Machine learning for new materials design

An introductory course on machine learning for predicting the properties of solids.

What it does?

This resource is more of a reference tool. It will help you navigate the methods that can be used to build machine learning-based models to predict material properties.

Quick start

Initial requirements includes several specific python libraries. To navigate it might be helpfull to use "Search docs" field.

To use the Materials Project data base, you need to register (login) on the [website](https://legacy.materialsproject.org/open) and get an API key:

There are two versions of the API available on the Materials Project website. 
- The first [Next-Gen API](https://next-gen.materialsproject.org/api) is a version for developers. In the future, it will become the main one. 
- And the second - [legacy API](https://docs.materialsproject.org/methodology/materials-methodology). This version is most stable and functional.

An extremely useful section of [web-site](https://docs.materialsproject.org/methodology/materials-methodology) describes the methodology for populating the database.