Skip to content

Machine learning for determining band-gap from crystalline formula using a materials informatics approach.

Notifications You must be signed in to change notification settings

JovinRyan/BandGap_ML

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

BandGap_ML

Introduction

The band gap is a fundamental property in materials science which plays a significant role in determining the applications of a material in fields such as optoelectronics, quantum computing, etc. The process of determining the band gap of a material through direct means (opticalabsorption spectroscopy, photoluminescence spectroscopy, and photoconductivity measurements) or indirect means involving theoretical calculations (Density Functional Theory) requires significant time and capital investment.

Current Results

A simple Feed Forward Neural Network (FFNN) showed promising results in predicting the band gap of crystalline materials. Further experimentation with different models, learning rates and data samples is needed to improve accuracy to a sufficient degree.

FFNN loss curve

Losses are reported as average loss per batch, with batch size of 32 data points.

Data and Citations

Band gap data compiled in /DataFiles/UpdatedMaterialsDF.csv was taken from MaterialsProject.org

  1. Anubhav Jain, Shyue Ping Ong, Geoffroy Hautier, Wei Chen, William Davidson Richards, Stephen Dacek, Shreyas Cholia, Dan Gunter, David Skinner, Gerbrand Ceder, Kristin A. Persson; Commentary: The Materials Project: A materials genome approach to accelerating materials innovation. APL Mater. 1 July 2013; 1 (1): 011002. https://doi.org/10.1063/1.4812323

Elemental data used for featurization listed in /DataFiles/PeriodicTable.csv was taken from https://gist.github.com/GoodmanSciences/c2dd862cd38f21b0ad36b8f96b4bf1ee

About

Machine learning for determining band-gap from crystalline formula using a materials informatics approach.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published