hatlessman
hatlessman
I'm sure we could get together funds to train on some A100's, but the training data is the real problem. Is the data even able to be released? Legal issues?
I'm getting the same. I'm running a 1080 TI and a 5600X, Yuzu 3791. I love the 60fps, but the QTE's run at half speed? I'm using https://github.com/masagrator/Xenoblade2DynFPS/releases/tag/1.0.0-CLEAN
I dropped realtimetts down to 0.3.48 and removed pypiwin32 (I'm on Archlinux). That fixed some errors but then I got ``` ERROR: Cannot install -r requirements.txt (line 100), -r requirements.txt...
I removed some of the explicit versions in requirements.txt and installed torch/torchaudio 2.3.0 instead of 2.1.2. I'm still working on all the settings to get it to launch, but I...
More fun. Linguflex (or one of its dependencies) uses tkinter. I had to follow https://dev.to/xshapira/using-tkinter-with-pyenv-a-simple-two-step-guide-hh5 to get tkinter working inside a venv. I had to also install RealtimeTTS==0.4.1 (pip still...
Blargh! I'm running a 1080TI, a pascal card, doesn't support float16 :( ``` [2024-06-21 11:09:44.523] [ctranslate2] [thread 3455359] [warning] The compute type inferred from the saved model is float16, but...
Blargh! Spoke too soon. Thats just a warning. I just tried transformers==4.38.2. Got the same issue.