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 2018/10/13 19:28:39 UTC
[GitHub] ArmageddonKnight commented on issue #11402: Understanding Memory
Allocation in MXNet
ArmageddonKnight commented on issue #11402: Understanding Memory Allocation in MXNet
URL: https://github.com/apache/incubator-mxnet/issues/11402#issuecomment-429568849
@SunnyBeike Sorry but I am not sure what has happened in your case, but there should definitely be something printed. After all, MXNet needs to allocate certain amount of GPU memory to the inputs, outputs, and weight parameters (whether those sum up to be the total memory consumption reported by `nvidia-smi` is another problem). My suggestion goes as follows:
- Have you recompiled the core libraries (using `make`)?
- Are you sure that you are using the GPU implementation of the operators?
----------------------------------------------------------------
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