|
mxnet
|
A descriptor for an RNN operation. More...
#include <dnnl_types.h>
A descriptor for an RNN operation.
| dnnl_alg_kind_t dnnl_rnn_desc_t::activation_kind |
Activation function used for vanilla_rnn cell kind. Must be either dnnl_eltwise_relu or dnnl_eltwise_tanh.
| float dnnl_rnn_desc_t::alpha |
| float dnnl_rnn_desc_t::beta |
| dnnl_memory_desc_t dnnl_rnn_desc_t::bias_desc |
Bias memory descriptor.
| dnnl_alg_kind_t dnnl_rnn_desc_t::cell_kind |
RNN cell kind. Must be one of dnnl_vanilla_rnn, dnnl_vanilla_lstm, dnnl_vanilla_gru, or dnnl_lbr_gru.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_bias_desc |
Bias gradient memory descriptor.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_dst_iter_c_desc |
Destination gradient iteration memory descriptor for cell state.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_dst_iter_desc |
Destination gradient iteration memory descriptor for hidden state.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_dst_layer_desc |
Destination gradient layer memory descriptor.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_src_iter_c_desc |
Source gradient iter memory descriptor for cell state.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_src_iter_desc |
Source gradient iter memory descriptor for hidden state.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_src_layer_desc |
Source gradient layer memory descriptor.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_weights_iter_desc |
Weights gradient iter memory descriptor.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_weights_layer_desc |
Weights gradient layer memory descriptor.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_weights_peephole_desc |
Weights gradient peephole memory descriptor. This memory descriptor is equal to zero memory descriptor in case of non-peephole LSTMs and other non-LSTM RNNs.
| dnnl_memory_desc_t dnnl_rnn_desc_t::diff_weights_projection_desc |
Weights gradient projection memory descriptor. This memory descriptor is equal to zero memory descriptor in case of non-projection LSTMs and other non-LSTM RNNs.
| dnnl_rnn_direction_t dnnl_rnn_desc_t::direction |
The direction of RNN primitive execution.
| dnnl_memory_desc_t dnnl_rnn_desc_t::dst_iter_c_desc |
Destination iter memory descriptor for cell state.
| dnnl_memory_desc_t dnnl_rnn_desc_t::dst_iter_desc |
Destination iter memory descriptor for hidden state.
| dnnl_memory_desc_t dnnl_rnn_desc_t::dst_layer_desc |
Destination layer memory descriptor.
| unsigned int dnnl_rnn_desc_t::flags |
RNN cell flags.
| dnnl_primitive_kind_t dnnl_rnn_desc_t::primitive_kind |
The kind of primitive. Used for self-identifying the primitive descriptor. Must be dnnl_rnn.
| dnnl_prop_kind_t dnnl_rnn_desc_t::prop_kind |
The kind of propagation. Possible values: dnnl_forward_training, dnnl_forward_inference, and dnnl_backward.
| dnnl_memory_desc_t dnnl_rnn_desc_t::src_iter_c_desc |
Source iteration memory descriptor for cell state.
| dnnl_memory_desc_t dnnl_rnn_desc_t::src_iter_desc |
Source iteration memory descriptor for hidden state.
| dnnl_memory_desc_t dnnl_rnn_desc_t::src_layer_desc |
Source layer memory descriptor.
| dnnl_memory_desc_t dnnl_rnn_desc_t::weights_iter_desc |
Weights iteration memory descriptor.
| dnnl_memory_desc_t dnnl_rnn_desc_t::weights_layer_desc |
Weights layer memory descriptor.
| dnnl_memory_desc_t dnnl_rnn_desc_t::weights_peephole_desc |
Weights peephole memory descriptor. This memory descriptor is equal to zero memory descriptor in case of non-peephole LSTMs and other non-LSTM RNNs.
| dnnl_memory_desc_t dnnl_rnn_desc_t::weights_projection_desc |
Weights projection memory descriptor. This memory descriptor is equal to zero memory descriptor in case of non-projection LSTMs and other non-LSTM RNNs.
1.8.11
