oneAPI Deep Neural Network Library (oneDNN)
Performance library for Deep Learning
1.96.0
dnnl::convolution_forward::desc Struct Reference

Descriptor for a convolution forward propagation primitive. More...

#include <dnnl.hpp>

Collaboration diagram for dnnl::convolution_forward::desc:

Public Member Functions

 desc (prop_kind aprop_kind, algorithm aalgorithm, const memory::desc &src_desc, const memory::desc &weights_desc, const memory::desc &bias_desc, const memory::desc &dst_desc, const memory::dims &strides, const memory::dims &padding_l, const memory::dims &padding_r)
 Constructs a descriptor for a convolution forward propagation primitive with bias. More...
 
 desc (prop_kind aprop_kind, algorithm aalgorithm, const memory::desc &src_desc, const memory::desc &weights_desc, const memory::desc &dst_desc, const memory::dims &strides, const memory::dims &padding_l, const memory::dims &padding_r)
 Constructs a descriptor for a convolution forward propagation primitive without bias. More...
 
 desc (prop_kind aprop_kind, algorithm aalgorithm, const memory::desc &src_desc, const memory::desc &weights_desc, const memory::desc &bias_desc, const memory::desc &dst_desc, const memory::dims &strides, const memory::dims &dilates, const memory::dims &padding_l, const memory::dims &padding_r)
 Constructs a descriptor for a dilated convolution forward propagation primitive with bias. More...
 
 desc (prop_kind aprop_kind, algorithm aalgorithm, const memory::desc &src_desc, const memory::desc &weights_desc, const memory::desc &dst_desc, const memory::dims &strides, const memory::dims &dilates, const memory::dims &padding_l, const memory::dims &padding_r)
 Constructs a descriptor for a dilated convolution forward propagation primitive without bias. More...
 

Detailed Description

Descriptor for a convolution forward propagation primitive.

Examples:
cnn_inference_f32.cpp, cnn_inference_int8.cpp, cnn_training_bf16.cpp, cnn_training_f32.cpp, convolution.cpp, and performance_profiling.cpp.

Constructor & Destructor Documentation

◆ desc() [1/4]

dnnl::convolution_forward::desc::desc ( prop_kind  aprop_kind,
algorithm  aalgorithm,
const memory::desc src_desc,
const memory::desc weights_desc,
const memory::desc bias_desc,
const memory::desc dst_desc,
const memory::dims strides,
const memory::dims padding_l,
const memory::dims padding_r 
)
inline

Constructs a descriptor for a convolution forward propagation primitive with bias.

Note
All the memory descriptors may be initialized with the dnnl::memory::format_tag::any value of format_tag.

Arrays strides, padding_l, and padding_r contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width.

Parameters
aprop_kindPropagation kind. Possible values are dnnl::prop_kind::forward_training, and dnnl::prop_kind::forward_inference.
aalgorithmConvolution algorithm. Possible values are dnnl::algorithm::convolution_direct, dnnl::algorithm::convolution_winograd, and dnnl::algorithm::convolution_auto.
src_descSource memory descriptor.
weights_descWeights memory descriptor.
bias_descBias memory descriptor. Passing zero memory descriptor disables the bias term.
dst_descDestination memory descriptor.
stridesStrides for each spatial dimension.
padding_lVector of padding values for low indices for each spatial dimension ([[front,] top,] left).
padding_rVector of padding values for high indices for each spatial dimension ([[back,] bottom,] right).

◆ desc() [2/4]

dnnl::convolution_forward::desc::desc ( prop_kind  aprop_kind,
algorithm  aalgorithm,
const memory::desc src_desc,
const memory::desc weights_desc,
const memory::desc dst_desc,
const memory::dims strides,
const memory::dims padding_l,
const memory::dims padding_r 
)
inline

Constructs a descriptor for a convolution forward propagation primitive without bias.

Note
All the memory descriptors may be initialized with the dnnl::memory::format_tag::any value of format_tag.

Arrays strides, padding_l, and padding_r contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width.

Parameters
aprop_kindPropagation kind. Possible values are dnnl::prop_kind::forward_training, and dnnl::prop_kind::forward_inference.
aalgorithmConvolution algorithm. Possible values are dnnl::algorithm::convolution_direct, dnnl::algorithm::convolution_winograd, and dnnl::algorithm::convolution_auto.
src_descSource memory descriptor.
weights_descWeights memory descriptor.
dst_descDestination memory descriptor.
stridesStrides for each spatial dimension.
padding_lVector of padding values for low indices for each spatial dimension ([[front,] top,] left).
padding_rVector of padding values for high indices for each spatial dimension ([[back,] bottom,] right).

◆ desc() [3/4]

dnnl::convolution_forward::desc::desc ( prop_kind  aprop_kind,
algorithm  aalgorithm,
const memory::desc src_desc,
const memory::desc weights_desc,
const memory::desc bias_desc,
const memory::desc dst_desc,
const memory::dims strides,
const memory::dims dilates,
const memory::dims padding_l,
const memory::dims padding_r 
)
inline

Constructs a descriptor for a dilated convolution forward propagation primitive with bias.

Note
All the memory descriptors may be initialized with the dnnl::memory::format_tag::any value of format_tag.

Arrays strides, dilates, padding_l, and padding_r contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width.

Parameters
aprop_kindPropagation kind. Possible values are dnnl::prop_kind::forward_training, and dnnl::prop_kind::forward_inference.
aalgorithmConvolution algorithm. Possible values are dnnl::algorithm::convolution_direct, dnnl::algorithm::convolution_winograd, and dnnl::algorithm::convolution_auto.
src_descSource memory descriptor.
weights_descWeights memory descriptor.
bias_descBias memory descriptor. Passing zero memory descriptor disables the bias term.
dst_descDestination memory descriptor.
stridesStrides for each spatial dimension.
dilatesDilations for each spatial dimension. A zero value means no dilation in the corresponding dimension.
padding_lVector of padding values for low indices for each spatial dimension ([[front,] top,] left).
padding_rVector of padding values for high indices for each spatial dimension ([[back,] bottom,] right).

◆ desc() [4/4]

dnnl::convolution_forward::desc::desc ( prop_kind  aprop_kind,
algorithm  aalgorithm,
const memory::desc src_desc,
const memory::desc weights_desc,
const memory::desc dst_desc,
const memory::dims strides,
const memory::dims dilates,
const memory::dims padding_l,
const memory::dims padding_r 
)
inline

Constructs a descriptor for a dilated convolution forward propagation primitive without bias.

Note
All the memory descriptors may be initialized with the dnnl::memory::format_tag::any value of format_tag.

Arrays strides, dilates, padding_l, and padding_r contain values for spatial dimensions only and hence must have the same number of elements as there are spatial dimensions. The order of values is the same as in the tensor: depth (for 3D tensors), height (for 3D and 2D tensors), and width.

Parameters
aprop_kindPropagation kind. Possible values are dnnl::prop_kind::forward_training, and dnnl::prop_kind::forward_inference.
aalgorithmConvolution algorithm. Possible values are dnnl::algorithm::convolution_direct, dnnl::algorithm::convolution_winograd, and dnnl::algorithm::convolution_auto.
src_descSource memory descriptor.
weights_descWeights memory descriptor.
dst_descDestination memory descriptor.
stridesStrides for each spatial dimension.
dilatesDilations for each spatial dimension. A zero value means no dilation in the corresponding dimension.
padding_lVector of padding values for low indices for each spatial dimension ([[front,] top,] left).
padding_rVector of padding values for high indices for each spatial dimension ([[back,] bottom,] right).

The documentation for this struct was generated from the following file: