| לְהַפִּיל | Raise a exception to abort the process when called. |
| Abs <T extends TNumber > | Computes the absolute value of a tensor. |
| AccumulateN <T extends TType > | Returns the element-wise sum of a list of tensors. |
| AccumulatorApplyGradient | Applies a gradient to a given accumulator. |
| AccumulatorNumAccumulated | Returns the number of gradients aggregated in the given accumulators. |
| AccumulatorSetGlobalStep | Updates the accumulator with a new value for global_step. |
| AccumulatorTakeGradient <T extends TType > | Extracts the average gradient in the given ConditionalAccumulator. |
| Acos <T extends TType > | Computes acos of x element-wise. |
| Acosh <T extends TType > | Computes inverse hyperbolic cosine of x element-wise. |
| Add <T extends TType > | Returns x + y element-wise. |
| AddManySparseToTensorsMap | Add an `N`-minibatch `SparseTensor` to a `SparseTensorsMap`, return `N` handles. |
| AddN <T extends TType > | Add all input tensors element wise. |
| AddSparseToTensorsMap | Add a `SparseTensor` to a `SparseTensorsMap` return its handle. |
| AdjustContrast <T extends TNumber > | Adjust the contrast of one or more images. |
| AdjustHue <T extends TNumber > | Adjust the hue of one or more images. |
| AdjustSaturation <T extends TNumber > | Adjust the saturation of one or more images. |
| כֹּל | Computes the "logical and" of elements across dimensions of a tensor. |
| AllCandidateSampler | Generates labels for candidate sampling with a learned unigram distribution. |
| AllReduce <T extends TNumber > | Mutually reduces multiple tensors of identical type and shape. |
| AllToAll <T extends TType > | An Op to exchange data across TPU replicas. |
| Angle <U extends TNumber > | Returns the argument of a complex number. |
| AnonymousIterator | A container for an iterator resource. |
| AnonymousMemoryCache | |
| AnonymousMultiDeviceIterator | A container for a multi device iterator resource. |
| AnonymousRandomSeedGenerator | |
| AnonymousSeedGenerator | |
| כֹּל | Computes the "logical or" of elements across dimensions of a tensor. |
| ApplyAdaMax <T extends TType > | Update '*var' according to the AdaMax algorithm. |
| ApplyAdadelta <T extends TType > | Update '*var' according to the adadelta scheme. |
| ApplyAdagrad <T extends TType > | Update '*var' according to the adagrad scheme. |
| ApplyAdagradDa <T extends TType > | Update '*var' according to the proximal adagrad scheme. |
| ApplyAdagradV2 <T extends TType > | Update '*var' according to the adagrad scheme. |
| ApplyAdam <T extends TType > | Update '*var' according to the Adam algorithm. |
| ApplyAddSign <T extends TType > | Update '*var' according to the AddSign update. |
| ApplyCenteredRmsProp <T extends TType > | Update '*var' according to the centered RMSProp algorithm. |
| ApplyFtrl <T extends TType > | Update '*var' according to the Ftrl-proximal scheme. |
| ApplyGradientDescent <T extends TType > | Update '*var' by subtracting 'alpha' * 'delta' from it. |
| ApplyMomentum <T extends TType > | Update '*var' according to the momentum scheme. |
| ApplyPowerSign <T extends TType > | Update '*var' according to the AddSign update. |
| ApplyProximalAdagrad <T extends TType > | Update '*var' and '*accum' according to FOBOS with Adagrad learning rate. |
| ApplyProximalGradientDescent <T extends TType > | Update '*var' as FOBOS algorithm with fixed learning rate. |
| ApplyRmsProp <T extends TType > | Update '*var' according to the RMSProp algorithm. |
| ApproximateEqual | Returns the truth value of abs(xy) < tolerance element-wise. |
| ArgMax <V extends TNumber > | Returns the index with the largest value across dimensions of a tensor. |
| ArgMin <V extends TNumber > | Returns the index with the smallest value across dimensions of a tensor. |
| AsString | Converts each entry in the given tensor to strings. |
| Asin <T extends TType > | Computes the trignometric inverse sine of x element-wise. |
| Asinh <T extends TType > | Computes inverse hyperbolic sine of x element-wise. |
| AssertCardinalityDataset | |
| AssertNextDataset | |
| AssertThat | Asserts that the given condition is true. |
| Assign <T extends TType > | Update 'ref' by assigning 'value' to it. |
| AssignAdd <T extends TType > | Update 'ref' by adding 'value' to it. |
| AssignAddVariableOp | Adds a value to the current value of a variable. |
| AssignSub <T extends TType > | Update 'ref' by subtracting 'value' from it. |
| AssignSubVariableOp | Subtracts a value from the current value of a variable. |
| AssignVariableOp | Assigns a new value to a variable. |
| Atan <T extends TType > | Computes the trignometric inverse tangent of x element-wise. |
| Atan2 <T extends TNumber > | Computes arctangent of `y/x` element-wise, respecting signs of the arguments. |
| Atanh <T extends TType > | Computes inverse hyperbolic tangent of x element-wise. |
| AudioSpectrogram | Produces a visualization of audio data over time. |
| AudioSummary | Outputs a `Summary` protocol buffer with audio. |
| AutoShardDataset | Creates a dataset that shards the input dataset. |
| AvgPool <T extends TNumber > | Performs average pooling on the input. |
| AvgPool3d <T extends TNumber > | Performs 3D average pooling on the input. |
| AvgPool3dGrad <T extends TNumber > | Computes gradients of average pooling function. |
| AvgPoolGrad <T extends TNumber > | Computes gradients of the average pooling function. |
| BandPart <T extends TType > | Copy a tensor setting everything outside a central band in each innermost matrix to zero. |
| BandedTriangularSolve <T extends TType > | |
| מַחסוֹם | Defines a barrier that persists across different graph executions. |
| BarrierClose | Closes the given barrier. |
| BarrierIncompleteSize | Computes the number of incomplete elements in the given barrier. |
| BarrierInsertMany | For each key, assigns the respective value to the specified component. |
| BarrierReadySize | Computes the number of complete elements in the given barrier. |
| BarrierTakeMany | Takes the given number of completed elements from a barrier. |
| קְבוּצָה | Batches all input tensors nondeterministically. |
| BatchCholesky <T extends TNumber > | |
| BatchCholeskyGrad <T extends TNumber > | |
| BatchDataset | Creates a dataset that batches `batch_size` elements from `input_dataset`. |
| BatchFft | |
| BatchFft2d | |
| BatchFft3d | |
| BatchIfft | |
| BatchIfft2d | |
| BatchIfft3d | |
| BatchMatMul <T extends TType > | Multiplies slices of two tensors in batches. |
| BatchMatrixBandPart <T extends TType > | |
| BatchMatrixDeterminant <T extends TType > | |
| BatchMatrixDiag <T extends TType > | |
| BatchMatrixDiagPart <T extends TType > | |
| BatchMatrixInverse <T extends TNumber > | |
| BatchMatrixSetDiag <T extends TType > | |
| BatchMatrixSolve <T extends TNumber > | |
| BatchMatrixSolveLs <T extends TNumber > | |
| BatchMatrixTriangularSolve <T extends TNumber > | |
| BatchNormWithGlobalNormalization <T extends TType > | Batch normalization. |
| BatchNormWithGlobalNormalizationGrad <T extends TType > | Gradients for batch normalization. |
| BatchSelfAdjointEig <T extends TNumber > | |
| BatchSvd <T extends TType > | |
| BatchToSpace <T extends TType > | BatchToSpace for 4-D tensors of type T. |
| BatchToSpaceNd <T extends TType > | BatchToSpace for ND tensors of type T. |
| BesselI0 <T extends TNumber > | |
| BesselI0e <T extends TNumber > | |
| BesselI1 <T extends TNumber > | |
| BesselI1e <T extends TNumber > | |
| BesselJ0 <T extends TNumber > | |
| BesselJ1 <T extends TNumber > | |
| BesselK0 <T extends TNumber > | |
| BesselK0e <T extends TNumber > | |
| BesselK1 <T extends TNumber > | |
| BesselK1e <T extends TNumber > | |
| BesselY0 <T extends TNumber > | |
| BesselY1 <T extends TNumber > | |
| Betainc <T extends TNumber > | Compute the regularized incomplete beta integral \\(I_x(a, b)\\). |
| BiasAdd <T extends TType > | Adds `bias` to `value`. |
| BiasAddGrad <T extends TType > | The backward operation for "BiasAdd" on the "bias" tensor. |
| Bincount <T extends TNumber > | Counts the number of occurrences of each value in an integer array. |
| Bitcast <U extends TType > | Bitcasts a tensor from one type to another without copying data. |
| BitwiseAnd <T extends TNumber > | Elementwise computes the bitwise AND of `x` and `y`. |
| BitwiseOr <T extends TNumber > | Elementwise computes the bitwise OR of `x` and `y`. |
| BitwiseXor <T extends TNumber > | Elementwise computes the bitwise XOR of `x` and `y`. |
| BlockLSTM <T extends TNumber > | Computes the LSTM cell forward propagation for all the time steps. |
| BlockLSTMGrad <T extends TNumber > | Computes the LSTM cell backward propagation for the entire time sequence. |
| BoostedTreesAggregateStats | Aggregates the summary of accumulated stats for the batch. |
| BoostedTreesBucketize | Bucketize each feature based on bucket boundaries. |
| BoostedTreesCalculateBestFeatureSplit | Calculates gains for each feature and returns the best possible split information for the feature. |
| BoostedTreesCalculateBestFeatureSplitV2 | Calculates gains for each feature and returns the best possible split information for each node. |
| BoostedTreesCalculateBestGainsPerFeature | Calculates gains for each feature and returns the best possible split information for the feature. |
| BoostedTreesCenterBias | Calculates the prior from the training data (the bias) and fills in the first node with the logits' prior. |
| BoostedTreesCreateEnsemble | Creates a tree ensemble model and returns a handle to it. |
| BoostedTreesCreateQuantileStreamResource | Create the Resource for Quantile Streams. |
| BoostedTreesDeserializeEnsemble | Deserializes a serialized tree ensemble config and replaces current tree מִכלוֹל. |
| BoostedTreesEnsembleResourceHandleOp | Creates a handle to a BoostedTreesEnsembleResource |
| BoostedTreesExampleDebugOutputs | Debugging/model interpretability outputs for each example. |
| BoostedTreesFlushQuantileSummaries | Flush the quantile summaries from each quantile stream resource. |
| BoostedTreesGetEnsembleStates | Retrieves the tree ensemble resource stamp token, number of trees and growing statistics. |
| BoostedTreesMakeQuantileSummaries | Makes the summary of quantiles for the batch. |
| BoostedTreesMakeStatsSummary | Makes the summary of accumulated stats for the batch. |
| BoostedTreesPredict | Runs multiple additive regression ensemble predictors on input instances and computes the logits. |
| BoostedTreesQuantileStreamResourceAddSummaries | Add the quantile summaries to each quantile stream resource. |
| BoostedTreesQuantileStreamResourceDeserialize | Deserialize bucket boundaries and ready flag into current QuantileAccumulator. |
| BoostedTreesQuantileStreamResourceFlush | Flush the summaries for a quantile stream resource. |
| BoostedTreesQuantileStreamResourceGetBucketBoundaries | Generate the bucket boundaries for each feature based on accumulated summaries. |
| BoostedTreesQuantileStreamResourceHandleOp | Creates a handle to a BoostedTreesQuantileStreamResource. |
| BoostedTreesSerializeEnsemble | Serializes the tree ensemble to a proto. |
| BoostedTreesSparseAggregateStats | Aggregates the summary of accumulated stats for the batch. |
| BoostedTreesSparseCalculateBestFeatureSplit | Calculates gains for each feature and returns the best possible split information for the feature. |
| BoostedTreesTrainingPredict | Runs multiple additive regression ensemble predictors on input instances and computes the update to cached logits. |
| BoostedTreesUpdateEnsemble | Updates the tree ensemble by either adding a layer to the last tree being grown or by starting a new tree. |
| BoostedTreesUpdateEnsembleV2 | Updates the tree ensemble by adding a layer to the last tree being grown or by starting a new tree. |
| BroadcastDynamicShape <T extends TNumber > | Return the shape of s0 op s1 with broadcast. |
| BroadcastGradientArgs <T extends TNumber > | Return the reduction indices for computing gradients of s0 op s1 with broadcast. |
| BroadcastHelper <T extends TType > | Helper operator for performing XLA-style broadcasts Broadcasts `lhs` and `rhs` to the same rank, by adding size 1 dimensions to whichever of `lhs` and `rhs` has the lower rank, using XLA's broadcasting rules for binary operators. |
| BroadcastRecv <T extends TType > | Receives a tensor value broadcast from another device. |
| BroadcastSend <T extends TType > | Broadcasts a tensor value to one or more other devices. |
| BroadcastTo <T extends TType > | Broadcast an array for a compatible shape. |
| Bucketize | Bucketizes 'input' based on 'boundaries'. |
| BytesProducedStatsDataset | Records the bytes size of each element of `input_dataset` in a StatsAggregator. |
| CSRSparseMatrixComponents <T extends TType > | Reads out the CSR components at batch `index`. |
| CSRSparseMatrixToDense <T extends TType > | Convert a (possibly batched) CSRSparseMatrix to dense. |
