-
Notifications
You must be signed in to change notification settings - Fork 1.3k
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
ModuleNotFoundError: No module named 'torch' #253
Comments
try |
this cant be done within a setup.py file in my project... |
Maybe regressing to older version of flash-attn is sufficient. I succeed with flash-attn==1.0.4 |
I'm getting below with
|
We recommend the Pytorch container from Nvidia, which has all the required tools to install FlashAttention. |
Looks like the issue was that my anaconda install was in /anaconda and therefore required sudo. After reinstalling anaconda in ~/, Also, I installed Pytorch nighly build that works with CUDA 12.0. It would be GREAT to retain ability to build things outside one specific container! |
Seeing python needs more details about dependencies during build time and it's not being threaded through the entire project definition (and it's not great/safe to be calling other installed libraries during install time, etc). It looks like this is project borrowed a copy of the broken detectron2 setup script which they also haven't fixed for years so the infrastructure contagion is spreading 🫠 also see: |
I had to remove pyproject.toml for now since I couldn't find a way to add torch as a build dependencies that work for everyone. Hopefully installation would work for the newest version (1.0.8). |
Hi, the installation doesn't work for me with 1.0.8. Same error
|
|
This issue is a dupe of #246. |
torch actually installed
The text was updated successfully, but these errors were encountered: