dnnl::convolution_backward_weights::desc Struct Reference

Descriptor for a convolution weights gradient primitive. More...

#include <dnnl.hpp>

Collaboration diagram for dnnl::convolution_backward_weights::desc:

Public Member Functions
	desc (algorithm algorithm, const memory::desc &src_desc, const memory::desc &diff_weights_desc, const memory::desc &diff_bias_desc, const memory::desc &diff_dst_desc, const memory::dims &strides, const memory::dims &padding_l, const memory::dims &padding_r)
	Constructs a descriptor for a convolution weights gradient primitive with bias. More...
	desc (algorithm algorithm, const memory::desc &src_desc, const memory::desc &diff_weights_desc, const memory::desc &diff_dst_desc, const memory::dims &strides, const memory::dims &padding_l, const memory::dims &padding_r)
	Constructs a descriptor for a convolution weights gradient primitive without bias. More...
	desc (algorithm algorithm, const memory::desc &src_desc, const memory::desc &diff_weights_desc, const memory::desc &diff_bias_desc, const memory::desc &diff_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 weights gradient primitive with bias. More...
	desc (algorithm algorithm, const memory::desc &src_desc, const memory::desc &diff_weights_desc, const memory::desc &diff_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 weights gradient primitive without bias. More...

Detailed Description

Descriptor for a convolution weights gradient primitive.

Examples:

cnn_training_f32.cpp, and cpu_cnn_training_bf16.cpp.

Constructor & Destructor Documentation

desc() [1/4]

dnnl::convolution_backward_weights::desc::desc ( algorithm  algorithm, const memory::desc &  src_desc, const memory::desc &  diff_weights_desc, const memory::desc &  diff_bias_desc, const memory::desc &  diff_dst_desc, const memory::dims &  strides, const memory::dims &  padding_l, const memory::dims &  padding_r  )

inline

Constructs a descriptor for a convolution weights gradient primitive with bias.

Note

Memory descriptors can be initialized with dnnl::memory::format_tag::any value of format_tag.

Inputs:

• src (dnnl::primitive_desc_base::src_desc (0))
• diff_dst (dnnl::primitive_desc_base::diff_dst_desc (0))

Outputs:

• diff_weights (dnnl::primitive_desc_base::diff_weights_desc (0))
• diff_bias (dnnl::primitive_desc_base::diff_weights_desc (1))

Parameters

algorithm	Convolution algorithm. Possible values are dnnl::algorithm::convolution_direct, dnnl::algorithm::convolution_winograd, and dnnl::algorithm::convolution_auto.
src_desc	Source memory descriptor.
diff_weights_desc	Diff weights memory descriptor.
diff_bias_desc	Diff bias memory descriptor. Passing zero memory descriptor disables the bias term.
diff_dst_desc	Diff destination memory descriptor.
strides	Strides for each spatial dimension.
padding_l	Vector of padding values for low indices for each spatial dimension (front, top, left).
padding_r	Vector of padding values for high indices for each spatial dimension (back, bottom, right).

desc() [2/4]

dnnl::convolution_backward_weights::desc::desc ( algorithm  algorithm, const memory::desc &  src_desc, const memory::desc &  diff_weights_desc, const memory::desc &  diff_dst_desc, const memory::dims &  strides, const memory::dims &  padding_l, const memory::dims &  padding_r  )

inline

Constructs a descriptor for a convolution weights gradient primitive without bias.

Note

Memory descriptors can be initialized with dnnl::memory::format_tag::any value of format_tag.

Inputs:

• src (dnnl::primitive_desc_base::src_desc (0))
• diff_dst (dnnl::primitive_desc_base::diff_dst_desc (0))

Outputs:

• diff_weights (dnnl::primitive_desc_base::diff_weights_desc (0))

Parameters

algorithm	Convolution algorithm. Possible values are dnnl::algorithm::convolution_direct, dnnl::algorithm::convolution_winograd, and dnnl::algorithm::convolution_auto.
src_desc	Source memory descriptor.
diff_weights_desc	Diff weights memory descriptor.
diff_dst_desc	Diff destination memory descriptor.
strides	Strides for each spatial dimension.
padding_l	Vector of padding values for low indices for each spatial dimension (front, top, left).
padding_r	Vector of padding values for high indices for each spatial dimension (back, bottom, right).

desc() [3/4]

dnnl::convolution_backward_weights::desc::desc ( algorithm  algorithm, const memory::desc &  src_desc, const memory::desc &  diff_weights_desc, const memory::desc &  diff_bias_desc, const memory::desc &  diff_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 weights gradient primitive with bias.

Note

Memory descriptors can be initialized with dnnl::memory::format_tag::any value of format_tag.

Inputs:

• src (dnnl::primitive_desc_base::src_desc (0))
• diff_dst (dnnl::primitive_desc_base::diff_dst_desc (0))

Outputs:

• diff_weights (dnnl::primitive_desc_base::diff_weights_desc (0))
• diff_bias (dnnl::primitive_desc_base::diff_weights_desc (1))

Parameters

algorithm	Convolution algorithm. Possible values are dnnl::algorithm::convolution_direct, dnnl::algorithm::convolution_winograd, and dnnl::algorithm::convolution_auto.
src_desc	Source memory descriptor.
diff_weights_desc	Diff weights memory descriptor.
diff_bias_desc	Diff bias memory descriptor. Passing zero memory descriptor disables the bias term.
diff_dst_desc	Diff destination memory descriptor.
strides	Strides for each spatial dimension.
dilates	Dilations for each spatial dimension. A zero value means no dilation in the corresponding dimension.
padding_l	Vector of padding values for low indices for each spatial dimension (front, top, left).
padding_r	Vector of padding values for high indices for each spatial dimension (back, bottom, right).

desc() [4/4]

dnnl::convolution_backward_weights::desc::desc ( algorithm  algorithm, const memory::desc &  src_desc, const memory::desc &  diff_weights_desc, const memory::desc &  diff_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 weights gradient primitive without bias.

Note

Memory descriptors can be initialized with dnnl::memory::format_tag::any value of format_tag.

Inputs:

• src (dnnl::primitive_desc_base::src_desc (0))
• diff_dst (dnnl::primitive_desc_base::diff_dst_desc (0))

Outputs:

• diff_weights (dnnl::primitive_desc_base::diff_weights_desc (0))

Parameters

algorithm	Convolution algorithm. Possible values are dnnl::algorithm::convolution_direct, dnnl::algorithm::convolution_winograd, and dnnl::algorithm::convolution_auto.
src_desc	Source memory descriptor.
diff_weights_desc	Diff weights memory descriptor.
diff_dst_desc	Diff destination memory descriptor.
strides	Strides for each spatial dimension.
dilates	Dilations for each spatial dimension. A zero value means no dilation in the corresponding dimension.
padding_l	Vector of padding values for low indices for each spatial dimension (front, top, left).
padding_r	Vector of padding values for high indices for each spatial dimension (back, bottom, right).

---

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

• include/dnnl.hpp