[GitHub] [incubator-mxnet] starimpact commented on issue #15102: the grad of lars should be scaled in lbsgd

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

starimpact commented on issue #15102: the grad of lars should be scaled in lbsgd
URL: https://github.com/apache/incubator-mxnet/issues/15102#issuecomment-542476256
 
 
   _l2norm is time consuming for a large parameter, so I suggest it should be:
   ```python
       def _l2norm(self, v):
           "inner product implementation"
           #for big local parameter
           v = v.reshape(-1)
           if len(v) > 100000:
               step = len(v)/100000+1
               v = v[::step]
           norm = multiply(v, v).asnumpy().sum()
           #norm = (multiply(v, v).sum()).asnumpy()
           norm = math.sqrt(norm)
           return norm
   ```

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