Class RNNDescriptor

Inheritance

System.Object

RNNDescriptor

Implements

System.IDisposable

Inherited Members

System.Object.Equals(System.Object)

System.Object.Equals(System.Object, System.Object)

System.Object.GetHashCode()

System.Object.GetType()

System.Object.MemberwiseClone()

System.Object.ReferenceEquals(System.Object, System.Object)

System.Object.ToString()

Namespace: ManagedCuda.CudaDNN

Assembly: CudaDNN.dll

Syntax

public class RNNDescriptor : IDisposable

Constructors

RNNDescriptor(CudaDNNContext)

Declaration

public RNNDescriptor(CudaDNNContext context)

Parameters

Type	Name	Description
CudaDNNContext	context

Properties

Desc

Returns the inner handle.

Declaration

public cudnnRNNDescriptor Desc { get; }

Property Value

Type	Description
cudnnRNNDescriptor

MathType

The math type specified in a given RNN descriptor.

Declaration

public cudnnMathType MathType { set; }

Property Value

Type	Description
cudnnMathType

Methods

cudnnGetRNNParamsSize(TensorDescriptor, ref SizeT, cudnnDataType)

This function is used to query the amount of parameter space required to execute the RNN described by rnnDesc with inputs dimensions defined by xDesc.

Declaration

public void cudnnGetRNNParamsSize(TensorDescriptor xDesc, ref SizeT sizeInBytes, cudnnDataType dataType)

Parameters

Type	Name	Description
TensorDescriptor	xDesc	A fully packed tensor descriptor describing the input to one recurrent iteration.
SizeT	sizeInBytes	Minimum amount of GPU memory needed as parameter space to be able to execute an RNN with the specified descriptor and input tensors.
cudnnDataType	dataType	The data type of the parameters.

Dispose()

Dispose

Declaration

public void Dispose()

Dispose(Boolean)

For IDisposable

Declaration

protected virtual void Dispose(bool fDisposing)

Parameters

Type	Name	Description
System.Boolean	fDisposing

Finalize()

For dispose

Declaration

protected void Finalize()

GetRNNLinLayerBiasParams(Int32, TensorDescriptor, FilterDescriptor, CudaDeviceVariable<Double>, Int32, FilterDescriptor, CudaDeviceVariable<SizeT>)

This function is used to obtain a pointer and descriptor for the bias parameters in layer within the RNN described by rnnDesc with inputs dimensions defined by xDesc.

Declaration

public void GetRNNLinLayerBiasParams(int layer, TensorDescriptor xDesc, FilterDescriptor wDesc, CudaDeviceVariable<double> w, int linLayerID, FilterDescriptor linLayerBiasDesc, CudaDeviceVariable<SizeT> linLayerBias)

Parameters

Type	Name	Description
System.Int32	layer	The layer to query.
TensorDescriptor	xDesc	A fully packed tensor descriptor describing the input to one recurrent iteration.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Double>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
System.Int32	linLayerID	The linear layer to obtain information about: If mode in rnnDesc was set to CUDNN_RNN_RELU or CUDNN_RNN_TANH a value of 0 references the bias applied to the input from the previous layer, a value of 1 references the bias applied to the recurrent input. If mode in rnnDesc was set to CUDNN_LSTM values of 0, 1, 2 and 3 reference bias applied to the input from the previous layer, value of 4, 5, 6 and 7 reference bias applied to the recurrent input. ‣ Values 0 and 4 reference the input gate. ‣ Values 1 and 5 reference the forget gate. ‣ Values 2 and 6 reference the new memory gate. ‣ Values 3 and 7 reference the output gate. If mode in rnnDesc was set to CUDNN_GRU values of 0, 1 and 2 reference bias applied to the input from the previous layer, value of 3, 4 and 5 reference bias applied to the recurrent input. ‣ Values 0 and 3 reference the reset gate. ‣ Values 1 and 4 reference the update gate. ‣ Values 2 and 5 reference the new memory gate.
FilterDescriptor	linLayerBiasDesc	Handle to a previously created filter descriptor.
CudaDeviceVariable<SizeT>	linLayerBias	Data pointer to GPU memory associated with the filter descriptor linLayerMatDesc.

GetRNNLinLayerBiasParams(Int32, TensorDescriptor, FilterDescriptor, CudaDeviceVariable<Single>, Int32, FilterDescriptor, CudaDeviceVariable<SizeT>)

This function is used to obtain a pointer and descriptor for the bias parameters in layer within the RNN described by rnnDesc with inputs dimensions defined by xDesc.

