySparse operation to the graph
<T> SerializeSparse<String>	serializeSparse(Operand<Long> sparseIndices, Operand<T> sparseValues, Operand<Long> sparseShape) Adds an SerializeSparse operation to the graph
<U,T> SerializeSparse<U>	serializeSparse(Operand<Long> sparseIndices, Operand<T> sparseValues, Operand<Long> sparseShape, Class<U> outType) Adds an SerializeSparse operation to the graph
<T> SerializeTensor	serializeTensor(Operand<T> tensor) Adds an SerializeTensor operation to the graph
<T> SetDiff1D<T,Integer>	setDiff1D(Operand<T> x, Operand<T> y) Adds an SetDiff1D operation to the graph
<T,U extends Number> SetDiff1D<T,U>	setDiff1D(Operand<T> x, Operand<T> y, Class<U> outIdx) Adds an SetDiff1D operation to the graph
<T> SetSize	setSize(Operand<Long> setIndices, Operand<T> setValues, Operand<Long> setShape, SetSize.Options... options) Adds an SetSize operation to the graph
<T> Shape<Integer>	shape(Operand<T> input) Adds an Shape operation to the graph
<U extends Number,T> Shape<U>	shape(Operand<T> input, Class<U> outType) Adds an Shape operation to the graph
<T> ShapeN<Integer>	shapeN(Operand<T> input) Adds an ShapeN operation to the graph
<U extends Number,T> ShapeN<U>	shapeN(Operand<T> input, Class<U> outType) Adds an ShapeN operation to the graph
ShardedFilename	shardedFilename(Operand<String> basename, Operand<Integer> shard, Operand<Integer> numShards) Adds an ShardedFilename operation to the graph
ShardedFilespec	shardedFilespec(Operand<String> basename, Operand<Integer> numShards) Adds an ShardedFilespec operation to the graph
<T> Sigmoid<T>	sigmoid(Operand<T> x) Adds an Sigmoid operation to the graph
<T> Sign<T>	sign(Operand<T> x) Adds an Sign operation to the graph
<T> Sin<T>	sin(Operand<T> x) Adds an Sin operation to the graph
<T> Sinh<T>	sinh(Operand<T> x) Adds an Sinh operation to the graph
<T> Size<Integer>	size(Operand<T> input) Adds an Size operation to the graph
<U extends Number,T> Size<U>	size(Operand<T> input, Class<U> outType) Adds an Size operation to the graph
Skipgram	skipgram(String filename, Long batchSize, Skipgram.Options... options) Adds an Skipgram operation to the graph
<T,U extends Number> Slice<T>	slice(Operand<T> input, Operand<U> begin, Operand<U> size) Adds an Slice operation to the graph
<T> Snapshot<T>	snapshot(Operand<T> input) Adds an Snapshot operation to the graph
<T extends Number> Softmax<T>	softmax(Operand<T> logits) Adds an Softmax operation to the graph
<T extends Number> SoftmaxCrossEntropyWithLogits<T>	softmaxCrossEntropyWithLogits(Operand<T> features, Operand<T> labels) Adds an SoftmaxCrossEntropyWithLogits operation to the graph
<T extends Number> Softplus<T>	softplus(Operand<T> features) Adds an Softplus operation to the graph
<T extends Number> Softsign<T>	softsign(Operand<T> features) Adds an Softsign operation to the graph
<T,U extends Number> SpaceToBatch<T>	spaceToBatch(Operand<T> input, Operand<U> paddings, Long blockSize) Adds an SpaceToBatch operation to the graph
<T,U extends Number,V extends Number> SpaceToBatchND<T>	spaceToBatchND(Operand<T> input, Operand<U> blockShape, Operand<V> paddings) Adds an SpaceToBatchND operation to the graph
<T> SpaceToDepth<T>	spaceToDepth(Operand<T> input, Long blockSize, SpaceToDepth.Options... options) Adds an SpaceToDepth operation to the graph
<T> SparseAccumulatorApplyGradient	sparseAccumulatorApplyGradient(Operand<String> handle, Operand<Long> localStep, Operand<Long> gradientIndices, Operand<T> gradientValues, Operand<Long> gradientShape, Boolean hasKnownShape) Adds an SparseAccumulatorApplyGradient operation to the graph
<T> SparseAccumulatorTakeGradient<T>	sparseAccumulatorTakeGradient(Operand<String> handle, Operand<Integer> numRequired, Class<T> dtype) Adds an SparseAccumulatorTakeGradient operation to the graph
<T,U extends Number> SparseAdd<T>	sparseAdd(Operand<Long> aIndices, Operand<T> aValues, Operand<Long> aShape, Operand<Long> bIndices, Operand<T> bValues, Operand<Long> bShape, Operand<U> thresh) Adds an SparseAdd operation to the graph
<T> SparseAddGrad<T>	sparseAddGrad(Operand<T> backpropValGrad, Operand<Long> aIndices, Operand<Long> bIndices, Operand<Long> sumIndices) Adds an SparseAddGrad operation to the graph
<T,U extends Number> SparseApplyAdadelta<T>	sparseApplyAdadelta(Operand<T> var, Operand<T> accum, Operand<T> accumUpdate, Operand<T> lr, Operand<T> rho, Operand<T> epsilon, Operand<T> grad, Operand<U> indices, SparseApplyAdadelta.Options... options) Adds an SparseApplyAdadelta operation to the graph
<T,U extends Number> SparseApplyAdagrad<T>	sparseApplyAdagrad(Operand<T> var, Operand<T> accum, Operand<T> lr, Operand<T> grad, Operand<U> indices, SparseApplyAdagrad.Options... options) Adds an SparseApplyAdagrad operation to the graph
<T,U extends Number> SparseApplyAdagradDA<T>	sparseApplyAdagradDA(Operand<T> var, Operand<T> gradientAccumulator, Operand<T> gradientSquaredAccumulator, Operand<T> grad, Operand<U> indices, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<Long> globalStep, SparseApplyAdagradDA.Options... options) Adds an SparseApplyAdagradDA operation to the graph
<T,U extends Number> SparseApplyCenteredRMSProp<T>	sparseApplyCenteredRMSProp(Operand<T> var, Operand<T> mg, Operand<T> ms, Operand<T> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, Operand<U> indices, SparseApplyCenteredRMSProp.Options... options) Adds an SparseApplyCenteredRMSProp operation to the graph
<T,U extends Number> SparseApplyFtrl<T>	sparseApplyFtrl(Operand<T> var, Operand<T> accum, Operand<T> linear, Operand<T> grad, Operand<U> indices, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> lrPower, SparseApplyFtrl.Options... options) Adds an SparseApplyFtrl operation to the graph
<T,U extends Number> SparseApplyFtrlV2<T>	sparseApplyFtrlV2(Operand<T> var, Operand<T> accum, Operand<T> linear, Operand<T> grad, Operand<U> indices, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> l2Shrinkage, Operand<T> lrPower, SparseApplyFtrlV2.Options... options) Adds an SparseApplyFtrlV2 operation to the graph
<T,U extends Number> SparseApplyMomentum<T>	sparseApplyMomentum(Operand<T> var, Operand<T> accum, Operand<T> lr, Operand<T> grad, Operand<U> indices, Operand<T> momentum, SparseApplyMomentum.Options... options) Adds an SparseApplyMomentum operation to the graph
<T,U extends Number> SparseApplyProximalAdagrad<T>	sparseApplyProximalAdagrad(Operand<T> var, Operand<T> accum, Operand<T> lr, Operand<T> l1, Operand<T> l2, Operand<T> grad, Operand<U> indices, SparseApplyProximalAdagrad.Options... options) Adds an SparseApplyProximalAdagrad operation to the graph
<T,U extends Number> SparseApplyProximalGradientDescent<T>	sparseApplyProximalGradientDescent(Operand<T> var, Operand<T> alpha, Operand<T> l1, Operand<T> l2, Operand<T> grad, Operand<U> indices, SparseApplyProximalGradientDescent.Options... options) Adds an SparseApplyProximalGradientDescent operation to the graph
<T,U extends Number> SparseApplyRMSProp<T>	sparseApplyRMSProp(Operand<T> var, Operand<T> ms, Operand<T> mom, Operand<T> lr, Operand<T> rho, Operand<T> momentum, Operand<T> epsilon, Operand<T> grad, Operand<U> indices, SparseApplyRMSProp.Options... options) Adds an SparseApplyRMSProp operation to the graph
<T> SparseConcat<T>	sparseConcat(Iterable<Operand<Long>> indices, Operand<T> values, Iterable<Operand<Long>> shapes, Long concatDim) Adds an SparseConcat operation to the graph
<T> SparseConditionalAccumulator	sparseConditionalAccumulator(Class<T> dtype, Shape shape, SparseConditionalAccumulator.Options... options) Adds an SparseConditionalAccumulator operation to the graph
<T,U> SparseCross<T>	sparseCross(Iterable<Operand<Long>> indices, Iterable<Operand<?>> values, Iterable<Operand<Long>> shapes, Iterable<Operand<?>> denseInputs, Boolean hashedOutput, Long numBuckets, Long hashKey, Class<T> outType, Class<U> internalType) Adds an SparseCross operation to the graph
<T> SparseDenseCwiseAdd<T>	sparseDenseCwiseAdd(Operand<Long> spIndices, Operand<T> spValues, Operand<Long> spShape, Operand<T> dense) Adds an SparseDenseCwiseAdd operation to the graph
<T> SparseDenseCwiseDiv<T>	sparseDenseCwiseDiv(Operand<Long> spIndices, Operand<T> spValues, Operand<Long> spShape, Operand<T> dense) Adds an SparseDenseCwiseDiv operation to the graph
<T> SparseDenseCwiseMul<T>	sparseDenseCwiseMul(Operand<Long> spIndices, Operand<T> spValues, Operand<Long> spShape, Operand<T> dense) Adds an SparseDenseCwiseMul operation to the graph
<T> SparseFillEmptyRows<T>	sparseFillEmptyRows(Operand<Long> indices, Operand<T> values, Operand<Long> denseShape, Operand<T> defaultValue) Adds an SparseFillEmptyRows operation to the graph
<T> SparseFillEmptyRowsGrad<T>	sparseFillEmptyRowsGrad(Operand<Long> reverseIndexMap, Operand<T> gradValues) Adds an SparseFillEmptyRowsGrad operation to the graph
<T extends Number,U extends Number> SparseMatMul	sparseMatMul(Operand<T> a, Operand<U> b, SparseMatMul.Options... options) Adds an SparseMatMul operation to the graph
<T extends Number> SparseReduceMax<T>	sparseReduceMax(Operand<Long> inputIndices, Operand<T> inputValues, Operand<Long> inputShape, Operand<Integer> reductionAxes, SparseReduceMax.Options... options) Adds an SparseReduceMax operation to the graph
<T extends Number> SparseReduceMaxSparse<T>	sparseReduceMaxSparse(Operand<Long> inputIndices, Operand<T> inputValues, Operand<Long> inputShape, Operand<Integer> reductionAxes, SparseReduceMaxSparse.Options... options) Adds an SparseReduceMaxSparse operation to the graph
<T> SparseReduceSum<T>	sparseReduceSum(Operand<Long> inputIndices, Operand<T> inputValues, Operand<Long> inputShape, Operand<Integer> reductionAxes, SparseReduceSum.Options... options) Adds an SparseReduceSum operation to the graph
<T> SparseReduceSumSparse<T>	sparseReduceSumSparse(Operand<Long> inputIndices, Operand<T> inputValues, Operand<Long> inputShape, Operand<Integer> reductionAxes, SparseReduceSumSparse.Options... options) Adds an SparseReduceSumSparse operation to the graph
<T> SparseReorder<T>	sparseReorder(Operand<Long> inputIndices, Operand<T> inputValues, Operand<Long> inputShape) Adds an SparseReorder operation to the graph
SparseReshape	sparseReshape(Operand<Long> inputIndices, Operand<Long> inputShape, Operand<Long> newShape) Adds an SparseReshape operation to the graph
<T extends Number,U extends Number> SparseSegmentMean<T>	sparseSegmentMean(Operand<T> data, Operand<U> indices, Operand<Integer> segmentIds) Adds an SparseSegmentMean operation to the graph
<T extends Number,U extends Number> SparseSegmentMeanGrad<T>	sparseSegmentMeanGrad(Operand<T> grad, Operand<U> indices, Operand<Integer> segmentIds, Operand<Integer> outputDim0) Adds an SparseSegmentMeanGrad operation to the graph
<T extends Number,U extends Number,V extends Number> SparseSegmentMeanWithNumSegments<T>	sparseSegmentMeanWithNumSegments(Operand<T> data, Operand<U> indices, Operand<Integer> segmentIds, Operand<V> numSegments) Adds an SparseSegmentMeanWithNumSegments operation to the graph
<T extends Number,U extends Number> SparseSegmentSqrtN<T>	sparseSegmentSqrtN(Operand<T> data, Operand<U> indices, Operand<Integer> segmentIds) Adds an SparseSegmentSqrtN operation to the graph
<T extends Number,U extends Number> SparseSegmentSqrtNGrad<T>	sparseSegmentSqrtNGrad(Operand<T> grad, Operand<U> indices, Operand<Integer> segmentIds, Operand<Integer> outputDim0) Adds an SparseSegmentSqrtNGrad operation to the graph
<T extends Number,U extends Number,V extends Number> SparseSegmentSqrtNWithNumSegments<T>	sparseSegmentSqrtNWithNumSegments(Operand<T> data, Operand<U> indices, Operand<Integer> segmentIds, Operand<V> numSegments) Adds an SparseSegmentSqrtNWithNumSegments operation to the graph
<T extends Number,U extends Number> SparseSegmentSum<T>	sparseSegmentSum(Operand<T> data, Operand<U> indices, Operand<Integer> segmentIds) Adds an SparseSegmentSum operation to the graph
<T extends Number,U extends Number,V extends Number> SparseSegmentSumWithNumSegments<T>	sparseSegmentSumWithNumSegments(Operand<T> data, Operand<U> indices, Operand<Integer> segmentIds, Operand<V> numSegments) Adds an SparseSegmentSumWithNumSegments operation to the graph
<T> SparseSlice<T>	sparseSlice(Operand<Long> indices, Operand<T> values, Operand<Long> shape, Operand<Long> start, Operand<Long> size) Adds an SparseSlice operation to the graph
<T> SparseSliceGrad<T>	sparseSliceGrad(Operand<T> backpropValGrad, Operand<Long> inputIndices, Operand<Long> inputStart, Operand<Long> outputIndices) Adds an SparseSliceGrad operation to the graph
<T extends Number> SparseSoftmax<T>	sparseSoftmax(Operand<Long> spIndices, Operand<T> spValues, Operand<Long> spShape) Adds an SparseSoftmax operation to the graph
<T extends Number,U extends Number> SparseSoftmaxCrossEntropyWithLogits<T>	sparseSoftmaxCrossEntropyWithLogits(Operand<T> features, Operand<U> labels) Adds an SparseSoftmaxCrossEntropyWithLogits operation to the graph
<T extends Number> SparseSparseMaximum<T>	sparseSparseMaximum(Operand<Long> aIndices, Operand<T> aValues, Operand<Long> aShape, Operand<Long> bIndices, Operand<T> bValues, Operand<Long> bShape) Adds an SparseSparseMaximum operation to the graph
<T> SparseSparseMinimum<T>	sparseSparseMinimum(Operand<Long> aIndices, Operand<T> aValues, Operand<Long> aShape, Operand<Long> bIndices, Operand<T> bValues, Operand<Long> bShape) Adds an SparseSparseMinimum operation to the graph
<T> SparseSplit<T>	sparseSplit(Operand<Long> splitDim, Operand<Long> indices, Operand<T> values, Operand<Long> shape, Long numSplit) Adds an SparseSplit operation to the graph
<U,T extends Number> SparseTensorDenseAdd<U>	sparseTensorDenseAdd(Operand<T> aIndices, Operand<U> aValues, Operand<T> aShape, Operand<U> b) Adds an SparseTensorDenseAdd operation to the graph
<U,T extends Number> SparseTensorDenseMatMul<U>	sparseTensorDenseMatMul(Operand<T> aIndices, Operand<U> aValues, Operand<Long> aShape, Operand<U> b, SparseTensorDenseMatMul.Options... options) Adds an SparseTensorDenseMatMul operation to the graph
<U,T extends Number> SparseToDense<U>	sparseToDense(Operand<T> sparseIndices, Operand<T> outputShape, Operand<U> sparseValues, Operand<U> defaultValue, SparseToDense.Options... options) Adds an SparseToDense operation to the graph
<T> SparseToSparseSetOperation<T>	sparseToSparseSetOperation(Operand<Long> set1Indices, Operand<T> set1Values, Operand<Long> set1Shape, Operand<Long> set2Indices, Operand<T> set2Values, Operand<Long> set2Shape, String setOperation, SparseToSparseSetOperation.Options... options) Adds an SparseToSparseSetOperation operation to the graph
<T> Split<T>	split(Operand<Integer> axis, Operand<T> value, Long numSplit) Adds an Split operation to the graph
<T,U extends Number> SplitV<T>	splitV(Operand<T> value, Operand<U> sizeSplits, Operand<Integer> axis, Long numSplit) Adds an SplitV operation to the graph
<T> Sqrt<T>	sqrt(Operand<T> x) Adds an Sqrt operation to the graph
<T> Square<T>	square(Operand<T> x) Adds an Square operation to the graph
<T> SquaredDifference<T>	squaredDifference(Operand<T> x, Operand<T> y) Adds an SquaredDifference operation to the graph
<T> Squeeze<T>	squeeze(Operand<T> input, Squeeze.Options... options) Adds an Squeeze operation to the graph
<T> Stack<T>	stack(Operand<T> values, Stack.Options... options) Adds an Stack operation to the graph
Stage	stage(Iterable<Operand<?>> values, Stage.Options... options) Adds an Stage operation to the graph
StageClear	stageClear(List<Class<?>> dtypes, StageClear.Options... options) Adds an StageClear operation to the graph
StagePeek	stagePeek(Operand<Integer> index, List<Class<?>> dtypes, StagePeek.Options... options) Adds an StagePeek operation to the graph
StageSize	stageSize(List<Class<?>> dtypes, StageSize.Options... options) Adds an StageSize operation to the graph
<T extends Number,U extends Number> StatelessMultinomial<Long>	statelessMultinomial(Operand<T> logits, Operand<Integer> numSamples, Operand<U> seed) Adds an StatelessMultinomial operation to the graph
<V extends Number,T extends Number,U extends Number> StatelessMultinomial<V>	statelessMultinomial(Operand<T> logits, Operand<Integer> numSamples, Operand<U> seed, Class<V> outputDtype) Adds an StatelessMultinomial operation to the graph
<T extends Number,U extends Number> StatelessRandomNormal<Float>	statelessRandomNormal(Operand<T> shape, Operand<U> seed) Adds an StatelessRandomNormal operation to the graph
<V extends Number,T extends Number,U extends Number> StatelessRandomNormal<V>	statelessRandomNormal(Operand<T> shape, Operand<U> seed, Class<V> dtype) Adds an StatelessRandomNormal operation to the graph
<T extends Number,U extends Number> StatelessRandomUniform<Float>	statelessRandomUniform(Operand<T> shape, Operand<U> seed) Adds an StatelessRandomUniform operation to the graph
<V extends Number,T extends Number,U extends Number> StatelessRandomUniform<V>	statelessRandomUniform(Operand<T> shape, Operand<U> seed, Class<V> dtype) Adds an StatelessRandomUniform operation to the graph
<T extends Number,U extends Number> StatelessTruncatedNormal<Float>	statelessTruncatedNormal(Operand<T> shape, Operand<U> seed) Adds an StatelessTruncatedNormal operation to the graph
<V extends Number,T extends Number,U extends Number> StatelessTruncatedNormal<V>	statelessTruncatedNormal(Operand<T> shape, Operand<U> seed, Class<V> dtype) Adds an StatelessTruncatedNormal operation to the graph
<T> StopGradient<T>	stopGradient(Operand<T> input) Adds an StopGradient operation to the graph
<T,U extends Number> StridedSlice<T>	stridedSlice(Operand<T> input, Operand<U> begin, Operand<U> end, Operand<U> strides, StridedSlice.Options... options) Adds an StridedSlice operation to the graph
<T,U extends Number> StridedSliceAssign<T>	stridedSliceAssign(Operand<T> ref, Operand<U> begin, Operand<U> end, Operand<U> strides, Operand<T> value, StridedSliceAssign.Options... options) Adds an StridedSliceAssign operation to the graph
<U,T extends Number> StridedSliceGrad<U>	stridedSliceGrad(Operand<T> shape, Operand<T> begin, Operand<T> end, Operand<T> strides, Operand<U> dy, StridedSliceGrad.Options... options) Adds an StridedSliceGrad operation to the graph
StringFormat	stringFormat(Iterable<Operand<?>> inputs, StringFormat.Options... options) Adds an StringFormat operation to the graph
StringJoin	stringJoin(Iterable<Operand<String>> inputs, StringJoin.Options... options) Adds an StringJoin operation to the graph
StringLength	stringLength(Operand<String> input, StringLength.Options... options) Adds an StringLength operation to the graph
StringSplit	stringSplit(Operand<String> input, Operand<String> delimiter, StringSplit.Options... options) Adds an StringSplit operation to the graph
StringSplitV2	stringSplitV2(Operand<String> input, Operand<String> sep, StringSplitV2.Options... options) Adds an StringSplitV2 operation to the graph
StringStrip	stringStrip(Operand<String> input) Adds an StringStrip operation to the graph
StringToHashBucket	stringToHashBucket(Operand<String> stringTensor, Long numBuckets) Adds an StringToHashBucket operation to the graph
StringToHashBucketFast	stringToHashBucketFast(Operand<String> input, Long numBuckets) Adds an StringToHashBucketFast operation to the graph
StringToHashBucketStrong	stringToHashBucketStrong(Operand<String> input, Long numBuckets, List<Long> key) Adds an StringToHashBucketStrong operation to the graph
StringToNumber<Float>	stringToNumber(Operand<String> stringTensor) Adds an StringToNumber operation to the graph
<T extends Number> StringToNumber<T>	stringToNumber(Operand<String> stringTensor, Class<T> outType) Adds an StringToNumber operation to the graph
<T> Sub<T>	sub(Operand<T> x, Operand<T> y) Adds an Sub operation to the graph
<T extends Number> Substr	substr(Operand<String> input, Operand<T> pos, Operand<T> len, Substr.Options... options) Adds an Substr operation to the graph
<T> Subtract<T>	subtract(Operand<T> x, Operand<T> y) Adds an Subtract operation to the graph
<T,U extends Number> Sum<T>	sum(Operand<T> input, Operand<U> axis, Sum.Options... options) Adds an Sum operation to the graph
<T> Svd<T>	svd(Operand<T> input, Svd.Options... options) Adds an Svd operation to the graph
<T> TakeManySparseFromTensorsMap<T>	takeManySparseFromTensorsMap(Operand<Long> sparseHandles, Class<T> dtype, TakeManySparseFromTensorsMap.Options... options) Adds an TakeManySparseFromTensorsMap operation to the graph
<T> Tan<T>	tan(Operand<T> x) Adds an Tan operation to the graph
<T> Tanh<T>	tanh(Operand<T> x) Adds an Tanh operation to the graph
<T> TemporaryVariable<T>	temporaryVariable(Shape shape, Class<T> dtype, TemporaryVariable.Options... options) Adds an TemporaryVariable operation to the graph
<T> TensorArray	tensorArray(Operand<Integer> size, Class<T> dtype, TensorArray.Options... options) Adds an TensorArray operation to the graph
TensorArrayClose	tensorArrayClose(Operand<?> handle) Adds an TensorArrayClose operation to the graph
<T> TensorArrayConcat<T>	tensorArrayConcat(Operand<?> handle, Operand<Float> flowIn, Class<T> dtype, TensorArrayConcat.Options... options) Adds an TensorArrayConcat operation to the graph
<T> TensorArrayGather<T>	tensorArrayGather(Operand<?> handle, Operand<Integer> indices, Operand<Float> flowIn, Class<T> dtype, TensorArrayGather.Options... options) Adds an TensorArrayGather operation to the graph
TensorArrayGrad	tensorArrayGrad(Operand<?> handle, Operand<Float> flowIn, String source) Adds an TensorArrayGrad operation to the graph
TensorArrayGradWithShape	tensorArrayGradWithShape(Operand<?> handle, Operand<Float> flowIn, Operand<Integer> shapeToPrepend, String source) Adds an TensorArrayGradWithShape operation to the graph
<T> TensorArrayPack<T>	tensorArrayPack(Operand<String> handle, Operand<Float> flowIn, Class<T> dtype, TensorArrayPack.Options... options) Adds an TensorArrayPack operation to the graph
<T> TensorArrayRead<T>	tensorArrayRead(Operand<?> handle, Operand<Integer> index, Operand<Float> flowIn, Class<T> dtype) Adds an TensorArrayRead operation to the graph
<T> TensorArrayScatter	tensorArrayScatter(Operand<?> handle, Operand<Integer> indices, Operand<T> value, Operand<Float> flowIn) Adds an TensorArrayScatter operation to the graph
TensorArraySize	tensorArraySize(Operand<?> handle, Operand<Float> flowIn) Adds an TensorArraySize operation to the graph
<T> TensorArraySplit	tensorArraySplit(Operand<?> handle, Operand<T> value, Operand<Long> lengths, Operand<Float> flowIn) Adds an TensorArraySplit operation to the graph
<T> TensorArrayUnpack	tensorArrayUnpack(Operand<String> handle, Operand<T> value, Operand<Float> flowIn) Adds an TensorArrayUnpack operation to the graph
<T> TensorArrayWrite	tensorArrayWrite(Operand<?> handle, Operand<Integer> index, Operand<T> value, Operand<Float> flowIn) Adds an TensorArrayWrite operation to the graph
<T> TensorListConcat<T>	tensorListConcat(Operand<?> inputHandle, Class<T> elementDtype) Adds an TensorListConcat operation to the graph
<T> TensorListConcatLists	tensorListConcatLists(Operand<?> inputA, Operand<?> inputB, Class<T> elementDtype) Adds an TensorListConcatLists operation to the graph
<T extends Number> TensorListElementShape<T>	tensorListElementShape(Operand<?> inputHandle, Class<T> shapeType) Adds an TensorListElementShape operation to the graph
<T,U extends Number> TensorListFromTensor	tensorListFromTensor(Operand<T> tensor, Operand<U> elementShape) Adds an TensorListFromTensor operation to the graph
<T> TensorListGather<T>	tensorListGather(Operand<?> inputHandle, Operand<Integer> indices, Class<T> elementDtype) Adds an TensorListGather operation to the graph
<T> TensorListGetItem<T>	tensorListGetItem(Operand<?> inputHandle, Operand<Integer> index, Class<T> elementDtype) Adds an TensorListGetItem operation to the graph
TensorListLength	tensorListLength(Operand<?> inputHandle) Adds an TensorListLength operation to the graph
<T> TensorListPopBack<T>	tensorListPopBack(Operand<?> inputHandle, Class<T> elementDtype) Adds an TensorListPopBack operation to the graph
<T> TensorListPushBack	tensorListPushBack(Operand<?> inputHandle, Operand<T> tensor) Adds an TensorListPushBack operation to the graph
<T> TensorListPushBackBatch	tensorListPushBackBatch(Operand<?> inputHandles, Operand<T> tensor) Adds an TensorListPushBackBatch operation to the graph
<T extends Number,U> TensorListReserve	tensorListReserve(Operand<T> elementShape, Operand<Integer> numElements, Class<U> elementDtype) Adds an TensorListReserve operation to the graph
<T,U extends Number> TensorListScatter	tensorListScatter(Operand<T> tensor, Operand<Integer> indices, Operand<U> elementShape) Adds an TensorListScatter operation to the graph
<T> TensorListSetItem	tensorListSetItem(Operand<?> inputHandle, Operand<Integer> index, Operand<T> item) Adds an TensorListSetItem operation to the graph
<T,U extends Number> TensorListSplit	tensorListSplit(Operand<T> tensor, Operand<U> elementShape, Operand<Long> lengths) Adds an TensorListSplit operation to the graph
<T> TensorListStack<T>	tensorListStack(Operand<?> inputHandle, Class<T> elementDtype, TensorListStack.Options... options) Adds an TensorListStack operation to the graph
<T,U extends Number> TensorScatterAdd<T>	tensorScatterAdd(Operand<T> tensor, Operand<U> indices, Operand<T> updates) Adds an TensorScatterAdd operation to the graph
<T,U extends Number> TensorScatterSub<T>	tensorScatterSub(Operand<T> tensor, Operand<U> indices, Operand<T> updates) Adds an TensorScatterSub operation to the graph
<T,U extends Number> TensorScatterUpdate<T>	tensorScatterUpdate(Operand<T> tensor, Operand<U> indices, Operand<T> updates) Adds an TensorScatterUpdate operation to the graph
<T> TensorSummary	tensorSummary(Operand<T> tensor, TensorSummary.Options... options) Adds an TensorSummary operation to the graph
<T> TensorSummaryV2	tensorSummaryV2(Operand<String> tag, Operand<T> tensor, Operand<String> serializedSummaryMetadata) Adds an TensorSummaryV2 operation to the graph
TextLineReader	textLineReader(TextLineReader.Options... options) Adds an TextLineReader operation to the graph
TFRecordReader	tFRecordReader(TFRecordReader.Options... options) Adds an TFRecordReader operation to the graph
<T,U extends Number> Tile<T>	tile(Operand<T> input, Operand<U> multiples) Adds an Tile operation to the graph
<T> TileGrad<T>	tileGrad(Operand<T> input, Operand<Integer> multiples) Adds an TileGrad operation to the graph
Timestamp	timestamp() Adds an Timestamp operation to the graph
<T extends Number> TopK<T>	topK(Operand<T> input, Operand<Integer> k, TopK.Options... options) Adds an TopK operation to the graph
<T,U extends Number> Transpose<T>	transpose(Operand<T> x, Operand<U> perm) Adds an Transpose operation to the graph
<T> TruncateDiv<T>	truncateDiv(Operand<T> x, Operand<T> y) Adds an TruncateDiv operation to the graph
<U extends Number,T extends Number> TruncatedNormal<U>	truncatedNormal(Operand<T> shape, Class<U> dtype, TruncatedNormal.Options... options) Adds an TruncatedNormal operation to the graph
<T extends Number> TruncateMod<T>	truncateMod(Operand<T> x, Operand<T> y) Adds an TruncateMod operation to the graph
TryRpc	tryRpc(Operand<String> address, Operand<String> method, Operand<String> request, TryRpc.Options... options) Adds an TryRpc operation to the graph
<T> Unbatch<T>	unbatch(Operand<T> batchedTensor, Operand<Long> batchIndex, Operand<Long> id, Long timeoutMicros, Unbatch.Options... options) Adds an Unbatch operation to the graph
<T> UnbatchGrad<T>	unbatchGrad(Operand<T> originalInput, Operand<Long> batchIndex, Operand<T> grad, Operand<Long> id, UnbatchGrad.Options... options) Adds an UnbatchGrad operation to the graph
UnicodeScript	unicodeScript(Operand<Integer> input) Adds an UnicodeScript operation to the graph
UnicodeTranscode	unicodeTranscode(Operand<String> input, String inputEncoding, String outputEncoding, UnicodeTranscode.Options... options) Adds an UnicodeTranscode operation to the graph
UniformCandidateSampler	uniformCandidateSampler(Operand<Long> trueClasses, Long numTrue, Long numSampled, Boolean unique, Long rangeMax, UniformCandidateSampler.Options... options) Adds an UniformCandidateSampler operation to the graph
<T> Unique<T,Integer>	unique(Operand<T> x) Adds an Unique operation to the graph
<T,U extends Number> Unique<T,U>	unique(Operand<T> x, Class<U> outIdx) Adds an Unique operation to the graph
<T,U extends Number> UniqueV2<T,Integer>	uniqueV2(Operand<T> x, Operand<U> axis) Adds an UniqueV2 operation to the graph
<T,V extends Number,U extends Number> UniqueV2<T,V>	uniqueV2(Operand<T> x, Operand<U> axis, Class<V> outIdx) Adds an UniqueV2 operation to the graph
<T> UniqueWithCounts<T,Integer>	uniqueWithCounts(Operand<T> x) Adds an UniqueWithCounts operation to the graph
<T,U extends Number> UniqueWithCounts<T,U>	uniqueWithCounts(Operand<T> x, Class<U> outIdx) Adds an UniqueWithCounts operation to the graph
<T,U extends Number> UniqueWithCountsV2<T,Integer>	uniqueWithCountsV2(Operand<T> x, Operand<U> axis) Adds an UniqueWithCountsV2 operation to the graph
<T,V extends Number,U extends Number> UniqueWithCountsV2<T,V>	uniqueWithCountsV2(Operand<T> x, Operand<U> axis, Class<V> outIdx) Adds an UniqueWithCountsV2 operation to the graph
<T extends Number> UnravelIndex<T>	unravelIndex(Operand<T> indices, Operand<T> dims) Adds an UnravelIndex operation to the graph
<T extends Number,U extends Number,V extends Number> UnsortedSegmentMax<T>	unsortedSegmentMax(Operand<T> data, Operand<U> segmentIds, Operand<V> numSegments) Adds an UnsortedSegmentMax operation to the graph
<T extends Number,U extends Number,V extends Number> UnsortedSegmentMin<T>	unsortedSegmentMin(Operand<T> data, Operand<U> segmentIds, Operand<V> numSegments) Adds an UnsortedSegmentMin operation to the graph
<T,U extends Number,V extends Number> UnsortedSegmentProd<T>	unsortedSegmentProd(Operand<T> data, Operand<U> segmentIds, Operand<V> numSegments) Adds an UnsortedSegmentProd operation to the graph
<T,U extends Number,V extends Number> UnsortedSegmentSum<T>	unsortedSegmentSum(Operand<T> data, Operand<U> segmentIds, Operand<V> numSegments) Adds an UnsortedSegmentSum operation to the graph
<T> Unstack<T>	unstack(Operand<T> value, Long num, Unstack.Options... options) Adds an Unstack operation to the graph
Unstage	unstage(List<Class<?>> dtypes, Unstage.Options... options) Adds an Unstage operation to the graph
<T> VarHandleOp	varHandleOp(Class<T> dtype, Shape shape, VarHandleOp.Options... options) Adds an VarHandleOp operation to the graph
<T> Variable<T>	variable(Shape shape, Class<T> dtype, Variable.Options... options) Adds an Variable operation to the graph
VariableShape<Integer>	variableShape(Operand<?> input) Adds an VariableShape operation to the graph
<T extends Number> VariableShape<T>	variableShape(Operand<?> input, Class<T> outType) Adds an VariableShape operation to the graph
VarIsInitializedOp	varIsInitializedOp(Operand<?> resource) Adds an VarIsInitializedOp operation to the graph
<T> Where	where(Operand<T> condition) Adds an Where operation to the graph
<T> Where3<T>	where3(Operand<Boolean> condition, Operand<T> x, Operand<T> y) Adds an Where3 operation to the graph
WholeFileReader	wholeFileReader(WholeFileReader.Options... options) Adds an WholeFileReader operation to the graph
Ops	withName(String opName) Returns an API that uses the provided name for an op.
Ops	withSubScope(String childScopeName) Returns an API that adds operations to the graph with the provided name prefix.
WriteFile	writeFile(Operand<String> filename, Operand<String> contents) Adds an WriteFile operation to the graph
<T> Xdivy<T>	xdivy(Operand<T> x, Operand<T> y) Adds an Xdivy operation to the graph
<T> Xlogy<T>	xlogy(Operand<T> x, Operand<T> y) Adds an Xlogy operation to the graph
<T,U extends Number> Zeros<T>	zeros(Operand<U> dims, Class<T> type) Adds an Zeros operation to the graph
<T> ZerosLike<T>	zerosLike(Operand<T> x) Adds an ZerosLike operation to the graph
<T extends Number> Zeta<T>	zeta(Operand<T> x, Operand<T> q) Adds an Zeta operation to the graph

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

gatherV2

public <T,U extends Number,V extends Number> GatherV2<T> gatherV2(Operand<T> params,
                                                                  Operand<U> indices,
                                                                  Operand<V> axis)

Adds an GatherV2 operation to the graph

Parameters:

params - The tensor from which to gather values. Must be at least rank

indices - Index tensor. Must be in range `[0, params.shape[axis])`.

axis - The axis in `params` to gather `indices` from. Defaults to the first

Returns:

a new instance of GatherV2

See Also:

GatherV2

sparseSegmentSumWithNumSegments

public <T extends Number,U extends Number,V extends Number> SparseSegmentSumWithNumSegments<T> sparseSegmentSumWithNumSegments(Operand<T> data,
                                                                                                                               Operand<U> indices,
                                                                                                                               Operand<Integer> segmentIds,
                                                                                                                               Operand<V> numSegments)

Adds an SparseSegmentSumWithNumSegments operation to the graph

Parameters:

data -

indices - A 1-D tensor. Has same rank as `segment_ids`.

segmentIds - A 1-D tensor. Values should be sorted and can be repeated.

numSegments - Should equal the number of distinct segment IDs.

Returns:

a new instance of SparseSegmentSumWithNumSegments

See Also:

SparseSegmentSumWithNumSegments

assign

public <T> Assign<T> assign(Operand<T> ref,
                            Operand<T> value,
                            Assign.Options... options)

Adds an Assign operation to the graph

Parameters:

ref - Should be from a `Variable` node. May be uninitialized.

value - The value to be assigned to the variable.

options - carries optional attributes values

Returns:

a new instance of Assign

See Also:

Assign

isFinite

public <T extends Number> IsFinite isFinite(Operand<T> x)

Adds an IsFinite operation to the graph

Parameters:

x -

Returns:

a new instance of IsFinite

See Also:

IsFinite

conv2DBackpropInput

public <T extends Number> Conv2DBackpropInput<T> conv2DBackpropInput(Operand<Integer> inputSizes,
                                                                     Operand<T> filter,
                                                                     Operand<T> outBackprop,
                                                                     List<Long> strides,
                                                                     String padding,
                                                                     Conv2DBackpropInput.Options... options)

Adds an Conv2DBackpropInput operation to the graph

Parameters:

inputSizes - An integer vector representing the shape of `input`,

filter - 4-D with shape

outBackprop - 4-D with shape `[batch, out_height, out_width, out_channels]`.

strides - The stride of the sliding window for each dimension of the input

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of Conv2DBackpropInput

See Also:

Conv2DBackpropInput

split

public <T> Split<T> split(Operand<Integer> axis,
                          Operand<T> value,
                          Long numSplit)

Adds an Split operation to the graph

Parameters:

axis - 0-D. The dimension along which to split. Must be in the range

value - The tensor to split.

numSplit - The number of ways to split. Must evenly divide

Returns:

a new instance of Split

See Also:

Split

cTCLoss

public CTCLoss cTCLoss(Operand<Float> inputs,
                       Operand<Long> labelsIndices,
                       Operand<Integer> labelsValues,
                       Operand<Integer> sequenceLength,
                       CTCLoss.Options... options)

Adds an CTCLoss operation to the graph

Parameters:

inputs - 3-D, shape: `(max_time x batch_size x num_classes)`, the logits.

labelsIndices - The indices of a `SparseTensor`.

labelsValues - The values (labels) associated with the given batch and time.

sequenceLength - A vector containing sequence lengths (batch).

options - carries optional attributes values

Returns:

a new instance of CTCLoss

See Also:

CTCLoss

sigmoid

public <T> Sigmoid<T> sigmoid(Operand<T> x)

Adds an Sigmoid operation to the graph

Parameters:

x -

Returns:

a new instance of Sigmoid

See Also:

Sigmoid

resourceStridedSliceAssign

public <T extends Number,U> ResourceStridedSliceAssign resourceStridedSliceAssign(Operand<?> ref,
                                                                                  Operand<T> begin,
                                                                                  Operand<T> end,
                                                                                  Operand<T> strides,
                                                                                  Operand<U> value,
                                                                                  ResourceStridedSliceAssign.Options... options)

Adds an ResourceStridedSliceAssign operation to the graph

Parameters:

ref -

begin -

end -

strides -

value -

options - carries optional attributes values

Returns:

a new instance of ResourceStridedSliceAssign

See Also:

ResourceStridedSliceAssign

batchCholeskyGrad

public <T extends Number> BatchCholeskyGrad<T> batchCholeskyGrad(Operand<T> l,
                                                                 Operand<T> grad)

Adds an BatchCholeskyGrad operation to the graph

Parameters:

l -

grad -

Returns:

a new instance of BatchCholeskyGrad

See Also:

BatchCholeskyGrad

priorityQueue

public PriorityQueue priorityQueue(List<Class<?>> componentTypes,
                                   List<Shape> shapes,
                                   PriorityQueue.Options... options)

Adds an PriorityQueue operation to the graph

Parameters:

componentTypes - The type of each component in a value.

shapes - The shape of each component in a value. The length of this attr must

options - carries optional attributes values

Returns:

a new instance of PriorityQueue

See Also:

PriorityQueue

pad

public <T,U extends Number> Pad<T> pad(Operand<T> input,
                                       Operand<U> paddings)

Adds an Pad operation to the graph

Parameters:

input -

paddings -

Returns:

a new instance of Pad

See Also:

Pad

constant

public Constant<Float> constant(float[][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

floorDiv

public <T> FloorDiv<T> floorDiv(Operand<T> x,
                                Operand<T> y)

Adds an FloorDiv operation to the graph

Parameters:

x -

y -

Returns:

a new instance of FloorDiv

See Also:

FloorDiv

quantizedReluX

public <U,T> QuantizedReluX<U> quantizedReluX(Operand<T> features,
                                              Operand<Float> maxValue,
                                              Operand<Float> minFeatures,
                                              Operand<Float> maxFeatures,
                                              Class<U> outType)

Adds an QuantizedReluX operation to the graph

Parameters:

features -

maxValue -

minFeatures - The float value that the lowest quantized value represents.

maxFeatures - The float value that the highest quantized value represents.

outType -

Returns:

a new instance of QuantizedReluX

See Also:

QuantizedReluX

unravelIndex

public <T extends Number> UnravelIndex<T> unravelIndex(Operand<T> indices,
                                                       Operand<T> dims)

Adds an UnravelIndex operation to the graph

Parameters:

indices - An 0-D or 1-D `int` Tensor whose elements are indices into the

dims - An 1-D `int` Tensor. The shape of the array to use for unraveling

Returns:

a new instance of UnravelIndex

See Also:

UnravelIndex

resourceScatterAdd

public <T extends Number,U> ResourceScatterAdd resourceScatterAdd(Operand<?> resource,
                                                                  Operand<T> indices,
                                                                  Operand<U> updates)

Adds an ResourceScatterAdd operation to the graph

Parameters:

resource - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to add to `ref`.

Returns:

a new instance of ResourceScatterAdd

See Also:

ResourceScatterAdd

restoreSlice

public <T> RestoreSlice<T> restoreSlice(Operand<String> filePattern,
                                        Operand<String> tensorName,
                                        Operand<String> shapeAndSlice,
                                        Class<T> dt,
                                        RestoreSlice.Options... options)

Adds an RestoreSlice operation to the graph

Parameters:

filePattern - Must have a single element. The pattern of the files from

tensorName - Must have a single element. The name of the tensor to be

shapeAndSlice - Scalar. The shapes and slice specifications to use when

dt - The type of the tensor to be restored.

options - carries optional attributes values

Returns:

a new instance of RestoreSlice

See Also:

RestoreSlice

accumulateNV2

public <T> AccumulateNV2<T> accumulateNV2(Operand<T> inputs,
                                          Shape shape)

Adds an AccumulateNV2 operation to the graph

Parameters:

inputs - A list of `Tensor` objects, each with same shape and type.

shape - Shape of elements of `inputs`.

Returns:

a new instance of AccumulateNV2

See Also:

AccumulateNV2

scatterNdAdd

public <T,U extends Number> ScatterNdAdd<T> scatterNdAdd(Operand<T> ref,
                                                         Operand<U> indices,
                                                         Operand<T> updates,
                                                         ScatterNdAdd.Options... options)

Adds an ScatterNdAdd operation to the graph

Parameters:

ref - A mutable Tensor. Should be from a Variable node.

indices - A Tensor. Must be one of the following types: int32, int64.

updates - A Tensor. Must have the same type as ref. A tensor of updated values

options - carries optional attributes values

Returns:

a new instance of ScatterNdAdd

See Also:

ScatterNdAdd

constant

public Constant<Integer> constant(int[][][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

addN

public <T> AddN<T> addN(Operand<T> inputs)

Adds an AddN operation to the graph

Parameters:

inputs - Must all be the same size and shape.

Returns:

a new instance of AddN

See Also:

AddN

sum

public <T,U extends Number> Sum<T> sum(Operand<T> input,
                                       Operand<U> axis,
                                       Sum.Options... options)

Adds an Sum operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of Sum

See Also:

Sum

placeholderWithDefault

public <T> PlaceholderWithDefault<T> placeholderWithDefault(Operand<T> input,
                                                            Shape shape)

Adds an PlaceholderWithDefault operation to the graph

Parameters:

input - The default value to produce when `output` is not fed.

shape - The (possibly partial) shape of the tensor.

Returns:

a new instance of PlaceholderWithDefault

See Also:

PlaceholderWithDefault

abort

public Abort abort(Abort.Options... options)

Adds an Abort operation to the graph

Parameters:

options - carries optional attributes values

Returns:

a new instance of Abort

See Also:

Abort

avgPool

public <T extends Number> AvgPool<T> avgPool(Operand<T> value,
                                             List<Long> ksize,
                                             List<Long> strides,
                                             String padding,
                                             AvgPool.Options... options)

Adds an AvgPool operation to the graph

Parameters:

value - 4-D with shape `[batch, height, width, channels]`.

ksize - The size of the sliding window for each dimension of `value`.

strides - The stride of the sliding window for each dimension of `value`.