| CSRSparseMatrixToSparseTensor <T extends TType > | Converts a (possibly batched) CSRSparesMatrix to a SparseTensor. |
| CSVDataset | |
| CSVDatasetV2 | |
| CTCLossV2 | Calculates the CTC Loss (log probability) for each batch entry. |
| CacheDataset | Creates a dataset that caches elements from `input_dataset`. |
| CacheDatasetV2 | |
| Cast <U extends TType > | Cast x of type SrcT to y of DstT. |
| Ceil <T extends TNumber > | Returns element-wise smallest integer not less than x. |
| CheckNumerics <T extends TNumber > | Checks a tensor for NaN, -Inf and +Inf values. |
| Cholesky <T extends TType > | Computes the Cholesky decomposition of one or more square matrices. |
| CholeskyGrad <T extends TNumber > | Computes the reverse mode backpropagated gradient of the Cholesky algorithm. |
| ChooseFastestDataset | |
| ClipByValue <T extends TType > | Clips tensor values to a specified min and max. |
| CloseSummaryWriter | |
| ClusterOutput <T extends TType > | Operator that connects the output of an XLA computation to other consumer graph nodes. |
| CollectiveGather <T extends TNumber > | Mutually accumulates multiple tensors of identical type and shape. |
| CollectivePermute <T extends TType > | An Op to permute tensors across replicated TPU instances. |
| CombinedNonMaxSuppression | Greedily selects a subset of bounding boxes in descending order of score, This operation performs non_max_suppression on the inputs per batch, across all classes. |
| CompareAndBitpack | Compare values of `input` to `threshold` and pack resulting bits into a `uint8`. |
| CompilationResult | Returns the result of a TPU compilation. |
| CompileSucceededAssert | Asserts that compilation succeeded. |
| Complex <U extends TType > | Converts two real numbers to a complex number. |
| ComplexAbs <U extends TNumber > | Computes the complex absolute value of a tensor. |
| CompressElement | Compresses a dataset element. |
| ComputeAccidentalHits | Computes the ids of the positions in sampled_candidates that match true_labels. |
| ComputeBatchSize | Computes the static batch size of a dataset sans partial batches. |
| Concat <T extends TType > | Concatenates tensors along one dimension. |
| ConcatenateDataset | Creates a dataset that concatenates `input_dataset` with `another_dataset`. |
| ConditionalAccumulator | A conditional accumulator for aggregating gradients. |
| ConfigureDistributedTPU | Sets up the centralized structures for a distributed TPU system. |
| ConfigureTPUEmbedding | Sets up TPUEmbedding in a distributed TPU system. |
| Conj <T extends TType > | Returns the complex conjugate of a complex number. |
| ConjugateTranspose <T extends TType > | Shuffle dimensions of x according to a permutation and conjugate the result. |
| Constant <T extends TType > | An operator producing a constant value. |
| ConsumeMutexLock | This op consumes a lock created by `MutexLock`. |
| ControlTrigger | Does nothing. |
| Conv <T extends TType > | Wraps the XLA ConvGeneralDilated operator, documented at https://www.tensorflow.org/performance/xla/operation_semantics#conv_convolution . |
| Conv2d <T extends TNumber > | Computes a 2-D convolution given 4-D `input` and `filter` tensors. |
| Conv2dBackpropFilter <T extends TNumber > | Computes the gradients of convolution with respect to the filter. |
| Conv2dBackpropInput <T extends TNumber > | Computes the gradients of convolution with respect to the input. |
| Conv3d <T extends TNumber > | Computes a 3-D convolution given 5-D `input` and `filter` tensors. |
| Conv3dBackpropFilter <T extends TNumber > | Computes the gradients of 3-D convolution with respect to the filter. |
| Conv3dBackpropInput <U extends TNumber > | Computes the gradients of 3-D convolution with respect to the input. |
| Copy <T extends TType > | Copy a tensor from CPU-to-CPU or GPU-to-GPU. |
| CopyHost <T extends TType > | Copy a tensor to host. |
| Cos <T extends TType > | Computes cos of x element-wise. |
| Cosh <T extends TType > | Computes hyperbolic cosine of x element-wise. |
| CountUpTo <T extends TNumber > | Increments 'ref' until it reaches 'limit'. |
| CreateSummaryDbWriter | |
| CreateSummaryFileWriter | |
| CropAndResize | Extracts crops from the input image tensor and resizes them. |
| CropAndResizeGradBoxes | Computes the gradient of the crop_and_resize op wrt the input boxes tensor. |
| CropAndResizeGradImage <T extends TNumber > | Computes the gradient of the crop_and_resize op wrt the input image tensor. |
| Cross <T extends TNumber > | Compute the pairwise cross product. |
| CrossReplicaSum <T extends TNumber > | An Op to sum inputs across replicated TPU instances. |
| CtcBeamSearchDecoder <T extends TNumber > | Performs beam search decoding on the logits given in input. |
| CtcGreedyDecoder <T extends TNumber > | Performs greedy decoding on the logits given in inputs. |
| CtcLoss <T extends TNumber > | Calculates the CTC Loss (log probability) for each batch entry. |
| CudnnRNN <T extends TNumber > | A RNN backed by cuDNN. |
| CudnnRNNBackprop <T extends TNumber > | Backprop step of CudnnRNNV3. |
| CudnnRNNCanonicalToParams <T extends TNumber > | Converts CudnnRNN params from canonical form to usable form. |
| CudnnRNNParamsToCanonical <T extends TNumber > | Retrieves CudnnRNN params in canonical form. |
| CudnnRnnParamsSize <U extends TNumber > | Computes size of weights that can be used by a Cudnn RNN model. |
| Cumprod <T extends TType > | Compute the cumulative product of the tensor `x` along `axis`. |
| Cumsum <T extends TType > | Compute the cumulative sum of the tensor `x` along `axis`. |
| CumulativeLogsumexp <T extends TNumber > | Compute the cumulative product of the tensor `x` along `axis`. |
| DataFormatDimMap <T extends TNumber > | Returns the dimension index in the destination data format given the one in the source data format. |
| DataFormatVecPermute <T extends TNumber > | Permute input tensor from `src_format` to `dst_format`. |
| DataServiceDataset | |
| DatasetCardinality | Returns the cardinality of `input_dataset`. |
| DatasetFromGraph | Creates a dataset from the given `graph_def`. |
| DatasetToGraph | Returns a serialized GraphDef representing `input_dataset`. |
| DatasetToSingleElement | Outputs the single element from the given dataset. |
| DatasetToTFRecord | Writes the given dataset to the given file using the TFRecord format. |
| DatasetToTfRecord | Writes the given dataset to the given file using the TFRecord format. |
| Dawsn <T extends TNumber > | |
| DebugGradientIdentity <T extends TType > | Identity op for gradient debugging. |
| DebugGradientRefIdentity <T extends TType > | Identity op for gradient debugging. |
| DebugIdentity <T extends TType > | Debug Identity V2 Op. |
| DebugNanCount | Debug NaN Value Counter Op. |
| DebugNumericsSummary <U extends TNumber > | Debug Numeric Summary V2 Op. |
| DecodeAndCropJpeg | Decode and Crop a JPEG-encoded image to a uint8 tensor. |
| DecodeBase64 | Decode web-safe base64-encoded strings. |
| DecodeBmp | Decode the first frame of a BMP-encoded image to a uint8 tensor. |
| DecodeCompressed | Decompress strings. |
| DecodeCsv | Convert CSV records to tensors. |
| DecodeGif | Decode the frame(s) of a GIF-encoded image to a uint8 tensor. |
| DecodeImage <T extends TNumber > | Function for decode_bmp, decode_gif, decode_jpeg, and decode_png. |
| DecodeJpeg | Decode a JPEG-encoded image to a uint8 tensor. |
| DecodeJsonExample | Convert JSON-encoded Example records to binary protocol buffer strings. |
| DecodePaddedRaw <T extends TNumber > | Reinterpret the bytes of a string as a vector of numbers. |
| DecodePng <T extends TNumber > | Decode a PNG-encoded image to a uint8 or uint16 tensor. |
| DecodeProto | The op extracts fields from a serialized protocol buffers message into tensors. |
| DecodeRaw <T extends TType > | Reinterpret the bytes of a string as a vector of numbers. |
| DecodeWav | Decode a 16-bit PCM WAV file to a float tensor. |
| DeepCopy <T extends TType > | Makes a copy of `x`. |
| DeleteIterator | A container for an iterator resource. |
| DeleteMemoryCache | |
| DeleteMultiDeviceIterator | A container for an iterator resource. |
| DeleteRandomSeedGenerator | |
| DeleteSeedGenerator | |
| DeleteSessionTensor | Delete the tensor specified by its handle in the session. |
| DenseBincount <U extends TNumber > | Counts the number of occurrences of each value in an integer array. |
| DenseCountSparseOutput <U extends TNumber > | Performs sparse-output bin counting for a tf.tensor input. |
| DenseToCSRSparseMatrix | Converts a dense tensor to a (possibly batched) CSRSparseMatrix. |
| DenseToDenseSetOperation <T extends TType > | Applies set operation along last dimension of 2 `Tensor` inputs. |
| DenseToSparseBatchDataset | Creates a dataset that batches input elements into a SparseTensor. |
| DenseToSparseSetOperation <T extends TType > | Applies set operation along last dimension of `Tensor` and `SparseTensor`. |
| DepthToSpace <T extends TType > | DepthToSpace for tensors of type T. |
| DepthwiseConv2dNative <T extends TNumber > | Computes a 2-D depthwise convolution given 4-D `input` and `filter` tensors. |
| DepthwiseConv2dNativeBackpropFilter <T extends TNumber > | Computes the gradients of depthwise convolution with respect to the filter. |
| DepthwiseConv2dNativeBackpropInput <T extends TNumber > | Computes the gradients of depthwise convolution with respect to the input. |
| Dequantize | Takes the packed uint32 input and unpacks the input to uint8 to do Dequantization on device. |
| DeserializeIterator | Converts the given variant tensor to an iterator and stores it in the given resource. |
| DeserializeManySparse <T extends TType > | Deserialize and concatenate `SparseTensors` from a serialized minibatch. |
| DeserializeSparse <U extends TType > | Deserialize `SparseTensor` objects. |
| DestroyResourceOp | Deletes the resource specified by the handle. |
| DestroyTemporaryVariable <T extends TType > | Destroys the temporary variable and returns its final value. |
| Det <T extends TType > | Computes the determinant of one or more square matrices. |
| DeviceIndex | Return the index of device the op runs. |
| Digamma <T extends TNumber > | Computes Psi, the derivative of Lgamma (the log of the absolute value of `Gamma(x)`), element-wise. |
| Dilation2d <T extends TNumber > | Computes the grayscale dilation of 4-D `input` and 3-D `filter` tensors. |
| Dilation2dBackpropFilter <T extends TNumber > | Computes the gradient of morphological 2-D dilation with respect to the filter. |
| Dilation2dBackpropInput <T extends TNumber > | Computes the gradient of morphological 2-D dilation with respect to the input. |
| DirectedInterleaveDataset | A substitute for `InterleaveDataset` on a fixed list of `N` datasets. |
| Div <T extends TType > | Returns x / y element-wise. |
| DivNoNan <T extends TType > | Returns 0 if the denominator is zero. |
| Dot <T extends TType > | Wraps the XLA DotGeneral operator, documented at https://www.tensorflow.org/performance/xla/operation_semantics#dotgeneral . |
| DrawBoundingBoxes <T extends TNumber > | Draw bounding boxes on a batch of images. |
| DummyIterationCounter | |
| DummyMemoryCache | |
| DummySeedGenerator | |
| DynamicPartition <T extends TType > | Partitions `data` into `num_partitions` tensors using indices from `partitions`. |
| DynamicSlice <T extends TType > | Wraps the XLA DynamicSlice operator, documented at https://www.tensorflow.org/performance/xla/operation_semantics#dynamicslice . |
| DynamicStitch <T extends TType > | Interleave the values from the `data` tensors into a single tensor. |
| DynamicUpdateSlice <T extends TType > | Wraps the XLA DynamicUpdateSlice operator, documented at https://www.tensorflow.org/performance/xla/operation_semantics#dynamicupdateslice . |
| EditDistance | Computes the (possibly normalized) Levenshtein Edit Distance. |
| Eig <U extends TType > | Computes the eigen decomposition of one or more square matrices. |
| Einsum <T extends TType > | An op which supports basic einsum op with 2 inputs and 1 output. |
| Elu <T extends TNumber > | Computes exponential linear: `exp(features) - 1` if < 0, `features` otherwise. |
| EluGrad <T extends TNumber > | Computes gradients for the exponential linear (Elu) operation. |
| EmbeddingActivations | An op enabling differentiation of TPU Embeddings. |
| Empty <T extends TType > | Creates a tensor with the given shape. |
| EmptyTensorList | Creates and returns an empty tensor list. |
| EmptyTensorMap | Creates and returns an empty tensor map. |
| EncodeBase64 | Encode strings into web-safe base64 format. |
| EncodeJpeg | JPEG-encode an image. |
| EncodeJpegVariableQuality | JPEG encode input image with provided compression quality. |
| EncodePng | PNG-encode an image. |
| EncodeProto | The op serializes protobuf messages provided in the input tensors. |
| EncodeWav | Encode audio data using the WAV file format. |
