-
Notifications
You must be signed in to change notification settings - Fork 4.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
XTTS v1.1 GPT Trainer #3086
XTTS v1.1 GPT Trainer #3086
Conversation
2e029c1
to
affaf11
Compare
1cf1eba
to
ec7f547
Compare
b0a046c
to
bb7f9d1
Compare
bb7f9d1
to
de1d521
Compare
d503d7a
to
8af3d2d
Compare
Looks good to me. Waiting for @WeberJulian |
I'm not 100% sure that we should use those values BATCH_SIZE 3 and GRAD_ACUMM_STEPS 84 for fine-tuning. In some extreme cases (small datasets) a single step will be more than an epoch. Why not use the values we use internally 8 and 1 ? |
Otherwise, it looks good to me too. Great work. In future PRs, we can focus on reducing the VRAM footprint of fine-tuning to something more reasonable that you can do on collab with 16Gb of VRAM. |
That's interesting that you use 8/1 internally. The comments say you need an effective batch size of like 230 or something for good quality. Regardless, it would seem difficult to use a small dataset, in my limited experience, the model overfits pretty quickly, even with a much reduced learning rate, even with a decent sized data set. I'm still trying to get anywhere near a good voice match before it overfits. |
XTTS GPT Trainer for XTTS v1.1.
ToDos: