pixel-rnn-tensorflow icon indicating copy to clipboard operation
pixel-rnn-tensorflow copied to clipboard

Shouldn't we use tanh instead of sigmoid with this implementation?

Open ghost opened this issue 7 years ago • 0 comments

https://github.com/carpedm20/pixel-rnn-tensorflow/blob/1d98c28c62ed9c9584a342b63e4612dd9e5ae688/ops.py#L315

I have found at L315 that we have used the sigmoid function either to calculate the new inputs g, is that possible? Shouldn't we use the tanh instead?

Thank you

ghost avatar Aug 24 '18 22:08 ghost