Namespace ManagedCuda.CudaDNN

Classes

ConvolutionDescriptor

An opaque structure holding the description of a convolution operation.

CudaDNNContext

An opaque structure holding the cuDNN library context.

The cuDNN library context must be created using cudnnCreate() and the returned handle must be passed to all subsequent library function calls. The context should be destroyed at the end using cudnnDestroy(). The context is associated with only one GPU device, the current device at the time of the call to cudnnCreate(). However multiple contexts can be created on the same GPU device.

CudaDNNException

An CudaDNNException is thrown, if any wrapped call to the cudnn-library does not return Success.

CudaDNNNativeMethods

DropoutDescriptor

An opaque structure holding the description of a generic n-D dataset.

FilterDescriptor

An opaque structure holding the description of a filter dataset.

LRNDescriptor

OpTensorDescriptor

An opaque structure holding the description of a generic n-D dataset.

PersistentRNNPlan

PersistentRNNPlan is a pointer to an opaque structure holding a plan to execute a dynamic persistent RNN. cudnnCreatePersistentRNNPlan() is used to create and initialize one instance.

PoolingDescriptor

An opaque structure holding the description of a pooling operation.

ReduceTensorDescriptor

ReduceTensorDescriptor is a pointer to an opaque structure holding the description of a tensor reduction operation, used as a parameter to cudnnReduceTensor(). cudnnCreateReduceTensorDescriptor() is used to create one instance, and cudnnSetReduceTensorDescriptor() must be used to initialize this instance.

RNNDescriptor

SpatialTransformerDescriptor

An opaque structure holding the description of a generic n-D dataset.

TensorDescriptor

An opaque structure holding the description of a generic n-D dataset.

Structs

BNConstants

Constant values for BN

cudnnActivationDescriptor

cudnnActivationDescriptor is a pointer to an opaque structure holding the description of a activation operation.

cudnnConvolutionBwdDataAlgoPerf

cudnnConvolutionBwdDataAlgoPerf is a structure containing performance results returned by cudnnFindConvolutionBackwardDataAlgorithm().

cudnnConvolutionBwdFilterAlgoPerf

cudnnConvolutionBwdFilterAlgoPerf is a structure containing performance results returned by cudnnFindConvolutionBackwardFilterAlgorithm().

cudnnConvolutionDescriptor

cudnnConvolutionDescriptor is a pointer to an opaque structure holding the description of a convolution operation.

cudnnConvolutionFwdAlgoPerf

cudnnConvolutionFwdAlgoPerf is a structure containing performance results returned by cudnnFindConvolutionForwardAlgorithm().

cudnnCTCLossDescriptor

cudnnDropoutDescriptor

cudnnDropoutDescriptor_t is a pointer to an opaque structure holding the description of a dropout operation. cudnnCreateDropoutDescriptor() is used to create one instance, cudnnSetDropoutDescriptor() is be used to initialize this instance, cudnnDestroyDropoutDescriptor() is be used to destroy this instance.

cudnnFilterDescriptor

cudnnFilterDescriptor is a pointer to an opaque structure holding the description of a filter dataset.

cudnnHandle

cudnnHandle is a pointer to an opaque structure holding the cuDNN library context.

cudnnLRNDescriptor

cudnnLRNDescriptor is a pointer to an opaque structure holding the description of a local response normalization operation.

cudnnOpTensorDescriptor

cudnnOpTensorDescriptor is a pointer to an opaque structure holding the description of a tensor operation, used as a parameter to cudnnOpTensor(). cudnnCreateOpTensorDescriptor() is used to create one instance, and cudnnSetOpTensorDescriptor() must be used to initialize this instance.

cudnnPersistentRNNPlan

cudnnPersistentRNNPlan_t is a pointer to an opaque structure holding a plan to execute a dynamic persistent RNN.cudnnCreatePersistentRNNPlan() is used to create and initialize one instance.

cudnnPoolingDescriptor

cudnnPoolingDescriptor is a pointer to an opaque structure holding the description of a pooling operation.

cudnnReduceTensorDescriptor

cudnnReduceTensorDescriptor_t is a pointer to an opaque structure holding the description of a tensor reduction operation, used as a parameter to cudnnReduceTensor(). cudnnCreateReduceTensorDescriptor() is used to create one instance, and cudnnSetReduceTensorDescriptor() must be used to initialize this instance.

cudnnRNNDescriptor

cudnnRNNDescriptor_t is a pointer to an opaque structure holding the description of an RNN operation. cudnnCreateRNNDescriptor() is used to create one instance, and cudnnSetRNNDescriptor() must be used to initialize this instance.

cudnnRuntimeTag

Forward definition in this version only

cudnnSpatialTransformerDescriptor

cudnnSpatialTransformerDescriptor_t is a pointer to an opaque structure holding the description of a spatial transformation operation. cudnnCreateSpatialTransformerDescriptor() is used to create one instance, cudnnSetSpatialTransformerNdDescriptor() is used to initialize this instance, cudnnDestroySpatialTransformerDescriptor() is used to destroy this instance.

cudnnTensorDescriptor

cudnnCreateTensorDescriptor is a pointer to an opaque structure holding the description of a generic n-D dataset.

