jessebrizzi edited a comment on issue #10867: Scala Module API resize is leaking memory on the native size. URL: https://github.com/apache/incubator-mxnet/issues/10867#issuecomment-458285899 @zachgk Sorry for the late response, I had to find some time to run through my tests. I modified my repo (https://github.com/jessebrizzi/MXNet-Bug) with the example code for the memory leak bug to use the snapshots published in the Nightly Repo https://repository.apache.org/#nexus-search;gav~org.apache.mxnet and include a Dockerfile that sets up an image to test my code with the nightly builds (You have to change the Cuda version in the docker image to switch between `1.4.0-SNAPSHOT` and `1.5.0-SNAPSHOT`) I think I am still seeing the bug (native memory climbing when JVM and GPU memory stays stable), but I am running into a crashing issue when running my "bugged" reproduction code https://github.com/jessebrizzi/MXNet-Bug/blob/master/scala/TestBug.scala In CPU mode the script will freeze after a few hundred loops with all CPU's pegged and in GPU mode it freezes with the GPU/CPU idling. In the non-bugged version (https://github.com/jessebrizzi/MXNet-Bug/blob/master/scala/TestNoBug.scala) of my repoduction code this does not happen, it runs fine since it avoids the resize call. ~~I'm currently trying to get a repo/test using the current 1.3.1 published release, but I have to work through some other blocking issues with that (Mainly the 1.3.1 published scala jar for Linux-GPU is built against opencv 3.4 instead of the opencv 2.4 provided by Ubuntu 16.04 https://packages.ubuntu.com/xenial/libopencv-dev), and I'll comment again when I have that working.~~
---------------------------------------------------------------- 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
