Nathan Fradet

Hey guys, thanks for reviewing, its my pleasure to contribute considering how useful transformers have been to me ! 😃 @gante 1. Noted. I have put them back. 2. Sounds...

That's totally understandable. You guys are already managing this ecosystem really well, and make a huge impact ! 🙌 I created the `__init__` method, overriding `model.generation_config`. Indeed the code is...

Hey @gante, Thanks, the last changes are done. I'll take the instructions for the rebase, I just didn't do it right

Good, evaluate and predict are back as original, it should be good now

Suggestions applied, sorry for these typos (copy / paste ...) 😅

Sure, here it is

Good point (CLI)! In that can a json file could work, and alternatively the argument could maybe accept both paths to this file and a `GenerationConfig` object ?

Hey @gante, yep, just clearing my backlog, it should be done by the week-end

Closing as micro timings will be implemented in another PR

Hi @Kinyugo , thank you for the report! :) I'll inspect the script very soon, tomorrow at the latest. Now indeed `tokenize_midi_dataset` doesn't handle multiprocessing, which was actually already discussed...