JARVIS ML Training Data

Various properties of 24,759 bulk and 2D materials computed with the OptB88vdW and TBmBJ functionals taken from the JARVIS DFT database.

This dataset was modified from the JARVIS ML training set developed by NIST (1-2). The custom descriptors have been removed, the column naming scheme revised, and a composition column created. This leaves the training set as a dataset of composition and structure descriptors mapped to a diverse set of materials properties.

Available as Monty Encoder encoded JSON and as the source Monty Encoder encoded JSON file. Recommended access method is with the matminer Python package using the datasets module.

Note on citations: If you found this dataset useful and would like to cite it in your work, please be sure to cite its original sources below rather than or in addition to this page.

Dataset discussed in:

Machine learning with force-field-inspired descriptors for materials: Fast screening and mapping energy landscape Kamal Choudhary, Brian DeCost, and Francesca Tavazza Phys. Rev. Materials 2, 083801

Original Data file sourced from: