gumbel

gumbel copied to clipboard

zhuqunxi

Correct loss

1

As seen here: https://github.com/ericjang/gumbel-softmax/blob/master/Categorical%20VAE.ipynb `loss=tf.reduce_mean(-elbo)`

adammenges

Minimizing KL (not maximizing)

1

Thanks for uploading this. However, there is an error. In the loss computation, you have `elbo = data_dim * bce(x, x_hat) - KL`. It should be `elbo = data_dim *...

sherjilozair

Continuous target

3

Hi Eder, Thanks for the Keras implementation! I'm excited about this idea as well. (Hopefully) quick question: I'm interested in learning an autoencoder for the output features of some other...

bkj