mxnet
|
A descriptor of a Softmax operation. More...
#include <dnnl_types.h>
Public Attributes | |
dnnl_primitive_kind_t | primitive_kind |
dnnl_prop_kind_t | prop_kind |
dnnl_memory_desc_t | src_desc |
Source memory descriptor. More... | |
dnnl_memory_desc_t | diff_src_desc |
Source gradient memory descriptor. More... | |
int | softmax_axis |
The axis along which to perform the softmax. More... | |
dnnl_alg_kind_t | alg_kind |
dnnl_memory_desc_t | dst_desc |
Destination memory descriptor. More... | |
dnnl_memory_desc_t | diff_dst_desc |
Destination gradient memory descriptor. More... | |
A descriptor of a Softmax operation.
dnnl_alg_kind_t dnnl_softmax_v2_desc_t::alg_kind |
Softmax algorithm. Possible values: dnnl_softmax_accurate and dnnl_softmax_log.
dnnl_memory_desc_t dnnl_softmax_v2_desc_t::diff_dst_desc |
Destination gradient memory descriptor.
dnnl_memory_desc_t dnnl_softmax_v2_desc_t::diff_src_desc |
Source gradient memory descriptor.
dnnl_memory_desc_t dnnl_softmax_v2_desc_t::dst_desc |
Destination memory descriptor.
dnnl_primitive_kind_t dnnl_softmax_v2_desc_t::primitive_kind |
The kind of primitive. Used for self-identifying the primitive descriptor. Must be dnnl_softmax_v2.
dnnl_prop_kind_t dnnl_softmax_v2_desc_t::prop_kind |
The kind of propagation. Possible values: dnnl_forward_training, dnnl_forward_inference, and dnnl_backward_data.
int dnnl_softmax_v2_desc_t::softmax_axis |
The axis along which to perform the softmax.
dnnl_memory_desc_t dnnl_softmax_v2_desc_t::src_desc |
Source memory descriptor.