.. _installation:

Installation
============

torch-pgn either be installed from PyPi using the pip command or from source. We assume that all users are using conda,
if you do not have conda, please install Miniconda (see the link below). Currently torch_pgn required python 3.7 due to
some dependencies not being upaded ; however, we are working to provide a release compatible with python 3.x.

.. admonition:: Key Links

    | pip: https://pypi.org/project/torch-pgn/
    | git repo: https://github.com/keiserlab/torch_pgn
    | miniconda: https://docs.conda.io/en/latest/miniconda.html

Option 1: Installation using pip (cpu only)
-------------------------------------------

.. code-block::

    conda create --name torch_pgn python=3.7
    conda activate torch_pgn
    pip install torch_pgn
    conda install pytorch-sparse -c pyg
    conda install -c conda-forge openbabel

Option 2: Installation using pip (cuda)
---------------------------------------

.. code-block::

    conda create --name torch_pgn python=3.7
    conda activate torch_pgn
    conda install pytorch==1.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia
    conda install pyg -c pyg
    conda install pytorch-sparse -c pyg
    conda install -c conda-forge openbabel
    pip install torch_pgn


Option 3: Installing from source using environment.yml
-------------------------------------------------------

.. code-block::

    git clone https://github.com/keiserlab/torch_pgn/torch_pgn.git
    cd torch_pgn
    conda env create -f environment.yml
    conda activate torch_pgn
    pip install -e