dnnl::inner_product_forward::desc Struct Reference

Descriptor for an inner product forward propagation primitive. More...

#include <dnnl.hpp>

Collaboration diagram for dnnl::inner_product_forward::desc:

Public Member Functions
	desc (prop_kind prop_kind, const memory::desc &src_desc, const memory::desc &weights_desc, const memory::desc &bias_desc, const memory::desc &dst_desc)
	Constructs a descriptor for an inner product forward propagation primitive with bias.
	desc (prop_kind prop_kind, const memory::desc &src_desc, const memory::desc &weights_desc, const memory::desc &dst_desc)
	Constructs a descriptor for an inner product forward propagation primitive without bias.

Detailed Description

Descriptor for an inner product forward propagation primitive.

Examples

cnn_inference_f32.cpp, and inner_product.cpp.

Constructor & Destructor Documentation

desc() [1/2]

dnnl::inner_product_forward::desc::desc ( prop_kind prop_kind, const memory::desc & src_desc, const memory::desc & weights_desc, const memory::desc & bias_desc, const memory::desc & dst_desc )

inline

Constructs a descriptor for an inner product forward propagation primitive with bias.

Inputs:

• src (dnnl::primitive_desc_base::src_desc(0))
• weights (dnnl::primitive_desc_base::weights_desc(0))
• bias (dnnl::primitive_desc_base::weights_desc(1))

Outputs:

• dst (dnnl::primitive_desc_base::dst_desc(0))

Note

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

Parameters

prop_kind	Propagation kind. Possible values are dnnl::prop_kind::forward_training, and dnnl::prop_kind::forward_inference.
src_desc	Memory descriptor for src.
weights_desc	Memory descriptor for diff weights.
bias_desc	Memory descriptor for diff bias.
dst_desc	Memory descriptor for diff dst.

desc() [2/2]

dnnl::inner_product_forward::desc::desc ( prop_kind prop_kind, const memory::desc & src_desc, const memory::desc & weights_desc, const memory::desc & dst_desc )

inline

Constructs a descriptor for an inner product forward propagation primitive without bias.

Inputs:

• src (dnnl::primitive_desc_base::src_desc(0))
• weights (dnnl::primitive_desc_base::weights_desc(0))

Outputs:

• dst (dnnl::primitive_desc_base::dst_desc(0))

Note

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

Parameters

prop_kind	Propagation kind. Possible values are dnnl::prop_kind::forward_training, and dnnl::prop_kind::forward_inference.
src_desc	Memory descriptor for src.
weights_desc	Memory descriptor for diff weights.
dst_desc	Memory descriptor for dst.

---

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

• include/dnnl.hpp