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of AvgPool

See Also:

AvgPool

batchIFFT

public BatchIFFT batchIFFT(Operand<?> input)

Adds an BatchIFFT operation to the graph

Parameters:

input -

Returns:

a new instance of BatchIFFT

See Also:

BatchIFFT

nthElement

public <T extends Number> NthElement<T> nthElement(Operand<T> input,
                                                   Operand<Integer> n,
                                                   NthElement.Options... options)

Adds an NthElement operation to the graph

Parameters:

input - 1-D or higher with last dimension at least `n+1`.

n - 0-D. Position of sorted vector to select along the last dimension (along

options - carries optional attributes values

Returns:

a new instance of NthElement

See Also:

NthElement

encodeJpeg

public EncodeJpeg encodeJpeg(Operand<UInt8> image,
                             EncodeJpeg.Options... options)

Adds an EncodeJpeg operation to the graph

Parameters:

image - 3-D with shape `[height, width, channels]`.

options - carries optional attributes values

Returns:

a new instance of EncodeJpeg

See Also:

EncodeJpeg

segmentSum

public <T,U extends Number> SegmentSum<T> segmentSum(Operand<T> data,
                                                     Operand<U> segmentIds)

Adds an SegmentSum operation to the graph

Parameters:

data -

segmentIds - A 1-D tensor whose size is equal to the size of `data`'s

Returns:

a new instance of SegmentSum

See Also:

SegmentSum

maxPoolV2

public <T> MaxPoolV2<T> maxPoolV2(Operand<T> input,
                                  Operand<Integer> ksize,
                                  Operand<Integer> strides,
                                  String padding,
                                  MaxPoolV2.Options... options)

Adds an MaxPoolV2 operation to the graph

Parameters:

input - 4-D input to pool over.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of MaxPoolV2

See Also:

MaxPoolV2

matMul

public <T> MatMul<T> matMul(Operand<T> a,
                            Operand<T> b,
                            MatMul.Options... options)

Adds an MatMul operation to the graph

Parameters:

a -

b -

options - carries optional attributes values

Returns:

a new instance of MatMul

See Also:

MatMul

acos

public <T> Acos<T> acos(Operand<T> x)

Adds an Acos operation to the graph

Parameters:

x -

Returns:

a new instance of Acos

See Also:

Acos

tensorListScatter

public <T,U extends Number> TensorListScatter tensorListScatter(Operand<T> tensor,
                                                                Operand<Integer> indices,
                                                                Operand<U> elementShape)

Adds an TensorListScatter operation to the graph

Parameters:

tensor -

indices -

elementShape -

Returns:

a new instance of TensorListScatter

See Also:

TensorListScatter

batchFFT2D

public BatchFFT2D batchFFT2D(Operand<?> input)

Adds an BatchFFT2D operation to the graph

Parameters:

input -

Returns:

a new instance of BatchFFT2D

See Also:

BatchFFT2D

requantize

public <U,T> Requantize<U> requantize(Operand<T> input,
                                      Operand<Float> inputMin,
                                      Operand<Float> inputMax,
                                      Operand<Float> requestedOutputMin,
                                      Operand<Float> requestedOutputMax,
                                      Class<U> outType)

Adds an Requantize operation to the graph

Parameters:

input -

inputMin - The float value that the minimum quantized input value represents.

inputMax - The float value that the maximum quantized input value represents.

requestedOutputMin - The float value that the minimum quantized output value represents.

requestedOutputMax - The float value that the maximum quantized output value represents.

outType - The type of the output. Should be a lower bit depth than Tinput.

Returns:

a new instance of Requantize

See Also:

Requantize

all

public <T extends Number> All all(Operand<Boolean> input,
                                  Operand<T> axis,
                                  All.Options... options)

Adds an All operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of All

See Also:

All

stridedSliceAssign

public <T,U extends Number> StridedSliceAssign<T> stridedSliceAssign(Operand<T> ref,
                                                                     Operand<U> begin,
                                                                     Operand<U> end,
                                                                     Operand<U> strides,
                                                                     Operand<T> value,
                                                                     StridedSliceAssign.Options... options)

Adds an StridedSliceAssign operation to the graph

Parameters:

ref -

begin -

end -

strides -

value -

options - carries optional attributes values

Returns:

a new instance of StridedSliceAssign

See Also:

StridedSliceAssign

orderedMapUnstage

public OrderedMapUnstage orderedMapUnstage(Operand<Long> key,
                                           Operand<Integer> indices,
                                           List<Class<?>> dtypes,
                                           OrderedMapUnstage.Options... options)

Adds an OrderedMapUnstage operation to the graph

Parameters:

key -

indices -

dtypes -

options - carries optional attributes values

Returns:

a new instance of OrderedMapUnstage

See Also:

OrderedMapUnstage

sparseReshape

public SparseReshape sparseReshape(Operand<Long> inputIndices,
                                   Operand<Long> inputShape,
                                   Operand<Long> newShape)

Adds an SparseReshape operation to the graph

Parameters:

inputIndices - 2-D. `N x R_in` matrix with the indices of non-empty values in a

inputShape - 1-D. `R_in` vector with the input SparseTensor's dense shape.

newShape - 1-D. `R_out` vector with the requested new dense shape.

Returns:

a new instance of SparseReshape

See Also:

SparseReshape

assignAddVariableOp

public <T> AssignAddVariableOp assignAddVariableOp(Operand<?> resource,
                                                   Operand<T> value)

Adds an AssignAddVariableOp operation to the graph

Parameters:

resource - handle to the resource in which to store the variable.

value - the value by which the variable will be incremented.

Returns:

a new instance of AssignAddVariableOp

See Also:

AssignAddVariableOp

dataFormatDimMap

public <T extends Number> DataFormatDimMap<T> dataFormatDimMap(Operand<T> x,
                                                               DataFormatDimMap.Options... options)

Adds an DataFormatDimMap operation to the graph

Parameters:

x - A Tensor with each element as a dimension index in source data format.

options - carries optional attributes values

Returns:

a new instance of DataFormatDimMap

See Also:

DataFormatDimMap

scatterNdSub

public <T,U extends Number> ScatterNdSub<T> scatterNdSub(Operand<T> ref,
                                                         Operand<U> indices,
                                                         Operand<T> updates,
                                                         ScatterNdSub.Options... options)

Adds an ScatterNdSub operation to the graph

Parameters:

ref - A mutable Tensor. Should be from a Variable node.

indices - A Tensor. Must be one of the following types: int32, int64.

updates - A Tensor. Must have the same type as ref. A tensor of updated values

options - carries optional attributes values

Returns:

a new instance of ScatterNdSub

See Also:

ScatterNdSub

stringToHashBucketStrong

public StringToHashBucketStrong stringToHashBucketStrong(Operand<String> input,
                                                         Long numBuckets,
                                                         List<Long> key)

Adds an StringToHashBucketStrong operation to the graph

Parameters:

input - The strings to assign a hash bucket.

numBuckets - The number of buckets.

key - The key for the keyed hash function passed as a list of two uint64

Returns:

a new instance of StringToHashBucketStrong

See Also:

StringToHashBucketStrong

polygamma

public <T extends Number> Polygamma<T> polygamma(Operand<T> a,
                                                 Operand<T> x)

Adds an Polygamma operation to the graph

Parameters:

a -

x -

Returns:

a new instance of Polygamma

See Also:

Polygamma

bigQueryReader

public BigQueryReader bigQueryReader(String projectId,
                                     String datasetId,
                                     String tableId,
                                     List<String> columns,
                                     Long timestampMillis,
                                     BigQueryReader.Options... options)

Adds an BigQueryReader operation to the graph

Parameters:

projectId - GCP project ID.

datasetId - BigQuery Dataset ID.

tableId - Table to read.

columns - List of columns to read. Leave empty to read all columns.

timestampMillis - Table snapshot timestamp in millis since epoch. Relative

options - carries optional attributes values

Returns:

a new instance of BigQueryReader

See Also:

BigQueryReader

sparseApplyFtrl

public <T,U extends Number> SparseApplyFtrl<T> sparseApplyFtrl(Operand<T> var,
                                                               Operand<T> accum,
                                                               Operand<T> linear,
                                                               Operand<T> grad,
                                                               Operand<U> indices,
                                                               Operand<T> lr,
                                                               Operand<T> l1,
                                                               Operand<T> l2,
                                                               Operand<T> lrPower,
                                                               SparseApplyFtrl.Options... options)

Adds an SparseApplyFtrl operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

linear - Should be from a Variable().

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

lr - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

lrPower - Scaling factor. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of SparseApplyFtrl

See Also:

SparseApplyFtrl

sparseSegmentSqrtN

public <T extends Number,U extends Number> SparseSegmentSqrtN<T> sparseSegmentSqrtN(Operand<T> data,
                                                                                    Operand<U> indices,
                                                                                    Operand<Integer> segmentIds)

Adds an SparseSegmentSqrtN operation to the graph

Parameters:

data -

indices - A 1-D tensor. Has same rank as `segment_ids`.

segmentIds - A 1-D tensor. Values should be sorted and can be repeated.

Returns:

a new instance of SparseSegmentSqrtN

See Also:

SparseSegmentSqrtN

assignVariableOp

public <T> AssignVariableOp assignVariableOp(Operand<?> resource,
                                             Operand<T> value)

Adds an AssignVariableOp operation to the graph

Parameters:

resource - handle to the resource in which to store the variable.

value - the value to set the new tensor to use.

Returns:

a new instance of AssignVariableOp

See Also:

AssignVariableOp

preventGradient

public <T> PreventGradient<T> preventGradient(Operand<T> input,
                                              PreventGradient.Options... options)

Adds an PreventGradient operation to the graph

Parameters:

input - any tensor.

options - carries optional attributes values

Returns:

a new instance of PreventGradient

See Also:

PreventGradient

fakeQuantWithMinMaxVars

public FakeQuantWithMinMaxVars fakeQuantWithMinMaxVars(Operand<Float> inputs,
                                                       Operand<Float> min,
                                                       Operand<Float> max,
                                                       FakeQuantWithMinMaxVars.Options... options)

Adds an FakeQuantWithMinMaxVars operation to the graph

Parameters:

inputs -

min -

max -

options - carries optional attributes values

Returns:

a new instance of FakeQuantWithMinMaxVars

See Also:

FakeQuantWithMinMaxVars

tensorListPushBackBatch

public <T> TensorListPushBackBatch tensorListPushBackBatch(Operand<?> inputHandles,
                                                           Operand<T> tensor)

Adds an TensorListPushBackBatch operation to the graph

Parameters:

inputHandles -

tensor -

Returns:

a new instance of TensorListPushBackBatch

See Also:

TensorListPushBackBatch

besselI0e

public <T extends Number> BesselI0e<T> besselI0e(Operand<T> x)

Adds an BesselI0e operation to the graph

Parameters:

x -

Returns:

a new instance of BesselI0e

See Also:

BesselI0e

rpc

public Rpc rpc(Operand<String> address,
               Operand<String> method,
               Operand<String> request,
               Rpc.Options... options)

Adds an Rpc operation to the graph

Parameters:

address - `0-D` or `1-D`. The address (i.e. host_name:port) of the RPC server.

method - `0-D` or `1-D`. The method address on the RPC server.

request - `0-D` or `1-D`. Serialized proto strings: the rpc request argument.

options - carries optional attributes values

Returns:

a new instance of Rpc

See Also:

Rpc

resizeArea

public <T extends Number> ResizeArea resizeArea(Operand<T> images,
                                                Operand<Integer> size,
                                                ResizeArea.Options... options)

Adds an ResizeArea operation to the graph

Parameters:

images - 4-D with shape `[batch, height, width, channels]`.

size - = A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The

options - carries optional attributes values

Returns:

a new instance of ResizeArea

See Also:

ResizeArea

constant

public Constant<Integer> constant(int[][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

asin

public <T> Asin<T> asin(Operand<T> x)

Adds an Asin operation to the graph

Parameters:

x -

Returns:

a new instance of Asin

See Also:

Asin

sparseSoftmax

public <T extends Number> SparseSoftmax<T> sparseSoftmax(Operand<Long> spIndices,
                                                         Operand<T> spValues,
                                                         Operand<Long> spShape)

Adds an SparseSoftmax operation to the graph

Parameters:

spIndices - 2-D. `NNZ x R` matrix with the indices of non-empty values in a

spValues - 1-D. `NNZ` non-empty values corresponding to `sp_indices`.

spShape - 1-D. Shape of the input SparseTensor.

Returns:

a new instance of SparseSoftmax

See Also:

SparseSoftmax

constant

public Constant<Boolean> constant(boolean data)

Adds an Constant operation to the graph

Parameters:

data - The value to put into the new constant.

Returns:

a boolean constant

See Also:

Constant

variableShape

public VariableShape<Integer> variableShape(Operand<?> input)

Adds an VariableShape operation to the graph

Parameters:

input -

Returns:

a new instance of VariableShape

See Also:

VariableShape

tensorArrayGradWithShape

public TensorArrayGradWithShape tensorArrayGradWithShape(Operand<?> handle,
                                                         Operand<Float> flowIn,
                                                         Operand<Integer> shapeToPrepend,
                                                         String source)

Adds an TensorArrayGradWithShape operation to the graph

Parameters:

handle - The handle to the forward TensorArray.

flowIn - A float scalar that enforces proper chaining of operations.

shapeToPrepend - An int32 vector representing a shape. Elements in the gradient accumulator will

source - The gradient source string, used to decide which gradient TensorArray

Returns:

a new instance of TensorArrayGradWithShape

See Also:

TensorArrayGradWithShape

applyAdagradDA

public <T> ApplyAdagradDA<T> applyAdagradDA(Operand<T> var,
                                            Operand<T> gradientAccumulator,
                                            Operand<T> gradientSquaredAccumulator,
                                            Operand<T> grad,
                                            Operand<T> lr,
                                            Operand<T> l1,
                                            Operand<T> l2,
                                            Operand<Long> globalStep,
                                            ApplyAdagradDA.Options... options)

Adds an ApplyAdagradDA operation to the graph

Parameters:

var - Should be from a Variable().

gradientAccumulator - Should be from a Variable().

gradientSquaredAccumulator - Should be from a Variable().

grad - The gradient.

lr - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

globalStep - Training step number. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ApplyAdagradDA

See Also:

ApplyAdagradDA

angle

public <U extends Number,T> Angle<U> angle(Operand<T> input,
                                           Class<U> Tout)

Adds an Angle operation to the graph

Parameters:

input -

Tout -

Returns:

a new instance of Angle

See Also:

Angle

betainc

public <T extends Number> Betainc<T> betainc(Operand<T> a,
                                             Operand<T> b,
                                             Operand<T> x)

Adds an Betainc operation to the graph

Parameters:

a -

b -

x -

Returns:

a new instance of Betainc

See Also:

Betainc

mutableHashTable

public <T,U> MutableHashTable mutableHashTable(Class<T> keyDtype,
                                               Class<U> valueDtype,
                                               MutableHashTable.Options... options)

Adds an MutableHashTable operation to the graph

Parameters:

keyDtype - Type of the table keys.

valueDtype - Type of the table values.

options - carries optional attributes values

Returns:

a new instance of MutableHashTable

See Also:

MutableHashTable

iFFT3D

public <T> IFFT3D<T> iFFT3D(Operand<T> input)

Adds an IFFT3D operation to the graph

Parameters:

input - A complex64 tensor.

Returns:

a new instance of IFFT3D

See Also:

IFFT3D

erfc

public <T extends Number> Erfc<T> erfc(Operand<T> x)

Adds an Erfc operation to the graph

Parameters:

x -

Returns:

a new instance of Erfc

See Also:

Erfc

applyRMSProp

public <T> ApplyRMSProp<T> applyRMSProp(Operand<T> var,
                                        Operand<T> ms,
                                        Operand<T> mom,
                                        Operand<T> lr,
                                        Operand<T> rho,
                                        Operand<T> momentum,
                                        Operand<T> epsilon,
                                        Operand<T> grad,
                                        ApplyRMSProp.Options... options)

Adds an ApplyRMSProp operation to the graph

Parameters:

var - Should be from a Variable().

ms - Should be from a Variable().

mom - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay rate. Must be a scalar.

momentum -

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ApplyRMSProp

See Also:

ApplyRMSProp

constant

public Constant<Long> constant(long[][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

applyGradientDescent

public <T> ApplyGradientDescent<T> applyGradientDescent(Operand<T> var,
                                                        Operand<T> alpha,
                                                        Operand<T> delta,
                                                        ApplyGradientDescent.Options... options)

Adds an ApplyGradientDescent operation to the graph

Parameters:

var - Should be from a Variable().

alpha - Scaling factor. Must be a scalar.

delta - The change.

options - carries optional attributes values

Returns:

a new instance of ApplyGradientDescent

See Also:

ApplyGradientDescent

atanh

public <T> Atanh<T> atanh(Operand<T> x)

Adds an Atanh operation to the graph

Parameters:

x -

Returns:

a new instance of Atanh

See Also:

Atanh

spaceToBatchND

public <T,U extends Number,V extends Number> SpaceToBatchND<T> spaceToBatchND(Operand<T> input,
                                                                              Operand<U> blockShape,
                                                                              Operand<V> paddings)

Adds an SpaceToBatchND operation to the graph

Parameters:

input - N-D with shape `input_shape = [batch] + spatial_shape + remaining_shape`,

blockShape - 1-D with shape `[M]`, all values must be >= 1.

paddings - 2-D with shape `[M, 2]`, all values must be >= 0.

Returns:

a new instance of SpaceToBatchND

See Also:

SpaceToBatchND

tanh

public <T> Tanh<T> tanh(Operand<T> x)

Adds an Tanh operation to the graph

Parameters:

x -

Returns:

a new instance of Tanh

See Also:

Tanh

tensorArrayUnpack

public <T> TensorArrayUnpack tensorArrayUnpack(Operand<String> handle,
                                               Operand<T> value,
                                               Operand<Float> flowIn)

Adds an TensorArrayUnpack operation to the graph

Parameters:

handle -

value -

flowIn -

Returns:

a new instance of TensorArrayUnpack

See Also:

TensorArrayUnpack

barrierInsertMany

public <T> BarrierInsertMany barrierInsertMany(Operand<String> handle,
                                               Operand<String> keys,
                                               Operand<T> values,
                                               Long componentIndex)

Adds an BarrierInsertMany operation to the graph

Parameters:

handle - The handle to a barrier.

keys - A one-dimensional tensor of keys, with length n.

values - An any-dimensional tensor of values, which are associated with the

componentIndex - The component of the barrier elements that is being assigned.

Returns:

a new instance of BarrierInsertMany

See Also:

BarrierInsertMany

bitwiseXor

public <T extends Number> BitwiseXor<T> bitwiseXor(Operand<T> x,
                                                   Operand<T> y)

Adds an BitwiseXor operation to the graph

Parameters:

x -

y -

Returns:

a new instance of BitwiseXor

See Also:

BitwiseXor

scalarSummary

public <T extends Number> ScalarSummary scalarSummary(Operand<String> tags,
                                                      Operand<T> values)

Adds an ScalarSummary operation to the graph

Parameters:

tags - Tags for the summary.

values - Same shape as `tags. Values for the summary.

Returns:

a new instance of ScalarSummary

See Also:

ScalarSummary

applyCenteredRMSProp

public <T> ApplyCenteredRMSProp<T> applyCenteredRMSProp(Operand<T> var,
                                                        Operand<T> mg,
                                                        Operand<T> ms,
                                                        Operand<T> mom,
                                                        Operand<T> lr,
                                                        Operand<T> rho,
                                                        Operand<T> momentum,
                                                        Operand<T> epsilon,
                                                        Operand<T> grad,
                                                        ApplyCenteredRMSProp.Options... options)

Adds an ApplyCenteredRMSProp operation to the graph

Parameters:

var - Should be from a Variable().

mg - Should be from a Variable().

ms - Should be from a Variable().

mom - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay rate. Must be a scalar.

momentum -

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ApplyCenteredRMSProp

See Also:

ApplyCenteredRMSProp

constant

public <T> Constant<T> constant(Object object,
                                Class<T> type)

Adds an Constant operation to the graph

Parameters:

object - a Java object representing the constant.

Returns:

a constant of type `type`

See Also:

Tensor.create, Constant

maxPoolWithArgmax

public <T extends Number> MaxPoolWithArgmax<T,Long> maxPoolWithArgmax(Operand<T> input,
                                                                      List<Long> ksize,
                                                                      List<Long> strides,
                                                                      String padding)

Adds an MaxPoolWithArgmax operation to the graph

Parameters:

input - 4-D with shape `[batch, height, width, channels]`. Input to pool over.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the

padding - The type of padding algorithm to use.

Returns:

a new instance of MaxPoolWithArgmax

See Also:

MaxPoolWithArgmax

sparseToDense

public <U,T extends Number> SparseToDense<U> sparseToDense(Operand<T> sparseIndices,
                                                           Operand<T> outputShape,
                                                           Operand<U> sparseValues,
                                                           Operand<U> defaultValue,
                                                           SparseToDense.Options... options)

Adds an SparseToDense operation to the graph

Parameters:

sparseIndices - 0-D, 1-D, or 2-D. `sparse_indices[i]` contains the complete

outputShape - 1-D. Shape of the dense output tensor.

sparseValues - 1-D. Values corresponding to each row of `sparse_indices`,

defaultValue - Scalar value to set for indices not specified in

options - carries optional attributes values

Returns:

a new instance of SparseToDense

See Also:

SparseToDense

stringStrip

public StringStrip stringStrip(Operand<String> input)

Adds an StringStrip operation to the graph

Parameters:

input - A string `Tensor` of any shape.

Returns:

a new instance of StringStrip

See Also:

StringStrip

deserializeSparse

public <U,T> DeserializeSparse<U> deserializeSparse(Operand<T> serializedSparse,
                                                    Class<U> dtype)

Adds an DeserializeSparse operation to the graph

Parameters:

serializedSparse - The serialized `SparseTensor` objects. The last dimension

dtype - The `dtype` of the serialized `SparseTensor` objects.

Returns:

a new instance of DeserializeSparse

See Also:

DeserializeSparse

sparseConditionalAccumulator

public <T> SparseConditionalAccumulator sparseConditionalAccumulator(Class<T> dtype,
                                                                     Shape shape,
                                                                     SparseConditionalAccumulator.Options... options)

Adds an SparseConditionalAccumulator operation to the graph

Parameters:

dtype - The type of the value being accumulated.

shape - The shape of the values.

options - carries optional attributes values

Returns:

a new instance of SparseConditionalAccumulator

See Also:

SparseConditionalAccumulator

textLineReader

public TextLineReader textLineReader(TextLineReader.Options... options)

Adds an TextLineReader operation to the graph

Parameters:

options - carries optional attributes values

Returns:

a new instance of TextLineReader

See Also:

TextLineReader

unique

public <T> Unique<T,Integer> unique(Operand<T> x)

Adds an Unique operation to the graph

Parameters:

x - 1-D.

Returns:

a new instance of Unique

See Also:

Unique

divNoNan

public <T extends Number> DivNoNan<T> divNoNan(Operand<T> x,
                                               Operand<T> y)

Adds an DivNoNan operation to the graph

Parameters:

x -

y -

Returns:

a new instance of DivNoNan

See Also:

DivNoNan

diag

public <T> Diag<T> diag(Operand<T> diagonal)

Adds an Diag operation to the graph

Parameters:

diagonal - Rank k tensor where k is at most 1.

Returns:

a new instance of Diag

See Also:

Diag

resourceScatterMax

public <T extends Number,U> ResourceScatterMax resourceScatterMax(Operand<?> resource,
                                                                  Operand<T> indices,
                                                                  Operand<U> updates)

Adds an ResourceScatterMax operation to the graph

Parameters:

resource - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to add to `ref`.

Returns:

a new instance of ResourceScatterMax

See Also:

ResourceScatterMax

lMDBReader

public LMDBReader lMDBReader(LMDBReader.Options... options)

Adds an LMDBReader operation to the graph

Parameters:

options - carries optional attributes values

Returns:

a new instance of LMDBReader

See Also:

LMDBReader

decodeCompressed

public DecodeCompressed decodeCompressed(Operand<String> bytes,
                                         DecodeCompressed.Options... options)

Adds an DecodeCompressed operation to the graph

Parameters:

bytes - A Tensor of string which is compressed.

options - carries optional attributes values

Returns:

a new instance of DecodeCompressed

See Also:

DecodeCompressed

queueDequeueUpTo

public QueueDequeueUpTo queueDequeueUpTo(Operand<?> handle,
                                         Operand<Integer> n,
                                         List<Class<?>> componentTypes,
                                         QueueDequeueUpTo.Options... options)

Adds an QueueDequeueUpTo operation to the graph

Parameters:

handle - The handle to a queue.

n - The number of tuples to dequeue.

componentTypes - The type of each component in a tuple.

options - carries optional attributes values

Returns:

a new instance of QueueDequeueUpTo

See Also:

QueueDequeueUpTo

topK

public <T extends Number> TopK<T> topK(Operand<T> input,
                                       Operand<Integer> k,
                                       TopK.Options... options)

Adds an TopK operation to the graph

Parameters:

input - 1-D or higher with last dimension at least `k`.

k - 0-D. Number of top elements to look for along the last dimension (along each

options - carries optional attributes values

Returns:

a new instance of TopK

See Also:

TopK

resourceApplyProximalGradientDescent

public <T> ResourceApplyProximalGradientDescent resourceApplyProximalGradientDescent(Operand<?> var,
                                                                                     Operand<T> alpha,
                                                                                     Operand<T> l1,
                                                                                     Operand<T> l2,
                                                                                     Operand<T> delta,
                                                                                     ResourceApplyProximalGradientDescent.Options... options)

Adds an ResourceApplyProximalGradientDescent operation to the graph

Parameters:

var - Should be from a Variable().

alpha - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

delta - The change.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyProximalGradientDescent

See Also:

ResourceApplyProximalGradientDescent

tFRecordReader

public TFRecordReader tFRecordReader(TFRecordReader.Options... options)

Adds an TFRecordReader operation to the graph

Parameters:

options - carries optional attributes values

Returns:

a new instance of TFRecordReader

See Also:

TFRecordReader

stage

public Stage stage(Iterable<Operand<?>> values,
                   Stage.Options... options)

Adds an Stage operation to the graph

Parameters:

values - a list of tensors

options - carries optional attributes values

Returns:

a new instance of Stage

See Also:

Stage

tensorArrayClose

public TensorArrayClose tensorArrayClose(Operand<?> handle)

Adds an TensorArrayClose operation to the graph

Parameters:

handle - The handle to a TensorArray (output of TensorArray or TensorArrayGrad).

Returns:

a new instance of TensorArrayClose

See Also:

TensorArrayClose

batchSelfAdjointEigV2

public <T extends Number> BatchSelfAdjointEigV2<T> batchSelfAdjointEigV2(Operand<T> input,
                                                                         BatchSelfAdjointEigV2.Options... options)

Adds an BatchSelfAdjointEigV2 operation to the graph

Parameters:

input -

options - carries optional attributes values

Returns:

a new instance of BatchSelfAdjointEigV2

See Also:

BatchSelfAdjointEigV2

histogramFixedWidth

public <T extends Number> HistogramFixedWidth<Integer> histogramFixedWidth(Operand<T> values,
                                                                           Operand<T> valueRange,
                                                                           Operand<Integer> nbins)

Adds an HistogramFixedWidth operation to the graph

Parameters:

values - Numeric `Tensor`.

valueRange - Shape [2] `Tensor` of same `dtype` as `values`.

nbins - Scalar `int32 Tensor`. Number of histogram bins.

Returns:

a new instance of HistogramFixedWidth

See Also:

HistogramFixedWidth

negTrain

public NegTrain negTrain(Operand<Float> wIn,
                         Operand<Float> wOut,
                         Operand<Integer> examples,
                         Operand<Integer> labels,
                         Operand<Float> lr,
                         List<Long> vocabCount,
                         Long numNegativeSamples)

Adds an NegTrain operation to the graph

Parameters:

wIn - input word embedding.

wOut - output word embedding.

examples - A vector of word ids.

labels - A vector of word ids.

lr -

vocabCount - Count of words in the vocabulary.

numNegativeSamples - Number of negative samples per example.

Returns:

a new instance of NegTrain

See Also:

NegTrain

setSize

public <T> SetSize setSize(Operand<Long> setIndices,
                           Operand<T> setValues,
                           Operand<Long> setShape,
                           SetSize.Options... options)

Adds an SetSize operation to the graph

Parameters:

setIndices - 2D `Tensor`, indices of a `SparseTensor`.

setValues - 1D `Tensor`, values of a `SparseTensor`.

setShape - 1D `Tensor`, shape of a `SparseTensor`.

options - carries optional attributes values

Returns:

a new instance of SetSize

See Also:

SetSize

initializeTable

public <T,U> InitializeTable initializeTable(Operand<?> tableHandle,
                                             Operand<T> keys,
                                             Operand<U> values)

Adds an InitializeTable operation to the graph

Parameters:

tableHandle - Handle to a table which will be initialized.

keys - Keys of type Tkey.

values - Values of type Tval.

Returns:

a new instance of InitializeTable

See Also:

InitializeTable

parameterizedTruncatedNormal

public <U extends Number,T extends Number> ParameterizedTruncatedNormal<U> parameterizedTruncatedNormal(Operand<T> shape,
                                                                                                        Operand<U> means,
                                                                                                        Operand<U> stdevs,
                                                                                                        Operand<U> minvals,
                                                                                                        Operand<U> maxvals,
                                                                                                        ParameterizedTruncatedNormal.Options... options)

Adds an ParameterizedTruncatedNormal operation to the graph

Parameters:

shape - The shape of the output tensor. Batches are indexed by the 0th dimension.

means - The mean parameter of each batch.

stdevs - The standard deviation parameter of each batch. Must be greater than 0.

minvals - The minimum cutoff. May be -infinity.

maxvals - The maximum cutoff. May be +infinity, and must be more than the minval

options - carries optional attributes values

Returns:

a new instance of ParameterizedTruncatedNormal

See Also:

ParameterizedTruncatedNormal

scatterMul

public <T,U extends Number> ScatterMul<T> scatterMul(Operand<T> ref,
                                                     Operand<U> indices,
                                                     Operand<T> updates,
                                                     ScatterMul.Options... options)

Adds an ScatterMul operation to the graph

Parameters:

ref - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to multiply to `ref`.

options - carries optional attributes values

Returns:

a new instance of ScatterMul

See Also:

ScatterMul

scatterUpdate

public <T,U extends Number> ScatterUpdate<T> scatterUpdate(Operand<T> ref,
                                                           Operand<U> indices,
                                                           Operand<T> updates,
                                                           ScatterUpdate.Options... options)

Adds an ScatterUpdate operation to the graph

Parameters:

ref - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to store in `ref`.

options - carries optional attributes values

Returns:

a new instance of ScatterUpdate

See Also:

ScatterUpdate

stageSize

public StageSize stageSize(List<Class<?>> dtypes,
                           StageSize.Options... options)

Adds an StageSize operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of StageSize

See Also:

StageSize

stringLength

public StringLength stringLength(Operand<String> input,
                                 StringLength.Options... options)

Adds an StringLength operation to the graph

Parameters:

input - The string for which to compute the length.

options - carries optional attributes values

Returns:

a new instance of StringLength

See Also:

StringLength

quantizedInstanceNorm

public <T> QuantizedInstanceNorm<T> quantizedInstanceNorm(Operand<T> x,
                                                          Operand<Float> xMin,
                                                          Operand<Float> xMax,
                                                          QuantizedInstanceNorm.Options... options)

Adds an QuantizedInstanceNorm operation to the graph

Parameters:

x - A 4D input Tensor.

xMin - The value represented by the lowest quantized input.

xMax - The value represented by the highest quantized input.

options - carries optional attributes values

Returns:

a new instance of QuantizedInstanceNorm

See Also:

QuantizedInstanceNorm

matchingFiles

public MatchingFiles matchingFiles(Operand<String> pattern)

Adds an MatchingFiles operation to the graph

Parameters:

pattern - Shell wildcard pattern(s). Scalar or vector of type string.

Returns:

a new instance of MatchingFiles

See Also:

MatchingFiles

resourceSparseApplyKerasMomentum

public <T,U extends Number> ResourceSparseApplyKerasMomentum resourceSparseApplyKerasMomentum(Operand<?> var,
                                                                                              Operand<?> accum,
                                                                                              Operand<T> lr,
                                                                                              Operand<T> grad,
                                                                                              Operand<U> indices,
                                                                                              Operand<T> momentum,
                                                                                              ResourceSparseApplyKerasMomentum.Options... options)

Adds an ResourceSparseApplyKerasMomentum operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Learning rate. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

momentum - Momentum. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyKerasMomentum

See Also:

ResourceSparseApplyKerasMomentum

queueDequeueMany

public QueueDequeueMany queueDequeueMany(Operand<?> handle,
                                         Operand<Integer> n,
                                         List<Class<?>> componentTypes,
                                         QueueDequeueMany.Options... options)

Adds an QueueDequeueMany operation to the graph

Parameters:

handle - The handle to a queue.

n - The number of tuples to dequeue.

componentTypes - The type of each component in a tuple.

options - carries optional attributes values

Returns:

a new instance of QueueDequeueMany

See Also:

QueueDequeueMany

assignAdd

public <T> AssignAdd<T> assignAdd(Operand<T> ref,
                                  Operand<T> value,
                                  AssignAdd.Options... options)

Adds an AssignAdd operation to the graph

Parameters:

ref - Should be from a `Variable` node.

value - The value to be added to the variable.

options - carries optional attributes values

Returns:

a new instance of AssignAdd

See Also:

AssignAdd

stringToHashBucket

public StringToHashBucket stringToHashBucket(Operand<String> stringTensor,
                                             Long numBuckets)

Adds an StringToHashBucket operation to the graph

Parameters:

stringTensor -

numBuckets - The number of buckets.

Returns:

a new instance of StringToHashBucket

See Also:

StringToHashBucket

refSelect

public <T> RefSelect<T> refSelect(Operand<Integer> index,
                                  Operand<T> inputs)

Adds an RefSelect operation to the graph

Parameters:

index - A scalar that determines the input that gets selected.

inputs - A list of ref tensors, one of which will be forwarded to `output`.

Returns:

a new instance of RefSelect

See Also:

RefSelect

destroyTemporaryVariable

public <T> DestroyTemporaryVariable<T> destroyTemporaryVariable(Operand<T> ref,
                                                                String varName)

Adds an DestroyTemporaryVariable operation to the graph

Parameters:

ref - A reference to the temporary variable tensor.

varName - Name of the temporary variable, usually the name of the matching

Returns:

a new instance of DestroyTemporaryVariable

See Also:

DestroyTemporaryVariable

sparseSparseMinimum

public <T> SparseSparseMinimum<T> sparseSparseMinimum(Operand<Long> aIndices,
                                                      Operand<T> aValues,
                                                      Operand<Long> aShape,
                                                      Operand<Long> bIndices,
                                                      Operand<T> bValues,
                                                      Operand<Long> bShape)

Adds an SparseSparseMinimum operation to the graph

Parameters:

aIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

aValues - 1-D. `N` non-empty values corresponding to `a_indices`.

aShape - 1-D. Shape of the input SparseTensor.

bIndices - counterpart to `a_indices` for the other operand.

bValues - counterpart to `a_values` for the other operand; must be of the same dtype.

bShape - counterpart to `a_shape` for the other operand; the two shapes must be equal.

Returns:

a new instance of SparseSparseMinimum

See Also:

SparseSparseMinimum

batchMatrixTriangularSolve

public <T extends Number> BatchMatrixTriangularSolve<T> batchMatrixTriangularSolve(Operand<T> matrix,
                                                                                   Operand<T> rhs,
                                                                                   BatchMatrixTriangularSolve.Options... options)

Adds an BatchMatrixTriangularSolve operation to the graph

Parameters:

matrix -

rhs -

options - carries optional attributes values

Returns:

a new instance of BatchMatrixTriangularSolve

See Also:

BatchMatrixTriangularSolve

erf

public <T extends Number> Erf<T> erf(Operand<T> x)

Adds an Erf operation to the graph

Parameters:

x -

Returns:

a new instance of Erf

See Also:

Erf

gcsConfigureCredentials

public GcsConfigureCredentials gcsConfigureCredentials(Operand<String> json)

Adds an GcsConfigureCredentials operation to the graph

Parameters:

json -

Returns:

a new instance of GcsConfigureCredentials

See Also:

GcsConfigureCredentials

statelessMultinomial

public <V extends Number,T extends Number,U extends Number> StatelessMultinomial<V> statelessMultinomial(Operand<T> logits,
                                                                                                         Operand<Integer> numSamples,
                                                                                                         Operand<U> seed,
                                                                                                         Class<V> outputDtype)

Adds an StatelessMultinomial operation to the graph

Parameters:

logits - 2-D Tensor with shape `[batch_size, num_classes]`. Each slice `[i, :]`

numSamples - 0-D. Number of independent samples to draw for each row slice.

seed - 2 seeds (shape [2]).

outputDtype -

Returns:

a new instance of StatelessMultinomial

See Also:

StatelessMultinomial

fact

public Fact fact()

Adds an Fact operation to the graph

Returns:

a new instance of Fact

See Also:

Fact

resourceApplyAdagrad

public <T> ResourceApplyAdagrad resourceApplyAdagrad(Operand<?> var,
                                                     Operand<?> accum,
                                                     Operand<T> lr,
                                                     Operand<T> grad,
                                                     ResourceApplyAdagrad.Options... options)

Adds an ResourceApplyAdagrad operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyAdagrad

See Also:

ResourceApplyAdagrad

tryRpc

public TryRpc tryRpc(Operand<String> address,
                     Operand<String> method,
                     Operand<String> request,
                     TryRpc.Options... options)

Adds an TryRpc operation to the graph

Parameters:

address - `0-D` or `1-D`. The address (i.e. host_name:port) of the RPC server.

method - `0-D` or `1-D`. The method address on the RPC server.

request - `0-D` or `1-D`. Serialized proto strings: the rpc request argument.

options - carries optional attributes values

Returns:

a new instance of TryRpc

See Also:

TryRpc

resourceApplyFtrl

public <T> ResourceApplyFtrl resourceApplyFtrl(Operand<?> var,
                                               Operand<?> accum,
                                               Operand<?> linear,
                                               Operand<T> grad,
                                               Operand<T> lr,
                                               Operand<T> l1,
                                               Operand<T> l2,
                                               Operand<T> lrPower,
                                               ResourceApplyFtrl.Options... options)

Adds an ResourceApplyFtrl operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

linear - Should be from a Variable().

grad - The gradient.

lr - Scaling factor. Must be a scalar.

l1 - L1 regulariation. Must be a scalar.

l2 - L2 regulariation. Must be a scalar.

lrPower - Scaling factor. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyFtrl

See Also:

ResourceApplyFtrl

applyProximalAdagrad

public <T> ApplyProximalAdagrad<T> applyProximalAdagrad(Operand<T> var,
                                                        Operand<T> accum,
                                                        Operand<T> lr,
                                                        Operand<T> l1,
                                                        Operand<T> l2,
                                                        Operand<T> grad,
                                                        ApplyProximalAdagrad.Options... options)

Adds an ApplyProximalAdagrad operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ApplyProximalAdagrad

See Also:

ApplyProximalAdagrad

broadcastDynamicShape

public <T extends Number> BroadcastDynamicShape<T> broadcastDynamicShape(Operand<T> s0,
                                                                         Operand<T> s1)

Adds an BroadcastDynamicShape operation to the graph

Parameters:

s0 -

s1 -

Returns:

a new instance of BroadcastDynamicShape

See Also:

BroadcastDynamicShape

sparseSegmentMeanWithNumSegments

public <T extends Number,U extends Number,V extends Number> SparseSegmentMeanWithNumSegments<T> sparseSegmentMeanWithNumSegments(Operand<T> data,
                                                                                                                                 Operand<U> indices,
                                                                                                                                 Operand<Integer> segmentIds,
                                                                                                                                 Operand<V> numSegments)

Adds an SparseSegmentMeanWithNumSegments operation to the graph

Parameters:

data -

indices - A 1-D tensor. Has same rank as `segment_ids`.

segmentIds - A 1-D tensor. Values should be sorted and can be repeated.

numSegments - Should equal the number of distinct segment IDs.

Returns:

a new instance of SparseSegmentMeanWithNumSegments

See Also:

SparseSegmentMeanWithNumSegments

shardedFilespec

public ShardedFilespec shardedFilespec(Operand<String> basename,
                                       Operand<Integer> numShards)

Adds an ShardedFilespec operation to the graph

Parameters:

basename -

numShards -

Returns:

a new instance of ShardedFilespec

See Also:

ShardedFilespec

iteratorGetNextAsOptional

public IteratorGetNextAsOptional iteratorGetNextAsOptional(Operand<?> iterator,
                                                           List<Class<?>> outputTypes,
                                                           List<Shape> outputShapes)

Adds an IteratorGetNextAsOptional operation to the graph

Parameters:

iterator -

outputTypes -

outputShapes -

Returns:

a new instance of IteratorGetNextAsOptional

See Also:

IteratorGetNextAsOptional

serializeSparse

public <T> SerializeSparse<String> serializeSparse(Operand<Long> sparseIndices,
                                                   Operand<T> sparseValues,
                                                   Operand<Long> sparseShape)

Adds an SerializeSparse operation to the graph

Parameters:

sparseIndices - 2-D. The `indices` of the `SparseTensor`.

sparseValues - 1-D. The `values` of the `SparseTensor`.

sparseShape - 1-D. The `shape` of the `SparseTensor`.

Returns:

a new instance of SerializeSparse

See Also:

SerializeSparse

maxPoolGradGrad

public <T extends Number> MaxPoolGradGrad<T> maxPoolGradGrad(Operand<T> origInput,
                                                             Operand<T> origOutput,
                                                             Operand<T> grad,
                                                             List<Long> ksize,
                                                             List<Long> strides,
                                                             String padding,
                                                             MaxPoolGradGrad.Options... options)

Adds an MaxPoolGradGrad operation to the graph

Parameters:

origInput - The original input tensor.

origOutput - The original output tensor.

grad - 4-D. Gradients of gradients w.r.t. the input of `max_pool`.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of MaxPoolGradGrad

See Also:

MaxPoolGradGrad

quantizeAndDequantizeV2

public <T extends Number> QuantizeAndDequantizeV2<T> quantizeAndDequantizeV2(Operand<T> input,
                                                                             Operand<T> inputMin,
                                                                             Operand<T> inputMax,
                                                                             QuantizeAndDequantizeV2.Options... options)

Adds an QuantizeAndDequantizeV2 operation to the graph

Parameters:

input - Tensor to quantize and then dequantize.

inputMin - If `range_given == True`, this specifies the minimum input value that needs to

inputMax - If `range_given == True`, this specifies the maximum input value that needs to

options - carries optional attributes values

Returns:

a new instance of QuantizeAndDequantizeV2

See Also:

QuantizeAndDequantizeV2

greaterEqual

public <T extends Number> GreaterEqual greaterEqual(Operand<T> x,
                                                    Operand<T> y)

Adds an GreaterEqual operation to the graph

Parameters:

x -

y -

Returns:

a new instance of GreaterEqual

See Also:

GreaterEqual

sqrt

public <T> Sqrt<T> sqrt(Operand<T> x)

Adds an Sqrt operation to the graph

Parameters:

x -

Returns:

a new instance of Sqrt

See Also:

Sqrt

decodeWav

public DecodeWav decodeWav(Operand<String> contents,
                           DecodeWav.Options... options)

Adds an DecodeWav operation to the graph

Parameters:

contents - The WAV-encoded audio, usually from a file.

options - carries optional attributes values

Returns:

a new instance of DecodeWav

See Also:

DecodeWav

sparseSlice

public <T> SparseSlice<T> sparseSlice(Operand<Long> indices,
                                      Operand<T> values,
                                      Operand<Long> shape,
                                      Operand<Long> start,
                                      Operand<Long> size)

Adds an SparseSlice operation to the graph

Parameters:

indices - 2-D tensor represents the indices of the sparse tensor.

values - 1-D tensor represents the values of the sparse tensor.

shape - 1-D. tensor represents the shape of the sparse tensor.

start - 1-D. tensor represents the start of the slice.

size - 1-D. tensor represents the size of the slice.

Returns:

a new instance of SparseSlice

See Also:

SparseSlice

decodeProtoV2

public DecodeProtoV2 decodeProtoV2(Operand<String> bytes,
                                   String messageType,
                                   List<String> fieldNames,
                                   List<Class<?>> outputTypes,
                                   DecodeProtoV2.Options... options)

Adds an DecodeProtoV2 operation to the graph

Parameters:

bytes - Tensor of serialized protos with shape `batch_shape`.

messageType - Name of the proto message type to decode.

fieldNames - List of strings containing proto field names.

outputTypes - List of TF types to use for the respective field in field_names.

options - carries optional attributes values

Returns:

a new instance of DecodeProtoV2

See Also:

DecodeProtoV2

iteratorGetNext

public IteratorGetNext iteratorGetNext(Operand<?> iterator,
                                       List<Class<?>> outputTypes,
                                       List<Shape> outputShapes)

Adds an IteratorGetNext operation to the graph

Parameters:

iterator -

outputTypes -

outputShapes -

Returns:

a new instance of IteratorGetNext

See Also:

IteratorGetNext

batchCholesky

public <T extends Number> BatchCholesky<T> batchCholesky(Operand<T> input)

Adds an BatchCholesky operation to the graph

Parameters:

input -

Returns:

a new instance of BatchCholesky

See Also:

BatchCholesky

mergeV2Checkpoints

public MergeV2Checkpoints mergeV2Checkpoints(Operand<String> checkpointPrefixes,
                                             Operand<String> destinationPrefix,
                                             MergeV2Checkpoints.Options... options)

Adds an MergeV2Checkpoints operation to the graph

Parameters:

checkpointPrefixes - prefixes of V2 checkpoints to merge.

destinationPrefix - scalar. The desired final prefix. Allowed to be the same

options - carries optional attributes values

Returns:

a new instance of MergeV2Checkpoints

See Also:

MergeV2Checkpoints

temporaryVariable

public <T> TemporaryVariable<T> temporaryVariable(Shape shape,
                                                  Class<T> dtype,
                                                  TemporaryVariable.Options... options)

Adds an TemporaryVariable operation to the graph

Parameters:

shape - The shape of the variable tensor.

dtype - The type of elements in the variable tensor.

options - carries optional attributes values

Returns:

a new instance of TemporaryVariable

See Also:

TemporaryVariable

applyFtrlV2

public <T> ApplyFtrlV2<T> applyFtrlV2(Operand<T> var,
                                      Operand<T> accum,
                                      Operand<T> linear,
                                      Operand<T> grad,
                                      Operand<T> lr,
                                      Operand<T> l1,
                                      Operand<T> l2,
                                      Operand<T> l2Shrinkage,
                                      Operand<T> lrPower,
                                      ApplyFtrlV2.Options... options)

Adds an ApplyFtrlV2 operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

linear - Should be from a Variable().

grad - The gradient.

lr - Scaling factor. Must be a scalar.

l1 - L1 regulariation. Must be a scalar.

l2 - L2 shrinkage regulariation. Must be a scalar.

l2Shrinkage -

lrPower - Scaling factor. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ApplyFtrlV2

See Also:

ApplyFtrlV2

maxPool3D

public <T extends Number> MaxPool3D<T> maxPool3D(Operand<T> input,
                                                 List<Long> ksize,
                                                 List<Long> strides,
                                                 String padding,
                                                 MaxPool3D.Options... options)

Adds an MaxPool3D operation to the graph

Parameters:

input - Shape `[batch, depth, rows, cols, channels]` tensor to pool over.

ksize - 1-D tensor of length 5. The size of the window for each dimension of

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of MaxPool3D

See Also:

MaxPool3D

fusedBatchNorm

public <T extends Number> FusedBatchNorm<T> fusedBatchNorm(Operand<T> x,
                                                           Operand<T> scale,
                                                           Operand<T> offset,
                                                           Operand<T> mean,
                                                           Operand<T> variance,
                                                           FusedBatchNorm.Options... options)

Adds an FusedBatchNorm operation to the graph

Parameters:

x - A 4D Tensor for input data.

scale - A 1D Tensor for scaling factor, to scale the normalized x.

offset - A 1D Tensor for offset, to shift to the normalized x.

mean - A 1D Tensor for population mean. Used for inference only;

variance - A 1D Tensor for population variance. Used for inference only;

options - carries optional attributes values

Returns:

a new instance of FusedBatchNorm

See Also:

FusedBatchNorm

segmentMin

public <T extends Number,U extends Number> SegmentMin<T> segmentMin(Operand<T> data,
                                                                    Operand<U> segmentIds)

Adds an SegmentMin operation to the graph

Parameters:

data -

segmentIds - A 1-D tensor whose size is equal to the size of `data`'s

Returns:

a new instance of SegmentMin

See Also:

SegmentMin

printV2

public PrintV2 printV2(Operand<String> input,
                       PrintV2.Options... options)

Adds an PrintV2 operation to the graph

Parameters:

input - The string scalar to print.

options - carries optional attributes values

Returns:

a new instance of PrintV2

See Also:

PrintV2

gcsConfigureBlockCache

public GcsConfigureBlockCache gcsConfigureBlockCache(Operand<?> maxCacheSize,
                                                     Operand<?> blockSize,
                                                     Operand<?> maxStaleness)

Adds an GcsConfigureBlockCache operation to the graph

Parameters:

maxCacheSize -

blockSize -

maxStaleness -

Returns:

a new instance of GcsConfigureBlockCache

See Also:

GcsConfigureBlockCache

biasAdd

public <T> BiasAdd<T> biasAdd(Operand<T> value,
                              Operand<T> bias,
                              BiasAdd.Options... options)

Adds an BiasAdd operation to the graph

Parameters:

value - Any number of dimensions.

bias - 1-D with size the last dimension of `value`.

options - carries optional attributes values

Returns:

a new instance of BiasAdd

See Also:

BiasAdd

matrixSquareRoot

public <T> MatrixSquareRoot<T> matrixSquareRoot(Operand<T> input)

Adds an MatrixSquareRoot operation to the graph

Parameters:

input - Shape is `[..., M, M]`.

