Deep Neural Network Library (DNNL)  1.1.3
Performance library for Deep Learning
Functions
Inner product

A primitive to compute an inner product. More...

Functions

dnnl_status_t DNNL_API dnnl_inner_product_forward_desc_init (dnnl_inner_product_desc_t *ip_desc, dnnl_prop_kind_t prop_kind, const dnnl_memory_desc_t *src_desc, const dnnl_memory_desc_t *weights_desc, const dnnl_memory_desc_t *bias_desc, const dnnl_memory_desc_t *dst_desc)
 Initializes an inner product descriptor ip_desc for forward propagation using prop_kind (possible values are dnnl_forward_training and dnnl_forward_inference) and memory descriptors. More...
 
dnnl_status_t DNNL_API dnnl_inner_product_backward_data_desc_init (dnnl_inner_product_desc_t *ip_desc, const dnnl_memory_desc_t *diff_src_desc, const dnnl_memory_desc_t *weights_desc, const dnnl_memory_desc_t *diff_dst_desc)
 Initializes an inner product descriptor ip_desc for backward propagation with respect to data using memory descriptors. More...
 
dnnl_status_t DNNL_API dnnl_inner_product_backward_weights_desc_init (dnnl_inner_product_desc_t *ip_desc, const dnnl_memory_desc_t *src_desc, const dnnl_memory_desc_t *diff_weights_desc, const dnnl_memory_desc_t *diff_bias_desc, const dnnl_memory_desc_t *diff_dst_desc)
 Initializes an inner product descriptor ip_desc for backward propagation with respect to weights using memory descriptors. More...
 

Detailed Description

A primitive to compute an inner product.

See also
Inner Product in developer guide
Inner Product in C++ API

Function Documentation

◆ dnnl_inner_product_forward_desc_init()

dnnl_status_t DNNL_API dnnl_inner_product_forward_desc_init ( dnnl_inner_product_desc_t ip_desc,
dnnl_prop_kind_t  prop_kind,
const dnnl_memory_desc_t src_desc,
const dnnl_memory_desc_t weights_desc,
const dnnl_memory_desc_t bias_desc,
const dnnl_memory_desc_t dst_desc 
)

Initializes an inner product descriptor ip_desc for forward propagation using prop_kind (possible values are dnnl_forward_training and dnnl_forward_inference) and memory descriptors.

In order to create an inner product without bias, bias_desc should be either NULL or a pointer to a descriptor with memory format kind equals dnnl_format_kind_undef.

Note
Memory descriptors are allowed to be initialized with dnnl_format_kind_any value of format_kind.

Inputs:

Outputs:

◆ dnnl_inner_product_backward_data_desc_init()

dnnl_status_t DNNL_API dnnl_inner_product_backward_data_desc_init ( dnnl_inner_product_desc_t ip_desc,
const dnnl_memory_desc_t diff_src_desc,
const dnnl_memory_desc_t weights_desc,
const dnnl_memory_desc_t diff_dst_desc 
)

Initializes an inner product descriptor ip_desc for backward propagation with respect to data using memory descriptors.

Note
Memory descriptors are allowed to be initialized with dnnl_format_kind_any value of format_kind.

Inputs:

Outputs:

◆ dnnl_inner_product_backward_weights_desc_init()

dnnl_status_t DNNL_API dnnl_inner_product_backward_weights_desc_init ( dnnl_inner_product_desc_t ip_desc,
const dnnl_memory_desc_t src_desc,
const dnnl_memory_desc_t diff_weights_desc,
const dnnl_memory_desc_t diff_bias_desc,
const dnnl_memory_desc_t diff_dst_desc 
)

Initializes an inner product descriptor ip_desc for backward propagation with respect to weights using memory descriptors.

Note
Memory descriptors are allowed to be initialized with dnnl_format_kind_any value of format_kind.

Inputs:

Outputs: