newcomer520 / tf-dagmm Goto Github PK

Hello
Thank for your great implementation. I just used your code to run, but finding that the sample energy is negative. Can you explain it?

Hi,

Is there a way to easily reproduce the paper using this implementation? Using the datasets mentioned in the paper and getting the same metrics.

Are modifications necessary for larger camera images? Does your code expect any multiples of image sizes?

The network worked well for you, according to the paper content?

Thanks for your implementation and the good readme.

Hello,
Thank you for your great implementation !
But can you please show some of your images which generated from the AE model ?

Hi,
Is there any guidelines for choosing latent space dimensions? Thanks!