Sampling of alpha in wgan_gp.py

[ghost]

Hi,

in the gradient_penalty function in the WGAN class, alpha is sampled from a normal distribution (tf.random.normal) with mean 0.0 and std 1.0.

In the "Improved Training of Wasserstein GANs" paper/code, and all other implementations I have seen, this is sampled from a uniform distribution with min 0.0 and max 1.0.

I cannot find any discussion on sampling this from alternative distributions to what was proposed, but it clearly still works. I just wonder whether anyone can explain the motivation for this deviation from the original model?

[pcoet]

For folks who might be interested in this question, please feel free to raise it under the Keras category of the TensorFlow forum. Thanks!