Installation

Requirements

Requirements:

• Python >= 3.9

• pytorch >= 1.9

• numpy

Optional Dependencies:

• numba (recommended, for accelerating performance)

• cupy (also for accelerating performance)

• ASE (for usage with ase)

• matplotlib (for plotting)

• tqdm (for progress bars)

• graphviz (for viewing model graphs as figures)

• h5py (for ani-h5 datasets)

• pyanitools (for ani-h5 datasets)

Interfacing codes:

• ASE

• PYSEQM

• LAMMPS

Installation Instructions

Conda

Install using conda:

conda install -c conda-forge hippynn

Pip

Install using pip:

pip install hippynn

Install from source:

Clone the hippynn repository and navigate into it, e.g.:

$ git clone https://github.com/lanl/hippynn.git
$ cd hippynn

Note

If you wish to do a cpu-only install, you may need to comment out cupy from the conda_requirements.txt file.

Dependencies using conda

Install dependencies from conda using recommended channels:

$ conda install -c pytorch -c conda-forge --file conda_requirements.txt

Dependencies using pip

Minimum dependencies using pip:

$ pip install -e .

If you feel like tinkering, do an editable install:

$ pip install -e .

If you would like to get all optional dependencies from pip:

$ pip install -e .[full]

Notes

• Install dependencies with pip from requirements.txt .

• Install dependencies with conda from conda_requirements.txt .

• If you don’t want pip to install them, conda install from file before installing hippynn. You may want to use -c pytorch for the pytorch channel. For ase and cupy, you probably want to use -c conda-forge.

• Optional dependencies are in optional_dependencies.txt