Declaration

public void GetRNNLinLayerBiasParams(int layer, TensorDescriptor xDesc, FilterDescriptor wDesc, CudaDeviceVariable<float> w, int linLayerID, FilterDescriptor linLayerBiasDesc, CudaDeviceVariable<SizeT> linLayerBias)

Parameters

Type	Name	Description
System.Int32	layer	The layer to query.
TensorDescriptor	xDesc	A fully packed tensor descriptor describing the input to one recurrent iteration.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Single>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
System.Int32	linLayerID	The linear layer to obtain information about: If mode in rnnDesc was set to CUDNN_RNN_RELU or CUDNN_RNN_TANH a value of 0 references the bias applied to the input from the previous layer, a value of 1 references the bias applied to the recurrent input. If mode in rnnDesc was set to CUDNN_LSTM values of 0, 1, 2 and 3 reference bias applied to the input from the previous layer, value of 4, 5, 6 and 7 reference bias applied to the recurrent input. ‣ Values 0 and 4 reference the input gate. ‣ Values 1 and 5 reference the forget gate. ‣ Values 2 and 6 reference the new memory gate. ‣ Values 3 and 7 reference the output gate. If mode in rnnDesc was set to CUDNN_GRU values of 0, 1 and 2 reference bias applied to the input from the previous layer, value of 3, 4 and 5 reference bias applied to the recurrent input. ‣ Values 0 and 3 reference the reset gate. ‣ Values 1 and 4 reference the update gate. ‣ Values 2 and 5 reference the new memory gate.
FilterDescriptor	linLayerBiasDesc	Handle to a previously created filter descriptor.
CudaDeviceVariable<SizeT>	linLayerBias	Data pointer to GPU memory associated with the filter descriptor linLayerMatDesc.

GetRNNLinLayerMatrixParams(Int32, TensorDescriptor, FilterDescriptor, CudaDeviceVariable<Double>, Int32, FilterDescriptor, CudaDeviceVariable<SizeT>)

This function is used to obtain a pointer and descriptor for the matrix parameters in layer within the RNN described by rnnDesc with inputs dimensions defined by xDesc.

Declaration

public void GetRNNLinLayerMatrixParams(int layer, TensorDescriptor xDesc, FilterDescriptor wDesc, CudaDeviceVariable<double> w, int linLayerID, FilterDescriptor linLayerMatDesc, CudaDeviceVariable<SizeT> linLayerMat)

Parameters

Type	Name	Description
System.Int32	layer	The layer to query.
TensorDescriptor	xDesc	A fully packed tensor descriptor describing the input to one recurrent iteration.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Double>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
System.Int32	linLayerID	The linear layer to obtain information about: If mode in rnnDesc was set to CUDNN_RNN_RELU or CUDNN_RNN_TANH a value of 0 references the matrix multiplication applied to the input from the previous layer, a value of 1 references the matrix multiplication applied to the recurrent input. If mode in rnnDesc was set to CUDNN_LSTM values of 0-3 reference matrix multiplications applied to the input from the previous layer, value of 4-7 reference matrix multiplications applied to the recurrent input. ‣ Values 0 and 4 reference the input gate. ‣ Values 1 and 5 reference the forget gate. ‣ Values 2 and 6 reference the new memory gate. ‣ Values 3 and 7 reference the output gate. If mode in rnnDesc was set to CUDNN_GRU values of 0-2 reference matrix multiplications applied to the input from the previous layer, value of 3-5 reference matrix multiplications applied to the recurrent input. ‣ Values 0 and 3 reference the reset gate. ‣ Values 1 and 4 reference the update gate. ‣ Values 2 and 5 reference the new memory gate.
FilterDescriptor	linLayerMatDesc	Handle to a previously created filter descriptor.
CudaDeviceVariable<SizeT>	linLayerMat	Data pointer to GPU memory associated with the filter descriptor linLayerMatDesc.

GetRNNLinLayerMatrixParams(Int32, TensorDescriptor, FilterDescriptor, CudaDeviceVariable<Single>, Int32, FilterDescriptor, CudaDeviceVariable<SizeT>)

This function is used to obtain a pointer and descriptor for the matrix parameters in layer within the RNN described by rnnDesc with inputs dimensions defined by xDesc.

Declaration

public void GetRNNLinLayerMatrixParams(int layer, TensorDescriptor xDesc, FilterDescriptor wDesc, CudaDeviceVariable<float> w, int linLayerID, FilterDescriptor linLayerMatDesc, CudaDeviceVariable<SizeT> linLayerMat)

Parameters

Type	Name	Description
System.Int32	layer	The layer to query.
TensorDescriptor	xDesc	A fully packed tensor descriptor describing the input to one recurrent iteration.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Single>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
System.Int32	linLayerID	The linear layer to obtain information about: If mode in rnnDesc was set to CUDNN_RNN_RELU or CUDNN_RNN_TANH a value of 0 references the matrix multiplication applied to the input from the previous layer, a value of 1 references the matrix multiplication applied to the recurrent input. If mode in rnnDesc was set to CUDNN_LSTM values of 0-3 reference matrix multiplications applied to the input from the previous layer, value of 4-7 reference matrix multiplications applied to the recurrent input. ‣ Values 0 and 4 reference the input gate. ‣ Values 1 and 5 reference the forget gate. ‣ Values 2 and 6 reference the new memory gate. ‣ Values 3 and 7 reference the output gate. If mode in rnnDesc was set to CUDNN_GRU values of 0-2 reference matrix multiplications applied to the input from the previous layer, value of 3-5 reference matrix multiplications applied to the recurrent input. ‣ Values 0 and 3 reference the reset gate. ‣ Values 1 and 4 reference the update gate. ‣ Values 2 and 5 reference the new memory gate.
FilterDescriptor	linLayerMatDesc	Handle to a previously created filter descriptor.
CudaDeviceVariable<SizeT>	linLayerMat	Data pointer to GPU memory associated with the filter descriptor linLayerMatDesc.

GetRNNTrainingReserveSize(Int32, TensorDescriptor[], ref SizeT)

This function is used to query the amount of reserved space required for training the RNN described by rnnDesc with inputs dimensions defined by xDesc. The same reserve space must be passed to cudnnRNNForwardTraining, cudnnRNNBackwardData and cudnnRNNBackwardWeights.

Declaration

public void GetRNNTrainingReserveSize(int seqLength, TensorDescriptor[] xDesc, ref SizeT sizeInBytes)

Parameters

Type	Name	Description
System.Int32	seqLength	Number of iterations to unroll over.
TensorDescriptor[]	xDesc	An array of tensor descriptors describing the input to each recurrent iteration.
SizeT	sizeInBytes	Minimum amount of GPU memory needed as reserve space to be able to train an RNN with the specified descriptor and input tensors.

GetRNNWorkspaceSize(Int32, TensorDescriptor[], ref SizeT)

This function is used to query the amount of work space required to execute the RNN described by rnnDesc with inputs dimensions defined by xDesc.

Declaration

public void GetRNNWorkspaceSize(int seqLength, TensorDescriptor[] xDesc, ref SizeT sizeInBytes)

Parameters

Type	Name	Description
System.Int32	seqLength	Number of iterations to unroll over.
TensorDescriptor[]	xDesc	An array of tensor descriptors describing the input to each recurrent iteration.
SizeT	sizeInBytes	Minimum amount of GPU memory needed as workspace to be able to execute an RNN with the specified descriptor and input tensors.

RNNBackwardData(TensorDescriptor[], CudaDeviceVariable<Double>, TensorDescriptor[], CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, FilterDescriptor, CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor[], CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, CudaDeviceVariable<Byte>, SizeT, CudaDeviceVariable<Byte>, SizeT)

This routine executes the recurrent neural network described by rnnDesc with output gradients dy, dhy, dhc, weights w and input gradients dx, dhx, dcx. workspace is required for intermediate storage. The data in reserveSpace must have previously been generated by cudnnRNNForwardTraining. The same reserveSpace data must be used for future calls to cudnnRNNBackwardWeights if they execute on the same input data.

Declaration

public void RNNBackwardData(TensorDescriptor[] yDesc, CudaDeviceVariable<double> y, TensorDescriptor[] dyDesc, CudaDeviceVariable<double> dy, TensorDescriptor dhyDesc, CudaDeviceVariable<double> dhy, TensorDescriptor dcyDesc, CudaDeviceVariable<double> dcy, FilterDescriptor wDesc, CudaDeviceVariable<double> w, TensorDescriptor hxDesc, CudaDeviceVariable<double> hx, TensorDescriptor cxDesc, CudaDeviceVariable<double> cx, TensorDescriptor[] dxDesc, CudaDeviceVariable<double> dx, TensorDescriptor dhxDesc, CudaDeviceVariable<double> dhx, TensorDescriptor dcxDesc, CudaDeviceVariable<double> dcx, CudaDeviceVariable<byte> workspace, SizeT workSpaceSizeInBytes, CudaDeviceVariable<byte> reserveSpace, SizeT reserveSpaceSizeInBytes)

Parameters

Type	Name	Description
TensorDescriptor[]	yDesc	An array of tensor descriptors describing the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in dyDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	y	Data pointer to GPU memory associated with the output tensor descriptor yDesc.
TensorDescriptor[]	dyDesc	An array of tensor descriptors describing the gradient at the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in dxDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	dy	Data pointer to GPU memory associated with the tensor descriptors in the array dyDesc.
TensorDescriptor	dhyDesc	Handle to a previously initialized tensor descriptor describing the gradients at the final hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in dyDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	dhy	Data pointer to GPU memory associated with the tensor descriptor dhyDesc. If a NULL pointer is passed, the gradients at the final hidden state of the network will be initialized to zero.
TensorDescriptor	dcyDesc	Handle to a previously initialized tensor descriptor describing the gradients at the final cell state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in dyDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	dcy	Data pointer to GPU memory associated with the tensor descriptor dcyDesc. If a NULL pointer is passed, the gradients at the final cell state of the network will be initialized to zero.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Double>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
TensorDescriptor	hxDesc	Handle to a previously initialized tensor descriptor describing the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	hx	Data pointer to GPU memory associated with the tensor descriptor hxDesc. If a NULL pointer is passed, the initial hidden state of the network will be initialized to zero.
TensorDescriptor	cxDesc	Handle to a previously initialized tensor descriptor describing the initial cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	cx	Data pointer to GPU memory associated with the tensor descriptor cxDesc. If a NULL pointer is passed, the initial cell state of the network will be initialized to zero.
TensorDescriptor[]	dxDesc	An array of tensor descriptors describing the gradient at the input of each recurrent iteration. Each tensor descriptor must have the same first dimension. The second dimension of the tensors may decrease from element n to element n+1 but may not increase. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	dx	Data pointer to GPU memory associated with the tensor descriptors in the array dxDesc.
TensorDescriptor	dhxDesc	Handle to a previously initialized tensor descriptor describing the gradient at the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	dhx	Data pointer to GPU memory associated with the tensor descriptor dhxDesc. If a NULL pointer is passed, the gradient at the hidden input of the network will not be set.
TensorDescriptor	dcxDesc	Handle to a previously initialized tensor descriptor describing the gradient at the initial cell state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	dcx	Data pointer to GPU memory associated with the tensor descriptor dcxDesc. If a NULL pointer is passed, the gradient at the cell input of the network will not be set.
CudaDeviceVariable<System.Byte>	workspace	Data pointer to GPU memory to be used as a workspace for this call.
SizeT	workSpaceSizeInBytes	Specifies the size in bytes of the provided workspace.
CudaDeviceVariable<System.Byte>	reserveSpace	Data pointer to GPU memory to be used as a reserve space for this call.
SizeT	reserveSpaceSizeInBytes	Specifies the size in bytes of the provided reserveSpace.

