Git Product home page Git Product logo

judftteam / best-of-atomistic-machine-learning Goto Github PK

View Code? Open in Web Editor NEW
275.0 5.0 21.0 4.1 MB

πŸ† A ranked list of awesome atomistic machine learning projects βš›οΈπŸ§¬πŸ’Ž.

License: Creative Commons Attribution Share Alike 4.0 International

atomistic-machine-learning awesome-list best-of-list computational-chemistry computational-materials-science condensed-matter density-functional-theory drug-discovery electronic-structure interatomic-potentials

best-of-atomistic-machine-learning's Introduction

Best of Atomistic Machine Learning βš›οΈπŸ§¬πŸ’Ž

πŸ†Β  A ranked list of awesome atomistic machine learning (AML) projects. Updated quarterly.

DOI

This curated list contains 360 awesome open-source projects with a total of 180K stars grouped into 22 categories. All projects are ranked by a project-quality score, which is calculated based on various metrics automatically collected from GitHub and different package managers. If you like to add or update projects, feel free to open an issue, submit a pull request, or directly edit the projects.yaml.

The current focus of this list is more on simulation data rather than experimental data, and more on materials rather than drug design. Nevertheless, contributions from other fields are warmly welcome!

πŸ§™β€β™‚οΈ Discover other best-of lists or create your own.

Contents

Explanation

  • πŸ₯‡πŸ₯ˆπŸ₯‰Β  Combined project-quality score
  • ⭐️  Star count from GitHub
  • 🐣  New project (less than 6 months old)
  • πŸ’€Β  Inactive project (6 months no activity)
  • πŸ’€Β  Dead project (12 months no activity)
  • πŸ“ˆπŸ“‰Β  Project is trending up or down
  • βž•Β  Project was recently added
  • πŸ‘¨β€πŸ’»Β  Contributors count from GitHub
  • πŸ”€Β  Fork count from GitHub
  • πŸ“‹Β  Issue count from GitHub
  • ⏱️  Last update timestamp on package manager
  • πŸ“₯Β  Download count from package manager
  • πŸ“¦Β  Number of dependent projects

Active learning

Back to top

Projects that focus on enabling active learning, iterative learning schemes for atomistic ML.

FLARE (πŸ₯‡20 Β· ⭐ 280) - An open-source Python package for creating fast and accurate interatomic potentials. MIT C++ ML-IAP
  • GitHub (πŸ‘¨β€πŸ’» 36 Β· πŸ”€ 63 Β· πŸ“₯ 7 Β· πŸ“¦ 11 Β· πŸ“‹ 200 - 15% open Β· ⏱️ 11.06.2024):

     git clone https://github.com/mir-group/flare
    
IPSuite (πŸ₯ˆ16 Β· ⭐ 16) - A Python toolkit for FAIR development and deployment of machine-learned interatomic potentials. EPL-2.0 ML-IAP MD workflows HTC
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 7 Β· πŸ“¦ 2 Β· πŸ“‹ 120 - 49% open Β· ⏱️ 26.06.2024):

     git clone https://github.com/zincware/IPSuite
    
  • PyPi (πŸ“₯ 59 / month):

     pip install ipsuite
    
Finetuna (πŸ₯‰10 Β· ⭐ 42) - Active Learning for Machine Learning Potentials. MIT
  • GitHub (πŸ‘¨β€πŸ’» 11 Β· πŸ”€ 11 Β· πŸ“‹ 20 - 25% open Β· ⏱️ 15.05.2024):

     git clone https://github.com/ulissigroup/finetuna
    
ACEHAL (πŸ₯‰5 Β· ⭐ 11 Β· πŸ’€) - Hyperactive Learning (HAL) Python interface for building Atomic Cluster Expansion potentials. Unlicensed Julia
  • GitHub (πŸ‘¨β€πŸ’» 3 Β· πŸ”€ 6 Β· πŸ“‹ 10 - 40% open Β· ⏱️ 21.09.2023):

     git clone https://github.com/ACEsuit/ACEHAL
    
Show 1 hidden projects...
  • flare++ (πŸ₯‰10 Β· ⭐ 35 Β· πŸ’€) - A many-body extension of the FLARE code. MIT C++ ML-IAP

Biomolecules

Back to top

Projects that focus on biomolecules, protein structure, protein folding, etc. using atomistic ML.

AlphaFold (πŸ₯‡22 Β· ⭐ 12K Β· πŸ“‰) - Open source code for AlphaFold. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 20 Β· πŸ”€ 2.1K Β· πŸ“¦ 14 Β· πŸ“‹ 840 - 28% open Β· ⏱️ 08.05.2024):

     git clone https://github.com/deepmind/alphafold
    
Uni-Fold (πŸ₯‰16 Β· ⭐ 360) - An open-source platform for developing protein models beyond AlphaFold. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 64 Β· πŸ“₯ 3.5K Β· πŸ“‹ 70 - 27% open Β· ⏱️ 08.01.2024):

     git clone https://github.com/dptech-corp/Uni-Fold
    

Community resources

Back to top

Projects that collect atomistic ML resources or foster communication within community.

πŸ”—Β AI for Science Map - Interactive mindmap of the AI4Science research field, including atomistic machine learning, including papers,..

πŸ”—Β Atomic Cluster Expansion - Atomic Cluster Expansion (ACE) community homepage.

πŸ”—Β CrystaLLM - Generate a crystal structure from a composition. language-models generative pre-trained transformer

πŸ”—Β matsci.org - A community forum for the discussion of anything materials science, with a focus on computational materials science..

πŸ”—Β Matter Modeling Stack Exchange - Machine Learning - Forum StackExchange, site Matter Modeling, ML-tagged questions.

Best-of Machine Learning with Python (πŸ₯‡22 Β· ⭐ 16K) - A ranked list of awesome machine learning Python libraries. Updated weekly. CC-BY-4.0 general-ml Python
  • GitHub (πŸ‘¨β€πŸ’» 45 Β· πŸ”€ 2.2K Β· πŸ“‹ 53 - 35% open Β· ⏱️ 07.06.2024):

     git clone https://github.com/ml-tooling/best-of-ml-python
    
MatBench (πŸ₯‡18 Β· ⭐ 100) - Matbench: Benchmarks for materials science property prediction. MIT datasets benchmarking
  • GitHub (πŸ‘¨β€πŸ’» 25 Β· πŸ”€ 41 Β· πŸ“¦ 15 Β· πŸ“‹ 57 - 54% open Β· ⏱️ 20.01.2024):

     git clone https://github.com/materialsproject/matbench
    
  • PyPi (πŸ“₯ 730 / month):

     pip install matbench
    
GT4SD - Generative Toolkit for Scientific Discovery (πŸ₯ˆ17 Β· ⭐ 320) - Gradio apps of generative models in GT4SD. MIT generative pre-trained drug-discovery
  • GitHub (πŸ‘¨β€πŸ’» 20 Β· πŸ”€ 66 Β· πŸ“‹ 99 - 1% open Β· ⏱️ 04.07.2024):

     git clone https://github.com/GT4SD/gt4sd-core
    
Graph-based Deep Learning Literature (πŸ₯ˆ16 Β· ⭐ 4.7K) - links to conference publications in graph-based deep learning. MIT general-ml rep-learn
  • GitHub (πŸ‘¨β€πŸ’» 12 Β· πŸ”€ 740 Β· ⏱️ 30.03.2024):

     git clone https://github.com/naganandy/graph-based-deep-learning-literature
    
MatBench Discovery (πŸ₯ˆ16 Β· ⭐ 77) - An evaluation framework for machine learning models simulating high-throughput materials discovery. MIT datasets benchmarking
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 9 Β· πŸ“¦ 2 Β· πŸ“‹ 33 - 6% open Β· ⏱️ 08.07.2024):

     git clone https://github.com/janosh/matbench-discovery
    
  • PyPi (πŸ“₯ 86 / month):