Returns:

a new instance of MatrixSquareRoot

See Also:

MatrixSquareRoot

abs

public <T extends Number> Abs<T> abs(Operand<T> x)

Adds an Abs operation to the graph

Parameters:

x -

Returns:

a new instance of Abs

See Also:

Abs

dynamicStitch

public <T> DynamicStitch<T> dynamicStitch(Iterable<Operand<Integer>> indices,
                                          Operand<T> data)

Adds an DynamicStitch operation to the graph

Parameters:

indices -

data -

Returns:

a new instance of DynamicStitch

See Also:

DynamicStitch

substr

public <T extends Number> Substr substr(Operand<String> input,
                                        Operand<T> pos,
                                        Operand<T> len,
                                        Substr.Options... options)

Adds an Substr operation to the graph

Parameters:

input - Tensor of strings

pos - Scalar defining the position of first character in each substring

len - Scalar defining the number of characters to include in each substring

options - carries optional attributes values

Returns:

a new instance of Substr

See Also:

Substr

resizeBilinear

public <T extends Number> ResizeBilinear resizeBilinear(Operand<T> images,
                                                        Operand<Integer> size,
                                                        ResizeBilinear.Options... options)

Adds an ResizeBilinear operation to the graph

Parameters:

images - 4-D with shape `[batch, height, width, channels]`.

size - = A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The

options - carries optional attributes values

Returns:

a new instance of ResizeBilinear

See Also:

ResizeBilinear

acosh

public <T> Acosh<T> acosh(Operand<T> x)

Adds an Acosh operation to the graph

Parameters:

x -

Returns:

a new instance of Acosh

See Also:

Acosh

constant

public Constant<Long> constant(long[][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

emptyTensorList

public <T extends Number,U> EmptyTensorList emptyTensorList(Operand<T> elementShape,
                                                            Operand<Integer> maxNumElements,
                                                            Class<U> elementDtype)

Adds an EmptyTensorList operation to the graph

Parameters:

elementShape -

maxNumElements -

elementDtype -

Returns:

a new instance of EmptyTensorList

See Also:

EmptyTensorList

queueIsClosedV2

public QueueIsClosedV2 queueIsClosedV2(Operand<?> handle)

Adds an QueueIsClosedV2 operation to the graph

Parameters:

handle - The handle to a queue.

Returns:

a new instance of QueueIsClosedV2

See Also:

QueueIsClosedV2

selu

public <T extends Number> Selu<T> selu(Operand<T> features)

Adds an Selu operation to the graph

Parameters:

features -

Returns:

a new instance of Selu

See Also:

Selu

quantizeAndDequantizeV3

public <T extends Number> QuantizeAndDequantizeV3<T> quantizeAndDequantizeV3(Operand<T> input,
                                                                             Operand<T> inputMin,
                                                                             Operand<T> inputMax,
                                                                             Operand<Integer> numBits,
                                                                             QuantizeAndDequantizeV3.Options... options)

Adds an QuantizeAndDequantizeV3 operation to the graph

Parameters:

input -

inputMin -

inputMax -

numBits -

options - carries optional attributes values

Returns:

a new instance of QuantizeAndDequantizeV3

See Also:

QuantizeAndDequantizeV3

sparseCross

public <T,U> SparseCross<T> sparseCross(Iterable<Operand<Long>> indices,
                                        Iterable<Operand<?>> values,
                                        Iterable<Operand<Long>> shapes,
                                        Iterable<Operand<?>> denseInputs,
                                        Boolean hashedOutput,
                                        Long numBuckets,
                                        Long hashKey,
                                        Class<T> outType,
                                        Class<U> internalType)

Adds an SparseCross operation to the graph

Parameters:

indices - 2-D. Indices of each input `SparseTensor`.

values - 1-D. values of each `SparseTensor`.

shapes - 1-D. Shapes of each `SparseTensor`.

denseInputs - 2-D. Columns represented by dense `Tensor`.

hashedOutput - If true, returns the hash of the cross instead of the string.

numBuckets - It is used if hashed_output is true.

hashKey - Specify the hash_key that will be used by the `FingerprintCat64`

outType -

internalType -

Returns:

a new instance of SparseCross

See Also:

SparseCross

constant

public Constant<Boolean> constant(boolean[] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

resourceSparseApplyMomentum

public <T,U extends Number> ResourceSparseApplyMomentum resourceSparseApplyMomentum(Operand<?> var,
                                                                                    Operand<?> accum,
                                                                                    Operand<T> lr,
                                                                                    Operand<T> grad,
                                                                                    Operand<U> indices,
                                                                                    Operand<T> momentum,
                                                                                    ResourceSparseApplyMomentum.Options... options)

Adds an ResourceSparseApplyMomentum operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Learning rate. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

momentum - Momentum. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyMomentum

See Also:

ResourceSparseApplyMomentum

iRFFT

public IRFFT iRFFT(Operand<?> input,
                   Operand<Integer> fftLength)

Adds an IRFFT operation to the graph

Parameters:

input - A complex64 tensor.

fftLength - An int32 tensor of shape [1]. The FFT length.

Returns:

a new instance of IRFFT

See Also:

IRFFT

getSessionTensor

public <T> GetSessionTensor<T> getSessionTensor(Operand<String> handle,
                                                Class<T> dtype)

Adds an GetSessionTensor operation to the graph

Parameters:

handle - The handle for a tensor stored in the session state.

dtype - The type of the output value.

Returns:

a new instance of GetSessionTensor

See Also:

GetSessionTensor

zeros

public <T,U extends Number> Zeros<T> zeros(Operand<U> dims,
                                           Class<T> type)

Adds an Zeros operation to the graph

Parameters:

dims - a 1-D operand that represents the shape of the output tensor

type - the output tensor datatype

Returns:

a constant tensor initialized with zeros

Throws:

IllegalArgumentException - if the tensor type or shape cannot be initialized with zeros.

See Also:

Zeros

exp

public <T> Exp<T> exp(Operand<T> x)

Adds an Exp operation to the graph

Parameters:

x -

Returns:

a new instance of Exp

See Also:

Exp

padV2

public <T,U extends Number> PadV2<T> padV2(Operand<T> input,
                                           Operand<U> paddings,
                                           Operand<T> constantValues)

Adds an PadV2 operation to the graph

Parameters:

input -

paddings -

constantValues -

Returns:

a new instance of PadV2

See Also:

PadV2

mul

public <T> Mul<T> mul(Operand<T> x,
                      Operand<T> y)

Adds an Mul operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Mul

See Also:

Mul

resourceSparseApplyProximalGradientDescent

public <T,U extends Number> ResourceSparseApplyProximalGradientDescent resourceSparseApplyProximalGradientDescent(Operand<?> var,
                                                                                                                  Operand<T> alpha,
                                                                                                                  Operand<T> l1,
                                                                                                                  Operand<T> l2,
                                                                                                                  Operand<T> grad,
                                                                                                                  Operand<U> indices,
                                                                                                                  ResourceSparseApplyProximalGradientDescent.Options... options)

Adds an ResourceSparseApplyProximalGradientDescent operation to the graph

Parameters:

var - Should be from a Variable().

alpha - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyProximalGradientDescent

See Also:

ResourceSparseApplyProximalGradientDescent

constant

public Constant<Boolean> constant(boolean[][][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

iFFT

public <T> IFFT<T> iFFT(Operand<T> input)

Adds an IFFT operation to the graph

Parameters:

input - A complex tensor.

Returns:

a new instance of IFFT

See Also:

IFFT

constant

public Constant<Float> constant(float[] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

iRFFT3D

public IRFFT3D iRFFT3D(Operand<?> input,
                       Operand<Integer> fftLength)

Adds an IRFFT3D operation to the graph

Parameters:

input - A complex64 tensor.

fftLength - An int32 tensor of shape [3]. The FFT length for each dimension.

Returns:

a new instance of IRFFT3D

See Also:

IRFFT3D

randomUniformInt

public <U extends Number,T extends Number> RandomUniformInt<U> randomUniformInt(Operand<T> shape,
                                                                                Operand<U> minval,
                                                                                Operand<U> maxval,
                                                                                RandomUniformInt.Options... options)

Adds an RandomUniformInt operation to the graph

Parameters:

shape - The shape of the output tensor.

minval - 0-D. Inclusive lower bound on the generated integers.

maxval - 0-D. Exclusive upper bound on the generated integers.

options - carries optional attributes values

Returns:

a new instance of RandomUniformInt

See Also:

RandomUniformInt

reduceJoin

public ReduceJoin reduceJoin(Operand<String> inputs,
                             Operand<Integer> reductionIndices,
                             ReduceJoin.Options... options)

Adds an ReduceJoin operation to the graph

Parameters:

inputs - The input to be joined. All reduced indices must have non-zero size.

reductionIndices - The dimensions to reduce over. Dimensions are reduced in the

options - carries optional attributes values

Returns:

a new instance of ReduceJoin

See Also:

ReduceJoin

uniformCandidateSampler

public UniformCandidateSampler uniformCandidateSampler(Operand<Long> trueClasses,
                                                       Long numTrue,
                                                       Long numSampled,
                                                       Boolean unique,
                                                       Long rangeMax,
                                                       UniformCandidateSampler.Options... options)

Adds an UniformCandidateSampler operation to the graph

Parameters:

trueClasses - A batch_size * num_true matrix, in which each row contains the

numTrue - Number of true labels per context.

numSampled - Number of candidates to randomly sample.

unique - If unique is true, we sample with rejection, so that all sampled

rangeMax - The sampler will sample integers from the interval [0, range_max).

options - carries optional attributes values

Returns:

a new instance of UniformCandidateSampler

See Also:

UniformCandidateSampler

debugGradientRefIdentity

public <T> DebugGradientRefIdentity<T> debugGradientRefIdentity(Operand<T> input)

Adds an DebugGradientRefIdentity operation to the graph

Parameters:

input -

Returns:

a new instance of DebugGradientRefIdentity

See Also:

DebugGradientRefIdentity

dilation2DBackpropInput

public <T extends Number> Dilation2DBackpropInput<T> dilation2DBackpropInput(Operand<T> input,
                                                                             Operand<T> filter,
                                                                             Operand<T> outBackprop,
                                                                             List<Long> strides,
                                                                             List<Long> rates,
                                                                             String padding)

Adds an Dilation2DBackpropInput operation to the graph

Parameters:

input - 4-D with shape `[batch, in_height, in_width, depth]`.

filter - 3-D with shape `[filter_height, filter_width, depth]`.

outBackprop - 4-D with shape `[batch, out_height, out_width, depth]`.

strides - 1-D of length 4. The stride of the sliding window for each dimension of

rates - 1-D of length 4. The input stride for atrous morphological dilation.

padding - The type of padding algorithm to use.

Returns:

a new instance of Dilation2DBackpropInput

See Also:

Dilation2DBackpropInput

getSessionHandle

public <T> GetSessionHandle getSessionHandle(Operand<T> value)

Adds an GetSessionHandle operation to the graph

Parameters:

value - The tensor to be stored.

Returns:

a new instance of GetSessionHandle

See Also:

GetSessionHandle

applyAddSign

public <T> ApplyAddSign<T> applyAddSign(Operand<T> var,
                                        Operand<T> m,
                                        Operand<T> lr,
                                        Operand<T> alpha,
                                        Operand<T> signDecay,
                                        Operand<T> beta,
                                        Operand<T> grad,
                                        ApplyAddSign.Options... options)

Adds an ApplyAddSign operation to the graph

Parameters:

var - Should be from a Variable().

m - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

alpha - Must be a scalar.

signDecay - Must be a scalar.

beta - Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ApplyAddSign

See Also:

ApplyAddSign

readerRead

public ReaderRead readerRead(Operand<?> readerHandle,
                             Operand<?> queueHandle)

Adds an ReaderRead operation to the graph

Parameters:

readerHandle - Handle to a Reader.

queueHandle - Handle to a Queue, with string work items.

Returns:

a new instance of ReaderRead

See Also:

ReaderRead

iteratorFromStringHandle

public IteratorFromStringHandle iteratorFromStringHandle(Operand<String> stringHandle,
                                                         List<Class<?>> outputTypes,
                                                         IteratorFromStringHandle.Options... options)

Adds an IteratorFromStringHandle operation to the graph

Parameters:

stringHandle - A string representation of the given handle.

outputTypes - If specified, defines the type of each tuple component in an

options - carries optional attributes values

Returns:

a new instance of IteratorFromStringHandle

See Also:

IteratorFromStringHandle

scatterMax

public <T extends Number,U extends Number> ScatterMax<T> scatterMax(Operand<T> ref,
                                                                    Operand<U> indices,
                                                                    Operand<T> updates,
                                                                    ScatterMax.Options... options)

Adds an ScatterMax operation to the graph

Parameters:

ref - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to reduce into `ref`.

options - carries optional attributes values

Returns:

a new instance of ScatterMax

See Also:

ScatterMax

fixedUnigramCandidateSampler

public FixedUnigramCandidateSampler fixedUnigramCandidateSampler(Operand<Long> trueClasses,
                                                                 Long numTrue,
                                                                 Long numSampled,
                                                                 Boolean unique,
                                                                 Long rangeMax,
                                                                 FixedUnigramCandidateSampler.Options... options)

Adds an FixedUnigramCandidateSampler operation to the graph

Parameters:

trueClasses - A batch_size * num_true matrix, in which each row contains the

numTrue - Number of true labels per context.

numSampled - Number of candidates to randomly sample.

unique - If unique is true, we sample with rejection, so that all sampled

rangeMax - The sampler will sample integers from the interval [0, range_max).

options - carries optional attributes values

Returns:

a new instance of FixedUnigramCandidateSampler

See Also:

FixedUnigramCandidateSampler

mapUnstageNoKey

public MapUnstageNoKey mapUnstageNoKey(Operand<Integer> indices,
                                       List<Class<?>> dtypes,
                                       MapUnstageNoKey.Options... options)

Adds an MapUnstageNoKey operation to the graph

Parameters:

indices -

dtypes -

options - carries optional attributes values

Returns:

a new instance of MapUnstageNoKey

See Also:

MapUnstageNoKey

addSparseToTensorsMap

public <T> AddSparseToTensorsMap addSparseToTensorsMap(Operand<Long> sparseIndices,
                                                       Operand<T> sparseValues,
                                                       Operand<Long> sparseShape,
                                                       AddSparseToTensorsMap.Options... options)

Adds an AddSparseToTensorsMap operation to the graph

Parameters:

sparseIndices - 2-D. The `indices` of the `SparseTensor`.

sparseValues - 1-D. The `values` of the `SparseTensor`.

sparseShape - 1-D. The `shape` of the `SparseTensor`.

options - carries optional attributes values

Returns:

a new instance of AddSparseToTensorsMap

See Also:

AddSparseToTensorsMap

loadAndRemapMatrix

public LoadAndRemapMatrix loadAndRemapMatrix(Operand<String> ckptPath,
                                             Operand<String> oldTensorName,
                                             Operand<Long> rowRemapping,
                                             Operand<Long> colRemapping,
                                             Operand<Float> initializingValues,
                                             Long numRows,
                                             Long numCols,
                                             LoadAndRemapMatrix.Options... options)

Adds an LoadAndRemapMatrix operation to the graph

Parameters:

ckptPath - Path to the TensorFlow checkpoint (version 2, `TensorBundle`) from

oldTensorName - Name of the 2-D `Tensor` to load from checkpoint.

rowRemapping - An int `Tensor` of row remappings (generally created by

colRemapping - An int `Tensor` of column remappings (generally created by

initializingValues - A float `Tensor` containing values to fill in for cells

numRows - Number of rows (length of the 1st dimension) in the output matrix.

numCols - Number of columns (length of the 2nd dimension) in the output matrix.

options - carries optional attributes values

Returns:

a new instance of LoadAndRemapMatrix

See Also:

LoadAndRemapMatrix

scatterNd

public <U,T extends Number> ScatterNd<U> scatterNd(Operand<T> indices,
                                                   Operand<U> updates,
                                                   Operand<T> shape)

Adds an ScatterNd operation to the graph

Parameters:

indices - Index tensor.

updates - Updates to scatter into output.

shape - 1-D. The shape of the resulting tensor.

Returns:

a new instance of ScatterNd

See Also:

ScatterNd

tensorListSetItem

public <T> TensorListSetItem tensorListSetItem(Operand<?> inputHandle,
                                               Operand<Integer> index,
                                               Operand<T> item)

Adds an TensorListSetItem operation to the graph

Parameters:

inputHandle -

index -

item -

Returns:

a new instance of TensorListSetItem

See Also:

TensorListSetItem

refSwitch

public <T> RefSwitch<T> refSwitch(Operand<T> data,
                                  Operand<Boolean> pred)

Adds an RefSwitch operation to the graph

Parameters:

data - The ref tensor to be forwarded to the appropriate output.

pred - A scalar that specifies which output port will receive data.

Returns:

a new instance of RefSwitch

See Also:

RefSwitch

sdcaOptimizer

public SdcaOptimizer sdcaOptimizer(Iterable<Operand<Long>> sparseExampleIndices,
                                   Iterable<Operand<Long>> sparseFeatureIndices,
                                   Iterable<Operand<Float>> sparseFeatureValues,
                                   Iterable<Operand<Float>> denseFeatures,
                                   Operand<Float> exampleWeights,
                                   Operand<Float> exampleLabels,
                                   Iterable<Operand<Long>> sparseIndices,
                                   Iterable<Operand<Float>> sparseWeights,
                                   Iterable<Operand<Float>> denseWeights,
                                   Operand<Float> exampleStateData,
                                   String lossType,
                                   Float l1,
                                   Float l2,
                                   Long numLossPartitions,
                                   Long numInnerIterations,
                                   SdcaOptimizer.Options... options)

Adds an SdcaOptimizer operation to the graph

Parameters:

sparseExampleIndices - a list of vectors which contain example indices.

sparseFeatureIndices - a list of vectors which contain feature indices.

sparseFeatureValues - a list of vectors which contains feature value

denseFeatures - a list of matrices which contains the dense feature values.

exampleWeights - a vector which contains the weight associated with each

exampleLabels - a vector which contains the label/target associated with each

sparseIndices - a list of vectors where each value is the indices which has

sparseWeights - a list of vectors where each value is the weight associated with

denseWeights - a list of vectors where the values are the weights associated

exampleStateData - a list of vectors containing the example state data.

lossType - Type of the primal loss. Currently SdcaSolver supports logistic,

l1 - Symmetric l1 regularization strength.

l2 - Symmetric l2 regularization strength.

numLossPartitions - Number of partitions of the global loss function.

numInnerIterations - Number of iterations per mini-batch.

options - carries optional attributes values

Returns:

a new instance of SdcaOptimizer

See Also:

SdcaOptimizer

sparseFillEmptyRowsGrad

public <T> SparseFillEmptyRowsGrad<T> sparseFillEmptyRowsGrad(Operand<Long> reverseIndexMap,
                                                              Operand<T> gradValues)

Adds an SparseFillEmptyRowsGrad operation to the graph

Parameters:

reverseIndexMap - 1-D. The reverse index map from SparseFillEmptyRows.

gradValues - 1-D. The gradients from backprop.

Returns:

a new instance of SparseFillEmptyRowsGrad

See Also:

SparseFillEmptyRowsGrad

identity

public <T> Identity<T> identity(Operand<T> input)

Adds an Identity operation to the graph

Parameters:

input -

Returns:

a new instance of Identity

See Also:

Identity

where3

public <T> Where3<T> where3(Operand<Boolean> condition,
                            Operand<T> x,
                            Operand<T> y)

Adds an Where3 operation to the graph

Parameters:

condition -

x - = A `Tensor` which may have the same shape as `condition`.

y - = A `Tensor` with the same type and shape as `x`.

Returns:

a new instance of Where3

See Also:

Where3

readerReadUpTo

public ReaderReadUpTo readerReadUpTo(Operand<?> readerHandle,
                                     Operand<?> queueHandle,
                                     Operand<Long> numRecords)

Adds an ReaderReadUpTo operation to the graph

Parameters:

readerHandle - Handle to a `Reader`.

queueHandle - Handle to a `Queue`, with string work items.

numRecords - number of records to read from `Reader`.

Returns:

a new instance of ReaderReadUpTo

See Also:

ReaderReadUpTo

lRN

public <T extends Number> LRN<T> lRN(Operand<T> input,
                                     LRN.Options... options)

Adds an LRN operation to the graph

Parameters:

input - 4-D.

options - carries optional attributes values

Returns:

a new instance of LRN

See Also:

LRN

serializeManySparse

public <U,T> SerializeManySparse<U> serializeManySparse(Operand<Long> sparseIndices,
                                                        Operand<T> sparseValues,
                                                        Operand<Long> sparseShape,
                                                        Class<U> outType)

Adds an SerializeManySparse operation to the graph

Parameters:

sparseIndices - 2-D. The `indices` of the minibatch `SparseTensor`.

sparseValues - 1-D. The `values` of the minibatch `SparseTensor`.

sparseShape - 1-D. The `shape` of the minibatch `SparseTensor`.

outType - The `dtype` to use for serialization; the supported types are `string`

Returns:

a new instance of SerializeManySparse

See Also:

SerializeManySparse

audioSpectrogram

public AudioSpectrogram audioSpectrogram(Operand<Float> input,
                                         Long windowSize,
                                         Long stride,
                                         AudioSpectrogram.Options... options)

Adds an AudioSpectrogram operation to the graph

Parameters:

input - Float representation of audio data.

windowSize - How wide the input window is in samples. For the highest efficiency

stride - How widely apart the center of adjacent sample windows should be.

options - carries optional attributes values

Returns:

a new instance of AudioSpectrogram

See Also:

AudioSpectrogram

sparseApplyProximalAdagrad

public <T,U extends Number> SparseApplyProximalAdagrad<T> sparseApplyProximalAdagrad(Operand<T> var,
                                                                                     Operand<T> accum,
                                                                                     Operand<T> lr,
                                                                                     Operand<T> l1,
                                                                                     Operand<T> l2,
                                                                                     Operand<T> grad,
                                                                                     Operand<U> indices,
                                                                                     SparseApplyProximalAdagrad.Options... options)

Adds an SparseApplyProximalAdagrad operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Learning rate. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

options - carries optional attributes values

Returns:

a new instance of SparseApplyProximalAdagrad

See Also:

SparseApplyProximalAdagrad

quantizedRelu6

public <U,T> QuantizedRelu6<U> quantizedRelu6(Operand<T> features,
                                              Operand<Float> minFeatures,
                                              Operand<Float> maxFeatures,
                                              Class<U> outType)

Adds an QuantizedRelu6 operation to the graph

Parameters:

features -

minFeatures - The float value that the lowest quantized value represents.

maxFeatures - The float value that the highest quantized value represents.

outType -

Returns:

a new instance of QuantizedRelu6

See Also:

QuantizedRelu6

greater

public <T extends Number> Greater greater(Operand<T> x,
                                          Operand<T> y)

Adds an Greater operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Greater

See Also:

Greater

encodeBase64

public EncodeBase64 encodeBase64(Operand<String> input,
                                 EncodeBase64.Options... options)

Adds an EncodeBase64 operation to the graph

Parameters:

input - Strings to be encoded.

options - carries optional attributes values

Returns:

a new instance of EncodeBase64

See Also:

EncodeBase64

square

public <T> Square<T> square(Operand<T> x)

Adds an Square operation to the graph

Parameters:

x -

Returns:

a new instance of Square

See Also:

Square

encodeWav

public EncodeWav encodeWav(Operand<Float> audio,
                           Operand<Integer> sampleRate)

Adds an EncodeWav operation to the graph

Parameters:

audio - 2-D with shape `[length, channels]`.

sampleRate - Scalar containing the sample frequency.

Returns:

a new instance of EncodeWav

See Also:

EncodeWav

resourceApplyFtrlV2

public <T> ResourceApplyFtrlV2 resourceApplyFtrlV2(Operand<?> var,
                                                   Operand<?> accum,
                                                   Operand<?> linear,
                                                   Operand<T> grad,
                                                   Operand<T> lr,
                                                   Operand<T> l1,
                                                   Operand<T> l2,
                                                   Operand<T> l2Shrinkage,
                                                   Operand<T> lrPower,
                                                   ResourceApplyFtrlV2.Options... options)

Adds an ResourceApplyFtrlV2 operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

linear - Should be from a Variable().

grad - The gradient.

lr - Scaling factor. Must be a scalar.

l1 - L1 regulariation. Must be a scalar.

l2 - L2 shrinkage regulariation. Must be a scalar.

l2Shrinkage -

lrPower - Scaling factor. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyFtrlV2

See Also:

ResourceApplyFtrlV2

sparseToSparseSetOperation

public <T> SparseToSparseSetOperation<T> sparseToSparseSetOperation(Operand<Long> set1Indices,
                                                                    Operand<T> set1Values,
                                                                    Operand<Long> set1Shape,
                                                                    Operand<Long> set2Indices,
                                                                    Operand<T> set2Values,
                                                                    Operand<Long> set2Shape,
                                                                    String setOperation,
                                                                    SparseToSparseSetOperation.Options... options)

Adds an SparseToSparseSetOperation operation to the graph

Parameters:

set1Indices - 2D `Tensor`, indices of a `SparseTensor`. Must be in row-major

set1Values - 1D `Tensor`, values of a `SparseTensor`. Must be in row-major

set1Shape - 1D `Tensor`, shape of a `SparseTensor`. `set1_shape[0...n-1]` must

set2Indices - 2D `Tensor`, indices of a `SparseTensor`. Must be in row-major

set2Values - 1D `Tensor`, values of a `SparseTensor`. Must be in row-major

set2Shape - 1D `Tensor`, shape of a `SparseTensor`. `set2_shape[0...n-1]` must

setOperation -

options - carries optional attributes values

Returns:

a new instance of SparseToSparseSetOperation

See Also:

SparseToSparseSetOperation

timestamp

public Timestamp timestamp()

Adds an Timestamp operation to the graph

Returns:

a new instance of Timestamp

See Also:

Timestamp

batchMatrixSolveLs

public <T extends Number> BatchMatrixSolveLs<T> batchMatrixSolveLs(Operand<T> matrix,
                                                                   Operand<T> rhs,
                                                                   Operand<Double> l2Regularizer,
                                                                   BatchMatrixSolveLs.Options... options)

Adds an BatchMatrixSolveLs operation to the graph

Parameters:

matrix -

rhs -

l2Regularizer -

options - carries optional attributes values

Returns:

a new instance of BatchMatrixSolveLs

See Also:

BatchMatrixSolveLs

statelessRandomUniform

public <T extends Number,U extends Number> StatelessRandomUniform<Float> statelessRandomUniform(Operand<T> shape,
                                                                                                Operand<U> seed)

Adds an StatelessRandomUniform operation to the graph

Parameters:

shape - The shape of the output tensor.

seed - 2 seeds (shape [2]).

Returns:

a new instance of StatelessRandomUniform

See Also:

StatelessRandomUniform

constant

public Constant<Boolean> constant(boolean[][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

sparseAdd

public <T,U extends Number> SparseAdd<T> sparseAdd(Operand<Long> aIndices,
                                                   Operand<T> aValues,
                                                   Operand<Long> aShape,
                                                   Operand<Long> bIndices,
                                                   Operand<T> bValues,
                                                   Operand<Long> bShape,
                                                   Operand<U> thresh)

Adds an SparseAdd operation to the graph

Parameters:

aIndices - 2-D. The `indices` of the first `SparseTensor`, size `[nnz, ndims]` Matrix.

aValues - 1-D. The `values` of the first `SparseTensor`, size `[nnz]` Vector.

aShape - 1-D. The `shape` of the first `SparseTensor`, size `[ndims]` Vector.

bIndices - 2-D. The `indices` of the second `SparseTensor`, size `[nnz, ndims]` Matrix.

bValues - 1-D. The `values` of the second `SparseTensor`, size `[nnz]` Vector.

bShape - 1-D. The `shape` of the second `SparseTensor`, size `[ndims]` Vector.

thresh - 0-D. The magnitude threshold that determines if an output value/index

Returns:

a new instance of SparseAdd

See Also:

SparseAdd

linSpace

public <T extends Number,U extends Number> LinSpace<T> linSpace(Operand<T> start,
                                                                Operand<T> stop,
                                                                Operand<U> num)

Adds an LinSpace operation to the graph

Parameters:

start - 0-D tensor. First entry in the range.

stop - 0-D tensor. Last entry in the range.

num - 0-D tensor. Number of values to generate.

Returns:

a new instance of LinSpace

See Also:

LinSpace

addManySparseToTensorsMap

public <T> AddManySparseToTensorsMap addManySparseToTensorsMap(Operand<Long> sparseIndices,
                                                               Operand<T> sparseValues,
                                                               Operand<Long> sparseShape,
                                                               AddManySparseToTensorsMap.Options... options)

Adds an AddManySparseToTensorsMap operation to the graph

Parameters:

sparseIndices - 2-D. The `indices` of the minibatch `SparseTensor`.

sparseValues - 1-D. The `values` of the minibatch `SparseTensor`.

sparseShape - 1-D. The `shape` of the minibatch `SparseTensor`.

options - carries optional attributes values

Returns:

a new instance of AddManySparseToTensorsMap

See Also:

AddManySparseToTensorsMap

constant

public Constant<String> constant(String data)

Adds an Constant operation to the graph

Parameters:

data - The string to put into the new constant.

Returns:

a string constant

See Also:

Constant

inplaceSub

public <T> InplaceSub<T> inplaceSub(Operand<T> x,
                                    Operand<Integer> i,
                                    Operand<T> v)

Adds an InplaceSub operation to the graph

Parameters:

x - A `Tensor` of type T.

i - A vector. Indices into the left-most dimension of `x`.

v - A `Tensor` of type T. Same dimension sizes as x except the first dimension, which must be the same as i's size.

Returns:

a new instance of InplaceSub

See Also:

InplaceSub

cumsum

public <T,U extends Number> Cumsum<T> cumsum(Operand<T> x,
                                             Operand<U> axis,
                                             Cumsum.Options... options)

Adds an Cumsum operation to the graph

Parameters:

x - A `Tensor`. Must be one of the following types: `float32`, `float64`,

axis - A `Tensor` of type `int32` (default: 0). Must be in the range

options - carries optional attributes values

Returns:

a new instance of Cumsum

See Also:

Cumsum

stringToNumber

public StringToNumber<Float> stringToNumber(Operand<String> stringTensor)

Adds an StringToNumber operation to the graph

Parameters:

stringTensor -

Returns:

a new instance of StringToNumber

See Also:

StringToNumber

debugGradientIdentity

public <T> DebugGradientIdentity<T> debugGradientIdentity(Operand<T> input)

Adds an DebugGradientIdentity operation to the graph

Parameters:

input -

Returns:

a new instance of DebugGradientIdentity

See Also:

DebugGradientIdentity

applyProximalGradientDescent

public <T> ApplyProximalGradientDescent<T> applyProximalGradientDescent(Operand<T> var,
                                                                        Operand<T> alpha,
                                                                        Operand<T> l1,
                                                                        Operand<T> l2,
                                                                        Operand<T> delta,
                                                                        ApplyProximalGradientDescent.Options... options)

Adds an ApplyProximalGradientDescent operation to the graph

Parameters:

var - Should be from a Variable().

alpha - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

delta - The change.

options - carries optional attributes values

Returns:

a new instance of ApplyProximalGradientDescent

See Also:

ApplyProximalGradientDescent

sdcaFprint

public SdcaFprint sdcaFprint(Operand<String> input)

Adds an SdcaFprint operation to the graph

Parameters:

input - vector of strings to compute fingerprints on.

Returns:

a new instance of SdcaFprint

See Also:

SdcaFprint

fFT

public <T> FFT<T> fFT(Operand<T> input)

Adds an FFT operation to the graph

Parameters:

input - A complex tensor.

Returns:

a new instance of FFT

See Also:

FFT

optionalNone

public OptionalNone optionalNone()

Adds an OptionalNone operation to the graph

Returns:

a new instance of OptionalNone

See Also:

OptionalNone

mapStage

public MapStage mapStage(Operand<Long> key,
                         Operand<Integer> indices,
                         Iterable<Operand<?>> values,
                         List<Class<?>> dtypes,
                         MapStage.Options... options)

Adds an MapStage operation to the graph

Parameters:

key - int64

indices -

values - a list of tensors

dtypes -

options - carries optional attributes values

Returns:

a new instance of MapStage

See Also:

MapStage

sparseApplyProximalGradientDescent

public <T,U extends Number> SparseApplyProximalGradientDescent<T> sparseApplyProximalGradientDescent(Operand<T> var,
                                                                                                     Operand<T> alpha,
                                                                                                     Operand<T> l1,
                                                                                                     Operand<T> l2,
                                                                                                     Operand<T> grad,
                                                                                                     Operand<U> indices,
                                                                                                     SparseApplyProximalGradientDescent.Options... options)

Adds an SparseApplyProximalGradientDescent operation to the graph

Parameters:

var - Should be from a Variable().

alpha - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

options - carries optional attributes values

Returns:

a new instance of SparseApplyProximalGradientDescent

See Also:

SparseApplyProximalGradientDescent

immutableConst

public <T> ImmutableConst<T> immutableConst(Class<T> dtype,
                                            Shape shape,
                                            String memoryRegionName)

Adds an ImmutableConst operation to the graph

Parameters:

dtype - Type of the returned tensor.

shape - Shape of the returned tensor.

memoryRegionName - Name of readonly memory region used by the tensor, see

Returns:

a new instance of ImmutableConst

See Also:

ImmutableConst

resizeNearestNeighbor

public <T extends Number> ResizeNearestNeighbor<T> resizeNearestNeighbor(Operand<T> images,
                                                                         Operand<Integer> size,
                                                                         ResizeNearestNeighbor.Options... options)

Adds an ResizeNearestNeighbor operation to the graph

Parameters:

images - 4-D with shape `[batch, height, width, channels]`.

size - = A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The

options - carries optional attributes values

Returns:

a new instance of ResizeNearestNeighbor

See Also:

ResizeNearestNeighbor

constant

public Constant<String> constant(byte[] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. String elements are

See Also:

Constant

max

public <T,U extends Number> Max<T> max(Operand<T> input,
                                       Operand<U> axis,
                                       Max.Options... options)

Adds an Max operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of Max

See Also:

Max

sparseReduceMaxSparse

public <T extends Number> SparseReduceMaxSparse<T> sparseReduceMaxSparse(Operand<Long> inputIndices,
                                                                         Operand<T> inputValues,
                                                                         Operand<Long> inputShape,
                                                                         Operand<Integer> reductionAxes,
                                                                         SparseReduceMaxSparse.Options... options)

Adds an SparseReduceMaxSparse operation to the graph

Parameters:

inputIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

inputValues - 1-D. `N` non-empty values corresponding to `input_indices`.

inputShape - 1-D. Shape of the input SparseTensor.

reductionAxes - 1-D. Length-`K` vector containing the reduction axes.

options - carries optional attributes values

Returns:

a new instance of SparseReduceMaxSparse

See Also:

SparseReduceMaxSparse

size

public <U extends Number,T> Size<U> size(Operand<T> input,
                                         Class<U> outType)

Adds an Size operation to the graph

Parameters:

input -

outType -

Returns:

a new instance of Size

See Also:

Size

tensorArraySize

public TensorArraySize tensorArraySize(Operand<?> handle,
                                       Operand<Float> flowIn)

Adds an TensorArraySize operation to the graph

Parameters:

handle - The handle to a TensorArray (output of TensorArray or TensorArrayGrad).

flowIn - A float scalar that enforces proper chaining of operations.

Returns:

a new instance of TensorArraySize

See Also:

TensorArraySize

tensorArrayWrite

public <T> TensorArrayWrite tensorArrayWrite(Operand<?> handle,
                                             Operand<Integer> index,
                                             Operand<T> value,
                                             Operand<Float> flowIn)

Adds an TensorArrayWrite operation to the graph

Parameters:

handle - The handle to a TensorArray.

index - The position to write to inside the TensorArray.

value - The tensor to write to the TensorArray.

flowIn - A float scalar that enforces proper chaining of operations.

Returns:

a new instance of TensorArrayWrite

See Also:

TensorArrayWrite

where

public <T> Where where(Operand<T> condition)

Adds an Where operation to the graph

Parameters:

condition -

Returns:

a new instance of Where

See Also:

Where

sparseSegmentSqrtNWithNumSegments

public <T extends Number,U extends Number,V extends Number> SparseSegmentSqrtNWithNumSegments<T> sparseSegmentSqrtNWithNumSegments(Operand<T> data,
                                                                                                                                   Operand<U> indices,
                                                                                                                                   Operand<Integer> segmentIds,
                                                                                                                                   Operand<V> numSegments)

Adds an SparseSegmentSqrtNWithNumSegments operation to the graph

Parameters:

data -

indices - A 1-D tensor. Has same rank as `segment_ids`.

segmentIds - A 1-D tensor. Values should be sorted and can be repeated.

numSegments - Should equal the number of distinct segment IDs.

Returns:

a new instance of SparseSegmentSqrtNWithNumSegments

See Also:

SparseSegmentSqrtNWithNumSegments

splitV

public <T,U extends Number> SplitV<T> splitV(Operand<T> value,
                                             Operand<U> sizeSplits,
                                             Operand<Integer> axis,
                                             Long numSplit)

Adds an SplitV operation to the graph

Parameters:

value - The tensor to split.

sizeSplits - list containing the sizes of each output tensor along the split

axis - 0-D. The dimension along which to split. Must be in the range

numSplit -

Returns:

a new instance of SplitV

See Also:

SplitV

tensorScatterUpdate

public <T,U extends Number> TensorScatterUpdate<T> tensorScatterUpdate(Operand<T> tensor,
                                                                       Operand<U> indices,
                                                                       Operand<T> updates)

Adds an TensorScatterUpdate operation to the graph

Parameters:

tensor - Tensor to copy/update.

indices - Index tensor.

updates - Updates to scatter into output.

Returns:

a new instance of TensorScatterUpdate

See Also:

TensorScatterUpdate

orderedMapSize

public OrderedMapSize orderedMapSize(List<Class<?>> dtypes,
                                     OrderedMapSize.Options... options)

Adds an OrderedMapSize operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of OrderedMapSize

See Also:

OrderedMapSize

queueSize

public QueueSize queueSize(Operand<?> handle)

Adds an QueueSize operation to the graph

Parameters:

handle - The handle to a queue.

Returns:

a new instance of QueueSize

See Also:

QueueSize

depthToSpace

public <T> DepthToSpace<T> depthToSpace(Operand<T> input,
                                        Long blockSize,
                                        DepthToSpace.Options... options)

Adds an DepthToSpace operation to the graph

Parameters:

input -

blockSize - The size of the spatial block, same as in Space2Depth.

options - carries optional attributes values

Returns:

a new instance of DepthToSpace

See Also:

DepthToSpace

decodeBmp

public DecodeBmp decodeBmp(Operand<String> contents,
                           DecodeBmp.Options... options)

Adds an DecodeBmp operation to the graph

Parameters:

contents - 0-D. The BMP-encoded image.

options - carries optional attributes values

Returns:

a new instance of DecodeBmp

See Also:

DecodeBmp

less

public <T extends Number> Less less(Operand<T> x,
                                    Operand<T> y)

Adds an Less operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Less

See Also:

Less

resourceApplyAdam

public <T> ResourceApplyAdam resourceApplyAdam(Operand<?> var,
                                               Operand<?> m,
                                               Operand<?> v,
                                               Operand<T> beta1Power,
                                               Operand<T> beta2Power,
                                               Operand<T> lr,
                                               Operand<T> beta1,
                                               Operand<T> beta2,
                                               Operand<T> epsilon,
                                               Operand<T> grad,
                                               ResourceApplyAdam.Options... options)

Adds an ResourceApplyAdam operation to the graph

Parameters:

var - Should be from a Variable().

m - Should be from a Variable().

v - Should be from a Variable().

beta1Power - Must be a scalar.

beta2Power - Must be a scalar.

lr - Scaling factor. Must be a scalar.

beta1 - Momentum factor. Must be a scalar.

beta2 - Momentum factor. Must be a scalar.

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyAdam

See Also:

ResourceApplyAdam

floor

public <T extends Number> Floor<T> floor(Operand<T> x)

Adds an Floor operation to the graph

Parameters:

x -

Returns:

a new instance of Floor

See Also:

Floor

resourceApplyMomentum

public <T> ResourceApplyMomentum resourceApplyMomentum(Operand<?> var,
                                                       Operand<?> accum,
                                                       Operand<T> lr,
                                                       Operand<T> grad,
                                                       Operand<T> momentum,
                                                       ResourceApplyMomentum.Options... options)

Adds an ResourceApplyMomentum operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

grad - The gradient.

momentum - Momentum. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyMomentum

See Also:

ResourceApplyMomentum

sparseReduceMax

public <T extends Number> SparseReduceMax<T> sparseReduceMax(Operand<Long> inputIndices,
                                                             Operand<T> inputValues,
                                                             Operand<Long> inputShape,
                                                             Operand<Integer> reductionAxes,
                                                             SparseReduceMax.Options... options)

Adds an SparseReduceMax operation to the graph

Parameters:

inputIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

inputValues - 1-D. `N` non-empty values corresponding to `input_indices`.

inputShape - 1-D. Shape of the input SparseTensor.

reductionAxes - 1-D. Length-`K` vector containing the reduction axes.

options - carries optional attributes values

Returns:

a new instance of SparseReduceMax

See Also:

SparseReduceMax

placeholder

public <T> Placeholder<T> placeholder(Class<T> dtype,
                                      Placeholder.Options... options)

Adds an Placeholder operation to the graph

Parameters:

dtype - The type of elements in the tensor.

options - carries optional attributes values

Returns:

a new instance of Placeholder

See Also:

Placeholder

reverseSequence

public <T,U extends Number> ReverseSequence<T> reverseSequence(Operand<T> input,
                                                               Operand<U> seqLengths,
                                                               Long seqDim,
                                                               ReverseSequence.Options... options)

Adds an ReverseSequence operation to the graph

Parameters:

input - The input to reverse.

seqLengths - 1-D with length `input.dims(batch_dim)` and

seqDim - The dimension which is partially reversed.

options - carries optional attributes values

Returns:

a new instance of ReverseSequence

See Also:

ReverseSequence

fusedBatchNormV2

public <T extends Number,U extends Number> FusedBatchNormV2<T,U> fusedBatchNormV2(Operand<T> x,
                                                                                  Operand<U> scale,
                                                                                  Operand<U> offset,
                                                                                  Operand<U> mean,
                                                                                  Operand<U> variance,
                                                                                  FusedBatchNormV2.Options... options)

Adds an FusedBatchNormV2 operation to the graph

Parameters:

x - A 4D Tensor for input data.

scale - A 1D Tensor for scaling factor, to scale the normalized x.

offset - A 1D Tensor for offset, to shift to the normalized x.

mean - A 1D Tensor for population mean. Used for inference only;

variance - A 1D Tensor for population variance. Used for inference only;

options - carries optional attributes values

Returns:

a new instance of FusedBatchNormV2

See Also:

FusedBatchNormV2

tensorListGather

public <T> TensorListGather<T> tensorListGather(Operand<?> inputHandle,
                                                Operand<Integer> indices,
                                                Class<T> elementDtype)

Adds an TensorListGather operation to the graph

Parameters:

inputHandle -

indices -

elementDtype -

Returns:

a new instance of TensorListGather

See Also:

TensorListGather

maxPool3DGrad

public <U extends Number,T extends Number> MaxPool3DGrad<U> maxPool3DGrad(Operand<T> origInput,
                                                                          Operand<T> origOutput,
                                                                          Operand<U> grad,
                                                                          List<Long> ksize,
                                                                          List<Long> strides,
                                                                          String padding,
                                                                          MaxPool3DGrad.Options... options)

Adds an MaxPool3DGrad operation to the graph

Parameters:

origInput - The original input tensor.

origOutput - The original output tensor.

grad - Output backprop of shape `[batch, depth, rows, cols, channels]`.

ksize - 1-D tensor of length 5. The size of the window for each dimension of

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of MaxPool3DGrad

See Also:

MaxPool3DGrad

decodeCSV

public DecodeCSV decodeCSV(Operand<String> records,
                           Iterable<Operand<?>> recordDefaults,
                           DecodeCSV.Options... options)

Adds an DecodeCSV operation to the graph

Parameters:

records - Each string is a record/row in the csv and all records should have

recordDefaults - One tensor per column of the input record, with either a

options - carries optional attributes values

Returns:

a new instance of DecodeCSV

See Also:

DecodeCSV

resourceCountUpTo

public <T extends Number> ResourceCountUpTo<T> resourceCountUpTo(Operand<?> resource,
                                                                 Long limit,
                                                                 Class<T> T)

Adds an ResourceCountUpTo operation to the graph

Parameters:

resource - Should be from a scalar `Variable` node.

limit - If incrementing ref would bring it above limit, instead generates an

T -

Returns:

a new instance of ResourceCountUpTo

See Also:

ResourceCountUpTo

tile

public <T,U extends Number> Tile<T> tile(Operand<T> input,
                                         Operand<U> multiples)

Adds an Tile operation to the graph

Parameters:

input - 1-D or higher.

multiples - 1-D. Length must be the same as the number of dimensions in `input`

Returns:

a new instance of Tile

See Also:

Tile

cholesky

public <T> Cholesky<T> cholesky(Operand<T> input)

Adds an Cholesky operation to the graph

Parameters:

input - Shape is `[..., M, M]`.

Returns:

a new instance of Cholesky

See Also:

Cholesky

conv3DBackpropInputV2

public <U extends Number,T extends Number> Conv3DBackpropInputV2<U> conv3DBackpropInputV2(Operand<T> inputSizes,
                                                                                          Operand<U> filter,
                                                                                          Operand<U> outBackprop,
                                                                                          List<Long> strides,
                                                                                          String padding,
                                                                                          Conv3DBackpropInputV2.Options... options)

Adds an Conv3DBackpropInputV2 operation to the graph

Parameters:

inputSizes - An integer vector representing the tensor shape of `input`,

filter - Shape `[depth, rows, cols, in_channels, out_channels]`.

outBackprop - Backprop signal of shape `[batch, out_depth, out_rows, out_cols,

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of Conv3DBackpropInputV2

See Also:

Conv3DBackpropInputV2

sparseAddGrad

public <T> SparseAddGrad<T> sparseAddGrad(Operand<T> backpropValGrad,
                                          Operand<Long> aIndices,
                                          Operand<Long> bIndices,
                                          Operand<Long> sumIndices)

Adds an SparseAddGrad operation to the graph

Parameters:

backpropValGrad - 1-D with shape `[nnz(sum)]`. The gradient with respect to

aIndices - 2-D. The `indices` of the `SparseTensor` A, size `[nnz(A), ndims]`.

bIndices - 2-D. The `indices` of the `SparseTensor` B, size `[nnz(B), ndims]`.

sumIndices - 2-D. The `indices` of the sum `SparseTensor`, size

Returns:

a new instance of SparseAddGrad

See Also:

SparseAddGrad

rank

public <T> Rank rank(Operand<T> input)

Adds an Rank operation to the graph

Parameters:

input -

Returns:

a new instance of Rank

See Also:

Rank

negate

public <T> Negate<T> negate(Operand<T> x)

Adds an Negate operation to the graph

Parameters:

x -

Returns:

a new instance of Negate

See Also:

Negate

queueEnqueueMany

public QueueEnqueueMany queueEnqueueMany(Operand<?> handle,
                                         Iterable<Operand<?>> components,
                                         QueueEnqueueMany.Options... options)

Adds an QueueEnqueueMany operation to the graph

Parameters:

handle - The handle to a queue.

components - One or more tensors from which the enqueued tensors should

options - carries optional attributes values

Returns:

a new instance of QueueEnqueueMany

See Also:

QueueEnqueueMany

multinomial

public <T extends Number> Multinomial<Long> multinomial(Operand<T> logits,
                                                        Operand<Integer> numSamples,
                                                        Multinomial.Options... options)

Adds an Multinomial operation to the graph

Parameters:

logits - 2-D Tensor with shape `[batch_size, num_classes]`. Each slice `[i, :]`

numSamples - 0-D. Number of independent samples to draw for each row slice.

options - carries optional attributes values

Returns:

a new instance of Multinomial

See Also:

Multinomial

constant

public Constant<String> constant(String data,
                                 Charset charset)

Adds an Constant operation to the graph

Parameters:

charset - The encoding from String to bytes.

data - The string to put into the new constant.

Returns:

a string constant

See Also:

Constant

spaceToDepth

public <T> SpaceToDepth<T> spaceToDepth(Operand<T> input,
                                        Long blockSize,
                                        SpaceToDepth.Options... options)

Adds an SpaceToDepth operation to the graph

Parameters:

input -

blockSize - The size of the spatial block.

options - carries optional attributes values

Returns:

a new instance of SpaceToDepth

See Also:

SpaceToDepth

mapSize

public MapSize mapSize(List<Class<?>> dtypes,
                       MapSize.Options... options)

Adds an MapSize operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of MapSize

See Also:

MapSize

constant

public Constant<Integer> constant(int[][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

orderedMapIncompleteSize

public OrderedMapIncompleteSize orderedMapIncompleteSize(List<Class<?>> dtypes,
                                                         OrderedMapIncompleteSize.Options... options)

Adds an OrderedMapIncompleteSize operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of OrderedMapIncompleteSize

See Also:

OrderedMapIncompleteSize

sparseApplyCenteredRMSProp

public <T,U extends Number> SparseApplyCenteredRMSProp<T> sparseApplyCenteredRMSProp(Operand<T> var,
                                                                                     Operand<T> mg,
                                                                                     Operand<T> ms,
                                                                                     Operand<T> mom,
                                                                                     Operand<T> lr,
                                                                                     Operand<T> rho,
                                                                                     Operand<T> momentum,
                                                                                     Operand<T> epsilon,
                                                                                     Operand<T> grad,
                                                                                     Operand<U> indices,
                                                                                     SparseApplyCenteredRMSProp.Options... options)

Adds an SparseApplyCenteredRMSProp operation to the graph

Parameters:

var - Should be from a Variable().

mg - Should be from a Variable().

ms - Should be from a Variable().

mom - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay rate. Must be a scalar.

momentum -

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var, ms and mom.

options - carries optional attributes values

Returns:

a new instance of SparseApplyCenteredRMSProp

See Also:

SparseApplyCenteredRMSProp

randomUniform

public <U extends Number,T extends Number> RandomUniform<U> randomUniform(Operand<T> shape,
                                                                          Class<U> dtype,
                                                                          RandomUniform.Options... options)

Adds an RandomUniform operation to the graph

Parameters:

shape - The shape of the output tensor.

dtype - The type of the output.

options - carries optional attributes values

Returns:

a new instance of RandomUniform

See Also:

RandomUniform

resourceSparseApplyRMSProp

public <T,U extends Number> ResourceSparseApplyRMSProp resourceSparseApplyRMSProp(Operand<?> var,
                                                                                  Operand<?> ms,
                                                                                  Operand<?> mom,
                                                                                  Operand<T> lr,
                                                                                  Operand<T> rho,
                                                                                  Operand<T> momentum,
                                                                                  Operand<T> epsilon,
                                                                                  Operand<T> grad,
                                                                                  Operand<U> indices,
                                                                                  ResourceSparseApplyRMSProp.Options... options)

Adds an ResourceSparseApplyRMSProp operation to the graph

Parameters:

var - Should be from a Variable().

ms - Should be from a Variable().

mom - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay rate. Must be a scalar.

momentum -

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var, ms and mom.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyRMSProp

See Also:

ResourceSparseApplyRMSProp

isVariableInitialized

public <T> IsVariableInitialized isVariableInitialized(Operand<T> ref)

Adds an IsVariableInitialized operation to the graph

Parameters:

ref - Should be from a `Variable` node. May be uninitialized.

Returns:

a new instance of IsVariableInitialized

See Also:

IsVariableInitialized

unsortedSegmentMin

public <T extends Number,U extends Number,V extends Number> UnsortedSegmentMin<T> unsortedSegmentMin(Operand<T> data,
                                                                                                     Operand<U> segmentIds,
                                                                                                     Operand<V> numSegments)

Adds an UnsortedSegmentMin operation to the graph

Parameters:

data -

segmentIds - A tensor whose shape is a prefix of `data.shape`.

numSegments -

Returns:

a new instance of UnsortedSegmentMin

See Also:

UnsortedSegmentMin

igamma

public <T extends Number> Igamma<T> igamma(Operand<T> a,
                                           Operand<T> x)

Adds an Igamma operation to the graph

Parameters:

a -

x -

Returns:

a new instance of Igamma

See Also:

Igamma

rFFT

public RFFT rFFT(Operand<Float> input,
                 Operand<Integer> fftLength)

Adds an RFFT operation to the graph

Parameters:

input - A float32 tensor.

fftLength - An int32 tensor of shape [1]. The FFT length.

Returns:

a new instance of RFFT

See Also:

RFFT

varHandleOp

public <T> VarHandleOp varHandleOp(Class<T> dtype,
                                   Shape shape,
                                   VarHandleOp.Options... options)

Adds an VarHandleOp operation to the graph

Parameters:

dtype - the type of this variable. Must agree with the dtypes

shape - The (possibly partially specified) shape of this variable.

options - carries optional attributes values

Returns:

a new instance of VarHandleOp

See Also:

VarHandleOp

queueDequeue

public QueueDequeue queueDequeue(Operand<?> handle,
                                 List<Class<?>> componentTypes,
                                 QueueDequeue.Options... options)

Adds an QueueDequeue operation to the graph

Parameters:

handle - The handle to a queue.

componentTypes - The type of each component in a tuple.

options - carries optional attributes values

Returns:

a new instance of QueueDequeue

See Also:

QueueDequeue

populationCount

public <T extends Number> PopulationCount populationCount(Operand<T> x)

Adds an PopulationCount operation to the graph

Parameters:

x -

Returns:

a new instance of PopulationCount

See Also:

PopulationCount

resourceApplyAdamWithAmsgrad

public <T> ResourceApplyAdamWithAmsgrad resourceApplyAdamWithAmsgrad(Operand<?> var,
                                                                     Operand<?> m,
                                                                     Operand<?> v,
                                                                     Operand<?> vhat,
                                                                     Operand<T> beta1Power,
                                                                     Operand<T> beta2Power,
                                                                     Operand<T> lr,
                                                                     Operand<T> beta1,
                                                                     Operand<T> beta2,
                                                                     Operand<T> epsilon,
                                                                     Operand<T> grad,
                                                                     ResourceApplyAdamWithAmsgrad.Options... options)

Adds an ResourceApplyAdamWithAmsgrad operation to the graph

Parameters:

var - Should be from a Variable().

m - Should be from a Variable().

v - Should be from a Variable().

vhat - Should be from a Variable().

beta1Power - Must be a scalar.

beta2Power - Must be a scalar.

lr - Scaling factor. Must be a scalar.

beta1 - Momentum factor. Must be a scalar.

beta2 - Momentum factor. Must be a scalar.

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyAdamWithAmsgrad

See Also:

ResourceApplyAdamWithAmsgrad

parseSequenceExample

public ParseSequenceExample parseSequenceExample(Operand<String> serialized,
                                                 Operand<String> debugName,
                                                 Iterable<Operand<?>> contextDenseDefaults,
                                                 List<String> featureListDenseMissingAssumedEmpty,
                                                 List<String> contextSparseKeys,
                                                 List<String> contextDenseKeys,
                                                 List<String> featureListSparseKeys,
                                                 List<String> featureListDenseKeys,
                                                 List<Class<?>> contextSparseTypes,
                                                 List<Class<?>> featureListDenseTypes,
                                                 List<Class<?>> featureListSparseTypes,
                                                 ParseSequenceExample.Options... options)

Adds an ParseSequenceExample operation to the graph

Parameters:

serialized - A vector containing binary serialized SequenceExample protos.

debugName - A vector containing the names of the serialized protos.

contextDenseDefaults - A list of Ncontext_dense Tensors (some may be empty).

featureListDenseMissingAssumedEmpty - A vector listing the

contextSparseKeys - A list of Ncontext_sparse string Tensors (scalars).

contextDenseKeys - A list of Ncontext_dense string Tensors (scalars).

featureListSparseKeys - A list of Nfeature_list_sparse string Tensors

featureListDenseKeys - A list of Nfeature_list_dense string Tensors (scalars).

contextSparseTypes - A list of Ncontext_sparse types; the data types of data in

featureListDenseTypes -

featureListSparseTypes - A list of Nfeature_list_sparse types; the data types

options - carries optional attributes values

Returns:

a new instance of ParseSequenceExample

See Also:

ParseSequenceExample

multinomial

public <U extends Number,T extends Number> Multinomial<U> multinomial(Operand<T> logits,
                                                                      Operand<Integer> numSamples,
                                                                      Class<U> outputDtype,
                                                                      Multinomial.Options... options)

Adds an Multinomial operation to the graph

Parameters:

logits - 2-D Tensor with shape `[batch_size, num_classes]`. Each slice `[i, :]`

numSamples - 0-D. Number of independent samples to draw for each row slice.

outputDtype -

options - carries optional attributes values

Returns:

a new instance of Multinomial

See Also:

Multinomial

tensorListStack

public <T> TensorListStack<T> tensorListStack(Operand<?> inputHandle,
                                              Class<T> elementDtype,
                                              TensorListStack.Options... options)

Adds an TensorListStack operation to the graph

Parameters:

inputHandle -

elementDtype -

options - carries optional attributes values

Returns:

a new instance of TensorListStack

See Also:

TensorListStack

tensorListFromTensor

public <T,U extends Number> TensorListFromTensor tensorListFromTensor(Operand<T> tensor,
                                                                      Operand<U> elementShape)

Adds an TensorListFromTensor operation to the graph

Parameters:

tensor -

elementShape -

Returns:

a new instance of TensorListFromTensor

See Also:

TensorListFromTensor

constant

public Constant<Long> constant(long[][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

scatterSub

public <T,U extends Number> ScatterSub<T> scatterSub(Operand<T> ref,
                                                     Operand<U> indices,
                                                     Operand<T> updates,
                                                     ScatterSub.Options... options)

Adds an ScatterSub operation to the graph

Parameters:

ref - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to subtract from `ref`.

options - carries optional attributes values

Returns:

a new instance of ScatterSub

See Also:

ScatterSub

rint

public <T extends Number> Rint<T> rint(Operand<T> x)

Adds an Rint operation to the graph

Parameters:

x -

Returns:

a new instance of Rint

See Also:

Rint

quantizedAdd

public <V,T,U> QuantizedAdd<V> quantizedAdd(Operand<T> x,
                                            Operand<U> y,
                                            Operand<Float> minX,
                                            Operand<Float> maxX,
                                            Operand<Float> minY,
                                            Operand<Float> maxY,
                                            Class<V> Toutput)

Adds an QuantizedAdd operation to the graph

Parameters:

x -

y -

minX - The float value that the lowest quantized `x` value represents.

maxX - The float value that the highest quantized `x` value represents.

minY - The float value that the lowest quantized `y` value represents.

maxY - The float value that the highest quantized `y` value represents.

Toutput -

Returns:

a new instance of QuantizedAdd

See Also:

QuantizedAdd

stopGradient

public <T> StopGradient<T> stopGradient(Operand<T> input)

Adds an StopGradient operation to the graph

Parameters:

input -

Returns:

a new instance of StopGradient

See Also:

StopGradient

assignSub

public <T> AssignSub<T> assignSub(Operand<T> ref,
                                  Operand<T> value,
                                  AssignSub.Options... options)

Adds an AssignSub operation to the graph

Parameters:

ref - Should be from a `Variable` node.

value - The value to be subtracted to the variable.

options - carries optional attributes values

Returns:

a new instance of AssignSub

See Also:

AssignSub

avgPool3D

public <T extends Number> AvgPool3D<T> avgPool3D(Operand<T> input,
                                                 List<Long> ksize,
                                                 List<Long> strides,
                                                 String padding,
                                                 AvgPool3D.Options... options)

Adds an AvgPool3D operation to the graph

Parameters:

input - Shape `[batch, depth, rows, cols, channels]` tensor to pool over.

ksize - 1-D tensor of length 5. The size of the window for each dimension of

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of AvgPool3D

See Also:

AvgPool3D

inTopK

public <T extends Number> InTopK inTopK(Operand<Float> predictions,
                                        Operand<T> targets,
                                        Long k)

Adds an InTopK operation to the graph

Parameters:

predictions - A `batch_size` x `classes` tensor.

targets - A `batch_size` vector of class ids.

k - Number of top elements to look at for computing precision.

Returns:

a new instance of InTopK

See Also:

InTopK

restore

public <T> Restore<T> restore(Operand<String> filePattern,
                              Operand<String> tensorName,
                              Class<T> dt,
                              Restore.Options... options)

Adds an Restore operation to the graph

Parameters:

filePattern - Must have a single element. The pattern of the files from

tensorName - Must have a single element. The name of the tensor to be

dt - The type of the tensor to be restored.

options - carries optional attributes values

Returns:

a new instance of Restore

See Also:

Restore

matrixTriangularSolve

public <T> MatrixTriangularSolve<T> matrixTriangularSolve(Operand<T> matrix,
                                                          Operand<T> rhs,
                                                          MatrixTriangularSolve.Options... options)

Adds an MatrixTriangularSolve operation to the graph

Parameters:

matrix - Shape is `[..., M, M]`.

rhs - Shape is `[..., M, K]`.

options - carries optional attributes values

Returns:

a new instance of MatrixTriangularSolve

See Also:

MatrixTriangularSolve

batchSelfAdjointEig

public <T extends Number> BatchSelfAdjointEig<T> batchSelfAdjointEig(Operand<T> input)

Adds an BatchSelfAdjointEig operation to the graph

Parameters:

input -

Returns:

a new instance of BatchSelfAdjointEig

See Also:

BatchSelfAdjointEig

matrixDiag

public <T> MatrixDiag<T> matrixDiag(Operand<T> diagonal)

Adds an MatrixDiag operation to the graph

Parameters:

diagonal - Rank `k`, where `k >= 1`.

Returns:

a new instance of MatrixDiag

See Also:

MatrixDiag

constant

public Constant<Boolean> constant(boolean[][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

audioSummary

public AudioSummary audioSummary(Operand<String> tag,
                                 Operand<Float> tensor,
                                 Operand<Float> sampleRate,
                                 AudioSummary.Options... options)

Adds an AudioSummary operation to the graph

Parameters:

tag - Scalar. Used to build the `tag` attribute of the summary values.

tensor - 2-D of shape `[batch_size, frames]`.

sampleRate - The sample rate of the signal in hertz.

options - carries optional attributes values

Returns:

a new instance of AudioSummary

See Also:

AudioSummary

allCandidateSampler

public AllCandidateSampler allCandidateSampler(Operand<Long> trueClasses,
                                               Long numTrue,
                                               Long numSampled,
                                               Boolean unique,
                                               AllCandidateSampler.Options... options)

Adds an AllCandidateSampler operation to the graph

Parameters:

trueClasses - A batch_size * num_true matrix, in which each row contains the

numTrue - Number of true labels per context.

numSampled - Number of candidates to produce.

unique - If unique is true, we sample with rejection, so that all sampled

options - carries optional attributes values

Returns:

a new instance of AllCandidateSampler

See Also:

AllCandidateSampler

extractVolumePatches

public <T extends Number> ExtractVolumePatches<T> extractVolumePatches(Operand<T> input,
                                                                       List<Long> ksizes,
                                                                       List<Long> strides,
                                                                       String padding)

Adds an ExtractVolumePatches operation to the graph

Parameters:

input - 5-D Tensor with shape `[batch, in_planes, in_rows, in_cols, depth]`.

ksizes - The size of the sliding window for each dimension of `input`.

strides - 1-D of length 5. How far the centers of two consecutive patches are in

padding - The type of padding algorithm to use.

Returns:

a new instance of ExtractVolumePatches

See Also:

ExtractVolumePatches

decodePng

public <T extends Number> DecodePng<T> decodePng(Operand<String> contents,
                                                 Class<T> dtype,
                                                 DecodePng.Options... options)

Adds an DecodePng operation to the graph

Parameters:

contents - 0-D. The PNG-encoded image.

dtype -

options - carries optional attributes values

Returns:

a new instance of DecodePng

See Also:

DecodePng

applyPowerSign

public <T> ApplyPowerSign<T> applyPowerSign(Operand<T> var,
                                            Operand<T> m,
                                            Operand<T> lr,
                                            Operand<T> logbase,
                                            Operand<T> signDecay,
                                            Operand<T> beta,
                                            Operand<T> grad,
                                            ApplyPowerSign.Options... options)

Adds an ApplyPowerSign operation to the graph

Parameters:

var - Should be from a Variable().

m - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

logbase - Must be a scalar.

signDecay - Must be a scalar.

beta - Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ApplyPowerSign

See Also:

ApplyPowerSign

decodePng

public DecodePng<UInt8> decodePng(Operand<String> contents,
                                  DecodePng.Options... options)

Adds an DecodePng operation to the graph

Parameters:

contents - 0-D. The PNG-encoded image.

options - carries optional attributes values

Returns:

a new instance of DecodePng

See Also:

DecodePng

variableShape

public <T extends Number> VariableShape<T> variableShape(Operand<?> input,
                                                         Class<T> outType)

Adds an VariableShape operation to the graph

Parameters:

input -

outType -

Returns:

a new instance of VariableShape

See Also:

VariableShape

extractImagePatches

public <T extends Number> ExtractImagePatches<T> extractImagePatches(Operand<T> images,
                                                                     List<Long> ksizes,
                                                                     List<Long> strides,
                                                                     List<Long> rates,
                                                                     String padding)

Adds an ExtractImagePatches operation to the graph

Parameters:

images - 4-D Tensor with shape `[batch, in_rows, in_cols, depth]`.

ksizes - The size of the sliding window for each dimension of `images`.

strides - 1-D of length 4. How far the centers of two consecutive patches are in

rates - 1-D of length 4. Must be: `[1, rate_rows, rate_cols, 1]`. This is the

padding - The type of padding algorithm to use.

Returns:

a new instance of ExtractImagePatches

See Also:

ExtractImagePatches

quantizeDownAndShrinkRange

public <U,T> QuantizeDownAndShrinkRange<U> quantizeDownAndShrinkRange(Operand<T> input,
                                                                      Operand<Float> inputMin,
                                                                      Operand<Float> inputMax,
                                                                      Class<U> outType)

Adds an QuantizeDownAndShrinkRange operation to the graph

Parameters:

input -

inputMin - The float value that the minimum quantized input value represents.

inputMax - The float value that the maximum quantized input value represents.

outType - The type of the output. Should be a lower bit depth than Tinput.

Returns:

a new instance of QuantizeDownAndShrinkRange

See Also:

QuantizeDownAndShrinkRange

decodeJSONExample

public DecodeJSONExample decodeJSONExample(Operand<String> jsonExamples)

Adds an DecodeJSONExample operation to the graph

Parameters:

jsonExamples - Each string is a JSON object serialized according to the JSON

Returns:

a new instance of DecodeJSONExample

See Also:

DecodeJSONExample

bincount

public <T extends Number> Bincount<T> bincount(Operand<Integer> arr,
                                               Operand<Integer> size,
                                               Operand<T> weights)

Adds an Bincount operation to the graph

Parameters:

arr - int32 `Tensor`.

size - non-negative int32 scalar `Tensor`.

weights - is an int32, int64, float32, or float64 `Tensor` with the same

Returns:

a new instance of Bincount

See Also:

Bincount

matrixBandPart

public <T,U extends Number> MatrixBandPart<T> matrixBandPart(Operand<T> input,
                                                             Operand<U> numLower,
                                                             Operand<U> numUpper)

Adds an MatrixBandPart operation to the graph

Parameters:

input - Rank `k` tensor.

numLower - 0-D tensor. Number of subdiagonals to keep. If negative, keep entire

numUpper - 0-D tensor. Number of superdiagonals to keep. If negative, keep

Returns:

a new instance of MatrixBandPart

See Also:

MatrixBandPart

quantizedBatchNormWithGlobalNormalization

public <U,T> QuantizedBatchNormWithGlobalNormalization<U> quantizedBatchNormWithGlobalNormalization(Operand<T> t,
                                                                                                    Operand<Float> tMin,
                                                                                                    Operand<Float> tMax,
                                                                                                    Operand<T> m,
                                                                                                    Operand<Float> mMin,
                                                                                                    Operand<Float> mMax,
                                                                                                    Operand<T> v,
                                                                                                    Operand<Float> vMin,
                                                                                                    Operand<Float> vMax,
                                                                                                    Operand<T> beta,
                                                                                                    Operand<Float> betaMin,
                                                                                                    Operand<Float> betaMax,
                                                                                                    Operand<T> gamma,
                                                                                                    Operand<Float> gammaMin,
                                                                                                    Operand<Float> gammaMax,
                                                                                                    Class<U> outType,
                                                                                                    Float varianceEpsilon,
                                                                                                    Boolean scaleAfterNormalization)

Adds an QuantizedBatchNormWithGlobalNormalization operation to the graph

Parameters:

t - A 4D input Tensor.

tMin - The value represented by the lowest quantized input.

tMax - The value represented by the highest quantized input.

m - A 1D mean Tensor with size matching the last dimension of t.

mMin - The value represented by the lowest quantized mean.

mMax - The value represented by the highest quantized mean.

v - A 1D variance Tensor with size matching the last dimension of t.

vMin - The value represented by the lowest quantized variance.

vMax - The value represented by the highest quantized variance.

beta - A 1D beta Tensor with size matching the last dimension of t.

betaMin - The value represented by the lowest quantized offset.

betaMax - The value represented by the highest quantized offset.

gamma - A 1D gamma Tensor with size matching the last dimension of t.

gammaMin - The value represented by the lowest quantized gamma.

gammaMax - The value represented by the highest quantized gamma.

outType -

varianceEpsilon - A small float number to avoid dividing by 0.

scaleAfterNormalization - A bool indicating whether the resulted tensor

Returns:

a new instance of QuantizedBatchNormWithGlobalNormalization

See Also:

QuantizedBatchNormWithGlobalNormalization

resourceSparseApplyAdagradDA

public <T,U extends Number> ResourceSparseApplyAdagradDA resourceSparseApplyAdagradDA(Operand<?> var,
                                                                                      Operand<?> gradientAccumulator,
                                                                                      Operand<?> gradientSquaredAccumulator,
                                                                                      Operand<T> grad,
                                                                                      Operand<U> indices,
                                                                                      Operand<T> lr,
                                                                                      Operand<T> l1,
                                                                                      Operand<T> l2,
                                                                                      Operand<Long> globalStep,
                                                                                      ResourceSparseApplyAdagradDA.Options... options)

Adds an ResourceSparseApplyAdagradDA operation to the graph

Parameters:

var - Should be from a Variable().

gradientAccumulator - Should be from a Variable().

gradientSquaredAccumulator - Should be from a Variable().

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

lr - Learning rate. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

globalStep - Training step number. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyAdagradDA

See Also:

ResourceSparseApplyAdagradDA

mapIncompleteSize

public MapIncompleteSize mapIncompleteSize(List<Class<?>> dtypes,
                                           MapIncompleteSize.Options... options)

Adds an MapIncompleteSize operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of MapIncompleteSize

See Also:

MapIncompleteSize

stridedSliceGrad

public <U,T extends Number> StridedSliceGrad<U> stridedSliceGrad(Operand<T> shape,
                                                                 Operand<T> begin,
                                                                 Operand<T> end,
                                                                 Operand<T> strides,
                                                                 Operand<U> dy,
                                                                 StridedSliceGrad.Options... options)

Adds an StridedSliceGrad operation to the graph

Parameters:

shape -

begin -

end -

strides -

dy -

options - carries optional attributes values

Returns:

a new instance of StridedSliceGrad

See Also:

StridedSliceGrad

maxPoolGradV2

public <T extends Number> MaxPoolGradV2<T> maxPoolGradV2(Operand<T> origInput,
                                                         Operand<T> origOutput,
                                                         Operand<T> grad,
                                                         Operand<Integer> ksize,
                                                         Operand<Integer> strides,
                                                         String padding,
                                                         MaxPoolGradV2.Options... options)

Adds an MaxPoolGradV2 operation to the graph

Parameters:

origInput - The original input tensor.

origOutput - The original output tensor.

grad - 4-D. Gradients w.r.t. the output of `max_pool`.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of MaxPoolGradV2

See Also:

MaxPoolGradV2

stridedSlice

public <T,U extends Number> StridedSlice<T> stridedSlice(Operand<T> input,
                                                         Operand<U> begin,
                                                         Operand<U> end,
                                                         Operand<U> strides,
                                                         StridedSlice.Options... options)

Adds an StridedSlice operation to the graph

Parameters:

input -

begin - `begin[k]` specifies the offset into the `k`th range specification.

end - `end[i]` is like `begin` with the exception that `end_mask` is

strides - `strides[i]` specifies the increment in the `i`th specification

options - carries optional attributes values

Returns:

a new instance of StridedSlice

See Also:

StridedSlice

depthwiseConv2dNative

public <T extends Number> DepthwiseConv2dNative<T> depthwiseConv2dNative(Operand<T> input,
                                                                         Operand<T> filter,
                                                                         List<Long> strides,
                                                                         String padding,
                                                                         DepthwiseConv2dNative.Options... options)

Adds an DepthwiseConv2dNative operation to the graph

Parameters:

input -

filter -

strides - 1-D of length 4. The stride of the sliding window for each dimension

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of DepthwiseConv2dNative

See Also:

DepthwiseConv2dNative

resourceScatterMul

public <T extends Number,U> ResourceScatterMul resourceScatterMul(Operand<?> resource,
                                                                  Operand<T> indices,
                                                                  Operand<U> updates)

Adds an ResourceScatterMul operation to the graph

Parameters:

resource - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to add to `ref`.

Returns:

a new instance of ResourceScatterMul

See Also:

ResourceScatterMul

controlTrigger

public ControlTrigger controlTrigger()

Adds an ControlTrigger operation to the graph

Returns:

a new instance of ControlTrigger

See Also:

ControlTrigger

cudnnRNNBackprop

public <T extends Number> CudnnRNNBackprop<T> cudnnRNNBackprop(Operand<T> input,
                                                               Operand<T> inputH,
                                                               Operand<T> inputC,
                                                               Operand<T> params,
                                                               Operand<T> output,
                                                               Operand<T> outputH,
                                                               Operand<T> outputC,
                                                               Operand<T> outputBackprop,
                                                               Operand<T> outputHBackprop,
                                                               Operand<T> outputCBackprop,
                                                               Operand<T> reserveSpace,
                                                               CudnnRNNBackprop.Options... options)

Adds an CudnnRNNBackprop operation to the graph

Parameters:

input -

inputH -

inputC -

params -

output -

outputH -

outputC -

outputBackprop -

outputHBackprop -

outputCBackprop -

reserveSpace -

options - carries optional attributes values

Returns:

a new instance of CudnnRNNBackprop

See Also:

CudnnRNNBackprop

shape

public <U extends Number,T> Shape<U> shape(Operand<T> input,
                                           Class<U> outType)

Adds an Shape operation to the graph

Parameters:

input -

outType -

Returns:

a new instance of Shape

See Also:

Shape

tensorListPushBack

public <T> TensorListPushBack tensorListPushBack(Operand<?> inputHandle,
                                                 Operand<T> tensor)

Adds an TensorListPushBack operation to the graph

Parameters:

inputHandle -

tensor -

Returns:

a new instance of TensorListPushBack

See Also:

TensorListPushBack

identityN

public IdentityN identityN(Iterable<Operand<?>> input)

Adds an IdentityN operation to the graph

Parameters:

input -

Returns:

a new instance of IdentityN

See Also:

IdentityN

reciprocal

public <T> Reciprocal<T> reciprocal(Operand<T> x)

Adds an Reciprocal operation to the graph

Parameters:

x -

Returns:

a new instance of Reciprocal

See Also:

Reciprocal

denseToDenseSetOperation

public <T> DenseToDenseSetOperation<T> denseToDenseSetOperation(Operand<T> set1,
                                                                Operand<T> set2,
                                                                String setOperation,
                                                                DenseToDenseSetOperation.Options... options)

Adds an DenseToDenseSetOperation operation to the graph

Parameters:

set1 - `Tensor` with rank `n`. 1st `n-1` dimensions must be the same as `set2`.

set2 - `Tensor` with rank `n`. 1st `n-1` dimensions must be the same as `set1`.

setOperation -

options - carries optional attributes values

Returns:

a new instance of DenseToDenseSetOperation

See Also:

DenseToDenseSetOperation

reshape

public <T,U extends Number> Reshape<T> reshape(Operand<T> tensor,
                                               Operand<U> shape)

Adds an Reshape operation to the graph

Parameters:

tensor -

shape - Defines the shape of the output tensor.

Returns:

a new instance of Reshape

See Also:

Reshape

sparseSparseMaximum

public <T extends Number> SparseSparseMaximum<T> sparseSparseMaximum(Operand<Long> aIndices,
                                                                     Operand<T> aValues,
                                                                     Operand<Long> aShape,
                                                                     Operand<Long> bIndices,
                                                                     Operand<T> bValues,
                                                                     Operand<Long> bShape)

Adds an SparseSparseMaximum operation to the graph

Parameters:

aIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

aValues - 1-D. `N` non-empty values corresponding to `a_indices`.

aShape - 1-D. Shape of the input SparseTensor.

bIndices - counterpart to `a_indices` for the other operand.

bValues - counterpart to `a_values` for the other operand; must be of the same dtype.

bShape - counterpart to `a_shape` for the other operand; the two shapes must be equal.

Returns:

a new instance of SparseSparseMaximum

See Also:

SparseSparseMaximum

resourceApplyRMSProp

public <T> ResourceApplyRMSProp resourceApplyRMSProp(Operand<?> var,
                                                     Operand<?> ms,
                                                     Operand<?> mom,
                                                     Operand<T> lr,
                                                     Operand<T> rho,
                                                     Operand<T> momentum,
                                                     Operand<T> epsilon,
                                                     Operand<T> grad,
                                                     ResourceApplyRMSProp.Options... options)

Adds an ResourceApplyRMSProp operation to the graph

Parameters:

var - Should be from a Variable().

ms - Should be from a Variable().

mom - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay rate. Must be a scalar.

momentum -

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyRMSProp

See Also:

ResourceApplyRMSProp

destroyResourceOp

public DestroyResourceOp destroyResourceOp(Operand<?> resource,
                                           DestroyResourceOp.Options... options)

Adds an DestroyResourceOp operation to the graph

Parameters:

resource - handle to the resource to delete.

options - carries optional attributes values

Returns:

a new instance of DestroyResourceOp

See Also:

DestroyResourceOp

drawBoundingBoxes

public <T extends Number> DrawBoundingBoxes<T> drawBoundingBoxes(Operand<T> images,
                                                                 Operand<Float> boxes)

Adds an DrawBoundingBoxes operation to the graph

Parameters:

images - 4-D with shape `[batch, height, width, depth]`. A batch of images.

boxes - 3-D with shape `[batch, num_bounding_boxes, 4]` containing bounding

Returns:

a new instance of DrawBoundingBoxes

See Also:

DrawBoundingBoxes

lookupTableExport

public <T,U> LookupTableExport<T,U> lookupTableExport(Operand<?> tableHandle,
                                                      Class<T> Tkeys,
                                                      Class<U> Tvalues)

Adds an LookupTableExport operation to the graph

Parameters:

tableHandle - Handle to the table.

Tkeys -

Tvalues -

Returns:

a new instance of LookupTableExport

See Also:

LookupTableExport

sparseTensorDenseAdd

public <U,T extends Number> SparseTensorDenseAdd<U> sparseTensorDenseAdd(Operand<T> aIndices,
                                                                         Operand<U> aValues,
                                                                         Operand<T> aShape,
                                                                         Operand<U> b)

Adds an SparseTensorDenseAdd operation to the graph

Parameters:

aIndices - 2-D. The `indices` of the `SparseTensor`, with shape `[nnz, ndims]`.

aValues - 1-D. The `values` of the `SparseTensor`, with shape `[nnz]`.

aShape - 1-D. The `shape` of the `SparseTensor`, with shape `[ndims]`.

b - `ndims`-D Tensor. With shape `a_shape`.

Returns:

a new instance of SparseTensorDenseAdd

See Also:

SparseTensorDenseAdd

unsortedSegmentMax

public <T extends Number,U extends Number,V extends Number> UnsortedSegmentMax<T> unsortedSegmentMax(Operand<T> data,
                                                                                                     Operand<U> segmentIds,
                                                                                                     Operand<V> numSegments)

Adds an UnsortedSegmentMax operation to the graph

Parameters:

data -

segmentIds - A tensor whose shape is a prefix of `data.shape`.END

numSegments -

Returns:

a new instance of UnsortedSegmentMax

See Also:

UnsortedSegmentMax

stringToNumber

public <T extends Number> StringToNumber<T> stringToNumber(Operand<String> stringTensor,
                                                           Class<T> outType)

Adds an StringToNumber operation to the graph

Parameters:

stringTensor -

outType - The numeric type to interpret each string in `string_tensor` as.

Returns:

a new instance of StringToNumber

See Also:

StringToNumber

tensorArrayRead

public <T> TensorArrayRead<T> tensorArrayRead(Operand<?> handle,
                                              Operand<Integer> index,
                                              Operand<Float> flowIn,
                                              Class<T> dtype)

Adds an TensorArrayRead operation to the graph

Parameters:

handle - The handle to a TensorArray.

index -

flowIn - A float scalar that enforces proper chaining of operations.

dtype - The type of the elem that is returned.

Returns:

a new instance of TensorArrayRead

See Also:

TensorArrayRead

quantizedMaxPool

public <T> QuantizedMaxPool<T> quantizedMaxPool(Operand<T> input,
                                                Operand<Float> minInput,
                                                Operand<Float> maxInput,
                                                List<Long> ksize,
                                                List<Long> strides,
                                                String padding)

Adds an QuantizedMaxPool operation to the graph

Parameters:

input - The 4D (batch x rows x cols x depth) Tensor to MaxReduce over.

minInput - The float value that the lowest quantized input value represents.

maxInput - The float value that the highest quantized input value represents.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the input

padding - The type of padding algorithm to use.

Returns:

a new instance of QuantizedMaxPool

See Also:

QuantizedMaxPool

serializeIterator

public SerializeIterator serializeIterator(Operand<?> resourceHandle)

Adds an SerializeIterator operation to the graph

Parameters:

resourceHandle - A handle to an iterator resource.

Returns:

a new instance of SerializeIterator

See Also:

SerializeIterator

constant

public Constant<Float> constant(long[] shape,
                                FloatBuffer data)

Adds an Constant operation to the graph

Parameters:

shape - the tensor shape.

data - a buffer containing the tensor data.

Returns:

a float constant

Throws:

IllegalArgumentException - If the tensor shape is not compatible with the buffer

See Also:

Constant

cropAndResizeGradImage

public <T extends Number> CropAndResizeGradImage<T> cropAndResizeGradImage(Operand<Float> grads,
                                                                           Operand<Float> boxes,
                                                                           Operand<Integer> boxInd,
                                                                           Operand<Integer> imageSize,
                                                                           Class<T> T,
                                                                           CropAndResizeGradImage.Options... options)

Adds an CropAndResizeGradImage operation to the graph

Parameters:

grads - A 4-D tensor of shape `[num_boxes, crop_height, crop_width, depth]`.

boxes - A 2-D tensor of shape `[num_boxes, 4]`. The `i`-th row of the tensor

boxInd - A 1-D tensor of shape `[num_boxes]` with int32 values in `[0, batch)`.

imageSize - A 1-D tensor with value `[batch, image_height, image_width, depth]`

T -

options - carries optional attributes values

Returns:

a new instance of CropAndResizeGradImage

See Also:

CropAndResizeGradImage

sparseAccumulatorApplyGradient

public <T> SparseAccumulatorApplyGradient sparseAccumulatorApplyGradient(Operand<String> handle,
                                                                         Operand<Long> localStep,
                                                                         Operand<Long> gradientIndices,
                                                                         Operand<T> gradientValues,
                                                                         Operand<Long> gradientShape,
                                                                         Boolean hasKnownShape)

Adds an SparseAccumulatorApplyGradient operation to the graph

Parameters:

handle - The handle to a accumulator.

localStep - The local_step value at which the sparse gradient was computed.

gradientIndices - Indices of the sparse gradient to be accumulated. Must be a

gradientValues - Values are the non-zero slices of the gradient, and must have

gradientShape - Shape of the sparse gradient to be accumulated.

hasKnownShape - Boolean indicating whether gradient_shape is unknown, in which

Returns:

a new instance of SparseAccumulatorApplyGradient

See Also:

SparseAccumulatorApplyGradient

sparseSegmentSum

public <T extends Number,U extends Number> SparseSegmentSum<T> sparseSegmentSum(Operand<T> data,
                                                                                Operand<U> indices,
                                                                                Operand<Integer> segmentIds)

Adds an SparseSegmentSum operation to the graph

Parameters:

data -

indices - A 1-D tensor. Has same rank as `segment_ids`.

segmentIds - A 1-D tensor. Values should be sorted and can be repeated.

Returns:

a new instance of SparseSegmentSum

See Also:

SparseSegmentSum

queueIsClosed

public QueueIsClosed queueIsClosed(Operand<String> handle)

Adds an QueueIsClosed operation to the graph

Parameters:

handle - The handle to a queue.

Returns:

a new instance of QueueIsClosed

See Also:

QueueIsClosed

lu

public <T,U extends Number> Lu<T,U> lu(Operand<T> input,
                                       Class<U> outputIdxType)

Adds an Lu operation to the graph

Parameters:

input - A tensor of shape `[..., M, M]` whose inner-most 2 dimensions form matrices of

outputIdxType -

Returns:

a new instance of Lu

See Also:

Lu

segmentProd

public <T,U extends Number> SegmentProd<T> segmentProd(Operand<T> data,
                                                       Operand<U> segmentIds)

Adds an SegmentProd operation to the graph

Parameters:

data -

segmentIds - A 1-D tensor whose size is equal to the size of `data`'s

Returns:

a new instance of SegmentProd

See Also:

SegmentProd

iteratorToStringHandle

public IteratorToStringHandle iteratorToStringHandle(Operand<?> resourceHandle)

Adds an IteratorToStringHandle operation to the graph

Parameters:

resourceHandle - A handle to an iterator resource.

Returns:

a new instance of IteratorToStringHandle

See Also:

IteratorToStringHandle

sdcaShrinkL1

public SdcaShrinkL1 sdcaShrinkL1(Iterable<Operand<Float>> weights,
                                 Float l1,
                                 Float l2)

Adds an SdcaShrinkL1 operation to the graph

Parameters:

weights - a list of vectors where each value is the weight associated with a

l1 - Symmetric l1 regularization strength.

l2 - Symmetric l2 regularization strength. Should be a positive float.

Returns:

a new instance of SdcaShrinkL1

See Also:

SdcaShrinkL1

batchMatrixDiagPart

public <T> BatchMatrixDiagPart<T> batchMatrixDiagPart(Operand<T> input)

Adds an BatchMatrixDiagPart operation to the graph

Parameters:

input -

Returns:

a new instance of BatchMatrixDiagPart

See Also:

BatchMatrixDiagPart

prod

public <T,U extends Number> Prod<T> prod(Operand<T> input,
                                         Operand<U> axis,
                                         Prod.Options... options)

Adds an Prod operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of Prod

See Also:

Prod

conv3D

public <T extends Number> Conv3D<T> conv3D(Operand<T> input,
                                           Operand<T> filter,
                                           List<Long> strides,
                                           String padding,
                                           Conv3D.Options... options)

Adds an Conv3D operation to the graph

Parameters:

input - Shape `[batch, in_depth, in_height, in_width, in_channels]`.

filter - Shape `[filter_depth, filter_height, filter_width, in_channels,

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of Conv3D

See Also:

Conv3D

adjustSaturation

public AdjustSaturation adjustSaturation(Operand<Float> images,
                                         Operand<Float> scale)

Adds an AdjustSaturation operation to the graph

Parameters:

images - Images to adjust. At least 3-D.

scale - A float scale to add to the saturation.

Returns:

a new instance of AdjustSaturation

See Also:

AdjustSaturation

logMatrixDeterminant

public <T> LogMatrixDeterminant<T> logMatrixDeterminant(Operand<T> input)

Adds an LogMatrixDeterminant operation to the graph

Parameters:

input - Shape is `[N, M, M]`.

Returns:

a new instance of LogMatrixDeterminant

See Also:

LogMatrixDeterminant

resourceScatterSub

public <T extends Number,U> ResourceScatterSub resourceScatterSub(Operand<?> resource,
                                                                  Operand<T> indices,
                                                                  Operand<U> updates)

Adds an ResourceScatterSub operation to the graph

Parameters:

resource - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to add to `ref`.

Returns:

a new instance of ResourceScatterSub

See Also:

ResourceScatterSub

readerReset

public ReaderReset readerReset(Operand<?> readerHandle)

Adds an ReaderReset operation to the graph

Parameters:

readerHandle - Handle to a Reader.

Returns:

a new instance of ReaderReset

See Also:

ReaderReset

relu

public <T> Relu<T> relu(Operand<T> features)

Adds an Relu operation to the graph

Parameters:

features -

Returns:

a new instance of Relu

See Also:

Relu

hSVToRGB

public <T extends Number> HSVToRGB<T> hSVToRGB(Operand<T> images)

Adds an HSVToRGB operation to the graph

Parameters:

images - 1-D or higher rank. HSV data to convert. Last dimension must be size 3.

Returns:

a new instance of HSVToRGB

See Also:

HSVToRGB

scatterMin

public <T extends Number,U extends Number> ScatterMin<T> scatterMin(Operand<T> ref,
                                                                    Operand<U> indices,
                                                                    Operand<T> updates,
                                                                    ScatterMin.Options... options)

Adds an ScatterMin operation to the graph

Parameters:

ref - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to reduce into `ref`.

options - carries optional attributes values

Returns:

a new instance of ScatterMin

See Also:

ScatterMin

matrixInverse

public <T> MatrixInverse<T> matrixInverse(Operand<T> input,
                                          MatrixInverse.Options... options)

Adds an MatrixInverse operation to the graph

Parameters:

input - Shape is `[..., M, M]`.

options - carries optional attributes values

Returns:

a new instance of MatrixInverse

See Also:

MatrixInverse

stack

public <T> Stack<T> stack(Operand<T> values,
                          Stack.Options... options)

Adds an Stack operation to the graph

Parameters:

values - Must be of same shape and type.

options - carries optional attributes values

Returns:

a new instance of Stack

See Also:

Stack

add

public <T> Add<T> add(Operand<T> x,
                      Operand<T> y)

Adds an Add operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Add

See Also:

Add

cTCGreedyDecoder

public CTCGreedyDecoder cTCGreedyDecoder(Operand<Float> inputs,
                                         Operand<Integer> sequenceLength,
                                         CTCGreedyDecoder.Options... options)

Adds an CTCGreedyDecoder operation to the graph

Parameters:

inputs - 3-D, shape: `(max_time x batch_size x num_classes)`, the logits.

sequenceLength - A vector containing sequence lengths, size `(batch_size)`.

options - carries optional attributes values

Returns:

a new instance of CTCGreedyDecoder

See Also:

CTCGreedyDecoder

lookupTableImport

public <T,U> LookupTableImport lookupTableImport(Operand<?> tableHandle,
                                                 Operand<T> keys,
                                                 Operand<U> values)

Adds an LookupTableImport operation to the graph

Parameters:

tableHandle - Handle to the table.

keys - Any shape. Keys to look up.

values - Values to associate with keys.

Returns:

a new instance of LookupTableImport

See Also:

LookupTableImport

isInf

public <T extends Number> IsInf isInf(Operand<T> x)

Adds an IsInf operation to the graph

Parameters:

x -

Returns:

a new instance of IsInf

See Also:

IsInf

constant

public Constant<Double> constant(double[][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

tensorArraySplit

public <T> TensorArraySplit tensorArraySplit(Operand<?> handle,
                                             Operand<T> value,
                                             Operand<Long> lengths,
                                             Operand<Float> flowIn)

Adds an TensorArraySplit operation to the graph

Parameters:

handle - The handle to a TensorArray.

value - The concatenated tensor to write to the TensorArray.

lengths - The vector of lengths, how to split the rows of value into the

flowIn - A float scalar that enforces proper chaining of operations.

Returns:

a new instance of TensorArraySplit

See Also:

TensorArraySplit

extractJpegShape

public ExtractJpegShape<Integer> extractJpegShape(Operand<String> contents)

Adds an ExtractJpegShape operation to the graph

Parameters:

contents - 0-D. The JPEG-encoded image.

Returns:

a new instance of ExtractJpegShape

See Also:

ExtractJpegShape

truncatedNormal

public <U extends Number,T extends Number> TruncatedNormal<U> truncatedNormal(Operand<T> shape,
                                                                              Class<U> dtype,
                                                                              TruncatedNormal.Options... options)

Adds an TruncatedNormal operation to the graph

Parameters:

shape - The shape of the output tensor.

dtype - The type of the output.

options - carries optional attributes values

Returns:

a new instance of TruncatedNormal

See Also:

TruncatedNormal

mergeSummary

public MergeSummary mergeSummary(Iterable<Operand<String>> inputs)

Adds an MergeSummary operation to the graph

Parameters:

inputs - Can be of any shape. Each must contain serialized `Summary` protocol

Returns:

a new instance of MergeSummary

See Also:

MergeSummary

conv2DBackpropFilter

public <T extends Number> Conv2DBackpropFilter<T> conv2DBackpropFilter(Operand<T> input,
                                                                       Operand<Integer> filterSizes,
                                                                       Operand<T> outBackprop,
                                                                       List<Long> strides,
                                                                       String padding,
                                                                       Conv2DBackpropFilter.Options... options)

Adds an Conv2DBackpropFilter operation to the graph

Parameters:

input - 4-D with shape `[batch, in_height, in_width, in_channels]`.

filterSizes - An integer vector representing the tensor shape of `filter`,

outBackprop - 4-D with shape `[batch, out_height, out_width, out_channels]`.

strides - The stride of the sliding window for each dimension of the input

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of Conv2DBackpropFilter

See Also:

Conv2DBackpropFilter

maxPoolWithArgmax

public <T extends Number,U extends Number> MaxPoolWithArgmax<T,U> maxPoolWithArgmax(Operand<T> input,
                                                                                    List<Long> ksize,
                                                                                    List<Long> strides,
                                                                                    Class<U> Targmax,
                                                                                    String padding)

Adds an MaxPoolWithArgmax operation to the graph

Parameters:

input - 4-D with shape `[batch, height, width, channels]`. Input to pool over.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the

Targmax -

padding - The type of padding algorithm to use.

Returns:

a new instance of MaxPoolWithArgmax

See Also:

MaxPoolWithArgmax

regexFullMatch

public RegexFullMatch regexFullMatch(Operand<String> input,
                                     Operand<String> pattern)

Adds an RegexFullMatch operation to the graph

Parameters:

input - A string tensor of the text to be processed.

pattern - A scalar string tensor containing the regular expression to match the input.

Returns:

a new instance of RegexFullMatch

See Also:

RegexFullMatch

cropAndResize

public <T extends Number> CropAndResize cropAndResize(Operand<T> image,
                                                      Operand<Float> boxes,
                                                      Operand<Integer> boxInd,
                                                      Operand<Integer> cropSize,
                                                      CropAndResize.Options... options)

Adds an CropAndResize operation to the graph

Parameters:

image - A 4-D tensor of shape `[batch, image_height, image_width, depth]`.

boxes - A 2-D tensor of shape `[num_boxes, 4]`. The `i`-th row of the tensor

boxInd - A 1-D tensor of shape `[num_boxes]` with int32 values in `[0, batch)`.

cropSize - A 1-D tensor of 2 elements, `size = [crop_height, crop_width]`. All

options - carries optional attributes values

Returns:

a new instance of CropAndResize

See Also:

CropAndResize

quantizedConv2D

public <V,T,U> QuantizedConv2D<V> quantizedConv2D(Operand<T> input,
                                                  Operand<U> filter,
                                                  Operand<Float> minInput,
                                                  Operand<Float> maxInput,
                                                  Operand<Float> minFilter,
                                                  Operand<Float> maxFilter,
                                                  Class<V> outType,
                                                  List<Long> strides,
                                                  String padding,
                                                  QuantizedConv2D.Options... options)

Adds an QuantizedConv2D operation to the graph

Parameters:

input -

filter - filter's input_depth dimension must match input's depth dimensions.

minInput - The float value that the lowest quantized input value represents.

maxInput - The float value that the highest quantized input value represents.

minFilter - The float value that the lowest quantized filter value represents.

maxFilter - The float value that the highest quantized filter value represents.

outType -

strides - The stride of the sliding window for each dimension of the input

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of QuantizedConv2D

See Also:

QuantizedConv2D

barrierClose

public BarrierClose barrierClose(Operand<String> handle,
                                 BarrierClose.Options... options)

Adds an BarrierClose operation to the graph

Parameters:

handle - The handle to a barrier.

options - carries optional attributes values

Returns:

a new instance of BarrierClose

See Also:

BarrierClose

constant

public Constant<Double> constant(double[] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

spaceToBatch

public <T,U extends Number> SpaceToBatch<T> spaceToBatch(Operand<T> input,
                                                         Operand<U> paddings,
                                                         Long blockSize)

Adds an SpaceToBatch operation to the graph

Parameters:

input - 4-D with shape `[batch, height, width, depth]`.

paddings - 2-D tensor of non-negative integers with shape `[2, 2]`. It specifies

blockSize -

Returns:

a new instance of SpaceToBatch

See Also:

SpaceToBatch

quantizedAvgPool

public <T> QuantizedAvgPool<T> quantizedAvgPool(Operand<T> input,
                                                Operand<Float> minInput,
                                                Operand<Float> maxInput,
                                                List<Long> ksize,
                                                List<Long> strides,
                                                String padding)

Adds an QuantizedAvgPool operation to the graph

Parameters:

input - 4-D with shape `[batch, height, width, channels]`.

minInput - The float value that the lowest quantized input value represents.

maxInput - The float value that the highest quantized input value represents.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the input

padding - The type of padding algorithm to use.

Returns:

a new instance of QuantizedAvgPool

See Also:

QuantizedAvgPool

decodeAndCropJpeg

public DecodeAndCropJpeg decodeAndCropJpeg(Operand<String> contents,
                                           Operand<Integer> cropWindow,
                                           DecodeAndCropJpeg.Options... options)

Adds an DecodeAndCropJpeg operation to the graph

Parameters:

contents - 0-D. The JPEG-encoded image.

cropWindow - 1-D. The crop window: [crop_y, crop_x, crop_height, crop_width].

options - carries optional attributes values

Returns:

a new instance of DecodeAndCropJpeg

See Also:

DecodeAndCropJpeg

histogramFixedWidth

public <U extends Number,T extends Number> HistogramFixedWidth<U> histogramFixedWidth(Operand<T> values,
                                                                                      Operand<T> valueRange,
                                                                                      Operand<Integer> nbins,
                                                                                      Class<U> dtype)

Adds an HistogramFixedWidth operation to the graph

Parameters:

values - Numeric `Tensor`.

valueRange - Shape [2] `Tensor` of same `dtype` as `values`.

nbins - Scalar `int32 Tensor`. Number of histogram bins.

dtype -

Returns:

a new instance of HistogramFixedWidth

See Also:

HistogramFixedWidth

sampleDistortedBoundingBoxV2

public <T extends Number> SampleDistortedBoundingBoxV2<T> sampleDistortedBoundingBoxV2(Operand<T> imageSize,
                                                                                       Operand<Float> boundingBoxes,
                                                                                       Operand<Float> minObjectCovered,
                                                                                       SampleDistortedBoundingBoxV2.Options... options)

Adds an SampleDistortedBoundingBoxV2 operation to the graph

Parameters:

imageSize - 1-D, containing `[height, width, channels]`.

boundingBoxes - 3-D with shape `[batch, N, 4]` describing the N bounding boxes

minObjectCovered - The cropped area of the image must contain at least this

options - carries optional attributes values

Returns:

a new instance of SampleDistortedBoundingBoxV2

See Also:

SampleDistortedBoundingBoxV2

mfcc

public Mfcc mfcc(Operand<Float> spectrogram,
                 Operand<Integer> sampleRate,
                 Mfcc.Options... options)

Adds an Mfcc operation to the graph

Parameters:

spectrogram - Typically produced by the Spectrogram op, with magnitude_squared

sampleRate - How many samples per second the source audio used.

options - carries optional attributes values

Returns:

a new instance of Mfcc

See Also:

Mfcc

inplaceUpdate

public <T> InplaceUpdate<T> inplaceUpdate(Operand<T> x,
                                          Operand<Integer> i,
                                          Operand<T> v)

Adds an InplaceUpdate operation to the graph

Parameters:

x - A tensor of type `T`.

i - A vector. Indices into the left-most dimension of `x`.

v - A `Tensor` of type T. Same dimension sizes as x except the first dimension, which must be the same as i's size.

Returns:

a new instance of InplaceUpdate

See Also:

InplaceUpdate

inplaceAdd

public <T> InplaceAdd<T> inplaceAdd(Operand<T> x,
                                    Operand<Integer> i,
                                    Operand<T> v)

Adds an InplaceAdd operation to the graph

Parameters:

x - A `Tensor` of type T.

i - A vector. Indices into the left-most dimension of `x`.

v - A `Tensor` of type T. Same dimension sizes as x except the first dimension, which must be the same as i's size.

Returns:

a new instance of InplaceAdd

See Also:

InplaceAdd

argMin

public <T,U extends Number> ArgMin<Long> argMin(Operand<T> input,
                                                Operand<U> dimension)

Adds an ArgMin operation to the graph

Parameters:

input -

dimension - int32 or int64, must be in the range `[-rank(input), rank(input))`.

Returns:

a new instance of ArgMin

See Also:

ArgMin

scatterNdUpdate

public <T,U extends Number> ScatterNdUpdate<T> scatterNdUpdate(Operand<T> ref,
                                                               Operand<U> indices,
                                                               Operand<T> updates,
                                                               ScatterNdUpdate.Options... options)

Adds an ScatterNdUpdate operation to the graph

Parameters:

ref - A mutable Tensor. Should be from a Variable node.

indices - A Tensor. Must be one of the following types: int32, int64.

updates - A Tensor. Must have the same type as ref. A tensor of updated

options - carries optional attributes values

Returns:

a new instance of ScatterNdUpdate

See Also:

ScatterNdUpdate

sparseConcat

public <T> SparseConcat<T> sparseConcat(Iterable<Operand<Long>> indices,
                                        Operand<T> values,
                                        Iterable<Operand<Long>> shapes,
                                        Long concatDim)

Adds an SparseConcat operation to the graph

Parameters:

indices - 2-D. Indices of each input `SparseTensor`.

values - 1-D. Non-empty values of each `SparseTensor`.

shapes - 1-D. Shapes of each `SparseTensor`.

concatDim - Dimension to concatenate along. Must be in range [-rank, rank),

Returns:

a new instance of SparseConcat

See Also:

SparseConcat

gather

public <T,U extends Number> Gather<T> gather(Operand<T> params,
                                             Operand<U> indices,
                                             Gather.Options... options)

Adds an Gather operation to the graph

Parameters:

params -

indices -

options - carries optional attributes values

Returns:

a new instance of Gather

See Also:

Gather

segmentMax

public <T extends Number,U extends Number> SegmentMax<T> segmentMax(Operand<T> data,
                                                                    Operand<U> segmentIds)

Adds an SegmentMax operation to the graph

Parameters:

data -

segmentIds - A 1-D tensor whose size is equal to the size of `data`'s

Returns:

a new instance of SegmentMax

See Also:

SegmentMax

batchMatrixInverse

public <T extends Number> BatchMatrixInverse<T> batchMatrixInverse(Operand<T> input,
                                                                   BatchMatrixInverse.Options... options)

Adds an BatchMatrixInverse operation to the graph

Parameters:

input -

options - carries optional attributes values

Returns:

a new instance of BatchMatrixInverse

See Also:

BatchMatrixInverse

bucketize

public <T extends Number> Bucketize bucketize(Operand<T> input,
                                              List<Float> boundaries)

Adds an Bucketize operation to the graph

Parameters:

input - Any shape of Tensor contains with int or float type.

boundaries - A sorted list of floats gives the boundary of the buckets.

Returns:

a new instance of Bucketize

See Also:

Bucketize

constant

public Constant<Integer> constant(int[][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

iRFFT2D

public IRFFT2D iRFFT2D(Operand<?> input,
                       Operand<Integer> fftLength)

Adds an IRFFT2D operation to the graph

Parameters:

input - A complex64 tensor.

fftLength - An int32 tensor of shape [2]. The FFT length for each dimension.

Returns:

a new instance of IRFFT2D

See Also:

IRFFT2D

readerRestoreState

public ReaderRestoreState readerRestoreState(Operand<?> readerHandle,
                                             Operand<String> state)

Adds an ReaderRestoreState operation to the graph

Parameters:

readerHandle - Handle to a Reader.

state - Result of a ReaderSerializeState of a Reader with type

Returns:

a new instance of ReaderRestoreState

See Also:

ReaderRestoreState

noOp

public NoOp noOp()

Adds an NoOp operation to the graph

Returns:

a new instance of NoOp

See Also:

NoOp

conjugateTranspose

public <T,U extends Number> ConjugateTranspose<T> conjugateTranspose(Operand<T> x,
                                                                     Operand<U> perm)

Adds an ConjugateTranspose operation to the graph

Parameters:

x -

perm -

Returns:

a new instance of ConjugateTranspose

See Also:

ConjugateTranspose

serializeSparse

public <U,T> SerializeSparse<U> serializeSparse(Operand<Long> sparseIndices,
                                                Operand<T> sparseValues,
                                                Operand<Long> sparseShape,
                                                Class<U> outType)

Adds an SerializeSparse operation to the graph

Parameters:

sparseIndices - 2-D. The `indices` of the `SparseTensor`.

sparseValues - 1-D. The `values` of the `SparseTensor`.

sparseShape - 1-D. The `shape` of the `SparseTensor`.

outType - The `dtype` to use for serialization; the supported types are `string`

Returns:

a new instance of SerializeSparse

See Also:

SerializeSparse

round

public <T> Round<T> round(Operand<T> x)

Adds an Round operation to the graph

Parameters:

x -

Returns:

a new instance of Round

See Also:

Round

writeFile

public WriteFile writeFile(Operand<String> filename,
                           Operand<String> contents)

Adds an WriteFile operation to the graph

Parameters:

filename - scalar. The name of the file to which we write the contents.

contents - scalar. The content to be written to the output file.

Returns:

a new instance of WriteFile

See Also:

WriteFile

matrixSolve

public <T> MatrixSolve<T> matrixSolve(Operand<T> matrix,
                                      Operand<T> rhs,
                                      MatrixSolve.Options... options)

Adds an MatrixSolve operation to the graph

Parameters:

matrix - Shape is `[..., M, M]`.

rhs - Shape is `[..., M, K]`.

options - carries optional attributes values

Returns:

a new instance of MatrixSolve

See Also:

MatrixSolve

nonMaxSuppression

public NonMaxSuppression nonMaxSuppression(Operand<Float> boxes,
                                           Operand<Float> scores,
                                           Operand<Integer> maxOutputSize,
                                           NonMaxSuppression.Options... options)

Adds an NonMaxSuppression operation to the graph

Parameters:

boxes - A 2-D float tensor of shape `[num_boxes, 4]`.

scores - A 1-D float tensor of shape `[num_boxes]` representing a single

maxOutputSize - A scalar integer tensor representing the maximum number of

options - carries optional attributes values

Returns:

a new instance of NonMaxSuppression

See Also:

NonMaxSuppression

maxPool3DGradGrad

public <T extends Number> MaxPool3DGradGrad<T> maxPool3DGradGrad(Operand<T> origInput,
                                                                 Operand<T> origOutput,
                                                                 Operand<T> grad,
                                                                 List<Long> ksize,
                                                                 List<Long> strides,
                                                                 String padding,
                                                                 MaxPool3DGradGrad.Options... options)

Adds an MaxPool3DGradGrad operation to the graph

Parameters:

origInput - The original input tensor.

origOutput - The original output tensor.

grad - Output backprop of shape `[batch, depth, rows, cols, channels]`.

ksize - 1-D tensor of length 5. The size of the window for each dimension of

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of MaxPool3DGradGrad

See Also:

MaxPool3DGradGrad

batchMatrixDeterminant

public <T> BatchMatrixDeterminant<T> batchMatrixDeterminant(Operand<T> input)

Adds an BatchMatrixDeterminant operation to the graph

Parameters:

input -

Returns:

a new instance of BatchMatrixDeterminant

See Also:

BatchMatrixDeterminant

cross

public <T extends Number> Cross<T> cross(Operand<T> a,
                                         Operand<T> b)

Adds an Cross operation to the graph

Parameters:

a - A tensor containing 3-element vectors.

b - Another tensor, of same type and shape as `a`.

Returns:

a new instance of Cross

See Also:

Cross

anonymousIterator

public AnonymousIterator anonymousIterator(List<Class<?>> outputTypes,
                                           List<Shape> outputShapes)

Adds an AnonymousIterator operation to the graph

Parameters:

outputTypes -

outputShapes -

Returns:

a new instance of AnonymousIterator

See Also:

AnonymousIterator

quantizedBiasAdd

public <V,T,U> QuantizedBiasAdd<V> quantizedBiasAdd(Operand<T> input,
                                                    Operand<U> bias,
                                                    Operand<Float> minInput,
                                                    Operand<Float> maxInput,
                                                    Operand<Float> minBias,
                                                    Operand<Float> maxBias,
                                                    Class<V> outType)

Adds an QuantizedBiasAdd operation to the graph

Parameters:

input -

bias - A 1D bias Tensor with size matching the last dimension of 'input'.

minInput - The float value that the lowest quantized input value represents.

maxInput - The float value that the highest quantized input value represents.

minBias - The float value that the lowest quantized bias value represents.

maxBias - The float value that the highest quantized bias value represents.

outType -

Returns:

a new instance of QuantizedBiasAdd

See Also:

QuantizedBiasAdd

cudnnRNN

public <T extends Number> CudnnRNN<T> cudnnRNN(Operand<T> input,
                                               Operand<T> inputH,
                                               Operand<T> inputC,
                                               Operand<T> params,
                                               CudnnRNN.Options... options)

Adds an CudnnRNN operation to the graph

Parameters:

input -

inputH -

inputC -

params -

options - carries optional attributes values

Returns:

a new instance of CudnnRNN

See Also:

CudnnRNN

decodeRaw

public <T extends Number> DecodeRaw<T> decodeRaw(Operand<String> bytes,
                                                 Class<T> outType,
                                                 DecodeRaw.Options... options)

Adds an DecodeRaw operation to the graph

Parameters:

bytes - All the elements must have the same length.

outType -

options - carries optional attributes values

Returns:

a new instance of DecodeRaw

See Also:

DecodeRaw

sparseAccumulatorTakeGradient

public <T> SparseAccumulatorTakeGradient<T> sparseAccumulatorTakeGradient(Operand<String> handle,
                                                                          Operand<Integer> numRequired,
                                                                          Class<T> dtype)

Adds an SparseAccumulatorTakeGradient operation to the graph

Parameters:

handle - The handle to a SparseConditionalAccumulator.

numRequired - Number of gradients required before we return an aggregate.

dtype - The data type of accumulated gradients. Needs to correspond to the type

Returns:

a new instance of SparseAccumulatorTakeGradient

See Also:

SparseAccumulatorTakeGradient

imag

public <U extends Number,T> Imag<U> imag(Operand<T> input,
                                         Class<U> Tout)

Adds an Imag operation to the graph

Parameters:

input -

Tout -

Returns:

a new instance of Imag

See Also:

Imag

realDiv

public <T> RealDiv<T> realDiv(Operand<T> x,
                              Operand<T> y)

Adds an RealDiv operation to the graph

Parameters:

x -

y -

Returns:

a new instance of RealDiv

See Also:

RealDiv

depthwiseConv2dNativeBackpropInput

public <T extends Number> DepthwiseConv2dNativeBackpropInput<T> depthwiseConv2dNativeBackpropInput(Operand<Integer> inputSizes,
                                                                                                   Operand<T> filter,
                                                                                                   Operand<T> outBackprop,
                                                                                                   List<Long> strides,
                                                                                                   String padding,
                                                                                                   DepthwiseConv2dNativeBackpropInput.Options... options)

Adds an DepthwiseConv2dNativeBackpropInput operation to the graph

Parameters:

inputSizes - An integer vector representing the shape of `input`, based

filter - 4-D with shape

outBackprop - 4-D with shape based on `data_format`.

strides - The stride of the sliding window for each dimension of the input

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of DepthwiseConv2dNativeBackpropInput

See Also:

DepthwiseConv2dNativeBackpropInput

conj

public <T> Conj<T> conj(Operand<T> input)

Adds an Conj operation to the graph

Parameters:

input -

Returns:

a new instance of Conj

See Also:

Conj

nonMaxSuppressionV3

public <T extends Number> NonMaxSuppressionV3 nonMaxSuppressionV3(Operand<T> boxes,
                                                                  Operand<T> scores,
                                                                  Operand<Integer> maxOutputSize,
                                                                  Operand<Float> iouThreshold,
                                                                  Operand<Float> scoreThreshold)

Adds an NonMaxSuppressionV3 operation to the graph

Parameters:

boxes - A 2-D float tensor of shape `[num_boxes, 4]`.

scores - A 1-D float tensor of shape `[num_boxes]` representing a single

maxOutputSize - A scalar integer tensor representing the maximum number of

iouThreshold - A 0-D float tensor representing the threshold for deciding whether

scoreThreshold - A 0-D float tensor representing the threshold for deciding when to remove

Returns:

a new instance of NonMaxSuppressionV3

See Also:

NonMaxSuppressionV3

tensorListReserve

public <T extends Number,U> TensorListReserve tensorListReserve(Operand<T> elementShape,
                                                                Operand<Integer> numElements,
                                                                Class<U> elementDtype)

Adds an TensorListReserve operation to the graph

Parameters:

elementShape -

numElements -

elementDtype -

Returns:

a new instance of TensorListReserve

See Also:

TensorListReserve

barrierIncompleteSize

public BarrierIncompleteSize barrierIncompleteSize(Operand<String> handle)

Adds an BarrierIncompleteSize operation to the graph

Parameters:

handle - The handle to a barrier.

Returns:

a new instance of BarrierIncompleteSize

See Also:

BarrierIncompleteSize

constant

public Constant<Double> constant(double[][][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

constant

public Constant<String> constant(byte[][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. String elements are

See Also:

Constant

parseTensor

public <T> ParseTensor<T> parseTensor(Operand<String> serialized,
                                      Class<T> outType)

Adds an ParseTensor operation to the graph

Parameters:

serialized - A scalar string containing a serialized TensorProto proto.

outType - The type of the serialized tensor. The provided type must match the

Returns:

a new instance of ParseTensor

See Also:

ParseTensor

countUpTo

public <T extends Number> CountUpTo<T> countUpTo(Operand<T> ref,
                                                 Long limit)

Adds an CountUpTo operation to the graph

Parameters:

ref - Should be from a scalar `Variable` node.

limit - If incrementing ref would bring it above limit, instead generates an

Returns:

a new instance of CountUpTo

See Also:

CountUpTo

mod

public <T extends Number> Mod<T> mod(Operand<T> x,
                                     Operand<T> y)

Adds an Mod operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Mod

See Also:

Mod

sparseApplyMomentum

public <T,U extends Number> SparseApplyMomentum<T> sparseApplyMomentum(Operand<T> var,
                                                                       Operand<T> accum,
                                                                       Operand<T> lr,
                                                                       Operand<T> grad,
                                                                       Operand<U> indices,
                                                                       Operand<T> momentum,
                                                                       SparseApplyMomentum.Options... options)

Adds an SparseApplyMomentum operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Learning rate. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

momentum - Momentum. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of SparseApplyMomentum

See Also:

SparseApplyMomentum

fractionalAvgPool

public <T extends Number> FractionalAvgPool<T> fractionalAvgPool(Operand<T> value,
                                                                 List<Float> poolingRatio,
                                                                 FractionalAvgPool.Options... options)

Adds an FractionalAvgPool operation to the graph

Parameters:

value - 4-D with shape `[batch, height, width, channels]`.

poolingRatio - Pooling ratio for each dimension of `value`, currently only

options - carries optional attributes values

Returns:

a new instance of FractionalAvgPool

See Also:

FractionalAvgPool

uniqueWithCountsV2

public <T,U extends Number> UniqueWithCountsV2<T,Integer> uniqueWithCountsV2(Operand<T> x,
                                                                             Operand<U> axis)

Adds an UniqueWithCountsV2 operation to the graph

Parameters:

x - A `Tensor`.

axis - A `Tensor` of type `int32` (default: None). The axis of the Tensor to

Returns:

a new instance of UniqueWithCountsV2

See Also:

UniqueWithCountsV2

stageClear

public StageClear stageClear(List<Class<?>> dtypes,
                             StageClear.Options... options)

Adds an StageClear operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of StageClear

See Also:

StageClear

fill

public <U,T extends Number> Fill<U> fill(Operand<T> dims,
                                         Operand<U> value)

Adds an Fill operation to the graph

Parameters:

dims - 1-D. Represents the shape of the output tensor.

value - 0-D (scalar). Value to fill the returned tensor.

Returns:

a new instance of Fill

See Also:

Fill

isNan

public <T extends Number> IsNan isNan(Operand<T> x)

Adds an IsNan operation to the graph

Parameters:

x -

Returns:

a new instance of IsNan

See Also:

IsNan

varIsInitializedOp

public VarIsInitializedOp varIsInitializedOp(Operand<?> resource)

Adds an VarIsInitializedOp operation to the graph

Parameters:

resource - the input resource handle.

Returns:

a new instance of VarIsInitializedOp

See Also:

VarIsInitializedOp

merge

public <T> Merge<T> merge(Operand<T> inputs)

Adds an Merge operation to the graph

Parameters:

inputs - The input tensors, exactly one of which will become available.

Returns:

a new instance of Merge

See Also:

Merge

depthwiseConv2dNativeBackpropFilter

public <T extends Number> DepthwiseConv2dNativeBackpropFilter<T> depthwiseConv2dNativeBackpropFilter(Operand<T> input,
                                                                                                     Operand<Integer> filterSizes,
                                                                                                     Operand<T> outBackprop,
                                                                                                     List<Long> strides,
                                                                                                     String padding,
                                                                                                     DepthwiseConv2dNativeBackpropFilter.Options... options)

Adds an DepthwiseConv2dNativeBackpropFilter operation to the graph

Parameters:

input - 4-D with shape based on `data_format`. For example, if

filterSizes - An integer vector representing the tensor shape of `filter`,

outBackprop - 4-D with shape based on `data_format`.

strides - The stride of the sliding window for each dimension of the input

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of DepthwiseConv2dNativeBackpropFilter

See Also:

DepthwiseConv2dNativeBackpropFilter

batchNormWithGlobalNormalizationGrad

public <T> BatchNormWithGlobalNormalizationGrad<T> batchNormWithGlobalNormalizationGrad(Operand<T> t,
                                                                                        Operand<T> m,
                                                                                        Operand<T> v,
                                                                                        Operand<T> gamma,
                                                                                        Operand<T> backprop,
                                                                                        Float varianceEpsilon,
                                                                                        Boolean scaleAfterNormalization)

Adds an BatchNormWithGlobalNormalizationGrad operation to the graph

Parameters:

t - A 4D input Tensor.

m - A 1D mean Tensor with size matching the last dimension of t.

v - A 1D variance Tensor with size matching the last dimension of t.

gamma - A 1D gamma Tensor with size matching the last dimension of t.

backprop - 4D backprop Tensor.

varianceEpsilon - A small float number to avoid dividing by 0.

scaleAfterNormalization - A bool indicating whether the resulted tensor

Returns:

a new instance of BatchNormWithGlobalNormalizationGrad

See Also:

BatchNormWithGlobalNormalizationGrad

cudnnRNNCanonicalToParams

public <T extends Number> CudnnRNNCanonicalToParams<T> cudnnRNNCanonicalToParams(Operand<Integer> numLayers,
                                                                                 Operand<Integer> numUnits,
                                                                                 Operand<Integer> inputSize,
                                                                                 Operand<T> weights,
                                                                                 Iterable<Operand<T>> biases,
                                                                                 CudnnRNNCanonicalToParams.Options... options)

Adds an CudnnRNNCanonicalToParams operation to the graph

Parameters:

numLayers -

numUnits -

inputSize -

weights -

biases -

options - carries optional attributes values

Returns:

a new instance of CudnnRNNCanonicalToParams

See Also:

CudnnRNNCanonicalToParams

dynamicPartition

public <T> DynamicPartition<T> dynamicPartition(Operand<T> data,
                                                Operand<Integer> partitions,
                                                Long numPartitions)

Adds an DynamicPartition operation to the graph

Parameters:

data -

partitions - Any shape. Indices in the range `[0, num_partitions)`.

numPartitions - The number of partitions to output.

Returns:

a new instance of DynamicPartition

See Also:

DynamicPartition

maxPool

public <T> MaxPool<T> maxPool(Operand<T> input,
                              List<Long> ksize,
                              List<Long> strides,
                              String padding,
                              MaxPool.Options... options)

Adds an MaxPool operation to the graph

Parameters:

input - 4-D input to pool over.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of MaxPool

See Also:

MaxPool

setDiff1D

public <T> SetDiff1D<T,Integer> setDiff1D(Operand<T> x,
                                          Operand<T> y)

Adds an SetDiff1D operation to the graph

Parameters:

x - 1-D. Values to keep.

y - 1-D. Values to remove.

Returns:

a new instance of SetDiff1D

See Also:

SetDiff1D

constant

public Constant<Double> constant(long[] shape,
                                 DoubleBuffer data)

Adds an Constant operation to the graph

Parameters:

shape - the tensor shape.

data - a buffer containing the tensor data.

Returns:

a double constant

Throws:

IllegalArgumentException - If the tensor shape is not compatible with the buffer

See Also:

Constant

neg

public <T> Neg<T> neg(Operand<T> x)

Adds an Neg operation to the graph

Parameters:

x -

Returns:

a new instance of Neg

See Also:

Neg

broadcastTo

public <T,U extends Number> BroadcastTo<T> broadcastTo(Operand<T> input,
                                                       Operand<U> shape)

Adds an BroadcastTo operation to the graph

Parameters:

input - A Tensor to broadcast.

shape - An 1-D `int` Tensor. The shape of the desired output.

Returns:

a new instance of BroadcastTo

See Also:

BroadcastTo

maximum

public <T extends Number> Maximum<T> maximum(Operand<T> x,
                                             Operand<T> y)

Adds an Maximum operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Maximum

See Also:

Maximum

getSessionHandleV2

public <T> GetSessionHandleV2 getSessionHandleV2(Operand<T> value)

Adds an GetSessionHandleV2 operation to the graph

Parameters:

value - The tensor to be stored.

Returns:

a new instance of GetSessionHandleV2

See Also:

GetSessionHandleV2

resourceApplyCenteredRMSProp

public <T> ResourceApplyCenteredRMSProp resourceApplyCenteredRMSProp(Operand<?> var,
                                                                     Operand<?> mg,
                                                                     Operand<?> ms,
                                                                     Operand<?> mom,
                                                                     Operand<T> lr,
                                                                     Operand<T> rho,
                                                                     Operand<T> momentum,
                                                                     Operand<T> epsilon,
                                                                     Operand<T> grad,
                                                                     ResourceApplyCenteredRMSProp.Options... options)

Adds an ResourceApplyCenteredRMSProp operation to the graph

Parameters:

var - Should be from a Variable().

mg - Should be from a Variable().

ms - Should be from a Variable().

mom - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay rate. Must be a scalar.

momentum -

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyCenteredRMSProp

See Also:

ResourceApplyCenteredRMSProp

unsortedSegmentSum

public <T,U extends Number,V extends Number> UnsortedSegmentSum<T> unsortedSegmentSum(Operand<T> data,
                                                                                      Operand<U> segmentIds,
                                                                                      Operand<V> numSegments)

Adds an UnsortedSegmentSum operation to the graph

Parameters:

data -

segmentIds - A tensor whose shape is a prefix of `data.shape`.

numSegments -

Returns:

a new instance of UnsortedSegmentSum

See Also:

UnsortedSegmentSum

fixedLengthRecordReader

public FixedLengthRecordReader fixedLengthRecordReader(Long recordBytes,
                                                       FixedLengthRecordReader.Options... options)

Adds an FixedLengthRecordReader operation to the graph

Parameters:

recordBytes - Number of bytes in the record.

options - carries optional attributes values

Returns:

a new instance of FixedLengthRecordReader

See Also:

FixedLengthRecordReader

accumulatorApplyGradient

public <T> AccumulatorApplyGradient accumulatorApplyGradient(Operand<String> handle,
                                                             Operand<Long> localStep,
                                                             Operand<T> gradient)

Adds an AccumulatorApplyGradient operation to the graph

Parameters:

handle - The handle to a accumulator.

localStep - The local_step value at which the gradient was computed.

gradient - A tensor of the gradient to be accumulated.

Returns:

a new instance of AccumulatorApplyGradient

See Also:

AccumulatorApplyGradient

tensorArray

public <T> TensorArray tensorArray(Operand<Integer> size,
                                   Class<T> dtype,
                                   TensorArray.Options... options)

Adds an TensorArray operation to the graph

Parameters:

size - The size of the array.

dtype - The type of the elements on the tensor_array.

options - carries optional attributes values

Returns:

a new instance of TensorArray

See Also:

TensorArray

gradients

public Gradients gradients(Iterable<? extends Operand<?>> y,
                           Iterable<? extends Operand<?>> x,
                           Gradients.Options... options)

Adds an Gradients operation to the graph

Parameters:

y - outputs of the function to derive

x - inputs of the function for which partial derivatives are computed

options - carries optional attributes values

Returns:

a new instance of Gradients

See Also:

Gradients

snapshot

public <T> Snapshot<T> snapshot(Operand<T> input)

Adds an Snapshot operation to the graph

Parameters:

input -

Returns:

a new instance of Snapshot

See Also:

Snapshot

fractionalMaxPool

public <T extends Number> FractionalMaxPool<T> fractionalMaxPool(Operand<T> value,
                                                                 List<Float> poolingRatio,
                                                                 FractionalMaxPool.Options... options)

Adds an FractionalMaxPool operation to the graph

Parameters:

value - 4-D with shape `[batch, height, width, channels]`.

poolingRatio - Pooling ratio for each dimension of `value`, currently only

options - carries optional attributes values

Returns:

a new instance of FractionalMaxPool

See Also:

FractionalMaxPool

computeAccidentalHits

public ComputeAccidentalHits computeAccidentalHits(Operand<Long> trueClasses,
                                                   Operand<Long> sampledCandidates,
                                                   Long numTrue,
                                                   ComputeAccidentalHits.Options... options)

Adds an ComputeAccidentalHits operation to the graph

Parameters:

trueClasses - The true_classes output of UnpackSparseLabels.

sampledCandidates - The sampled_candidates output of CandidateSampler.

numTrue - Number of true labels per context.

options - carries optional attributes values

Returns:

a new instance of ComputeAccidentalHits

See Also:

ComputeAccidentalHits

checkNumerics

public <T extends Number> CheckNumerics<T> checkNumerics(Operand<T> tensor,
                                                         String message)

Adds an CheckNumerics operation to the graph

Parameters:

tensor -

message - Prefix of the error message.

Returns:

a new instance of CheckNumerics

See Also:

CheckNumerics

choleskyGrad

public <T extends Number> CholeskyGrad<T> choleskyGrad(Operand<T> l,
                                                       Operand<T> grad)

Adds an CholeskyGrad operation to the graph

Parameters:

l - Output of batch Cholesky algorithm l = cholesky(A). Shape is `[..., M, M]`.

grad - df/dl where f is some scalar function. Shape is `[..., M, M]`.

Returns:

a new instance of CholeskyGrad

See Also:

CholeskyGrad

logSoftmax

public <T extends Number> LogSoftmax<T> logSoftmax(Operand<T> logits)

Adds an LogSoftmax operation to the graph

Parameters:

logits - 2-D with shape `[batch_size, num_classes]`.

Returns:

a new instance of LogSoftmax

See Also:

LogSoftmax

gradients

public Gradients gradients(Operand<?> y,
                           Iterable<? extends Operand<?>> x,
                           Gradients.Options... options)

Adds an Gradients operation to the graph

Parameters:

y - output of the function to derive

x - inputs of the function for which partial derivatives are computed

options - carries optional attributes values

Returns:

a new instance of Gradients

See Also:

Gradients

tensorScatterAdd

public <T,U extends Number> TensorScatterAdd<T> tensorScatterAdd(Operand<T> tensor,
                                                                 Operand<U> indices,
                                                                 Operand<T> updates)

Adds an TensorScatterAdd operation to the graph

Parameters:

tensor - Tensor to copy/update.

indices - Index tensor.

updates - Updates to scatter into output.

Returns:

a new instance of TensorScatterAdd

See Also:

TensorScatterAdd

saveSlices

public SaveSlices saveSlices(Operand<String> filename,
                             Operand<String> tensorNames,
                             Operand<String> shapesAndSlices,
                             Iterable<Operand<?>> data)

Adds an SaveSlices operation to the graph

Parameters:

filename - Must have a single element. The name of the file to which we write the

tensorNames - Shape `[N]`. The names of the tensors to be saved.

shapesAndSlices - Shape `[N]`. The shapes and slice specifications to use when

data - `N` tensors to save.

Returns:

a new instance of SaveSlices

See Also:

SaveSlices

sparseTensorDenseMatMul

public <U,T extends Number> SparseTensorDenseMatMul<U> sparseTensorDenseMatMul(Operand<T> aIndices,
                                                                               Operand<U> aValues,
                                                                               Operand<Long> aShape,
                                                                               Operand<U> b,
                                                                               SparseTensorDenseMatMul.Options... options)

Adds an SparseTensorDenseMatMul operation to the graph

Parameters:

aIndices - 2-D. The `indices` of the `SparseTensor`, size `[nnz, 2]` Matrix.

aValues - 1-D. The `values` of the `SparseTensor`, size `[nnz]` Vector.

aShape - 1-D. The `shape` of the `SparseTensor`, size `[2]` Vector.

b - 2-D. A dense Matrix.

options - carries optional attributes values

Returns:

a new instance of SparseTensorDenseMatMul

See Also:

SparseTensorDenseMatMul

constant

public Constant<Float> constant(float data)

Adds an Constant operation to the graph

Parameters:

data - The value to put into the new constant.

Returns:

a float constant

See Also:

Constant

sparseDenseCwiseAdd

public <T> SparseDenseCwiseAdd<T> sparseDenseCwiseAdd(Operand<Long> spIndices,
                                                      Operand<T> spValues,
                                                      Operand<Long> spShape,
                                                      Operand<T> dense)

Adds an SparseDenseCwiseAdd operation to the graph

Parameters:

spIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

spValues - 1-D. `N` non-empty values corresponding to `sp_indices`.

spShape - 1-D. Shape of the input SparseTensor.

dense - `R`-D. The dense Tensor operand.

Returns:

a new instance of SparseDenseCwiseAdd

See Also:

SparseDenseCwiseAdd

resourceSparseApplyFtrl

public <T,U extends Number> ResourceSparseApplyFtrl resourceSparseApplyFtrl(Operand<?> var,
                                                                            Operand<?> accum,
                                                                            Operand<?> linear,
                                                                            Operand<T> grad,
                                                                            Operand<U> indices,
                                                                            Operand<T> lr,
                                                                            Operand<T> l1,
                                                                            Operand<T> l2,
                                                                            Operand<T> lrPower,
                                                                            ResourceSparseApplyFtrl.Options... options)

Adds an ResourceSparseApplyFtrl operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

linear - Should be from a Variable().

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

lr - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

lrPower - Scaling factor. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyFtrl

See Also:

ResourceSparseApplyFtrl

tensorListPopBack

public <T> TensorListPopBack<T> tensorListPopBack(Operand<?> inputHandle,
                                                  Class<T> elementDtype)

Adds an TensorListPopBack operation to the graph

Parameters:

inputHandle -

elementDtype -

Returns:

a new instance of TensorListPopBack

See Also:

TensorListPopBack

nonMaxSuppressionV2

public <T extends Number> NonMaxSuppressionV2 nonMaxSuppressionV2(Operand<T> boxes,
                                                                  Operand<T> scores,
                                                                  Operand<Integer> maxOutputSize,
                                                                  Operand<Float> iouThreshold)

Adds an NonMaxSuppressionV2 operation to the graph

Parameters:

boxes - A 2-D float tensor of shape `[num_boxes, 4]`.

scores - A 1-D float tensor of shape `[num_boxes]` representing a single

maxOutputSize - A scalar integer tensor representing the maximum number of

iouThreshold - A 0-D float tensor representing the threshold for deciding whether

Returns:

a new instance of NonMaxSuppressionV2

See Also:

NonMaxSuppressionV2

variable

public <T> Variable<T> variable(Shape shape,
                                Class<T> dtype,
                                Variable.Options... options)

Adds an Variable operation to the graph

Parameters:

shape - The shape of the variable tensor.

dtype - The type of elements in the variable tensor.

options - carries optional attributes values

Returns:

a new instance of Variable

See Also:

Variable

tensorArrayConcat

public <T> TensorArrayConcat<T> tensorArrayConcat(Operand<?> handle,
                                                  Operand<Float> flowIn,
                                                  Class<T> dtype,
                                                  TensorArrayConcat.Options... options)

Adds an TensorArrayConcat operation to the graph

Parameters:

handle - The handle to a TensorArray.

flowIn - A float scalar that enforces proper chaining of operations.

dtype - The type of the elem that is returned.

options - carries optional attributes values

Returns:

a new instance of TensorArrayConcat

See Also:

TensorArrayConcat

randomGamma

public <U extends Number,T extends Number> RandomGamma<U> randomGamma(Operand<T> shape,
                                                                      Operand<U> alpha,
                                                                      RandomGamma.Options... options)

Adds an RandomGamma operation to the graph

Parameters:

shape - 1-D integer tensor. Shape of independent samples to draw from each

alpha - A tensor in which each scalar is a "shape" parameter describing the

options - carries optional attributes values

Returns:

a new instance of RandomGamma

See Also:

RandomGamma

reverse

public <T,U extends Number> Reverse<T> reverse(Operand<T> tensor,
                                               Operand<U> axis)

Adds an Reverse operation to the graph

Parameters:

tensor - Up to 8-D.

axis - 1-D. The indices of the dimensions to reverse. Must be in the range

Returns:

a new instance of Reverse

See Also:

Reverse

batchToSpaceND

public <T,U extends Number,V extends Number> BatchToSpaceND<T> batchToSpaceND(Operand<T> input,
                                                                              Operand<U> blockShape,
                                                                              Operand<V> crops)

Adds an BatchToSpaceND operation to the graph

Parameters:

input - N-D with shape `input_shape = [batch] + spatial_shape + remaining_shape`,

blockShape - 1-D with shape `[M]`, all values must be >= 1.

crops - 2-D with shape `[M, 2]`, all values must be >= 0.

Returns:

a new instance of BatchToSpaceND

See Also:

BatchToSpaceND

constant

public Constant<String> constant(byte[][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. String elements are

See Also:

Constant

takeManySparseFromTensorsMap

public <T> TakeManySparseFromTensorsMap<T> takeManySparseFromTensorsMap(Operand<Long> sparseHandles,
                                                                        Class<T> dtype,
                                                                        TakeManySparseFromTensorsMap.Options... options)

Adds an TakeManySparseFromTensorsMap operation to the graph

Parameters:

sparseHandles - 1-D, The `N` serialized `SparseTensor` objects.

dtype - The `dtype` of the `SparseTensor` objects stored in the

options - carries optional attributes values

Returns:

a new instance of TakeManySparseFromTensorsMap

See Also:

TakeManySparseFromTensorsMap

constant

public Constant<String> constant(byte[][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. String elements are

See Also:

Constant

consumeMutexLock

public ConsumeMutexLock consumeMutexLock(Operand<?> mutexLock)

Adds an ConsumeMutexLock operation to the graph

Parameters:

mutexLock - A tensor returned by `MutexLock`.

Returns:

a new instance of ConsumeMutexLock

See Also:

ConsumeMutexLock

softplus

public <T extends Number> Softplus<T> softplus(Operand<T> features)

Adds an Softplus operation to the graph

Parameters:

features -

Returns:

a new instance of Softplus

See Also:

Softplus

cTCBeamSearchDecoder

public CTCBeamSearchDecoder cTCBeamSearchDecoder(Operand<Float> inputs,
                                                 Operand<Integer> sequenceLength,
                                                 Long beamWidth,
                                                 Long topPaths,
                                                 CTCBeamSearchDecoder.Options... options)

Adds an CTCBeamSearchDecoder operation to the graph

Parameters:

inputs - 3-D, shape: `(max_time x batch_size x num_classes)`, the logits.

sequenceLength - A vector containing sequence lengths, size `(batch)`.

beamWidth - A scalar >= 0 (beam search beam width).

topPaths - A scalar >= 0, <= beam_width (controls output size).

options - carries optional attributes values

Returns:

a new instance of CTCBeamSearchDecoder

See Also:

CTCBeamSearchDecoder

applyAdagrad

public <T> ApplyAdagrad<T> applyAdagrad(Operand<T> var,
                                        Operand<T> accum,
                                        Operand<T> lr,
                                        Operand<T> grad,
                                        ApplyAdagrad.Options... options)

Adds an ApplyAdagrad operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ApplyAdagrad

See Also:

ApplyAdagrad

sparseApplyAdagradDA

public <T,U extends Number> SparseApplyAdagradDA<T> sparseApplyAdagradDA(Operand<T> var,
                                                                         Operand<T> gradientAccumulator,
                                                                         Operand<T> gradientSquaredAccumulator,
                                                                         Operand<T> grad,
                                                                         Operand<U> indices,
                                                                         Operand<T> lr,
                                                                         Operand<T> l1,
                                                                         Operand<T> l2,
                                                                         Operand<Long> globalStep,
                                                                         SparseApplyAdagradDA.Options... options)

Adds an SparseApplyAdagradDA operation to the graph

Parameters:

var - Should be from a Variable().

gradientAccumulator - Should be from a Variable().

gradientSquaredAccumulator - Should be from a Variable().

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

lr - Learning rate. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

globalStep - Training step number. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of SparseApplyAdagradDA

See Also:

SparseApplyAdagradDA

complexAbs

public <U extends Number,T> ComplexAbs<U> complexAbs(Operand<T> x,
                                                     Class<U> Tout)

Adds an ComplexAbs operation to the graph

Parameters:

x -

Tout -

Returns:

a new instance of ComplexAbs

See Also:

ComplexAbs

makeIterator

public MakeIterator makeIterator(Operand<?> dataset,
                                 Operand<?> iterator)

Adds an MakeIterator operation to the graph

Parameters:

dataset -

iterator -

Returns:

a new instance of MakeIterator

See Also:

MakeIterator

sparseSplit

public <T> SparseSplit<T> sparseSplit(Operand<Long> splitDim,
                                      Operand<Long> indices,
                                      Operand<T> values,
                                      Operand<Long> shape,
                                      Long numSplit)

Adds an SparseSplit operation to the graph

Parameters:

splitDim - 0-D. The dimension along which to split. Must be in the range

indices - 2-D tensor represents the indices of the sparse tensor.

values - 1-D tensor represents the values of the sparse tensor.

shape - 1-D. tensor represents the shape of the sparse tensor.

numSplit - The number of ways to split.

