search

enhuiz / vall-e

An unofficial PyTorch implementation of the audio LM VALL-E
MIT License
2.97k stars 419 forks source link

vall-e How to run on macos m1 #100

Open johnny-yu opened 1 year ago

johnny-yu commented 1 year ago
  1. AssertionError: Torch not compiled with CUDA enabled vi /usr/local/lib/python3.10/site-packages/vall_e/main.py on line 18 parser.add_argument("--device", default="cuda") --> parser.add_argument("--device", default="cpu")

_pickle.UnpicklingError: could not find MARK

johnny-yu commented 1 year ago

The above issue has been resolved, it is a relative directory issue； cuda change cpu ~/config ~/data ~/nltk_data vi ~/data/test/ar.yml. data_dirs: [data/test] parameter should be equal to the file path where it is located 1.ok 2.ok 3.ok but 、5.python -m vall_e.train yaml=data/test/ar.yml on the macOS m1 RuntimeError: Distributed package doesn't have NCCL built in How to solve it ？