Jhosimar George Arias Figueroa
Jhosimar George Arias Figueroa
Sorry if the name of the loss was misleading. I called it categorical loss because the terms involved in this loss where based on the categorical latent variable that represents...
@nikhilrayaprolu On the blog the categorical loss is given in the following line of code: ``` # loss and train ops kl_tmp = tf.reshape(q_y*(log_q_y-tf.log(1.0/K)),[-1,N,K]) KL = tf.reduce_sum(kl_tmp,[1,2]) ... ``` On...
I'm closing this issue due to inactivity
Apparently the author deleted the post related to gmvae from his blog. The following paper is related but does not contain the detailed information that was given in the blog....