U
- data type for paramsSize()
output@Operator(group="nn") public final class CudnnRnnParamsSize<U extends Number> extends PrimitiveOp implements Operand<U>
Return the params size that can be used by the Cudnn RNN model. Subsequent weight allocation and initialization should use this size.
num_layers: Specifies the number of layers in the RNN model. num_units: Specifies the size of the hidden state. input_size: Specifies the size of the input state. rnn_mode: Indicates the type of the RNN model. input_mode: Indicate whether there is a linear projection between the input and The actual computation before the first layer. 'skip_input' is only allowed when input_size == num_units; 'auto_select' implies 'skip_input' when input_size == num_units; otherwise, it implies 'linear_input'. direction: Indicates whether a bidirectional model will be used. dir = (direction == bidirectional) ? 2 : 1 dropout: dropout probability. When set to 0., dropout is disabled. seed: the 1st part of a seed to initialize dropout. seed2: the 2nd part of a seed to initialize dropout. params_size: The size of the params buffer that should be allocated and initialized for this RNN model. Note that this params buffer may not be compatible across GPUs. Please use CudnnRNNParamsWeights and CudnnRNNParamsBiases to save and restore them in a way that is compatible across different runs.
Modifier and Type | Class and Description |
---|---|
static class |
CudnnRnnParamsSize.Options
Optional attributes for
CudnnRnnParamsSize |
operation
Modifier and Type | Method and Description |
---|---|
Output<U> |
asOutput()
Returns the symbolic handle of a tensor.
|
static <U extends Number,T extends Number> |
create(Scope scope,
Operand<Integer> numLayers,
Operand<Integer> numUnits,
Operand<Integer> inputSize,
Class<T> T,
Class<U> S,
CudnnRnnParamsSize.Options... options)
Factory method to create a class wrapping a new CudnnRnnParamsSize operation.
|
static CudnnRnnParamsSize.Options |
direction(String direction) |
static CudnnRnnParamsSize.Options |
dropout(Float dropout) |
static CudnnRnnParamsSize.Options |
inputMode(String inputMode) |
static CudnnRnnParamsSize.Options |
numProj(Long numProj) |
Output<U> |
paramsSize() |
static CudnnRnnParamsSize.Options |
rnnMode(String rnnMode) |
static CudnnRnnParamsSize.Options |
seed(Long seed) |
static CudnnRnnParamsSize.Options |
seed2(Long seed2) |
equals, hashCode, op, toString
public static <U extends Number,T extends Number> CudnnRnnParamsSize<U> create(Scope scope, Operand<Integer> numLayers, Operand<Integer> numUnits, Operand<Integer> inputSize, Class<T> T, Class<U> S, CudnnRnnParamsSize.Options... options)
scope
- current scopenumLayers
- numUnits
- inputSize
- T
- S
- options
- carries optional attributes valuespublic static CudnnRnnParamsSize.Options rnnMode(String rnnMode)
rnnMode
- public static CudnnRnnParamsSize.Options inputMode(String inputMode)
inputMode
- public static CudnnRnnParamsSize.Options direction(String direction)
direction
- public static CudnnRnnParamsSize.Options dropout(Float dropout)
dropout
- public static CudnnRnnParamsSize.Options seed(Long seed)
seed
- public static CudnnRnnParamsSize.Options seed2(Long seed2)
seed2
- public static CudnnRnnParamsSize.Options numProj(Long numProj)
numProj
- public Output<U> asOutput()
Operand
Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.
asOutput
in interface Operand<U extends Number>
OperationBuilder.addInput(Output)
Copyright © 2022. All rights reserved.