This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| ## Works for Della cluster at Princeton University as of 6/27/2024 | |
| ssh della-vis1 # use for installation | |
| cd /directory-to-download-your-source-code-to/ | |
| ## DeePMD-kit installation | |
| git clone https://github.com/deepmodeling/deepmd-kit.git deepmd-kit-new # can choose folder name | |
| cd deepmd-kit-new |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| ## Works as tested on Princeton Della Cluster on April 16, 2023. | |
| module load anaconda3/2021.5 | |
| conda create -n dpdev python=3.10 # Customize a name (e.g. dpdev, deepmd, ...) for this conda environment | |
| conda activate dpdev | |
| export CONDA_OVERRIDE_CUDA=11.6 | |
| # install libtensorflow_cc gpu version (Note that only libtensorflow_cc in the deepmodeling channel enables GPU. The libtensorflow_cc library in the conda-forge channel can only be used on GPU.) | |
| conda install -c deepmodeling libtensorflow_cc=*=cuda11* |