Data

Tools

Indexes

Applications

Experiments

Open Source Science

matex

Code for extrapolation in materials property prediction as proposed in "Known Unknowns: Out-of-Distribution Property Prediction in Materials and Molecules".

https://github.com/learningmatter-mit/matex

Science Score: 62.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ○
    .zenodo.json file
  • ✓
    DOI references
    Found 3 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: arxiv.org, zenodo.org
  • ○
    Academic email domains
  • ✓
    Institutional organization owner
    Organization learningmatter-mit has institutional domain (gomezbombarelli.mit.edu)
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (12.2%) to scientific vocabulary

Scientific Fields

Engineering Computer Science - 60% confidence
Artificial Intelligence and Machine Learning Computer Science - 40% confidence
Last synced: 4 months ago · JSON representation ·

Repository

Code for extrapolation in materials property prediction as proposed in "Known Unknowns: Out-of-Distribution Property Prediction in Materials and Molecules".

Basic Info
  • Host: GitHub
  • Owner: learningmatter-mit
  • License: mit
  • Language: Python
  • Default Branch: main
  • Size: 4.68 MB
Statistics
  • Stars: 11
  • Watchers: 4
  • Forks: 0
  • Open Issues: 0
  • Releases: 1
Created 11 months ago · Last pushed 10 months ago
Metadata Files
Readme License Citation

README.md

Known Unknowns: Out-of-Distribution Property Prediction in Materials and Molecules

License: MIT DOI

Code for extrapolation in materials property prediction as proposed in Known Unknowns: Out-of-Distribution Property Prediction in Materials and Molecules.

Setup

Clone the repository git clone https://github.com/learningmatter-mit/matex.git Create and activate a virtual environment conda create -n blt-matex python=3.9.16 conda activate blt-matex Install requirements pip install -r requirements.txt

Environment Setup

Update hyperparameters in blt/configs/materials.yml. Run the following command where path_to_dir is the parent directory of matex. export PYTHONPATH="${PYTHONPATH}:path_to_dir"

Data

Run the following script to process the data. Raw data is provided in blt/data. Processed data will be saved under blt/data as pkl files. bash data_modules/create_data.sh

Training and Evaluation

Run the following script to train, evaluate and save the model cd blt bash train_eval.sh

Run the following script to create and save distribution and correlation plots python plot_maker/plots.py

Cite

If you use this code in your research, please consider citing @inproceedings{segal2024known, title={Known Unknowns: Out-of-Distribution Property Prediction in Materials and Molecules}, author={Segal, Nofit and Netanyahu, Aviv and Greenman, Kevin and Agrawal, Pulkit and Gómez-Bombarelli, Rafael}, booktitle={Workshop on AI for Accelerated Materials Design at Advances in Neural Information Processing Systems}, year={2024} }

Acknowledgements

The implementation is derived from Bilinear Transduction. The datasets are derived from AFLOW, Matbench, Materials Project, and MoleculeNet. The data processing and feature extraction are derived from Can machine learning find extraordinary materials?, Modnet and Deepchem.

Research was sponsored by the Department of the Air Force Artificial Intelligence Accelerator and was accomplished under Cooperative Agreement Number FA8750-19-2-1000. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Department of the Air Force or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein.

Owner

  • Name: Learning Matter @ MIT
  • Login: learningmatter-mit
  • Kind: organization
  • Email: rafagb@mit.edu

Rafael Gomez-Bombarelli Group @ MIT

Citation (CITATION.bib) 

@inproceedings{segal2024known,
  title={Known Unknowns: Out-of-Distribution Property Prediction in Materials and Molecules},
  author={Segal, Nofit and Netanyahu, Aviv and Greenman, Kevin and Agrawal, Pulkit and Gómez-Bombarelli, Rafael},
  booktitle={Workshop on AI for Accelerated Materials Design at Advances in Neural Information Processing Systems},
  year={2024}
}

GitHub Events

Total
  • Create event: 3
  • Issues event: 2
  • Release event: 1
  • Watch event: 16
  • Member event: 2
  • Push event: 7
  • Fork event: 2
Last Year
  • Create event: 3
  • Issues event: 2
  • Release event: 1
  • Watch event: 16
  • Member event: 2
  • Push event: 7
  • Fork event: 2