org.bytedeco.pytorch

Class LSTMImplBase

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.pytorch.Module

org.bytedeco.pytorch.LSTMImplCloneable

org.bytedeco.pytorch.LSTMImplBase

All Implemented Interfaces:

AutoCloseable

Direct Known Subclasses:

LSTMImpl

@Name(value="torch::nn::detail::RNNImplBase<torch::nn::LSTMImpl>")
 @NoOffset
 @Properties(inherit=torch.class)
public class LSTMImplBase
extends LSTMImplCloneable

Nested Class Summary

Nested classes/interfaces inherited from class org.bytedeco.javacpp.Pointer

Pointer.CustomDeallocator, Pointer.Deallocator, Pointer.NativeDeallocator, Pointer.ReferenceCounter

Field Summary

Fields inherited from class org.bytedeco.javacpp.Pointer

address, capacity, limit, position

Constructor Summary

Constructors

Constructor and Description
LSTMImplBase(Pointer p) Pointer cast constructor.
LSTMImplBase(RNNOptionsBase options_)

Method Summary

Modifier and Type	Method and Description
TensorVector	all_weights()
void	flatten_parameters() Modifies the internal storage of weights for optimization purposes.
RNNOptionsBase	options_base() The RNN's options.
LSTMImplBase	options_base(RNNOptionsBase setter)
void	pretty_print(Pointer stream) Pretty prints the RNN module into the given stream.
void	reset_parameters()
void	reset() Initializes the parameters of the RNN module.
void	to(Device device)
void	to(Device device, boolean non_blocking) Recursively moves all parameters to the given device.
void	to(Device device, torch.ScalarType dtype)
void	to(Device device, torch.ScalarType dtype, boolean non_blocking) Overrides nn::Module::to() to call flatten_parameters() after the original operation.
void	to(torch.ScalarType dtype)
void	to(torch.ScalarType dtype, boolean non_blocking) Recursively casts all parameters to the given dtype.

Methods inherited from class org.bytedeco.pytorch.LSTMImplCloneable

clone, clone

Methods inherited from class org.bytedeco.pytorch.Module

apply, apply, apply, apply, apply, apply, apply, apply, asAdaptiveAvgPool1d, asAdaptiveAvgPool2d, asAdaptiveAvgPool3d, asAdaptiveLogSoftmaxWithLoss, asAdaptiveMaxPool1d, asAdaptiveMaxPool2d, asAdaptiveMaxPool3d, asAlphaDropout, asAvgPool1d, asAvgPool2d, asAvgPool3d, asBatchNorm1d, asBatchNorm2d, asBatchNorm3d, asBCELoss, asBCEWithLogitsLoss, asBilinear, asCELU, asConstantPad1d, asConstantPad2d, asConstantPad3d, asConv1d, asConv2d, asConv3d, asConvTranspose1d, asConvTranspose2d, asConvTranspose3d, asCosineEmbeddingLoss, asCosineSimilarity, asCrossEntropyLoss, asCrossMapLRN2d, asCTCLoss, asDropout, asDropout2d, asDropout3d, asELU, asEmbedding, asEmbeddingBag, asFeatureAlphaDropout, asFlatten, asFold, asFractionalMaxPool2d, asFractionalMaxPool3d, asGELU, asGLU, asGroupNorm, asGRU, asGRUCell, asHardshrink, asHardtanh, asHingeEmbeddingLoss, asHuberLoss, asIdentity, asInstanceNorm1d, asInstanceNorm2d, asInstanceNorm3d, asKLDivLoss, asL1Loss, asLayerNorm, asLeakyReLU, asLinear, asLocalResponseNorm, asLogSigmoid, asLogSoftmax, asLPPool1d, asLPPool2d, asLPPool3d, asLSTM, asLSTMCell, asMarginRankingLoss, asMaxPool1d, asMaxPool2d, asMaxPool3d, asMaxUnpool1d, asMaxUnpool2d, asMaxUnpool3d, asMish, asModuleDict, asModuleList, asMSELoss, asMultiheadAttention, asMultiLabelMarginLoss, asMultiLabelSoftMarginLoss, asMultiMarginLoss, asNLLLoss, asPairwiseDistance, asParameterDict, asParameterList, asPixelShuffle, asPixelUnshuffle, asPoissonNLLLoss, asPReLU, asReflectionPad1d, asReflectionPad2d, asReflectionPad3d, asReLU, asReLU6, asReplicationPad1d, asReplicationPad2d, asReplicationPad3d, asRNN, asRNNCell, asRReLU, asSELU, asSequential, asSigmoid, asSiLU, asSmoothL1Loss, asSoftMarginLoss, asSoftmax, asSoftmax2d, asSoftmin, asSoftplus, asSoftshrink, asSoftsign, asTanh, asTanhshrink, asThreshold, asTransformer, asTransformerDecoder, asTransformerDecoderLayer, asTransformerEncoder, asTransformerEncoderLayer, asTripletMarginLoss, asTripletMarginWithDistanceLoss, asUnflatten, asUnfold, asUpsample, asZeroPad1d, asZeroPad2d, asZeroPad3d, buffers, buffers, children, eval, is_serializable, is_training, load, modules, modules, name, named_buffers, named_buffers, named_children, named_modules, named_modules, named_modules, named_parameters, named_parameters, parameters, parameters, put, register_buffer, register_buffer, register_module, register_module, register_parameter, register_parameter, register_parameter, register_parameter, save, shiftLeft, train, unregister_module, unregister_module, zero_grad

Methods inherited from class org.bytedeco.javacpp.Pointer

address, asBuffer, asByteBuffer, availablePhysicalBytes, calloc, capacity, capacity, close, deallocate, deallocate, deallocateReferences, deallocator, deallocator, equals, fill, formatBytes, free, getDirectBufferAddress, getPointer, getPointer, getPointer, getPointer, hashCode, interruptDeallocatorThread, isNull, isNull, limit, limit, malloc, maxBytes, maxPhysicalBytes, memchr, memcmp, memcpy, memmove, memset, offsetAddress, offsetof, offsetof, parseBytes, physicalBytes, physicalBytesInaccurate, position, position, put, realloc, referenceCount, releaseReference, retainReference, setNull, sizeof, sizeof, toString, totalBytes, totalCount, totalPhysicalBytes, withDeallocator, zero

Methods inherited from class java.lang.Object

finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

LSTMImplBase

public LSTMImplBase(Pointer p)

Pointer cast constructor. Invokes Pointer(Pointer).

LSTMImplBase

public LSTMImplBase(@Const @ByRef
                    RNNOptionsBase options_)

Method Detail

reset

public void reset()

Initializes the parameters of the RNN module.

Overrides:

reset in class LSTMImplCloneable

reset_parameters

public void reset_parameters()

to

public void to(@ByVal
               Device device,
               torch.ScalarType dtype,
               @Cast(value="bool")
               boolean non_blocking)

Overrides nn::Module::to() to call flatten_parameters() after the original operation.

Overrides:

to in class Module

to

public void to(@ByVal
               Device device,
               torch.ScalarType dtype)

to

public void to(torch.ScalarType dtype,
               @Cast(value="bool")
               boolean non_blocking)

Description copied from class: Module

Recursively casts all parameters to the given dtype. If non_blocking is true and the source is in pinned memory and destination is on the GPU or vice versa, the copy is performed asynchronously with respect to the host. Otherwise, the argument has no effect.

Overrides:

to in class Module

to

public void to(torch.ScalarType dtype)

to

public void to(@ByVal
               Device device,
               @Cast(value="bool")
               boolean non_blocking)

Description copied from class: Module

Recursively moves all parameters to the given device. If non_blocking is true and the source is in pinned memory and destination is on the GPU or vice versa, the copy is performed asynchronously with respect to the host. Otherwise, the argument has no effect.

Overrides:

to in class Module

to

public void to(@ByVal
               Device device)

pretty_print

public void pretty_print(@Cast(value="std::ostream*") @ByRef
                         Pointer stream)

Pretty prints the RNN module into the given stream.

Overrides:

pretty_print in class Module

flatten_parameters

public void flatten_parameters()

Modifies the internal storage of weights for optimization purposes. On CPU, this method should be called if any of the weight or bias vectors are changed (i.e. weights are added or removed). On GPU, it should be called __any time the storage of any parameter is modified__, e.g. any time a parameter is assigned a new value. This allows using the fast path in cuDNN implementations of respective RNN forward() methods. It is called once upon construction, inside reset().

all_weights

@ByVal
public TensorVector all_weights()

options_base

@ByRef
public RNNOptionsBase options_base()

The RNN's options.

options_base

public LSTMImplBase options_base(RNNOptionsBase setter)