Dnnl_api_prelu

Collaboration diagram for Dnnl_api_prelu:

Classes
struct	dnnl_prelu_desc_t

Functions
dnnl_status_t DNNL_API	dnnl_prelu_forward_desc_init (dnnl_prelu_desc_t *prelu_desc, dnnl_prop_kind_t prop_kind, const dnnl_memory_desc_t *data_desc, const dnnl_memory_desc_t *weights_desc)
dnnl_status_t DNNL_API	dnnl_prelu_backward_desc_init (dnnl_prelu_desc_t *prelu_desc, const dnnl_memory_desc_t *data_desc, const dnnl_memory_desc_t *weights_desc, const dnnl_memory_desc_t *diff_data_desc, const dnnl_memory_desc_t *diff_weights_desc)

Detailed Description

Function Documentation

dnnl_prelu_backward_desc_init()

dnnl_status_t DNNL_API dnnl_prelu_backward_desc_init	(	dnnl_prelu_desc_t *	prelu_desc,
		const dnnl_memory_desc_t *	data_desc,
		const dnnl_memory_desc_t *	weights_desc,
		const dnnl_memory_desc_t *	diff_data_desc,
		const dnnl_memory_desc_t *	diff_weights_desc
	)

Initializes a descriptor for PReLU (leaky ReLU with trainable alpha parameter) backward propagation primitive.

Note

weights descriptor and diff_weights descriptor are allowed to be initialized with dnnl_format_tag_any or with format_kind set to dnnl_format_kind_any.

Parameters

prelu_desc	Output descriptor for a prelu primitive.
data_desc	Source and destination memory descriptor.
weights_desc	Alpha parameters memory descriptor.
diff_data_desc	Diff source and destination memory descriptor.
diff_weights_desc	Diff alpha parameters memory descriptor.

Returns

dnnl_success on success and a status describing the error otherwise.

dnnl_prelu_forward_desc_init()

dnnl_status_t DNNL_API dnnl_prelu_forward_desc_init	(	dnnl_prelu_desc_t *	prelu_desc,
		dnnl_prop_kind_t	prop_kind,
		const dnnl_memory_desc_t *	data_desc,
		const dnnl_memory_desc_t *	weights_desc
	)

Initializes a descriptor for PReLU (leaky ReLU with trainable alpha parameter) forward propagation primitive.

Note

weights descriptor is allowed to be initialized with dnnl_format_tag_any or with format_kind set to dnnl_format_kind_any.

Parameters

prelu_desc	Output descriptor for a prelu primitive.
prop_kind	Propagation kind. Possible values are dnnl_forward_training and dnnl_forward_inference.
data_desc	Source and destination memory descriptor.
weights_desc	Alpha parameters memory descriptor.

Returns

dnnl_success on success and a status describing the error otherwise.