Modifier and Type | Class and Description |
---|---|
class |
AudioSpectrogram
Produces a visualization of audio data over time.
|
class |
DecodeWav
Decode a 16-bit PCM WAV file to a float tensor.
|
class |
EncodeWav
Encode audio data using the WAV file format.
|
class |
Mfcc
Transforms a spectrogram into a form that's useful for speech recognition.
|
Modifier and Type | Class and Description |
---|---|
class |
BitwiseAnd<T extends Number>
Elementwise computes the bitwise AND of `x` and `y`.
|
class |
BitwiseOr<T extends Number>
Elementwise computes the bitwise OR of `x` and `y`.
|
class |
BitwiseXor<T extends Number>
Elementwise computes the bitwise XOR of `x` and `y`.
|
class |
Invert<T extends Number>
Invert (flip) each bit of supported types; for example, type `uint8` value 01010101 becomes 10101010.
|
class |
LeftShift<T extends Number>
Elementwise computes the bitwise left-shift of `x` and `y`.
|
class |
RightShift<T extends Number>
Elementwise computes the bitwise right-shift of `x` and `y`.
|
Modifier and Type | Class and Description |
---|---|
class |
AllReduce<T extends Number>
Mutually reduces multiple tensors of identical type and shape.
|
class |
BroadcastRecv<T extends Number>
Receives a tensor value broadcast from another device.
|
class |
BroadcastSend<T extends Number>
Broadcasts a tensor value to one or more other devices.
|
Modifier and Type | Class and Description |
---|---|
class |
Abort
Raise a exception to abort the process when called.
|
class |
All
Computes the "logical and" of elements across dimensions of a tensor.
|
class |
AllToAll<T>
An Op to exchange data across TPU replicas.
|
class |
AnonymousIteratorV2
A container for an iterator resource.
|
class |
AnonymousMemoryCache |
class |
AnonymousMultiDeviceIterator
A container for a multi device iterator resource.
|
class |
AnonymousRandomSeedGenerator |
class |
Any
Computes the "logical or" of elements across dimensions of a tensor.
|
class |
ApplyAdagradV2<T>
Update '*var' according to the adagrad scheme.
|
class |
AssertNextDataset
A transformation that asserts which transformations happen next.
|
class |
AssertThat
Asserts that the given condition is true.
|
class |
Assign<T>
Update 'ref' by assigning 'value' to it.
|
class |
AssignAdd<T>
Update 'ref' by adding 'value' to it.
|
class |
AssignAddVariableOp
Adds a value to the current value of a variable.
|
class |
AssignSub<T>
Update 'ref' by subtracting 'value' from it.
|
class |
AssignSubVariableOp
Subtracts a value from the current value of a variable.
|
class |
AssignVariableOp
Assigns a new value to a variable.
|
class |
AutoShardDataset
Creates a dataset that shards the input dataset.
|
class |
Barrier
Defines a barrier that persists across different graph executions.
|
class |
BarrierClose
Closes the given barrier.
|
class |
BarrierIncompleteSize
Computes the number of incomplete elements in the given barrier.
|
class |
BarrierInsertMany
For each key, assigns the respective value to the specified component.
|
class |
BarrierReadySize
Computes the number of complete elements in the given barrier.
|
class |
BarrierTakeMany
Takes the given number of completed elements from a barrier.
|
class |
Batch
Batches all input tensors nondeterministically.
|
class |
BatchMatMulV2<T>
Multiplies slices of two tensors in batches.
|
class |
BatchToSpace<T>
BatchToSpace for 4-D tensors of type T.
|
class |
BatchToSpaceNd<T>
BatchToSpace for N-D tensors of type T.
|
class |
Bitcast<U>
Bitcasts a tensor from one type to another without copying data.
|
class |
BlockLSTM<T extends Number>
Computes the LSTM cell forward propagation for all the time steps.
|
class |
BlockLSTMGrad<T extends Number>
Computes the LSTM cell backward propagation for the entire time sequence.
|
class |
BlockLSTMGradV2<T extends Number>
Computes the LSTM cell backward propagation for the entire time sequence.
|
class |
BlockLSTMV2<T extends Number>
Computes the LSTM cell forward propagation for all the time steps.
|
class |
BoostedTreesAggregateStats
Aggregates the summary of accumulated stats for the batch.
|
class |
BoostedTreesBucketize
Bucketize each feature based on bucket boundaries.
|
class |
BoostedTreesCalculateBestFeatureSplit
Calculates gains for each feature and returns the best possible split information for the feature.
|
class |
BoostedTreesCalculateBestGainsPerFeature
Calculates gains for each feature and returns the best possible split information for the feature.
|
class |
BoostedTreesCenterBias
Calculates the prior from the training data (the bias) and fills in the first node with the logits' prior.
|
class |
BoostedTreesCreateEnsemble
Creates a tree ensemble model and returns a handle to it.
|
class |
BoostedTreesCreateQuantileStreamResource
Create the Resource for Quantile Streams.
|
class |
BoostedTreesDeserializeEnsemble
Deserializes a serialized tree ensemble config and replaces current tree
|
class |
BoostedTreesEnsembleResourceHandleOp
Creates a handle to a BoostedTreesEnsembleResource
|
class |
BoostedTreesExampleDebugOutputs
Debugging/model interpretability outputs for each example.
|
class |
BoostedTreesFlushQuantileSummaries
Flush the quantile summaries from each quantile stream resource.
|
class |
BoostedTreesGetEnsembleStates
Retrieves the tree ensemble resource stamp token, number of trees and growing statistics.
|
class |
BoostedTreesMakeQuantileSummaries
Makes the summary of quantiles for the batch.
|
class |
BoostedTreesMakeStatsSummary
Makes the summary of accumulated stats for the batch.
|
class |
BoostedTreesPredict
Runs multiple additive regression ensemble predictors on input instances and
|
class |
BoostedTreesQuantileStreamResourceAddSummaries
Add the quantile summaries to each quantile stream resource.
|
class |
BoostedTreesQuantileStreamResourceDeserialize
Deserialize bucket boundaries and ready flag into current QuantileAccumulator.
|
class |
BoostedTreesQuantileStreamResourceFlush
Flush the summaries for a quantile stream resource.
|
class |
BoostedTreesQuantileStreamResourceGetBucketBoundaries
Generate the bucket boundaries for each feature based on accumulated summaries.
|
class |
BoostedTreesQuantileStreamResourceHandleOp
Creates a handle to a BoostedTreesQuantileStreamResource.
|
class |
BoostedTreesSerializeEnsemble
Serializes the tree ensemble to a proto.
|
class |
BoostedTreesSparseAggregateStats
Aggregates the summary of accumulated stats for the batch.
|
class |
BoostedTreesSparseCalculateBestFeatureSplit
Calculates gains for each feature and returns the best possible split information for the feature.
|
class |
BoostedTreesTrainingPredict
Runs multiple additive regression ensemble predictors on input instances and
|
class |
BoostedTreesUpdateEnsemble
Updates the tree ensemble by either adding a layer to the last tree being grown
|
class |
BoostedTreesUpdateEnsembleV2
Updates the tree ensemble by adding a layer to the last tree being grown
|
class |
BroadcastDynamicShape<T extends Number>
Return the shape of s0 op s1 with broadcast.
|
class |
BroadcastGradientArgs<T extends Number>
Return the reduction indices for computing gradients of s0 op s1 with broadcast.
|
class |
BroadcastTo<T>
Broadcast an array for a compatible shape.
|
class |
Bucketize
Bucketizes 'input' based on 'boundaries'.
|
class |
CacheDatasetV2 |
class |
ChooseFastestDataset |
class |
ClipByValue<T>
Clips tensor values to a specified min and max.
|
class |
CollectiveGather<T extends Number>
Mutually accumulates multiple tensors of identical type and shape.
|
class |
CollectivePermute<T>
An Op to permute tensors across replicated TPU instances.
|
class |
CombinedNonMaxSuppression
Greedily selects a subset of bounding boxes in descending order of score,
|
class |
Concat<T>
Concatenates tensors along one dimension.
|
class |
ConfigureDistributedTPU
Sets up the centralized structures for a distributed TPU system.
|
class |
ConfigureTPUEmbedding
Sets up TPUEmbedding in a distributed TPU system.
|
class |
Constant<T>
An operator producing a constant value.
|
class |
ConsumeMutexLock
This op consumes a lock created by `MutexLock`.
|
class |
ControlTrigger
Does nothing.
|
class |
CountUpTo<T extends Number>
Increments 'ref' until it reaches 'limit'.
|
class |
CrossReplicaSum<T extends Number>
An Op to sum inputs across replicated TPU instances.
|
class |
CSVDataset |
class |
CudnnRNNBackpropV3<T extends Number>
Backprop step of CudnnRNNV3.
|
class |
CudnnRNNCanonicalToParamsV2<T extends Number>
Converts CudnnRNN params from canonical form to usable form.
|
class |
CudnnRNNParamsToCanonicalV2<T extends Number>
Retrieves CudnnRNN params in canonical form.
|
class |
CudnnRNNV3<T extends Number>
A RNN backed by cuDNN.
|
class |
CumulativeLogsumexp<T extends Number>
Compute the cumulative product of the tensor `x` along `axis`.
|
class |
DatasetCardinality
Returns the cardinality of `input_dataset`.
|
class |
DatasetFromGraph
Creates a dataset from the given `graph_def`.
|
class |
DebugGradientIdentity<T>
Identity op for gradient debugging.
|
class |
DebugGradientRefIdentity<T>
Identity op for gradient debugging.
|
class |
DecodePaddedRaw<T extends Number>
Reinterpret the bytes of a string as a vector of numbers.
|
class |
DecodeProto
The op extracts fields from a serialized protocol buffers message into tensors.
|
class |
DeepCopy<T>
Makes a copy of `x`.
|
class |
DeleteIterator
A container for an iterator resource.
|
class |
DeleteMemoryCache |
class |
DeleteMultiDeviceIterator
A container for an iterator resource.
|
class |
DeleteRandomSeedGenerator |
class |
DeleteSessionTensor
Delete the tensor specified by its handle in the session.
|
class |
DestroyResourceOp
Deletes the resource specified by the handle.
|
class |
DestroyTemporaryVariable<T>
Destroys the temporary variable and returns its final value.
|
class |
DirectedInterleaveDataset
A substitute for `InterleaveDataset` on a fixed list of `N` datasets.
|
class |
DrawBoundingBoxesV2<T extends Number>
Draw bounding boxes on a batch of images.
|
class |
DynamicPartition<T>
Partitions `data` into `num_partitions` tensors using indices from `partitions`.
|
class |
DynamicStitch<T>
Interleave the values from the `data` tensors into a single tensor.
|
class |
EditDistance
Computes the (possibly normalized) Levenshtein Edit Distance.
|
class |
Einsum<T>
Tensor contraction according to Einstein summation convention.
|
class |
Empty<T>
Creates a tensor with the given shape.
|
class |
EmptyTensorList
Creates and returns an empty tensor list.
|
class |
EncodeProto
The op serializes protobuf messages provided in the input tensors.
|
class |
EnqueueTPUEmbeddingIntegerBatch
An op that enqueues a list of input batch tensors to TPUEmbedding.
|
class |
EnqueueTPUEmbeddingSparseBatch
An op that enqueues TPUEmbedding input indices from a SparseTensor.
|
class |
EnqueueTPUEmbeddingSparseTensorBatch
Eases the porting of code that uses tf.nn.embedding_lookup_sparse().
|
class |
EnsureShape<T>
Ensures that the tensor's shape matches the expected shape.
|
class |
Enter<T>
Creates or finds a child frame, and makes `data` available to the child frame.
|
class |
EuclideanNorm<T>
Computes the euclidean norm of elements across dimensions of a tensor.
|
class |
Exit<T>
Exits the current frame to its parent frame.
|
class |
ExpandDims<T>
Inserts a dimension of 1 into a tensor's shape.
|
class |
ExperimentalAutoShardDataset
Creates a dataset that shards the input dataset.
|
class |
ExperimentalBytesProducedStatsDataset
Records the bytes size of each element of `input_dataset` in a StatsAggregator.
|
class |
ExperimentalChooseFastestDataset |
class |
ExperimentalDatasetCardinality
Returns the cardinality of `input_dataset`.
|
class |
ExperimentalDatasetToTFRecord
Writes the given dataset to the given file using the TFRecord format.
|
class |
ExperimentalDenseToSparseBatchDataset
Creates a dataset that batches input elements into a SparseTensor.
|
class |
ExperimentalLatencyStatsDataset
Records the latency of producing `input_dataset` elements in a StatsAggregator.
|
class |
ExperimentalMatchingFilesDataset |
class |
ExperimentalMaxIntraOpParallelismDataset
Creates a dataset that overrides the maximum intra-op parallelism.
|
class |
ExperimentalParseExampleDataset
Transforms `input_dataset` containing `Example` protos as vectors of DT_STRING into a dataset of `Tensor` or `SparseTensor` objects representing the parsed features.
|
class |
ExperimentalPrivateThreadPoolDataset
Creates a dataset that uses a custom thread pool to compute `input_dataset`.
|
class |
ExperimentalRandomDataset
Creates a Dataset that returns pseudorandom numbers.
|
class |
ExperimentalRebatchDataset
Creates a dataset that changes the batch size.
|
class |
ExperimentalSetStatsAggregatorDataset |
class |
ExperimentalSlidingWindowDataset
Creates a dataset that passes a sliding window over `input_dataset`.
|
class |
ExperimentalSqlDataset
Creates a dataset that executes a SQL query and emits rows of the result set.
|
class |
ExperimentalStatsAggregatorHandle
Creates a statistics manager resource.
|
class |
ExperimentalStatsAggregatorSummary
Produces a summary of any statistics recorded by the given statistics manager.
|
class |
ExperimentalUnbatchDataset
A dataset that splits the elements of its input into multiple elements.
|
class |
ExtractVolumePatches<T extends Number>
Extract `patches` from `input` and put them in the "depth" output dimension.
|
class |
Fill<U>
Creates a tensor filled with a scalar value.
|
class |
Fingerprint
Generates fingerprint values.
|
class |
FusedBatchNormGradV3<T extends Number,U extends Number>
Gradient for batch normalization.
|
class |
FusedBatchNormV3<T extends Number,U extends Number>
Batch normalization.
|
class |
Gather<T>
Gather slices from `params` axis `axis` according to `indices`.
|
class |
GatherNd<T>
Gather slices from `params` into a Tensor with shape specified by `indices`.
|
class |
GetSessionHandle
Store the input tensor in the state of the current session.
|
class |
GetSessionTensor<T>
Get the value of the tensor specified by its handle.
|
class |
GRUBlockCell<T extends Number>
Computes the GRU cell forward propagation for 1 time step.
|
class |
GRUBlockCellGrad<T extends Number>
Computes the GRU cell back-propagation for 1 time step.
|
class |
GuaranteeConst<T>
Gives a guarantee to the TF runtime that the input tensor is a constant.
|
class |
HashTable
Creates a non-initialized hash table.
|
class |
HistogramFixedWidth<U extends Number>
Return histogram of values.
|
class |
Identity<T>
Return a tensor with the same shape and contents as the input tensor or value.
|
class |
IdentityN
Returns a list of tensors with the same shapes and contents as the input
|
class |
IgnoreErrorsDataset
Creates a dataset that contains the elements of `input_dataset` ignoring errors.
|
class |
ImmutableConst<T>
Returns immutable tensor from memory region.
|
class |
InfeedDequeue<T>
A placeholder op for a value that will be fed into the computation.
|
class |
InfeedDequeueTuple
Fetches multiple values from infeed as an XLA tuple.
|
class |
InfeedEnqueue
An op which feeds a single Tensor value into the computation.
|
class |
InfeedEnqueuePrelinearizedBuffer
An op which enqueues prelinearized buffer into TPU infeed.
|
class |
InfeedEnqueueTuple
Feeds multiple Tensor values into the computation as an XLA tuple.
|
class |
InitializeTable
Table initializer that takes two tensors for keys and values respectively.
|
class |
InitializeTableFromTextFile
Initializes a table from a text file.
|
class |
InplaceAdd<T>
Adds v into specified rows of x.
|
class |
InplaceSub<T>
Subtracts `v` into specified rows of `x`.
|
class |
InplaceUpdate<T>
Updates specified rows with values in `v`.
|
class |
IsBoostedTreesEnsembleInitialized
Checks whether a tree ensemble has been initialized.
|
class |
IsBoostedTreesQuantileStreamResourceInitialized
Checks whether a quantile stream has been initialized.
|
class |
IsVariableInitialized
Checks whether a tensor has been initialized.
|
class |
IteratorGetDevice
Returns the name of the device on which `resource` has been placed.
|
class |
KMC2ChainInitialization
Returns the index of a data point that should be added to the seed set.
|
class |
KmeansPlusPlusInitialization
Selects num_to_sample rows of input using the KMeans++ criterion.
|
class |
LinSpace<T extends Number>
Generates values in an interval.
|
class |
LMDBDataset |
class |
LoadTPUEmbeddingAdadeltaParameters
Load Adadelta embedding parameters.
|
class |
LoadTPUEmbeddingAdadeltaParametersGradAccumDebug
Load Adadelta parameters with debug support.
|
class |
LoadTPUEmbeddingAdagradParameters
Load Adagrad embedding parameters.
|
class |
LoadTPUEmbeddingAdagradParametersGradAccumDebug
Load Adagrad embedding parameters with debug support.
|
class |
LoadTPUEmbeddingADAMParameters
Load ADAM embedding parameters.
|
class |
LoadTPUEmbeddingADAMParametersGradAccumDebug
Load ADAM embedding parameters with debug support.
|
class |
LoadTPUEmbeddingCenteredRMSPropParameters
Load centered RMSProp embedding parameters.
|
class |
LoadTPUEmbeddingFTRLParameters
Load FTRL embedding parameters.
|
class |
LoadTPUEmbeddingFTRLParametersGradAccumDebug
Load FTRL embedding parameters with debug support.
|
class |
LoadTPUEmbeddingMDLAdagradLightParameters
Load MDL Adagrad Light embedding parameters.
|
class |
LoadTPUEmbeddingMomentumParameters
Load Momentum embedding parameters.
|
class |
LoadTPUEmbeddingMomentumParametersGradAccumDebug
Load Momentum embedding parameters with debug support.
|
class |
LoadTPUEmbeddingProximalAdagradParameters
Load proximal Adagrad embedding parameters.
|
class |
LoadTPUEmbeddingProximalAdagradParametersGradAccumDebug
Load proximal Adagrad embedding parameters with debug support.
|
class |
LoadTPUEmbeddingRMSPropParameters
Load RMSProp embedding parameters.
|
class |
LoadTPUEmbeddingRMSPropParametersGradAccumDebug
Load RMSProp embedding parameters with debug support.
|
class |
LoadTPUEmbeddingStochasticGradientDescentParameters
Load SGD embedding parameters.
|
class |
LookupTableExport<T,U>
Outputs all keys and values in the table.
|
class |
LookupTableFind<U>
Looks up keys in a table, outputs the corresponding values.
|
class |
LookupTableImport
Replaces the contents of the table with the specified keys and values.
|
class |
LookupTableInsert
Updates the table to associates keys with values.
|
class |
LookupTableRemove
Removes keys and its associated values from a table.
|
class |
LookupTableSize
Computes the number of elements in the given table.
|
class |
LoopCond
Forwards the input to the output.
|
class |
LowerBound<U extends Number>
Applies lower_bound(sorted_search_values, values) along each row.
|
class |
LSTMBlockCell<T extends Number>
Computes the LSTM cell forward propagation for 1 time step.
|
class |
LSTMBlockCellGrad<T extends Number>
Computes the LSTM cell backward propagation for 1 timestep.
|
class |
Lu<T,U extends Number>
Computes the LU decomposition of one or more square matrices.
|
class |
MapClear
Op removes all elements in the underlying container.
|
class |
MapIncompleteSize
Op returns the number of incomplete elements in the underlying container.
|
class |
MapPeek
Op peeks at the values at the specified key.
|
class |
MapSize
Op returns the number of elements in the underlying container.
|
class |
MapStage
Stage (key, values) in the underlying container which behaves like a hashtable.
|
class |
MapUnstage
Op removes and returns the values associated with the key
|
class |
MapUnstageNoKey
Op removes and returns a random (key, value)
|
class |
MatrixDiagPartV2<T>
Returns the batched diagonal part of a batched tensor.
|
class |
MatrixDiagV2<T>
Returns a batched diagonal tensor with given batched diagonal values.
|
class |
MatrixSetDiagV2<T>
Returns a batched matrix tensor with new batched diagonal values.
|
class |
Max<T>
Computes the maximum of elements across dimensions of a tensor.
|
class |
MaxIntraOpParallelismDataset
Creates a dataset that overrides the maximum intra-op parallelism.
|
class |
Merge<T>
Forwards the value of an available tensor from `inputs` to `output`.
|
class |
Min<T>
Computes the minimum of elements across dimensions of a tensor.
|
class |
MirrorPad<T>
Pads a tensor with mirrored values.
|
class |
MirrorPadGrad<T>
Gradient op for `MirrorPad` op.
|
class |
MulNoNan<T>
Returns x * y element-wise.
|
class |
MutableDenseHashTable
Creates an empty hash table that uses tensors as the backing store.
|
class |
MutableHashTable
Creates an empty hash table.
|
class |
MutableHashTableOfTensors
Creates an empty hash table.
|
class |
Mutex
Creates a Mutex resource that can be locked by `MutexLock`.
|
class |
MutexLock
Locks a mutex resource.
|
class |
NcclAllReduce<T extends Number>
Outputs a tensor containing the reduction across all input tensors.
|
class |
NcclBroadcast<T extends Number>
Sends `input` to all devices that are connected to the output.
|
class |
NcclReduce<T extends Number>
Reduces `input` from `num_devices` using `reduction` to a single device.
|
class |
NearestNeighbors
Selects the k nearest centers for each point.
|
class |
NextAfter<T extends Number>
Returns the next representable value of `x1` in the direction of `x2`, element-wise.
|
class |
NextIteration<T>
Makes its input available to the next iteration.
|
class |
NonDeterministicInts<U>
Non-deterministically generates some integers.
|
class |
NonMaxSuppressionV5<T extends Number>
Greedily selects a subset of bounding boxes in descending order of score,
|
class |
NonSerializableDataset |
class |
NoOp
Does nothing.
|
class |
OneHot<U>
Returns a one-hot tensor.
|
class |
OnesLike<T>
Returns a tensor of ones with the same shape and type as x.
|
class |
OrderedMapClear
Op removes all elements in the underlying container.
|
class |
OrderedMapIncompleteSize
Op returns the number of incomplete elements in the underlying container.
|
class |
OrderedMapPeek
Op peeks at the values at the specified key.
|
class |
OrderedMapSize
Op returns the number of elements in the underlying container.
|
class |
OrderedMapStage
Stage (key, values) in the underlying container which behaves like a ordered
|
class |
OrderedMapUnstage
Op removes and returns the values associated with the key
|
class |
OrderedMapUnstageNoKey
Op removes and returns the (key, value) element with the smallest
|
class |
OutfeedDequeue<T>
Retrieves a single tensor from the computation outfeed.
|
class |
OutfeedDequeueTuple
Retrieve multiple values from the computation outfeed.
|
class |
OutfeedEnqueue
Enqueue a Tensor on the computation outfeed.
|
class |
OutfeedEnqueueTuple
Enqueue multiple Tensor values on the computation outfeed.
|
class |
Pad<T>
Pads a tensor.
|
class |
ParallelConcat<T>
Concatenates a list of `N` tensors along the first dimension.
|
class |
ParallelDynamicStitch<T>
Interleave the values from the `data` tensors into a single tensor.
|
class |
Placeholder<T>
A placeholder op for a value that will be fed into the computation.
|
class |
PlaceholderWithDefault<T>
A placeholder op that passes through `input` when its output is not fed.
|
class |
Prelinearize
An op which linearizes one Tensor value to an opaque variant tensor.
|
class |
PrelinearizeTuple
An op which linearizes multiple Tensor values to an opaque variant tensor.
|
class |
Print
Prints a string scalar.
