dmas-at-wiris opened a new issue #10073: NaN in loss when using gluon ELU block URL: https://github.com/apache/incubator-mxnet/issues/10073 I have a CNN that uses ELUs as activation function. When I use the gluon implementation (`F.where(x > 0, x, self._alpha * (F.exp(x) - 1.0))` from [here](https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/gluon/nn/activations.py#L161)) I obtain NaN values in the loss after few epochs. However, if I create a block with this `F.LeakyReLU(x, act_type='elu', alpha)` the loss doesn't diverge. Any idea? I'll try to provide a minimal working example asap, if necessary.
---------------------------------------------------------------- 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: us...@infra.apache.org With regards, Apache Git Services