Prafulla Dhariwal

Just pushed a fix, can you check again if it works now?

Yea, thats probably because the previously the links were in Google cloud and since the colab notebook is also in Google cloud the download was fast. It'll only have to...

I updated the link to download from a CDN, that should improve download speeds. To try it you'll need to restart the notebook.

You want to add an entry for `restore_prior="path/to/checkpoint"` inside the corresponding hps dictionary, in your case "small_prior". For example, if your checkpoint is stored at "/root/jukebox/logs/small_prior/checkpoint_latest.pth.tar" and you're going to...

Can you print the data types of all variables in this step as well as input to the function adam_step?

The project does require a GPU to run, it could work on CPU but hasn't been tested and will almost surely be very slow. @maraoz The NCCL error you see...

A way to fine-tune from our models would be to add a new embedding(s) for your new artist(s), and initialise them from the aritst_id = 0 ie "unknown" artist embedding.

Uploaded pretrained model files and logs for the Ablation experiments and the LSUN experiments, links are in README. Hopefully they'll be useful to help reproduce results! Check the demo folder...

Yes we trained with 40 GPU's for about a week, but samples did start to look good after a couple of days. If you're getting invertibility errors with small batch...

The change I did was to replace the label returned (ie `y`) with `attr` in `parse_tf_record`, and not add it as an extra output. Then the rest of the code...