adrianloy commented on issue #8959: Loss becomes nan when using correlation loss with MakeLoss on 2 images. URL: https://github.com/apache/incubator-mxnet/issues/8959#issuecomment-349603363 @piiswrong the cross correlation should never be NaN, right? It should always be between -1 and 1 or something like that I thought. -Edit- Just realized MakeLoss gives us the gradient, which can be nan for cross correlation I guess.
---------------------------------------------------------------- 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