|
class |
PrivateThreadPoolDataset
Creates a dataset that uses a custom thread pool to compute `input_dataset`.
|
class |
Prod<T>
Computes the product of elements across dimensions of a tensor.
|
class |
QuantizedConcat<T>
Concatenates quantized tensors along one dimension.
|
class |
QuantizedConcatV2<T> |
class |
QuantizedConv2DAndRelu<V> |
class |
QuantizedConv2DAndReluAndRequantize<V> |
class |
QuantizedConv2DAndRequantize<V> |
class |
QuantizedConv2DPerChannel<V>
Computes QuantizedConv2D per channel.
|
class |
QuantizedConv2DWithBias<V> |
class |
QuantizedConv2DWithBiasAndRelu<V> |
class |
QuantizedConv2DWithBiasAndReluAndRequantize<W> |
class |
QuantizedConv2DWithBiasAndRequantize<W> |
class |
QuantizedConv2DWithBiasSignedSumAndReluAndRequantize<X> |
class |
QuantizedConv2DWithBiasSumAndRelu<V> |
class |
QuantizedConv2DWithBiasSumAndReluAndRequantize<X> |
class |
QuantizedDepthwiseConv2D<V>
Computes quantized depthwise Conv2D.
|
class |
QuantizedDepthwiseConv2DWithBias<V>
Computes quantized depthwise Conv2D with Bias.
|
class |
QuantizedDepthwiseConv2DWithBiasAndRelu<V>
Computes quantized depthwise Conv2D with Bias and Relu.
|
class |
QuantizedDepthwiseConv2DWithBiasAndReluAndRequantize<W>
Computes quantized depthwise Conv2D with Bias, Relu and Requantize.
|
class |
QuantizedMatMulWithBias<W>
Performs a quantized matrix multiplication of `a` by the matrix `b` with bias
add.
|
class |
QuantizedMatMulWithBiasAndRelu<V>
Perform a quantized matrix multiplication of `a` by the matrix `b` with bias
add and relu fusion.
|
class |
QuantizedMatMulWithBiasAndReluAndRequantize<W>
Perform a quantized matrix multiplication of `a` by the matrix `b` with bias
add and relu and requantize fusion.
|
class |
QuantizedReshape<T>
Reshapes a quantized tensor as per the Reshape op.
|
class |
RaggedGather<T extends Number,U>
Gather ragged slices from `params` axis `0` according to `indices`.
|
class |
RaggedRange<U extends Number,T extends Number>
Returns a `RaggedTensor` containing the specified sequences of numbers.
|
class |
RaggedTensorFromVariant<U extends Number,T>
Decodes a `variant` Tensor into a `RaggedTensor`.
|
class |
RaggedTensorToSparse<U>
Converts a `RaggedTensor` into a `SparseTensor` with the same values.
|
class |
RaggedTensorToTensor<U>
Create a dense tensor from a ragged tensor, possibly altering its shape.
|
class |
RaggedTensorToVariant
Encodes a `RaggedTensor` into a `variant` Tensor.
|
class |
Range<T extends Number>
Creates a sequence of numbers.
|
class |
Rank
Returns the rank of a tensor.
|
class |
ReadVariableOp<T>
Reads the value of a variable.
|
class |
RebatchDataset
Creates a dataset that changes the batch size.
|
class |
RecvTPUEmbeddingActivations
An op that receives embedding activations on the TPU.
|
class |
ReduceAll
Computes the "logical and" of elements across dimensions of a tensor.
|
class |
ReduceAny
Computes the "logical or" of elements across dimensions of a tensor.
|
class |
ReduceMax<T>
Computes the maximum of elements across dimensions of a tensor.
|
class |
ReduceMin<T>
Computes the minimum of elements across dimensions of a tensor.
|
class |
ReduceProd<T>
Computes the product of elements across dimensions of a tensor.
|
class |
ReduceSum<T>
Computes the sum of elements across dimensions of a tensor.
|
class |
RefEnter<T>
Creates or finds a child frame, and makes `data` available to the child frame.
|
class |
RefExit<T>
Exits the current frame to its parent frame.
|
class |
RefIdentity<T>
Return the same ref tensor as the input ref tensor.
|
class |
RefMerge<T>
Forwards the value of an available tensor from `inputs` to `output`.
|
class |
RefNextIteration<T>
Makes its input available to the next iteration.
|
class |
RefSelect<T>
Forwards the `index`th element of `inputs` to `output`.
|
class |
RefSwitch<T>
Forwards the ref tensor `data` to the output port determined by `pred`.
|
class |
RemoteFusedGraphExecute
Execute a sub graph on a remote processor.
|
class |
RequantizationRangePerChannel
Computes requantization range per channel.
|
class |
RequantizePerChannel<U>
Requantizes input with min and max values known per channel.
|
class |
Reshape<T>
Reshapes a tensor.
|
class |
ResourceAccumulatorApplyGradient
Applies a gradient to a given accumulator.
|
class |
ResourceAccumulatorNumAccumulated
Returns the number of gradients aggregated in the given accumulators.
|
class |
ResourceAccumulatorSetGlobalStep
Updates the accumulator with a new value for global_step.
|
class |
ResourceAccumulatorTakeGradient<T>
Extracts the average gradient in the given ConditionalAccumulator.
|
class |
ResourceApplyAdagradV2
Update '*var' according to the adagrad scheme.
|
class |
ResourceApplyAdamWithAmsgrad
Update '*var' according to the Adam algorithm.
|
class |
ResourceApplyKerasMomentum
Update '*var' according to the momentum scheme.
|
class |
ResourceConditionalAccumulator
A conditional accumulator for aggregating gradients.
|
class |
ResourceCountUpTo<T extends Number>
Increments variable pointed to by 'resource' until it reaches 'limit'.
|
class |
ResourceGather<U>
Gather slices from the variable pointed to by `resource` according to `indices`.
|
class |
ResourceGatherNd<U> |
class |
ResourceScatterAdd
Adds sparse updates to the variable referenced by `resource`.
|
class |
ResourceScatterDiv
Divides sparse updates into the variable referenced by `resource`.
|
class |
ResourceScatterMax
Reduces sparse updates into the variable referenced by `resource` using the `max` operation.
|
class |
ResourceScatterMin
Reduces sparse updates into the variable referenced by `resource` using the `min` operation.
|
class |
ResourceScatterMul
Multiplies sparse updates into the variable referenced by `resource`.
|
class |
ResourceScatterNdAdd
Applies sparse addition to individual values or slices in a Variable.
|
class |
ResourceScatterNdSub
Applies sparse subtraction to individual values or slices in a Variable.
|
class |
ResourceScatterNdUpdate
Applies sparse `updates` to individual values or slices within a given
|
class |
ResourceScatterSub
Subtracts sparse updates from the variable referenced by `resource`.
|
class |
ResourceScatterUpdate
Assigns sparse updates to the variable referenced by `resource`.
|
class |
ResourceSparseApplyAdagradV2
Update relevant entries in '*var' and '*accum' according to the adagrad scheme.
|
class |
ResourceSparseApplyKerasMomentum
Update relevant entries in '*var' and '*accum' according to the momentum scheme.
|
class |
ResourceStridedSliceAssign
Assign `value` to the sliced l-value reference of `ref`.
|
class |
RetrieveTPUEmbeddingAdadeltaParameters
Retrieve Adadelta embedding parameters.
|
class |
RetrieveTPUEmbeddingAdadeltaParametersGradAccumDebug
Retrieve Adadelta embedding parameters with debug support.
|
class |
RetrieveTPUEmbeddingAdagradParameters
Retrieve Adagrad embedding parameters.
|
class |
RetrieveTPUEmbeddingAdagradParametersGradAccumDebug
Retrieve Adagrad embedding parameters with debug support.
|
class |
RetrieveTPUEmbeddingADAMParameters
Retrieve ADAM embedding parameters.
|
class |
RetrieveTPUEmbeddingADAMParametersGradAccumDebug
Retrieve ADAM embedding parameters with debug support.
|
class |
RetrieveTPUEmbeddingCenteredRMSPropParameters
Retrieve centered RMSProp embedding parameters.
|
class |
RetrieveTPUEmbeddingFTRLParameters
Retrieve FTRL embedding parameters.
|
class |
RetrieveTPUEmbeddingFTRLParametersGradAccumDebug
Retrieve FTRL embedding parameters with debug support.
|
class |
RetrieveTPUEmbeddingMDLAdagradLightParameters
Retrieve MDL Adagrad Light embedding parameters.
|
class |
RetrieveTPUEmbeddingMomentumParameters
Retrieve Momentum embedding parameters.
|
class |
RetrieveTPUEmbeddingMomentumParametersGradAccumDebug
Retrieve Momentum embedding parameters with debug support.
|
class |
RetrieveTPUEmbeddingProximalAdagradParameters
Retrieve proximal Adagrad embedding parameters.
|
class |
RetrieveTPUEmbeddingProximalAdagradParametersGradAccumDebug
Retrieve proximal Adagrad embedding parameters with debug support.
|
class |
RetrieveTPUEmbeddingRMSPropParameters
Retrieve RMSProp embedding parameters.
|
class |
RetrieveTPUEmbeddingRMSPropParametersGradAccumDebug
Retrieve RMSProp embedding parameters with debug support.
|
class |
RetrieveTPUEmbeddingStochasticGradientDescentParameters
Retrieve SGD embedding parameters.
|
class |
Reverse<T>
Reverses specific dimensions of a tensor.
|
class |
ReverseSequence<T>
Reverses variable length slices.
|
class |
RngSkip
Advance the counter of a counter-based RNG.
|
class |
Roll<T>
Rolls the elements of a tensor along an axis.
|
class |
Rpc
Perform batches of RPC requests.
|
class |
SamplingDataset
Creates a dataset that takes a Bernoulli sample of the contents of another dataset.
|
class |
ScaleAndTranslate |
class |
ScaleAndTranslateGrad<T extends Number> |
class |
ScatterAdd<T>
Adds sparse updates to a variable reference.
|
class |
ScatterDiv<T>
Divides a variable reference by sparse updates.
|
class |
ScatterMax<T extends Number>
Reduces sparse updates into a variable reference using the `max` operation.
|
class |
ScatterMin<T extends Number>
Reduces sparse updates into a variable reference using the `min` operation.
|
class |
ScatterMul<T>
Multiplies sparse updates into a variable reference.
|
class |
ScatterNd<U>
Scatter `updates` into a new tensor according to `indices`.
|
class |
ScatterNdAdd<T>
Applies sparse addition to individual values or slices in a Variable.
|
class |
ScatterNdNonAliasingAdd<T>
Applies sparse addition to `input` using individual values or slices
|
class |
ScatterNdSub<T>
Applies sparse subtraction to individual values or slices in a Variable.
|
class |
ScatterNdUpdate<T>
Applies sparse `updates` to individual values or slices within a given
|
class |
ScatterSub<T>
Subtracts sparse updates to a variable reference.
|
class |
ScatterUpdate<T>
Applies sparse updates to a variable reference.
|
class |
SelectV2<T> |
class |
SendTPUEmbeddingGradients
Performs gradient updates of embedding tables.
|
class |
SetDiff1d<T,U extends Number>
Computes the difference between two lists of numbers or strings.
|
class |
SetSize
Number of unique elements along last dimension of input `set`.
|
class |
Shape<U extends Number>
Returns the shape of a tensor.
|
class |
ShapeN<U extends Number>
Returns shape of tensors.
|
class |
ShardDataset
Creates a `Dataset` that includes only 1/`num_shards` of this dataset.
|
class |
ShuffleDatasetV2 |
class |
ShutdownDistributedTPU
Shuts down a running distributed TPU system.
|
class |
Size<U extends Number>
Returns the size of a tensor.
|
class |
Skipgram
Parses a text file and creates a batch of examples.
|
class |
SleepDataset |
class |
Slice<T>
Return a slice from 'input'.
|
class |
SlidingWindowDataset
Creates a dataset that passes a sliding window over `input_dataset`.
|
class |
Snapshot<T>
Returns a copy of the input tensor.
|
class |
SnapshotDataset
Creates a dataset that will write to / read from a snapshot.
|
class |
SpaceToBatchNd<T>
SpaceToBatch for N-D tensors of type T.
|
class |
SparseApplyAdagradV2<T>
Update relevant entries in '*var' and '*accum' according to the adagrad scheme.
|
class |
Split<T>
Splits a tensor into `num_split` tensors along one dimension.
|
class |
SplitV<T>
Splits a tensor into `num_split` tensors along one dimension.
|
class |
Squeeze<T>
Removes dimensions of size 1 from the shape of a tensor.
|
class |
Stack<T>
Packs a list of `N` rank-`R` tensors into one rank-`(R+1)` tensor.
|
class |
Stage
Stage values similar to a lightweight Enqueue.
|
class |
StageClear
Op removes all elements in the underlying container.
|
class |
StagePeek
Op peeks at the values at the specified index.
|
class |
StageSize
Op returns the number of elements in the underlying container.
|
class |
StatefulRandomBinomial<V extends Number> |
class |
StatefulStandardNormal<U>
Outputs random values from a normal distribution.
|
class |
StatefulStandardNormalV2<U>
Outputs random values from a normal distribution.
|
class |
StatefulTruncatedNormal<U>
Outputs random values from a truncated normal distribution.
|
class |
StatefulUniform<U>
Outputs random values from a uniform distribution.
|
class |
StatefulUniformFullInt<U>
Outputs random integers from a uniform distribution.
|
class |
StatefulUniformInt<U>
Outputs random integers from a uniform distribution.
|
class |
StatsAggregatorHandleV2 |
class |
StatsAggregatorSetSummaryWriter
Set a summary_writer_interface to record statistics using given stats_aggregator.
|
class |
StopGradient<T>
Stops gradient computation.
|
class |
StridedSlice<T>
Return a strided slice from `input`.
|
class |
StridedSliceAssign<T>
Assign `value` to the sliced l-value reference of `ref`.
|
class |
StridedSliceGrad<U>
Returns the gradient of `StridedSlice`.
|
class |
StringLower |
class |
StringNGrams<T extends Number>
Creates ngrams from ragged string data.
|
class |
StringUpper |
class |
Sum<T>
Computes the sum of elements across dimensions of a tensor.
|
class |
SwitchCond<T>
Forwards `data` to the output port determined by `pred`.
|
class |
TemporaryVariable<T>
Returns a tensor that may be mutated, but only persists within a single step.
|
class |
TensorArray
An array of Tensors of given size.
|
class |
TensorArrayClose
Delete the TensorArray from its resource container.
|
class |
TensorArrayConcat<T>
Concat the elements from the TensorArray into value `value`.
|
class |
TensorArrayGather<T>
Gather specific elements from the TensorArray into output `value`.
|
class |
TensorArrayGrad
Creates a TensorArray for storing the gradients of values in the given handle.
|
class |
TensorArrayGradWithShape
Creates a TensorArray for storing multiple gradients of values in the given handle.
|
class |
TensorArrayPack<T> |
class |
TensorArrayRead<T>
Read an element from the TensorArray into output `value`.
|
class |
TensorArrayScatter
Scatter the data from the input value into specific TensorArray elements.
|
class |
TensorArraySize
Get the current size of the TensorArray.
|
class |
TensorArraySplit
Split the data from the input value into TensorArray elements.
|
class |
TensorArrayUnpack |
class |
TensorArrayWrite
Push an element onto the tensor_array.
|
class |
TensorForestCreateTreeVariable
Creates a tree resource and returns a handle to it.
|
class |
TensorForestTreeDeserialize
Deserializes a proto into the tree handle
|
class |
TensorForestTreeIsInitializedOp
Checks whether a tree has been initialized.
|
class |
TensorForestTreePredict
Output the logits for the given input data
|
class |
TensorForestTreeResourceHandleOp
Creates a handle to a TensorForestTreeResource
|
class |
TensorForestTreeSerialize
Serializes the tree handle to a proto
|
class |
TensorForestTreeSize
Get the number of nodes in a tree
|
class |
TensorListConcat<T>
Concats all tensors in the list along the 0th dimension.
|
class |
TensorListConcatLists |
class |
TensorListConcatV2<U>
Concats all tensors in the list along the 0th dimension.
|
class |
TensorListElementShape<T extends Number>
The shape of the elements of the given list, as a tensor.
|
class |
TensorListFromTensor
Creates a TensorList which, when stacked, has the value of `tensor`.
|
class |
TensorListGather<T>
Creates a Tensor by indexing into the TensorList.
|
class |
TensorListGetItem<T> |
class |
TensorListLength
Returns the number of tensors in the input tensor list.
|
class |
TensorListPopBack<T>
Returns the last element of the input list as well as a list with all but that element.
|
class |
TensorListPushBack
Returns a list which has the passed-in `Tensor` as last element and the other elements of the given list in `input_handle`.
|
class |
TensorListPushBackBatch |
class |
TensorListReserve
List of the given size with empty elements.
|
class |
TensorListResize
Resizes the list.
|
class |
TensorListScatter
Creates a TensorList by indexing into a Tensor.
|
class |
TensorListScatterIntoExistingList
Scatters tensor at indices in an input list.
|
class |
TensorListScatterV2
Creates a TensorList by indexing into a Tensor.
|
class |
TensorListSetItem |
class |
TensorListSplit
Splits a tensor into a list.
|
class |
TensorListStack<T>
Stacks all tensors in the list.
|
class |
TensorScatterAdd<T>
Adds sparse `updates` to an existing tensor according to `indices`.
|
class |
TensorScatterSub<T>
Subtracts sparse `updates` from an existing tensor according to `indices`.
|
class |
TensorScatterUpdate<T>
Scatter `updates` into an existing tensor according to `indices`.
|
class |
TensorStridedSliceUpdate<T>
Assign `value` to the sliced l-value reference of `input`.
|
class |
ThreadPoolDataset
Creates a dataset that uses a custom thread pool to compute `input_dataset`.
|
class |
ThreadPoolHandle
Creates a dataset that uses a custom thread pool to compute `input_dataset`.
|
class |
Tile<T>
Constructs a tensor by tiling a given tensor.
|
class |
Timestamp
Provides the time since epoch in seconds.
|
class |
TPUCompilationResult
CompilationResultProto indicating the status of the TPU compilation.
|
class |
TPUEmbeddingActivations
An op enabling differentiation of TPU Embeddings.
|
class |
TPUOrdinalSelector
A TPU core selector Op.
|
class |
TPUReplicatedInput<T>
Connects N inputs to an N-way replicated TPU computation.
|
class |
TPUReplicatedOutput<T>
Connects outputs of an N-way replicated computation to N outputs.
|
class |
TPUReplicateMetadata
Metadata indicaitng how the TPU computation should be replicated.
|
class |
TridiagonalMatMul<T>
Calculate product with tridiagonal matrix.
|
class |
TridiagonalSolve<T>
Solves tridiagonal systems of equations.
|
class |
TryRpc
Perform batches of RPC requests.
|
class |
Unbatch<T>
Reverses the operation of Batch for a single output Tensor.
|
class |
UnbatchGrad<T>
Gradient of Unbatch.
|
class |
UnicodeDecode<T extends Number>
Decodes each string in `input` into a sequence of Unicode code points.
|
class |
UnicodeEncode
Encode a tensor of ints into unicode strings.
|
class |
Unique<T,V extends Number>
Finds unique elements along an axis of a tensor.
|
class |
UniqueDataset
Creates a dataset that contains the unique elements of `input_dataset`.
|
class |
UniqueWithCounts<T,V extends Number>
Finds unique elements along an axis of a tensor.
|
class |
UnravelIndex<T extends Number>
Converts an array of flat indices into a tuple of coordinate arrays.
|
class |
UnsortedSegmentJoin
Joins the elements of `inputs` based on `segment_ids`.
|
class |
Unstack<T>
Unpacks a given dimension of a rank-`R` tensor into `num` rank-`(R-1)` tensors.
|
class |
Unstage
Op is similar to a lightweight Dequeue.
|
class |
UnwrapDatasetVariant |
class |
UpperBound<U extends Number>
Applies upper_bound(sorted_search_values, values) along each row.
|
class |
VarHandleOp
Creates a handle to a Variable resource.
|
class |
Variable<T>
Holds state in the form of a tensor that persists across steps.
|
class |
VariableShape<T extends Number>
Returns the shape of the variable pointed to by `resource`.
|
class |
VarIsInitializedOp
Checks whether a resource handle-based variable has been initialized.
|
class |
Where
Returns locations of nonzero / true values in a tensor.
|
class |
Where3<T>
Selects elements from `x` or `y`, depending on `condition`.
|
class |
WorkerHeartbeat
Worker heartbeat op.
|
class |
WrapDatasetVariant |
class |
WriteRawProtoSummary |
class |
ZerosLike<T>
Returns a tensor of zeros with the same shape and type as x.
|
Modifier and Type | Class and Description |
---|---|
class |
AnonymousIterator
A container for an iterator resource.
|
class |
BatchDataset
Creates a dataset that batches `batch_size` elements from `input_dataset`.
|
class |
BytesProducedStatsDataset
Records the bytes size of each element of `input_dataset` in a StatsAggregator.
|
class |
CacheDataset
Creates a dataset that caches elements from `input_dataset`.
|
class |
ConcatenateDataset
Creates a dataset that concatenates `input_dataset` with `another_dataset`.
|
class |
DatasetToGraph
Returns a serialized GraphDef representing `input_dataset`.
|
class |
DatasetToSingleElement
Outputs the single element from the given dataset.
|
class |
DatasetToTfRecord
Writes the given dataset to the given file using the TFRecord format.
|
class |
DenseToSparseBatchDataset
Creates a dataset that batches input elements into a SparseTensor.
|
class |
DeserializeIterator
Converts the given variant tensor to an iterator and stores it in the given resource.
|
class |
ExperimentalAssertNextDataset |
class |
ExperimentalCsvDataset |
class |
ExperimentalDirectedInterleaveDataset
A substitute for `InterleaveDataset` on a fixed list of `N` datasets.
|
class |
ExperimentalIgnoreErrorsDataset
Creates a dataset that contains the elements of `input_dataset` ignoring errors.
|
class |
ExperimentalIteratorGetDevice
Returns the name of the device on which `resource` has been placed.
|
class |
ExperimentalLmdbDataset |
class |
ExperimentalNonSerializableDataset |
class |
ExperimentalSleepDataset |
class |
ExperimentalThreadPoolDataset
Creates a dataset that uses a custom thread pool to compute `input_dataset`.
|
class |
ExperimentalThreadPoolHandle
Creates a dataset that uses a custom thread pool to compute `input_dataset`.
|
class |
ExperimentalUniqueDataset
Creates a dataset that contains the unique elements of `input_dataset`.
|
class |
FilterByLastComponentDataset
Creates a dataset containing elements of first component of `input_dataset` having true in the last component.
|
class |
FixedLengthRecordDataset |
class |
Iterator |
class |
IteratorFromStringHandle |
class |
IteratorGetNext
Gets the next output from the given iterator .
|
class |
IteratorGetNextAsOptional
Gets the next output from the given iterator as an Optional variant.
|
class |
IteratorGetNextSync
Gets the next output from the given iterator.
|
class |
IteratorToStringHandle
Converts the given `resource_handle` representing an iterator to a string.
|
class |
LatencyStatsDataset
Records the latency of producing `input_dataset` elements in a StatsAggregator.
|
class |
LeakyReluGrad<T extends Number>
Computes rectified linear gradients for a LeakyRelu operation.
|
class |
MakeIterator
Makes a new iterator from the given `dataset` and stores it in `iterator`.
|
class |
MatchingFilesDataset |
class |
ModelDataset
Identity transformation that models performance.
|
class |
MultiDeviceIterator
Creates a MultiDeviceIterator resource.
|
class |
MultiDeviceIteratorFromStringHandle
Generates a MultiDeviceIterator resource from its provided string handle.
|
class |
MultiDeviceIteratorGetNextFromShard
Gets next element for the provided shard number.
|
class |
MultiDeviceIteratorInit
Initializes the multi device iterator with the given dataset.
|
class |
MultiDeviceIteratorToStringHandle
Produces a string handle for the given MultiDeviceIterator.
|
class |
OptimizeDataset
Creates a dataset by applying optimizations to `input_dataset`.
