Pooling

Collaboration diagram for Pooling:

Functions
mkldnn_status_t MKLDNN_API	mkldnn_pooling_forward_desc_init (mkldnn_pooling_desc_t *pool_desc, mkldnn_prop_kind_t prop_kind, mkldnn_alg_kind_t alg_kind, const mkldnn_memory_desc_t *src_desc, const mkldnn_memory_desc_t *dst_desc, const mkldnn_dims_t strides, const mkldnn_dims_t kernel, const mkldnn_dims_t padding_l, const mkldnn_dims_t padding_r)
mkldnn_status_t MKLDNN_API	mkldnn_pooling_backward_desc_init (mkldnn_pooling_desc_t *pool_desc, mkldnn_alg_kind_t alg_kind, const mkldnn_memory_desc_t *diff_src_desc, const mkldnn_memory_desc_t *diff_dst_desc, const mkldnn_dims_t strides, const mkldnn_dims_t kernel, const mkldnn_dims_t padding_l, const mkldnn_dims_t padding_r)

Detailed Description

A primitive to perform max or average pooling.

See also

dev_guide_pooling in developer guide

cpp_api_pooling in cpp_api

Function Documentation

mkldnn_status_t MKLDNN_API mkldnn_pooling_backward_desc_init	(	mkldnn_pooling_desc_t *	pool_desc,
		mkldnn_alg_kind_t	alg_kind,
		const mkldnn_memory_desc_t *	diff_src_desc,
		const mkldnn_memory_desc_t *	diff_dst_desc,
		const mkldnn_dims_t	strides,
		const mkldnn_dims_t	kernel,
		const mkldnn_dims_t	padding_l,
		const mkldnn_dims_t	padding_r
	)

Initializes a pooling descriptor pool_desc for backward propagation using alg_kind, memory descriptors, and pooling parameters in the spatial domain: strides, kernel sizes, padding_l, and padding_r.

Note

If padding_r is NULL, the padding is supposed to be symmetric.

Inputs:

• diff_dst (mkldnn_query_diff_dst_md, 0)
• workspace (mkldnn_query_workspace_md, 0), if alg_kind = mkldnn_pooling_max

Outputs:

• diff_src (mkldnn_query_diff_src_md, 0)

mkldnn_status_t MKLDNN_API mkldnn_pooling_forward_desc_init	(	mkldnn_pooling_desc_t *	pool_desc,
		mkldnn_prop_kind_t	prop_kind,
		mkldnn_alg_kind_t	alg_kind,
		const mkldnn_memory_desc_t *	src_desc,
		const mkldnn_memory_desc_t *	dst_desc,
		const mkldnn_dims_t	strides,
		const mkldnn_dims_t	kernel,
		const mkldnn_dims_t	padding_l,
		const mkldnn_dims_t	padding_r
	)

Initializes a pooling descriptor pool_desc for forward propagation using prop_kind (possible values are mkldnn_forward_training and mkldnn_forward_inference), alg_kind, memory descriptors, and pooling parameters in the spatial domain: strides, kernel sizes, padding_l, and padding_r.

Note

If padding_r is NULL, the padding is supposed to be symmetric.

Inputs:

• src (mkldnn_query_src_md, 0)

Outputs:

• dst (mkldnn_query_dst_md, 0)
• workspace (mkldnn_query_workspace_md, 0), if alg_kind = mkldnn_pooling_max and prop_kind = mkldnn_forward_training