szha commented on pull request #20065: URL: https://github.com/apache/incubator-mxnet/pull/20065#issuecomment-803638870
@khaotik that might have been the host running out of memory. There are some additional errors related to AMP: https://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Fcentos-gpu/detail/PR-20065/1/pipeline/#step-170-log-891. The way to fix it is to include the new operators in [this list](https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/amp/lists/symbol_fp16.py#L40), which signifies that when AMP is enabled, these operators need not to be casted. -- 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: [email protected]
