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 2017/12/04 21:53:30 UTC

[GitHub] bejjani opened a new issue #8940: Embedding layer - add default behavior for out of bound index to documentation

bejjani opened a new issue #8940: Embedding layer - add default behavior for out of bound index to documentation
URL: https://github.com/apache/incubator-mxnet/issues/8940
 
 
   https://github.com/apache/incubator-mxnet/blob/master/src/operator/tensor/indexing_op.h#L613
   
   The current default/only behavior of the embedding layer in case of an out of bound index is to "clip" to max index of the embedding layer.
   The documentation does not mention it.
   
   Another suggestion is to add the option to assign a random embedding during prediction for out of bound indexes. That makes more sense to me compared to clipping.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on 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