RNNBackwardData(TensorDescriptor[], CudaDeviceVariable<Single>, TensorDescriptor[], CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, FilterDescriptor, CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor[], CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, CudaDeviceVariable<Byte>, SizeT, CudaDeviceVariable<Byte>, SizeT)

This routine executes the recurrent neural network described by rnnDesc with output gradients dy, dhy, dhc, weights w and input gradients dx, dhx, dcx. workspace is required for intermediate storage. The data in reserveSpace must have previously been generated by cudnnRNNForwardTraining. The same reserveSpace data must be used for future calls to cudnnRNNBackwardWeights if they execute on the same input data.

Declaration

public void RNNBackwardData(TensorDescriptor[] yDesc, CudaDeviceVariable<float> y, TensorDescriptor[] dyDesc, CudaDeviceVariable<float> dy, TensorDescriptor dhyDesc, CudaDeviceVariable<float> dhy, TensorDescriptor dcyDesc, CudaDeviceVariable<float> dcy, FilterDescriptor wDesc, CudaDeviceVariable<float> w, TensorDescriptor hxDesc, CudaDeviceVariable<float> hx, TensorDescriptor cxDesc, CudaDeviceVariable<float> cx, TensorDescriptor[] dxDesc, CudaDeviceVariable<float> dx, TensorDescriptor dhxDesc, CudaDeviceVariable<float> dhx, TensorDescriptor dcxDesc, CudaDeviceVariable<float> dcx, CudaDeviceVariable<byte> workspace, SizeT workSpaceSizeInBytes, CudaDeviceVariable<byte> reserveSpace, SizeT reserveSpaceSizeInBytes)

Parameters

Type	Name	Description
TensorDescriptor[]	yDesc	An array of tensor descriptors describing the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in dyDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	y	Data pointer to GPU memory associated with the output tensor descriptor yDesc.
TensorDescriptor[]	dyDesc	An array of tensor descriptors describing the gradient at the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in dxDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	dy	Data pointer to GPU memory associated with the tensor descriptors in the array dyDesc.
TensorDescriptor	dhyDesc	Handle to a previously initialized tensor descriptor describing the gradients at the final hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in dyDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	dhy	Data pointer to GPU memory associated with the tensor descriptor dhyDesc. If a NULL pointer is passed, the gradients at the final hidden state of the network will be initialized to zero.
TensorDescriptor	dcyDesc	Handle to a previously initialized tensor descriptor describing the gradients at the final cell state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in dyDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	dcy	Data pointer to GPU memory associated with the tensor descriptor dcyDesc. If a NULL pointer is passed, the gradients at the final cell state of the network will be initialized to zero.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Single>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
TensorDescriptor	hxDesc	Handle to a previously initialized tensor descriptor describing the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	hx	Data pointer to GPU memory associated with the tensor descriptor hxDesc. If a NULL pointer is passed, the initial hidden state of the network will be initialized to zero.
TensorDescriptor	cxDesc	Handle to a previously initialized tensor descriptor describing the initial cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	cx	Data pointer to GPU memory associated with the tensor descriptor cxDesc. If a NULL pointer is passed, the initial cell state of the network will be initialized to zero.
TensorDescriptor[]	dxDesc	An array of tensor descriptors describing the gradient at the input of each recurrent iteration. Each tensor descriptor must have the same first dimension. The second dimension of the tensors may decrease from element n to element n+1 but may not increase. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	dx	Data pointer to GPU memory associated with the tensor descriptors in the array dxDesc.
TensorDescriptor	dhxDesc	Handle to a previously initialized tensor descriptor describing the gradient at the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	dhx	Data pointer to GPU memory associated with the tensor descriptor dhxDesc. If a NULL pointer is passed, the gradient at the hidden input of the network will not be set.
TensorDescriptor	dcxDesc	Handle to a previously initialized tensor descriptor describing the gradient at the initial cell state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	dcx	Data pointer to GPU memory associated with the tensor descriptor dcxDesc. If a NULL pointer is passed, the gradient at the cell input of the network will not be set.
CudaDeviceVariable<System.Byte>	workspace	Data pointer to GPU memory to be used as a workspace for this call.
SizeT	workSpaceSizeInBytes	Specifies the size in bytes of the provided workspace.
CudaDeviceVariable<System.Byte>	reserveSpace	Data pointer to GPU memory to be used as a reserve space for this call.
SizeT	reserveSpaceSizeInBytes	Specifies the size in bytes of the provided reserveSpace.

RNNBackwardWeights(TensorDescriptor[], CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor[], CudaDeviceVariable<Double>, CudaDeviceVariable<Byte>, SizeT, FilterDescriptor, CudaDeviceVariable<Double>, CudaDeviceVariable<Byte>, SizeT)

This routine accumulates weight gradients dw from the recurrent neural network described by rnnDesc with inputs x, hx, and outputs y. The mode of operation in this case is additive, the weight gradients calculated will be added to those already existing in dw. workspace is required for intermediate storage. The data in reserveSpace must have previously been generated by cudnnRNNBackwardData.

Declaration

public void RNNBackwardWeights(TensorDescriptor[] xDesc, CudaDeviceVariable<double> x, TensorDescriptor hxDesc, CudaDeviceVariable<double> hx, TensorDescriptor[] yDesc, CudaDeviceVariable<double> y, CudaDeviceVariable<byte> workspace, SizeT workSpaceSizeInBytes, FilterDescriptor dwDesc, CudaDeviceVariable<double> dw, CudaDeviceVariable<byte> reserveSpace, SizeT reserveSpaceSizeInBytes)

Parameters

Type	Name	Description
TensorDescriptor[]	xDesc	An array of tensor descriptors describing the input to each recurrent iteration. Each tensor descriptor must have the same first dimension. The second dimension of the tensors may decrease from element n to element n+1 but may not increase. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	x	Data pointer to GPU memory associated with the tensor descriptors in the array xDesc.
TensorDescriptor	hxDesc	Handle to a previously initialized tensor descriptor describing the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	hx	Data pointer to GPU memory associated with the tensor descriptor hxDesc. If a NULL pointer is passed, the initial hidden state of the network will be initialized to zero.
TensorDescriptor[]	yDesc	An array of tensor descriptors describing the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in dyDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	y	Data pointer to GPU memory associated with the output tensor descriptor yDesc.
CudaDeviceVariable<System.Byte>	workspace	Data pointer to GPU memory to be used as a workspace for this call.
SizeT	workSpaceSizeInBytes	Specifies the size in bytes of the provided workspace.
FilterDescriptor	dwDesc	Handle to a previously initialized filter descriptor describing the gradients of the weights for the RNN.
CudaDeviceVariable<System.Double>	dw	Data pointer to GPU memory associated with the filter descriptor dwDesc.
CudaDeviceVariable<System.Byte>	reserveSpace	Data pointer to GPU memory to be used as a reserve space for this call.
SizeT	reserveSpaceSizeInBytes	Specifies the size in bytes of the provided reserveSpace.

RNNBackwardWeights(TensorDescriptor[], CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor[], CudaDeviceVariable<Single>, CudaDeviceVariable<Byte>, SizeT, FilterDescriptor, CudaDeviceVariable<Single>, CudaDeviceVariable<Byte>, SizeT)

This routine accumulates weight gradients dw from the recurrent neural network described by rnnDesc with inputs x, hx, and outputs y. The mode of operation in this case is additive, the weight gradients calculated will be added to those already existing in dw. workspace is required for intermediate storage. The data in reserveSpace must have previously been generated by cudnnRNNBackwardData.

Declaration

public void RNNBackwardWeights(TensorDescriptor[] xDesc, CudaDeviceVariable<float> x, TensorDescriptor hxDesc, CudaDeviceVariable<float> hx, TensorDescriptor[] yDesc, CudaDeviceVariable<float> y, CudaDeviceVariable<byte> workspace, SizeT workSpaceSizeInBytes, FilterDescriptor dwDesc, CudaDeviceVariable<float> dw, CudaDeviceVariable<byte> reserveSpace, SizeT reserveSpaceSizeInBytes)

Parameters

Type	Name	Description
TensorDescriptor[]	xDesc	An array of tensor descriptors describing the input to each recurrent iteration. Each tensor descriptor must have the same first dimension. The second dimension of the tensors may decrease from element n to element n+1 but may not increase. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	x	Data pointer to GPU memory associated with the tensor descriptors in the array xDesc.
TensorDescriptor	hxDesc	Handle to a previously initialized tensor descriptor describing the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	hx	Data pointer to GPU memory associated with the tensor descriptor hxDesc. If a NULL pointer is passed, the initial hidden state of the network will be initialized to zero.
TensorDescriptor[]	yDesc	An array of tensor descriptors describing the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in dyDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	y	Data pointer to GPU memory associated with the output tensor descriptor yDesc.
CudaDeviceVariable<System.Byte>	workspace	Data pointer to GPU memory to be used as a workspace for this call.
SizeT	workSpaceSizeInBytes	Specifies the size in bytes of the provided workspace.
FilterDescriptor	dwDesc	Handle to a previously initialized filter descriptor describing the gradients of the weights for the RNN.
CudaDeviceVariable<System.Single>	dw	Data pointer to GPU memory associated with the filter descriptor dwDesc.
CudaDeviceVariable<System.Byte>	reserveSpace	Data pointer to GPU memory to be used as a reserve space for this call.
SizeT	reserveSpaceSizeInBytes	Specifies the size in bytes of the provided reserveSpace.