Returns:

a new instance of SparseSplit

See Also:

SparseSplit

readerNumWorkUnitsCompleted

public ReaderNumWorkUnitsCompleted readerNumWorkUnitsCompleted(Operand<?> readerHandle)

Adds an ReaderNumWorkUnitsCompleted operation to the graph

Parameters:

readerHandle - Handle to a Reader.

Returns:

a new instance of ReaderNumWorkUnitsCompleted

See Also:

ReaderNumWorkUnitsCompleted

batchMatrixBandPart

public <T> BatchMatrixBandPart<T> batchMatrixBandPart(Operand<T> input,
                                                      Operand<Long> numLower,
                                                      Operand<Long> numUpper)

Adds an BatchMatrixBandPart operation to the graph

Parameters:

input -

numLower -

numUpper -

Returns:

a new instance of BatchMatrixBandPart

See Also:

BatchMatrixBandPart

optionalFromValue

public OptionalFromValue optionalFromValue(Iterable<Operand<?>> components)

Adds an OptionalFromValue operation to the graph

Parameters:

components -

Returns:

a new instance of OptionalFromValue

See Also:

OptionalFromValue

reduceAll

public <T extends Number> ReduceAll reduceAll(Operand<Boolean> input,
                                              Operand<T> axis,
                                              ReduceAll.Options... options)

Adds an ReduceAll operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of ReduceAll

See Also:

ReduceAll

div

public <T> Div<T> div(Operand<T> x,
                      Operand<T> y)

Adds an Div operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Div

See Also:

Div

concat

public <T,U extends Number> Concat<T> concat(Operand<T> values,
                                             Operand<U> axis)

Adds an Concat operation to the graph

Parameters:

values - List of `N` Tensors to concatenate. Their ranks and types must match,

axis - 0-D. The dimension along which to concatenate. Must be in the

Returns:

a new instance of Concat

See Also:

Concat

resourceApplyAdadelta

public <T> ResourceApplyAdadelta resourceApplyAdadelta(Operand<?> var,
                                                       Operand<?> accum,
                                                       Operand<?> accumUpdate,
                                                       Operand<T> lr,
                                                       Operand<T> rho,
                                                       Operand<T> epsilon,
                                                       Operand<T> grad,
                                                       ResourceApplyAdadelta.Options... options)

Adds an ResourceApplyAdadelta operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

accumUpdate - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay factor. Must be a scalar.

epsilon - Constant factor. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyAdadelta

See Also:

ResourceApplyAdadelta

randomShuffle

public <T> RandomShuffle<T> randomShuffle(Operand<T> value,
                                          RandomShuffle.Options... options)

Adds an RandomShuffle operation to the graph

Parameters:

value - The tensor to be shuffled.

options - carries optional attributes values

Returns:

a new instance of RandomShuffle

See Also:

RandomShuffle

mutableHashTableOfTensors

public <T,U> MutableHashTableOfTensors mutableHashTableOfTensors(Class<T> keyDtype,
                                                                 Class<U> valueDtype,
                                                                 MutableHashTableOfTensors.Options... options)

Adds an MutableHashTableOfTensors operation to the graph

Parameters:

keyDtype - Type of the table keys.

valueDtype - Type of the table values.

options - carries optional attributes values

Returns:

a new instance of MutableHashTableOfTensors

See Also:

MutableHashTableOfTensors

sparseDenseCwiseDiv

public <T> SparseDenseCwiseDiv<T> sparseDenseCwiseDiv(Operand<Long> spIndices,
                                                      Operand<T> spValues,
                                                      Operand<Long> spShape,
                                                      Operand<T> dense)

Adds an SparseDenseCwiseDiv operation to the graph

Parameters:

spIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

spValues - 1-D. `N` non-empty values corresponding to `sp_indices`.

spShape - 1-D. Shape of the input SparseTensor.

dense - `R`-D. The dense Tensor operand.

Returns:

a new instance of SparseDenseCwiseDiv

See Also:

SparseDenseCwiseDiv

encodeProto

public EncodeProto encodeProto(Operand<Integer> sizes,
                               Iterable<Operand<?>> values,
                               List<String> fieldNames,
                               String messageType,
                               EncodeProto.Options... options)

Adds an EncodeProto operation to the graph

Parameters:

sizes - Tensor of int32 with shape `[batch_shape, len(field_names)]`.

values - List of tensors containing values for the corresponding field.

fieldNames - List of strings containing proto field names.

messageType - Name of the proto message type to decode.

options - carries optional attributes values

Returns:

a new instance of EncodeProto

See Also:

EncodeProto

sparseReorder

public <T> SparseReorder<T> sparseReorder(Operand<Long> inputIndices,
                                          Operand<T> inputValues,
                                          Operand<Long> inputShape)

Adds an SparseReorder operation to the graph

Parameters:

inputIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

inputValues - 1-D. `N` non-empty values corresponding to `input_indices`.

inputShape - 1-D. Shape of the input SparseTensor.

Returns:

a new instance of SparseReorder

See Also:

SparseReorder

uniqueWithCountsV2

public <T,V extends Number,U extends Number> UniqueWithCountsV2<T,V> uniqueWithCountsV2(Operand<T> x,
                                                                                        Operand<U> axis,
                                                                                        Class<V> outIdx)

Adds an UniqueWithCountsV2 operation to the graph

Parameters:

x - A `Tensor`.

axis - A `Tensor` of type `int32` (default: None). The axis of the Tensor to

outIdx -

Returns:

a new instance of UniqueWithCountsV2

See Also:

UniqueWithCountsV2

batchMatMul

public <T> BatchMatMul<T> batchMatMul(Operand<T> x,
                                      Operand<T> y,
                                      BatchMatMul.Options... options)

Adds an BatchMatMul operation to the graph

Parameters:

x - 2-D or higher with shape `[..., r_x, c_x]`.

y - 2-D or higher with shape `[..., r_y, c_y]`.

options - carries optional attributes values

Returns:

a new instance of BatchMatMul

See Also:

BatchMatMul

unicodeTranscode

public UnicodeTranscode unicodeTranscode(Operand<String> input,
                                         String inputEncoding,
                                         String outputEncoding,
                                         UnicodeTranscode.Options... options)

Adds an UnicodeTranscode operation to the graph

Parameters:

input - The text to be processed. Can have any shape.

inputEncoding - Text encoding of the input strings. This is any of the encodings supported

outputEncoding - The unicode encoding to use in the output. Must be one of

options - carries optional attributes values

Returns:

a new instance of UnicodeTranscode

See Also:

UnicodeTranscode

resourceSparseApplyAdagrad

public <T,U extends Number> ResourceSparseApplyAdagrad resourceSparseApplyAdagrad(Operand<?> var,
                                                                                  Operand<?> accum,
                                                                                  Operand<T> lr,
                                                                                  Operand<T> grad,
                                                                                  Operand<U> indices,
                                                                                  ResourceSparseApplyAdagrad.Options... options)

Adds an ResourceSparseApplyAdagrad operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Learning rate. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyAdagrad

See Also:

ResourceSparseApplyAdagrad

squeeze

public <T> Squeeze<T> squeeze(Operand<T> input,
                              Squeeze.Options... options)

Adds an Squeeze operation to the graph

Parameters:

input - The `input` to squeeze.

options - carries optional attributes values

Returns:

a new instance of Squeeze

See Also:

Squeeze

deepCopy

public <T> DeepCopy<T> deepCopy(Operand<T> x)

Adds an DeepCopy operation to the graph

Parameters:

x - The source tensor of type `T`.

Returns:

a new instance of DeepCopy

See Also:

DeepCopy

constant

public Constant<Integer> constant(int data)

Adds an Constant operation to the graph

Parameters:

data - The value to put into the new constant.

Returns:

an integer constant

See Also:

Constant

conv3DBackpropFilterV2

public <T extends Number> Conv3DBackpropFilterV2<T> conv3DBackpropFilterV2(Operand<T> input,
                                                                           Operand<Integer> filterSizes,
                                                                           Operand<T> outBackprop,
                                                                           List<Long> strides,
                                                                           String padding,
                                                                           Conv3DBackpropFilterV2.Options... options)

Adds an Conv3DBackpropFilterV2 operation to the graph

Parameters:

input - Shape `[batch, depth, rows, cols, in_channels]`.

filterSizes - An integer vector representing the tensor shape of `filter`,

outBackprop - Backprop signal of shape `[batch, out_depth, out_rows, out_cols,

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of Conv3DBackpropFilterV2

See Also:

Conv3DBackpropFilterV2

constant

public Constant<Long> constant(long[] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

constant

public Constant<String> constant(byte[][][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. String elements are

See Also:

Constant

reduceMin

public <T,U extends Number> ReduceMin<T> reduceMin(Operand<T> input,
                                                   Operand<U> axis,
                                                   ReduceMin.Options... options)

Adds an ReduceMin operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of ReduceMin

See Also:

ReduceMin

regexReplace

public RegexReplace regexReplace(Operand<String> input,
                                 Operand<String> pattern,
                                 Operand<String> rewrite,
                                 RegexReplace.Options... options)

Adds an RegexReplace operation to the graph

Parameters:

input - The text to be processed.

pattern - The regular expression to match the input.

rewrite - The rewrite to be applied to the matched expresion.

options - carries optional attributes values

Returns:

a new instance of RegexReplace

See Also:

RegexReplace

quantizedMatMul

public <V,T,U,W> QuantizedMatMul<V> quantizedMatMul(Operand<T> a,
                                                    Operand<U> b,
                                                    Operand<Float> minA,
                                                    Operand<Float> maxA,
                                                    Operand<Float> minB,
                                                    Operand<Float> maxB,
                                                    Class<V> Toutput,
                                                    Class<W> Tactivation,
                                                    QuantizedMatMul.Options... options)

Adds an QuantizedMatMul operation to the graph

Parameters:

a - Must be a two-dimensional tensor.

b - Must be a two-dimensional tensor.

minA - The float value that the lowest quantized `a` value represents.

maxA - The float value that the highest quantized `a` value represents.

minB - The float value that the lowest quantized `b` value represents.

maxB - The float value that the highest quantized `b` value represents.

Toutput -

Tactivation - The type of output produced by activation function

options - carries optional attributes values

Returns:

a new instance of QuantizedMatMul

See Also:

QuantizedMatMul

deleteSessionTensor

public DeleteSessionTensor deleteSessionTensor(Operand<String> handle)

Adds an DeleteSessionTensor operation to the graph

Parameters:

handle - The handle for a tensor stored in the session state.

Returns:

a new instance of DeleteSessionTensor

See Also:

DeleteSessionTensor

resourceSparseApplyCenteredRMSProp

public <T,U extends Number> ResourceSparseApplyCenteredRMSProp resourceSparseApplyCenteredRMSProp(Operand<?> var,
                                                                                                  Operand<?> mg,
                                                                                                  Operand<?> ms,
                                                                                                  Operand<?> mom,
                                                                                                  Operand<T> lr,
                                                                                                  Operand<T> rho,
                                                                                                  Operand<T> momentum,
                                                                                                  Operand<T> epsilon,
                                                                                                  Operand<T> grad,
                                                                                                  Operand<U> indices,
                                                                                                  ResourceSparseApplyCenteredRMSProp.Options... options)

Adds an ResourceSparseApplyCenteredRMSProp operation to the graph

Parameters:

var - Should be from a Variable().

mg - Should be from a Variable().

ms - Should be from a Variable().

mom - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay rate. Must be a scalar.

momentum -

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var, ms and mom.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyCenteredRMSProp

See Also:

ResourceSparseApplyCenteredRMSProp

matrixDeterminant

public <T> MatrixDeterminant<T> matrixDeterminant(Operand<T> input)

Adds an MatrixDeterminant operation to the graph

Parameters:

input - Shape is `[..., M, M]`.

Returns:

a new instance of MatrixDeterminant

See Also:

MatrixDeterminant

tileGrad

public <T> TileGrad<T> tileGrad(Operand<T> input,
                                Operand<Integer> multiples)

Adds an TileGrad operation to the graph

Parameters:

input -

multiples -

Returns:

a new instance of TileGrad

See Also:

TileGrad

rightShift

public <T extends Number> RightShift<T> rightShift(Operand<T> x,
                                                   Operand<T> y)

Adds an RightShift operation to the graph

Parameters:

x -

y -

Returns:

a new instance of RightShift

See Also:

RightShift

readerSerializeState

public ReaderSerializeState readerSerializeState(Operand<?> readerHandle)

Adds an ReaderSerializeState operation to the graph

Parameters:

readerHandle - Handle to a Reader.

Returns:

a new instance of ReaderSerializeState

See Also:

ReaderSerializeState

save

public Save save(Operand<String> filename,
                 Operand<String> tensorNames,
                 Iterable<Operand<?>> data)

Adds an Save operation to the graph

Parameters:

filename - Must have a single element. The name of the file to which we write

tensorNames - Shape `[N]`. The names of the tensors to be saved.

data - `N` tensors to save.

Returns:

a new instance of Save

See Also:

Save

xlogy

public <T> Xlogy<T> xlogy(Operand<T> x,
                          Operand<T> y)

Adds an Xlogy operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Xlogy

See Also:

Xlogy

tensorListConcat

public <T> TensorListConcat<T> tensorListConcat(Operand<?> inputHandle,
                                                Class<T> elementDtype)

Adds an TensorListConcat operation to the graph

Parameters:

inputHandle -

elementDtype -

Returns:

a new instance of TensorListConcat

See Also:

TensorListConcat

imageSummary

public <T extends Number> ImageSummary imageSummary(Operand<String> tag,
                                                    Operand<T> tensor,
                                                    ImageSummary.Options... options)

Adds an ImageSummary operation to the graph

Parameters:

tag - Scalar. Used to build the `tag` attribute of the summary values.

tensor - 4-D of shape `[batch_size, height, width, channels]` where

options - carries optional attributes values

Returns:

a new instance of ImageSummary

See Also:

ImageSummary

tensorListSplit

public <T,U extends Number> TensorListSplit tensorListSplit(Operand<T> tensor,
                                                            Operand<U> elementShape,
                                                            Operand<Long> lengths)

Adds an TensorListSplit operation to the graph

Parameters:

tensor -

elementShape -

lengths -

Returns:

a new instance of TensorListSplit

See Also:

TensorListSplit

nextIteration

public <T> NextIteration<T> nextIteration(Operand<T> data)

Adds an NextIteration operation to the graph

Parameters:

data - The tensor to be made available to the next iteration.

Returns:

a new instance of NextIteration

See Also:

NextIteration

mutableDenseHashTable

public <T,U> MutableDenseHashTable mutableDenseHashTable(Operand<T> emptyKey,
                                                         Operand<T> deletedKey,
                                                         Class<U> valueDtype,
                                                         MutableDenseHashTable.Options... options)

Adds an MutableDenseHashTable operation to the graph

Parameters:

emptyKey - The key used to represent empty key buckets internally. Must not

deletedKey -

valueDtype - Type of the table values.

options - carries optional attributes values

Returns:

a new instance of MutableDenseHashTable

See Also:

MutableDenseHashTable

constant

public Constant<Integer> constant(long[] shape,
                                  IntBuffer data)

Adds an Constant operation to the graph

Parameters:

shape - the tensor shape.

data - a buffer containing the tensor data.

Returns:

an integer constant

Throws:

IllegalArgumentException - If the tensor shape is not compatible with the buffer

See Also:

Constant

quantizedResizeBilinear

public <T> QuantizedResizeBilinear<T> quantizedResizeBilinear(Operand<T> images,
                                                              Operand<Integer> size,
                                                              Operand<Float> min,
                                                              Operand<Float> max,
                                                              QuantizedResizeBilinear.Options... options)

Adds an QuantizedResizeBilinear operation to the graph

Parameters:

images - 4-D with shape `[batch, height, width, channels]`.

size - = A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The

min -

max -

options - carries optional attributes values

Returns:

a new instance of QuantizedResizeBilinear

See Also:

QuantizedResizeBilinear

randomCrop

public <T extends Number> RandomCrop<T> randomCrop(Operand<T> image,
                                                   Operand<Long> size,
                                                   RandomCrop.Options... options)

Adds an RandomCrop operation to the graph

Parameters:

image - 3-D of shape `[height, width, channels]`.

size - 1-D of length 2 containing: `crop_height`, `crop_width`..

options - carries optional attributes values

Returns:

a new instance of RandomCrop

See Also:

RandomCrop

constant

public Constant<Double> constant(double[][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

selfAdjointEig

public <T> SelfAdjointEig<T> selfAdjointEig(Operand<T> input,
                                            SelfAdjointEig.Options... options)

Adds an SelfAdjointEig operation to the graph

Parameters:

input - `Tensor` input of shape `[N, N]`.

options - carries optional attributes values

Returns:

a new instance of SelfAdjointEig

See Also:

SelfAdjointEig

maxPoolGradGradWithArgmax

public <T extends Number,U extends Number> MaxPoolGradGradWithArgmax<T> maxPoolGradGradWithArgmax(Operand<T> input,
                                                                                                  Operand<T> grad,
                                                                                                  Operand<U> argmax,
                                                                                                  List<Long> ksize,
                                                                                                  List<Long> strides,
                                                                                                  String padding)

Adds an MaxPoolGradGradWithArgmax operation to the graph

Parameters:

input - The original input.

grad - 4-D with shape `[batch, height, width, channels]`. Gradients w.r.t. the

argmax - The indices of the maximum values chosen for each output of `max_pool`.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the

padding - The type of padding algorithm to use.

Returns:

a new instance of MaxPoolGradGradWithArgmax

See Also:

MaxPoolGradGradWithArgmax

minimum

public <T extends Number> Minimum<T> minimum(Operand<T> x,
                                             Operand<T> y)

Adds an Minimum operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Minimum

See Also:

Minimum

constant

public Constant<Integer> constant(int[][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

atan2

public <T extends Number> Atan2<T> atan2(Operand<T> y,
                                         Operand<T> x)

Adds an Atan2 operation to the graph

Parameters:

y -

x -

Returns:

a new instance of Atan2

See Also:

Atan2

empty

public <T> Empty<T> empty(Operand<Integer> shape,
                          Class<T> dtype,
                          Empty.Options... options)

Adds an Empty operation to the graph

Parameters:

shape - 1-D. Represents the shape of the output tensor.

dtype -

options - carries optional attributes values

Returns:

a new instance of Empty

See Also:

Empty

resourceSparseApplyFtrlV2

public <T,U extends Number> ResourceSparseApplyFtrlV2 resourceSparseApplyFtrlV2(Operand<?> var,
                                                                                Operand<?> accum,
                                                                                Operand<?> linear,
                                                                                Operand<T> grad,
                                                                                Operand<U> indices,
                                                                                Operand<T> lr,
                                                                                Operand<T> l1,
                                                                                Operand<T> l2,
                                                                                Operand<T> l2Shrinkage,
                                                                                Operand<T> lrPower,
                                                                                ResourceSparseApplyFtrlV2.Options... options)

Adds an ResourceSparseApplyFtrlV2 operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

linear - Should be from a Variable().

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

lr - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 shrinkage regulariation. Must be a scalar.

l2Shrinkage -

lrPower - Scaling factor. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyFtrlV2

See Also:

ResourceSparseApplyFtrlV2

fFT2D

public <T> FFT2D<T> fFT2D(Operand<T> input)

Adds an FFT2D operation to the graph

Parameters:

input - A complex tensor.

Returns:

a new instance of FFT2D

See Also:

FFT2D

statelessRandomNormal

public <T extends Number,U extends Number> StatelessRandomNormal<Float> statelessRandomNormal(Operand<T> shape,
                                                                                              Operand<U> seed)

Adds an StatelessRandomNormal operation to the graph

Parameters:

shape - The shape of the output tensor.

seed - 2 seeds (shape [2]).

Returns:

a new instance of StatelessRandomNormal

See Also:

StatelessRandomNormal

mapPeek

public MapPeek mapPeek(Operand<Long> key,
                       Operand<Integer> indices,
                       List<Class<?>> dtypes,
                       MapPeek.Options... options)

Adds an MapPeek operation to the graph

Parameters:

key -

indices -

dtypes -

options - carries optional attributes values

Returns:

a new instance of MapPeek

See Also:

MapPeek

orderedMapUnstageNoKey

public OrderedMapUnstageNoKey orderedMapUnstageNoKey(Operand<Integer> indices,
                                                     List<Class<?>> dtypes,
                                                     OrderedMapUnstageNoKey.Options... options)

Adds an OrderedMapUnstageNoKey operation to the graph

Parameters:

indices -

dtypes -

options - carries optional attributes values

Returns:

a new instance of OrderedMapUnstageNoKey

See Also:

OrderedMapUnstageNoKey

dequantize

public <T> Dequantize dequantize(Operand<T> input,
                                 Operand<Float> minRange,
                                 Operand<Float> maxRange,
                                 Dequantize.Options... options)

Adds an Dequantize operation to the graph

Parameters:

input -

minRange - The minimum scalar value possibly produced for the input.

maxRange - The maximum scalar value possibly produced for the input.

options - carries optional attributes values

Returns:

a new instance of Dequantize

See Also:

Dequantize

sparseSegmentMean

public <T extends Number,U extends Number> SparseSegmentMean<T> sparseSegmentMean(Operand<T> data,
                                                                                  Operand<U> indices,
                                                                                  Operand<Integer> segmentIds)

Adds an SparseSegmentMean operation to the graph

Parameters:

data -

indices - A 1-D tensor. Has same rank as `segment_ids`.

segmentIds - A 1-D tensor. Values should be sorted and can be repeated.

Returns:

a new instance of SparseSegmentMean

See Also:

SparseSegmentMean

accumulatorTakeGradient

public <T> AccumulatorTakeGradient<T> accumulatorTakeGradient(Operand<String> handle,
                                                              Operand<Integer> numRequired,
                                                              Class<T> dtype)

Adds an AccumulatorTakeGradient operation to the graph

Parameters:

handle - The handle to an accumulator.

numRequired - Number of gradients required before we return an aggregate.

dtype - The data type of accumulated gradients. Needs to correspond to the type

Returns:

a new instance of AccumulatorTakeGradient

See Also:

AccumulatorTakeGradient

slice

public <T,U extends Number> Slice<T> slice(Operand<T> input,
                                           Operand<U> begin,
                                           Operand<U> size)

Adds an Slice operation to the graph

Parameters:

input -

begin - begin[i] specifies the offset into the 'i'th dimension of

size - size[i] specifies the number of elements of the 'i'th dimension

Returns:

a new instance of Slice

See Also:

Slice

skipgram

public Skipgram skipgram(String filename,
                         Long batchSize,
                         Skipgram.Options... options)

Adds an Skipgram operation to the graph

Parameters:

filename - The corpus's text file name.

batchSize - The size of produced batch.

options - carries optional attributes values

Returns:

a new instance of Skipgram

See Also:

Skipgram

generateVocabRemapping

public GenerateVocabRemapping generateVocabRemapping(Operand<String> newVocabFile,
                                                     Operand<String> oldVocabFile,
                                                     Long newVocabOffset,
                                                     Long numNewVocab,
                                                     GenerateVocabRemapping.Options... options)

Adds an GenerateVocabRemapping operation to the graph

Parameters:

newVocabFile - Path to the new vocab file.

oldVocabFile - Path to the old vocab file.

newVocabOffset - How many entries into the new vocab file to start reading.

numNewVocab - Number of entries in the new vocab file to remap.

options - carries optional attributes values

Returns:

a new instance of GenerateVocabRemapping

See Also:

GenerateVocabRemapping

notEqual

public <T> NotEqual notEqual(Operand<T> x,
                             Operand<T> y)

Adds an NotEqual operation to the graph

Parameters:

x -

y -

Returns:

a new instance of NotEqual

See Also:

NotEqual

uniqueWithCounts

public <T> UniqueWithCounts<T,Integer> uniqueWithCounts(Operand<T> x)

Adds an UniqueWithCounts operation to the graph

Parameters:

x - 1-D.

Returns:

a new instance of UniqueWithCounts

See Also:

UniqueWithCounts

fakeQuantWithMinMaxArgsGradient

public FakeQuantWithMinMaxArgsGradient fakeQuantWithMinMaxArgsGradient(Operand<Float> gradients,
                                                                       Operand<Float> inputs,
                                                                       FakeQuantWithMinMaxArgsGradient.Options... options)

Adds an FakeQuantWithMinMaxArgsGradient operation to the graph

Parameters:

gradients - Backpropagated gradients above the FakeQuantWithMinMaxArgs operation.

inputs - Values passed as inputs to the FakeQuantWithMinMaxArgs operation.

options - carries optional attributes values

Returns:

a new instance of FakeQuantWithMinMaxArgsGradient

See Also:

FakeQuantWithMinMaxArgsGradient

resourceScatterNdAdd

public <T extends Number,U> ResourceScatterNdAdd resourceScatterNdAdd(Operand<?> ref,
                                                                      Operand<T> indices,
                                                                      Operand<U> updates,
                                                                      ResourceScatterNdAdd.Options... options)

Adds an ResourceScatterNdAdd operation to the graph

Parameters:

ref - A resource handle. Must be from a VarHandleOp.

indices - A Tensor. Must be one of the following types: int32, int64.

updates - A Tensor. Must have the same type as ref. A tensor of

options - carries optional attributes values

Returns:

a new instance of ResourceScatterNdAdd

See Also:

ResourceScatterNdAdd

orderedMapStage

public OrderedMapStage orderedMapStage(Operand<Long> key,
                                       Operand<Integer> indices,
                                       Iterable<Operand<?>> values,
                                       List<Class<?>> dtypes,
                                       OrderedMapStage.Options... options)

Adds an OrderedMapStage operation to the graph

Parameters:

key - int64

indices -

values - a list of tensors

dtypes -

options - carries optional attributes values

Returns:

a new instance of OrderedMapStage

See Also:

OrderedMapStage

statelessRandomUniform

public <V extends Number,T extends Number,U extends Number> StatelessRandomUniform<V> statelessRandomUniform(Operand<T> shape,
                                                                                                             Operand<U> seed,
                                                                                                             Class<V> dtype)

Adds an StatelessRandomUniform operation to the graph

Parameters:

shape - The shape of the output tensor.

seed - 2 seeds (shape [2]).

dtype - The type of the output.

Returns:

a new instance of StatelessRandomUniform

See Also:

StatelessRandomUniform

softmaxCrossEntropyWithLogits

public <T extends Number> SoftmaxCrossEntropyWithLogits<T> softmaxCrossEntropyWithLogits(Operand<T> features,
                                                                                         Operand<T> labels)

Adds an SoftmaxCrossEntropyWithLogits operation to the graph

Parameters:

features - batch_size x num_classes matrix

labels - batch_size x num_classes matrix

Returns:

a new instance of SoftmaxCrossEntropyWithLogits

See Also:

SoftmaxCrossEntropyWithLogits

shapeN

public <U extends Number,T> ShapeN<U> shapeN(Operand<T> input,
                                             Class<U> outType)

Adds an ShapeN operation to the graph

Parameters:

input -

outType -

Returns:

a new instance of ShapeN

See Also:

ShapeN

extractJpegShape

public <T extends Number> ExtractJpegShape<T> extractJpegShape(Operand<String> contents,
                                                               Class<T> outputType)

Adds an ExtractJpegShape operation to the graph

Parameters:

contents - 0-D. The JPEG-encoded image.

outputType - (Optional) The output type of the operation (int32 or int64).

Returns:

a new instance of ExtractJpegShape

See Also:

ExtractJpegShape

print

public <T> Print<T> print(Operand<T> input,
                          Iterable<Operand<?>> data,
                          Print.Options... options)

Adds an Print operation to the graph

Parameters:

input - The tensor passed to `output`

data - A list of tensors to print out when op is evaluated.

options - carries optional attributes values

Returns:

a new instance of Print

See Also:

Print

resourceApplyProximalAdagrad

public <T> ResourceApplyProximalAdagrad resourceApplyProximalAdagrad(Operand<?> var,
                                                                     Operand<?> accum,
                                                                     Operand<T> lr,
                                                                     Operand<T> l1,
                                                                     Operand<T> l2,
                                                                     Operand<T> grad,
                                                                     ResourceApplyProximalAdagrad.Options... options)

Adds an ResourceApplyProximalAdagrad operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyProximalAdagrad

See Also:

ResourceApplyProximalAdagrad

identityReader

public IdentityReader identityReader(IdentityReader.Options... options)

Adds an IdentityReader operation to the graph

Parameters:

options - carries optional attributes values

Returns:

a new instance of IdentityReader

See Also:

IdentityReader

unsortedSegmentProd

public <T,U extends Number,V extends Number> UnsortedSegmentProd<T> unsortedSegmentProd(Operand<T> data,
                                                                                        Operand<U> segmentIds,
                                                                                        Operand<V> numSegments)

Adds an UnsortedSegmentProd operation to the graph

Parameters:

data -

segmentIds - A tensor whose shape is a prefix of `data.shape`.

numSegments -

Returns:

a new instance of UnsortedSegmentProd

See Also:

UnsortedSegmentProd

tan

public <T> Tan<T> tan(Operand<T> x)

Adds an Tan operation to the graph

Parameters:

x -

Returns:

a new instance of Tan

See Also:

Tan

loopCond

public LoopCond loopCond(Operand<Boolean> input)

Adds an LoopCond operation to the graph

Parameters:

input - A boolean scalar, representing the branch predicate of the Switch op.

Returns:

a new instance of LoopCond

See Also:

LoopCond

argMin

public <V extends Number,T,U extends Number> ArgMin<V> argMin(Operand<T> input,
                                                              Operand<U> dimension,
                                                              Class<V> outputType)

Adds an ArgMin operation to the graph

Parameters:

input -

dimension - int32 or int64, must be in the range `[-rank(input), rank(input))`.

outputType -

Returns:

a new instance of ArgMin

See Also:

ArgMin

batchToSpace

public <T,U extends Number> BatchToSpace<T> batchToSpace(Operand<T> input,
                                                         Operand<U> crops,
                                                         Long blockSize)

Adds an BatchToSpace operation to the graph

Parameters:

input - 4-D tensor with shape

crops - 2-D tensor of non-negative integers with shape `[2, 2]`. It specifies

blockSize -

Returns:

a new instance of BatchToSpace

See Also:

BatchToSpace

quantizedRelu

public <U,T> QuantizedRelu<U> quantizedRelu(Operand<T> features,
                                            Operand<Float> minFeatures,
                                            Operand<Float> maxFeatures,
                                            Class<U> outType)

Adds an QuantizedRelu operation to the graph

Parameters:

features -

minFeatures - The float value that the lowest quantized value represents.

maxFeatures - The float value that the highest quantized value represents.

outType -

Returns:

a new instance of QuantizedRelu

See Also:

QuantizedRelu

barrierReadySize

public BarrierReadySize barrierReadySize(Operand<String> handle)

Adds an BarrierReadySize operation to the graph

Parameters:

handle - The handle to a barrier.

Returns:

a new instance of BarrierReadySize

See Also:

BarrierReadySize

segmentMean

public <T,U extends Number> SegmentMean<T> segmentMean(Operand<T> data,
                                                       Operand<U> segmentIds)

Adds an SegmentMean operation to the graph

Parameters:

data -

segmentIds - A 1-D tensor whose size is equal to the size of `data`'s

Returns:

a new instance of SegmentMean

See Also:

SegmentMean

conv3DBackpropInput

public <T extends Number> Conv3DBackpropInput<T> conv3DBackpropInput(Operand<T> input,
                                                                     Operand<T> filter,
                                                                     Operand<T> outBackprop,
                                                                     List<Long> strides,
                                                                     String padding,
                                                                     Conv3DBackpropInput.Options... options)

Adds an Conv3DBackpropInput operation to the graph

Parameters:

input - Shape `[batch, depth, rows, cols, in_channels]`.

filter - Shape `[depth, rows, cols, in_channels, out_channels]`.

outBackprop - Backprop signal of shape `[batch, out_depth, out_rows, out_cols,

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of Conv3DBackpropInput

See Also:

Conv3DBackpropInput

cropAndResizeGradBoxes

public <T extends Number> CropAndResizeGradBoxes cropAndResizeGradBoxes(Operand<Float> grads,
                                                                        Operand<T> image,
                                                                        Operand<Float> boxes,
                                                                        Operand<Integer> boxInd,
                                                                        CropAndResizeGradBoxes.Options... options)

Adds an CropAndResizeGradBoxes operation to the graph

Parameters:

grads - A 4-D tensor of shape `[num_boxes, crop_height, crop_width, depth]`.

image - A 4-D tensor of shape `[batch, image_height, image_width, depth]`.

boxes - A 2-D tensor of shape `[num_boxes, 4]`. The `i`-th row of the tensor

boxInd - A 1-D tensor of shape `[num_boxes]` with int32 values in `[0, batch)`.

options - carries optional attributes values

Returns:

a new instance of CropAndResizeGradBoxes

See Also:

CropAndResizeGradBoxes

igammac

public <T extends Number> Igammac<T> igammac(Operand<T> a,
                                             Operand<T> x)

Adds an Igammac operation to the graph

Parameters:

a -

x -

Returns:

a new instance of Igammac

See Also:

Igammac

barrier

public Barrier barrier(List<Class<?>> componentTypes,
                       Barrier.Options... options)

Adds an Barrier operation to the graph

Parameters:

componentTypes - The type of each component in a value.

options - carries optional attributes values

Returns:

a new instance of Barrier

See Also:

Barrier

scatterAdd

public <T,U extends Number> ScatterAdd<T> scatterAdd(Operand<T> ref,
                                                     Operand<U> indices,
                                                     Operand<T> updates,
                                                     ScatterAdd.Options... options)

Adds an ScatterAdd operation to the graph

Parameters:

ref - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to add to `ref`.

options - carries optional attributes values

Returns:

a new instance of ScatterAdd

See Also:

ScatterAdd

constant

public Constant<Float> constant(float[][][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

shardedFilename

public ShardedFilename shardedFilename(Operand<String> basename,
                                       Operand<Integer> shard,
                                       Operand<Integer> numShards)

Adds an ShardedFilename operation to the graph

Parameters:

basename -

shard -

numShards -

Returns:

a new instance of ShardedFilename

See Also:

ShardedFilename

mean

public <T,U extends Number> Mean<T> mean(Operand<T> input,
                                         Operand<U> axis,
                                         Mean.Options... options)

Adds an Mean operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of Mean

See Also:

Mean

adjustHue

public AdjustHue adjustHue(Operand<Float> images,
                           Operand<Float> delta)

Adds an AdjustHue operation to the graph

Parameters:

images - Images to adjust. At least 3-D.

delta - A float delta to add to the hue.

Returns:

a new instance of AdjustHue

See Also:

AdjustHue

parallelConcat

public <T> ParallelConcat<T> parallelConcat(Operand<T> values,
                                            Shape shape)

Adds an ParallelConcat operation to the graph

Parameters:

values - Tensors to be concatenated. All must have size 1 in the first dimension

shape - the final shape of the result; should be equal to the shapes of any input

Returns:

a new instance of ParallelConcat

See Also:

ParallelConcat

resourceGather

public <U,T extends Number> ResourceGather<U> resourceGather(Operand<?> resource,
                                                             Operand<T> indices,
                                                             Class<U> dtype,
                                                             ResourceGather.Options... options)

Adds an ResourceGather operation to the graph

Parameters:

resource -

indices -

dtype -

options - carries optional attributes values

Returns:

a new instance of ResourceGather

See Also:

ResourceGather

constant

public Constant<Boolean> constant(boolean[][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

logicalAnd

public LogicalAnd logicalAnd(Operand<Boolean> x,
                             Operand<Boolean> y)

Adds an LogicalAnd operation to the graph

Parameters:

x -

y -

Returns:

a new instance of LogicalAnd

See Also:

LogicalAnd

decodeBase64

public DecodeBase64 decodeBase64(Operand<String> input)

Adds an DecodeBase64 operation to the graph

Parameters:

input - Base64 strings to decode.

Returns:

a new instance of DecodeBase64

See Also:

DecodeBase64

shapeN

public <T> ShapeN<Integer> shapeN(Operand<T> input)

Adds an ShapeN operation to the graph

Parameters:

input -

Returns:

a new instance of ShapeN

See Also:

ShapeN

requantizationRange

public <T> RequantizationRange requantizationRange(Operand<T> input,
                                                   Operand<Float> inputMin,
                                                   Operand<Float> inputMax)

Adds an RequantizationRange operation to the graph

Parameters:

input -

inputMin - The float value that the minimum quantized input value represents.

inputMax - The float value that the maximum quantized input value represents.

Returns:

a new instance of RequantizationRange

See Also:

RequantizationRange

conditionalAccumulator

public <T> ConditionalAccumulator conditionalAccumulator(Class<T> dtype,
                                                         Shape shape,
                                                         ConditionalAccumulator.Options... options)

Adds an ConditionalAccumulator operation to the graph

Parameters:

dtype - The type of the value being accumulated.

shape - The shape of the values, can be [], in which case shape is unknown.

options - carries optional attributes values

Returns:

a new instance of ConditionalAccumulator

See Also:

ConditionalAccumulator

applyFtrl

public <T> ApplyFtrl<T> applyFtrl(Operand<T> var,
                                  Operand<T> accum,
                                  Operand<T> linear,
                                  Operand<T> grad,
                                  Operand<T> lr,
                                  Operand<T> l1,
                                  Operand<T> l2,
                                  Operand<T> lrPower,
                                  ApplyFtrl.Options... options)

Adds an ApplyFtrl operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

linear - Should be from a Variable().

grad - The gradient.

lr - Scaling factor. Must be a scalar.

l1 - L1 regulariation. Must be a scalar.

l2 - L2 regulariation. Must be a scalar.

lrPower - Scaling factor. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ApplyFtrl

See Also:

ApplyFtrl

truncateMod

public <T extends Number> TruncateMod<T> truncateMod(Operand<T> x,
                                                     Operand<T> y)

Adds an TruncateMod operation to the graph

Parameters:

x -

y -

Returns:

a new instance of TruncateMod

See Also:

TruncateMod

constant

public Constant<Float> constant(float[][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

sparseMatMul

public <T extends Number,U extends Number> SparseMatMul sparseMatMul(Operand<T> a,
                                                                     Operand<U> b,
                                                                     SparseMatMul.Options... options)

Adds an SparseMatMul operation to the graph

Parameters:

a -

b -

options - carries optional attributes values

Returns:

a new instance of SparseMatMul

See Also:

SparseMatMul

unbatchGrad

public <T> UnbatchGrad<T> unbatchGrad(Operand<T> originalInput,
                                      Operand<Long> batchIndex,
                                      Operand<T> grad,
                                      Operand<Long> id,
                                      UnbatchGrad.Options... options)

Adds an UnbatchGrad operation to the graph

Parameters:

originalInput -

batchIndex -

grad -

id -

options - carries optional attributes values

Returns:

a new instance of UnbatchGrad

See Also:

UnbatchGrad

cudnnRNNParamsSize

public <U extends Number,T extends Number> CudnnRNNParamsSize<U> cudnnRNNParamsSize(Operand<Integer> numLayers,
                                                                                    Operand<Integer> numUnits,
                                                                                    Operand<Integer> inputSize,
                                                                                    Class<T> T,
                                                                                    Class<U> S,
                                                                                    CudnnRNNParamsSize.Options... options)

Adds an CudnnRNNParamsSize operation to the graph

Parameters:

numLayers -

numUnits -

inputSize -

T -

S -

options - carries optional attributes values

Returns:

a new instance of CudnnRNNParamsSize

See Also:

CudnnRNNParamsSize

quantizedConcat

public <T> QuantizedConcat<T> quantizedConcat(Operand<Integer> concatDim,
                                              Operand<T> values,
                                              Iterable<Operand<Float>> inputMins,
                                              Iterable<Operand<Float>> inputMaxes)

Adds an QuantizedConcat operation to the graph

Parameters:

concatDim - 0-D. The dimension along which to concatenate. Must be in the

values - The `N` Tensors to concatenate. Their ranks and types must match,

inputMins - The minimum scalar values for each of the input tensors.

inputMaxes - The maximum scalar values for each of the input tensors.

Returns:

a new instance of QuantizedConcat

See Also:

QuantizedConcat

xdivy

public <T> Xdivy<T> xdivy(Operand<T> x,
                          Operand<T> y)

Adds an Xdivy operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Xdivy

See Also:

Xdivy

batchMatrixSetDiag

public <T> BatchMatrixSetDiag<T> batchMatrixSetDiag(Operand<T> input,
                                                    Operand<T> diagonal)

Adds an BatchMatrixSetDiag operation to the graph

Parameters:

input -

diagonal -

Returns:

a new instance of BatchMatrixSetDiag

See Also:

BatchMatrixSetDiag

qr

public <T> Qr<T> qr(Operand<T> input,
                    Qr.Options... options)

Adds an Qr operation to the graph

Parameters:

input - A tensor of shape `[..., M, N]` whose inner-most 2 dimensions

options - carries optional attributes values

Returns:

a new instance of Qr

See Also:

Qr

log

public <T> Log<T> log(Operand<T> x)

Adds an Log operation to the graph

Parameters:

x -

Returns:

a new instance of Log

See Also:

Log

tensorScatterSub

public <T,U extends Number> TensorScatterSub<T> tensorScatterSub(Operand<T> tensor,
                                                                 Operand<U> indices,
                                                                 Operand<T> updates)

Adds an TensorScatterSub operation to the graph

Parameters:

tensor - Tensor to copy/update.

indices - Index tensor.

updates - Updates to scatter into output.

Returns:

a new instance of TensorScatterSub

See Also:

TensorScatterSub

invertPermutation

public <T extends Number> InvertPermutation<T> invertPermutation(Operand<T> x)

Adds an InvertPermutation operation to the graph

Parameters:

x - 1-D.

Returns:

a new instance of InvertPermutation

See Also:

InvertPermutation

applyAdam

public <T> ApplyAdam<T> applyAdam(Operand<T> var,
                                  Operand<T> m,
                                  Operand<T> v,
                                  Operand<T> beta1Power,
                                  Operand<T> beta2Power,
                                  Operand<T> lr,
                                  Operand<T> beta1,
                                  Operand<T> beta2,
                                  Operand<T> epsilon,
                                  Operand<T> grad,
                                  ApplyAdam.Options... options)

Adds an ApplyAdam operation to the graph

Parameters:

var - Should be from a Variable().

m - Should be from a Variable().

v - Should be from a Variable().

beta1Power - Must be a scalar.

beta2Power - Must be a scalar.

lr - Scaling factor. Must be a scalar.

beta1 - Momentum factor. Must be a scalar.

beta2 - Momentum factor. Must be a scalar.

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ApplyAdam

See Also:

ApplyAdam

fakeQuantWithMinMaxVarsGradient

public FakeQuantWithMinMaxVarsGradient fakeQuantWithMinMaxVarsGradient(Operand<Float> gradients,
                                                                       Operand<Float> inputs,
                                                                       Operand<Float> min,
                                                                       Operand<Float> max,
                                                                       FakeQuantWithMinMaxVarsGradient.Options... options)

Adds an FakeQuantWithMinMaxVarsGradient operation to the graph

Parameters:

gradients - Backpropagated gradients above the FakeQuantWithMinMaxVars operation.

inputs - Values passed as inputs to the FakeQuantWithMinMaxVars operation.

min -

max -

options - carries optional attributes values

Returns:

a new instance of FakeQuantWithMinMaxVarsGradient

See Also:

FakeQuantWithMinMaxVarsGradient

sparseFillEmptyRows

public <T> SparseFillEmptyRows<T> sparseFillEmptyRows(Operand<Long> indices,
                                                      Operand<T> values,
                                                      Operand<Long> denseShape,
                                                      Operand<T> defaultValue)

Adds an SparseFillEmptyRows operation to the graph

Parameters:

indices - 2-D. the indices of the sparse tensor.

values - 1-D. the values of the sparse tensor.

denseShape - 1-D. the shape of the sparse tensor.

defaultValue - 0-D. default value to insert into location `[row, 0, ..., 0]`

Returns:

a new instance of SparseFillEmptyRows

See Also:

SparseFillEmptyRows

sampleDistortedBoundingBox

public <T extends Number> SampleDistortedBoundingBox<T> sampleDistortedBoundingBox(Operand<T> imageSize,
                                                                                   Operand<Float> boundingBoxes,
                                                                                   SampleDistortedBoundingBox.Options... options)

Adds an SampleDistortedBoundingBox operation to the graph

Parameters:

imageSize - 1-D, containing `[height, width, channels]`.

boundingBoxes - 3-D with shape `[batch, N, 4]` describing the N bounding boxes

options - carries optional attributes values

Returns:

a new instance of SampleDistortedBoundingBox

See Also:

SampleDistortedBoundingBox

ceil

public <T extends Number> Ceil<T> ceil(Operand<T> x)

Adds an Ceil operation to the graph

Parameters:

x -

Returns:

a new instance of Ceil

See Also:

Ceil

complexAbs

public <T> ComplexAbs<Float> complexAbs(Operand<T> x)

Adds an ComplexAbs operation to the graph

Parameters:

x -

Returns:

a new instance of ComplexAbs

See Also:

ComplexAbs

stringJoin

public StringJoin stringJoin(Iterable<Operand<String>> inputs,
                             StringJoin.Options... options)

Adds an StringJoin operation to the graph

Parameters:

inputs - A list of string tensors. The tensors must all have the same shape,

options - carries optional attributes values

Returns:

a new instance of StringJoin

See Also:

StringJoin

readFile

public ReadFile readFile(Operand<String> filename)

Adds an ReadFile operation to the graph

Parameters:

filename -

Returns:

a new instance of ReadFile

See Also:

ReadFile

mutexV2

public MutexV2 mutexV2(MutexV2.Options... options)

Adds an MutexV2 operation to the graph

Parameters:

options - carries optional attributes values

Returns:

a new instance of MutexV2

See Also:

MutexV2

invert

public <T extends Number> Invert<T> invert(Operand<T> x)

Adds an Invert operation to the graph

Parameters:

x -

Returns:

a new instance of Invert

See Also:

Invert

sparseDenseCwiseMul

public <T> SparseDenseCwiseMul<T> sparseDenseCwiseMul(Operand<Long> spIndices,
                                                      Operand<T> spValues,
                                                      Operand<Long> spShape,
                                                      Operand<T> dense)

Adds an SparseDenseCwiseMul operation to the graph

Parameters:

spIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

spValues - 1-D. `N` non-empty values corresponding to `sp_indices`.

spShape - 1-D. Shape of the input SparseTensor.

dense - `R`-D. The dense Tensor operand.

Returns:

a new instance of SparseDenseCwiseMul

See Also:

SparseDenseCwiseMul

sparseApplyAdagrad

public <T,U extends Number> SparseApplyAdagrad<T> sparseApplyAdagrad(Operand<T> var,
                                                                     Operand<T> accum,
                                                                     Operand<T> lr,
                                                                     Operand<T> grad,
                                                                     Operand<U> indices,
                                                                     SparseApplyAdagrad.Options... options)

Adds an SparseApplyAdagrad operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Learning rate. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

options - carries optional attributes values

Returns:

a new instance of SparseApplyAdagrad

See Also:

SparseApplyAdagrad

reduceProd

public <T,U extends Number> ReduceProd<T> reduceProd(Operand<T> input,
                                                     Operand<U> axis,
                                                     ReduceProd.Options... options)

Adds an ReduceProd operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of ReduceProd

See Also:

ReduceProd

resourceScatterDiv

public <T extends Number,U> ResourceScatterDiv resourceScatterDiv(Operand<?> resource,
                                                                  Operand<T> indices,
                                                                  Operand<U> updates)

Adds an ResourceScatterDiv operation to the graph

Parameters:

resource - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to add to `ref`.

Returns:

a new instance of ResourceScatterDiv

See Also:

ResourceScatterDiv

randomPoissonV2

public <V extends Number,T extends Number,U extends Number> RandomPoissonV2<V> randomPoissonV2(Operand<T> shape,
                                                                                               Operand<U> rate,
                                                                                               Class<V> dtype,
                                                                                               RandomPoissonV2.Options... options)

Adds an RandomPoissonV2 operation to the graph

Parameters:

shape - 1-D integer tensor. Shape of independent samples to draw from each

rate - A tensor in which each scalar is a "rate" parameter describing the

dtype -

options - carries optional attributes values

Returns:

a new instance of RandomPoissonV2

See Also:

RandomPoissonV2

conv3DBackpropFilter

public <T extends Number> Conv3DBackpropFilter<T> conv3DBackpropFilter(Operand<T> input,
                                                                       Operand<T> filter,
                                                                       Operand<T> outBackprop,
                                                                       List<Long> strides,
                                                                       String padding,
                                                                       Conv3DBackpropFilter.Options... options)

Adds an Conv3DBackpropFilter operation to the graph

Parameters:

input - Shape `[batch, depth, rows, cols, in_channels]`.

filter - Shape `[depth, rows, cols, in_channels, out_channels]`.

outBackprop - Backprop signal of shape `[batch, out_depth, out_rows, out_cols,

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of Conv3DBackpropFilter

See Also:

Conv3DBackpropFilter

paddingFIFOQueue

public PaddingFIFOQueue paddingFIFOQueue(List<Class<?>> componentTypes,
                                         PaddingFIFOQueue.Options... options)

Adds an PaddingFIFOQueue operation to the graph

Parameters:

componentTypes - The type of each component in a value.

options - carries optional attributes values

Returns:

a new instance of PaddingFIFOQueue

See Also:

PaddingFIFOQueue

tensorSummary

public <T> TensorSummary tensorSummary(Operand<T> tensor,
                                       TensorSummary.Options... options)

Adds an TensorSummary operation to the graph

Parameters:

tensor - A tensor to serialize.

options - carries optional attributes values

Returns:

a new instance of TensorSummary

See Also:

TensorSummary

parallelDynamicStitch

public <T> ParallelDynamicStitch<T> parallelDynamicStitch(Iterable<Operand<Integer>> indices,
                                                          Operand<T> data)

Adds an ParallelDynamicStitch operation to the graph

Parameters:

indices -

data -

Returns:

a new instance of ParallelDynamicStitch

See Also:

ParallelDynamicStitch

lgamma

public <T extends Number> Lgamma<T> lgamma(Operand<T> x)

Adds an Lgamma operation to the graph

Parameters:

x -

Returns:

a new instance of Lgamma

See Also:

Lgamma

applyAdadelta

public <T> ApplyAdadelta<T> applyAdadelta(Operand<T> var,
                                          Operand<T> accum,
                                          Operand<T> accumUpdate,
                                          Operand<T> lr,
                                          Operand<T> rho,
                                          Operand<T> epsilon,
                                          Operand<T> grad,
                                          ApplyAdadelta.Options... options)

Adds an ApplyAdadelta operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

accumUpdate - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay factor. Must be a scalar.

epsilon - Constant factor. Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ApplyAdadelta

See Also:

ApplyAdadelta

resourceApplyKerasMomentum

public <T> ResourceApplyKerasMomentum resourceApplyKerasMomentum(Operand<?> var,
                                                                 Operand<?> accum,
                                                                 Operand<T> lr,
                                                                 Operand<T> grad,
                                                                 Operand<T> momentum,
                                                                 ResourceApplyKerasMomentum.Options... options)

Adds an ResourceApplyKerasMomentum operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

grad - The gradient.

momentum - Momentum. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyKerasMomentum

See Also:

ResourceApplyKerasMomentum

hashTable

public <T,U> HashTable hashTable(Class<T> keyDtype,
                                 Class<U> valueDtype,
                                 HashTable.Options... options)

Adds an HashTable operation to the graph

Parameters:

keyDtype - Type of the table keys.

valueDtype - Type of the table values.

options - carries optional attributes values

Returns:

a new instance of HashTable

See Also:

HashTable

constant

public Constant<Boolean> constant(boolean[][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

fusedResizeAndPadConv2D

public <T extends Number> FusedResizeAndPadConv2D<T> fusedResizeAndPadConv2D(Operand<T> input,
                                                                             Operand<Integer> size,
                                                                             Operand<Integer> paddings,
                                                                             Operand<T> filter,
                                                                             String mode,
                                                                             List<Long> strides,
                                                                             String padding,
                                                                             FusedResizeAndPadConv2D.Options... options)

Adds an FusedResizeAndPadConv2D operation to the graph

Parameters:

input - 4-D with shape `[batch, in_height, in_width, in_channels]`.

size - A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The

paddings - A two-column matrix specifying the padding sizes. The number of

filter - 4-D with shape

mode -

strides - 1-D of length 4. The stride of the sliding window for each dimension

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of FusedResizeAndPadConv2D

See Also:

FusedResizeAndPadConv2D

nonMaxSuppressionV4

public <T extends Number> NonMaxSuppressionV4 nonMaxSuppressionV4(Operand<T> boxes,
                                                                  Operand<T> scores,
                                                                  Operand<Integer> maxOutputSize,
                                                                  Operand<Float> iouThreshold,
                                                                  Operand<Float> scoreThreshold,
                                                                  NonMaxSuppressionV4.Options... options)

Adds an NonMaxSuppressionV4 operation to the graph

Parameters:

boxes - A 2-D float tensor of shape `[num_boxes, 4]`.

scores - A 1-D float tensor of shape `[num_boxes]` representing a single

maxOutputSize - A scalar integer tensor representing the maximum number of

iouThreshold - A 0-D float tensor representing the threshold for deciding whether

scoreThreshold - A 0-D float tensor representing the threshold for deciding when to remove

options - carries optional attributes values

Returns:

a new instance of NonMaxSuppressionV4

See Also:

NonMaxSuppressionV4

unstack

public <T> Unstack<T> unstack(Operand<T> value,
                              Long num,
                              Unstack.Options... options)

Adds an Unstack operation to the graph

Parameters:

value - 1-D or higher, with `axis` dimension size equal to `num`.

num -

options - carries optional attributes values

Returns:

a new instance of Unstack

See Also:

Unstack

softmax

public <T extends Number> Softmax<T> softmax(Operand<T> logits)

Adds an Softmax operation to the graph

Parameters:

logits - 2-D with shape `[batch_size, num_classes]`.

Returns:

a new instance of Softmax

See Also:

Softmax

batchIFFT2D

public BatchIFFT2D batchIFFT2D(Operand<?> input)

Adds an BatchIFFT2D operation to the graph

Parameters:

input -

Returns:

a new instance of BatchIFFT2D

See Also:

BatchIFFT2D

expandDims

public <T,U extends Number> ExpandDims<T> expandDims(Operand<T> input,
                                                     Operand<U> axis)

Adds an ExpandDims operation to the graph

Parameters:

input -

axis - 0-D (scalar). Specifies the dimension index at which to

Returns:

a new instance of ExpandDims

See Also:

ExpandDims

reduceAny

public <T extends Number> ReduceAny reduceAny(Operand<Boolean> input,
                                              Operand<T> axis,
                                              ReduceAny.Options... options)

Adds an ReduceAny operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of ReduceAny

See Also:

ReduceAny

rFFT3D

public RFFT3D rFFT3D(Operand<Float> input,
                     Operand<Integer> fftLength)

Adds an RFFT3D operation to the graph

Parameters:

input - A float32 tensor.

fftLength - An int32 tensor of shape [3]. The FFT length for each dimension.

Returns:

a new instance of RFFT3D

See Also:

RFFT3D

queueEnqueue

public QueueEnqueue queueEnqueue(Operand<?> handle,
                                 Iterable<Operand<?>> components,
                                 QueueEnqueue.Options... options)

Adds an QueueEnqueue operation to the graph

Parameters:

handle - The handle to a queue.

components - One or more tensors from which the enqueued tensors should be taken.

options - carries optional attributes values

Returns:

a new instance of QueueEnqueue

See Also:

QueueEnqueue

multiply

public <T> Multiply<T> multiply(Operand<T> x,
                                Operand<T> y)

Adds an Multiply operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Multiply

See Also:

Multiply

reduceSum

public <T,U extends Number> ReduceSum<T> reduceSum(Operand<T> input,
                                                   Operand<U> axis,
                                                   ReduceSum.Options... options)

Adds an ReduceSum operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of ReduceSum

See Also:

ReduceSum

sparseApplyRMSProp

public <T,U extends Number> SparseApplyRMSProp<T> sparseApplyRMSProp(Operand<T> var,
                                                                     Operand<T> ms,
                                                                     Operand<T> mom,
                                                                     Operand<T> lr,
                                                                     Operand<T> rho,
                                                                     Operand<T> momentum,
                                                                     Operand<T> epsilon,
                                                                     Operand<T> grad,
                                                                     Operand<U> indices,
                                                                     SparseApplyRMSProp.Options... options)

Adds an SparseApplyRMSProp operation to the graph

Parameters:

var - Should be from a Variable().

ms - Should be from a Variable().

mom - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

rho - Decay rate. Must be a scalar.

momentum -

epsilon - Ridge term. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var, ms and mom.

options - carries optional attributes values

Returns:

a new instance of SparseApplyRMSProp

See Also:

SparseApplyRMSProp

logUniformCandidateSampler

public LogUniformCandidateSampler logUniformCandidateSampler(Operand<Long> trueClasses,
                                                             Long numTrue,
                                                             Long numSampled,
                                                             Boolean unique,
                                                             Long rangeMax,
                                                             LogUniformCandidateSampler.Options... options)

Adds an LogUniformCandidateSampler operation to the graph

Parameters:

trueClasses - A batch_size * num_true matrix, in which each row contains the

numTrue - Number of true labels per context.

numSampled - Number of candidates to randomly sample.

unique - If unique is true, we sample with rejection, so that all sampled

rangeMax - The sampler will sample integers from the interval [0, range_max).

options - carries optional attributes values

Returns:

a new instance of LogUniformCandidateSampler

See Also:

LogUniformCandidateSampler

stringFormat

public StringFormat stringFormat(Iterable<Operand<?>> inputs,
                                 StringFormat.Options... options)

Adds an StringFormat operation to the graph

Parameters:

inputs - The list of tensors to format into the placeholder string.

options - carries optional attributes values

Returns:

a new instance of StringFormat

See Also:

StringFormat

readVariableOp

public <T> ReadVariableOp<T> readVariableOp(Operand<?> resource,
                                            Class<T> dtype)

Adds an ReadVariableOp operation to the graph

Parameters:

resource - handle to the resource in which to store the variable.

dtype - the dtype of the value.

Returns:

a new instance of ReadVariableOp

See Also:

ReadVariableOp

statelessTruncatedNormal

public <V extends Number,T extends Number,U extends Number> StatelessTruncatedNormal<V> statelessTruncatedNormal(Operand<T> shape,
                                                                                                                 Operand<U> seed,
                                                                                                                 Class<V> dtype)

Adds an StatelessTruncatedNormal operation to the graph

Parameters:

shape - The shape of the output tensor.

seed - 2 seeds (shape [2]).

dtype - The type of the output.

Returns:

a new instance of StatelessTruncatedNormal

See Also:

StatelessTruncatedNormal

generateBigQueryReaderPartitions

public GenerateBigQueryReaderPartitions generateBigQueryReaderPartitions(String projectId,
                                                                         String datasetId,
                                                                         String tableId,
                                                                         List<String> columns,
                                                                         Long timestampMillis,
                                                                         Long numPartitions,
                                                                         GenerateBigQueryReaderPartitions.Options... options)

Adds an GenerateBigQueryReaderPartitions operation to the graph

Parameters:

projectId - GCP project ID.

datasetId - BigQuery Dataset ID.

tableId - Table to read.

columns - List of columns to read. Leave empty to read all columns.

timestampMillis - Table snapshot timestamp in millis since epoch. Relative

numPartitions - Number of partitions to split the table into.

options - carries optional attributes values

Returns:

a new instance of GenerateBigQueryReaderPartitions

See Also:

GenerateBigQueryReaderPartitions

fFT3D

public <T> FFT3D<T> fFT3D(Operand<T> input)

Adds an FFT3D operation to the graph

Parameters:

input - A complex64 tensor.

Returns:

a new instance of FFT3D

See Also:

FFT3D

matrixSetDiag

public <T> MatrixSetDiag<T> matrixSetDiag(Operand<T> input,
                                          Operand<T> diagonal)

Adds an MatrixSetDiag operation to the graph

Parameters:

input - Rank `k+1`, where `k >= 1`.

diagonal - Rank `k`, where `k >= 1`.

Returns:

a new instance of MatrixSetDiag

See Also:

MatrixSetDiag

logicalNot

public LogicalNot logicalNot(Operand<Boolean> x)

Adds an LogicalNot operation to the graph

Parameters:

x -

Returns:

a new instance of LogicalNot

See Also:

LogicalNot

adjustContrast

public AdjustContrast adjustContrast(Operand<Float> images,
                                     Operand<Float> contrastFactor)

Adds an AdjustContrast operation to the graph

Parameters:

images - Images to adjust. At least 3-D.

contrastFactor - A float multiplier for adjusting contrast.

Returns:

a new instance of AdjustContrast

See Also:

AdjustContrast

resourceApplyAdagradDA

public <T> ResourceApplyAdagradDA resourceApplyAdagradDA(Operand<?> var,
                                                         Operand<?> gradientAccumulator,
                                                         Operand<?> gradientSquaredAccumulator,
                                                         Operand<T> grad,
                                                         Operand<T> lr,
                                                         Operand<T> l1,
                                                         Operand<T> l2,
                                                         Operand<Long> globalStep,
                                                         ResourceApplyAdagradDA.Options... options)

Adds an ResourceApplyAdagradDA operation to the graph

Parameters:

var - Should be from a Variable().

gradientAccumulator - Should be from a Variable().

gradientSquaredAccumulator - Should be from a Variable().

grad - The gradient.

lr - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

globalStep - Training step number. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyAdagradDA

See Also:

ResourceApplyAdagradDA

restoreV2

public RestoreV2 restoreV2(Operand<String> prefix,
                           Operand<String> tensorNames,
                           Operand<String> shapeAndSlices,
                           List<Class<?>> dtypes)

Adds an RestoreV2 operation to the graph

Parameters:

prefix - Must have a single element. The prefix of a V2 checkpoint.

tensorNames - shape {N}. The names of the tensors to be restored.

shapeAndSlices - shape {N}. The slice specs of the tensors to be restored.

dtypes - shape {N}. The list of expected dtype for the tensors. Must match

Returns:

a new instance of RestoreV2

See Also:

RestoreV2

lookupTableSize

public LookupTableSize lookupTableSize(Operand<?> tableHandle)

Adds an LookupTableSize operation to the graph

Parameters:

tableHandle - Handle to the table.

Returns:

a new instance of LookupTableSize

See Also:

LookupTableSize

zeta

public <T extends Number> Zeta<T> zeta(Operand<T> x,
                                       Operand<T> q)

Adds an Zeta operation to the graph

Parameters:

x -

q -

Returns:

a new instance of Zeta

See Also:

Zeta

sparseSegmentMeanGrad

public <T extends Number,U extends Number> SparseSegmentMeanGrad<T> sparseSegmentMeanGrad(Operand<T> grad,
                                                                                          Operand<U> indices,
                                                                                          Operand<Integer> segmentIds,
                                                                                          Operand<Integer> outputDim0)

Adds an SparseSegmentMeanGrad operation to the graph

Parameters:

grad - gradient propagated to the SparseSegmentMean op.

indices - indices passed to the corresponding SparseSegmentMean op.

segmentIds - segment_ids passed to the corresponding SparseSegmentMean op.

outputDim0 - dimension 0 of "data" passed to SparseSegmentMean op.

Returns:

a new instance of SparseSegmentMeanGrad

See Also:

SparseSegmentMeanGrad

digamma

public <T extends Number> Digamma<T> digamma(Operand<T> x)

Adds an Digamma operation to the graph

Parameters:

x -

Returns:

a new instance of Digamma

See Also:

Digamma

stringToHashBucketFast

public StringToHashBucketFast stringToHashBucketFast(Operand<String> input,
                                                     Long numBuckets)

Adds an StringToHashBucketFast operation to the graph

Parameters:

input - The strings to assign a hash bucket.

numBuckets - The number of buckets.

Returns:

a new instance of StringToHashBucketFast

See Also:

StringToHashBucketFast

floorMod

public <T extends Number> FloorMod<T> floorMod(Operand<T> x,
                                               Operand<T> y)

Adds an FloorMod operation to the graph

Parameters:

x -

y -

Returns:

a new instance of FloorMod

See Also:

FloorMod

parseSingleExample

public ParseSingleExample parseSingleExample(Operand<String> serialized,
                                             Iterable<Operand<?>> denseDefaults,
                                             Long numSparse,
                                             List<String> sparseKeys,
                                             List<String> denseKeys,
                                             List<Class<?>> sparseTypes,
                                             List<Shape> denseShapes)

Adds an ParseSingleExample operation to the graph

Parameters:

serialized - A vector containing a batch of binary serialized Example protos.

denseDefaults - A list of Tensors (some may be empty), whose length matches

numSparse - The number of sparse features to be parsed from the example. This

sparseKeys - A list of `num_sparse` strings.

denseKeys - The keys expected in the Examples' features associated with dense

sparseTypes - A list of `num_sparse` types; the data types of data in each

denseShapes - The shapes of data in each Feature given in dense_keys.

Returns:

a new instance of ParseSingleExample

See Also:

ParseSingleExample

gatherNd

public <T,U extends Number> GatherNd<T> gatherNd(Operand<T> params,
                                                 Operand<U> indices)

Adds an GatherNd operation to the graph

Parameters:

params - The tensor from which to gather values.

indices - Index tensor.

Returns:

a new instance of GatherNd

See Also:

GatherNd

accumulatorNumAccumulated

public AccumulatorNumAccumulated accumulatorNumAccumulated(Operand<String> handle)

Adds an AccumulatorNumAccumulated operation to the graph

Parameters:

handle - The handle to an accumulator.

Returns:

a new instance of AccumulatorNumAccumulated

See Also:

AccumulatorNumAccumulated

resizeBicubic

public <T extends Number> ResizeBicubic resizeBicubic(Operand<T> images,
                                                      Operand<Integer> size,
                                                      ResizeBicubic.Options... options)

Adds an ResizeBicubic operation to the graph

Parameters:

images - 4-D with shape `[batch, height, width, channels]`.

size - = A 1-D int32 Tensor of 2 elements: `new_height, new_width`. The

options - carries optional attributes values

Returns:

a new instance of ResizeBicubic

See Also:

ResizeBicubic

refNextIteration

public <T> RefNextIteration<T> refNextIteration(Operand<T> data)

Adds an RefNextIteration operation to the graph

Parameters:

data - The tensor to be made available to the next iteration.

Returns:

a new instance of RefNextIteration

See Also:

RefNextIteration

rsqrt

public <T> Rsqrt<T> rsqrt(Operand<T> x)

Adds an Rsqrt operation to the graph

Parameters:

x -

Returns:

a new instance of Rsqrt

See Also:

Rsqrt

fusedPadConv2D

public <T extends Number> FusedPadConv2D<T> fusedPadConv2D(Operand<T> input,
                                                           Operand<Integer> paddings,
                                                           Operand<T> filter,
                                                           String mode,
                                                           List<Long> strides,
                                                           String padding)

Adds an FusedPadConv2D operation to the graph

Parameters:

input - 4-D with shape `[batch, in_height, in_width, in_channels]`.

paddings - A two-column matrix specifying the padding sizes. The number of

filter - 4-D with shape

mode -

strides - 1-D of length 4. The stride of the sliding window for each dimension

padding - The type of padding algorithm to use.

Returns:

a new instance of FusedPadConv2D

See Also:

FusedPadConv2D

unstage

public Unstage unstage(List<Class<?>> dtypes,
                       Unstage.Options... options)

Adds an Unstage operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of Unstage

See Also:

Unstage

cast

public <U,T> Cast<U> cast(Operand<T> x,
                          Class<U> DstT,
                          Cast.Options... options)

Adds an Cast operation to the graph

Parameters:

x -

DstT -

options - carries optional attributes values

Returns:

a new instance of Cast

See Also:

Cast

editDistance

public <T> EditDistance editDistance(Operand<Long> hypothesisIndices,
                                     Operand<T> hypothesisValues,
                                     Operand<Long> hypothesisShape,
                                     Operand<Long> truthIndices,
                                     Operand<T> truthValues,
                                     Operand<Long> truthShape,
                                     EditDistance.Options... options)

Adds an EditDistance operation to the graph

Parameters:

hypothesisIndices - The indices of the hypothesis list SparseTensor.

hypothesisValues - The values of the hypothesis list SparseTensor.

hypothesisShape - The shape of the hypothesis list SparseTensor.

truthIndices - The indices of the truth list SparseTensor.

truthValues - The values of the truth list SparseTensor.

truthShape - truth indices, vector.

options - carries optional attributes values

Returns:

a new instance of EditDistance

See Also:

EditDistance

relu6

public <T extends Number> Relu6<T> relu6(Operand<T> features)

Adds an Relu6 operation to the graph

Parameters:

features -

Returns:

a new instance of Relu6

See Also:

Relu6

shape

public <T> Shape<Integer> shape(Operand<T> input)

Adds an Shape operation to the graph

Parameters:

input -

Returns:

a new instance of Shape

See Also:

Shape

mapUnstage

public MapUnstage mapUnstage(Operand<Long> key,
                             Operand<Integer> indices,
                             List<Class<?>> dtypes,
                             MapUnstage.Options... options)

Adds an MapUnstage operation to the graph

Parameters:

key -

indices -

dtypes -

options - carries optional attributes values

Returns:

a new instance of MapUnstage

See Also:

MapUnstage

orderedMapClear

public OrderedMapClear orderedMapClear(List<Class<?>> dtypes,
                                       OrderedMapClear.Options... options)

Adds an OrderedMapClear operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of OrderedMapClear

See Also:

OrderedMapClear

fakeQuantWithMinMaxVarsPerChannelGradient

public FakeQuantWithMinMaxVarsPerChannelGradient fakeQuantWithMinMaxVarsPerChannelGradient(Operand<Float> gradients,
                                                                                           Operand<Float> inputs,
                                                                                           Operand<Float> min,
                                                                                           Operand<Float> max,
                                                                                           FakeQuantWithMinMaxVarsPerChannelGradient.Options... options)

Adds an FakeQuantWithMinMaxVarsPerChannelGradient operation to the graph

Parameters:

gradients - Backpropagated gradients above the FakeQuantWithMinMaxVars operation,

inputs - Values passed as inputs to the FakeQuantWithMinMaxVars operation, shape

min -

max -

options - carries optional attributes values

Returns:

a new instance of FakeQuantWithMinMaxVarsPerChannelGradient

See Also:

FakeQuantWithMinMaxVarsPerChannelGradient

bitwiseOr

public <T extends Number> BitwiseOr<T> bitwiseOr(Operand<T> x,
                                                 Operand<T> y)

Adds an BitwiseOr operation to the graph

Parameters:

x -

y -

Returns:

a new instance of BitwiseOr

See Also:

BitwiseOr

readerNumRecordsProduced

public ReaderNumRecordsProduced readerNumRecordsProduced(Operand<?> readerHandle)

Adds an ReaderNumRecordsProduced operation to the graph

Parameters:

readerHandle - Handle to a Reader.

Returns:

a new instance of ReaderNumRecordsProduced

See Also:

ReaderNumRecordsProduced

tensorListElementShape

public <T extends Number> TensorListElementShape<T> tensorListElementShape(Operand<?> inputHandle,
                                                                           Class<T> shapeType)

Adds an TensorListElementShape operation to the graph

Parameters:

inputHandle -

shapeType -

Returns:

a new instance of TensorListElementShape

See Also:

TensorListElementShape

softsign

public <T extends Number> Softsign<T> softsign(Operand<T> features)

Adds an Softsign operation to the graph

Parameters:

features -

Returns:

a new instance of Softsign

See Also:

Softsign

squaredDifference

public <T> SquaredDifference<T> squaredDifference(Operand<T> x,
                                                  Operand<T> y)

Adds an SquaredDifference operation to the graph

Parameters:

x -

y -

Returns:

a new instance of SquaredDifference

See Also:

SquaredDifference

batchMatrixDiag

public <T> BatchMatrixDiag<T> batchMatrixDiag(Operand<T> diagonal)

Adds an BatchMatrixDiag operation to the graph

Parameters:

diagonal -

Returns:

a new instance of BatchMatrixDiag

See Also:

BatchMatrixDiag

scatterDiv

public <T,U extends Number> ScatterDiv<T> scatterDiv(Operand<T> ref,
                                                     Operand<U> indices,
                                                     Operand<T> updates,
                                                     ScatterDiv.Options... options)

Adds an ScatterDiv operation to the graph

Parameters:

ref - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of values that `ref` is divided by.

options - carries optional attributes values

Returns:

a new instance of ScatterDiv

See Also:

ScatterDiv

quantizedReshape

public <T,U extends Number> QuantizedReshape<T> quantizedReshape(Operand<T> tensor,
                                                                 Operand<U> shape,
                                                                 Operand<Float> inputMin,
                                                                 Operand<Float> inputMax)

Adds an QuantizedReshape operation to the graph

Parameters:

tensor -

shape - Defines the shape of the output tensor.

inputMin - The minimum value of the input.

inputMax - The maximum value of the input.

Returns:

a new instance of QuantizedReshape

See Also:

QuantizedReshape

logicalOr

public LogicalOr logicalOr(Operand<Boolean> x,
                           Operand<Boolean> y)

Adds an LogicalOr operation to the graph

Parameters:

x -

y -

Returns:

a new instance of LogicalOr

See Also:

LogicalOr

barrierTakeMany

public BarrierTakeMany barrierTakeMany(Operand<String> handle,
                                       Operand<Integer> numElements,
                                       List<Class<?>> componentTypes,
                                       BarrierTakeMany.Options... options)

Adds an BarrierTakeMany operation to the graph

Parameters:

handle - The handle to a barrier.

numElements - A single-element tensor containing the number of elements to

componentTypes - The type of each component in a value.

options - carries optional attributes values

Returns:

a new instance of BarrierTakeMany

See Also:

BarrierTakeMany

real

public <T> Real<Float> real(Operand<T> input)

Adds an Real operation to the graph

Parameters:

input -

Returns:

a new instance of Real

See Also:

Real

cumprod

public <T,U extends Number> Cumprod<T> cumprod(Operand<T> x,
                                               Operand<U> axis,
                                               Cumprod.Options... options)

Adds an Cumprod operation to the graph

Parameters:

x - A `Tensor`. Must be one of the following types: `float32`, `float64`,

axis - A `Tensor` of type `int32` (default: 0). Must be in the range

options - carries optional attributes values

Returns:

a new instance of Cumprod

See Also:

Cumprod

real

public <U extends Number,T> Real<U> real(Operand<T> input,
                                         Class<U> Tout)

Adds an Real operation to the graph

Parameters:

input -

Tout -

Returns:

a new instance of Real

See Also:

Real

resourceApplyGradientDescent

public <T> ResourceApplyGradientDescent resourceApplyGradientDescent(Operand<?> var,
                                                                     Operand<T> alpha,
                                                                     Operand<T> delta,
                                                                     ResourceApplyGradientDescent.Options... options)

Adds an ResourceApplyGradientDescent operation to the graph

Parameters:

var - Should be from a Variable().

alpha - Scaling factor. Must be a scalar.

delta - The change.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyGradientDescent

See Also:

ResourceApplyGradientDescent

maxPoolGradGradV2

public <T extends Number> MaxPoolGradGradV2<T> maxPoolGradGradV2(Operand<T> origInput,
                                                                 Operand<T> origOutput,
                                                                 Operand<T> grad,
                                                                 Operand<Integer> ksize,
                                                                 Operand<Integer> strides,
                                                                 String padding,
                                                                 MaxPoolGradGradV2.Options... options)

Adds an MaxPoolGradGradV2 operation to the graph

Parameters:

origInput - The original input tensor.

origOutput - The original output tensor.

grad - 4-D. Gradients of gradients w.r.t. the input of `max_pool`.

ksize - The size of the window for each dimension of the input tensor.

strides - The stride of the sliding window for each dimension of the

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of MaxPoolGradGradV2

See Also:

MaxPoolGradGradV2

optionalGetValue

public OptionalGetValue optionalGetValue(Operand<?> optional,
                                         List<Class<?>> outputTypes,
                                         List<Shape> outputShapes)

Adds an OptionalGetValue operation to the graph

Parameters:

optional -

outputTypes -

outputShapes -

Returns:

a new instance of OptionalGetValue

See Also:

OptionalGetValue

tensorListGetItem

public <T> TensorListGetItem<T> tensorListGetItem(Operand<?> inputHandle,
                                                  Operand<Integer> index,
                                                  Class<T> elementDtype)

Adds an TensorListGetItem operation to the graph

Parameters:

inputHandle -

index -

elementDtype -

Returns:

a new instance of TensorListGetItem

See Also:

TensorListGetItem

nonMaxSuppressionWithOverlaps

public NonMaxSuppressionWithOverlaps nonMaxSuppressionWithOverlaps(Operand<Float> overlaps,
                                                                   Operand<Float> scores,
                                                                   Operand<Integer> maxOutputSize,
                                                                   Operand<Float> overlapThreshold,
                                                                   Operand<Float> scoreThreshold)

Adds an NonMaxSuppressionWithOverlaps operation to the graph

Parameters:

overlaps - A 2-D float tensor of shape `[num_boxes, num_boxes]` representing

scores - A 1-D float tensor of shape `[num_boxes]` representing a single

maxOutputSize - A scalar integer tensor representing the maximum number of

overlapThreshold - A 0-D float tensor representing the threshold for deciding whether

scoreThreshold - A 0-D float tensor representing the threshold for deciding when to remove

Returns:

a new instance of NonMaxSuppressionWithOverlaps

See Also:

NonMaxSuppressionWithOverlaps

resourceApplyPowerSign

public <T> ResourceApplyPowerSign resourceApplyPowerSign(Operand<?> var,
                                                         Operand<?> m,
                                                         Operand<T> lr,
                                                         Operand<T> logbase,
                                                         Operand<T> signDecay,
                                                         Operand<T> beta,
                                                         Operand<T> grad,
                                                         ResourceApplyPowerSign.Options... options)

Adds an ResourceApplyPowerSign operation to the graph

Parameters:

var - Should be from a Variable().

m - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

logbase - Must be a scalar.

signDecay - Must be a scalar.

beta - Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyPowerSign

See Also:

ResourceApplyPowerSign

applyMomentum

public <T> ApplyMomentum<T> applyMomentum(Operand<T> var,
                                          Operand<T> accum,
                                          Operand<T> lr,
                                          Operand<T> grad,
                                          Operand<T> momentum,
                                          ApplyMomentum.Options... options)

Adds an ApplyMomentum operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

grad - The gradient.

momentum - Momentum. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of ApplyMomentum

See Also:

ApplyMomentum

tensorArrayScatter

public <T> TensorArrayScatter tensorArrayScatter(Operand<?> handle,
                                                 Operand<Integer> indices,
                                                 Operand<T> value,
                                                 Operand<Float> flowIn)

Adds an TensorArrayScatter operation to the graph

Parameters:

handle - The handle to a TensorArray.

indices - The locations at which to write the tensor elements.

value - The concatenated tensor to write to the TensorArray.

flowIn - A float scalar that enforces proper chaining of operations.

Returns:

a new instance of TensorArrayScatter

See Also:

TensorArrayScatter

batchIFFT3D

public BatchIFFT3D batchIFFT3D(Operand<?> input)

Adds an BatchIFFT3D operation to the graph

Parameters:

input -

Returns:

a new instance of BatchIFFT3D

See Also:

BatchIFFT3D

sparseSegmentSqrtNGrad

public <T extends Number,U extends Number> SparseSegmentSqrtNGrad<T> sparseSegmentSqrtNGrad(Operand<T> grad,
                                                                                            Operand<U> indices,
                                                                                            Operand<Integer> segmentIds,
                                                                                            Operand<Integer> outputDim0)

Adds an SparseSegmentSqrtNGrad operation to the graph

Parameters:

grad - gradient propagated to the SparseSegmentSqrtN op.

indices - indices passed to the corresponding SparseSegmentSqrtN op.

segmentIds - segment_ids passed to the corresponding SparseSegmentSqrtN op.

outputDim0 - dimension 0 of "data" passed to SparseSegmentSqrtN op.

Returns:

a new instance of SparseSegmentSqrtNGrad

See Also:

SparseSegmentSqrtNGrad

quantizedMul

public <V,T,U> QuantizedMul<V> quantizedMul(Operand<T> x,
                                            Operand<U> y,
                                            Operand<Float> minX,
                                            Operand<Float> maxX,
                                            Operand<Float> minY,
                                            Operand<Float> maxY,
                                            Class<V> Toutput)

Adds an QuantizedMul operation to the graph

Parameters:

x -

y -

minX - The float value that the lowest quantized `x` value represents.

maxX - The float value that the highest quantized `x` value represents.

minY - The float value that the lowest quantized `y` value represents.

maxY - The float value that the highest quantized `y` value represents.

Toutput -

Returns:

a new instance of QuantizedMul

See Also:

QuantizedMul

iFFT2D

public <T> IFFT2D<T> iFFT2D(Operand<T> input)

Adds an IFFT2D operation to the graph

Parameters:

input - A complex tensor.

Returns:

a new instance of IFFT2D

See Also:

IFFT2D

inv

public <T> Inv<T> inv(Operand<T> x)

Adds an Inv operation to the graph

Parameters:

x -

Returns:

a new instance of Inv

See Also:

Inv

extractGlimpse

public ExtractGlimpse extractGlimpse(Operand<Float> input,
                                     Operand<Integer> size,
                                     Operand<Float> offsets,
                                     ExtractGlimpse.Options... options)

Adds an ExtractGlimpse operation to the graph

Parameters:

input - A 4-D float tensor of shape `[batch_size, height, width, channels]`.

size - A 1-D tensor of 2 elements containing the size of the glimpses

offsets - A 2-D integer tensor of shape `[batch_size, 2]` containing

options - carries optional attributes values

Returns:

a new instance of ExtractGlimpse

See Also:

ExtractGlimpse

cudnnRNNParamsToCanonical

public <T extends Number> CudnnRNNParamsToCanonical<T> cudnnRNNParamsToCanonical(Operand<Integer> numLayers,
                                                                                 Operand<Integer> numUnits,
                                                                                 Operand<Integer> inputSize,
                                                                                 Operand<T> params,
                                                                                 Long numParams,
                                                                                 CudnnRNNParamsToCanonical.Options... options)

Adds an CudnnRNNParamsToCanonical operation to the graph

Parameters:

numLayers -

numUnits -

inputSize -

params -

numParams -

options - carries optional attributes values

Returns:

a new instance of CudnnRNNParamsToCanonical

See Also:

CudnnRNNParamsToCanonical

inTopKV2

public <T extends Number> InTopKV2 inTopKV2(Operand<Float> predictions,
                                            Operand<T> targets,
                                            Operand<T> k)

Adds an InTopKV2 operation to the graph

Parameters:

predictions - A `batch_size` x `classes` tensor.

targets - A `batch_size` vector of class ids.

k - Number of top elements to look at for computing precision.

Returns:

a new instance of InTopKV2

See Also:

InTopKV2

bitcast

public <U,T> Bitcast<U> bitcast(Operand<T> input,
                                Class<U> type)

Adds an Bitcast operation to the graph

Parameters:

input -

type -

Returns:

a new instance of Bitcast

See Also:

Bitcast

assignSubVariableOp

public <T> AssignSubVariableOp assignSubVariableOp(Operand<?> resource,
                                                   Operand<T> value)

Adds an AssignSubVariableOp operation to the graph

Parameters:

resource - handle to the resource in which to store the variable.

value - the value by which the variable will be incremented.

Returns:

a new instance of AssignSubVariableOp

See Also:

AssignSubVariableOp

roll

public <T,U extends Number,V extends Number> Roll<T> roll(Operand<T> input,
                                                          Operand<U> shift,
                                                          Operand<V> axis)

Adds an Roll operation to the graph

Parameters:

input -

shift - Dimension must be 0-D or 1-D. `shift[i]` specifies the number of places by which

axis - Dimension must be 0-D or 1-D. `axis[i]` specifies the dimension that the shift

Returns:

a new instance of Roll

See Also:

Roll

histogramSummary

public <T extends Number> HistogramSummary histogramSummary(Operand<String> tag,
                                                            Operand<T> values)

Adds an HistogramSummary operation to the graph

Parameters:

tag - Scalar. Tag to use for the `Summary.Value`.

values - Any shape. Values to use to build the histogram.

Returns:

a new instance of HistogramSummary

See Also:

HistogramSummary

fusedBatchNormGrad

public <T extends Number> FusedBatchNormGrad<T> fusedBatchNormGrad(Operand<T> yBackprop,
                                                                   Operand<T> x,
                                                                   Operand<T> scale,
                                                                   Operand<T> reserveSpace1,
                                                                   Operand<T> reserveSpace2,
                                                                   FusedBatchNormGrad.Options... options)

Adds an FusedBatchNormGrad operation to the graph

Parameters:

yBackprop - A 4D Tensor for the gradient with respect to y.

x - A 4D Tensor for input data.

scale - A 1D Tensor for scaling factor, to scale the normalized x.

reserveSpace1 - When is_training is True, a 1D Tensor for the computed batch

reserveSpace2 - When is_training is True, a 1D Tensor for the computed batch

options - carries optional attributes values

Returns:

a new instance of FusedBatchNormGrad

See Also:

FusedBatchNormGrad

encodePng

public <T extends Number> EncodePng encodePng(Operand<T> image,
                                              EncodePng.Options... options)

Adds an EncodePng operation to the graph

Parameters:

image - 3-D with shape `[height, width, channels]`.

options - carries optional attributes values

Returns:

a new instance of EncodePng

See Also:

EncodePng

sparseSoftmaxCrossEntropyWithLogits

public <T extends Number,U extends Number> SparseSoftmaxCrossEntropyWithLogits<T> sparseSoftmaxCrossEntropyWithLogits(Operand<T> features,
                                                                                                                      Operand<U> labels)

Adds an SparseSoftmaxCrossEntropyWithLogits operation to the graph

Parameters:

features - batch_size x num_classes matrix

labels - batch_size vector with values in [0, num_classes).

Returns:

a new instance of SparseSoftmaxCrossEntropyWithLogits

See Also:

SparseSoftmaxCrossEntropyWithLogits

deserializeManySparse

public <T> DeserializeManySparse<T> deserializeManySparse(Operand<String> serializedSparse,
                                                          Class<T> dtype)

Adds an DeserializeManySparse operation to the graph

Parameters:

serializedSparse - 2-D, The `N` serialized `SparseTensor` objects.

dtype - The `dtype` of the serialized `SparseTensor` objects.

Returns:

a new instance of DeserializeManySparse

See Also:

DeserializeManySparse

matrixSolveLs

public <T> MatrixSolveLs<T> matrixSolveLs(Operand<T> matrix,
                                          Operand<T> rhs,
                                          Operand<Double> l2Regularizer,
                                          MatrixSolveLs.Options... options)

Adds an MatrixSolveLs operation to the graph

Parameters:

matrix - Shape is `[..., M, N]`.

rhs - Shape is `[..., M, K]`.

l2Regularizer - Scalar tensor.

options - carries optional attributes values

Returns:

a new instance of MatrixSolveLs

See Also:

MatrixSolveLs

eagerPyFunc

public EagerPyFunc eagerPyFunc(Iterable<Operand<?>> input,
                               String token,
                               List<Class<?>> Tout)

Adds an EagerPyFunc operation to the graph

Parameters:

input -

token -

Tout -

Returns:

a new instance of EagerPyFunc

See Also:

EagerPyFunc

ensureShape

public <T> EnsureShape<T> ensureShape(Operand<T> input,
                                      Shape shape)

Adds an EnsureShape operation to the graph

Parameters:

input - A tensor, whose shape is to be validated.

shape - The expected (possibly partially specified) shape of the input tensor.

Returns:

a new instance of EnsureShape

See Also:

EnsureShape

subtract

public <T> Subtract<T> subtract(Operand<T> x,
                                Operand<T> y)

Adds an Subtract operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Subtract

See Also:

Subtract

randomShuffleQueue

public RandomShuffleQueue randomShuffleQueue(List<Class<?>> componentTypes,
                                             RandomShuffleQueue.Options... options)

Adds an RandomShuffleQueue operation to the graph

Parameters:

componentTypes - The type of each component in a value.

options - carries optional attributes values

Returns:

a new instance of RandomShuffleQueue

See Also:

RandomShuffleQueue

stringSplitV2

public StringSplitV2 stringSplitV2(Operand<String> input,
                                   Operand<String> sep,
                                   StringSplitV2.Options... options)

Adds an StringSplitV2 operation to the graph

Parameters:

input - `1-D` string `Tensor`, the strings to split.

sep - `0-D` string `Tensor`, the delimiter character.

options - carries optional attributes values

Returns:

a new instance of StringSplitV2

See Also:

StringSplitV2

sparseReduceSum

public <T> SparseReduceSum<T> sparseReduceSum(Operand<Long> inputIndices,
                                              Operand<T> inputValues,
                                              Operand<Long> inputShape,
                                              Operand<Integer> reductionAxes,
                                              SparseReduceSum.Options... options)

Adds an SparseReduceSum operation to the graph

Parameters:

inputIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

inputValues - 1-D. `N` non-empty values corresponding to `input_indices`.

inputShape - 1-D. Shape of the input SparseTensor.

reductionAxes - 1-D. Length-`K` vector containing the reduction axes.

options - carries optional attributes values

Returns:

a new instance of SparseReduceSum

See Also:

SparseReduceSum

equal

public <T> Equal equal(Operand<T> x,
                       Operand<T> y)

Adds an Equal operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Equal

See Also:

Equal

remoteFusedGraphExecute

public RemoteFusedGraphExecute remoteFusedGraphExecute(Iterable<Operand<?>> inputs,
                                                       List<Class<?>> Toutputs,
                                                       String serializedRemoteFusedGraphExecuteInfo)

Adds an RemoteFusedGraphExecute operation to the graph

Parameters:

inputs - Arbitrary number of tensors with arbitrary data types

Toutputs -

serializedRemoteFusedGraphExecuteInfo - Serialized protocol buffer

Returns:

a new instance of RemoteFusedGraphExecute

See Also:

RemoteFusedGraphExecute

mirrorPad

public <T,U extends Number> MirrorPad<T> mirrorPad(Operand<T> input,
                                                   Operand<U> paddings,
                                                   String mode)

Adds an MirrorPad operation to the graph

Parameters:

input - The input tensor to be padded.

paddings - A two-column matrix specifying the padding sizes. The number of

mode - Either `REFLECT` or `SYMMETRIC`. In reflect mode the padded regions

Returns:

a new instance of MirrorPad

See Also:

MirrorPad

resourceApplyAddSign

public <T> ResourceApplyAddSign resourceApplyAddSign(Operand<?> var,
                                                     Operand<?> m,
                                                     Operand<T> lr,
                                                     Operand<T> alpha,
                                                     Operand<T> signDecay,
                                                     Operand<T> beta,
                                                     Operand<T> grad,
                                                     ResourceApplyAddSign.Options... options)

Adds an ResourceApplyAddSign operation to the graph

Parameters:

var - Should be from a Variable().

m - Should be from a Variable().

lr - Scaling factor. Must be a scalar.

alpha - Must be a scalar.

signDecay - Must be a scalar.

beta - Must be a scalar.

grad - The gradient.

options - carries optional attributes values

Returns:

a new instance of ResourceApplyAddSign

See Also:

ResourceApplyAddSign

complex

public <U,T extends Number> Complex<U> complex(Operand<T> real,
                                               Operand<T> imag,
                                               Class<U> Tout)

Adds an Complex operation to the graph

Parameters:

real -

imag -

Tout -

Returns:

a new instance of Complex

See Also:

Complex

fakeQuantWithMinMaxVarsPerChannel

public FakeQuantWithMinMaxVarsPerChannel fakeQuantWithMinMaxVarsPerChannel(Operand<Float> inputs,
                                                                           Operand<Float> min,
                                                                           Operand<Float> max,
                                                                           FakeQuantWithMinMaxVarsPerChannel.Options... options)

Adds an FakeQuantWithMinMaxVarsPerChannel operation to the graph

Parameters:

inputs -

min -

max -

options - carries optional attributes values

Returns:

a new instance of FakeQuantWithMinMaxVarsPerChannel

See Also:

FakeQuantWithMinMaxVarsPerChannel

constant

public Constant<Long> constant(long[] shape,
                               LongBuffer data)

Adds an Constant operation to the graph

Parameters:

shape - the tensor shape.

data - a buffer containing the tensor data.

Returns:

a long constant

Throws:

IllegalArgumentException - If the tensor shape is not compatible with the buffer

See Also:

Constant

oneHot

public <U,T extends Number> OneHot<U> oneHot(Operand<T> indices,
                                             Operand<Integer> depth,
                                             Operand<U> onValue,
                                             Operand<U> offValue,
                                             OneHot.Options... options)

Adds an OneHot operation to the graph

Parameters:

indices - A tensor of indices.

depth - A scalar defining the depth of the one hot dimension.

onValue - A scalar defining the value to fill in output when `indices[j] = i`.

offValue - A scalar defining the value to fill in output when `indices[j] != i`.

options - carries optional attributes values

Returns:

a new instance of OneHot

See Also:

OneHot

sparseApplyFtrlV2

public <T,U extends Number> SparseApplyFtrlV2<T> sparseApplyFtrlV2(Operand<T> var,
                                                                   Operand<T> accum,
                                                                   Operand<T> linear,
                                                                   Operand<T> grad,
                                                                   Operand<U> indices,
                                                                   Operand<T> lr,
                                                                   Operand<T> l1,
                                                                   Operand<T> l2,
                                                                   Operand<T> l2Shrinkage,
                                                                   Operand<T> lrPower,
                                                                   SparseApplyFtrlV2.Options... options)

Adds an SparseApplyFtrlV2 operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

linear - Should be from a Variable().