LRNConstants

Constants for LRN, #define in cudnn.h

Enums

cudnnActivationMode

cudnnActivationMode is an enumerated type used to select the neuron activation function used in cudnnActivationForward() and cudnnActivationBackward().

cudnnAddMode

cudnnAddMode is an enumerated type used by cudnnAddTensor() to specify how a bias tensor is added to an input/output tensor.

cudnnBatchNormMode

cudnnBatchNormMode is an enumerated type used to specify the mode of operation in cudnnBatchNormalizationForwardInference(), cudnnBatchNormalizationForwardTraining(), cudnnBatchNormalizationBackward() and cudnnDeriveBNTensorDescriptor() routines.

cudnnConvolutionBwdDataAlgo

cudnnConvolutionBwdDataAlgo is an enumerated type that exposes the different algorithms available to execute the backward data convolution operation.

cudnnConvolutionBwdDataPreference

cudnnConvolutionBwdDataPreference is an enumerated type used by cudnnGetConvolutionBackwardDataAlgorithm() to help the choice of the algorithm used for the backward data convolution.

cudnnConvolutionBwdFilterAlgo

cudnnConvolutionBwdFilterAlgo is an enumerated type that exposes the different algorithms available to execute the backward filter convolution operation.

cudnnConvolutionBwdFilterPreference

cudnnConvolutionBwdFilterPreference is an enumerated type used by cudnnGetConvolutionBackwardFilterAlgorithm() to help the choice of the algorithm used for the backward filter convolution.

cudnnConvolutionFwdAlgo

cudnnConvolutionFwdAlgo is an enumerated type that exposes the different algorithms available to execute the forward convolution operation.

cudnnConvolutionFwdPreference

cudnnConvolutionFwdPreference is an enumerated type used by cudnnGetConvolutionForwardAlgorithm() to help the choice of the algorithm used for the forward convolution.

cudnnConvolutionMode

cudnnConvolutionMode is an enumerated type used by cudnnSetConvolutionDescriptor() to configure a convolution descriptor.

cudnnCTCLossAlgo

cudnnDataType

cudnnDataType is an enumerated type indicating the data type to which a tensor descriptor or filter descriptor refers.

cudnnDeterminism

cudnnDeterminism_t is an enumerated type used to indicate if the computed results are deterministic(reproducible). See section 2.5 (Reproducibility) for more details on determinism.

cudnnDirectionMode

cudnnDirectionMode_t is an enumerated type used to specify the recurrence pattern in the cudnnRNNForwardInference(), cudnnRNNForwardTraining(), cudnnRNNBackwardData() and cudnnRNNBackwardWeights() routines.

cudnnDivNormMode

cudnnDivNormMode is an enumerated type used to specify the mode of operation in cudnnDivisiveNormalizationForward() and cudnnDivisiveNormalizationBackward().

cudnnErrQueryMode

cudnnIndicesType

cudnnIndicesType_t is an enumerated type used to indicate the data type for the indices to be computed by the cudnnReduceTensor() routine. This enumerated type is used as a field for the cudnnReduceTensorDescriptor_t descriptor.

cudnnLRNMode

cudnnLRNMode is an enumerated type used to specify the mode of operation in cudnnLRNCrossChannelForward() and cudnnLRNCrossChannelBackward().

cudnnMathType

CUDNN math type

cudnnNanPropagation

cudnnNanPropagation is an enumerated type for the NanPropagation flag.

cudnnOpTensorOp

cudnnOpTensorOp is an enumerated type used to indicate the tensor operation to be used by the cudnnOpTensor() routine. This enumerated type is used as a field for the cudnnOpTensorDescriptor descriptor.

cudnnPoolingMode

cudnnPoolingMode is an enumerated type passed to cudnnSetPoolingDescriptor() to select the pooling method to be used by cudnnPoolingForward() and cudnnPoolingBackward().

cudnnReduceTensorIndices

cudnnReduceTensorIndices_t is an enumerated type used to indicate whether indices are to be computed by the cudnnReduceTensor() routine.This enumerated type is used as a field for the cudnnReduceTensorDescriptor_t descriptor.

cudnnReduceTensorOp

cudnnReduceTensorOp is an enumerated type used to indicate the tensor operation to be used by the cudnnReduceTensor() routine.This enumerated type is used as a field for the cudnnReduceTensorDescriptor_t descriptor.

cudnnRNNAlgo

cudnnRNNAlgo_t is an enumerated type used to specify the algorithm used in the cudnnRNNForwardInference(), cudnnRNNForwardTraining(), cudnnRNNBackwardData() and cudnnRNNBackwardWeights() routines.

cudnnRNNInputMode

cudnnRNNInputMode_t is an enumerated type used to specify the behavior of the first layer in the cudnnRNNForwardInference(), cudnnRNNForwardTraining(), cudnnRNNBackwardData() and cudnnRNNBackwardWeights() routines.

cudnnRNNMode

cudnnRNNMode_t is an enumerated type used to specify the type of network used in the cudnnRNNForwardInference(), cudnnRNNForwardTraining(), cudnnRNNBackwardData() and cudnnRNNBackwardWeights() routines.

cudnnSamplerType

cudnnSamplerType_t is an enumerated type passed to cudnnSetSpatialTransformerNdDescriptor() to select the sampler type to be used by cudnnSpatialTfSamplerForward() and cudnnSpatialTfSamplerBackward().

cudnnSoftmaxAlgorithm

cudnnSoftmaxAlgorithm is used to select an implementation of the softmax function used in cudnnSoftmaxForward() and cudnnSoftmaxBackward().

cudnnSoftmaxMode

cudnnSoftmaxMode is used to select over which data the cudnnSoftmaxForward() and cudnnSoftmaxBackward() are computing their results.

cudnnStatus

CUDNN return codes

cudnnTensorFormat

cudnnTensorFormat is an enumerated type used by cudnnSetTensor4dDescriptor() to create a tensor with a pre-defined layout.