stephenrawls commented on a change in pull request #14208: Add support for fast
variable-length LSTM
URL: https://github.com/apache/incubator-mxnet/pull/14208#discussion_r280663423
##########
File path: src/operator/rnn-inl.h
##########
@@ -1362,7 +1427,9 @@ class RNNOp {
std::vector<cudnnTensorDescriptor_t> x_desc_vec_, y_desc_vec_, dx_desc_vec_,
dy_desc_vec_;
#if USE_CUDNN_LSTM_PROJ
cudnnRNNDataDescriptor_t x_data_desc_, y_data_desc_, dx_data_desc_,
dy_data_desc_;
+ DType padding_fill_ = 0;
#endif
+ std::vector<IType> sequence_length_array_cpu;
Review comment:
Done.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