RNNForwardInference(TensorDescriptor[], CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, FilterDescriptor, CudaDeviceVariable<Double>, TensorDescriptor[], CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, CudaDeviceVariable<Byte>, SizeT)

This routine executes the recurrent neural network described by rnnDesc with inputs x, hx, cx, weights w and outputs y, hy, cy. workspace is required for intermediate storage. This function does not store data required for training; cudnnRNNForwardTraining should be used for that purpose.

Declaration

public void RNNForwardInference(TensorDescriptor[] xDesc, CudaDeviceVariable<double> x, TensorDescriptor hxDesc, CudaDeviceVariable<double> hx, TensorDescriptor cxDesc, CudaDeviceVariable<double> cx, FilterDescriptor wDesc, CudaDeviceVariable<double> w, TensorDescriptor[] yDesc, CudaDeviceVariable<double> y, TensorDescriptor hyDesc, CudaDeviceVariable<double> hy, TensorDescriptor cyDesc, CudaDeviceVariable<double> cy, CudaDeviceVariable<byte> workspace, SizeT workSpaceSizeInBytes)

Parameters

Type	Name	Description
TensorDescriptor[]	xDesc	An array of tensor descriptors describing the input to each recurrent iteration. Each tensor descriptor must have the same first dimension. The second dimension of the tensors may decrease from element n to element n+1 but may not increase. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	x	Data pointer to GPU memory associated with the tensor descriptors in the array xDesc. The data are expected to be packed contiguously with the first element of iteration n+1 following directly from the last element of iteration n.
TensorDescriptor	hxDesc	Handle to a previously initialized tensor descriptor describing the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	hx	Data pointer to GPU memory associated with the tensor descriptor hxDesc. If a NULL pointer is passed, the initial hidden state of the network will be initialized to zero.
TensorDescriptor	cxDesc	Handle to a previously initialized tensor descriptor describing the initial cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	cx	Data pointer to GPU memory associated with the tensor descriptor cxDesc. If a NULL pointer is passed, the initial cell state of the network will be initialized to zero.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Double>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
TensorDescriptor[]	yDesc	An array of tensor descriptors describing the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in xDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	y	Data pointer to GPU memory associated with the output tensor descriptor yDesc. The data are expected to be packed contiguously with the first element of iteration n+1 following directly from the last element of iteration n.
TensorDescriptor	hyDesc	Handle to a previously initialized tensor descriptor describing the final hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	hy	Data pointer to GPU memory associated with the tensor descriptor hyDesc. If a NULL pointer is passed, the final hidden state of the network will not be saved.
TensorDescriptor	cyDesc	Handle to a previously initialized tensor descriptor describing the final cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	cy	Data pointer to GPU memory associated with the tensor descriptor cyDesc. If a NULL pointer is passed, the final cell state of the network will be not be saved.
CudaDeviceVariable<System.Byte>	workspace	Data pointer to GPU memory to be used as a workspace for this call.
SizeT	workSpaceSizeInBytes	Specifies the size in bytes of the provided workspace.

RNNForwardInference(TensorDescriptor[], CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, FilterDescriptor, CudaDeviceVariable<Single>, TensorDescriptor[], CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, CudaDeviceVariable<Byte>, SizeT)

This routine executes the recurrent neural network described by rnnDesc with inputs x, hx, cx, weights w and outputs y, hy, cy. workspace is required for intermediate storage. This function does not store data required for training; cudnnRNNForwardTraining should be used for that purpose.

Declaration

public void RNNForwardInference(TensorDescriptor[] xDesc, CudaDeviceVariable<float> x, TensorDescriptor hxDesc, CudaDeviceVariable<float> hx, TensorDescriptor cxDesc, CudaDeviceVariable<float> cx, FilterDescriptor wDesc, CudaDeviceVariable<float> w, TensorDescriptor[] yDesc, CudaDeviceVariable<float> y, TensorDescriptor hyDesc, CudaDeviceVariable<float> hy, TensorDescriptor cyDesc, CudaDeviceVariable<float> cy, CudaDeviceVariable<byte> workspace, SizeT workSpaceSizeInBytes)