     pip install matbench-discovery
    
AI for Science Resources (πŸ₯ˆ12 Β· ⭐ 450) - List of resources for AI4Science research, including learning resources. GPL-3.0 license
  • GitHub (πŸ‘¨β€πŸ’» 28 Β· πŸ”€ 56 Β· πŸ“‹ 13 - 7% open Β· ⏱️ 14.06.2024):

     git clone https://github.com/divelab/AIRS
    
GNoME Explorer (πŸ₯‰10 Β· ⭐ 840 Β· πŸ’€) - Graph Networks for Materials Exploration Database. Apache-2 datasets materials-discovery
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 130 Β· πŸ“‹ 20 - 80% open Β· ⏱️ 02.12.2023):

     git clone https://github.com/google-deepmind/materials_discovery
    
Awesome Materials Informatics (πŸ₯‰9 Β· ⭐ 350) - Curated list of known efforts in materials informatics, i.e. in modern materials science. Custom
  • GitHub (πŸ‘¨β€πŸ’» 19 Β· πŸ”€ 79 Β· ⏱️ 22.06.2024):

     git clone https://github.com/tilde-lab/awesome-materials-informatics
    
MoLFormers UI (πŸ₯‰9 Β· ⭐ 220 Β· πŸ’€) - A family of foundation models trained on chemicals. Apache-2 transformer language-models pre-trained drug-discovery
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 39 Β· πŸ“‹ 18 - 44% open Β· ⏱️ 16.10.2023):

     git clone https://github.com/IBM/molformer
    
optimade.science (πŸ₯‰9 Β· ⭐ 8) - A sky-scanner Optimade browser-only GUI. MIT datasets
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 2 Β· πŸ“‹ 25 - 28% open Β· ⏱️ 10.06.2024):

     git clone https://github.com/tilde-lab/optimade.science
    
Awesome Neural Geometry (πŸ₯‰8 Β· ⭐ 870) - A curated collection of resources and research related to the geometry of representations in the brain, deep networks,.. Unlicensed educational rep-learn
  • GitHub (πŸ‘¨β€πŸ’» 11 Β· πŸ”€ 53 Β· ⏱️ 14.02.2024):

     git clone https://github.com/neurreps/awesome-neural-geometry
    
The Collection of Database and Dataset Resources in Materials Science (πŸ₯‰6 Β· ⭐ 240) - A list of databases, datasets and books/handbooks where you can find materials properties for machine learning.. Unlicensed datasets
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 40 Β· πŸ“‹ 2 - 50% open Β· ⏱️ 07.06.2024):

     git clone https://github.com/sedaoturak/data-resources-for-materials-science
    
Does this material exist? (πŸ₯‰5 Β· ⭐ 15 Β· πŸ“ˆ) - Vote on whether you think predicted crystal structures could be synthesised. MIT for-fun materials-discovery
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 3 Β· ⏱️ 10.04.2024):

     git clone https://github.com/ml-evs/this-material-does-not-exist
    
Show 3 hidden projects...

Datasets

Back to top

Datasets, databases and trained models for atomistic ML.

πŸ”—Β Catalysis Hub - A web-platform for sharing data and software for computational catalysis research!.

πŸ”—Β Citrination Datasets - AI-Powered Materials Data Platform. Open Citrination has been decommissioned.

πŸ”—Β crystals.ai - Curated datasets for reproducible AI in materials science.

πŸ”—Β DeepChem Models - DeepChem models on HuggingFace. pre-trained language-models

πŸ”—Β JARVIS-Leaderboard ( ⭐ 53) - Explore State-of-the-Art Materials Design Methods: https://www.nature.com/articles/s41524-024-01259-w. benchmarking

πŸ”—Β Materials Project - Charge Densities - Materials Project has started offering charge density information available for download via their public API.

πŸ”—Β matterverse.ai - Database of yet-to-be-sythesized materials predicted using state-of-the-art machine learning algorithms.

πŸ”—Β NRELMatDB - Computational materials database with the specific focus on materials for renewable energy applications including, but..

πŸ”—Β Quantum-Machine.org Datasets - Collection of datasets, including QM7, QM9, etc. MD, DFT. Small organic molecules, mostly.

πŸ”—Β sGDML Datasets - MD17, MD22, DFT datasets.

πŸ”—Β MoleculeNet - A Benchmark for Molecular Machine Learning. benchmarking

πŸ”—Β ZINC15 - A free database of commercially-available compounds for virtual screening. ZINC contains over 230 million purchasable.. graph biomolecules

πŸ”—Β ZINC20 - A free database of commercially-available compounds for virtual screening. ZINC contains over 230 million purchasable.. graph biomolecules

OPTIMADE Python tools (πŸ₯‡25 Β· ⭐ 63) - Tools for implementing and consuming OPTIMADE APIs in Python. MIT
  • GitHub (πŸ‘¨β€πŸ’» 28 Β· πŸ”€ 41 Β· πŸ“¦ 41 Β· πŸ“‹ 430 - 20% open Β· ⏱️ 08.07.2024):

     git clone https://github.com/Materials-Consortia/optimade-python-tools
    
  • PyPi (πŸ“₯ 7.4K / month):

     pip install optimade
    
  • Conda (πŸ“₯ 80K Β· ⏱️ 11.06.2024):

     conda install -c conda-forge optimade
    
MPContribs (πŸ₯‡23 Β· ⭐ 34) - Platform for materials scientists to contribute and disseminate their materials data through Materials Project. MIT
  • GitHub (πŸ‘¨β€πŸ’» 25 Β· πŸ”€ 19 Β· πŸ“¦ 38 Β· πŸ“‹ 98 - 20% open Β· ⏱️ 20.06.2024):

     git clone https://github.com/materialsproject/MPContribs
    
  • PyPi (πŸ“₯ 2.8K / month):

     pip install mpcontribs-client
    
FAIR Chemistry datasets (πŸ₯‡21 Β· ⭐ 710) - Datasets OC20, OC22, etc. Formerly known as Open Catalyst Project. MIT catalysis
  • GitHub (πŸ‘¨β€πŸ’» 39 Β· πŸ”€ 220 Β· πŸ“‹ 190 - 3% open Β· ⏱️ 11.07.2024):

     git clone https://github.com/FAIR-Chem/fairchem
    
Open Databases Integration for Materials Design (OPTIMADE) (πŸ₯ˆ18 Β· ⭐ 73) - Specification of a common REST API for access to materials databases. CC-BY-4.0
  • GitHub (πŸ‘¨β€πŸ’» 21 Β· πŸ”€ 35 Β· πŸ“‹ 230 - 27% open Β· ⏱️ 12.06.2024):

     git clone https://github.com/Materials-Consortia/OPTIMADE
    
QH9: A Quantum Hamiltonian Prediction Benchmark (πŸ₯ˆ12 Β· ⭐ 450) - Artificial Intelligence Research for Science (AIRS). CC-BY-NC-SA 4.0 ML-DFT
  • GitHub (πŸ‘¨β€πŸ’» 28 Β· πŸ”€ 56 Β· πŸ“‹ 13 - 7% open Β· ⏱️ 14.06.2024):

     git clone https://github.com/divelab/AIRS
    
SPICE (πŸ₯ˆ11 Β· ⭐ 130) - A collection of QM data for training potential functions. MIT ML-IAP MD
  • GitHub (πŸ”€ 7 Β· πŸ“₯ 240 Β· πŸ“‹ 58 - 25% open Β· ⏱️ 15.04.2024):

     git clone https://github.com/openmm/spice-dataset
    
Materials Data Facility (MDF) (πŸ₯ˆ10 Β· ⭐ 10) - A simple way to publish, discover, and access materials datasets. Publication of very large datasets supported (e.g.,.. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 1 Β· πŸ“‹ 7 - 14% open Β· ⏱️ 05.02.2024):

