tppppppppp commented on issue #8011: Memory used keeps arising during training.
URL: 
https://github.com/apache/incubator-mxnet/issues/8011#issuecomment-331895925
 
 
   Thanks. The gpu memory is steady after replacing curr_loss += loss with 
curr_loss += loss.asnumpy(). It looks as if the gpu memory of 'loss' is 
released after calling asnumpy. Would it not be realeased if I didn't call 
asnumpy? @piiswrong 
 
----------------------------------------------------------------
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:
[email protected]


With regards,
Apache Git Services

Reply via email to