[GitHub] [incubator-mxnet] lvpasquier commented on issue #17275: CosineEmbeddingLoss won't work after hybridize (ie. with symbol API)

[GitBox <gi...@apache.org>]

lvpasquier commented on issue #17275: CosineEmbeddingLoss won't work after hybridize (ie. with symbol API)
URL: https://github.com/apache/incubator-mxnet/issues/17275#issuecomment-573896943
 
 
   Hello, thanks for having a look.
   
   Actually shape issue seems to occur even before that when `F.norm` is called. 
   
   ```python
   import mxnet as mx
   from mxnet.gluon import HybridBlock
   
   
   class Foo(HybridBlock):
       def hybrid_forward(self, F, x, axis=-1):
           return F.norm(x, axis=axis).reshape(-1, 1)
   
   
   foo = Foo()
   x = mx.random.randn(2, 2)
   
   print(foo(x).shape)  # (2, 1)
   foo.hybridize()
   print(foo(x).shape)  # (2,)
   ```

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