|
class |
OptionalFromValue
Constructs an Optional variant from a tuple of tensors.
|
class |
OptionalGetValue
Returns the value stored in an Optional variant or raises an error if none exists.
|
class |
OptionalHasValue
Returns true if and only if the given Optional variant has a value.
|
class |
OptionalNone
Creates an Optional variant with no value.
|
class |
PaddedBatchDataset
Creates a dataset that batches and pads `batch_size` elements from the input.
|
class |
ParseExampleDataset
Transforms `input_dataset` containing `Example` protos as vectors of DT_STRING into a dataset of `Tensor` or `SparseTensor` objects representing the parsed features.
|
class |
PrefetchDataset
Creates a dataset that asynchronously prefetches elements from `input_dataset`.
|
class |
RandomDataset
Creates a Dataset that returns pseudorandom numbers.
|
class |
RangeDataset
Creates a dataset with a range of values.
|
class |
RepeatDataset
Creates a dataset that emits the outputs of `input_dataset` `count` times.
|
class |
SerializeIterator
Converts the given `resource_handle` representing an iterator to a variant tensor.
|
class |
SetStatsAggregatorDataset |
class |
ShuffleAndRepeatDataset
Creates a dataset that shuffles and repeats elements from `input_dataset`
|
class |
ShuffleDataset
Creates a dataset that shuffles elements from `input_dataset` pseudorandomly.
|
class |
SkipDataset
Creates a dataset that skips `count` elements from the `input_dataset`.
|
class |
SparseTensorSliceDataset
Creates a dataset that splits a SparseTensor into elements row-wise.
|
class |
SqlDataset
Creates a dataset that executes a SQL query and emits rows of the result set.
|
class |
StatsAggregatorHandle
Creates a statistics manager resource.
|
class |
TakeDataset
Creates a dataset that contains `count` elements from the `input_dataset`.
|
class |
TensorDataset
Creates a dataset that emits `components` as a tuple of tensors once.
|
class |
TensorSliceDataset
Creates a dataset that emits each dim-0 slice of `components` once.
|
class |
TextLineDataset
Creates a dataset that emits the lines of one or more text files.
|
class |
TfRecordDataset
Creates a dataset that emits the records from one or more TFRecord files.
|
class |
UnbatchDataset
A dataset that splits the elements of its input into multiple elements.
|
class |
WindowDataset
A dataset that creates window datasets from the input dataset.
|
class |
ZipDataset
Creates a dataset that zips together `input_datasets`.
|
Modifier and Type | Class and Description |
---|---|
class |
AsString
Converts each entry in the given tensor to strings.
|
class |
Cast<U>
Cast x of type SrcT to y of DstT.
|
class |
Complex<U>
Converts two real numbers to a complex number.
|
Modifier and Type | Class and Description |
---|---|
class |
AdjustContrast<T extends Number>
Adjust the contrast of one or more images.
|
class |
AdjustHue<T extends Number>
Adjust the hue of one or more images.
|
class |
AdjustSaturation<T extends Number>
Adjust the saturation of one or more images.
|
class |
CropAndResize
Extracts crops from the input image tensor and resizes them.
|
class |
CropAndResizeGradBoxes
Computes the gradient of the crop_and_resize op wrt the input boxes tensor.
|
class |
CropAndResizeGradImage<T extends Number>
Computes the gradient of the crop_and_resize op wrt the input image tensor.
|
class |
DecodeAndCropJpeg
Decode and Crop a JPEG-encoded image to a uint8 tensor.
|
class |
DecodeBmp
Decode the first frame of a BMP-encoded image to a uint8 tensor.
|
class |
DecodeGif
Decode the frame(s) of a GIF-encoded image to a uint8 tensor.
|
class |
DecodeJpeg
Decode a JPEG-encoded image to a uint8 tensor.
|
class |
DecodePng<T extends Number>
Decode a PNG-encoded image to a uint8 or uint16 tensor.
|
class |
DrawBoundingBoxes<T extends Number>
Draw bounding boxes on a batch of images.
|
class |
EncodeJpeg
JPEG-encode an image.
|
class |
EncodeJpegVariableQuality
JPEG encode input image with provided compression quality.
|
class |
EncodePng
PNG-encode an image.
|
class |
ExtractGlimpse
Extracts a glimpse from the input tensor.
|
class |
ExtractImagePatches<T extends Number>
Extract `patches` from `images` and put them in the "depth" output dimension.
|
class |
ExtractJpegShape<T extends Number>
Extract the shape information of a JPEG-encoded image.
|
class |
HsvToRgb<T extends Number>
Convert one or more images from HSV to RGB.
|
class |
NonMaxSuppression
Greedily selects a subset of bounding boxes in descending order of score,
|
class |
NonMaxSuppressionWithOverlaps
Greedily selects a subset of bounding boxes in descending order of score,
|
class |
QuantizedResizeBilinear<T>
Resize quantized `images` to `size` using quantized bilinear interpolation.
|
class |
RandomCrop<T extends Number>
Randomly crop `image`.
|
class |
ResizeArea
Resize `images` to `size` using area interpolation.
|
class |
ResizeBicubic
Resize `images` to `size` using bicubic interpolation.
|
class |
ResizeBicubicGrad<T extends Number>
Computes the gradient of bicubic interpolation.
|
class |
ResizeBilinear
Resize `images` to `size` using bilinear interpolation.
|
class |
ResizeBilinearGrad<T extends Number>
Computes the gradient of bilinear interpolation.
|
class |
ResizeNearestNeighbor<T extends Number>
Resize `images` to `size` using nearest neighbor interpolation.
|
class |
ResizeNearestNeighborGrad<T extends Number>
Computes the gradient of nearest neighbor interpolation.
|
class |
RgbToHsv<T extends Number>
Converts one or more images from RGB to HSV.
|
class |
SampleDistortedBoundingBox<T extends Number>
Generate a single randomly distorted bounding box for an image.
|
Modifier and Type | Class and Description |
---|---|
class |
DecodeBase64
Decode web-safe base64-encoded strings.
|
class |
DecodeCompressed
Decompress strings.
|
class |
DecodeCsv
Convert CSV records to tensors.
|
class |
DecodeJsonExample
Convert JSON-encoded Example records to binary protocol buffer strings.
|
class |
DecodeRaw<T>
Reinterpret the bytes of a string as a vector of numbers.
|
class |
DeserializeManySparse<T>
Deserialize and concatenate `SparseTensors` from a serialized minibatch.
|
class |
EncodeBase64
Encode strings into web-safe base64 format.
|
class |
FifoQueue
A queue that produces elements in first-in first-out order.
|
class |
FixedLengthRecordReader
A Reader that outputs fixed-length records from a file.
|
class |
IdentityReader
A Reader that outputs the queued work as both the key and value.
|
class |
LmdbReader
A Reader that outputs the records from a LMDB file.
|
class |
MatchingFiles
Returns the set of files matching one or more glob patterns.
|
class |
PaddingFifoQueue
A queue that produces elements in first-in first-out order.
|
class |
ParseExample
Transforms a vector of brain.Example protos (as strings) into typed tensors.
|
class |
ParseSequenceExample
Transforms a vector of brain.SequenceExample protos (as strings) into typed tensors.
|
class |
ParseSingleExample
Transforms a tf.Example proto (as a string) into typed tensors.
|
class |
ParseSingleSequenceExample
Transforms a scalar brain.SequenceExample proto (as strings) into typed tensors.
|
class |
ParseTensor<T>
Transforms a serialized tensorflow.TensorProto proto into a Tensor.
|
class |
PriorityQueue
A queue that produces elements sorted by the first component value.
|
class |
QueueClose
Closes the given queue.
|
class |
QueueDequeue
Dequeues a tuple of one or more tensors from the given queue.
|
class |
QueueDequeueMany
Dequeues `n` tuples of one or more tensors from the given queue.
|
class |
QueueDequeueUpTo
Dequeues `n` tuples of one or more tensors from the given queue.
|
class |
QueueEnqueue
Enqueues a tuple of one or more tensors in the given queue.
|
class |
QueueEnqueueMany
Enqueues zero or more tuples of one or more tensors in the given queue.
|
class |
QueueIsClosed
Returns true if queue is closed.
|
class |
QueueSize
Computes the number of elements in the given queue.
|
class |
RandomShuffleQueue
A queue that randomizes the order of elements.
|
class |
ReaderNumRecordsProduced
Returns the number of records this Reader has produced.
|
class |
ReaderNumWorkUnitsCompleted
Returns the number of work units this Reader has finished processing.
|
class |
ReaderRead
Returns the next record (key, value pair) produced by a Reader.
|
class |
ReaderReadUpTo
Returns up to `num_records` (key, value) pairs produced by a Reader.
|
class |
ReaderReset
Restore a Reader to its initial clean state.
|
class |
ReaderRestoreState
Restore a reader to a previously saved state.
|
class |
ReaderSerializeState
Produce a string tensor that encodes the state of a Reader.
|
class |
ReadFile
Reads and outputs the entire contents of the input filename.
|
class |
SerializeManySparse<U>
Serialize an `N`-minibatch `SparseTensor` into an `[N, 3]` `Tensor` object.
|
class |
SerializeSparse<U>
Serialize a `SparseTensor` into a `[3]` `Tensor` object.
|
class |
SerializeTensor
Transforms a Tensor into a serialized TensorProto proto.
|
class |
ShardedFilename
Generate a sharded filename.
|
class |
ShardedFilespec
Generate a glob pattern matching all sharded file names.
|
class |
TextLineReader
A Reader that outputs the lines of a file delimited by '\n'.
|
class |
TfRecordReader
A Reader that outputs the records from a TensorFlow Records file.
|
class |
WholeFileReader
A Reader that outputs the entire contents of a file as a value.
|
class |
WriteFile
Writes contents to the file at input filename.
|
Modifier and Type | Class and Description |
---|---|
class |
BandPart<T>
Copy a tensor setting everything outside a central band in each innermost matrix
|
class |
BatchCholesky<T extends Number> |
class |
BatchCholeskyGrad<T extends Number> |
class |
BatchMatMul<T>
Multiplies slices of two tensors in batches.
|
class |
BatchMatrixBandPart<T> |
class |
BatchMatrixDeterminant<T> |
class |
BatchMatrixDiag<T> |
class |
BatchMatrixDiagPart<T> |
class |
BatchMatrixInverse<T extends Number> |
class |
BatchMatrixSetDiag<T> |
class |
BatchMatrixSolve<T extends Number> |
class |
BatchMatrixSolveLs<T extends Number> |
class |
BatchMatrixTriangularSolve<T extends Number> |
class |
BatchSelfAdjointEig<T extends Number> |
class |
BatchSvd<T> |
class |
Cholesky<T>
Computes the Cholesky decomposition of one or more square matrices.
|
class |
CholeskyGrad<T extends Number>
Computes the reverse mode backpropagated gradient of the Cholesky algorithm.
|
class |
ConjugateTranspose<T>
Shuffle dimensions of x according to a permutation and conjugate the result.
|
class |
Cross<T extends Number>
Compute the pairwise cross product.
|
class |
Det<T>
Computes the determinant of one or more square matrices.
|
class |
Diag<T>
Returns a batched diagonal tensor with a given batched diagonal values.
|
class |
DiagPart<T>
Returns the batched diagonal part of a batched tensor.
|
class |
Inv<T>
Computes the inverse of one or more square invertible matrices or their
|
class |
LoadAndRemapMatrix
Loads a 2-D (matrix) `Tensor` with name `old_tensor_name` from the checkpoint
|
class |
LogMatrixDeterminant<T>
Computes the sign and the log of the absolute value of the determinant of
|
class |
MatMul<T>
Multiply the matrix "a" by the matrix "b".
|
class |
MatrixLogarithm<T>
Computes the matrix logarithm of one or more square matrices:
|
class |
MatrixSolveLs<T>
Solves one or more linear least-squares problems.
|
class |
Qr<T>
Computes the QR decompositions of one or more matrices.
|
class |
QuantizedMatMul<V>
Perform a quantized matrix multiplication of `a` by the matrix `b`.
|
class |
SelfAdjointEig<T>
Computes the eigen decomposition of one or more square self-adjoint matrices.
|
class |
SetDiag<T>
Returns a batched matrix tensor with new batched diagonal values.
|
class |
Solve<T>
Solves systems of linear equations.
|
class |
Sqrtm<T>
Computes the matrix square root of one or more square matrices:
|
class |
Svd<T>
Computes the singular value decompositions of one or more matrices.
|
class |
TensorDiag<T>
Returns a diagonal tensor with a given diagonal values.
|
class |
TensorDiagPart<T>
Returns the diagonal part of the tensor.
|
class |
Transpose<T>
Shuffle dimensions of x according to a permutation.
|
class |
TriangularSolve<T>
Solves systems of linear equations with upper or lower triangular matrices by backsubstitution.
|
Modifier and Type | Class and Description |
---|---|
class |
Abs<T extends Number>
Computes the absolute value of a tensor.
|
class |
AccumulateN<T>
Returns the element-wise sum of a list of tensors.
|
class |
Acos<T>
Computes acos of x element-wise.
|
class |
Acosh<T>
Computes inverse hyperbolic cosine of x element-wise.
|
class |
Add<T>
Returns x + y element-wise.
|
class |
AddN<T>
Add all input tensors element wise.
|
class |
Angle<U extends Number>
Returns the argument of a complex number.
|
class |
ApproximateEqual
Returns the truth value of abs(x-y) < tolerance element-wise.
|
class |
ArgMax<V extends Number>
Returns the index with the largest value across dimensions of a tensor.
|
class |
ArgMin<V extends Number>
Returns the index with the smallest value across dimensions of a tensor.
|
class |
Asin<T>
Computes the trignometric inverse sine of x element-wise.
|
class |
Asinh<T>
Computes inverse hyperbolic sine of x element-wise.
|
class |
Atan<T>
Computes the trignometric inverse tangent of x element-wise.
|
class |
Atan2<T extends Number>
Computes arctangent of `y/x` element-wise, respecting signs of the arguments.
|
class |
Atanh<T>
Computes inverse hyperbolic tangent of x element-wise.
|
class |
BesselI0e<T extends Number>
Computes the Bessel i0e function of `x` element-wise.
|
class |
BesselI1e<T extends Number>
Computes the Bessel i1e function of `x` element-wise.
|
class |
Betainc<T extends Number>
Compute the regularized incomplete beta integral \\(I_x(a, b)\\).
|
class |
Bincount<T extends Number>
Counts the number of occurrences of each value in an integer array.
|
class |
Ceil<T extends Number>
Returns element-wise smallest integer not less than x.
|
class |
CheckNumerics<T extends Number>
Checks a tensor for NaN and Inf values.
|
class |
CompareAndBitpack
Compare values of `input` to `threshold` and pack resulting bits into a `uint8`.
|
class |
ComplexAbs<U extends Number>
Computes the complex absolute value of a tensor.
|
class |
Conj<T>
Returns the complex conjugate of a complex number.
|
class |
Cos<T>
Computes cos of x element-wise.
|
class |
Cosh<T>
Computes hyperbolic cosine of x element-wise.
|
class |
Cumprod<T>
Compute the cumulative product of the tensor `x` along `axis`.
|
class |
Cumsum<T>
Compute the cumulative sum of the tensor `x` along `axis`.
|
class |
Digamma<T extends Number>
Computes Psi, the derivative of Lgamma (the log of the absolute value of
|
class |
Div<T>
Returns x / y element-wise.
|
class |
DivNoNan<T>
Returns 0 if the denominator is zero.
|
class |
Equal
Returns the truth value of (x == y) element-wise.
|
class |
Erf<T extends Number>
Computes the Gauss error function of `x` element-wise.
|
class |
Erfc<T extends Number>
Computes the complementary error function of `x` element-wise.
|
class |
Exp<T>
Computes exponential of x element-wise.
|
class |
Expm1<T>
Computes `exp(x) - 1` element-wise.
|
class |
Fact
Output a fact about factorials.
|
class |
Floor<T extends Number>
Returns element-wise largest integer not greater than x.
|
class |
FloorDiv<T>
Returns x // y element-wise.
|
class |
FloorMod<T extends Number>
Returns element-wise remainder of division.
|
class |
Greater
Returns the truth value of (x > y) element-wise.
|
class |
GreaterEqual
Returns the truth value of (x >= y) element-wise.
|
class |
Igamma<T extends Number>
Compute the lower regularized incomplete Gamma function `P(a, x)`.
|
class |
Igammac<T extends Number>
Compute the upper regularized incomplete Gamma function `Q(a, x)`.
|
class |
IgammaGradA<T extends Number>
Computes the gradient of `igamma(a, x)` wrt `a`.
|
class |
Imag<U extends Number>
Returns the imaginary part of a complex number.
|
class |
InvertPermutation<T extends Number>
Computes the inverse permutation of a tensor.
|
class |
IsFinite
Returns which elements of x are finite.
|
class |
IsInf
Returns which elements of x are Inf.
|
class |
IsNan
Returns which elements of x are NaN.
|
class |
Less
Returns the truth value of (x < y) element-wise.
|
class |
LessEqual
Returns the truth value of (x <= y) element-wise.
|
class |
Lgamma<T extends Number>
Computes the log of the absolute value of `Gamma(x)` element-wise.
|
class |
Log<T>
Computes natural logarithm of x element-wise.
|
class |
Log1p<T>
Computes natural logarithm of (1 + x) element-wise.
|
class |
LogicalAnd
Returns the truth value of x AND y element-wise.
|
class |
LogicalNot
Returns the truth value of NOT x element-wise.
|
class |
LogicalOr
Returns the truth value of x OR y element-wise.
|
class |
Maximum<T extends Number>
Returns the max of x and y (i.e.
|
class |
Mean<T>
Computes the mean of elements across dimensions of a tensor.
|
class |
Minimum<T extends Number>
Returns the min of x and y (i.e.
|
class |
Mod<T extends Number>
Returns element-wise remainder of division.
|
class |
Mul<T>
Returns x * y element-wise.
|
class |
Neg<T>
Computes numerical negative value element-wise.
|
class |
NotEqual
Returns the truth value of (x != y) element-wise.
|
class |
Polygamma<T extends Number>
Compute the polygamma function \\(\psi^{(n)}(x)\\).
|
class |
PopulationCount
Computes element-wise population count (a.k.a.
|
class |
Pow<T>
Computes the power of one value to another.
|
class |
QuantizedAdd<V>
Returns x + y element-wise, working on quantized buffers.
|
class |
QuantizedMul<V>
Returns x * y element-wise, working on quantized buffers.
|
class |
Real<U extends Number>
Returns the real part of a complex number.
|
class |
RealDiv<T>
Returns x / y element-wise for real types.
|
class |
Reciprocal<T>
Computes the reciprocal of x element-wise.
|
class |
ReciprocalGrad<T>
Computes the gradient for the inverse of `x` wrt its input.
|
class |
Rint<T extends Number>
Returns element-wise integer closest to x.
|
class |
Round<T>
Rounds the values of a tensor to the nearest integer, element-wise.
|
class |
Rsqrt<T>
Computes reciprocal of square root of x element-wise.
|
class |
RsqrtGrad<T>
Computes the gradient for the rsqrt of `x` wrt its input.
|
class |
SegmentMax<T extends Number>
Computes the maximum along segments of a tensor.
|
class |
SegmentMean<T>
Computes the mean along segments of a tensor.
|
class |
SegmentMin<T extends Number>
Computes the minimum along segments of a tensor.
|
class |
SegmentProd<T>
Computes the product along segments of a tensor.
|
class |
SegmentSum<T>
Computes the sum along segments of a tensor.
|
class |
Sigmoid<T>
Computes sigmoid of `x` element-wise.
|
class |
SigmoidGrad<T>
Computes the gradient of the sigmoid of `x` wrt its input.
|
class |
Sign<T>
Returns an element-wise indication of the sign of a number.
|
class |
Sin<T>
Computes sine of x element-wise.
|
class |
Sinh<T>
Computes hyperbolic sine of x element-wise.
|
class |
Softplus<T extends Number>
Computes softplus: `log(exp(features) + 1)`.
|
class |
SoftplusGrad<T extends Number>
Computes softplus gradients for a softplus operation.
|
class |
Sqrt<T>
Computes square root of x element-wise.
|
class |
SqrtGrad<T>
Computes the gradient for the sqrt of `x` wrt its input.
|
class |
Square<T>
Computes square of x element-wise.
|
class |
SquaredDifference<T>
Returns (x - y)(x - y) element-wise.
|
class |
Sub<T>
Returns x - y element-wise.
|
class |
Tan<T>
Computes tan of x element-wise.
|
class |
Tanh<T>
Computes hyperbolic tangent of `x` element-wise.
|
class |
TanhGrad<T>
Computes the gradient for the tanh of `x` wrt its input.
|
class |
TruncateDiv<T>
Returns x / y element-wise for integer types.
|
class |
TruncateMod<T extends Number>
Returns element-wise remainder of division.
|
class |
UnsortedSegmentMax<T extends Number>
Computes the maximum along segments of a tensor.
|
class |
UnsortedSegmentMin<T extends Number>
Computes the minimum along segments of a tensor.
|
class |
UnsortedSegmentProd<T>
Computes the product along segments of a tensor.
|
class |
UnsortedSegmentSum<T>
Computes the sum along segments of a tensor.
|
class |
Xdivy<T>
Returns 0 if x == 0, and x / y otherwise, elementwise.
|
class |
Xlogy<T>
Returns 0 if x == 0, and x * log(y) otherwise, elementwise.
|
class |
Zeta<T extends Number>
Compute the Hurwitz zeta function \\(\zeta(x, q)\\).
|
Modifier and Type | Class and Description |
---|---|
class |
AvgPool<T extends Number>
Performs average pooling on the input.
|
class |
AvgPool3d<T extends Number>
Performs 3D average pooling on the input.
|
class |
AvgPool3dGrad<T extends Number>
Computes gradients of average pooling function.
|
class |
AvgPoolGrad<T extends Number>
Computes gradients of the average pooling function.
|
class |
BatchNormWithGlobalNormalization<T>
Batch normalization.
|
class |
BatchNormWithGlobalNormalizationGrad<T>
Gradients for batch normalization.
|
class |
BiasAdd<T>
Adds `bias` to `value`.
|
class |
BiasAddGrad<T>
The backward operation for "BiasAdd" on the "bias" tensor.
|
class |
ComputeAccidentalHits
Computes the ids of the positions in sampled_candidates that match true_labels.
|
class |
Conv2d<T extends Number>
Computes a 2-D convolution given 4-D `input` and `filter` tensors.
|
class |
Conv2dBackpropFilter<T extends Number>
Computes the gradients of convolution with respect to the filter.
|
class |
Conv2dBackpropInput<T extends Number>
Computes the gradients of convolution with respect to the input.
|
class |
Conv3d<T extends Number>
Computes a 3-D convolution given 5-D `input` and `filter` tensors.
|
class |
Conv3dBackpropFilter<T extends Number>
Computes the gradients of 3-D convolution with respect to the filter.
|
class |
Conv3dBackpropInput<U extends Number>
Computes the gradients of 3-D convolution with respect to the input.
|
class |
CtcBeamSearchDecoder<T extends Number>
Performs beam search decoding on the logits given in input.
|
class |
CtcGreedyDecoder<T extends Number>
Performs greedy decoding on the logits given in inputs.
|
class |
CtcLoss<T extends Number>
Calculates the CTC Loss (log probability) for each batch entry.
|
class |
CudnnRnn<T extends Number>
A RNN backed by cuDNN.
|
class |
CudnnRnnBackprop<T extends Number>
Backprop step of CudnnRNN.
|
class |
CudnnRnnCanonicalToParams<T extends Number>
Converts CudnnRNN params from canonical form to usable form.
|
class |
CudnnRnnParamsSize<U extends Number>
Computes size of weights that can be used by a Cudnn RNN model.
|
class |
CudnnRnnParamsToCanonical<T extends Number>
Retrieves CudnnRNN params in canonical form.
|
class |
DataFormatDimMap<T extends Number>
Returns the dimension index in the destination data format given the one in
|
class |
DataFormatVecPermute<T extends Number>
Returns the permuted vector/tensor in the destination data format given the
|
class |
DepthToSpace<T>
DepthToSpace for tensors of type T.
