Successful environment setup #57
Description
I've made several attempts at setting up the environment, and when I followed these steps, the code ran successfully.
I'm sharing this to help others who might encounter similar issues.
First of all, the content of the "env_copy.yml" file is as follows:
channels:
- pytorch
- conda-forge
- defaults
- pyg
dependencies:
- ase=3.22
- autopep8
- cudatoolkit=10.2
- jupyterlab
- matminer=0.7.3
- matplotlib
- nglview
- pip
- pyg=2.0.1
- ipywidgets
- pylint
- pymatgen==2023.8.10
- python=3.8.13
- pytorch-lightning=1.3.8
- pytorch=1.9.0
- seaborn
- tqdm
- pip:
- hydra-core==1.1.0
- hydra-joblib-launcher==1.1.5
- p-tqdm==1.3.3
- pytest
- python-dotenv
- smact==2.2.1
- streamlit==0.79.0
- wandb==0.10.33
name: cdvae_copy
Save the above content to the "env_copy.yml" file and then execute the following.
conda env create -f env_copy.yml
conda activate cdvae_copy
If you encounter the error
ImportError: cannot import name 'get_num_classes' from 'torchmetrics.utilities.data'
, then try the following steps:
pip uninstall torchmetrics==0.9.0.dev0
pip install torchmetrics==0.5
If you encounter the error
AttributionError: module 'distutils' has no attribute 'version'
, then try:
pip install setuptools==59.5.0