matbench

Materials Science Benchmarking Tool

Provides tools and resources for testing machine learning performance on materials science data

Matbench: Benchmarks for materials science property prediction

GitHub

132 stars
8 watching
47 forks
Language: Python
last commit: 6 months ago
Linked from 1 awesome list

benchmarkchemistrycondensed-matterdata-sciencemachine-learningmachine-learning-algorithmsmaterials-sciencephysics

Backlinks from these awesome lists:

Related projects:

Repository Description Stars
materialsvirtuallab/maml A toolkit for machine learning in materials science, enabling the development of predictive models and simulations. 376
janosh/matbench-discovery An evaluation framework for machine learning models used in materials discovery 115
ppdebreuck/modnet A Python package implementing a machine learning framework for predicting material properties from composition or crystal structure data. 82
anthony-wang/crabnet A deep learning framework for predicting material properties from composition information. 94
materialsproject/emmet A framework for building and managing material properties databases 55
janosh/pymatviz A toolkit for visualizing data in materials science and informatics 177
sparks-baird/matbench-genmetrics Provides standardized benchmarks for evaluating the quality of generative models for crystal structures. 34
multimodal-art-projection/omnibench Evaluates and benchmarks multimodal language models' ability to process visual, acoustic, and textual inputs simultaneously. 15
materialsintelligence/mat2vec Unsupervised word embeddings capture latent knowledge from materials science literature 624
atomistic-machine-learning/schnetpack A toolbox for training and applying deep neural networks to predict atomistic properties of molecules and materials 795
catboost/benchmarks Comparative benchmarks of various machine learning algorithms 169
mlcommons/inference Measures the performance of deep learning models in various deployment scenarios. 1,256
materialsproject/atomate2 Automates complex materials science workflows using simple Python functions 169
szilard/benchm-ml A benchmark for evaluating machine learning algorithms' performance on large datasets 1,874
usnistgov/jarvis An open-source software package for data-driven materials design using atomistic simulations and machine learning techniques. 318