You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Needed to install cuda 12.1 which is latest supported by pytorch 2.1.2
Torch needed a specific installation using conda :
conda install pytorch==2.1.2 torchvision==0.16.2 torchaudio==2.1.2 pytorch-cuda=12.1 -c pytorch -c nvidia https://pytorch.org/get-started/previous-versions/
Xformer also needed a specific installation :
pip3 install "xformers<0.0.23" --index-url https://download.pytorch.org/whl/cu121
Running AudioTest.py told me that I was running Numpy > 2 and Applied pip install "numpy<2" anyway :
blis 1.0.1 requires numpy<3.0.0,>=2.0.0, but you have numpy 1.26.4 which is incompatible.
thinc 8.3.2 requires numpy<2.1.0,>=2.0.0; python_version >= "3.9", but you have numpy 1.26.4 which is incompatible.
Then it worked. You could very well update your install guide since it's really incomplete.
The text was updated successfully, but these errors were encountered:
Carlsans
changed the title
Did not build on python 3.9, worked with 3.11
Multiple errors compiling
Oct 3, 2024
Needed to install cuda 12.1 which is latest supported by pytorch 2.1.2
Torch needed a specific installation using conda :
conda install pytorch==2.1.2 torchvision==0.16.2 torchaudio==2.1.2 pytorch-cuda=12.1 -c pytorch -c nvidia
https://pytorch.org/get-started/previous-versions/
Xformer also needed a specific installation :
pip3 install "xformers<0.0.23" --index-url https://download.pytorch.org/whl/cu121
Running AudioTest.py told me that I was running Numpy > 2 and Applied pip install "numpy<2" anyway :
blis 1.0.1 requires numpy<3.0.0,>=2.0.0, but you have numpy 1.26.4 which is incompatible.
thinc 8.3.2 requires numpy<2.1.0,>=2.0.0; python_version >= "3.9", but you have numpy 1.26.4 which is incompatible.
Then it worked. You could very well update your install guide since it's really incomplete.
The text was updated successfully, but these errors were encountered: