shiqing cai

mixed precision

3

It seems that the package dose not support mixed precision training, since the Embedding layer needs add operating with output and embedding parameters, which dtype are 'float32' and 'float16' leading...

package version

3

Can you please give the exact version of each package on you envirnment?

loss problem

1

I'm doubt that why generator loss is defined as `gen_loss = K.mean(fake_output)`, maybe should it be `gen_loss = K.mean(real_output)` to confuse discriminator?

As described in this paper, the noise(sigma) increases, the respected L(W) decreases. But if we understand sigma as uncertainty of y, maybe it's better for L to be increase with...