GraphChem: Graph-based machine learning for fuel property prediction
GraphChem is an open source Python package for constructing graph-based machine learning models with a focus on fuel property prediction.
Future plans for GraphChem include:
- Robust hyper-parameter and model architecture tuning runtimes
- Molecule visualization via RDKit
- Extensive automated testing
Installation:
Prerequisites:
- Have Python 3.5+ installed
- Have RDKit installed (using Conda environments is highly recommended)
Method 1: pip
$ pip install graphchem
Method 2: From Source
$ git clone https://github.com/ecrl/graphchem
$ cd graphchem
$ python setup.py install
If any errors occur when installing dependencies, namely with PyTorch or torch-geometric, visit their installation pages and follow the installation instructions: PyTorch, PyTorch Geometric
Usage:
API documentation is coming in the future! In the meantime, take a look at some examples.
Contributing, Reporting Issues and Other Support:
To contribute to GraphChem, make a pull request. Contributions should include tests for new features added, as well as extensive documentation.
To report problems with the software or feature requests, file an issue. When reporting problems, include information such as error messages, your OS/environment and Python version.
For additional support/questions, contact Travis Kessler ([email protected]).