-
Notifications
You must be signed in to change notification settings - Fork 16
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
Unable to run in Colab or Lightning or GCP VMs - various dependency issues (tried many workarounds) #11
Comments
Hi, this likely indicates a version conflict. You could try using a slightly newer version of torch, such as 1.11.0 or 1.12.0, to see if it resolves the issue. Since HittER and Relphormer were both open-sourced in 2021, there's a possibility that many libraries do not match up with the latest ones available now. |
Are there requirements to use a specific version of CUDA? |
Same behavior. |
When I upgrade setup.py to PyTorch 2.0.0, which requires me to use Python 3.8 (min supported Python version for PyTorch 2), I get this cython error that appears again to relate to PyYAML when I run
|
I met the same problem, I changed cuda11.1 and torch 1.8.1, it works. |
I'm running into versioning/dependency issues when trying to get this running.
When I follow the instructions on the readme, I get:
When I pick one manually to to get beyond the error, I get blocked by a conflict between PyTorch Lightning 1.3.1 and PyYAML.
Both Lightning Studio and Colab use PyYAML 6.0.1, which was released in 2021, by the version of PyTorch Lightning is not compatible with this version.
I have tried manually removing the dependencies and installing from scratch in both environments, but then I get into even deeper dependency conflicts with cython that look like this:
What's the best path forward here?
The text was updated successfully, but these errors were encountered: