Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN)  0.21.0
Performance library for Deep Learning
Public Attributes | List of all members
mkldnn_rnn_cell_desc_t Struct Reference

#include <mkldnn_types.h>

Public Attributes

mkldnn_alg_kind_t cell_kind
 RNN cell kind. More...
 
mkldnn_alg_kind_t activation_kind
 Activation function used. More...
 
unsigned int flags
 RNN cell flags. More...
 
float alpha
 alpha is a negative slope parameter (used only if (flags & mkldnn_rnn_cell_with_relu) != 0) More...
 
float clipping
 clipping parameter (used only if (flags & mkldnn_rnn_cell_with_clipping) != 0) More...
 

Member Data Documentation

◆ cell_kind

mkldnn_alg_kind_t mkldnn_rnn_cell_desc_t::cell_kind

◆ activation_kind

mkldnn_alg_kind_t mkldnn_rnn_cell_desc_t::activation_kind

Activation function used.

Must be either mkldnn_eltwise_relu or mkldnn_eltwise_tanh.

◆ flags

unsigned int mkldnn_rnn_cell_desc_t::flags

RNN cell flags.

◆ alpha

float mkldnn_rnn_cell_desc_t::alpha

alpha is a negative slope parameter (used only if (flags & mkldnn_rnn_cell_with_relu) != 0)

◆ clipping

float mkldnn_rnn_cell_desc_t::clipping

clipping parameter (used only if (flags & mkldnn_rnn_cell_with_clipping) != 0)


The documentation for this struct was generated from the following file: