This tutorial contains an introduction to the use of the Materials Simulation Toolkit for Machine Learning (MAST-ML), a python package designed to broaden and accelerate the use of machine learning and data science methods for materials property prediction. Through hands-on activities, we will use MAST-ML to (1) import materials datasets from online databases and clean and examine our input data, (2) conduct feature engineering analysis, including generation, preprocessing, and selection of features, (3) construct, evaluate and compare the performance of different model types and data splitting techniques, and (4) conduct a preliminary assessment of model error analysis and uncertainty quantification (UQ).

MAST-ML Tool (on nanoHUB): Launch MAST-ML Tool

MAST-ML code: https://github.com/uw-cmg/MAST-ML

Publication: https://doi.org/10.1016/j.commatsci.2020.109544

MAST-ML first tutorial: https://colab.research.google.com/github/uw-cmg/MAST-ML/blob/master/examples/MASTML_Tutorial_1_GettingStarted.ipynb

Ryan Jacobs received a B.S. degree in Materials Science and Engineering from the University of Minnesota-Twin Cities in 2010. He then obtained an M.S. and Ph. D. in Materials Science from the University of Wisconsin-Madison in 2012 and 2015, respectively.

Dr. Jacobs is currently a Research Scientist in the Department of Materials Science and Engineering at the University of Wisconsin- Madison. His work focuses on using atomistic modeling, data science and machine learning (materials informatics) methods to understand the structure and properties of materials at the atomic scale in order to discover and design novel material compounds for specific technological applications. His main research application areas of interest comprise materials for energy technology, such as solid oxide and protonic fuel cells, batteries, and solar photovoltaics. Another main thrust of his research is the investigation of surface electronic and thermodynamic properties of metals and oxides used as electron emission cathodes.

Network for Computational Nanotechnology (NCN)

Researchers should cite this work as follows:

• Ryan Jacobs (2022), "The Materials Simulation Toolkit for Machine Learning (MAST-ML): Automating Development and Evaluation of Machine Learning Models for Materials Property Prediction," https://nanohub.org/resources/36490.

  BibTex | EndNote

1. computational materials science
2. data-driven materials discovery
3. data science
4. hands on
5. Jupyter notebooks
6. machine learning
7. MAST-ML
8. material properties
9. materials science
10. model error analysis
11. NCN Group - Machine Learning Hands-on
12. uncertainty quantification (UQ)