| EnqueueTPUEmbeddingIntegerBatch | An op that enqueues a list of input batch tensors to TPUEmbedding. |
| EnqueueTPUEmbeddingRaggedTensorBatch | Eases the porting of code that uses tf.nn.embedding_lookup(). |
| EnqueueTPUEmbeddingSparseBatch | An op that enqueues TPUEmbedding input indices from a SparseTensor. |
| EnqueueTPUEmbeddingSparseTensorBatch | Eases the porting of code that uses tf.nn.embedding_lookup_sparse(). |
| EnsureShape <T extends TType > | Ensures that the tensor's shape matches the expected shape. |
| Enter <T extends TType > | Creates or finds a child frame, and makes `data` available to the child frame. |
| לְהִשְׁתַווֹת | Returns the truth value of (x == y) element-wise. |
| Erf <T extends TNumber > | Computes the Gauss error function of `x` element-wise. |
| Erfc <T extends TNumber > | Computes the complementary error function of `x` element-wise. |
| EuclideanNorm <T extends TType > | Computes the euclidean norm of elements across dimensions of a tensor. |
| לְבַצֵעַ | Op that loads and executes a TPU program on a TPU device. |
| ExecuteAndUpdateVariables | Op that executes a program with optional in-place variable updates. |
| Exit <T extends TType > | Exits the current frame to its parent frame. |
| Exp <T extends TType > | Computes exponential of x element-wise. |
| ExpandDims <T extends TType > | Inserts a dimension of 1 into a tensor's shape. |
| Expint <T extends TNumber > | |
| Expm1 <T extends TType > | Computes `exp(x) - 1` element-wise. |
| ExtractGlimpse | Extracts a glimpse from the input tensor. |
| ExtractImagePatches <T extends TType > | Extract `patches` from `images` and put them in the "depth" output dimension. |
| ExtractJpegShape <T extends TNumber > | Extract the shape information of a JPEG-encoded image. |
| ExtractVolumePatches <T extends TNumber > | Extract `patches` from `input` and put them in the `"depth"` output dimension. |
| עוּבדָה | Output a fact about factorials. |
| FakeQuantWithMinMaxArgs | Fake-quantize the 'inputs' tensor, type float to 'outputs' tensor of same type. |
| FakeQuantWithMinMaxArgsGradient | Compute gradients for a FakeQuantWithMinMaxArgs operation. |
| FakeQuantWithMinMaxVars | Fake-quantize the 'inputs' tensor of type float via global float scalars Fake-quantize the `inputs` tensor of type float via global float scalars `min` and `max` to `outputs` tensor of same shape as `inputs`. |
| FakeQuantWithMinMaxVarsGradient | Compute gradients for a FakeQuantWithMinMaxVars operation. |
| FakeQuantWithMinMaxVarsPerChannel | Fake-quantize the 'inputs' tensor of type float via per-channel floats Fake-quantize the `inputs` tensor of type float per-channel and one of the shapes: `[d]`, `[b, d]` `[b, h, w, d]` via per-channel floats `min` and `max` of shape `[d]` to `outputs` tensor of same shape as `inputs`. |
| FakeQuantWithMinMaxVarsPerChannelGradient | Compute gradients for a FakeQuantWithMinMaxVarsPerChannel operation. |
| Fft <T extends TType > | Fast Fourier transform. |
| Fft2d <T extends TType > | 2D fast Fourier transform. |
| Fft3d <T extends TType > | 3D fast Fourier transform. |
| FifoQueue | A queue that produces elements in first-in first-out order. |
| Fill <U extends TType > | Creates a tensor filled with a scalar value. |
| FilterByLastComponentDataset | Creates a dataset containing elements of first component of `input_dataset` having true in the last component. |
| טְבִיעַת אֶצבָּעוֹת | Generates fingerprint values. |
| FixedLengthRecordDataset | |
| FixedLengthRecordReader | A Reader that outputs fixed-length records from a file. |
| FixedUnigramCandidateSampler | Generates labels for candidate sampling with a learned unigram distribution. |
| Floor <T extends TNumber > | Returns element-wise largest integer not greater than x. |
| FloorDiv <T extends TType > | Returns x // y element-wise. |
| FloorMod <T extends TNumber > | Returns element-wise remainder of division. |
| FlushSummaryWriter | |
| FractionalAvgPool <T extends TNumber > | Performs fractional average pooling on the input. |
| FractionalAvgPoolGrad <T extends TNumber > | Computes gradient of the FractionalAvgPool function. |
| FractionalMaxPool <T extends TNumber > | Performs fractional max pooling on the input. |
| FractionalMaxPoolGrad <T extends TNumber > | Computes gradient of the FractionalMaxPool function. |
| FresnelCos <T extends TNumber > | |
| FresnelSin <T extends TNumber > | |
| FusedBatchNorm <T extends TNumber , U extends TNumber > | Batch normalization. |
| FusedBatchNormGrad <T extends TNumber , U extends TNumber > | Gradient for batch normalization. |
| FusedPadConv2d <T extends TNumber > | Performs a padding as a preprocess during a convolution. |
| FusedResizeAndPadConv2d <T extends TNumber > | Performs a resize and padding as a preprocess during a convolution. |
| GRUBlockCell <T extends TNumber > | Computes the GRU cell forward propagation for 1 time step. |
| GRUBlockCellGrad <T extends TNumber > | Computes the GRU cell back-propagation for 1 time step. |
| Gather <T extends TType > | Wraps the XLA Gather operator documented at https://www.tensorflow.org/xla/operation_semantics#gather |
| GatherNd <T extends TType > | Gather slices from `params` into a Tensor with shape specified by `indices`. |
| GatherV2 <T extends TNumber > | Mutually accumulates multiple tensors of identical type and shape. |
| GenerateBoundingBoxProposals | This op produces Region of Interests from given bounding boxes(bbox_deltas) encoded wrt anchors according to eq.2 in arXiv:1506.01497 The op selects top `pre_nms_topn` scoring boxes, decodes them with respect to anchors, applies non-maximal suppression on overlapping boxes with higher than `nms_threshold` intersection-over-union (iou) value, discarding boxes where shorter side is less than `min_size`. |
| GenerateVocabRemapping | Given a path to new and old vocabulary files, returns a remapping Tensor of length `num_new_vocab`, where `remapping[i]` contains the row number in the old vocabulary that corresponds to row `i` in the new vocabulary (starting at line `new_vocab_offset` and up to `num_new_vocab` entities), or `-1` if entry `i` in the new vocabulary is not in the old vocabulary. |
| GetSessionHandle | Store the input tensor in the state of the current session. |
| GetSessionTensor <T extends TType > | Get the value of the tensor specified by its handle. |
| גדול יותר | Returns the truth value of (x > y) element-wise. |
| GreaterEqual | Returns the truth value of (x >= y) element-wise. |
| GuaranteeConst <T extends TType > | Gives a guarantee to the TF runtime that the input tensor is a constant. |
| HashTable | Creates a non-initialized hash table. |
| HistogramFixedWidth <U extends TNumber > | Return histogram of values. |
| HistogramSummary | Outputs a `Summary` protocol buffer with a histogram. |
| HsvToRgb <T extends TNumber > | Convert one or more images from HSV to RGB. |
| Identity <T extends TType > | Return a tensor with the same shape and contents as the input tensor or value. |
| IdentityN | Returns a list of tensors with the same shapes and contents as the input tensors. |
| IdentityReader | A Reader that outputs the queued work as both the key and value. |
| Ifft <T extends TType > | Inverse fast Fourier transform. |
| Ifft2d <T extends TType > | Inverse 2D fast Fourier transform. |
| Ifft3d <T extends TType > | Inverse 3D fast Fourier transform. |
| Igamma <T extends TNumber > | Compute the lower regularized incomplete Gamma function `P(a, x)`. |
| IgammaGradA <T extends TNumber > | Computes the gradient of `igamma(a, x)` wrt `a`. |
| Igammac <T extends TNumber > | Compute the upper regularized incomplete Gamma function `Q(a, x)`. |
| IgnoreErrorsDataset | Creates a dataset that contains the elements of `input_dataset` ignoring errors. |
| Imag <U extends TNumber > | Returns the imaginary part of a complex number. |
| ImageProjectiveTransformV2 <T extends TNumber > | Applies the given transform to each of the images. |
| ImageProjectiveTransformV3 <T extends TNumber > | Applies the given transform to each of the images. |
| ImageSummary | Outputs a `Summary` protocol buffer with images. |
| ImmutableConst <T extends TType > | Returns immutable tensor from memory region. |
| ImportEvent | |
| InTopK | Says whether the targets are in the top `K` predictions. |
| InfeedDequeue <T extends TType > | A placeholder op for a value that will be fed into the computation. |
| InfeedDequeueTuple | Fetches multiple values from infeed as an XLA tuple. |
| InfeedEnqueue | An op which feeds a single Tensor value into the computation. |
| InfeedEnqueuePrelinearizedBuffer | An op which enqueues prelinearized buffer into TPU infeed. |
| InfeedEnqueueTuple | Feeds multiple Tensor values into the computation as an XLA tuple. |
| Init | |
| InitializeTable | Table initializer that takes two tensors for keys and values respectively. |
| InitializeTableFromDataset | |
| InitializeTableFromTextFile | Initializes a table from a text file. |
| InplaceAdd <T extends TType > | Adds v into specified rows of x. |
| InplaceSub <T extends TType > | Subtracts `v` into specified rows of `x`. |
| InplaceUpdate <T extends TType > | Updates specified rows 'i' with values 'v'. |
| Inv <T extends TType > | Computes the inverse of one or more square invertible matrices or their adjoints (conjugate transposes). |
| InvGrad <T extends TType > | Computes the gradient for the inverse of `x` wrt its input. |
| Invert <T extends TNumber > | Invert (flip) each bit of supported types; for example, type `uint8` value 01010101 becomes 10101010. |
| InvertPermutation <T extends TNumber > | Computes the inverse permutation of a tensor. |
| Irfft <U extends TNumber > | Inverse real-valued fast Fourier transform. |
| Irfft2d <U extends TNumber > | Inverse 2D real-valued fast Fourier transform. |
| Irfft3d <U extends TNumber > | Inverse 3D real-valued fast Fourier transform. |
| IsBoostedTreesEnsembleInitialized | Checks whether a tree ensemble has been initialized. |
| IsBoostedTreesQuantileStreamResourceInitialized | Checks whether a quantile stream has been initialized. |
| IsFinite | Returns which elements of x are finite. |
| IsInf | Returns which elements of x are Inf. |
| IsNan | Returns which elements of x are NaN. |
| IsVariableInitialized | Checks whether a tensor has been initialized. |
| IsotonicRegression <U extends TNumber > | Solves a batch of isotonic regression problems. |
| Iterator | |
| IteratorFromStringHandle | |
| IteratorGetDevice | Returns the name of the device on which `resource` has been placed. |
| IteratorGetNext | Gets the next output from the given iterator . |
| IteratorGetNextAsOptional | Gets the next output from the given iterator as an Optional variant. |
| IteratorGetNextSync | Gets the next output from the given iterator. |
| IteratorToStringHandle | Converts the given `resource_handle` representing an iterator to a string. |
| לְהִצְטַרֵף | Joins the strings in the given list of string tensors into one tensor; with the given separator (default is an empty separator). |
| KMC2ChainInitialization | Returns the index of a data point that should be added to the seed set. |
| KeyValueSort <T extends TNumber , U extends TType > | Wraps the XLA Sort operator, documented at https://www.tensorflow.org/performance/xla/operation_semantics#sort . |
| KmeansPlusPlusInitialization | Selects num_to_sample rows of input using the KMeans++ criterion. |
| KthOrderStatistic | Computes the Kth order statistic of a data set. |
| L2Loss <T extends TNumber > | L2 Loss. |
| LMDBDataset | Creates a dataset that emits the key-value pairs in one or more LMDB files. |
| LSTMBlockCell <T extends TNumber > | Computes the LSTM cell forward propagation for 1 time step. |
| LSTMBlockCellGrad <T extends TNumber > | Computes the LSTM cell backward propagation for 1 timestep. |
| LatencyStatsDataset | Records the latency of producing `input_dataset` elements in a StatsAggregator. |
| LeakyRelu <T extends TNumber > | Computes rectified linear: `max(features, features * alpha)`. |
| LeakyReluGrad <T extends TNumber > | Computes rectified linear gradients for a LeakyRelu operation. |
| LearnedUnigramCandidateSampler | Generates labels for candidate sampling with a learned unigram distribution. |
| LeftShift <T extends TNumber > | Elementwise computes the bitwise left-shift of `x` and `y`. |
| פָּחוֹת | Returns the truth value of (x < y) element-wise. |
| LessEqual | Returns the truth value of (x <= y) element-wise. |
| Lgamma <T extends TNumber > | Computes the log of the absolute value of `Gamma(x)` element-wise. |
| LinSpace <T extends TNumber > | Generates values in an interval. |
| LmdbDataset | |
| LmdbReader | A Reader that outputs the records from a LMDB file. |
| LoadAndRemapMatrix | Loads a 2-D (matrix) `Tensor` with name `old_tensor_name` from the checkpoint at `ckpt_path` and potentially reorders its rows and columns using the specified remappings. |
| LoadTPUEmbeddingADAMParameters | Load ADAM embedding parameters. |
| LoadTPUEmbeddingADAMParametersGradAccumDebug | Load ADAM embedding parameters with debug support. |
| LoadTPUEmbeddingAdadeltaParameters | Load Adadelta embedding parameters. |
| LoadTPUEmbeddingAdadeltaParametersGradAccumDebug | Load Adadelta parameters with debug support. |
| LoadTPUEmbeddingAdagradParameters | Load Adagrad embedding parameters. |
| LoadTPUEmbeddingAdagradParametersGradAccumDebug | Load Adagrad embedding parameters with debug support. |
| LoadTPUEmbeddingCenteredRMSPropParameters | Load centered RMSProp embedding parameters. |
| LoadTPUEmbeddingFTRLParameters | Load FTRL embedding parameters. |
| LoadTPUEmbeddingFTRLParametersGradAccumDebug | Load FTRL embedding parameters with debug support. |
| LoadTPUEmbeddingMDLAdagradLightParameters | Load MDL Adagrad Light embedding parameters. |
| LoadTPUEmbeddingMomentumParameters | Load Momentum embedding parameters. |
| LoadTPUEmbeddingMomentumParametersGradAccumDebug | Load Momentum embedding parameters with debug support. |
| LoadTPUEmbeddingProximalAdagradParameters | Load proximal Adagrad embedding parameters. |
| LoadTPUEmbeddingProximalAdagradParametersGradAccumDebug | Load proximal Adagrad embedding parameters with debug support. |
| LoadTPUEmbeddingProximalYogiParameters | |
| LoadTPUEmbeddingProximalYogiParametersGradAccumDebug | |
| LoadTPUEmbeddingRMSPropParameters | Load RMSProp embedding parameters. |
| LoadTPUEmbeddingRMSPropParametersGradAccumDebug | Load RMSProp embedding parameters with debug support. |
| LoadTPUEmbeddingStochasticGradientDescentParameters | Load SGD embedding parameters. |
| LoadTPUEmbeddingStochasticGradientDescentParametersGradAccumDebug | Load SGD embedding parameters. |
| LocalResponseNormalization <T extends TNumber > | Local Response Normalization. |
| LocalResponseNormalizationGrad <T extends TNumber > | Gradients for Local Response Normalization. |
| Log <T extends TType > | Computes natural logarithm of x element-wise. |
| Log1p <T extends TType > | Computes natural logarithm of (1 + x) element-wise. |
| LogMatrixDeterminant <T extends TType > | Computes the sign and the log of the absolute value of the determinant of one or more square matrices. |
| LogSoftmax <T extends TNumber > | Computes log softmax activations. |
| LogUniformCandidateSampler | Generates labels for candidate sampling with a log-uniform distribution. |
| LogicalAnd | Returns the truth value of x AND y element-wise. |
| LogicalNot | Returns the truth value of `NOT x` element-wise. |
| LogicalOr | Returns the truth value of x OR y element-wise. |
| LookupTableExport <T extends TType , U extends TType > | Outputs all keys and values in the table. |
| LookupTableFind <U extends TType > | Looks up keys in a table, outputs the corresponding values. |
| LookupTableImport | Replaces the contents of the table with the specified keys and values. |
| LookupTableInsert | Updates the table to associates keys with values. |
| LookupTableRemove | Removes keys and its associated values from a table. |
| LookupTableSize | Computes the number of elements in the given table. |
| LoopCond | Forwards the input to the output. |
| לְהוֹרִיד | Converts all uppercase characters into their respective lowercase replacements. |
| LowerBound <U extends TNumber > | Applies lower_bound(sorted_search_values, values) along each row. |
| Lu <T extends TType , U extends TNumber > | Computes the LU decomposition of one or more square matrices. |
| MakeIterator | Makes a new iterator from the given `dataset` and stores it in `iterator`. |
| MakeUnique | Make all elements in the non-Batch dimension unique, but \"close\" to their initial value. |
| MapClear | Op removes all elements in the underlying container. |
| MapIncompleteSize | Op returns the number of incomplete elements in the underlying container. |
| MapPeek | Op peeks at the values at the specified key. |
| MapSize | Op returns the number of elements in the underlying container. |
| MapStage | Stage (key, values) in the underlying container which behaves like a hashtable. |
| MapUnstage | Op removes and returns the values associated with the key from the underlying container. |
| MapUnstageNoKey | Op removes and returns a random (key, value) from the underlying container. |
| MatMul <T extends TType > | Multiply the matrix "a" by the matrix "b". |
| MatchingFiles | Returns the set of files matching one or more glob patterns. |
| MatchingFilesDataset | |
| MatrixDiag <T extends TType > | Returns a batched diagonal tensor with given batched diagonal values. |
| MatrixDiagPart <T extends TType > | Returns the batched diagonal part of a batched tensor. |
| MatrixDiagPartV3 <T extends TType > | Returns the batched diagonal part of a batched tensor. |
| MatrixDiagV3 <T extends TType > | Returns a batched diagonal tensor with given batched diagonal values. |
| MatrixLogarithm <T extends TType > | Computes the matrix logarithm of one or more square matrices: \\(log(exp(A)) = A\\) This op is only defined for complex matrices. |
| MatrixSetDiag <T extends TType > | Returns a batched matrix tensor with new batched diagonal values. |
| MatrixSolveLs <T extends TType > | Solves one or more linear least-squares problems. |
| Max <T extends TType > | Computes the maximum of elements across dimensions of a tensor. |
| MaxIntraOpParallelismDataset | Creates a dataset that overrides the maximum intra-op parallelism. |
| MaxPool <T extends TType > | Performs max pooling on the input. |
| MaxPool3d <T extends TNumber > | Performs 3D max pooling on the input. |
| MaxPool3dGrad <U extends TNumber > | Computes gradients of 3D max pooling function. |
| MaxPool3dGradGrad <T extends TNumber > | Computes second-order gradients of the maxpooling function. |
| MaxPoolGrad <T extends TNumber > | Computes gradients of the maxpooling function. |
| MaxPoolGradGrad <T extends TNumber > | Computes second-order gradients of the maxpooling function. |
| MaxPoolGradGradWithArgmax <T extends TNumber > | Computes second-order gradients of the maxpooling function. |
| MaxPoolGradWithArgmax <T extends TNumber > | Computes gradients of the maxpooling function. |
| MaxPoolWithArgmax <T extends TNumber , U extends TNumber > | Performs max pooling on the input and outputs both max values and indices. |
| Maximum <T extends TNumber > | Returns the max of x and y (ie |
| Mean <T extends TType > | Computes the mean of elements across dimensions of a tensor. |
| Merge <T extends TType > | Forwards the value of an available tensor from `inputs` to `output`. |
| MergeSummary | Merges summaries. |
| MergeV2Checkpoints | V2 format specific: merges the metadata files of sharded checkpoints. |
| Mfcc | Transforms a spectrogram into a form that's useful for speech recognition. |
| Min <T extends TType > | Computes the minimum of elements across dimensions of a tensor. |
| Minimum <T extends TNumber > | Returns the min of x and y (ie |
| MirrorPad <T extends TType > | Pads a tensor with mirrored values. |
| MirrorPadGrad <T extends TType > | Gradient op for `MirrorPad` op. |
| MlirPassthroughOp | Wraps an arbitrary MLIR computation expressed as a module with a main() function. |
| Mod <T extends TNumber > | Returns element-wise remainder of division. |
| ModelDataset | Identity transformation that models performance. |
| Mul <T extends TType > | Returns x * y element-wise. |
| MulNoNan <T extends TType > | Returns x * y element-wise. |
| MultiDeviceIterator | Creates a MultiDeviceIterator resource. |
| MultiDeviceIteratorFromStringHandle | Generates a MultiDeviceIterator resource from its provided string handle. |
| MultiDeviceIteratorGetNextFromShard | Gets next element for the provided shard number. |
| MultiDeviceIteratorInit | Initializes the multi device iterator with the given dataset. |
| MultiDeviceIteratorToStringHandle | Produces a string handle for the given MultiDeviceIterator. |
| Multinomial <U extends TNumber > | Draws samples from a multinomial distribution. |
| MutableDenseHashTable | Creates an empty hash table that uses tensors as the backing store. |
| MutableHashTable | Creates an empty hash table. |
| MutableHashTableOfTensors | Creates an empty hash table. |
| Mutex | Creates a Mutex resource that can be locked by `MutexLock`. |
| MutexLock | Locks a mutex resource. |
| NcclAllReduce <T extends TNumber > | Outputs a tensor containing the reduction across all input tensors. |
| NcclBroadcast <T extends TNumber > | Sends `input` to all devices that are connected to the output. |
| NcclReduce <T extends TNumber > | Reduces `input` from `num_devices` using `reduction` to a single device. |
| Ndtri <T extends TNumber > | |
| NearestNeighbors | Selects the k nearest centers for each point. |
| Neg <T extends TType > | Computes numerical negative value element-wise. |
| NegTrain | Training via negative sampling. |
| NextAfter <T extends TNumber > | Returns the next representable value of `x1` in the direction of `x2`, element-wise. |
| NextIteration <T extends TType > | Makes its input available to the next iteration. |
| NoOp | Does nothing. |
| NonDeterministicInts <U extends TType > | Non-deterministically generates some integers. |
| NonMaxSuppression <T extends TNumber > | Greedily selects a subset of bounding boxes in descending order of score, pruning away boxes that have high intersection-over-union (IOU) overlap with previously selected boxes. |
| NonMaxSuppressionWithOverlaps | Greedily selects a subset of bounding boxes in descending order of score, pruning away boxes that have high overlaps with previously selected boxes. |
| NonSerializableDataset | |
| NotEqual | Returns the truth value of (x != y) element-wise. |
| NthElement <T extends TNumber > | Finds values of the `n`-th order statistic for the last dimension. |
| OneHot <U extends TType > | Returns a one-hot tensor. |
| OnesLike <T extends TType > | Returns a tensor of ones with the same shape and type as x. |
| OptimizeDataset | Creates a dataset by applying optimizations to `input_dataset`. |
| OptimizeDatasetV2 | Creates a dataset by applying related optimizations to `input_dataset`. |
| OptionalFromValue | Constructs an Optional variant from a tuple of tensors. |
| OptionalGetValue | Returns the value stored in an Optional variant or raises an error if none exists. |
| OptionalHasValue | Returns true if and only if the given Optional variant has a value. |
| OptionalNone | Creates an Optional variant with no value. |
| OrderedMapClear | Op removes all elements in the underlying container. |
| OrderedMapIncompleteSize | Op returns the number of incomplete elements in the underlying container. |
| OrderedMapPeek | Op peeks at the values at the specified key. |
| OrderedMapSize | Op returns the number of elements in the underlying container. |
| OrderedMapStage | Stage (key, values) in the underlying container which behaves like a ordered associative container. |
| OrderedMapUnstage | Op removes and returns the values associated with the key from the underlying container. |
| OrderedMapUnstageNoKey | Op removes and returns the (key, value) element with the smallest key from the underlying container. |
| OrdinalSelector | A TPU core selector Op. |
| OutfeedDequeue <T extends TType > | Retrieves a single tensor from the computation outfeed. |
| OutfeedDequeueTuple | Retrieve multiple values from the computation outfeed. |
| OutfeedDequeueTupleV2 | Retrieve multiple values from the computation outfeed. |
| OutfeedDequeueV2 <T extends TType > | Retrieves a single tensor from the computation outfeed. |
| OutfeedEnqueue | Enqueue a Tensor on the computation outfeed. |
| OutfeedEnqueueTuple | Enqueue multiple Tensor values on the computation outfeed. |
| Pad <T extends TType > | Wraps the XLA Pad operator, documented at https://www.tensorflow.org/performance/xla/operation_semantics#pad . |
| PaddedBatchDataset | Creates a dataset that batches and pads `batch_size` elements from the input. |
| PaddingFifoQueue | A queue that produces elements in first-in first-out order. |
| ParallelConcat <T extends TType > | Concatenates a list of `N` tensors along the first dimension. |
| ParallelDynamicStitch <T extends TType > | Interleave the values from the `data` tensors into a single tensor. |
| ParameterizedTruncatedNormal <U extends TNumber > | Outputs random values from a normal distribution. |
| ParseExample | Transforms a vector of tf.Example protos (as strings) into typed tensors. |
| ParseExampleDataset | Transforms `input_dataset` containing `Example` protos as vectors of DT_STRING into a dataset of `Tensor` or `SparseTensor` objects representing the parsed features. |
| ParseSequenceExample | Transforms a vector of tf.io.SequenceExample protos (as strings) into typed tensors. |
| ParseSingleExample | Transforms a tf.Example proto (as a string) into typed tensors. |
| ParseSingleSequenceExample | Transforms a scalar brain.SequenceExample proto (as strings) into typed tensors. |
| ParseTensor <T extends TType > | Transforms a serialized tensorflow.TensorProto proto into a Tensor. |
| PartitionedInput <T extends TType > | An op that groups a list of partitioned inputs together. |
| PartitionedOutput <T extends TType > | An op that demultiplexes a tensor to be sharded by XLA to a list of partitioned outputs outside the XLA computation. |
| Placeholder <T extends TType > | A placeholder op for a value that will be fed into the computation. |
| PlaceholderWithDefault <T extends TType > | A placeholder op that passes through `input` when its output is not fed. |
| Polygamma <T extends TNumber > | Compute the polygamma function \\(\psi^{(n)}(x)\\). |
| PopulationCount | Computes element-wise population count (aka |
| Pow <T extends TType > | Computes the power of one value to another. |
| PrefetchDataset | Creates a dataset that asynchronously prefetches elements from `input_dataset`. |
| Prelinearize | An op which linearizes one Tensor value to an opaque variant tensor. |
| PrelinearizeTuple | An op which linearizes multiple Tensor values to an opaque variant tensor. |
| PreventGradient <T extends TType > | An identity op that triggers an error if a gradient is requested. |
| הֶדפֵּס | Prints a string scalar. |
| PriorityQueue | A queue that produces elements sorted by the first component value. |
| PrivateThreadPoolDataset | Creates a dataset that uses a custom thread pool to compute `input_dataset`. |
| Prod <T extends TType > | Computes the product of elements across dimensions of a tensor. |
| Qr <T extends TType > | Computes the QR decompositions of one or more matrices. |
| Quantize <T extends TType > | Quantize the 'input' tensor of type float to 'output' tensor of type 'T'. |
| QuantizeAndDequantize <T extends TNumber > | Quantizes then dequantizes a tensor. |
| QuantizeAndDequantizeV3 <T extends TNumber > | Quantizes then dequantizes a tensor. |
| QuantizeAndDequantizeV4 <T extends TNumber > | Returns the gradient of `quantization.QuantizeAndDequantizeV4`. |
| QuantizeAndDequantizeV4Grad <T extends TNumber > | Returns the gradient of `QuantizeAndDequantizeV4`. |
| QuantizeDownAndShrinkRange <U extends TType > | Convert the quantized 'input' tensor into a lower-precision 'output', using the actual distribution of the values to maximize the usage of the lower bit depth and adjusting the output min and max ranges accordingly. |
| QuantizedAdd <V extends TType > | Returns x + y element-wise, working on quantized buffers. |
| QuantizedAvgPool <T extends TType > | Produces the average pool of the input tensor for quantized types. |
| QuantizedBatchNormWithGlobalNormalization <U extends TType > | Quantized Batch normalization. |
| QuantizedBiasAdd <V extends TType > | Adds Tensor 'bias' to Tensor 'input' for Quantized types. |
| QuantizedConcat <T extends TType > | Concatenates quantized tensors along one dimension. |
| QuantizedConv2DAndRelu <V extends TType > | |
| QuantizedConv2DAndReluAndRequantize <V extends TType > | |
| QuantizedConv2DAndRequantize <V extends TType > | |
| QuantizedConv2DPerChannel <V extends TType > | Computes QuantizedConv2D per channel. |
| QuantizedConv2DWithBias <V extends TType > | |
| QuantizedConv2DWithBiasAndRelu <V extends TType > | |
| QuantizedConv2DWithBiasAndReluAndRequantize <W extends TType > | |
| QuantizedConv2DWithBiasAndRequantize <W extends TType > | |
| QuantizedConv2DWithBiasSignedSumAndReluAndRequantize <X extends TType > | |
| QuantizedConv2DWithBiasSumAndRelu <V extends TType > | |
| QuantizedConv2DWithBiasSumAndReluAndRequantize <X extends TType > | |
| QuantizedConv2d <V extends TType > | Computes a 2D convolution given quantized 4D input and filter tensors. |
| QuantizedDepthwiseConv2D <V extends TType > | Computes quantized depthwise Conv2D. |
| QuantizedDepthwiseConv2DWithBias <V extends TType > | Computes quantized depthwise Conv2D with Bias. |
| QuantizedDepthwiseConv2DWithBiasAndRelu <V extends TType > | Computes quantized depthwise Conv2D with Bias and Relu. |
| QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize <W extends TType > | Computes quantized depthwise Conv2D with Bias, Relu and Requantize. |
| QuantizedInstanceNorm <T extends TType > | Quantized Instance normalization. |
| QuantizedMatMul <V extends TType > | Perform a quantized matrix multiplication of `a` by the matrix `b`. |
| QuantizedMatMulWithBias <W extends TType > | Performs a quantized matrix multiplication of `a` by the matrix `b` with bias add. |
| QuantizedMatMulWithBiasAndDequantize <W extends TNumber > | |
| QuantizedMatMulWithBiasAndRelu <V extends TType > | Perform a quantized matrix multiplication of `a` by the matrix `b` with bias add and relu fusion. |
| QuantizedMatMulWithBiasAndReluAndRequantize <W extends TType > | Perform a quantized matrix multiplication of `a` by the matrix `b` with bias add and relu and requantize fusion. |
| QuantizedMatMulWithBiasAndRequantize <W extends TType > | |
| QuantizedMaxPool <T extends TType > | Produces the max pool of the input tensor for quantized types. |
| QuantizedMul <V extends TType > | Returns x * y element-wise, working on quantized buffers. |
| QuantizedRelu <U extends TType > | Computes Quantized Rectified Linear: `max(features, 0)` |
| QuantizedRelu6 <U extends TType > | Computes Quantized Rectified Linear 6: `min(max(features, 0), 6)` |
| QuantizedReluX <U extends TType > | Computes Quantized Rectified Linear X: `min(max(features, 0), max_value)` |
| QuantizedReshape <T extends TType > | Reshapes a quantized tensor as per the Reshape op. |
| QuantizedResizeBilinear <T extends TType > | Resize quantized `images` to `size` using quantized bilinear interpolation. |
| QueueClose | Closes the given queue. |
| QueueDequeue | Dequeues a tuple of one or more tensors from the given queue. |
| QueueDequeueMany | Dequeues `n` tuples of one or more tensors from the given queue. |
| QueueDequeueUpTo | Dequeues `n` tuples of one or more tensors from the given queue. |
| QueueEnqueue | Enqueues a tuple of one or more tensors in the given queue. |
| QueueEnqueueMany | Enqueues zero or more tuples of one or more tensors in the given queue. |
| QueueIsClosed | Returns true if queue is closed. |
| QueueSize | Computes the number of elements in the given queue. |
| RaggedBincount <U extends TNumber > | Counts the number of occurrences of each value in an integer array. |
| RaggedCountSparseOutput <U extends TNumber > | Performs sparse-output bin counting for a ragged tensor input. |
| RaggedCross <T extends TType , U extends TNumber > | Generates a feature cross from a list of tensors, and returns it as a RaggedTensor. |
| RaggedGather <T extends TNumber , U extends TType > | Gather ragged slices from `params` axis `0` according to `indices`. |
| RaggedRange <U extends TNumber , T extends TNumber > | Returns a `RaggedTensor` containing the specified sequences of numbers. |
| RaggedTensorFromVariant <U extends TNumber , T extends TType > | Decodes a `variant` Tensor into a `RaggedTensor`. |
| RaggedTensorToSparse <U extends TType > | Converts a `RaggedTensor` into a `SparseTensor` with the same values. |
| RaggedTensorToTensor <U extends TType > | Create a dense tensor from a ragged tensor, possibly altering its shape. |
| RaggedTensorToVariant | Encodes a `RaggedTensor` into a `variant` Tensor. |
| RaggedTensorToVariantGradient <U extends TType > | Helper used to compute the gradient for `RaggedTensorToVariant`. |
| RandomCrop <T extends TNumber > | Randomly crop `image`. |
| RandomDataset | Creates a Dataset that returns pseudorandom numbers. |
| RandomGamma <U extends TNumber > | Outputs random values from the Gamma distribution(s) described by alpha. |
| RandomGammaGrad <T extends TNumber > | Computes the derivative of a Gamma random sample wrt |
| RandomPoisson <V extends TNumber > | Outputs random values from the Poisson distribution(s) described by rate. |
| RandomShuffle <T extends TType > | Randomly shuffles a tensor along its first dimension. |
| RandomShuffleQueue | A queue that randomizes the order of elements. |
| RandomStandardNormal <U extends TNumber > | Outputs random values from a normal distribution. |
| RandomUniform <U extends TNumber > | Outputs random values from a uniform distribution. |
| RandomUniformInt <U extends TNumber > | Outputs random integers from a uniform distribution. |
| Range <T extends TNumber > | Creates a sequence of numbers. |
| RangeDataset | Creates a dataset with a range of values. |
| דַרגָה | Returns the rank of a tensor. |
| ReadFile | Reads and outputs the entire contents of the input filename. |
| ReadVariableOp <T extends TType > | Reads the value of a variable. |
| ReaderNumRecordsProduced | Returns the number of records this Reader has produced. |
| ReaderNumWorkUnitsCompleted | Returns the number of work units this Reader has finished processing. |
| ReaderRead | Returns the next record (key, value pair) produced by a Reader. |
| ReaderReadUpTo | Returns up to `num_records` (key, value) pairs produced by a Reader. |
| ReaderReset | Restore a Reader to its initial clean state. |
| ReaderRestoreState | Restore a reader to a previously saved state. |
| ReaderSerializeState | Produce a string tensor that encodes the state of a Reader. |
| Real <U extends TNumber > | Returns the real part of a complex number. |
| RealDiv <T extends TType > | Returns x / y element-wise for real types. |
| RebatchDataset | Creates a dataset that changes the batch size. |
| RebatchDatasetV2 | Creates a dataset that changes the batch size. |
| Reciprocal <T extends TType > | Computes the reciprocal of x element-wise. |
| ReciprocalGrad <T extends TType > | Computes the gradient for the inverse of `x` wrt its input. |
| RecordInput | Emits randomized records. |
| Recv <T extends TType > | Receives the named tensor from another XLA computation. |
| RecvTPUEmbeddingActivations | An op that receives embedding activations on the TPU. |
| Reduce <T extends TNumber > | Mutually reduces multiple tensors of identical type and shape. |
| ReduceAll | Computes the "logical and" of elements across dimensions of a tensor. |
| ReduceAny | Computes the "logical or" of elements across dimensions of a tensor. |
| ReduceJoin | Joins a string Tensor across the given dimensions. |
| ReduceMax <T extends TType > | Computes the maximum of elements across dimensions of a tensor. |
| ReduceMin <T extends TType > | Computes the minimum of elements across dimensions of a tensor. |
| ReduceProd <T extends TType > | Computes the product of elements across dimensions of a tensor. |
| ReduceSum <T extends TType > | Computes the sum of elements across dimensions of a tensor. |
| ReduceV2 <T extends TNumber > | Mutually reduces multiple tensors of identical type and shape. |
| RefEnter <T extends TType > | Creates or finds a child frame, and makes `data` available to the child frame. |
| RefExit <T extends TType > | Exits the current frame to its parent frame. |
| RefIdentity <T extends TType > | Return the same ref tensor as the input ref tensor. |
| RefMerge <T extends TType > | Forwards the value of an available tensor from `inputs` to `output`. |
| RefNextIteration <T extends TType > | Makes its input available to the next iteration. |
| RefSelect <T extends TType > | Forwards the `index`th element of `inputs` to `output`. |
| RefSwitch <T extends TType > | Forwards the ref tensor `data` to the output port determined by `pred`. |
| RegexFullMatch | Check if the input matches the regex pattern. |
| RegexReplace | Replaces matches of the `pattern` regular expression in `input` with the replacement string provided in `rewrite`. |
| RegisterDataset | Registers a dataset with the tf.data service. |
| Relu <T extends TType > | Computes rectified linear: `max(features, 0)`. |
| Relu6 <T extends TNumber > | Computes rectified linear 6: `min(max(features, 0), 6)`. |
| Relu6Grad <T extends TNumber > | Computes rectified linear 6 gradients for a Relu6 operation. |
| ReluGrad <T extends TNumber > | Computes rectified linear gradients for a Relu operation. |
| RemoteFusedGraphExecute | Execute a sub graph on a remote processor. |
| RepeatDataset | Creates a dataset that emits the outputs of `input_dataset` `count` times. |
| ReplicaId | Replica ID. |
| ReplicateMetadata | Metadata indicating how the TPU computation should be replicated. |
| ReplicatedInput <T extends TType > | Connects N inputs to an N-way replicated TPU computation. |
| ReplicatedOutput <T extends TType > | Connects N outputs from an N-way replicated TPU computation. |
| RequantizationRange | Computes a range that covers the actual values present in a quantized tensor. |
| RequantizationRangePerChannel | Computes requantization range per channel. |
| Requantize <U extends TType > | Converts the quantized `input` tensor into a lower-precision `output`. |
| RequantizePerChannel <U extends TType > | Requantizes input with min and max values known per channel. |
| Reshape <T extends TType > | Reshapes a tensor. |
| ResizeArea | Resize `images` to `size` using area interpolation. |
| ResizeBicubic | Resize `images` to `size` using bicubic interpolation. |
| ResizeBicubicGrad <T extends TNumber > | Computes the gradient of bicubic interpolation. |
| ResizeBilinear | Resize `images` to `size` using bilinear interpolation. |
| ResizeBilinearGrad <T extends TNumber > | Computes the gradient of bilinear interpolation. |
| ResizeNearestNeighbor <T extends TNumber > | Resize `images` to `size` using nearest neighbor interpolation. |
| ResizeNearestNeighborGrad <T extends TNumber > | Computes the gradient of nearest neighbor interpolation. |
| ResourceAccumulatorApplyGradient | Applies a gradient to a given accumulator. |
| ResourceAccumulatorNumAccumulated | Returns the number of gradients aggregated in the given accumulators. |
| ResourceAccumulatorSetGlobalStep | Updates the accumulator with a new value for global_step. |
| ResourceAccumulatorTakeGradient <T extends TType > | Extracts the average gradient in the given ConditionalAccumulator. |
| ResourceApplyAdaMax | Update '*var' according to the AdaMax algorithm. |
| ResourceApplyAdadelta | Update '*var' according to the adadelta scheme. |
| ResourceApplyAdagrad | Update '*var' according to the adagrad scheme. |
| ResourceApplyAdagradDa | Update '*var' according to the proximal adagrad scheme. |
| ResourceApplyAdam | Update '*var' according to the Adam algorithm. |
| ResourceApplyAdamWithAmsgrad | Update '*var' according to the Adam algorithm. |
| ResourceApplyAddSign | Update '*var' according to the AddSign update. |
| ResourceApplyCenteredRmsProp | Update '*var' according to the centered RMSProp algorithm. |
| ResourceApplyFtrl | Update '*var' according to the Ftrl-proximal scheme. |
| ResourceApplyGradientDescent | Update '*var' by subtracting 'alpha' * 'delta' from it. |
| ResourceApplyKerasMomentum | Update '*var' according to the momentum scheme. |
| ResourceApplyMomentum | Update '*var' according to the momentum scheme. |
| ResourceApplyPowerSign | Update '*var' according to the AddSign update. |
| ResourceApplyProximalAdagrad | Update '*var' and '*accum' according to FOBOS with Adagrad learning rate. |
| ResourceApplyProximalGradientDescent | Update '*var' as FOBOS algorithm with fixed learning rate. |
| ResourceApplyRmsProp | Update '*var' according to the RMSProp algorithm. |
| ResourceConditionalAccumulator | A conditional accumulator for aggregating gradients. |
| ResourceCountUpTo <T extends TNumber > | Increments variable pointed to by 'resource' until it reaches 'limit'. |
| ResourceGather <U extends TType > | Gather slices from the variable pointed to by `resource` according to `indices`. |
| ResourceGatherNd <U extends TType > | |
| ResourceScatterAdd | Adds sparse updates to the variable referenced by `resource`. |
| ResourceScatterDiv | Divides sparse updates into the variable referenced by `resource`. |
| ResourceScatterMax | Reduces sparse updates into the variable referenced by `resource` using the `max` operation. |
| ResourceScatterMin | Reduces sparse updates into the variable referenced by `resource` using the `min` operation. |
| ResourceScatterMul | Multiplies sparse updates into the variable referenced by `resource`. |
| ResourceScatterNdAdd | Applies sparse addition to individual values or slices in a Variable. |
| ResourceScatterNdMax | |
| ResourceScatterNdMin | |
| ResourceScatterNdSub | Applies sparse subtraction to individual values or slices in a Variable. |
| ResourceScatterNdUpdate | Applies sparse `updates` to individual values or slices within a given variable according to `indices`. |
| ResourceScatterSub | Subtracts sparse updates from the variable referenced by `resource`. |
| ResourceScatterUpdate | Assigns sparse updates to the variable referenced by `resource`. |
| ResourceSparseApplyAdadelta | var: Should be from a Variable(). |
| ResourceSparseApplyAdagrad | Update relevant entries in '*var' and '*accum' according to the adagrad scheme. |
| ResourceSparseApplyAdagradDa | Update entries in '*var' and '*accum' according to the proximal adagrad scheme. |
| ResourceSparseApplyAdagradV2 | Update relevant entries in '*var' and '*accum' according to the adagrad scheme. |
| ResourceSparseApplyCenteredRmsProp | Update '*var' according to the centered RMSProp algorithm. |
| ResourceSparseApplyFtrl | Update relevant entries in '*var' according to the Ftrl-proximal scheme. |
| ResourceSparseApplyKerasMomentum | Update relevant entries in '*var' and '*accum' according to the momentum scheme. |
| ResourceSparseApplyMomentum | Update relevant entries in '*var' and '*accum' according to the momentum scheme. |
| ResourceSparseApplyProximalAdagrad | Sparse update entries in '*var' and '*accum' according to FOBOS algorithm. |
| ResourceSparseApplyProximalGradientDescent | Sparse update '*var' as FOBOS algorithm with fixed learning rate. |
| ResourceSparseApplyRmsProp | Update '*var' according to the RMSProp algorithm. |
| ResourceStridedSliceAssign | Assign `value` to the sliced l-value reference of `ref`. |
| לְשַׁחְזֵר | Restores tensors from a V2 checkpoint. |
| RestoreSlice <T extends TType > | Restores a tensor from checkpoint files. |
| RetrieveTPUEmbeddingADAMParameters | Retrieve ADAM embedding parameters. |
| RetrieveTPUEmbeddingADAMParametersGradAccumDebug | Retrieve ADAM embedding parameters with debug support. |
| RetrieveTPUEmbeddingAdadeltaParameters | Retrieve Adadelta embedding parameters. |
| RetrieveTPUEmbeddingAdadeltaParametersGradAccumDebug | Retrieve Adadelta embedding parameters with debug support. |
| RetrieveTPUEmbeddingAdagradParameters | Retrieve Adagrad embedding parameters. |
| RetrieveTPUEmbeddingAdagradParametersGradAccumDebug | Retrieve Adagrad embedding parameters with debug support. |
| RetrieveTPUEmbeddingCenteredRMSPropParameters | Retrieve centered RMSProp embedding parameters. |
| RetrieveTPUEmbeddingFTRLParameters | Retrieve FTRL embedding parameters. |
| RetrieveTPUEmbeddingFTRLParametersGradAccumDebug | Retrieve FTRL embedding parameters with debug support. |
| RetrieveTPUEmbeddingMDLAdagradLightParameters | Retrieve MDL Adagrad Light embedding parameters. |
| RetrieveTPUEmbeddingMomentumParameters | Retrieve Momentum embedding parameters. |
| RetrieveTPUEmbeddingMomentumParametersGradAccumDebug | Retrieve Momentum embedding parameters with debug support. |
| RetrieveTPUEmbeddingProximalAdagradParameters | Retrieve proximal Adagrad embedding parameters. |
| RetrieveTPUEmbeddingProximalAdagradParametersGradAccumDebug | Retrieve proximal Adagrad embedding parameters with debug support. |
| RetrieveTPUEmbeddingProximalYogiParameters | |
| RetrieveTPUEmbeddingProximalYogiParametersGradAccumDebug | |
| RetrieveTPUEmbeddingRMSPropParameters | Retrieve RMSProp embedding parameters. |
| RetrieveTPUEmbeddingRMSPropParametersGradAccumDebug | Retrieve RMSProp embedding parameters with debug support. |
| RetrieveTPUEmbeddingStochasticGradientDescentParameters | Retrieve SGD embedding parameters. |
| RetrieveTPUEmbeddingStochasticGradientDescentParametersGradAccumDebug | Retrieve SGD embedding parameters with debug support. |
| Reverse <T extends TType > | Reverses specific dimensions of a tensor. |
| ReverseSequence <T extends TType > | Reverses variable length slices. |
| Rfft <U extends TType > | Real-valued fast Fourier transform. |
| Rfft2d <U extends TType > | 2D real-valued fast Fourier transform. |
| Rfft3d <U extends TType > | 3D real-valued fast Fourier transform. |
| RgbToHsv <T extends TNumber > | Converts one or more images from RGB to HSV. |
| RightShift <T extends TNumber > | Elementwise computes the bitwise right-shift of `x` and `y`. |
| Rint <T extends TNumber > | Returns element-wise integer closest to x. |
| RngReadAndSkip | Advance the counter of a counter-based RNG. |
| RngSkip | Advance the counter of a counter-based RNG. |
| Roll <T extends TType > | Rolls the elements of a tensor along an axis. |
| Round <T extends TType > | Rounds the values of a tensor to the nearest integer, element-wise. |
| Rpc | Perform batches of RPC requests. |
| Rsqrt <T extends TType > | Computes reciprocal of square root of x element-wise. |
| RsqrtGrad <T extends TType > | Computes the gradient for the rsqrt of `x` wrt its input. |
| SampleDistortedBoundingBox <T extends TNumber > | Generate a single randomly distorted bounding box for an image. |
| SamplingDataset | Creates a dataset that takes a Bernoulli sample of the contents of another dataset. |
| לְהַצִיל | Saves tensors in V2 checkpoint format. |
| SaveSlices | Saves input tensors slices to disk. |
| ScalarSummary | Outputs a `Summary` protocol buffer with scalar values. |
| ScaleAndTranslate | |
| ScaleAndTranslateGrad <T extends TNumber > | |
| ScatterAdd <T extends TType > | Adds sparse updates to a variable reference. |
| ScatterDiv <T extends TType > | Divides a variable reference by sparse updates. |
| ScatterMax <T extends TNumber > | Reduces sparse updates into a variable reference using the `max` operation. |
| ScatterMin <T extends TNumber > | Reduces sparse updates into a variable reference using the `min` operation. |
| ScatterMul <T extends TType > | Multiplies sparse updates into a variable reference. |
| ScatterNd <U extends TType > | Scatter `updates` into a new tensor according to `indices`. |
| ScatterNdAdd <T extends TType > | Applies sparse addition to individual values or slices in a Variable. |
| ScatterNdMax <T extends TType > | Computes element-wise maximum. |
| ScatterNdMin <T extends TType > | Computes element-wise minimum. |
| ScatterNdNonAliasingAdd <T extends TType > | Applies sparse addition to `input` using individual values or slices from `updates` according to indices `indices`. |
| ScatterNdSub <T extends TType > | Applies sparse subtraction to individual values or slices in a Variable. |
| ScatterNdUpdate <T extends TType > | Applies sparse `updates` to individual values or slices within a given variable according to `indices`. |
| ScatterSub <T extends TType > | Subtracts sparse updates to a variable reference. |
| ScatterUpdate <T extends TType > | Applies sparse updates to a variable reference. |
| SdcaFprint | Computes fingerprints of the input strings. |
| SdcaOptimizer | Distributed version of Stochastic Dual Coordinate Ascent (SDCA) optimizer for linear models with L1 + L2 regularization. |
| SdcaShrinkL1 | Applies L1 regularization shrink step on the parameters. |
| SegmentMax <T extends TNumber > | Computes the maximum along segments of a tensor. |
| SegmentMean <T extends TType > | Computes the mean along segments of a tensor. |
| SegmentMin <T extends TNumber > | Computes the minimum along segments of a tensor. |
| SegmentProd <T extends TType > | Computes the product along segments of a tensor. |
| SegmentSum <T extends TType > | Computes the sum along segments of a tensor. |
| Select <T extends TType > | |
| SelfAdjointEig <T extends TType > | Computes the eigen decomposition of a batch of self-adjoint matrices (Note: Only real inputs are supported). |
| Selu <T extends TNumber > | Computes scaled exponential linear: `scale * alpha * (exp(features) - 1)` if < 0, `scale * features` otherwise. |
| SeluGrad <T extends TNumber > | Computes gradients for the scaled exponential linear (Selu) operation. |
| לִשְׁלוֹחַ | Sends the named tensor to another XLA computation. |
| SendTPUEmbeddingGradients | Performs gradient updates of embedding tables. |
| SerializeIterator | Converts the given `resource_handle` representing an iterator to a variant tensor. |
| SerializeManySparse <U extends TType > | Serialize an `N`-minibatch `SparseTensor` into an `[N, 3]` `Tensor` object. |
| SerializeSparse <U extends TType > | Serialize a `SparseTensor` into a `[3]` `Tensor` object. |
| SerializeTensor | Transforms a Tensor into a serialized TensorProto proto. |
| SetDiff1d <T extends TType , U extends TNumber > | Computes the difference between two lists of numbers or strings. |
| SetSize | Number of unique elements along last dimension of input `set`. |
| SetStatsAggregatorDataset | |
| Shape <U extends TNumber > | Returns the shape of a tensor. |
| ShapeN <U extends TNumber > | Returns shape of tensors. |
| ShardDataset | Creates a `Dataset` that includes only 1/`num_shards` of this dataset. |
| ShardedFilename | Generate a sharded filename. |
| ShardedFilespec | Generate a glob pattern matching all sharded file names. |
| Sharding <T extends TType > | An op which shards the input based on the given sharding attribute. |
| ShuffleAndRepeatDataset | |
| ShuffleDataset | |
| ShutdownDistributedTPU | Shuts down a running distributed TPU system. |
| Sigmoid <T extends TType > | Computes sigmoid of `x` element-wise. |
| SigmoidGrad <T extends TType > | Computes the gradient of the sigmoid of `x` wrt its input. |
| Sign <T extends TType > | Returns an element-wise indication of the sign of a number. |
| Sin <T extends TType > | Computes sine of x element-wise. |
| Sinh <T extends TType > | Computes hyperbolic sine of x element-wise. |
| Size <U extends TNumber > | Returns the size of a tensor. |
| SkipDataset | Creates a dataset that skips `count` elements from the `input_dataset`. |
| Skipgram | Parses a text file and creates a batch of examples. |
| SleepDataset | |
| Slice <T extends TType > | Return a slice from 'input'. |
| SlidingWindowDataset | Creates a dataset that passes a sliding window over `input_dataset`. |
| Snapshot <T extends TType > | Returns a copy of the input tensor. |
| SobolSample <T extends TNumber > | Generates points from the Sobol sequence. |
| Softmax <T extends TNumber > | Computes softmax activations. |
| SoftmaxCrossEntropyWithLogits <T extends TNumber > | Computes softmax cross entropy cost and gradients to backpropagate. |
| Softplus <T extends TNumber > | Computes softplus: `log(exp(features) + 1)`. |
| SoftplusGrad <T extends TNumber > | Computes softplus gradients for a softplus operation. |
| Softsign <T extends TNumber > | Computes softsign: `features / (abs(features) + 1)`. |
| SoftsignGrad <T extends TNumber > | Computes softsign gradients for a softsign operation. |
| Solve <T extends TType > | Solves systems of linear equations. |
| Sort <T extends TType > | Wraps the XLA Sort operator, documented at https://www.tensorflow.org/performance/xla/operation_semantics#sort . |
| SpaceToBatch <T extends TType > | SpaceToBatch for 4-D tensors of type T. |
| SpaceToBatchNd <T extends TType > | SpaceToBatch for ND tensors of type T. |
| SpaceToDepth <T extends TType > | SpaceToDepth for tensors of type T. |
| SparseAccumulatorApplyGradient | Applies a sparse gradient to a given accumulator. |
| SparseAccumulatorTakeGradient <T extends TType > | Extracts the average sparse gradient in a SparseConditionalAccumulator. |
| SparseAdd <T extends TType > | Adds two `SparseTensor` objects to produce another `SparseTensor`. |
| SparseAddGrad <T extends TType > | The gradient operator for the SparseAdd op. |
| SparseApplyAdadelta <T extends TType > | var: Should be from a Variable(). |
| SparseApplyAdagrad <T extends TType > | Update relevant entries in '*var' and '*accum' according to the adagrad scheme. |
| SparseApplyAdagradDa <T extends TType > | Update entries in '*var' and '*accum' according to the proximal adagrad scheme. |
| SparseApplyCenteredRmsProp <T extends TType > | Update '*var' according to the centered RMSProp algorithm. |
| SparseApplyFtrl <T extends TType > | Update relevant entries in '*var' according to the Ftrl-proximal scheme. |
| SparseApplyMomentum <T extends TType > | Update relevant entries in '*var' and '*accum' according to the momentum scheme. |
| SparseApplyProximalAdagrad <T extends TType > | Sparse update entries in '*var' and '*accum' according to FOBOS algorithm. |
| SparseApplyProximalGradientDescent <T extends TType > | Sparse update '*var' as FOBOS algorithm with fixed learning rate. |
| SparseApplyRmsProp <T extends TType > | Update '*var' according to the RMSProp algorithm. |
| SparseBincount <U extends TNumber > | Counts the number of occurrences of each value in an integer array. |
| SparseConcat <T extends TType > | Concatenates a list of `SparseTensor` along the specified dimension. |
| SparseConditionalAccumulator | A conditional accumulator for aggregating sparse gradients. |
| SparseCountSparseOutput <U extends TNumber > | Performs sparse-output bin counting for a sparse tensor input. |
| SparseCross | Generates sparse cross from a list of sparse and dense tensors. |
| SparseCrossHashed | Generates sparse cross from a list of sparse and dense tensors. |
| SparseDenseCwiseAdd <T extends TType > | Adds up a SparseTensor and a dense Tensor, using these special rules: (1) Broadcasts the dense side to have the same shape as the sparse side, if eligible; (2) Then, only the dense values pointed to by the indices of the SparseTensor participate in the cwise addition. |
| SparseDenseCwiseDiv <T extends TType > | Component-wise divides a SparseTensor by a dense Tensor. |
| SparseDenseCwiseMul <T extends TType > | Component-wise multiplies a SparseTensor by a dense Tensor. |
| SparseFillEmptyRows <T extends TType > | Fills empty rows in the input 2-D `SparseTensor` with a default value. |
| SparseFillEmptyRowsGrad <T extends TType > | The gradient of SparseFillEmptyRows. |
| SparseMatMul | Multiply matrix "a" by matrix "b". |
| SparseMatrixAdd | Sparse addition of two CSR matrices, C = alpha * A + beta * B. |
| SparseMatrixMatMul <T extends TType > | Matrix-multiplies a sparse matrix with a dense matrix. |
| SparseMatrixMul | Element-wise multiplication of a sparse matrix with a dense tensor. |
| SparseMatrixNNZ | Returns the number of nonzeroes of `sparse_matrix`. |
| SparseMatrixOrderingAMD | Computes the Approximate Minimum Degree (AMD) ordering of `input`. |
| SparseMatrixSoftmax | Calculates the softmax of a CSRSparseMatrix. |
| SparseMatrixSoftmaxGrad | Calculates the gradient of the SparseMatrixSoftmax op. |
| SparseMatrixSparseCholesky | Computes the sparse Cholesky decomposition of `input`. |
| SparseMatrixSparseMatMul | Sparse-matrix-multiplies two CSR matrices `a` and `b`. |
| SparseMatrixTranspose | Transposes the inner (matrix) dimensions of a CSRSparseMatrix. |
| SparseMatrixZeros | Creates an all-zeros CSRSparseMatrix with shape `dense_shape`. |
| SparseReduceMax <T extends TNumber > | Computes the max of elements across dimensions of a SparseTensor. |
| SparseReduceMaxSparse <T extends TNumber > | Computes the max of elements across dimensions of a SparseTensor. |
| SparseReduceSum <T extends TType > | Computes the sum of elements across dimensions of a SparseTensor. |
| SparseReduceSumSparse <T extends TType > | Computes the sum of elements across dimensions of a SparseTensor. |
| SparseReorder <T extends TType > | Reorders a SparseTensor into the canonical, row-major ordering. |
| SparseReshape | Reshapes a SparseTensor to represent values in a new dense shape. |
| SparseSegmentMean <T extends TNumber > | Computes the mean along sparse segments of a tensor. |
| SparseSegmentMeanGrad <T extends TNumber > | Computes gradients for SparseSegmentMean. |
| SparseSegmentMeanWithNumSegments <T extends TNumber > | Computes the mean along sparse segments of a tensor. |
| SparseSegmentSqrtN <T extends TNumber > | Computes the sum along sparse segments of a tensor divided by the sqrt of N. |
| SparseSegmentSqrtNGrad <T extends TNumber > | Computes gradients for SparseSegmentSqrtN. |
| SparseSegmentSqrtNWithNumSegments <T extends TNumber > | Computes the sum along sparse segments of a tensor divided by the sqrt of N. |
| SparseSegmentSum <T extends TNumber > | Computes the sum along sparse segments of a tensor. |
| SparseSegmentSumWithNumSegments <T extends TNumber > | Computes the sum along sparse segments of a tensor. |
| SparseSlice <T extends TType > | Slice a `SparseTensor` based on the `start` and `size`. |
| SparseSliceGrad <T extends TType > | The gradient operator for the SparseSlice op. |
| SparseSoftmax <T extends TNumber > | Applies softmax to a batched ND `SparseTensor`. |
| SparseSoftmaxCrossEntropyWithLogits <T extends TNumber > | Computes softmax cross entropy cost and gradients to backpropagate. |
| SparseSparseMaximum <T extends TNumber > | Returns the element-wise max of two SparseTensors. |
| SparseSparseMinimum <T extends TType > | Returns the element-wise min of two SparseTensors. |
| SparseSplit <T extends TType > | Split a `SparseTensor` into `num_split` tensors along one dimension. |
| SparseTensorDenseAdd <U extends TType > | Adds up a `SparseTensor` and a dense `Tensor`, producing a dense `Tensor`. |
| SparseTensorDenseMatMul <U extends TType > | Multiply SparseTensor (of rank 2) "A" by dense matrix "B". |
| SparseTensorSliceDataset | Creates a dataset that splits a SparseTensor into elements row-wise. |
| SparseTensorToCSRSparseMatrix | Converts a SparseTensor to a (possibly batched) CSRSparseMatrix. |
| SparseToDense <U extends TType > | Converts a sparse representation into a dense tensor. |
| SparseToSparseSetOperation <T extends TType > | Applies set operation along last dimension of 2 `SparseTensor` inputs. |
| Spence <T extends TNumber > | |
| Split <T extends TType > | Splits a tensor into `num_split` tensors along one dimension. |
| SplitV <T extends TType > | Splits a tensor into `num_split` tensors along one dimension. |
| SqlDataset | Creates a dataset that executes a SQL query and emits rows of the result set. |
| Sqrt <T extends TType > | Computes square root of x element-wise. |
| SqrtGrad <T extends TType > | Computes the gradient for the sqrt of `x` wrt its input. |
| Sqrtm <T extends TType > | Computes the matrix square root of one or more square matrices: matmul(sqrtm(A), sqrtm(A)) = A The input matrix should be invertible. |
| Square <T extends TType > | Computes square of x element-wise. |
| SquaredDifference <T extends TType > | Returns conj(x - y)(x - y) element-wise. |
| Squeeze <T extends TType > | Removes dimensions of size 1 from the shape of a tensor. |
| Stack <T extends TType > | Packs a list of `N` rank-`R` tensors into one rank-`(R+1)` tensor. |
| שָׁלָב | Stage values similar to a lightweight Enqueue. |
| StageClear | Op removes all elements in the underlying container. |
| StagePeek | Op peeks at the values at the specified index. |
| StageSize | Op returns the number of elements in the underlying container. |
| StatefulRandomBinomial <V extends TNumber > | |
| StatefulStandardNormal <U extends TType > | Outputs random values from a normal distribution. |
| StatefulTruncatedNormal <U extends TType > | Outputs random values from a truncated normal distribution. |
| StatefulUniform <U extends TType > | Outputs random values from a uniform distribution. |
| StatefulUniformFullInt <U extends TType > | Outputs random integers from a uniform distribution. |
| StatefulUniformInt <U extends TType > | Outputs random integers from a uniform distribution. |
| StatelessMultinomial <V extends TNumber > | Draws samples from a multinomial distribution. |
| StatelessParameterizedTruncatedNormal <V extends TNumber > | |
| StatelessRandomBinomial <W extends TNumber > | Outputs deterministic pseudorandom random numbers from a binomial distribution. |
| StatelessRandomGamma <V extends TNumber > | Outputs deterministic pseudorandom random numbers from a gamma distribution. |
| StatelessRandomGetKeyCounterAlg | Picks the best algorithm based on device, and scrambles seed into key and counter. |
| StatelessRandomNormal <V extends TNumber > | Outputs deterministic pseudorandom values from a normal distribution. |
| StatelessRandomNormalV2 <U extends TNumber > | Outputs deterministic pseudorandom values from a normal distribution. |
| StatelessRandomPoisson <W extends TNumber > | Outputs deterministic pseudorandom random numbers from a Poisson distribution. |
| StatelessRandomUniform <V extends TNumber > | Outputs deterministic pseudorandom random values from a uniform distribution. |
| StatelessRandomUniformFullInt <V extends TNumber > | Outputs deterministic pseudorandom random integers from a uniform distribution. |
| StatelessRandomUniformFullIntV2 <U extends TNumber > | Outputs deterministic pseudorandom random integers from a uniform distribution. |
| StatelessRandomUniformInt <V extends TNumber > | Outputs deterministic pseudorandom random integers from a uniform distribution. |
| StatelessRandomUniformIntV2 <U extends TNumber > | Outputs deterministic pseudorandom random integers from a uniform distribution. |
| StatelessRandomUniformV2 <U extends TNumber > | Outputs deterministic pseudorandom random values from a uniform distribution. |
| StatelessSampleDistortedBoundingBox <T extends TNumber > | Generate a randomly distorted bounding box for an image deterministically. |
| StatelessTruncatedNormal <V extends TNumber > | Outputs deterministic pseudorandom values from a truncated normal distribution. |
| StatelessTruncatedNormalV2 <U extends TNumber > | Outputs deterministic pseudorandom values from a truncated normal distribution. |
| StaticRegexFullMatch | Check if the input matches the regex pattern. |
| StaticRegexReplace | Replaces the match of pattern in input with rewrite. |
| StatsAggregatorHandle | |
| StatsAggregatorSetSummaryWriter | Set a summary_writer_interface to record statistics using given stats_aggregator. |
| StatsAggregatorSummary | Produces a summary of any statistics recorded by the given statistics manager. |
| StopGradient <T extends TType > | Stops gradient computation. |
| StridedSlice <T extends TType > | Return a strided slice from `input`. |
| StridedSliceAssign <T extends TType > | Assign `value` to the sliced l-value reference of `ref`. |
| StridedSliceGrad <U extends TType > | Returns the gradient of `StridedSlice`. |
| StringFormat | Formats a string template using a list of tensors. |
| StringLength | String lengths of `input`. |
| StringNGrams <T extends TNumber > | Creates ngrams from ragged string data. |
| StringSplit | Split elements of `source` based on `sep` into a `SparseTensor`. |
| לְהִתְפַּשֵׁט | Strip leading and trailing whitespaces from the Tensor. |
| Sub <T extends TType > | Returns x - y element-wise. |
| Substr | Return substrings from `Tensor` of strings. |
| Sum <T extends TType > | Computes the sum of elements across dimensions of a tensor. |
| SummaryWriter | |
| Svd <T extends TType > | Computes the eigen decomposition of a batch of self-adjoint matrices (Note: Only real inputs are supported). |
| SwitchCond <T extends TType > | Forwards `data` to the output port determined by `pred`. |
| TPUCompilationResult | Returns the result of a TPU compilation. |
| TPUEmbeddingActivations | An op enabling differentiation of TPU Embeddings. |
| TPUReplicateMetadata | Metadata indicating how the TPU computation should be replicated. |
| TPUReplicatedInput <T extends TType > | Connects N inputs to an N-way replicated TPU computation. |
| TPUReplicatedOutput <T extends TType > | Connects N outputs from an N-way replicated TPU computation. |
| TakeDataset | Creates a dataset that contains `count` elements from the `input_dataset`. |
| TakeManySparseFromTensorsMap <T extends TType > | Read `SparseTensors` from a `SparseTensorsMap` and concatenate them. |
| Tan <T extends TType > | Computes tan of x element-wise. |
| Tanh <T extends TType > | Computes hyperbolic tangent of `x` element-wise. |
| TanhGrad <T extends TType > | Computes the gradient for the tanh of `x` wrt its input. |
| TemporaryVariable <T extends TType > | Returns a tensor that may be mutated, but only persists within a single step. |
| TensorArray | An array of Tensors of given size. |
| TensorArrayClose | Delete the TensorArray from its resource container. |
| TensorArrayConcat <T extends TType > | Concat the elements from the TensorArray into value `value`. |
| TensorArrayGather <T extends TType > | Gather specific elements from the TensorArray into output `value`. |
| TensorArrayGrad | Creates a TensorArray for storing the gradients of values in the given handle. |
| TensorArrayGradWithShape | Creates a TensorArray for storing multiple gradients of values in the given handle. |
| TensorArrayPack <T extends TType > | |
| TensorArrayRead <T extends TType > | Read an element from the TensorArray into output `value`. |
| TensorArrayScatter | Scatter the data from the input value into specific TensorArray elements. |
| TensorArraySize | Get the current size of the TensorArray. |
| TensorArraySplit | Split the data from the input value into TensorArray elements. |
| TensorArrayUnpack | |
| TensorArrayWrite | Push an element onto the tensor_array. |
| TensorDataset | Creates a dataset that emits `components` as a tuple of tensors once. |
| TensorDiag <T extends TType > | Returns a diagonal tensor with a given diagonal values. |
| TensorDiagPart <T extends TType > | Returns the diagonal part of the tensor. |
| TensorForestCreateTreeVariable | Creates a tree resource and returns a handle to it. |
| TensorForestTreeDeserialize | Deserializes a proto into the tree handle |
| TensorForestTreeIsInitializedOp | Checks whether a tree has been initialized. |
| TensorForestTreePredict | Output the logits for the given input data |
| TensorForestTreeResourceHandleOp | Creates a handle to a TensorForestTreeResource |
| TensorForestTreeSerialize | Serializes the tree handle to a proto |
| TensorForestTreeSize | Get the number of nodes in a tree |
| TensorListConcat <U extends TType > | Concats all tensors in the list along the 0th dimension. |
| TensorListConcatLists | |
| TensorListElementShape <T extends TNumber > | The shape of the elements of the given list, as a tensor. |
| TensorListFromTensor | Creates a TensorList which, when stacked, has the value of `tensor`. |
| TensorListGather <T extends TType > | Creates a Tensor by indexing into the TensorList. |
| TensorListGetItem <T extends TType > | |
| TensorListLength | Returns the number of tensors in the input tensor list. |
| TensorListPopBack <T extends TType > | Returns the last element of the input list as well as a list with all but that element. |
| TensorListPushBack | Returns a list which has the passed-in `Tensor` as last element and the other elements of the given list in `input_handle`. |
| TensorListPushBackBatch | |
| TensorListReserve | List of the given size with empty elements. |
| TensorListResize | Resizes the list. |
| TensorListScatter | Creates a TensorList by indexing into a Tensor. |
| TensorListScatterIntoExistingList | Scatters tensor at indices in an input list. |
| TensorListSetItem | |
| TensorListSplit | Splits a tensor into a list. |
| TensorListStack <T extends TType > | Stacks all tensors in the list. |
| TensorMapErase | Returns a tensor map with item from given key erased. |
| TensorMapHasKey | Returns whether the given key exists in the map. |
| TensorMapInsert | Returns a map that is the 'input_handle' with the given key-value pair inserted. |
| TensorMapLookup <U extends TType > | Returns the value from a given key in a tensor map. |
| TensorMapSize | Returns the number of tensors in the input tensor map. |
| TensorMapStackKeys <T extends TType > | Returns a Tensor stack of all keys in a tensor map. |
| TensorScatterNdAdd <T extends TType > | Adds sparse `updates` to an existing tensor according to `indices`. |
| TensorScatterNdMax <T extends TType > | |
| TensorScatterNdMin <T extends TType > | |
| TensorScatterNdSub <T extends TType > | Subtracts sparse `updates` from an existing tensor according to `indices`. |
| TensorScatterNdUpdate <T extends TType > | Scatter `updates` into an existing tensor according to `indices`. |
| TensorSliceDataset | Creates a dataset that emits each dim-0 slice of `components` once. |
| TensorStridedSliceUpdate <T extends TType > | Assign `value` to the sliced l-value reference of `input`. |
| TensorSummary | Outputs a `Summary` protocol buffer with a tensor and per-plugin data. |
| TextLineDataset | Creates a dataset that emits the lines of one or more text files. |
| TextLineReader | A Reader that outputs the lines of a file delimited by '\n'. |
| TfRecordDataset | Creates a dataset that emits the records from one or more TFRecord files. |
| TfRecordReader | A Reader that outputs the records from a TensorFlow Records file. |
| ThreadPoolDataset | Creates a dataset that uses a custom thread pool to compute `input_dataset`. |
| ThreadPoolHandle | Creates a dataset that uses a custom thread pool to compute `input_dataset`. |
| Tile <T extends TType > | Constructs a tensor by tiling a given tensor. |
| TileGrad <T extends TType > | Returns the gradient of `Tile`. |
| Timestamp | Provides the time since epoch in seconds. |
| ToBool | Converts a tensor to a scalar predicate. |
| ToHashBucket | Converts each string in the input Tensor to its hash mod by a number of buckets. |
| ToHashBucketFast | Converts each string in the input Tensor to its hash mod by a number of buckets. |
| ToHashBucketStrong | Converts each string in the input Tensor to its hash mod by a number of buckets. |
| ToNumber <T extends TNumber > | Converts each string in the input Tensor to the specified numeric type. |
| TopK <T extends TNumber > | Finds values and indices of the `k` largest elements for the last dimension. |
| TopKUnique | Returns the TopK unique values in the array in sorted order. |
| TopKWithUnique | Returns the TopK values in the array in sorted order. |
| Transpose <T extends TType > | Shuffle dimensions of x according to a permutation. |
| TriangularSolve <T extends TType > | Solves systems of linear equations with upper or lower triangular matrices by backsubstitution. |
| TridiagonalMatMul <T extends TType > | Calculate product with tridiagonal matrix. |
| TridiagonalSolve <T extends TType > | Solves tridiagonal systems of equations. |
| TruncateDiv <T extends TType > | Returns x / y element-wise for integer types. |
| TruncateMod <T extends TNumber > | Returns element-wise remainder of division. |
| TruncatedNormal <U extends TNumber > | Outputs random values from a truncated normal distribution. |
| TryRpc | Perform batches of RPC requests. |
| Unbatch <T extends TType > | Reverses the operation of Batch for a single output Tensor. |
| UnbatchDataset | A dataset that splits the elements of its input into multiple elements. |
| UnbatchGrad <T extends TType > | Gradient of Unbatch. |
| UncompressElement | Uncompresses a compressed dataset element. |
| UnicodeDecode <T extends TNumber > | Decodes each string in `input` into a sequence of Unicode code points. |
| UnicodeDecodeWithOffsets <T extends TNumber > | Decodes each string in `input` into a sequence of Unicode code points. |
| UnicodeEncode | Encode a tensor of ints into unicode strings. |
| UnicodeScript | Determine the script codes of a given tensor of Unicode integer code points. |
| UnicodeTranscode | Transcode the input text from a source encoding to a destination encoding. |
| UniformCandidateSampler | Generates labels for candidate sampling with a uniform distribution. |
| Unique <T extends TType , V extends TNumber > | Finds unique elements along an axis of a tensor. |
| UniqueDataset | Creates a dataset that contains the unique elements of `input_dataset`. |
| UniqueWithCounts <T extends TType , V extends TNumber > | Finds unique elements along an axis of a tensor. |
| UnravelIndex <T extends TNumber > | Converts an array of flat indices into a tuple of coordinate arrays. |
| UnsortedSegmentJoin | Joins the elements of `inputs` based on `segment_ids`. |
| UnsortedSegmentMax <T extends TNumber > | Computes the maximum along segments of a tensor. |
| UnsortedSegmentMin <T extends TNumber > | Computes the minimum along segments of a tensor. |
| UnsortedSegmentProd <T extends TType > | Computes the product along segments of a tensor. |
| UnsortedSegmentSum <T extends TType > | Computes the sum along segments of a tensor. |
| Unstack <T extends TType > | Unpacks a given dimension of a rank-`R` tensor into `num` rank-`(R-1)` tensors. |
| Unstage | Op is similar to a lightweight Dequeue. |
| UnwrapDatasetVariant | |
| עֶלִיוֹן | Converts all lowercase characters into their respective uppercase replacements. |
| UpperBound <U extends TNumber > | Applies upper_bound(sorted_search_values, values) along each row. |
| VarHandleOp | Creates a handle to a Variable resource. |
| VarIsInitializedOp | Checks whether a resource handle-based variable has been initialized. |
| Variable <T extends TType > | Holds state in the form of a tensor that persists across steps. |
| VariableShape <T extends TNumber > | Returns the shape of the variable pointed to by `resource`. |
| אֵיפֹה | Returns locations of nonzero / true values in a tensor. |
| WholeFileReader | A Reader that outputs the entire contents of a file as a value. |
| WindowDataset | Combines (nests of) input elements into a dataset of (nests of) windows. |
| WorkerHeartbeat | Worker heartbeat op. |
| WrapDatasetVariant | |
| WriteAudioSummary | Writes an audio summary. |
| WriteFile | Writes contents to the file at input filename. |
| WriteGraphSummary | Writes a graph summary. |
| WriteHistogramSummary | Writes a histogram summary. |
| WriteImageSummary | Writes an image summary. |
| WriteRawProtoSummary | Writes a serialized proto summary. |
| WriteScalarSummary | Writes a scalar summary. |
| WriteSummary | Writes a tensor summary. |
| Xdivy <T extends TType > | Returns 0 if x == 0, and x / y otherwise, elementwise. |
| XlaRecvFromHost <T extends TType > | An op to receive a tensor from the host. |
| XlaSendToHost | An op to send a tensor to the host. |
| XlaSetBound | Set a bound for the given input value as a hint to Xla compiler, returns the same value. |
| XlaSpmdFullToShardShape <T extends TType > | An op used by XLA SPMD partitioner to switch from automatic partitioning to manual partitioning. |
| XlaSpmdShardToFullShape <T extends TType > | An op used by XLA SPMD partitioner to switch from manual partitioning to automatic partitioning. |
| Xlog1py <T extends TType > | Returns 0 if x == 0, and x * log1p(y) otherwise, elementwise. |
| Xlogy <T extends TType > | Returns 0 if x == 0, and x * log(y) otherwise, elementwise. |
| ZerosLike <T extends TType > | Returns a tensor of zeros with the same shape and type as x. |
| Zeta <T extends TNumber > | Compute the Hurwitz zeta function \\(\zeta(x, q)\\). |
| ZipDataset | Creates a dataset that zips together `input_datasets`. |
| erfinv <T extends TNumber > | |