DNXie opened a new issue #19860: URL: https://github.com/apache/incubator-mxnet/issues/19860
## Description `mxnet.gluon.nn.swish` outputs NaN if `beta` is a large value. Expect a grace exception message if the input is out of range instead of NaN as output. ### Error Message ~~~python [nan] <NDArray 1 @cpu(0)> ~~~ ## To Reproduce ~~~python import mxnet layer=mxnet.gluon.nn.Swish(beta=-1e+307) layer.initialize() out=layer(mxnet.nd.zeros((1))) ~~~ ## Environment OS: ubuntu 18.04 Python: 3.7.6 pip: 20.0.2 numpy: 1.18.5 mxnet: 1.6.0 ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
