Training bigger model - Githubissues

enhuiz / vall-e

An unofficial PyTorch implementation of the audio LM VALL-E

MIT License

2.94k stars 415 forks source link

Training bigger model #49

Open RyuGood0 opened 1 year ago

RyuGood0 commented 1 year ago

I don't understand how I can modify the amount of layers, heads, embedding size, etc...

CopyNinja1999 commented 1 year ago

The author has stated that in the readme which is creating yaml configuration for each dataset. You can check the config.py to check all the hyper-parameters as well.