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 2018/09/30 23:49:09 UTC

[GitHub] johnbroughton2017 opened a new issue #12707: well-trained model fails at training data

johnbroughton2017 opened a new issue #12707: well-trained model fails at training data
URL: https://github.com/apache/incubator-mxnet/issues/12707
 
 
   Hi all,
   
   For some reason I am encountering the following problem:
   I trained an alexnet type model until the training accuracy is about 100%, which is the output of the training process. To validate the model, I manually load the last checkpoint and then predict on the training data. But the accuracy drops to about 80%. I think something must be wrong. I tried adding or removing the mean RGB but it seems the same. 
   
   Anyone with experience on this issue? Any suggestion is highly appreciated. Thanks.
   
   -- J

----------------------------------------------------------------
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