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

lr - Scaling factor. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 shrinkage regulariation. Must be a scalar.

l2Shrinkage -

lrPower - Scaling factor. Must be a scalar.

options - carries optional attributes values

Returns:

a new instance of SparseApplyFtrlV2

See Also:

SparseApplyFtrlV2

decodeJpeg

public DecodeJpeg decodeJpeg(Operand<String> contents,
                             DecodeJpeg.Options... options)

Adds an DecodeJpeg operation to the graph

Parameters:

contents - 0-D. The JPEG-encoded image.

options - carries optional attributes values

Returns:

a new instance of DecodeJpeg

See Also:

DecodeJpeg

constant

public Constant<Float> constant(float[][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

biasAddGrad

public <T> BiasAddGrad<T> biasAddGrad(Operand<T> outBackprop,
                                      BiasAddGrad.Options... options)

Adds an BiasAddGrad operation to the graph

Parameters:

outBackprop - Any number of dimensions.

options - carries optional attributes values

Returns:

a new instance of BiasAddGrad

See Also:

BiasAddGrad

unicodeScript

public UnicodeScript unicodeScript(Operand<Integer> input)

Adds an UnicodeScript operation to the graph

Parameters:

input - A Tensor of int32 Unicode code points.

Returns:

a new instance of UnicodeScript

See Also:

UnicodeScript

constant

public Constant<Long> constant(long data)

Adds an Constant operation to the graph

Parameters:

data - The value to put into the new constant.

Returns:

a long constant

See Also:

Constant

optionalHasValue

public OptionalHasValue optionalHasValue(Operand<?> optional)

Adds an OptionalHasValue operation to the graph

Parameters:

optional -

Returns:

a new instance of OptionalHasValue

See Also:

OptionalHasValue

resourceScatterUpdate

public <T extends Number,U> ResourceScatterUpdate resourceScatterUpdate(Operand<?> resource,
                                                                        Operand<T> indices,
                                                                        Operand<U> updates)

Adds an ResourceScatterUpdate operation to the graph

Parameters:

resource - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to add to `ref`.

Returns:

a new instance of ResourceScatterUpdate

See Also:

ResourceScatterUpdate

imag

public <T> Imag<Float> imag(Operand<T> input)

Adds an Imag operation to the graph

Parameters:

input -

Returns:

a new instance of Imag

See Also:

Imag

diagPart

public <T> DiagPart<T> diagPart(Operand<T> input)

Adds an DiagPart operation to the graph

Parameters:

input - Rank k tensor where k is even and not zero.

Returns:

a new instance of DiagPart

See Also:

DiagPart

lookupTableInsert

public <T,U> LookupTableInsert lookupTableInsert(Operand<?> tableHandle,
                                                 Operand<T> keys,
                                                 Operand<U> values)

Adds an LookupTableInsert operation to the graph

Parameters:

tableHandle - Handle to the table.

keys - Any shape. Keys to look up.

values - Values to associate with keys.

Returns:

a new instance of LookupTableInsert

See Also:

LookupTableInsert

uniqueV2

public <T,V extends Number,U extends Number> UniqueV2<T,V> uniqueV2(Operand<T> x,
                                                                    Operand<U> axis,
                                                                    Class<V> outIdx)

Adds an UniqueV2 operation to the graph

Parameters:

x - A `Tensor`.

axis - A `Tensor` of type `int32` (default: None). The axis of the Tensor to

outIdx -

Returns:

a new instance of UniqueV2

See Also:

UniqueV2

quantizeV2

public <T> QuantizeV2<T> quantizeV2(Operand<Float> input,
                                    Operand<Float> minRange,
                                    Operand<Float> maxRange,
                                    Class<T> T,
                                    QuantizeV2.Options... options)

Adds an QuantizeV2 operation to the graph

Parameters:

input -

minRange - The minimum scalar value possibly produced for the input.

maxRange - The maximum scalar value possibly produced for the input.

T -

options - carries optional attributes values

Returns:

a new instance of QuantizeV2

See Also:

QuantizeV2

learnedUnigramCandidateSampler

public LearnedUnigramCandidateSampler learnedUnigramCandidateSampler(Operand<Long> trueClasses,
                                                                     Long numTrue,
                                                                     Long numSampled,
                                                                     Boolean unique,
                                                                     Long rangeMax,
                                                                     LearnedUnigramCandidateSampler.Options... options)

Adds an LearnedUnigramCandidateSampler operation to the graph

Parameters:

trueClasses - A batch_size * num_true matrix, in which each row contains the

numTrue - Number of true labels per context.

numSampled - Number of candidates to randomly sample.

unique - If unique is true, we sample with rejection, so that all sampled

rangeMax - The sampler will sample integers from the interval [0, range_max).

options - carries optional attributes values

Returns:

a new instance of LearnedUnigramCandidateSampler

See Also:

LearnedUnigramCandidateSampler

truncateDiv

public <T> TruncateDiv<T> truncateDiv(Operand<T> x,
                                      Operand<T> y)

Adds an TruncateDiv operation to the graph

Parameters:

x -

y -

Returns:

a new instance of TruncateDiv

See Also:

TruncateDiv

statelessRandomNormal

public <V extends Number,T extends Number,U extends Number> StatelessRandomNormal<V> statelessRandomNormal(Operand<T> shape,
                                                                                                           Operand<U> seed,
                                                                                                           Class<V> dtype)

Adds an StatelessRandomNormal operation to the graph

Parameters:

shape - The shape of the output tensor.

seed - 2 seeds (shape [2]).

dtype - The type of the output.

Returns:

a new instance of StatelessRandomNormal

See Also:

StatelessRandomNormal

constant

public Constant<Double> constant(double[][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

sparseApplyAdadelta

public <T,U extends Number> SparseApplyAdadelta<T> sparseApplyAdadelta(Operand<T> var,
                                                                       Operand<T> accum,
                                                                       Operand<T> accumUpdate,
                                                                       Operand<T> lr,
                                                                       Operand<T> rho,
                                                                       Operand<T> epsilon,
                                                                       Operand<T> grad,
                                                                       Operand<U> indices,
                                                                       SparseApplyAdadelta.Options... options)

Adds an SparseApplyAdadelta operation to the graph

Parameters:

var -

accum - Should be from a Variable().

accumUpdate - : Should be from a Variable().

lr - Learning rate. Must be a scalar.

rho - Decay factor. Must be a scalar.

epsilon - Constant factor. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

options - carries optional attributes values

Returns:

a new instance of SparseApplyAdadelta

See Also:

SparseApplyAdadelta

asString

public <T> AsString asString(Operand<T> input,
                             AsString.Options... options)

Adds an AsString operation to the graph

Parameters:

input -

options - carries optional attributes values

Returns:

a new instance of AsString

See Also:

AsString

addV2

public <T> AddV2<T> addV2(Operand<T> x,
                          Operand<T> y)

Adds an AddV2 operation to the graph

Parameters:

x -

y -

Returns:

a new instance of AddV2

See Also:

AddV2

fIFOQueue

public FIFOQueue fIFOQueue(List<Class<?>> componentTypes,
                           FIFOQueue.Options... options)

Adds an FIFOQueue operation to the graph

Parameters:

componentTypes - The type of each component in a value.

options - carries optional attributes values

Returns:

a new instance of FIFOQueue

See Also:

FIFOQueue

besselI1e

public <T extends Number> BesselI1e<T> besselI1e(Operand<T> x)

Adds an BesselI1e operation to the graph

Parameters:

x -

Returns:

a new instance of BesselI1e

See Also:

BesselI1e

bitwiseAnd

public <T extends Number> BitwiseAnd<T> bitwiseAnd(Operand<T> x,
                                                   Operand<T> y)

Adds an BitwiseAnd operation to the graph

Parameters:

x -

y -

Returns:

a new instance of BitwiseAnd

See Also:

BitwiseAnd

sub

public <T> Sub<T> sub(Operand<T> x,
                      Operand<T> y)

Adds an Sub operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Sub

See Also:

Sub

batchNormWithGlobalNormalization

public <T> BatchNormWithGlobalNormalization<T> batchNormWithGlobalNormalization(Operand<T> t,
                                                                                Operand<T> m,
                                                                                Operand<T> v,
                                                                                Operand<T> beta,
                                                                                Operand<T> gamma,
                                                                                Float varianceEpsilon,
                                                                                Boolean scaleAfterNormalization)

Adds an BatchNormWithGlobalNormalization operation to the graph

Parameters:

t - A 4D input Tensor.

m - A 1D mean Tensor with size matching the last dimension of t.

v - A 1D variance Tensor with size matching the last dimension of t.

beta - A 1D beta Tensor with size matching the last dimension of t.

gamma - A 1D gamma Tensor with size matching the last dimension of t.

varianceEpsilon - A small float number to avoid dividing by 0.

scaleAfterNormalization - A bool indicating whether the resulted tensor

Returns:

a new instance of BatchNormWithGlobalNormalization

See Also:

BatchNormWithGlobalNormalization

mapClear

public MapClear mapClear(List<Class<?>> dtypes,
                         MapClear.Options... options)

Adds an MapClear operation to the graph

Parameters:

dtypes -

options - carries optional attributes values

Returns:

a new instance of MapClear

See Also:

MapClear

serializeTensor

public <T> SerializeTensor serializeTensor(Operand<T> tensor)

Adds an SerializeTensor operation to the graph

Parameters:

tensor - A Tensor of type `T`.

Returns:

a new instance of SerializeTensor

See Also:

SerializeTensor

size

public <T> Size<Integer> size(Operand<T> input)

Adds an Size operation to the graph

Parameters:

input -

Returns:

a new instance of Size

See Also:

Size

serializeManySparse

public <T> SerializeManySparse<String> serializeManySparse(Operand<Long> sparseIndices,
                                                           Operand<T> sparseValues,
                                                           Operand<Long> sparseShape)

Adds an SerializeManySparse operation to the graph

Parameters:

sparseIndices - 2-D. The `indices` of the minibatch `SparseTensor`.

sparseValues - 1-D. The `values` of the minibatch `SparseTensor`.

sparseShape - 1-D. The `shape` of the minibatch `SparseTensor`.

Returns:

a new instance of SerializeManySparse

See Also:

SerializeManySparse

constant

public Constant<Float> constant(float[][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

stagePeek

public StagePeek stagePeek(Operand<Integer> index,
                           List<Class<?>> dtypes,
                           StagePeek.Options... options)

Adds an StagePeek operation to the graph

Parameters:

index -

dtypes -

options - carries optional attributes values

Returns:

a new instance of StagePeek

See Also:

StagePeek

iteratorGetNextSync

public IteratorGetNextSync iteratorGetNextSync(Operand<?> iterator,
                                               List<Class<?>> outputTypes,
                                               List<Shape> outputShapes)

Adds an IteratorGetNextSync operation to the graph

Parameters:

iterator -

outputTypes -

outputShapes -

Returns:

a new instance of IteratorGetNextSync

See Also:

IteratorGetNextSync

sign

public <T> Sign<T> sign(Operand<T> x)

Adds an Sign operation to the graph

Parameters:

x -

Returns:

a new instance of Sign

See Also:

Sign

zerosLike

public <T> ZerosLike<T> zerosLike(Operand<T> x)

Adds an ZerosLike operation to the graph

Parameters:

x - a tensor of type T.

Returns:

a new instance of ZerosLike

See Also:

ZerosLike

tensorArrayPack

public <T> TensorArrayPack<T> tensorArrayPack(Operand<String> handle,
                                              Operand<Float> flowIn,
                                              Class<T> dtype,
                                              TensorArrayPack.Options... options)

Adds an TensorArrayPack operation to the graph

Parameters:

handle -

flowIn -

dtype -

options - carries optional attributes values

Returns:

a new instance of TensorArrayPack

See Also:

TensorArrayPack

reduceMean

public <T,U extends Number> ReduceMean<T> reduceMean(Operand<T> input,
                                                     Operand<U> axis,
                                                     ReduceMean.Options... options)

Adds an ReduceMean operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of ReduceMean

See Also:

ReduceMean

batchFFT

public BatchFFT batchFFT(Operand<?> input)

Adds an BatchFFT operation to the graph

Parameters:

input -

Returns:

a new instance of BatchFFT

See Also:

BatchFFT

batchFFT3D

public BatchFFT3D batchFFT3D(Operand<?> input)

Adds an BatchFFT3D operation to the graph

Parameters:

input -

Returns:

a new instance of BatchFFT3D

See Also:

BatchFFT3D

log1p

public <T> Log1p<T> log1p(Operand<T> x)

Adds an Log1p operation to the graph

Parameters:

x -

Returns:

a new instance of Log1p

See Also:

Log1p

argMax

public <T,U extends Number> ArgMax<Long> argMax(Operand<T> input,
                                                Operand<U> dimension)

Adds an ArgMax operation to the graph

Parameters:

input -

dimension - int32 or int64, must be in the range `[-rank(input), rank(input))`.

Returns:

a new instance of ArgMax

See Also:

ArgMax

parseExample

public ParseExample parseExample(Operand<String> serialized,
                                 Operand<String> names,
                                 Iterable<Operand<String>> sparseKeys,
                                 Iterable<Operand<String>> denseKeys,
                                 Iterable<Operand<?>> denseDefaults,
                                 List<Class<?>> sparseTypes,
                                 List<Shape> denseShapes)

Adds an ParseExample operation to the graph

Parameters:

serialized - A vector containing a batch of binary serialized Example protos.

names - A vector containing the names of the serialized protos.

sparseKeys - A list of Nsparse string Tensors (scalars).

denseKeys - A list of Ndense string Tensors (scalars).

denseDefaults - A list of Ndense Tensors (some may be empty).

sparseTypes - A list of Nsparse types; the data types of data in each Feature

denseShapes - A list of Ndense shapes; the shapes of data in each Feature

Returns:

a new instance of ParseExample

See Also:

ParseExample

lu

public <T> Lu<T,Integer> lu(Operand<T> input)

Adds an Lu operation to the graph

Parameters:

input - A tensor of shape `[..., M, M]` whose inner-most 2 dimensions form matrices of

Returns:

a new instance of Lu

See Also:

Lu

fakeQuantWithMinMaxArgs

public FakeQuantWithMinMaxArgs fakeQuantWithMinMaxArgs(Operand<Float> inputs,
                                                       FakeQuantWithMinMaxArgs.Options... options)

Adds an FakeQuantWithMinMaxArgs operation to the graph

Parameters:

inputs -

options - carries optional attributes values

Returns:

a new instance of FakeQuantWithMinMaxArgs

See Also:

FakeQuantWithMinMaxArgs

pow

public <T> Pow<T> pow(Operand<T> x,
                      Operand<T> y)

Adds an Pow operation to the graph

Parameters:

x -

y -

Returns:

a new instance of Pow

See Also:

Pow

statelessMultinomial

public <T extends Number,U extends Number> StatelessMultinomial<Long> statelessMultinomial(Operand<T> logits,
                                                                                           Operand<Integer> numSamples,
                                                                                           Operand<U> seed)

Adds an StatelessMultinomial operation to the graph

Parameters:

logits - 2-D Tensor with shape `[batch_size, num_classes]`. Each slice `[i, :]`

numSamples - 0-D. Number of independent samples to draw for each row slice.

seed - 2 seeds (shape [2]).

Returns:

a new instance of StatelessMultinomial

See Also:

StatelessMultinomial

batchSvd

public <T> BatchSvd<T> batchSvd(Operand<T> input,
                                BatchSvd.Options... options)

Adds an BatchSvd operation to the graph

Parameters:

input -

options - carries optional attributes values

Returns:

a new instance of BatchSvd

See Also:

BatchSvd

queueClose

public QueueClose queueClose(Operand<?> handle,
                             QueueClose.Options... options)

Adds an QueueClose operation to the graph

Parameters:

handle - The handle to a queue.

options - carries optional attributes values

Returns:

a new instance of QueueClose

See Also:

QueueClose

atan

public <T> Atan<T> atan(Operand<T> x)

Adds an Atan operation to the graph

Parameters:

x -

Returns:

a new instance of Atan

See Also:

Atan

reduceMax

public <T,U extends Number> ReduceMax<T> reduceMax(Operand<T> input,
                                                   Operand<U> axis,
                                                   ReduceMax.Options... options)

Adds an ReduceMax operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of ReduceMax

See Also:

ReduceMax

tensorListLength

public TensorListLength tensorListLength(Operand<?> inputHandle)

Adds an TensorListLength operation to the graph

Parameters:

inputHandle -

Returns:

a new instance of TensorListLength

See Also:

TensorListLength

constant

public Constant<Integer> constant(int[] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

orderedMapPeek

public OrderedMapPeek orderedMapPeek(Operand<Long> key,
                                     Operand<Integer> indices,
                                     List<Class<?>> dtypes,
                                     OrderedMapPeek.Options... options)

Adds an OrderedMapPeek operation to the graph

Parameters:

key -

indices -

dtypes -

options - carries optional attributes values

Returns:

a new instance of OrderedMapPeek

See Also:

OrderedMapPeek

initializeTableFromTextFile

public InitializeTableFromTextFile initializeTableFromTextFile(Operand<?> tableHandle,
                                                               Operand<String> filename,
                                                               Long keyIndex,
                                                               Long valueIndex,
                                                               InitializeTableFromTextFile.Options... options)

Adds an InitializeTableFromTextFile operation to the graph

Parameters:

tableHandle - Handle to a table which will be initialized.

filename - Filename of a vocabulary text file.

keyIndex - Column index in a line to get the table `key` values from.

valueIndex - Column index that represents information of a line to get the table

options - carries optional attributes values

Returns:

a new instance of InitializeTableFromTextFile

See Also:

InitializeTableFromTextFile

expm1

public <T> Expm1<T> expm1(Operand<T> x)

Adds an Expm1 operation to the graph

Parameters:

x -

Returns:

a new instance of Expm1

See Also:

Expm1

l2Loss

public <T extends Number> L2Loss<T> l2Loss(Operand<T> t)

Adds an L2Loss operation to the graph

Parameters:

t - Typically 2-D, but may have any dimensions.

Returns:

a new instance of L2Loss

See Also:

L2Loss

any

public <T extends Number> Any any(Operand<Boolean> input,
                                  Operand<T> axis,
                                  Any.Options... options)

Adds an Any operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of Any

See Also:

Any

resourceScatterNdUpdate

public <T extends Number,U> ResourceScatterNdUpdate resourceScatterNdUpdate(Operand<?> ref,
                                                                            Operand<T> indices,
                                                                            Operand<U> updates,
                                                                            ResourceScatterNdUpdate.Options... options)

Adds an ResourceScatterNdUpdate operation to the graph

Parameters:

ref - A resource handle. Must be from a VarHandleOp.

indices - A Tensor. Must be one of the following types: int32, int64.

updates - A Tensor. Must have the same type as ref. A tensor of updated

options - carries optional attributes values

Returns:

a new instance of ResourceScatterNdUpdate

See Also:

ResourceScatterNdUpdate

uniqueV2

public <T,U extends Number> UniqueV2<T,Integer> uniqueV2(Operand<T> x,
                                                         Operand<U> axis)

Adds an UniqueV2 operation to the graph

Parameters:

x - A `Tensor`.

axis - A `Tensor` of type `int32` (default: None). The axis of the Tensor to

Returns:

a new instance of UniqueV2

See Also:

UniqueV2

randomNormal

public <U extends Number,T extends Number> RandomNormal<U> randomNormal(Operand<T> shape,
                                                                        Class<U> dtype,
                                                                        RandomNormal.Options... options)

Adds an RandomNormal operation to the graph

Parameters:

shape - The shape of the output tensor.

dtype - The type of the output.

options - carries optional attributes values

Returns:

a new instance of RandomNormal

See Also:

RandomNormal

clipByValue

public <T> ClipByValue<T> clipByValue(Operand<T> t,
                                      Operand<T> clipValueMin,
                                      Operand<T> clipValueMax)

Adds an ClipByValue operation to the graph

Parameters:

t - A `Tensor`.

clipValueMin - A 0-D (scalar) `Tensor`, or a `Tensor` with the same shape

clipValueMax - A 0-D (scalar) `Tensor`, or a `Tensor` with the same shape

Returns:

a new instance of ClipByValue

See Also:

ClipByValue

svd

public <T> Svd<T> svd(Operand<T> input,
                      Svd.Options... options)

Adds an Svd operation to the graph

Parameters:

input - A tensor of shape `[..., M, N]` whose inner-most 2 dimensions

options - carries optional attributes values

Returns:

a new instance of Svd

See Also:

Svd

sparseReduceSumSparse

public <T> SparseReduceSumSparse<T> sparseReduceSumSparse(Operand<Long> inputIndices,
                                                          Operand<T> inputValues,
                                                          Operand<Long> inputShape,
                                                          Operand<Integer> reductionAxes,
                                                          SparseReduceSumSparse.Options... options)

Adds an SparseReduceSumSparse operation to the graph

Parameters:

inputIndices - 2-D. `N x R` matrix with the indices of non-empty values in a

inputValues - 1-D. `N` non-empty values corresponding to `input_indices`.

inputShape - 1-D. Shape of the input SparseTensor.

reductionAxes - 1-D. Length-`K` vector containing the reduction axes.

options - carries optional attributes values

Returns:

a new instance of SparseReduceSumSparse

See Also:

SparseReduceSumSparse

asinh

public <T> Asinh<T> asinh(Operand<T> x)

Adds an Asinh operation to the graph

Parameters:

x -

Returns:

a new instance of Asinh

See Also:

Asinh

rGBToHSV

public <T extends Number> RGBToHSV<T> rGBToHSV(Operand<T> images)

Adds an RGBToHSV operation to the graph

Parameters:

images - 1-D or higher rank. RGB data to convert. Last dimension must be size 3.

Returns:

a new instance of RGBToHSV

See Also:

RGBToHSV

elu

public <T extends Number> Elu<T> elu(Operand<T> features)

Adds an Elu operation to the graph

Parameters:

features -

Returns:

a new instance of Elu

See Also:

Elu

batch

public Batch batch(Iterable<Operand<?>> inTensors,
                   Long numBatchThreads,
                   Long maxBatchSize,
                   Long batchTimeoutMicros,
                   Long gradTimeoutMicros,
                   Batch.Options... options)

Adds an Batch operation to the graph

Parameters:

inTensors -

numBatchThreads -

maxBatchSize -

batchTimeoutMicros -

gradTimeoutMicros -

options - carries optional attributes values

Returns:

a new instance of Batch

See Also:

Batch

recordInput

public RecordInput recordInput(String filePattern,
                               RecordInput.Options... options)

Adds an RecordInput operation to the graph

Parameters:

filePattern - Glob pattern for the data files.

options - carries optional attributes values

Returns:

a new instance of RecordInput

See Also:

RecordInput

dilation2DBackpropFilter

public <T extends Number> Dilation2DBackpropFilter<T> dilation2DBackpropFilter(Operand<T> input,
                                                                               Operand<T> filter,
                                                                               Operand<T> outBackprop,
                                                                               List<Long> strides,
                                                                               List<Long> rates,
                                                                               String padding)

Adds an Dilation2DBackpropFilter operation to the graph

Parameters:

input - 4-D with shape `[batch, in_height, in_width, depth]`.

filter - 3-D with shape `[filter_height, filter_width, depth]`.

outBackprop - 4-D with shape `[batch, out_height, out_width, depth]`.

strides - 1-D of length 4. The stride of the sliding window for each dimension of

rates - 1-D of length 4. The input stride for atrous morphological dilation.

padding - The type of padding algorithm to use.

Returns:

a new instance of Dilation2DBackpropFilter

See Also:

Dilation2DBackpropFilter

avgPool3DGrad

public <T extends Number> AvgPool3DGrad<T> avgPool3DGrad(Operand<Integer> origInputShape,
                                                         Operand<T> grad,
                                                         List<Long> ksize,
                                                         List<Long> strides,
                                                         String padding,
                                                         AvgPool3DGrad.Options... options)

Adds an AvgPool3DGrad operation to the graph

Parameters:

origInputShape - The original input dimensions.

grad - Output backprop of shape `[batch, depth, rows, cols, channels]`.

ksize - 1-D tensor of length 5. The size of the window for each dimension of

strides - 1-D tensor of length 5. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of AvgPool3DGrad

See Also:

AvgPool3DGrad

angle

public <T> Angle<Float> angle(Operand<T> input)

Adds an Angle operation to the graph

Parameters:

input -

Returns:

a new instance of Angle

See Also:

Angle

resourceScatterMin

public <T extends Number,U> ResourceScatterMin resourceScatterMin(Operand<?> resource,
                                                                  Operand<T> indices,
                                                                  Operand<U> updates)

Adds an ResourceScatterMin operation to the graph

Parameters:

resource - Should be from a `Variable` node.

indices - A tensor of indices into the first dimension of `ref`.

updates - A tensor of updated values to add to `ref`.

Returns:

a new instance of ResourceScatterMin

See Also:

ResourceScatterMin

sin

public <T> Sin<T> sin(Operand<T> x)

Adds an Sin operation to the graph

Parameters:

x -

Returns:

a new instance of Sin

See Also:

Sin

saveV2

public SaveV2 saveV2(Operand<String> prefix,
                     Operand<String> tensorNames,
                     Operand<String> shapeAndSlices,
                     Iterable<Operand<?>> tensors)

Adds an SaveV2 operation to the graph

Parameters:

prefix - Must have a single element. The prefix of the V2 checkpoint to which we

tensorNames - shape {N}. The names of the tensors to be saved.

shapeAndSlices - shape {N}. The slice specs of the tensors to be saved.

tensors - `N` tensors to save.

Returns:

a new instance of SaveV2

See Also:

SaveV2

resourceSparseApplyAdadelta

public <T,U extends Number> ResourceSparseApplyAdadelta resourceSparseApplyAdadelta(Operand<?> var,
                                                                                    Operand<?> accum,
                                                                                    Operand<?> accumUpdate,
                                                                                    Operand<T> lr,
                                                                                    Operand<T> rho,
                                                                                    Operand<T> epsilon,
                                                                                    Operand<T> grad,
                                                                                    Operand<U> indices,
                                                                                    ResourceSparseApplyAdadelta.Options... options)

Adds an ResourceSparseApplyAdadelta operation to the graph

Parameters:

var -

accum - Should be from a Variable().

accumUpdate - : Should be from a Variable().

lr - Learning rate. Must be a scalar.

rho - Decay factor. Must be a scalar.

epsilon - Constant factor. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyAdadelta

See Also:

ResourceSparseApplyAdadelta

randomPoisson

public <U extends Number,T extends Number> RandomPoisson<U> randomPoisson(Operand<T> shape,
                                                                          Operand<U> rate,
                                                                          RandomPoisson.Options... options)

Adds an RandomPoisson operation to the graph

Parameters:

shape -

rate -

options - carries optional attributes values

Returns:

a new instance of RandomPoisson

See Also:

RandomPoisson

scatterNdNonAliasingAdd

public <T,U extends Number> ScatterNdNonAliasingAdd<T> scatterNdNonAliasingAdd(Operand<T> input,
                                                                               Operand<U> indices,
                                                                               Operand<T> updates)

Adds an ScatterNdNonAliasingAdd operation to the graph

Parameters:

input - A Tensor.

indices - A Tensor. Must be one of the following types: `int32`, `int64`.

updates - A Tensor. Must have the same type as ref. A tensor of updated values

Returns:

a new instance of ScatterNdNonAliasingAdd

See Also:

ScatterNdNonAliasingAdd

tensorArrayGather

public <T> TensorArrayGather<T> tensorArrayGather(Operand<?> handle,
                                                  Operand<Integer> indices,
                                                  Operand<Float> flowIn,
                                                  Class<T> dtype,
                                                  TensorArrayGather.Options... options)

Adds an TensorArrayGather operation to the graph

Parameters:

handle - The handle to a TensorArray.

indices - The locations in the TensorArray from which to read tensor elements.

flowIn - A float scalar that enforces proper chaining of operations.

dtype - The type of the elem that is returned.

options - carries optional attributes values

Returns:

a new instance of TensorArrayGather

See Also:

TensorArrayGather

cosh

public <T> Cosh<T> cosh(Operand<T> x)

Adds an Cosh operation to the graph

Parameters:

x -

Returns:

a new instance of Cosh

See Also:

Cosh

unique

public <T,U extends Number> Unique<T,U> unique(Operand<T> x,
                                               Class<U> outIdx)

Adds an Unique operation to the graph

Parameters:

x - 1-D.

outIdx -

Returns:

a new instance of Unique

See Also:

Unique

tensorListConcatLists

public <T> TensorListConcatLists tensorListConcatLists(Operand<?> inputA,
                                                       Operand<?> inputB,
                                                       Class<T> elementDtype)

Adds an TensorListConcatLists operation to the graph

Parameters:

inputA -

inputB -

elementDtype -

Returns:

a new instance of TensorListConcatLists

See Also:

TensorListConcatLists

wholeFileReader

public WholeFileReader wholeFileReader(WholeFileReader.Options... options)

Adds an WholeFileReader operation to the graph

Parameters:

options - carries optional attributes values

Returns:

a new instance of WholeFileReader

See Also:

WholeFileReader

deserializeIterator

public DeserializeIterator deserializeIterator(Operand<?> resourceHandle,
                                               Operand<?> serialized)

Adds an DeserializeIterator operation to the graph

Parameters:

resourceHandle - A handle to an iterator resource.

serialized - A variant tensor storing the state of the iterator contained in the

Returns:

a new instance of DeserializeIterator

See Also:

DeserializeIterator

accumulatorSetGlobalStep

public AccumulatorSetGlobalStep accumulatorSetGlobalStep(Operand<String> handle,
                                                         Operand<Long> newGlobalStep)

Adds an AccumulatorSetGlobalStep operation to the graph

Parameters:

handle - The handle to an accumulator.

newGlobalStep - The new global_step value to set.

Returns:

a new instance of AccumulatorSetGlobalStep

See Also:

AccumulatorSetGlobalStep

lookupTableFind

public <U,T> LookupTableFind<U> lookupTableFind(Operand<?> tableHandle,
                                                Operand<T> keys,
                                                Operand<U> defaultValue)

Adds an LookupTableFind operation to the graph

Parameters:

tableHandle - Handle to the table.

keys - Any shape. Keys to look up.

defaultValue -

Returns:

a new instance of LookupTableFind

See Also:

LookupTableFind

range

public <T extends Number> Range<T> range(Operand<T> start,
                                         Operand<T> limit,
                                         Operand<T> delta)

Adds an Range operation to the graph

Parameters:

start - 0-D (scalar). First entry in the sequence.

limit - 0-D (scalar). Upper limit of sequence, exclusive.

delta - 0-D (scalar). Optional. Default is 1. Number that increments `start`.

Returns:

a new instance of Range

See Also:

Range

denseToSparseSetOperation

public <T> DenseToSparseSetOperation<T> denseToSparseSetOperation(Operand<T> set1,
                                                                  Operand<Long> set2Indices,
                                                                  Operand<T> set2Values,
                                                                  Operand<Long> set2Shape,
                                                                  String setOperation,
                                                                  DenseToSparseSetOperation.Options... options)

Adds an DenseToSparseSetOperation operation to the graph

Parameters:

set1 - `Tensor` with rank `n`. 1st `n-1` dimensions must be the same as `set2`.

set2Indices - 2D `Tensor`, indices of a `SparseTensor`. Must be in row-major

set2Values - 1D `Tensor`, values of a `SparseTensor`. Must be in row-major

set2Shape - 1D `Tensor`, shape of a `SparseTensor`. `set2_shape[0...n-1]` must

setOperation -

options - carries optional attributes values

Returns:

a new instance of DenseToSparseSetOperation

See Also:

DenseToSparseSetOperation

stringSplit

public StringSplit stringSplit(Operand<String> input,
                               Operand<String> delimiter,
                               StringSplit.Options... options)

Adds an StringSplit operation to the graph

Parameters:

input - 1-D. Strings to split.

delimiter - 0-D. Delimiter characters (bytes), or empty string.

options - carries optional attributes values

Returns:

a new instance of StringSplit

See Also:

StringSplit

tensorArrayGrad

public TensorArrayGrad tensorArrayGrad(Operand<?> handle,
                                       Operand<Float> flowIn,
                                       String source)

Adds an TensorArrayGrad operation to the graph

Parameters:

handle - The handle to the forward TensorArray.

flowIn - A float scalar that enforces proper chaining of operations.

source - The gradient source string, used to decide which gradient TensorArray

Returns:

a new instance of TensorArrayGrad

See Also:

TensorArrayGrad

decodeGif

public DecodeGif decodeGif(Operand<String> contents)

Adds an DecodeGif operation to the graph

Parameters:

contents - 0-D. The GIF-encoded image.

Returns:

a new instance of DecodeGif

See Also:

DecodeGif

tensorSummaryV2

public <T> TensorSummaryV2 tensorSummaryV2(Operand<String> tag,
                                           Operand<T> tensor,
                                           Operand<String> serializedSummaryMetadata)

Adds an TensorSummaryV2 operation to the graph

Parameters:

tag - A string attached to this summary. Used for organization in TensorBoard.

tensor - A tensor to serialize.

serializedSummaryMetadata - A serialized SummaryMetadata proto. Contains plugin

Returns:

a new instance of TensorSummaryV2

See Also:

TensorSummaryV2

setDiff1D

public <T,U extends Number> SetDiff1D<T,U> setDiff1D(Operand<T> x,
                                                     Operand<T> y,
                                                     Class<U> outIdx)

Adds an SetDiff1D operation to the graph

Parameters:

x - 1-D. Values to keep.

y - 1-D. Values to remove.

outIdx -

Returns:

a new instance of SetDiff1D

See Also:

SetDiff1D

constant

public <T> Constant<T> constant(Class<T> type,
                                long[] shape,
                                ByteBuffer data)

Adds an Constant operation to the graph

Parameters:

type - the tensor datatype.

shape - the tensor shape.

data - a buffer containing the tensor data.

Returns:

a constant of type `type`

Throws:

IllegalArgumentException - If the tensor datatype or shape is not compatible with the

See Also:

Constant

leftShift

public <T extends Number> LeftShift<T> leftShift(Operand<T> x,
                                                 Operand<T> y)

Adds an LeftShift operation to the graph

Parameters:

x -

y -

Returns:

a new instance of LeftShift

See Also:

LeftShift

constant

public Constant<String> constant(byte[][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. String elements are

See Also:

Constant

parseSingleSequenceExample

public ParseSingleSequenceExample parseSingleSequenceExample(Operand<String> serialized,
                                                             Operand<String> featureListDenseMissingAssumedEmpty,
                                                             Iterable<Operand<String>> contextSparseKeys,
                                                             Iterable<Operand<String>> contextDenseKeys,
                                                             Iterable<Operand<String>> featureListSparseKeys,
                                                             Iterable<Operand<String>> featureListDenseKeys,
                                                             Iterable<Operand<?>> contextDenseDefaults,
                                                             Operand<String> debugName,
                                                             List<Class<?>> contextSparseTypes,
                                                             List<Class<?>> featureListDenseTypes,
                                                             List<Class<?>> featureListSparseTypes,
                                                             ParseSingleSequenceExample.Options... options)

Adds an ParseSingleSequenceExample operation to the graph

Parameters:

serialized - A scalar containing a binary serialized SequenceExample proto.

featureListDenseMissingAssumedEmpty - A vector listing the

contextSparseKeys - A list of Ncontext_sparse string Tensors (scalars).

contextDenseKeys - A list of Ncontext_dense string Tensors (scalars).

featureListSparseKeys - A list of Nfeature_list_sparse string Tensors

featureListDenseKeys - A list of Nfeature_list_dense string Tensors (scalars).

contextDenseDefaults - A list of Ncontext_dense Tensors (some may be empty).

debugName - A scalar containing the name of the serialized proto.

contextSparseTypes - A list of Ncontext_sparse types; the data types of data in

featureListDenseTypes -

featureListSparseTypes - A list of Nfeature_list_sparse types; the data types

options - carries optional attributes values

Returns:

a new instance of ParseSingleSequenceExample

See Also:

ParseSingleSequenceExample

onesLike

public <T> OnesLike<T> onesLike(Operand<T> x)

Adds an OnesLike operation to the graph

Parameters:

x - a tensor of type T.

Returns:

a new instance of OnesLike

See Also:

OnesLike

lessEqual

public <T extends Number> LessEqual lessEqual(Operand<T> x,
                                              Operand<T> y)

Adds an LessEqual operation to the graph

Parameters:

x -

y -

Returns:

a new instance of LessEqual

See Also:

LessEqual

dilation2D

public <T extends Number> Dilation2D<T> dilation2D(Operand<T> input,
                                                   Operand<T> filter,
                                                   List<Long> strides,
                                                   List<Long> rates,
                                                   String padding)

Adds an Dilation2D operation to the graph

Parameters:

input - 4-D with shape `[batch, in_height, in_width, depth]`.

filter - 3-D with shape `[filter_height, filter_width, depth]`.

strides - The stride of the sliding window for each dimension of the input

rates - The input stride for atrous morphological dilation. Must be:

padding - The type of padding algorithm to use.

Returns:

a new instance of Dilation2D

See Also:

Dilation2D

unbatch

public <T> Unbatch<T> unbatch(Operand<T> batchedTensor,
                              Operand<Long> batchIndex,
                              Operand<Long> id,
                              Long timeoutMicros,
                              Unbatch.Options... options)

Adds an Unbatch operation to the graph

Parameters:

batchedTensor -

batchIndex -

id -

timeoutMicros -

options - carries optional attributes values

Returns:

a new instance of Unbatch

See Also:

Unbatch

constant

public Constant<Double> constant(double[][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

statelessTruncatedNormal

public <T extends Number,U extends Number> StatelessTruncatedNormal<Float> statelessTruncatedNormal(Operand<T> shape,
                                                                                                    Operand<U> seed)

Adds an StatelessTruncatedNormal operation to the graph

Parameters:

shape - The shape of the output tensor.

seed - 2 seeds (shape [2]).

Returns:

a new instance of StatelessTruncatedNormal

See Also:

StatelessTruncatedNormal

compareAndBitpack

public <T> CompareAndBitpack compareAndBitpack(Operand<T> input,
                                               Operand<T> threshold)

Adds an CompareAndBitpack operation to the graph

Parameters:

input - Values to compare against `threshold` and bitpack.

threshold - Threshold to compare against.

Returns:

a new instance of CompareAndBitpack

See Also:

CompareAndBitpack

constant

public Constant<Long> constant(long[][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

fusedBatchNormGradV2

public <T extends Number,U extends Number> FusedBatchNormGradV2<T,U> fusedBatchNormGradV2(Operand<T> yBackprop,
                                                                                          Operand<T> x,
                                                                                          Operand<Float> scale,
                                                                                          Operand<U> reserveSpace1,
                                                                                          Operand<U> reserveSpace2,
                                                                                          FusedBatchNormGradV2.Options... options)

Adds an FusedBatchNormGradV2 operation to the graph

Parameters:

yBackprop - A 4D Tensor for the gradient with respect to y.

x - A 4D Tensor for input data.

scale - A 1D Tensor for scaling factor, to scale the normalized x.

reserveSpace1 - When is_training is True, a 1D Tensor for the computed batch

reserveSpace2 - When is_training is True, a 1D Tensor for the computed batch

options - carries optional attributes values

Returns:

a new instance of FusedBatchNormGradV2

See Also:

FusedBatchNormGradV2

quantizeAndDequantize

public <T extends Number> QuantizeAndDequantize<T> quantizeAndDequantize(Operand<T> input,
                                                                         QuantizeAndDequantize.Options... options)

Adds an QuantizeAndDequantize operation to the graph

Parameters:

input -

options - carries optional attributes values

Returns:

a new instance of QuantizeAndDequantize

See Also:

QuantizeAndDequantize

randomPoissonV2

public <T extends Number,U extends Number> RandomPoissonV2<Long> randomPoissonV2(Operand<T> shape,
                                                                                 Operand<U> rate,
                                                                                 RandomPoissonV2.Options... options)

Adds an RandomPoissonV2 operation to the graph

Parameters:

shape - 1-D integer tensor. Shape of independent samples to draw from each

rate - A tensor in which each scalar is a "rate" parameter describing the

options - carries optional attributes values

Returns:

a new instance of RandomPoissonV2

See Also:

RandomPoissonV2

guaranteeConst

public <T> GuaranteeConst<T> guaranteeConst(Operand<T> input)

Adds an GuaranteeConst operation to the graph

Parameters:

input -

Returns:

a new instance of GuaranteeConst

See Also:

GuaranteeConst

sinh

public <T> Sinh<T> sinh(Operand<T> x)

Adds an Sinh operation to the graph

Parameters:

x -

Returns:

a new instance of Sinh

See Also:

Sinh

cos

public <T> Cos<T> cos(Operand<T> x)

Adds an Cos operation to the graph

Parameters:

x -

Returns:

a new instance of Cos

See Also:

Cos

mutexLock

public MutexLock mutexLock(Operand<?> mutex)

Adds an MutexLock operation to the graph

Parameters:

mutex - The mutex resource to lock.

Returns:

a new instance of MutexLock

See Also:

MutexLock

placeholderV2

public <T> PlaceholderV2<T> placeholderV2(Class<T> dtype,
                                          Shape shape)

Adds an PlaceholderV2 operation to the graph

Parameters:

dtype - The type of elements in the tensor.

shape - The shape of the tensor. The shape can be any partially-specified

Returns:

a new instance of PlaceholderV2

See Also:

PlaceholderV2

sparseSliceGrad

public <T> SparseSliceGrad<T> sparseSliceGrad(Operand<T> backpropValGrad,
                                              Operand<Long> inputIndices,
                                              Operand<Long> inputStart,
                                              Operand<Long> outputIndices)

Adds an SparseSliceGrad operation to the graph

Parameters:

backpropValGrad - 1-D. The gradient with respect to

inputIndices - 2-D. The `indices` of the input `SparseTensor`.

inputStart - 1-D. tensor represents the start of the slice.

outputIndices - 2-D. The `indices` of the sliced `SparseTensor`.

Returns:

a new instance of SparseSliceGrad

See Also:

SparseSliceGrad

approximateEqual

public <T> ApproximateEqual approximateEqual(Operand<T> x,
                                             Operand<T> y,
                                             ApproximateEqual.Options... options)

Adds an ApproximateEqual operation to the graph

Parameters:

x -

y -

options - carries optional attributes values

Returns:

a new instance of ApproximateEqual

See Also:

ApproximateEqual

argMax

public <V extends Number,T,U extends Number> ArgMax<V> argMax(Operand<T> input,
                                                              Operand<U> dimension,
                                                              Class<V> outputType)

Adds an ArgMax operation to the graph

Parameters:

input -

dimension - int32 or int64, must be in the range `[-rank(input), rank(input))`.

outputType -

Returns:

a new instance of ArgMax

See Also:

ArgMax

constant

public Constant<Long> constant(long[][][][][][] data)

Adds an Constant operation to the graph

Parameters:

data - An array containing the values to put into the new constant. The dimensions of the

See Also:

Constant

resourceSparseApplyProximalAdagrad

public <T,U extends Number> ResourceSparseApplyProximalAdagrad resourceSparseApplyProximalAdagrad(Operand<?> var,
                                                                                                  Operand<?> accum,
                                                                                                  Operand<T> lr,
                                                                                                  Operand<T> l1,
                                                                                                  Operand<T> l2,
                                                                                                  Operand<T> grad,
                                                                                                  Operand<U> indices,
                                                                                                  ResourceSparseApplyProximalAdagrad.Options... options)

Adds an ResourceSparseApplyProximalAdagrad operation to the graph

Parameters:

var - Should be from a Variable().

accum - Should be from a Variable().

lr - Learning rate. Must be a scalar.

l1 - L1 regularization. Must be a scalar.

l2 - L2 regularization. Must be a scalar.

grad - The gradient.

indices - A vector of indices into the first dimension of var and accum.

options - carries optional attributes values

Returns:

a new instance of ResourceSparseApplyProximalAdagrad

See Also:

ResourceSparseApplyProximalAdagrad

constant

public Constant<Double> constant(double data)

Adds an Constant operation to the graph

Parameters:

data - The value to put into the new constant.

Returns:

a double constant

See Also:

Constant

uniqueWithCounts

public <T,U extends Number> UniqueWithCounts<T,U> uniqueWithCounts(Operand<T> x,
                                                                   Class<U> outIdx)

Adds an UniqueWithCounts operation to the graph

Parameters:

x - 1-D.

outIdx -

Returns:

a new instance of UniqueWithCounts

See Also:

UniqueWithCounts

min

public <T,U extends Number> Min<T> min(Operand<T> input,
                                       Operand<U> axis,
                                       Min.Options... options)

Adds an Min operation to the graph

Parameters:

input - The tensor to reduce.

axis - The dimensions to reduce. Must be in the range

options - carries optional attributes values

Returns:

a new instance of Min

See Also:

Min

matrixDiagPart

public <T> MatrixDiagPart<T> matrixDiagPart(Operand<T> input)

Adds an MatrixDiagPart operation to the graph

Parameters:

input - Rank `k` tensor where `k >= 2`.

Returns:

a new instance of MatrixDiagPart

See Also:

MatrixDiagPart

dataFormatVecPermute

public <T extends Number> DataFormatVecPermute<T> dataFormatVecPermute(Operand<T> x,
                                                                       DataFormatVecPermute.Options... options)

Adds an DataFormatVecPermute operation to the graph

Parameters:

x - Vector of size 4 or Tensor of shape (4, 2) in source data format.

options - carries optional attributes values

Returns:

a new instance of DataFormatVecPermute

See Also:

DataFormatVecPermute

rFFT2D

public RFFT2D rFFT2D(Operand<Float> input,
                     Operand<Integer> fftLength)

Adds an RFFT2D operation to the graph

Parameters:

input - A float32 tensor.

fftLength - An int32 tensor of shape [2]. The FFT length for each dimension.

Returns:

a new instance of RFFT2D

See Also:

RFFT2D

batchMatrixSolve

public <T extends Number> BatchMatrixSolve<T> batchMatrixSolve(Operand<T> matrix,
                                                               Operand<T> rhs,
                                                               BatchMatrixSolve.Options... options)

Adds an BatchMatrixSolve operation to the graph

Parameters:

matrix -

rhs -

options - carries optional attributes values

Returns:

a new instance of BatchMatrixSolve

See Also:

BatchMatrixSolve

iterator

public Iterator iterator(String sharedName,
                         String container,
                         List<Class<?>> outputTypes,
                         List<Shape> outputShapes)

Adds an Iterator operation to the graph

Parameters:

sharedName -

container -

outputTypes -

outputShapes -

Returns:

a new instance of Iterator

See Also:

Iterator

transpose

public <T,U extends Number> Transpose<T> transpose(Operand<T> x,
                                                   Operand<U> perm)

Adds an Transpose operation to the graph

Parameters:

x -

perm -

Returns:

a new instance of Transpose

See Also:

Transpose

conv2D

public <T extends Number> Conv2D<T> conv2D(Operand<T> input,
                                           Operand<T> filter,
                                           List<Long> strides,
                                           String padding,
                                           Conv2D.Options... options)

Adds an Conv2D operation to the graph

Parameters:

input - A 4-D tensor. The dimension order is interpreted according to the value

filter - A 4-D tensor of shape

strides - 1-D tensor of length 4. The stride of the sliding window for each

padding - The type of padding algorithm to use.

options - carries optional attributes values

Returns:

a new instance of Conv2D

See Also:

Conv2D

withSubScope

public Ops withSubScope(String childScopeName)

Returns an API that adds operations to the graph with the provided name prefix.

See Also:

Scope#withSubScope(String)}

withName

public Ops withName(String opName)

Returns an API that uses the provided name for an op.

See Also:

Scope#withName(String)}

scope

public final Scope scope()

Returns the current scope of this API

create

public static Ops create(Graph graph)

Creates an API for adding operations to the provided graph