eric-haibin-lin commented on issue #12577: Training with fc and multi-gpu is much slower than single gpu URL: https://github.com/apache/incubator-mxnet/issues/12577#issuecomment-422117864 Did you scale the batch size linearly with the number of GPUs?
---------------------------------------------------------------- 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