     git clone https://github.com/materials-data-facility/connect_client
    
2DMD dataset (πŸ₯‰9 Β· ⭐ 5 Β· πŸ’€) - Code for Kazeev, N., Al-Maeeni, A.R., Romanov, I. et al. Sparse representation for machine learning the properties of.. Apache-2 material-defect
  • GitHub (πŸ‘¨β€πŸ’» 11 Β· πŸ”€ 3 Β· ⏱️ 21.11.2023):

     git clone https://github.com/HSE-LAMBDA/ai4material_design
    
3DSC Database (πŸ₯‰5 Β· ⭐ 14) - Repo for the paper publishing the superconductor database with 3D crystal structures. Custom superconductors materials-discovery
  • GitHub (πŸ”€ 4 Β· ⏱️ 08.01.2024):

     git clone https://github.com/aimat-lab/3DSC
    
SciGlass (πŸ₯‰5 Β· ⭐ 10 Β· πŸ’€) - The database contains a vast set of data on the properties of glass materials. MIT
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 3 Β· πŸ“₯ 16 Β· ⏱️ 27.08.2023):

     git clone https://github.com/drcassar/SciGlass
    
Show 12 hidden projects...
  • ATOM3D (πŸ₯ˆ18 Β· ⭐ 290 Β· πŸ’€) - ATOM3D: tasks on molecules in three dimensions. MIT biomolecules benchmarking
  • OpenKIM (πŸ₯ˆ10 Β· ⭐ 31 Β· πŸ’€) - The Open Knowledgebase of Interatomic Models (OpenKIM) aims to be an online resource for standardized testing, long-.. LGPL-2.1 knowledge-base pre-trained
  • ANI-1 Dataset (πŸ₯‰8 Β· ⭐ 95 Β· πŸ’€) - A data set of 20 million calculated off-equilibrium conformations for organic molecules. MIT
  • MoleculeNet Leaderboard (πŸ₯‰8 Β· ⭐ 84 Β· πŸ’€) - MIT benchmarking
  • GEOM (πŸ₯‰7 Β· ⭐ 180 Β· πŸ’€) - GEOM: Energy-annotated molecular conformations. Unlicensed drug-discovery
  • ANI-1x Datasets (πŸ₯‰6 Β· ⭐ 53 Β· πŸ’€) - The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for organic molecules. MIT
  • COMP6 Benchmark dataset (πŸ₯‰6 Β· ⭐ 39 Β· πŸ’€) - COMP6 Benchmark dataset for ML potentials. MIT
  • OPTIMADE providers dashboard (πŸ₯‰6 Β· ⭐ 1) - A dashboard of known providers. Unlicensed
  • Visual Graph Datasets (πŸ₯‰5 Β· ⭐ 1) - Datasets for the training of graph neural networks (GNNs) and subsequent visualization of attributional explanations.. MIT
  • linear-regression-benchmarks (πŸ₯‰5 Β· ⭐ 1 Β· πŸ’€) - Data sets used for linear regression benchmarks. MIT benchmarking single-paper
  • paper-data-redundancy (πŸ₯‰4 Β· ⭐ 6) - Repo for the paper Exploiting redundancy in large materials datasets for efficient machine learning with less data. BSD-3 small-data single-paper
  • nep-data (πŸ₯‰2 Β· ⭐ 12 Β· πŸ’€) - Data related to the NEP machine-learned potential of GPUMD. Unlicensed ML-IAP MD transport-phenomena

Data Structures

Back to top

Projects that focus on providing data structures used in atomistic machine learning.

dpdata (πŸ₯‡24 Β· ⭐ 190) - Manipulating multiple atomic simulation data formats, including DeePMD-kit, VASP, LAMMPS, ABACUS, etc. LGPL-3.0
  • GitHub (πŸ‘¨β€πŸ’» 60 Β· πŸ”€ 120 Β· πŸ“¦ 120 Β· πŸ“‹ 93 - 18% open Β· ⏱️ 06.06.2024):

     git clone https://github.com/deepmodeling/dpdata
    
  • PyPi (πŸ“₯ 51K / month):

     pip install dpdata
    
  • Conda (πŸ“₯ 210 Β· ⏱️ 27.09.2023):

     conda install -c deepmodeling dpdata
    
Metatensor (πŸ₯ˆ21 Β· ⭐ 44 Β· πŸ“ˆ) - Self-describing sparse tensor data format for atomistic machine learning and beyond. BSD-3 Rust C-lang C++ Python
  • GitHub (πŸ‘¨β€πŸ’» 21 Β· πŸ”€ 13 Β· πŸ“₯ 23K Β· πŸ“¦ 10 Β· πŸ“‹ 190 - 35% open Β· ⏱️ 10.07.2024):

     git clone https://github.com/lab-cosmo/metatensor
    
mp-pyrho (πŸ₯‰17 Β· ⭐ 35) - Tools for re-griding volumetric quantum chemistry data for machine-learning purposes. Custom ML-DFT
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 6 Β· πŸ“¦ 22 Β· πŸ“‹ 4 - 25% open Β· ⏱️ 23.02.2024):

     git clone https://github.com/materialsproject/pyrho
    
  • PyPi (πŸ“₯ 4.8K / month):

     pip install mp-pyrho
    
dlpack (πŸ₯‰15 Β· ⭐ 870) - common in-memory tensor structure. Apache-2 C++
  • GitHub (πŸ‘¨β€πŸ’» 23 Β· πŸ”€ 130 Β· πŸ“‹ 68 - 38% open Β· ⏱️ 26.03.2024):

     git clone https://github.com/dmlc/dlpack
    

Density functional theory (ML-DFT)

Back to top

Projects and models that focus on quantities of DFT, such as density functional approximations (ML-DFA), the charge density, density of states, the Hamiltonian, etc.

JAX-DFT (πŸ₯‡25 Β· ⭐ 33K) - This library provides basic building blocks that can construct DFT calculations as a differentiable program. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 790 Β· πŸ”€ 7.7K Β· πŸ“‹ 1.2K - 73% open Β· ⏱️ 10.07.2024):

     git clone https://github.com/google-research/google-research
    
MALA (πŸ₯‡20 Β· ⭐ 80) - Materials Learning Algorithms. A framework for machine learning materials properties from first-principles data. BSD-3
  • GitHub (πŸ‘¨β€πŸ’» 44 Β· πŸ”€ 23 Β· πŸ“¦ 1 Β· πŸ“‹ 260 - 11% open Β· ⏱️ 04.07.2024):

     git clone https://github.com/mala-project/mala
    
SALTED (πŸ₯ˆ14 Β· ⭐ 27) - Symmetry-Adapted Learning of Three-dimensional Electron Densities. GPL-3.0
  • GitHub (πŸ‘¨β€πŸ’» 17 Β· πŸ”€ 4 Β· πŸ“‹ 6 - 16% open Β· ⏱️ 08.07.2024):

     git clone https://github.com/andreagrisafi/SALTED
    
DeepH-pack (πŸ₯ˆ13 Β· ⭐ 190) - Deep neural networks for density functional theory Hamiltonian. LGPL-3.0 Julia
  • GitHub (πŸ‘¨β€πŸ’» 9 Β· πŸ”€ 35 Β· πŸ“‹ 50 - 24% open Β· ⏱️ 22.05.2024):

     git clone https://github.com/mzjb/DeepH-pack
    
QHNet (πŸ₯ˆ12 Β· ⭐ 450) - Artificial Intelligence Research for Science (AIRS). GPL-3.0 rep-learn
  • GitHub (πŸ‘¨β€πŸ’» 28 Β· πŸ”€ 56 Β· πŸ“‹ 13 - 7% open Β· ⏱️ 14.06.2024):

     git clone https://github.com/divelab/AIRS
    
DeePKS-kit (πŸ₯ˆ10 Β· ⭐ 96) - a package for developing machine learning-based chemically accurate energy and density functional models. LGPL-3.0
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 33 Β· πŸ“‹ 18 - 27% open Β· ⏱️ 13.04.2024):

