jeremiedb commented on issue #9358: Why does running 1 round of an MXNET model training produce Train-mse=NaN? URL: https://github.com/apache/incubator-mxnet/issues/9358#issuecomment-366400747 This looks like a bug. There should be a reset of the iterator prior to the forward-backward loop instead of the current reset that is made after exhausting the iterator. I can fix this in the current opened pull request. @alexmosc : this bug correction will however be applied on current master version, for which there's currently no recent Windows package built. Building from source on Windows is a bit of a pain, so easiest might be to run through Linux or use a docker image on Windows.
---------------------------------------------------------------- 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
