Developer Guide and Reference

  • 2022.1
  • 04/11/2022
  • Public Content
Contents

struct dnnl::inner_product_forward::desc

Overview

Descriptor for an inner product forward propagation primitive. More…
#include <dnnl.hpp> struct desc { // fields dnnl_inner_product_desc_t data; // construction desc( prop_kind aprop_kind, const memory::desc& src_desc, const memory::desc& weights_desc, const memory::desc& bias_desc, const memory::desc& dst_desc ); desc( prop_kind aprop_kind, const memory::desc& src_desc, const memory::desc& weights_desc, const memory::desc& dst_desc ); };

Detailed Documentation

Descriptor for an inner product forward propagation primitive.
Construction
desc( prop_kind aprop_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.
All the memory descriptors may be initialized with the dnnl::memory::format_tag::any value of
format_tag
.
Parameters:
aprop_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 weights.
bias_desc
Memory descriptor for bias.
dst_desc
Memory descriptor for dst.
desc( prop_kind aprop_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.
All the memory descriptors may be initialized with the dnnl::memory::format_tag::any value of
format_tag
.
Parameters:
aprop_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 weights.
dst_desc
Memory descriptor for dst.

Product and Performance Information

1

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.