Basics of machine learning for material science with python | ARC Centre of Excellence in Exciton Science

An enormous amount of data on materials are being pumped into freely accessible databases. Where there is data, there is machine learning. Therefore, machine learning has increasingly been employed to crunch all that data and extract new knowledge in several branches of material science.

In this tutorial, you are going to experience the power of using machine learning to predict materials properties. The tutorial covers a few basics of python that are key for doing data processing, and you will learn how to use our CrystalFeatures descriptors to generate a dataset of materials and train a machine learning model to predict material properties such as the bandgap.

The tutorial is delivered by Exciton Science Associate Investigator Dr Sherif Abbas of Deakin University. Click here to access the tutorial's Google Colab notebook.