Installing ARC - ReactionMechanismGenerator/ARC GitHub Wiki

Note: ARC was only tested Linux (Ubuntu 18.04.1 LTS) and Mac machines. We don't expect it to work on Windows.

1. Clone and setup path

  • Download and install the Anaconda Python Platform for Python 2.7 if you haven't already.
  • Get git if you don't have it already by typing sudo apt-get install git in a terminal.
  • Clone this repository to your local machine by typing the following command in the desired folder: git clone https://github.com/ReactionMechanismGenerator/ARC.git
  • Add ARC to your local path in .bashrc: export PYTHONPATH=$PYTHONPATH:~/Path/to/ARC/ (make sure to change ~/Path/to/ARC/ accordingly)

2. Install dependencies

  • Install the latest DEVELOPER version of RMG (which has Arkane). It is recommended to follow RMG's Developer installation by source using Anaconda instructions. Make sure to add RMG-Py to your PATH and PYTHONPATH as stated in RMG's documentation.
  • Create the Anaconda environment for ARC: conda env create -f environment.yml. Activate the ARC environment every time before you run ARC: source activate arc_env.

3. Generate RSA SSH keys and define servers

  • Generate RSA SSH keys for your favorite server/s on which relevant ESS is installed. Instructions for generating RSA keys could be found here.
  • Make sure that the server/s address/es and your username under un in the servers dictionary in ARC/arc/settings.py are all updated.
  • Update the submit scripts in ARC/arc/job/submit.py according to your servers' definitions.
  • Copy the RSA SSH key path/s on your local machine to ARC/arc/settings.py in the servers dictionary under keys.

4. Associating software with servers

ARC has an automated "radar" feature that scans the servers it has access to, and discovers relevant ESS (electronic structure software) it is familiar with. In order for this feature to function, just make sure your .bashrc file on the server\s does not have an interactive shell check. If it does, disable it.

You can check what the "radar" detects using the ARC ESS diagnostics notebook.

If this feature does not work for you as expected, you can simply bypass it by providing ARC an ess_settings dictionary.

6. Test

  • Run the ARCDemo.ipynb to test ARC. This demo also shows different methods to define species for thermoproperties calculations.
  • If you'd like, running unit tests is done by typing make test in the ARC directory (after activating the environment)

7. Add ARC aliases to your .bashrc (optional and convenient)

  • export arc_path=$HOME'~/Path/to/ARC/' (make sure to specify the correct path to ARC)
  • alias arce='source activate arc_env'
  • alias arc='source activate arc_env && python $arc_path/ARC.py input.yml'
  • alias arcrestart='source activate arc_env && python $arc_path/ARC.py restart.yml'
  • alias arcode='cd $arc_path'
  • alias arcj='cd $arc_path && source activate arc_env && jupyter notebook'