Open RifqiAW opened 5 years ago
+1!
On the same example: it seems that we are only training stage 1 of the paper. In order to perform generation we would also need to train the prior (using PixelCNN for example). Is this included somewhere?
This repository seems to implement VQ-VAE for video: https://github.com/ritheshkumar95/pytorch-vqvae
+1 for an updated example for raw audio!
Is it possible to update the example for vqvae to also include how to use it on raw audio and/or video data?
Thank you