-
https://git.ecker.tech/ aims to provide a place to share my efforts while maintaining true ownership of my code, as I do not trust GitHub.
XMR: 4B9TQdkAkBFYrbj5ztvTx89e5LpucPeTSPzemCihdDi9EBnx7btn8RDNZTBz2zihWsjMnDkzn5As1LU6gLv3KQy8BLsZ8SG
- Joined on
2022-10-10
[Errno 2] No such file or directory: './models/tortoise/dvae.pth'
To bluntly put it, you're doing something really, really wrong, if you managed to start the web UI without it downloading the…
d97639e138
whispercpp actually works now (language loading was weird, slicing needed to divide time by 100), transcribing audio checks for silence and discards them
Same as mrq/ai-voice-cloning#62: you should validate your training settings, because your gradient accumulation size is too large for the given batch size.
b8a620e8d7
actually accumulate derivatives when estimating milestones and final loss by using half of the log
Lazily wrapped in a try/catch block and one extra == 0 check in commit 35225a35daa5b346a132629a2efc0b1a0a7de067.
Well, I suppose it's better than nothing. At least, in my original scope of "I want TTS to unload when I train", I guess it does, but there's this lingering 500MiB I'm not sure what is for. It…
Make sure you click Validate Training Configuration for your given settings before saving. desu I pretty sure this is specifically because Batch Size / Gradient Accumulation Size > 2, so…
Haha, I wonder if that was the whole reason it never seemed to do anything. Fixed in cd8702ab0dab9bb156864e7b5da69060910ece3a.
desu, I don't expect it actually make a difference, since I added…