To install the server, you need one of:
If installing for the first time you need to create the conda environment:
conda env create
This downloads and installs the required dependencies in an isolated environment (i.e. does not interfere with your system's installation).
If you already have done the previous step, you might want to update the dependencies.
conda env update
After that you need to activate the environment:
conda activate gnn
If you have problems with PyTorch and pyg installation make sure you follow the following order (this worked on ubuntu22.04 LTS):
- Install CUDA manually. For example by:
sudo apt-get -y install nvidia-cuda-toolkit
- Install PyTorch (this also works with CUDA 11.5 instead of 11.3). Check version compatibility here: https://pytorch.org/get-started/previous-versions/
conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch
- Install PyTorch Geometric (pyg). Check version compatibility here: https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html
conda install pyg -c pyg
Also install Captum for pytorch
conda install captum -c pytorch
Make sure you also activate the conda environment in PyCharm, a guide can be found here.