[GitHub] edmBernard commented on issue #7964: Gradient accumulation of several sample

2017-09-23 Thread git 
edmBernard commented on issue #7964: Gradient accumulation of several sample
URL: 
https://github.com/apache/incubator-mxnet/issues/7964#issuecomment-331624308
 
 
   I was looking for a more regular way :(
   I will test to hack the optimizer thx
 

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:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] edmBernard commented on issue #7964: Gradient accumulation of several sample

2017-09-21 Thread git 
edmBernard commented on issue #7964: Gradient accumulation of several sample
URL: 
https://github.com/apache/incubator-mxnet/issues/7964#issuecomment-331126256
 
 
   I try but I can't run forward-backward on a batch of 100+ images, it consume 
too much memory and crash.
   That's why I ask if there is a way to accumulate gradient outside of batch 
usage.
 

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:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] edmBernard commented on issue #7964: Gradient accumulation of several sample

2017-09-20 Thread git 
edmBernard commented on issue #7964: Gradient accumulation of several sample
URL: 
https://github.com/apache/incubator-mxnet/issues/7964#issuecomment-331064787
 
 
   A bit more description of the process (it's for image retrieval-like task): 
   - compute triplet loss on 1 (query, relevant, non-relevant) images
   - sort these triplet to keep the 100 first with higher loss
   - compute gradient on these 100 samples
   - aggregate gradients
   - update weight
   
   I can't pack these 100 samples in one batch and make the learning process on 
this batch because a batch of 100*3 images take too much place in graphic card 
memory.
 

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:
us...@infra.apache.org


With regards,
Apache Git Services