Kamal Choudhary's Projects
Atomistic Line Graph Neural Network
Example usage of Exabyte.io platform through its RESTful API: programmatically create materials and modeling workflows, execute simulations on the cloud, analyze data and build machine learning models
Atomistic Calculations on Quantum Computers
Deep learning framework for atomistic image data
An SE(3)-invariant autoencoder for generating the periodic structure of materials [ICLR 2022]
Crystal graph convolutional neural networks for predicting material properties.
ChemNLP: A Natural Language Processing based Library for Materials Chemistry Text Data
The polls app from the official Django tutorial, that demonstrates how to build data-driven Python apps in Azure App Service.
Genetic algorithm for structure and phase prediction interfaced to GULP, LAMMPS and VASP.
These are the slides associated with the GNN tutorial at the APS March Meeting
A simple github action that automatically deploys Django app to heroku
JARVIS-Tools: an open-source software package for data-driven atomistic materials design
This project contains the data for evaluation of interatomic potentials/force-fields (used in Moecular-dynamics and Monte-carlo simulations). LAMMPS calculation were done using MPinterface code (https://github.com/JARVIS-Unifies/JARVIS-FF) and in.elastic script in LAMMPS/examples/ELASTIC folder (https://github.com/lammps/lammps/tree/master/examples/ELASTIC) on the structures downloaded from materials project (MP) using REST API (https://www.materialsproject.org/).Force-fields were downloaded from interatomic potential repository project(http://www.ctcms.nist.gov/potentials/) and LAMMPS (https://github.com/lammps/lammps/tree/master/potentials). The interactive plot was made with Bokeh (http://bokeh.pydata.org/en/0.10.0/docs/gallery/periodic.html). Please note that the starting lattice parameters were taken from density functional theory (DFT) and not from experiments. Generic minimization parameters were chosen for LAMMPS run rather than testing them for each individual case such as energy convergence criterion and so on. Hence, there are chances that the calculation gets trapped in a local energy minima. Please read carefully the assumptions taken during calculations in the in.elastic script and use the data at your own risk !
A conda-smithy repository for jarvis-tools.
This project provides benchmark-performances for materials science applications including Artificial Intelligence (AI), Electronic Structure (ES), Force-field (FF), Quantum Computation (QC) and Experiments (EXP) methods. https://arxiv.org/abs/2306.11688
Matbench: Benchmarks for materials science property prediction
Tools for implementing and consuming OPTIMADE APIs in Python
C++/CUDA package for parallelized simulation of image formation in Scanning Transmission Electron Microscopy (STEM) using the PRISM and multislice algorithms
This repository hosts the providers.json file for OPTIMADE that lists reserved database-specific prefixes and URLs to the index databases of all database providers that participate in the OPTIMADE network
Python Materials Genomics (pymatgen) is a robust materials analysis code that defines core object representations for structures and molecules with support for many electronic structure codes. It is currently the core analysis code powering the Materials Project.
A collection of Jupyter notebooks developed by the community showing how to use Qiskit
Quantum Nature
Automatic generation of crystal structure descriptions.