setup.md

Setup

If the setup.sh script does not work, then you can perform the following steps instead. These are the steps followed by setup.sh under the hood.

To create the barriers conda environment, use the following commands:

conda upgrade conda
conda env create -f environment.yml

This creates an environment called barriers.

To ensure that the barriers environment is accessible through Jupyter, add the the barriers display name:

python -m ipykernel install --user --name barriers --display-name "Python [conda env:barriers"]

Next, download the Neural Force Field repository, which is also managed by our group. You can either install it through pip, or clone it and put the folder in your path (see below). We recommend the latter, since NFF is under constant development, and often needs to be pulled from github when changes are made.

Lastly, put the azo_barriers repository (and possibly NFF) in your path by adding the following lines to ~/.bashrc (linux) or ~/.bash_profile (mac):

# add the `azo_barriers` path
export BARRIERS=<path to azo_barriers>
export PYTHONPATH=$BARRIERS:$PYTHONPATH

# if you're adding the Neural Force Field path as well
export NFFDIR=<path to NFF>
export PYTHONPATH=$NFFDIR:$PYTHONPATH

Make sure to run source ~/.bashrc or source ~/.bash_profile afterwards, so that your paths are updated.