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 2018/08/02 14:45:21 UTC

[GitHub] sbodenstein commented on issue #11984: WIP - Generalized broadcast_like operator

sbodenstein commented on issue #11984: WIP - Generalized broadcast_like operator
URL: https://github.com/apache/incubator-mxnet/pull/11984#issuecomment-409952182
 
 
   I find this design counterintuitive:
   ```
   s=mx.sym.broadcast_like(mx.sym.Variable("lhs"), mx.sym.Variable("rhs"), other_axes=(),input_axes=())
   s.infer_shape(lhs=(1,7,1,1), rhs=(3,7,9,4))[1]
   [(3, 7, 9, 4)]
   ```
   So `input_axes=()` is equivalent to `input_axes=(0, 2, 3)`. Surely an empty tuple should be interpreted as the identity? You could make `other_axes` and `input_axes` optional with default `None`.
   

----------------------------------------------------------------
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