|
class |
DepthwiseConv2dNative<T extends Number>
Computes a 2-D depthwise convolution given 4-D `input` and `filter` tensors.
|
class |
DepthwiseConv2dNativeBackpropFilter<T extends Number>
Computes the gradients of depthwise convolution with respect to the filter.
|
class |
DepthwiseConv2dNativeBackpropInput<T extends Number>
Computes the gradients of depthwise convolution with respect to the input.
|
class |
Dilation2d<T extends Number>
Computes the grayscale dilation of 4-D `input` and 3-D `filter` tensors.
|
class |
Dilation2dBackpropFilter<T extends Number>
Computes the gradient of morphological 2-D dilation with respect to the filter.
|
class |
Dilation2dBackpropInput<T extends Number>
Computes the gradient of morphological 2-D dilation with respect to the input.
|
class |
Elu<T extends Number>
Computes exponential linear: `exp(features) - 1` if < 0, `features` otherwise.
|
class |
EluGrad<T extends Number>
Computes gradients for the exponential linear (Elu) operation.
|
class |
FixedUnigramCandidateSampler
Generates labels for candidate sampling with a learned unigram distribution.
|
class |
FractionalAvgPool<T extends Number>
Performs fractional average pooling on the input.
|
class |
FractionalAvgPoolGrad<T extends Number>
Computes gradient of the FractionalAvgPool function.
|
class |
FractionalMaxPool<T extends Number>
Performs fractional max pooling on the input.
|
class |
FractionalMaxPoolGrad<T extends Number>
Computes gradient of the FractionalMaxPool function.
|
class |
FusedBatchNorm<T extends Number,U extends Number>
Batch normalization.
|
class |
FusedBatchNormGrad<T extends Number,U extends Number>
Gradient for batch normalization.
|
class |
FusedPadConv2d<T extends Number>
Performs a padding as a preprocess during a convolution.
|
class |
FusedResizeAndPadConv2d<T extends Number>
Performs a resize and padding as a preprocess during a convolution.
|
class |
InTopK
Says whether the targets are in the top `K` predictions.
|
class |
InvGrad<T>
Computes the gradient for the inverse of `x` wrt its input.
|
class |
L2Loss<T extends Number>
L2 Loss.
|
class |
LeakyRelu<T extends Number>
Computes rectified linear: `max(features, features * alpha)`.
|
class |
LearnedUnigramCandidateSampler
Generates labels for candidate sampling with a learned unigram distribution.
|
class |
LocalResponseNormalization<T extends Number>
Local Response Normalization.
|
class |
LocalResponseNormalizationGrad<T extends Number>
Gradients for Local Response Normalization.
|
class |
LogSoftmax<T extends Number>
Computes log softmax activations.
|
class |
MaxPool<T>
Performs max pooling on the input.
|
class |
MaxPool3d<T extends Number>
Performs 3D max pooling on the input.
|
class |
MaxPool3dGrad<U extends Number>
Computes gradients of max pooling function.
|
class |
MaxPool3dGradGrad<T extends Number>
Computes second-order gradients of the maxpooling function.
|
class |
MaxPoolGrad<T extends Number>
Computes gradients of the maxpooling function.
|
class |
MaxPoolGradGrad<T extends Number>
Computes second-order gradients of the maxpooling function.
|
class |
MaxPoolGradGradWithArgmax<T extends Number>
Computes second-order gradients of the maxpooling function.
|
class |
MaxPoolGradWithArgmax<T extends Number>
Computes gradients of the maxpooling function.
|
class |
MaxPoolWithArgmax<T extends Number,U extends Number>
Performs max pooling on the input and outputs both max values and indices.
|
class |
NthElement<T extends Number>
Finds values of the `n`-th order statistic for the last dimension.
|
class |
QuantizedAvgPool<T>
Produces the average pool of the input tensor for quantized types.
|
class |
QuantizedBatchNormWithGlobalNormalization<U>
Quantized Batch normalization.
|
class |
QuantizedBiasAdd<V>
Adds Tensor 'bias' to Tensor 'input' for Quantized types.
|
class |
QuantizedConv2d<V>
Computes a 2D convolution given quantized 4D input and filter tensors.
|
class |
QuantizedInstanceNorm<T>
Quantized Instance normalization.
|
class |
QuantizedMaxPool<T>
Produces the max pool of the input tensor for quantized types.
|
class |
QuantizedRelu<U>
Computes Quantized Rectified Linear: `max(features, 0)`
|
class |
QuantizedRelu6<U>
Computes Quantized Rectified Linear 6: `min(max(features, 0), 6)`
|
class |
QuantizedReluX<U>
Computes Quantized Rectified Linear X: `min(max(features, 0), max_value)`
|
class |
Relu<T>
Computes rectified linear: `max(features, 0)`.
|
class |
Relu6<T extends Number>
Computes rectified linear 6: `min(max(features, 0), 6)`.
|
class |
Relu6Grad<T extends Number>
Computes rectified linear 6 gradients for a Relu6 operation.
|
class |
ReluGrad<T extends Number>
Computes rectified linear gradients for a Relu operation.
|
class |
Selu<T extends Number>
Computes scaled exponential linear: `scale * alpha * (exp(features) - 1)`
|
class |
SeluGrad<T extends Number>
Computes gradients for the scaled exponential linear (Selu) operation.
|
class |
Softmax<T extends Number>
Computes softmax activations.
|
class |
SoftmaxCrossEntropyWithLogits<T extends Number>
Computes softmax cross entropy cost and gradients to backpropagate.
|
class |
Softsign<T extends Number>
Computes softsign: `features / (abs(features) + 1)`.
|
class |
SoftsignGrad<T extends Number>
Computes softsign gradients for a softsign operation.
|
class |
SpaceToBatch<T>
SpaceToBatch for 4-D tensors of type T.
|
class |
SpaceToDepth<T>
SpaceToDepth for tensors of type T.
|
class |
SparseSoftmaxCrossEntropyWithLogits<T extends Number>
Computes softmax cross entropy cost and gradients to backpropagate.
|
class |
TopK<T extends Number>
Finds values and indices of the `k` largest elements for the last dimension.
|
Modifier and Type | Class and Description |
---|---|
class |
Dequantize
Dequantize the 'input' tensor into a float Tensor.
|
class |
FakeQuantWithMinMaxArgs
Fake-quantize the 'inputs' tensor, type float to 'outputs' tensor of same type.
|
class |
FakeQuantWithMinMaxArgsGradient
Compute gradients for a FakeQuantWithMinMaxArgs operation.
|
class |
FakeQuantWithMinMaxVars
Fake-quantize the 'inputs' tensor of type float via global float scalars `min`
|
class |
FakeQuantWithMinMaxVarsGradient
Compute gradients for a FakeQuantWithMinMaxVars operation.
|
class |
FakeQuantWithMinMaxVarsPerChannel
Fake-quantize the 'inputs' tensor of type float and one of the shapes: `[d]`,
|
class |
FakeQuantWithMinMaxVarsPerChannelGradient
Compute gradients for a FakeQuantWithMinMaxVarsPerChannel operation.
|
class |
Quantize<T>
Quantize the 'input' tensor of type float to 'output' tensor of type 'T'.
|
class |
QuantizeAndDequantize<T extends Number>
Quantizes then dequantizes a tensor.
|
class |
QuantizeDownAndShrinkRange<U>
Convert the quantized 'input' tensor into a lower-precision 'output', using the
|
class |
RequantizationRange
Computes a range that covers the actual values present in a quantized tensor.
|
class |
Requantize<U>
Converts the quantized `input` tensor into a lower-precision `output`.
|
Modifier and Type | Class and Description |
---|---|
class |
AllCandidateSampler
Generates labels for candidate sampling with a learned unigram distribution.
|
class |
LogUniformCandidateSampler
Generates labels for candidate sampling with a log-uniform distribution.
|
class |
Multinomial<U extends Number>
Draws samples from a multinomial distribution.
|
class |
ParameterizedTruncatedNormal<U extends Number>
Outputs random values from a normal distribution.
|
class |
RandomGamma<U extends Number>
Outputs random values from the Gamma distribution(s) described by alpha.
|
class |
RandomGammaGrad<T extends Number>
Computes the derivative of a Gamma random sample w.r.t.
|
class |
RandomPoisson<V extends Number>
Outputs random values from the Poisson distribution(s) described by rate.
|
class |
RandomShuffle<T>
Randomly shuffles a tensor along its first dimension.
|
class |
RandomStandardNormal<U extends Number>
Outputs random values from a normal distribution.
|
class |
RandomUniform<U extends Number>
Outputs random values from a uniform distribution.
|
class |
RandomUniformInt<U extends Number>
Outputs random integers from a uniform distribution.
|
class |
RecordInput
Emits randomized records.
|
class |
StatelessMultinomial<V extends Number>
Draws samples from a multinomial distribution.
|
class |
StatelessRandomNormal<V extends Number>
Outputs deterministic pseudorandom values from a normal distribution.
|
class |
StatelessRandomUniform<V extends Number>
Outputs deterministic pseudorandom random values from a uniform distribution.
|
class |
StatelessRandomUniformInt<V extends Number>
Outputs deterministic pseudorandom random integers from a uniform distribution.
|
class |
StatelessTruncatedNormal<V extends Number>
Outputs deterministic pseudorandom values from a truncated normal distribution.
|
class |
TruncatedNormal<U extends Number>
Outputs random values from a truncated normal distribution.
|
class |
UniformCandidateSampler
Generates labels for candidate sampling with a uniform distribution.
|
Modifier and Type | Class and Description |
---|---|
class |
BatchFft |
class |
BatchFft2d |
class |
BatchFft3d |
class |
BatchIfft |
class |
BatchIfft2d |
class |
BatchIfft3d |
class |
Fft<T>
Fast Fourier transform.
|
class |
Fft2d<T>
2D fast Fourier transform.
|
class |
Fft3d<T>
3D fast Fourier transform.
|
class |
Ifft<T>
Inverse fast Fourier transform.
|
class |
Ifft2d<T>
Inverse 2D fast Fourier transform.
|
class |
Ifft3d<T>
Inverse 3D fast Fourier transform.
|
class |
Irfft
Inverse real-valued fast Fourier transform.
|
class |
Irfft2d
Inverse 2D real-valued fast Fourier transform.
|
class |
Irfft3d
Inverse 3D real-valued fast Fourier transform.
|
class |
Rfft
Real-valued fast Fourier transform.
|
class |
Rfft2d
2D real-valued fast Fourier transform.
|
class |
Rfft3d
3D real-valued fast Fourier transform.
|
Modifier and Type | Class and Description |
---|---|
class |
AddManySparseToTensorsMap
Add an `N`-minibatch `SparseTensor` to a `SparseTensorsMap`, return `N` handles.
|
class |
AddSparseToTensorsMap
Add a `SparseTensor` to a `SparseTensorsMap` return its handle.
|
class |
DenseToDenseSetOperation<T>
Applies set operation along last dimension of 2 `Tensor` inputs.
|
class |
DenseToSparseSetOperation<T>
Applies set operation along last dimension of `Tensor` and `SparseTensor`.
|
class |
DeserializeSparse<U>
Deserialize `SparseTensor` objects.
|
class |
SparseAccumulatorApplyGradient
Applies a sparse gradient to a given accumulator.
|
class |
SparseAccumulatorTakeGradient<T>
Extracts the average sparse gradient in a SparseConditionalAccumulator.
|
class |
SparseAdd<T>
Adds two `SparseTensor` objects to produce another `SparseTensor`.
|
class |
SparseAddGrad<T>
The gradient operator for the SparseAdd op.
|
class |
SparseConcat<T>
Concatenates a list of `SparseTensor` along the specified dimension.
|
class |
SparseConditionalAccumulator
A conditional accumulator for aggregating sparse gradients.
|
class |
SparseCross<T>
Generates sparse cross from a list of sparse and dense tensors.
|
class |
SparseDenseCwiseAdd<T>
Adds up a SparseTensor and a dense Tensor, using these special rules:
|
class |
SparseDenseCwiseDiv<T>
Component-wise divides a SparseTensor by a dense Tensor.
|
class |
SparseDenseCwiseMul<T>
Component-wise multiplies a SparseTensor by a dense Tensor.
|
class |
SparseFillEmptyRows<T>
Fills empty rows in the input 2-D `SparseTensor` with a default value.
|
class |
SparseFillEmptyRowsGrad<T>
The gradient of SparseFillEmptyRows.
|
class |
SparseMatMul
Multiply matrix "a" by matrix "b".
|
class |
SparseReduceMax<T extends Number>
Computes the max of elements across dimensions of a SparseTensor.
|
class |
SparseReduceMaxSparse<T extends Number>
Computes the max of elements across dimensions of a SparseTensor.
|
class |
SparseReduceSum<T>
Computes the sum of elements across dimensions of a SparseTensor.
|
class |
SparseReduceSumSparse<T>
Computes the sum of elements across dimensions of a SparseTensor.
|
class |
SparseReorder<T>
Reorders a SparseTensor into the canonical, row-major ordering.
|
class |
SparseReshape
Reshapes a SparseTensor to represent values in a new dense shape.
|
class |
SparseSegmentMean<T extends Number>
Computes the mean along sparse segments of a tensor.
|
class |
SparseSegmentMeanGrad<T extends Number>
Computes gradients for SparseSegmentMean.
|
class |
SparseSegmentMeanWithNumSegments<T extends Number>
Computes the mean along sparse segments of a tensor.
|
class |
SparseSegmentSqrtN<T extends Number>
Computes the sum along sparse segments of a tensor divided by the sqrt of N.
|
class |
SparseSegmentSqrtNGrad<T extends Number>
Computes gradients for SparseSegmentSqrtN.
|
class |
SparseSegmentSqrtNWithNumSegments<T extends Number>
Computes the sum along sparse segments of a tensor divided by the sqrt of N.
|
class |
SparseSegmentSum<T extends Number>
Computes the sum along sparse segments of a tensor.
|
class |
SparseSegmentSumWithNumSegments<T extends Number>
Computes the sum along sparse segments of a tensor.
|
class |
SparseSlice<T>
Slice a `SparseTensor` based on the `start` and `size`.
|
class |
SparseSliceGrad<T>
The gradient operator for the SparseSlice op.
|
class |
SparseSoftmax<T extends Number>
Applies softmax to a batched N-D `SparseTensor`.
|
class |
SparseSparseMaximum<T extends Number>
Returns the element-wise max of two SparseTensors.
|
class |
SparseSparseMinimum<T>
Returns the element-wise min of two SparseTensors.
|
class |
SparseSplit<T>
Split a `SparseTensor` into `num_split` tensors along one dimension.
|
class |
SparseTensorDenseAdd<U>
Adds up a `SparseTensor` and a dense `Tensor`, producing a dense `Tensor`.
|
class |
SparseTensorDenseMatMul<U>
Multiply SparseTensor (of rank 2) "A" by dense matrix "B".
|
class |
SparseToDense<U>
Converts a sparse representation into a dense tensor.
|
class |
SparseToSparseSetOperation<T>
Applies set operation along last dimension of 2 `SparseTensor` inputs.
|
class |
TakeManySparseFromTensorsMap<T>
Read `SparseTensors` from a `SparseTensorsMap` and concatenate them.
|
Modifier and Type | Class and Description |
---|---|
class |
Join
Joins the strings in the given list of string tensors into one tensor;
|
class |
ReduceJoin
Joins a string Tensor across the given dimensions.
|
class |
RegexFullMatch
Check if the input matches the regex pattern.
|
class |
RegexReplace
Replaces matches of the `pattern` regular expression in `input` with the
replacement string provided in `rewrite`.
|
class |
StaticRegexFullMatch
Check if the input matches the regex pattern.
|
class |
StaticRegexReplace
Replaces the match of pattern in input with rewrite.
|
class |
StringFormat
Formats a string template using a list of tensors.
|
class |
StringLength
String lengths of `input`.
|
class |
StringSplit
Split elements of `source` based on `sep` into a `SparseTensor`.
|
class |
Strip
Strip leading and trailing whitespaces from the Tensor.
|
class |
Substr
Return substrings from `Tensor` of strings.
|
class |
ToHashBucket
Converts each string in the input Tensor to its hash mod by a number of buckets.
|
class |
ToHashBucketFast
Converts each string in the input Tensor to its hash mod by a number of buckets.
|
class |
ToHashBucketStrong
Converts each string in the input Tensor to its hash mod by a number of buckets.
|
class |
ToNumber<T extends Number>
Converts each string in the input Tensor to the specified numeric type.
|
class |
UnicodeDecodeWithOffsets<T extends Number>
Decodes each string in `input` into a sequence of Unicode code points.
|
class |
UnicodeScript
Determine the script codes of a given tensor of Unicode integer code points.
|
class |
UnicodeTranscode
Transcode the input text from a source encoding to a destination encoding.
|
Modifier and Type | Class and Description |
---|---|
class |
AudioSummary
Outputs a `Summary` protocol buffer with audio.
|
class |
CloseSummaryWriter |
class |
CreateSummaryDbWriter |
class |
CreateSummaryFileWriter |
class |
FlushSummaryWriter |
class |
HistogramSummary
Outputs a `Summary` protocol buffer with a histogram.
|
class |
ImageSummary
Outputs a `Summary` protocol buffer with images.
|
class |
ImportEvent |
class |
MergeSummary
Merges summaries.
|
class |
ScalarSummary
Outputs a `Summary` protocol buffer with scalar values.
|
class |
StatsAggregatorSummary
Produces a summary of any statistics recorded by the given statistics manager.
|
class |
SummaryWriter |
class |
TensorSummary
Outputs a `Summary` protocol buffer with a tensor and per-plugin data.
|
class |
WriteAudioSummary |
class |
WriteGraphSummary |
class |
WriteHistogramSummary |
class |
WriteImageSummary |
class |
WriteScalarSummary |
class |
WriteSummary |
Modifier and Type | Class and Description |
---|---|
class |
AccumulatorApplyGradient
Applies a gradient to a given accumulator.
|
class |
AccumulatorNumAccumulated
Returns the number of gradients aggregated in the given accumulators.
|
class |
AccumulatorSetGlobalStep
Updates the accumulator with a new value for global_step.
|
class |
AccumulatorTakeGradient<T>
Extracts the average gradient in the given ConditionalAccumulator.
|
class |
ApplyAdadelta<T>
Update '*var' according to the adadelta scheme.
|
class |
ApplyAdagrad<T>
Update '*var' according to the adagrad scheme.
|
class |
ApplyAdagradDa<T>
Update '*var' according to the proximal adagrad scheme.
|
class |
ApplyAdam<T>
Update '*var' according to the Adam algorithm.
|
class |
ApplyAdaMax<T>
Update '*var' according to the AdaMax algorithm.
|
class |
ApplyAddSign<T>
Update '*var' according to the AddSign update.
|
class |
ApplyCenteredRmsProp<T>
Update '*var' according to the centered RMSProp algorithm.
|
class |
ApplyFtrl<T>
Update '*var' according to the Ftrl-proximal scheme.
|
class |
ApplyGradientDescent<T>
Update '*var' by subtracting 'alpha' * 'delta' from it.
|
class |
ApplyMomentum<T>
Update '*var' according to the momentum scheme.
|
class |
ApplyPowerSign<T>
Update '*var' according to the AddSign update.
|
class |
ApplyProximalAdagrad<T>
Update '*var' and '*accum' according to FOBOS with Adagrad learning rate.
|
class |
ApplyProximalGradientDescent<T>
Update '*var' as FOBOS algorithm with fixed learning rate.
|
class |
ApplyRmsProp<T>
Update '*var' according to the RMSProp algorithm.
|
class |
ConditionalAccumulator
A conditional accumulator for aggregating gradients.
|
class |
GenerateVocabRemapping
Given a path to new and old vocabulary files, returns a remapping Tensor of
|
class |
MergeV2Checkpoints
V2 format specific: merges the metadata files of sharded checkpoints.
|
class |
NegTrain
Training via negative sampling.
|
class |
PreventGradient<T>
An identity op that triggers an error if a gradient is requested.
|
class |
ResourceApplyAdadelta
Update '*var' according to the adadelta scheme.
|
class |
ResourceApplyAdagrad
Update '*var' according to the adagrad scheme.
|
class |
ResourceApplyAdagradDa
Update '*var' according to the proximal adagrad scheme.
|
class |
ResourceApplyAdam
Update '*var' according to the Adam algorithm.
|
class |
ResourceApplyAdaMax
Update '*var' according to the AdaMax algorithm.
|
class |
ResourceApplyAddSign
Update '*var' according to the AddSign update.
|
class |
ResourceApplyCenteredRmsProp
Update '*var' according to the centered RMSProp algorithm.
|
class |
ResourceApplyFtrl
Update '*var' according to the Ftrl-proximal scheme.
|
class |
ResourceApplyGradientDescent
Update '*var' by subtracting 'alpha' * 'delta' from it.
|
class |
ResourceApplyMomentum
Update '*var' according to the momentum scheme.
|
class |
ResourceApplyPowerSign
Update '*var' according to the AddSign update.
|
class |
ResourceApplyProximalAdagrad
Update '*var' and '*accum' according to FOBOS with Adagrad learning rate.
|
class |
ResourceApplyProximalGradientDescent
Update '*var' as FOBOS algorithm with fixed learning rate.
|
class |
ResourceApplyRmsProp
Update '*var' according to the RMSProp algorithm.
|
class |
ResourceSparseApplyAdadelta
var: Should be from a Variable().
|
class |
ResourceSparseApplyAdagrad
Update relevant entries in '*var' and '*accum' according to the adagrad scheme.
|
class |
ResourceSparseApplyAdagradDa
Update entries in '*var' and '*accum' according to the proximal adagrad scheme.
|
class |
ResourceSparseApplyCenteredRmsProp
Update '*var' according to the centered RMSProp algorithm.
|
class |
ResourceSparseApplyFtrl
Update relevant entries in '*var' according to the Ftrl-proximal scheme.
|
class |
ResourceSparseApplyMomentum
Update relevant entries in '*var' and '*accum' according to the momentum scheme.
|
class |
ResourceSparseApplyProximalAdagrad
Sparse update entries in '*var' and '*accum' according to FOBOS algorithm.
|
class |
ResourceSparseApplyProximalGradientDescent
Sparse update '*var' as FOBOS algorithm with fixed learning rate.
|
class |
ResourceSparseApplyRmsProp
Update '*var' according to the RMSProp algorithm.
|
class |
Restore
Restores tensors from a V2 checkpoint.
|
class |
RestoreSlice<T>
Restores a tensor from checkpoint files.
|
class |
Save
Saves tensors in V2 checkpoint format.
|
class |
SaveSlices
Saves input tensors slices to disk.
|
class |
SdcaFprint
Computes fingerprints of the input strings.
|
class |
SdcaOptimizer
Distributed version of Stochastic Dual Coordinate Ascent (SDCA) optimizer for
|
class |
SdcaShrinkL1
Applies L1 regularization shrink step on the parameters.
|
class |
SparseApplyAdadelta<T>
var: Should be from a Variable().
|
class |
SparseApplyAdagrad<T>
Update relevant entries in '*var' and '*accum' according to the adagrad scheme.
|
class |
SparseApplyAdagradDa<T>
Update entries in '*var' and '*accum' according to the proximal adagrad scheme.
|
class |
SparseApplyCenteredRmsProp<T>
Update '*var' according to the centered RMSProp algorithm.
|
class |
SparseApplyFtrl<T>
Update relevant entries in '*var' according to the Ftrl-proximal scheme.
|
class |
SparseApplyMomentum<T>
Update relevant entries in '*var' and '*accum' according to the momentum scheme.
|
class |
SparseApplyProximalAdagrad<T>
Sparse update entries in '*var' and '*accum' according to FOBOS algorithm.
|
class |
SparseApplyProximalGradientDescent<T>
Sparse update '*var' as FOBOS algorithm with fixed learning rate.
|
class |
SparseApplyRmsProp<T>
Update '*var' according to the RMSProp algorithm.
|
class |
TileGrad<T>
Returns the gradient of `Tile`.
|
Copyright © 2022. All rights reserved.