GYZHikari commented on issue #3032: training accuracy is very high but validation accuracy is very low URL: https://github.com/apache/incubator-mxnet/issues/3032#issuecomment-445124591 > @piiswrong ,yes, I used muti-GPU to train the model. Hey guys @leoxiaobin@piiswrong, I met the same problem using mxnet1.3.1, CUDNNv5.1. I set fix_gamma=True for training, use multi-gpu and kvstore='local'. While training, the gamma changes all the time. However, when I test use the same code, the results look good.
---------------------------------------------------------------- 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
