You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2019/03/08 01:40:28 UTC

[GitHub] [incubator-mxnet] leleamol commented on issue #14184: Cannot infer RNN symbol with mode=RNNMode::kRnn_tanh

leleamol commented on issue #14184: Cannot infer RNN symbol with mode=RNNMode::kRnn_tanh
URL: https://github.com/apache/incubator-mxnet/issues/14184#issuecomment-470770974
 
 
   Hi @e8035669 
   The "op.h" file in cpp-package is automatically generated. It contains the function declaration (such as the "Symbol RNN ()" mentioned in this issue.) that are derived by parsing the libmxnet.so file.
   Hence, it will not be 
   As per the documentation for the RNN operator, the "state_cell" parameter is only needed for LSTM network. However, due to the way how "op.h" is generated, it is not possible to separate between RNN variants and come up with matching function signatures.
   
   Therefore, users would be required to provide dummy state_cell symbol or invoke the RNN operator directly as you had mentioned.
   
   It is currently a limitation and would require re-design of how op.h is generated.
   Please let me know if this answers your question.
   
   @mxnet-label-bot add [Pending Requester Info]
   

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services