Parameters

Type	Name	Description
TensorDescriptor[]	xDesc	An array of tensor descriptors describing the input to each recurrent iteration. Each tensor descriptor must have the same first dimension. The second dimension of the tensors may decrease from element n to element n+1 but may not increase. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	x	Data pointer to GPU memory associated with the tensor descriptors in the array xDesc. The data are expected to be packed contiguously with the first element of iteration n+1 following directly from the last element of iteration n.
TensorDescriptor	hxDesc	Handle to a previously initialized tensor descriptor describing the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	hx	Data pointer to GPU memory associated with the tensor descriptor hxDesc. If a NULL pointer is passed, the initial hidden state of the network will be initialized to zero.
TensorDescriptor	cxDesc	Handle to a previously initialized tensor descriptor describing the initial cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	cx	Data pointer to GPU memory associated with the tensor descriptor cxDesc. If a NULL pointer is passed, the initial cell state of the network will be initialized to zero.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Single>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
TensorDescriptor[]	yDesc	An array of tensor descriptors describing the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in xDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	y	Data pointer to GPU memory associated with the output tensor descriptor yDesc. The data are expected to be packed contiguously with the first element of iteration n+1 following directly from the last element of iteration n.
TensorDescriptor	hyDesc	Handle to a previously initialized tensor descriptor describing the final hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	hy	Data pointer to GPU memory associated with the tensor descriptor hyDesc. If a NULL pointer is passed, the final hidden state of the network will not be saved.
TensorDescriptor	cyDesc	Handle to a previously initialized tensor descriptor describing the final cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	cy	Data pointer to GPU memory associated with the tensor descriptor cyDesc. If a NULL pointer is passed, the final cell state of the network will be not be saved.
CudaDeviceVariable<System.Byte>	workspace	Data pointer to GPU memory to be used as a workspace for this call.
SizeT	workSpaceSizeInBytes	Specifies the size in bytes of the provided workspace.

RNNForwardTraining(TensorDescriptor[], CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, FilterDescriptor, CudaDeviceVariable<Double>, TensorDescriptor[], CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, TensorDescriptor, CudaDeviceVariable<Double>, CudaDeviceVariable<Byte>, SizeT, CudaDeviceVariable<Byte>, SizeT)

This routine executes the recurrent neural network described by rnnDesc with inputs x, hx, cx, weights w and outputs y, hy, cy. workspace is required for intermediate storage. reserveSpace stores data required for training. The same reserveSpace data must be used for future calls to cudnnRNNBackwardData and cudnnRNNBackwardWeights if these execute on the same input data.

Declaration

public void RNNForwardTraining(TensorDescriptor[] xDesc, CudaDeviceVariable<double> x, TensorDescriptor hxDesc, CudaDeviceVariable<double> hx, TensorDescriptor cxDesc, CudaDeviceVariable<double> cx, FilterDescriptor wDesc, CudaDeviceVariable<double> w, TensorDescriptor[] yDesc, CudaDeviceVariable<double> y, TensorDescriptor hyDesc, CudaDeviceVariable<double> hy, TensorDescriptor cyDesc, CudaDeviceVariable<double> cy, CudaDeviceVariable<byte> workspace, SizeT workSpaceSizeInBytes, CudaDeviceVariable<byte> reserveSpace, SizeT reserveSpaceSizeInBytes)

Parameters

Type	Name	Description
TensorDescriptor[]	xDesc	An array of tensor descriptors describing the input to each recurrent iteration. Each tensor descriptor must have the same first dimension. The second dimension of the tensors may decrease from element n to element n+1 but may not increase. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	x	Data pointer to GPU memory associated with the tensor descriptors in the array xDesc.
TensorDescriptor	hxDesc	Handle to a previously initialized tensor descriptor describing the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	hx	Data pointer to GPU memory associated with the tensor descriptor hxDesc. If a NULL pointer is passed, the initial hidden state of the network will be initialized to zero.
TensorDescriptor	cxDesc	Handle to a previously initialized tensor descriptor describing the initial cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	cx	Data pointer to GPU memory associated with the tensor descriptor cxDesc. If a NULL pointer is passed, the initial cell state of the network will be initialized to zero.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Double>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
TensorDescriptor[]	yDesc	An array of tensor descriptors describing the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in xDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	y	Data pointer to GPU memory associated with the output tensor descriptor yDesc.
TensorDescriptor	hyDesc	Handle to a previously initialized tensor descriptor describing the final hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	hy	Data pointer to GPU memory associated with the tensor descriptor hyDesc. If a NULL pointer is passed, the final hidden state of the network will not be saved.
TensorDescriptor	cyDesc	Handle to a previously initialized tensor descriptor describing the final cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Double>	cy	Data pointer to GPU memory associated with the tensor descriptor cyDesc. If a NULL pointer is passed, the final cell state of the network will be not be saved.
CudaDeviceVariable<System.Byte>	workspace	Data pointer to GPU memory to be used as a workspace for this call.
SizeT	workSpaceSizeInBytes	Specifies the size in bytes of the provided workspace.
CudaDeviceVariable<System.Byte>	reserveSpace	Data pointer to GPU memory to be used as a reserve space for this call.
SizeT	reserveSpaceSizeInBytes	Specifies the size in bytes of the provided reserveSpace.

RNNForwardTraining(TensorDescriptor[], CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, FilterDescriptor, CudaDeviceVariable<Single>, TensorDescriptor[], CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, TensorDescriptor, CudaDeviceVariable<Single>, CudaDeviceVariable<Byte>, SizeT, CudaDeviceVariable<Byte>, SizeT)

This routine executes the recurrent neural network described by rnnDesc with inputs x, hx, cx, weights w and outputs y, hy, cy. workspace is required for intermediate storage. reserveSpace stores data required for training. The same reserveSpace data must be used for future calls to cudnnRNNBackwardData and cudnnRNNBackwardWeights if these execute on the same input data.

Declaration

public void RNNForwardTraining(TensorDescriptor[] xDesc, CudaDeviceVariable<float> x, TensorDescriptor hxDesc, CudaDeviceVariable<float> hx, TensorDescriptor cxDesc, CudaDeviceVariable<float> cx, FilterDescriptor wDesc, CudaDeviceVariable<float> w, TensorDescriptor[] yDesc, CudaDeviceVariable<float> y, TensorDescriptor hyDesc, CudaDeviceVariable<float> hy, TensorDescriptor cyDesc, CudaDeviceVariable<float> cy, CudaDeviceVariable<byte> workspace, SizeT workSpaceSizeInBytes, CudaDeviceVariable<byte> reserveSpace, SizeT reserveSpaceSizeInBytes)

Parameters

Type	Name	Description
TensorDescriptor[]	xDesc	An array of tensor descriptors describing the input to each recurrent iteration. Each tensor descriptor must have the same first dimension. The second dimension of the tensors may decrease from element n to element n+1 but may not increase. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	x	Data pointer to GPU memory associated with the tensor descriptors in the array xDesc.
TensorDescriptor	hxDesc	Handle to a previously initialized tensor descriptor describing the initial hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	hx	Data pointer to GPU memory associated with the tensor descriptor hxDesc. If a NULL pointer is passed, the initial hidden state of the network will be initialized to zero.
TensorDescriptor	cxDesc	Handle to a previously initialized tensor descriptor describing the initial cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	cx	Data pointer to GPU memory associated with the tensor descriptor cxDesc. If a NULL pointer is passed, the initial cell state of the network will be initialized to zero.
FilterDescriptor	wDesc	Handle to a previously initialized filter descriptor describing the weights for the RNN.
CudaDeviceVariable<System.Single>	w	Data pointer to GPU memory associated with the filter descriptor wDesc.
TensorDescriptor[]	yDesc	An array of tensor descriptors describing the output from each recurrent iteration. The first dimension of the tensor depends on the direction argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc: If direction is CUDNN_UNIDIRECTIONAL the first dimension should match the hiddenSize argument passed to cudnnSetRNNDescriptor. If direction is CUDNN_BIDIRECTIONAL the first dimension should match double the hiddenSize argument passed to cudnnSetRNNDescriptor. The second dimension of the tensor n must match the second dimension of the tensor n in xDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	y	Data pointer to GPU memory associated with the output tensor descriptor yDesc.
TensorDescriptor	hyDesc	Handle to a previously initialized tensor descriptor describing the final hidden state of the RNN. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	hy	Data pointer to GPU memory associated with the tensor descriptor hyDesc. If a NULL pointer is passed, the final hidden state of the network will not be saved.
TensorDescriptor	cyDesc	Handle to a previously initialized tensor descriptor describing the final cell state for LSTM networks. The first dimension of the tensor must match the hiddenSize argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The second dimension must match the second dimension of the first tensor described in xDesc. The third dimension must match the numLayers argument passed to the cudnnSetRNNDescriptor call used to initialize rnnDesc. The tensor must be fully packed.
CudaDeviceVariable<System.Single>	cy	Data pointer to GPU memory associated with the tensor descriptor cyDesc. If a NULL pointer is passed, the final cell state of the network will be not be saved.
CudaDeviceVariable<System.Byte>	workspace	Data pointer to GPU memory to be used as a workspace for this call.
SizeT	workSpaceSizeInBytes	Specifies the size in bytes of the provided workspace.
CudaDeviceVariable<System.Byte>	reserveSpace	Data pointer to GPU memory to be used as a reserve space for this call.
SizeT	reserveSpaceSizeInBytes	Specifies the size in bytes of the provided reserveSpace.

SetPersistentRNNPlan(PersistentRNNPlan)

This function sets the persistent RNN plan to be executed when using rnnDesc and CUDNN_RNN_ALGO_PERSIST_DYNAMIC algo.

Declaration

public void SetPersistentRNNPlan(PersistentRNNPlan plan)

Parameters

Type	Name	Description
PersistentRNNPlan	plan

SetRNNDescriptor(CudaDNNContext, Int32, Int32, DropoutDescriptor, cudnnRNNInputMode, cudnnDirectionMode, cudnnRNNMode, cudnnRNNAlgo, cudnnDataType)

This function initializes a previously created RNN descriptor object.

Declaration

public void SetRNNDescriptor(CudaDNNContext ctx, int hiddenSize, int numLayers, DropoutDescriptor dropoutDesc, cudnnRNNInputMode inputMode, cudnnDirectionMode direction, cudnnRNNMode mode, cudnnRNNAlgo algo, cudnnDataType dataType)

Parameters

Type	Name	Description
CudaDNNContext	ctx	Handle to a previously created cuDNN library descriptor.
System.Int32	hiddenSize	Size of the internal hidden state for each layer.
System.Int32	numLayers	Number of layers.
DropoutDescriptor	dropoutDesc	Handle to a previously created and initialized dropout descriptor.
cudnnRNNInputMode	inputMode	Specifies the behavior at the input to the first layer.
cudnnDirectionMode	direction	Specifies the recurrence pattern. (eg. bidirectional)
cudnnRNNMode	mode	The type of RNN to compute.
cudnnRNNAlgo	algo	Specifies which RNN algorithm should be used to compute the results.
cudnnDataType	dataType	Math precision.

Implements

System.IDisposable