I´m trying to run a project I got on github: https://github.com/ranahanocka/MeshCNN
First of all, I created an environment in anaconda with .yml files that come in ranahanocka's project, then I got errors related to the CUDA installation.
To solve it, I followed this tutorial to install CUDA and I got rid of that error:
conda install -c anaconda cudatoolkit=10.1
But now I get these errors:

In train.py:

In base_option.py:

And in __init__.py:

I don't know how to fix it, I have read several forums but no solution has helped me. I have tried installing this:
pip install torch===1.9.0+cu111 torchvision===0.10.0+cu111 torchaudio===0.9.0 -f https://download.pytorch.org/whl/cu111/torch_stable.html
But in the middle of downloading I get these errors:


I don't know what can happen, besides, I don't understand very well how CUDA works and its dependencies, so I'm sorry if I don't provide the necessary information.
It's likely that torch or cuda is not properly installed. In my case, someone has updated the Cuda version so I faced a similar issue.
In the command line you can check if you have Cuda available or not;
python
import torch
print(torch.cuda.is_available())
It should return True if not then you need to install Cuda or torch.
Check which Cuda version the GPU has:
nvidia-smi
I used this command to install the cuda and torch in my environment which solved my issue;
conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch-nightly -c nvidia
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With