     git clone https://github.com/deepmodeling/deepks-kit
    
Grad DFT (πŸ₯ˆ10 Β· ⭐ 70) - GradDFT is a JAX-based library enabling the differentiable design and experimentation of exchange-correlation.. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 4 Β· πŸ”€ 5 Β· πŸ“‹ 54 - 20% open Β· ⏱️ 13.02.2024):

     git clone https://github.com/XanaduAI/GradDFT
    
Show 19 hidden projects...
  • DM21 (πŸ₯‡20 Β· ⭐ 13K Β· πŸ’€) - This package provides a PySCF interface to the DM21 (DeepMind 21) family of exchange-correlation functionals described.. Apache-2
  • NeuralXC (πŸ₯ˆ11 Β· ⭐ 33 Β· πŸ’€) - Implementation of a machine learned density functional. BSD-3
  • ACEhamiltonians (πŸ₯ˆ11 Β· ⭐ 11 Β· πŸ’€) - Provides tools for constructing, fitting, and predicting self-consistent Hamiltonian and overlap matrices in solid-.. MIT Julia
  • PROPhet (πŸ₯ˆ9 Β· ⭐ 62 Β· πŸ’€) - PROPhet is a code to integrate machine learning techniques with first-principles quantum chemistry approaches. GPL-3.0 ML-IAP MD single-paper C++
  • Libnxc (πŸ₯‰7 Β· ⭐ 15 Β· πŸ’€) - A library for using machine-learned exchange-correlation functionals for density-functional theory. MPL-2.0 C++ Fortran
  • DeepH-E3 (πŸ₯‰6 Β· ⭐ 60 Β· πŸ’€) - General framework for E(3)-equivariant neural network representation of density functional theory Hamiltonian. MIT magnetism
  • DeepDFT (πŸ₯‰6 Β· ⭐ 54 Β· πŸ’€) - Official implementation of DeepDFT model. MIT
  • Mat2Spec (πŸ₯‰6 Β· ⭐ 26 Β· πŸ’€) - MIT spectroscopy
  • xDeepH (πŸ₯‰5 Β· ⭐ 29 Β· πŸ’€) - Extended DeepH (xDeepH) method for magnetic materials. LGPL-3.0 magnetism Julia
  • ML-DFT (πŸ₯‰5 Β· ⭐ 23 Β· πŸ’€) - A package for density functional approximation using machine learning. MIT
  • charge-density-models (πŸ₯‰4 Β· ⭐ 10 Β· πŸ’€) - Tools to build charge density models using fairchem. MIT rep-learn
  • gprep (πŸ₯‰4 Β· πŸ’€) - Fitting DFTB repulsive potentials with GPR. MIT single-paper
  • DeepCDP (πŸ₯‰3 Β· ⭐ 6 Β· πŸ’€) - DeepCDP: Deep learning Charge Density Prediction. Unlicensed
  • APET (πŸ₯‰3 Β· ⭐ 4 Β· πŸ’€) - Atomic Positional Embedding-based Transformer. GPL-3.0 density-of-states transformer
  • CSNN (πŸ₯‰3 Β· ⭐ 2 Β· πŸ’€) - Primary codebase of CSNN - Concentric Spherical Neural Network for 3D Representation Learning. BSD-3
  • MALADA (πŸ₯‰3 Β· ⭐ 1 Β· πŸ’€) - MALA Data Acquisition: Helpful tools to build data for MALA. BSD-3
  • A3MD (πŸ₯‰2 Β· ⭐ 8 Β· πŸ’€) - MPNN-like + Analytic Density Model = Accurate electron densities. Unlicensed representation-learning single-paper
  • kdft (πŸ₯‰1 Β· ⭐ 2 Β· πŸ’€) - The Kernel Density Functional (KDF) code allows generating ML based DFT functionals. Unlicensed
  • MLDensity (πŸ₯‰1 Β· ⭐ 2 Β· πŸ’€) - Linear Jacobi-Legendre expansion of the charge density for machine learning-accelerated electronic structure.. Unlicensed

Educational Resources

Back to top

Tutorials, guides, cookbooks, recipes, etc.

πŸ”—Β Quantum Chemistry in the Age of Machine Learning - Book, 2022.

πŸ”—Β AL4MS 2023 workshop tutorials active-learning

jarvis-tools-notebooks (πŸ₯‡12 Β· ⭐ 51) - A Google-Colab Notebook Collection for Materials Design: https://jarvis.nist.gov/. NIST
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 24 Β· ⏱️ 02.07.2024):

     git clone https://github.com/JARVIS-Materials-Design/jarvis-tools-notebooks
    
Deep Learning for Molecules and Materials Book (πŸ₯‡11 Β· ⭐ 590 Β· πŸ’€) - Deep learning for molecules and materials book. Custom
  • GitHub (πŸ‘¨β€πŸ’» 19 Β· πŸ”€ 110 Β· πŸ“‹ 160 - 17% open Β· ⏱️ 02.07.2023):

     git clone https://github.com/whitead/dmol-book
    
DSECOP (πŸ₯ˆ10 Β· ⭐ 43) - This repository contains data science educational materials developed by DSECOP Fellows. CCO-1.0
  • GitHub (πŸ‘¨β€πŸ’» 13 Β· πŸ”€ 25 Β· πŸ“‹ 8 - 12% open Β· ⏱️ 26.06.2024):

     git clone https://github.com/GDS-Education-Community-of-Practice/DSECOP
    
iam-notebooks (πŸ₯ˆ10 Β· ⭐ 24) - Jupyter notebooks for the lectures of the Introduction to Atomistic Modeling. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 5 Β· ⏱️ 26.06.2024):

     git clone https://github.com/ceriottm/iam-notebooks
    
OPTIMADE Tutorial Exercises (πŸ₯ˆ9 Β· ⭐ 14 Β· πŸ’€) - Tutorial exercises for the OPTIMADE API. MIT datasets
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 7 Β· ⏱️ 27.09.2023):

     git clone https://github.com/Materials-Consortia/optimade-tutorial-exercises
    
BestPractices (πŸ₯ˆ8 Β· ⭐ 160 Β· πŸ’€) - Things that you should (and should not) do in your Materials Informatics research. MIT
  • GitHub (πŸ‘¨β€πŸ’» 3 Β· πŸ”€ 68 Β· πŸ“‹ 7 - 71% open Β· ⏱️ 17.11.2023):

     git clone https://github.com/anthony-wang/BestPractices
    
MACE-tutorials (πŸ₯‰7 Β· ⭐ 29 Β· πŸ’€) - Another set of tutorials for the MACE interatomic potential by one of the authors. MIT ML-IAP rep-learn MD
  • GitHub (πŸ”€ 8 Β· ⏱️ 10.10.2023):

     git clone https://github.com/ilyes319/mace-tutorials
    
COSMO Software Cookbook (πŸ₯‰7 Β· ⭐ 10) - The COSMO cookbook contains recipes for atomic-scale modelling for materials and molecules. BSD-3
  • GitHub (πŸ‘¨β€πŸ’» 9 Β· πŸ”€ 1 Β· πŸ“‹ 12 - 16% open Β· ⏱️ 29.06.2024):

     git clone https://github.com/lab-cosmo/software-cookbook
    
Show 14 hidden projects...
  • Geometric GNN Dojo (πŸ₯‡13 Β· ⭐ 440 Β· πŸ’€) - New to geometric GNNs: try our practical notebook, prepared for MPhil students at the University of Cambridge. MIT rep-learn
  • DeepLearningLifeSciences (πŸ₯‡11 Β· ⭐ 340 Β· πŸ’€) - Example code from the book Deep Learning for the Life Sciences. MIT
  • RDKit Tutorials (πŸ₯ˆ8 Β· ⭐ 250 Β· πŸ’€) - Tutorials to learn how to work with the RDKit. Custom
  • MAChINE (πŸ₯‰7 Β· ⭐ 1 Β· πŸ’€) - Client-Server Web App to introduce usage of ML in materials science to beginners. MIT
  • Applied AI for Materials (πŸ₯‰6 Β· ⭐ 56 Β· πŸ’€) - Course materials for Applied AI for Materials Science and Engineering. Unlicensed
  • AI4Science101 (πŸ₯‰5 Β· ⭐ 82 Β· πŸ’€) - AI for Science. Unlicensed
  • Machine Learning for Materials Hard and Soft (πŸ₯‰5 Β· ⭐ 34 Β· πŸ’€) - ESI-DCAFM-TACO-VDSP Summer School on Machine Learning for Materials Hard and Soft. Unlicensed
  • Data Handling, DoE and Statistical Analysis for Material Chemists (πŸ₯‰5 Β· ⭐ 1 Β· πŸ’€) - Notebooks for workshops of DoE course, hosted by the Computational Materials Chemistry group at Uppsala University. GPL-3.0
  • ML-in-chemistry-101 (πŸ₯‰4 Β· ⭐ 65 Β· πŸ’€) - The course materials for Machine Learning in Chemistry 101. Unlicensed
  • chemrev-gpr (πŸ₯‰4 Β· ⭐ 6 Β· πŸ’€) - Notebooks accompanying the paper on GPR in materials and molecules in Chemical Reviews 2020. Unlicensed
  • MLDensity_tutorial (πŸ₯‰2 Β· ⭐ 7 Β· πŸ’€) - Tutorial files to work with ML for the charge density in molecules and solids. Unlicensed
  • LAMMPS-style pair potentials with GAP (πŸ₯‰2 Β· ⭐ 3 Β· πŸ’€) - A tutorial on how to create LAMMPS-style pair potentials and use them in combination with GAP potentials to run MD.. Unlicensed ML-IAP MD rep-eng
  • MALA Tutorial (πŸ₯‰2 Β· ⭐ 2 Β· πŸ’€) - A full MALA hands-on tutorial. Unlicensed
  • PiNN Lab (πŸ₯‰2 Β· ⭐ 2 Β· πŸ’€) - GPL-3.0

Explainable Artificial intelligence (XAI)

Back to top

Projects that focus on explainability and model interpretability in atomistic ML.

exmol (πŸ₯‡19 Β· ⭐ 280 Β· πŸ’€) - Explainer for black box models that predict molecule properties. MIT
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 42 Β· πŸ“¦ 20 Β· πŸ“‹ 69 - 15% open Β· ⏱️ 04.12.2023):

     git clone https://github.com/ur-whitelab/exmol
    
  • PyPi (πŸ“₯ 580 / month):

     pip install exmol
    
MEGAN: Multi Explanation Graph Attention Student (πŸ₯ˆ6 Β· ⭐ 5) - Minimal implementation of graph attention student model architecture. MIT
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 1 Β· ⏱️ 08.07.2024):

     git clone https://github.com/aimat-lab/graph_attention_student
    
MEGAN (πŸ₯ˆ6 Β· ⭐ 5) - Minimal implementation of graph attention student model architecture. MIT XAI rep-learn
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 1 Β· ⏱️ 08.07.2024):

     git clone https://github.com/aimat-lab/graph_attention_student
    
Show 1 hidden projects...
  • Linear vs blackbox (πŸ₯‰3 Β· ⭐ 2 Β· πŸ’€) - Code and data related to the publication: Interpretable models for extrapolation in scientific machine learning. MIT XAI single-paper rep-eng

Electronic structure methods (ML-ESM)

Back to top

Projects and models that focus on quantities of electronic structure methods, which do not fit into either of the categories ML-WFT or ML-DFT.

Show 3 hidden projects...

General Tools

Back to top

General tools for atomistic machine learning.

DeepChem (πŸ₯‡36 Β· ⭐ 5.3K) - Democratizing Deep-Learning for Drug Discovery, Quantum Chemistry, Materials Science and Biology. MIT
  • GitHub (πŸ‘¨β€πŸ’» 250 Β· πŸ”€ 1.6K Β· πŸ“¦ 390 Β· πŸ“‹ 1.7K - 26% open Β· ⏱️ 10.07.2024):

     git clone https://github.com/deepchem/deepchem
    
  • PyPi (πŸ“₯ 25K / month):

     pip install deepchem
    
  • Conda (πŸ“₯ 110K Β· ⏱️ 05.04.2024):

     conda install -c conda-forge deepchem
    
  • Docker Hub (πŸ“₯ 7.3K Β· ⭐ 5 Β· ⏱️ 10.07.2024):

     docker pull deepchemio/deepchem
    
RDKit (πŸ₯‡32 Β· ⭐ 2.5K) - BSD-3 C++
  • GitHub (πŸ‘¨β€πŸ’» 230 Β· πŸ”€ 830 Β· πŸ“₯ 1.3K Β· πŸ“¦ 3 Β· πŸ“‹ 3.2K - 28% open Β· ⏱️ 10.07.2024):

     git clone https://github.com/rdkit/rdkit
    
  • PyPi (πŸ“₯ 470K / month):

     pip install rdkit
    
  • Conda (πŸ“₯ 2.6M Β· ⏱️ 16.06.2023):

     conda install -c rdkit rdkit
    
Matminer (πŸ₯‡28 Β· ⭐ 450) - Data mining for materials science. Custom
  • GitHub (πŸ‘¨β€πŸ’» 54 Β· πŸ”€ 180 Β· πŸ“¦ 300 Β· πŸ“‹ 220 - 10% open Β· ⏱️ 27.05.2024):

     git clone https://github.com/hackingmaterials/matminer
    
  • PyPi (πŸ“₯ 7.8K / month):

     pip install matminer
    
  • Conda (πŸ“₯ 64K Β· ⏱️ 28.03.2024):

     conda install -c conda-forge matminer
    
QUIP (πŸ₯ˆ25 Β· ⭐ 350) - libAtoms/QUIP molecular dynamics framework: https://libatoms.github.io. GPL-2.0 MD ML-IAP rep-eng Fortran
  • GitHub (πŸ‘¨β€πŸ’» 83 Β· πŸ”€ 120 Β· πŸ“₯ 500 Β· πŸ“¦ 38 Β· πŸ“‹ 460 - 21% open Β· ⏱️ 04.07.2024):

     git clone https://github.com/libAtoms/QUIP
    
  • PyPi (πŸ“₯ 2.8K / month):

     pip install quippy-ase
    
  • Docker Hub (πŸ“₯ 9.9K Β· ⭐ 4 Β· ⏱️ 24.04.2023):

     docker pull libatomsquip/quip
    
MAML (πŸ₯ˆ24 Β· ⭐ 340) - Python for Materials Machine Learning, Materials Descriptors, Machine Learning Force Fields, Deep Learning, etc. BSD-3
  • GitHub (πŸ‘¨β€πŸ’» 32 Β· πŸ”€ 73 Β· πŸ“¦ 9 Β· πŸ“‹ 68 - 10% open Β· ⏱️ 03.07.2024):

     git clone https://github.com/materialsvirtuallab/maml
    
  • PyPi (πŸ“₯ 300 / month):

     pip install maml
    
JARVIS-Tools (πŸ₯ˆ24 Β· ⭐ 280) - JARVIS-Tools: an open-source software package for data-driven atomistic materials design. Publications:.. Custom
  • GitHub (πŸ‘¨β€πŸ’» 15 Β· πŸ”€ 120 Β· πŸ“¦ 90 Β· πŸ“‹ 89 - 49% open Β· ⏱️ 28.06.2024):

     git clone https://github.com/usnistgov/jarvis
    
  • PyPi (πŸ“₯ 13K / month):

     pip install jarvis-tools
    
  • Conda (πŸ“₯ 70K Β· ⏱️ 28.06.2024):

     conda install -c conda-forge jarvis-tools
    
MAST-ML (πŸ₯ˆ20 Β· ⭐ 97) - MAterials Simulation Toolkit for Machine Learning (MAST-ML). MIT
  • GitHub (πŸ‘¨β€πŸ’» 18 Β· πŸ”€ 56 Β· πŸ“₯ 95 Β· πŸ“¦ 43 Β· πŸ“‹ 210 - 10% open Β· ⏱️ 17.04.2024):

     git clone https://github.com/uw-cmg/MAST-ML
    
Scikit-Matter (πŸ₯ˆ19 Β· ⭐ 70) - A collection of scikit-learn compatible utilities that implement methods born out of the materials science and.. BSD-3 scikit-learn
  • GitHub (πŸ‘¨β€πŸ’» 15 Β· πŸ”€ 18 Β· πŸ“¦ 10 Β· πŸ“‹ 69 - 18% open Β· ⏱️ 15.06.2024):

     git clone https://github.com/scikit-learn-contrib/scikit-matter
    
  • PyPi (πŸ“₯ 1.7K / month):

     pip install skmatter
    
  • Conda (πŸ“₯ 1.8K Β· ⏱️ 24.08.2023):

     conda install -c conda-forge skmatter
    
XenonPy (πŸ₯‰14 Β· ⭐ 130) - XenonPy is a Python Software for Materials Informatics. BSD-3
  • GitHub (πŸ‘¨β€πŸ’» 10 Β· πŸ”€ 57 Β· πŸ“₯ 1.4K Β· πŸ“‹ 85 - 22% open Β· ⏱️ 21.04.2024):

     git clone https://github.com/yoshida-lab/XenonPy
    
  • PyPi (πŸ“₯ 250 / month):

     pip install xenonpy
    
Artificial Intelligence for Science (AIRS) (πŸ₯‰12 Β· ⭐ 450) - Artificial Intelligence Research for Science (AIRS). GPL-3.0 license rep-learn generative ML-IAP MD ML-DFT ML-WFT biomolecules
  • GitHub (πŸ‘¨β€πŸ’» 28 Β· πŸ”€ 56 Β· πŸ“‹ 13 - 7% open Β· ⏱️ 14.06.2024):

     git clone https://github.com/divelab/AIRS
    
AMPtorch (πŸ₯‰11 Β· ⭐ 58 Β· πŸ’€) - AMPtorch: Atomistic Machine Learning Package (AMP) - PyTorch. GPL-3.0
  • GitHub (πŸ‘¨β€πŸ’» 14 Β· πŸ”€ 32 Β· πŸ“‹ 33 - 21% open Β· ⏱️ 16.07.2023):

     git clone https://github.com/ulissigroup/amptorch
    
Equisolve (πŸ₯‰6 Β· ⭐ 5 Β· πŸ’€) - A ML toolkit package utilizing the metatensor data format to build models for the prediction of equivariant properties.. BSD-3 ML-IAP
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 1 Β· πŸ“‹ 23 - 82% open Β· ⏱️ 27.10.2023):

     git clone https://github.com/lab-cosmo/equisolve
    
Show 10 hidden projects...
  • QML (πŸ₯ˆ15 Β· ⭐ 200 Β· πŸ’€) - QML: Quantum Machine Learning. MIT
  • Automatminer (πŸ₯‰14 Β· ⭐ 130 Β· πŸ’€) - An automatic engine for predicting materials properties. Custom
  • OpenChem (πŸ₯‰10 Β· ⭐ 670 Β· πŸ’€) - OpenChem: Deep Learning toolkit for Computational Chemistry and Drug Design Research. MIT
  • JAXChem (πŸ₯‰7 Β· ⭐ 74 Β· πŸ’€) - JAXChem is a JAX-based deep learning library for complex and versatile chemical modeling. MIT
  • uncertainty_benchmarking (πŸ₯‰7 Β· ⭐ 38 Β· πŸ’€) - Various code/notebooks to benchmark different ways we could estimate uncertainty in ML predictions. Unlicensed benchmarking probabilistic
  • torchchem (πŸ₯‰7 Β· ⭐ 34 Β· πŸ’€) - An experimental repo for experimenting with PyTorch models. MIT
  • ACEatoms (πŸ₯‰4 Β· ⭐ 2 Β· πŸ’€) - Generic code for modelling atomic properties using ACE. Custom Julia
  • MLatom (πŸ₯‰4) - Machine learning for atomistic simulations. Custom
  • Magpie (πŸ₯‰3) - Materials Agnostic Platform for Informatics and Exploration (Magpie). MIT Java
  • quantum-structure-ml (πŸ₯‰2 Β· ⭐ 2 Β· πŸ’€) - Multi-class classification model for predicting the magnetic order of magnetic structures and a binary classification.. Unlicensed magnetism benchmarking

Generative Models

Back to top

Projects that implement generative models for atomistic ML.

GT4SD (πŸ₯‡19 Β· ⭐ 320) - GT4SD, an open-source library to accelerate hypothesis generation in the scientific discovery process. MIT pre-trained drug-discovery rep-learn
  • GitHub (πŸ‘¨β€πŸ’» 20 Β· πŸ”€ 66 Β· πŸ“‹ 99 - 1% open Β· ⏱️ 04.07.2024):

     git clone https://github.com/GT4SD/gt4sd-core
    
  • PyPi (πŸ“₯ 830 / month):

     pip install gt4sd
    
MoLeR (πŸ₯‡15 Β· ⭐ 250) - Implementation of MoLeR: a generative model of molecular graphs which supports scaffold-constrained generation. MIT
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 36 Β· πŸ“‹ 37 - 24% open Β· ⏱️ 03.01.2024):

     git clone https://github.com/microsoft/molecule-generation
    
  • PyPi (πŸ“₯ 300 / month):

     pip install molecule-generation
    
SchNetPack G-SchNet (πŸ₯ˆ14 Β· ⭐ 42) - G-SchNet extension for SchNetPack. MIT
  • GitHub (πŸ‘¨β€πŸ’» 3 Β· πŸ”€ 8 Β· ⏱️ 05.07.2024):

     git clone https://github.com/atomistic-machine-learning/schnetpack-gschnet
    
bVAE-IM (πŸ₯‰8 Β· ⭐ 11 Β· πŸ’€) - Implementation of Chemical Design with GPU-based Ising Machine. MIT QML single-paper
  • GitHub (πŸ”€ 3 Β· ⏱️ 11.07.2023):

     git clone https://github.com/tsudalab/bVAE-IM
    
COATI (πŸ₯‰5 Β· ⭐ 84) - COATI: multi-modal contrastive pre-training for representing and traversing chemical space. Apache-2 drug-discovery pre-trained rep-learn
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 5 Β· πŸ“‹ 3 - 33% open Β· ⏱️ 23.03.2024):

     git clone https://github.com/terraytherapeutics/COATI
    
Show 6 hidden projects...
  • synspace (πŸ₯ˆ12 Β· ⭐ 35 Β· πŸ’€) - Synthesis generative model. MIT
  • EDM (πŸ₯ˆ10 Β· ⭐ 410 Β· πŸ’€) - E(3) Equivariant Diffusion Model for Molecule Generation in 3D. MIT
  • G-SchNet (πŸ₯‰8 Β· ⭐ 130 Β· πŸ’€) - G-SchNet - a generative model for 3d molecular structures. MIT
  • cG-SchNet (πŸ₯‰8 Β· ⭐ 46 Β· πŸ’€) - cG-SchNet - a conditional generative neural network for 3d molecular structures. MIT
  • rxngenerator (πŸ₯‰5 Β· ⭐ 11 Β· πŸ’€) - A generative model for molecular generation via multi-step chemical reactions. MIT
  • MolSLEPA (πŸ₯‰5 Β· ⭐ 5 Β· πŸ’€) - Interpretable Fragment-based Molecule Design with Self-learning Entropic Population Annealing. MIT XAI

Interatomic Potentials (ML-IAP)

Back to top

Machine learning interatomic potentials (aka ML-IAP, MLIAP, MLIP, MLP) and force fields (ML-FF) for molecular dynamics.

DeePMD-kit (πŸ₯‡27 Β· ⭐ 1.4K Β· πŸ“‰) - A deep learning package for many-body potential energy representation and molecular dynamics. LGPL-3.0 C++
  • GitHub (πŸ‘¨β€πŸ’» 69 Β· πŸ”€ 470 Β· πŸ“₯ 37K Β· πŸ“¦ 16 Β· πŸ“‹ 720 - 9% open Β· ⏱️ 06.07.2024):

     git clone https://github.com/deepmodeling/deepmd-kit
    
  • PyPi (πŸ“₯ 1.6K / month):

     pip install deepmd-kit
    
  • Conda (πŸ“₯ 1.1K Β· ⏱️ 06.04.2024):

     conda install -c deepmodeling deepmd-kit
    
  • Docker Hub (πŸ“₯ 2.5K Β· ⭐ 1 Β· ⏱️ 08.07.2024):

     docker pull deepmodeling/deepmd-kit
    
NequIP (πŸ₯‡25 Β· ⭐ 560) - NequIP is a code for building E(3)-equivariant interatomic potentials. MIT
  • GitHub (πŸ‘¨β€πŸ’» 11 Β· πŸ”€ 120 Β· πŸ“¦ 19 Β· πŸ“‹ 83 - 26% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/mir-group/nequip
    
  • PyPi (πŸ“₯ 2.3K / month):

     pip install nequip
    
  • Conda (πŸ“₯ 4.8K Β· ⏱️ 10.07.2024):

     conda install -c conda-forge nequip
    
TorchANI (πŸ₯‡22 Β· ⭐ 440 Β· πŸ’€) - Accurate Neural Network Potential on PyTorch. MIT
  • GitHub (πŸ‘¨β€πŸ’» 17 Β· πŸ”€ 120 Β· πŸ“¦ 39 Β· πŸ“‹ 170 - 13% open Β· ⏱️ 14.11.2023):

     git clone https://github.com/aiqm/torchani
    
  • PyPi (πŸ“₯ 4.8K / month):

     pip install torchani
    
  • Conda (πŸ“₯ 360K Β· ⏱️ 31.05.2024):

     conda install -c conda-forge torchani
    
MACE (πŸ₯‡22 Β· ⭐ 420) - MACE - Fast and accurate machine learning interatomic potentials with higher order equivariant message passing. MIT
  • GitHub (πŸ‘¨β€πŸ’» 32 Β· πŸ”€ 160 Β· πŸ“‹ 200 - 20% open Β· ⏱️ 10.07.2024):

     git clone https://github.com/ACEsuit/mace
    
GPUMD (πŸ₯‡22 Β· ⭐ 420) - GPUMD is a highly efficient general-purpose molecular dynamic (MD) package and enables machine-learned potentials.. GPL-3.0 MD C++ electrostatics
  • GitHub (πŸ‘¨β€πŸ’» 29 Β· πŸ”€ 110 Β· πŸ“‹ 170 - 10% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/brucefan1983/GPUMD
    
TorchMD-NET (πŸ₯‡22 Β· ⭐ 300) - Training neural network potentials. MIT MD rep-learn transformer pre-trained
  • GitHub (πŸ‘¨β€πŸ’» 16 Β· πŸ”€ 67 Β· πŸ“‹ 100 - 19% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/torchmd/torchmd-net
    
  • Conda (πŸ“₯ 75K Β· ⏱️ 09.07.2024):

     conda install -c conda-forge torchmd-net
    
DP-GEN (πŸ₯‡22 Β· ⭐ 290) - The deep potential generator to generate a deep-learning based model of interatomic potential energy and force field. LGPL-3.0 workflows
  • GitHub (πŸ‘¨β€πŸ’» 64 Β· πŸ”€ 170 Β· πŸ“₯ 1.7K Β· πŸ“¦ 6 Β· πŸ“‹ 280 - 9% open Β· ⏱️ 10.04.2024):

     git clone https://github.com/deepmodeling/dpgen
    
  • PyPi (πŸ“₯ 430 / month):

     pip install dpgen
    
  • Conda (πŸ“₯ 200 Β· ⏱️ 16.06.2023):

     conda install -c deepmodeling dpgen
    
CHGNet (πŸ₯‡22 Β· ⭐ 210) - Pretrained universal neural network potential for charge-informed atomistic modeling https://chgnet.lbl.gov. Custom MD pre-trained electrostatics magnetism structure-relaxation
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 54 Β· πŸ“¦ 25 Β· πŸ“‹ 49 - 4% open Β· ⏱️ 07.07.2024):

     git clone https://github.com/CederGroupHub/chgnet
    
  • PyPi (πŸ“₯ 18K / month):

     pip install chgnet
    
fairchem (πŸ₯ˆ19 Β· ⭐ 710) - FAIR Chemistrys library of machine learning methods for chemistry. Formerly known as Open Catalyst Project (ocp). Unlicensed pre-trained rep-learn catalysis
  • GitHub (πŸ‘¨β€πŸ’» 39 Β· πŸ”€ 220 Β· πŸ“‹ 190 - 3% open Β· ⏱️ 11.07.2024):

     git clone https://github.com/FAIR-Chem/fairchem
    
Ultra-Fast Force Fields (UF3) (πŸ₯ˆ15 Β· ⭐ 55) - UF3: a python library for generating ultra-fast interatomic potentials. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 10 Β· πŸ”€ 20 Β· πŸ“¦ 1 Β· πŸ“‹ 42 - 30% open Β· ⏱️ 03.07.2024):

     git clone https://github.com/uf3/uf3
    
  • PyPi (πŸ“₯ 24 / month):

     pip install uf3
    
KLIFF (πŸ₯ˆ15 Β· ⭐ 34) - KIM-based Learning-Integrated Fitting Framework for interatomic potentials. LGPL-2.1 probabilistic workflows
  • GitHub (πŸ‘¨β€πŸ’» 9 Β· πŸ”€ 20 Β· πŸ“¦ 3 Β· πŸ“‹ 40 - 52% open Β· ⏱️ 06.07.2024):

     git clone https://github.com/openkim/kliff
    
  • PyPi (πŸ“₯ 62 / month):

     pip install kliff
    
  • Conda (πŸ“₯ 87K Β· ⏱️ 18.12.2023):

     conda install -c conda-forge kliff
    
wfl (πŸ₯ˆ15 Β· ⭐ 23) - Workflow is a Python toolkit for building interatomic potential creation and atomistic simulation workflows. GPL-2.0 workflows HTC
  • GitHub (πŸ‘¨β€πŸ’» 18 Β· πŸ”€ 16 Β· πŸ“¦ 1 Β· πŸ“‹ 150 - 42% open Β· ⏱️ 08.07.2024):

     git clone https://github.com/libAtoms/workflow
    
sGDML (πŸ₯ˆ14 Β· ⭐ 140 Β· πŸ’€) - sGDML - Reference implementation of the Symmetric Gradient Domain Machine Learning model. MIT
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 36 Β· πŸ“¦ 9 Β· πŸ“‹ 17 - 35% open Β· ⏱️ 31.08.2023):

     git clone https://github.com/stefanch/sGDML
    
  • PyPi (πŸ“₯ 120 / month):

     pip install sgdml
    
PyXtalFF (πŸ₯ˆ14 Β· ⭐ 85) - Machine Learning Interatomic Potential Predictions. MIT
  • GitHub (πŸ‘¨β€πŸ’» 9 Β· πŸ”€ 23 Β· πŸ“‹ 62 - 17% open Β· ⏱️ 07.01.2024):

     git clone https://github.com/MaterSim/PyXtal_FF
    
  • PyPi (πŸ“₯ 59 / month):

     pip install pyxtal_ff
    
apax (πŸ₯ˆ14 Β· ⭐ 12 Β· πŸ“‰) - A flexible and performant framework for training machine learning potentials. MIT
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 1 Β· πŸ“¦ 2 Β· πŸ“‹ 110 - 16% open Β· ⏱️ 17.05.2024):

     git clone https://github.com/apax-hub/apax
    
  • PyPi (πŸ“₯ 49 / month):

     pip install apax
    
NNPOps (πŸ₯ˆ13 Β· ⭐ 79) - High-performance operations for neural network potentials. MIT MD C++
  • GitHub (πŸ‘¨β€πŸ’» 9 Β· πŸ”€ 17 Β· πŸ“‹ 55 - 38% open Β· ⏱️ 10.07.2024):

     git clone https://github.com/openmm/NNPOps
    
  • Conda (πŸ“₯ 160K Β· ⏱️ 31.05.2024):

     conda install -c conda-forge nnpops
    
CCS_fit (πŸ₯ˆ13 Β· ⭐ 7) - Curvature Constrained Splines. GPL-3.0
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 10 Β· πŸ“₯ 400 Β· πŸ“‹ 14 - 57% open Β· ⏱️ 16.02.2024):

     git clone https://github.com/Teoroo-CMC/CCS
    
  • PyPi (πŸ“₯ 220 / month):

     pip install ccs_fit
    
Neural Force Field (πŸ₯ˆ12 Β· ⭐ 220) - Neural Network Force Field based on PyTorch. MIT pre-trained
  • GitHub (πŸ‘¨β€πŸ’» 10 Β· πŸ”€ 46 Β· ⏱️ 04.07.2024):

     git clone https://github.com/learningmatter-mit/NeuralForceField
    
ANI-1 (πŸ₯ˆ12 Β· ⭐ 220) - ANI-1 neural net potential with python interface (ASE). MIT
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 55 Β· πŸ“‹ 37 - 43% open Β· ⏱️ 11.03.2024):

     git clone https://github.com/isayev/ASE_ANI
    
DMFF (πŸ₯ˆ12 Β· ⭐ 140) - DMFF (Differentiable Molecular Force Field) is a Jax-based python package that provides a full differentiable.. LGPL-3.0
  • GitHub (πŸ‘¨β€πŸ’» 14 Β· πŸ”€ 40 Β· πŸ“‹ 26 - 38% open Β· ⏱️ 12.01.2024):

     git clone https://github.com/deepmodeling/DMFF
    
PiNN (πŸ₯ˆ11 Β· ⭐ 100) - A Python library for building atomic neural networks. BSD-3
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 30 Β· πŸ“‹ 6 - 16% open Β· ⏱️ 27.06.2024):

     git clone https://github.com/Teoroo-CMC/PiNN
    
  • Docker Hub (πŸ“₯ 240 Β· ⏱️ 27.06.2024):

     docker pull teoroo/pinn
    
Pacemaker (πŸ₯ˆ11 Β· ⭐ 57) - Python package for fitting atomic cluster expansion (ACE) potentials. Custom
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 15 Β· πŸ“‹ 45 - 28% open Β· ⏱️ 16.02.2024):

     git clone https://github.com/ICAMS/python-ace
    
  • PyPi (πŸ“₯ 19 / month):

     pip install python-ace
    
So3krates (MLFF) (πŸ₯ˆ11 Β· ⭐ 57) - Build neural networks for machine learning force fields with JAX. MIT
  • GitHub (πŸ‘¨β€πŸ’» 4 Β· πŸ”€ 13 Β· πŸ“‹ 9 - 44% open Β· ⏱️ 04.07.2024):

     git clone https://github.com/thorben-frank/mlff
    
tinker-hp (πŸ₯‰10 Β· ⭐ 77) - Tinker-HP: High-Performance Massively Parallel Evolution of Tinker on CPUs & GPUs. Custom
  • GitHub (πŸ‘¨β€πŸ’» 10 Β· πŸ”€ 21 Β· πŸ“‹ 19 - 15% open Β· ⏱️ 04.07.2024):

     git clone https://github.com/TinkerTools/tinker-hp
    
Point Edge Transformer (PET) (πŸ₯‰10 Β· ⭐ 16) - Point Edge Transformer. MIT rep-learn transformer
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 5 Β· ⏱️ 02.07.2024):

     git clone https://github.com/spozdn/pet
    
DimeNet (πŸ₯‰9 Β· ⭐ 280 Β· πŸ’€) - DimeNet and DimeNet++ models, as proposed in Directional Message Passing for Molecular Graphs (ICLR 2020) and Fast and.. Custom
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 57 Β· πŸ“¦ 1 Β· πŸ“‹ 31 - 3% open Β· ⏱️ 03.10.2023):

     git clone https://github.com/gasteigerjo/dimenet
    
TurboGAP (πŸ₯‰9 Β· ⭐ 16) - The TurboGAP code. Custom Fortran
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 8 Β· πŸ“‹ 8 - 62% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/mcaroba/turbogap
    
ACEfit (πŸ₯‰9 Β· ⭐ 8) - MIT Julia
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 5 Β· πŸ“‹ 55 - 40% open Β· ⏱️ 28.03.2024):

     git clone https://github.com/ACEsuit/ACEfit.jl
    
MACE-Jax (πŸ₯‰8 Β· ⭐ 51 Β· πŸ’€) - Equivariant machine learning interatomic potentials in JAX. MIT
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 3 Β· πŸ“‹ 5 - 40% open Β· ⏱️ 04.10.2023):

     git clone https://github.com/ACEsuit/mace-jax
    
PyNEP (πŸ₯‰8 Β· ⭐ 42) - A python interface of the machine learning potential NEP used in GPUMD. MIT
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 16 Β· πŸ“‹ 11 - 36% open Β· ⏱️ 01.06.2024):

     git clone https://github.com/bigd4/PyNEP
    
GAP (πŸ₯‰8 Β· ⭐ 39) - Gaussian Approximation Potential (GAP). Custom
  • GitHub (πŸ‘¨β€πŸ’» 13 Β· πŸ”€ 20 Β· ⏱️ 03.07.2024):

     git clone https://github.com/libAtoms/GAP
    
ACE1.jl (πŸ₯‰8 Β· ⭐ 20) - Atomic Cluster Expansion for Modelling Invariant Atomic Properties. Custom Julia
  • GitHub (πŸ‘¨β€πŸ’» 9 Β· πŸ”€ 7 Β· πŸ“‹ 46 - 47% open Β· ⏱️ 02.07.2024):

     git clone https://github.com/ACEsuit/ACE1.jl
    
ALF (πŸ₯‰7 Β· ⭐ 28) - A framework for performing active learning for training machine-learned interatomic potentials. Custom active-learning
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 11 Β· ⏱️ 28.05.2024):

     git clone https://github.com/lanl/alf
    
MLXDM (πŸ₯‰7 Β· ⭐ 5) - A Neural Network Potential with Rigorous Treatment of Long-Range Dispersion https://doi.org/10.1039/D2DD00150K. MIT long-range
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 2 Β· ⏱️ 08.07.2024):

     git clone https://github.com/RowleyGroup/MLXDM
    
ACE1Pack.jl (πŸ₯‰6 Β· πŸ’€) - Provides convenience functionality for the usage of ACE1.jl, ACEfit.jl, JuLIP.jl for fitting interatomic potentials.. MIT Julia
  • GitHub (πŸ‘¨β€πŸ’» 11 Β· ⏱️ 21.08.2023):

     git clone https://github.com/ACEsuit/ACE1pack.jl
    
Allegro-JAX ( ⭐ 16) - JAX implementation of the Allegro interatomic potential. Unlicensed
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 2 Β· ⏱️ 09.04.2024):

     git clone https://github.com/mariogeiger/allegro-jax
    
Show 29 hidden projects...
  • MEGNet (πŸ₯‡22 Β· ⭐ 490 Β· πŸ’€) - Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals. BSD-3
  • M3GNet (πŸ₯ˆ17 Β· ⭐ 220 Β· πŸ’€) - Materials graph network with 3-body interactions featuring a DFT surrogate crystal relaxer and a state-of-the-art.. BSD-3
  • n2p2 (πŸ₯ˆ13 Β· ⭐ 210 Β· πŸ’€) - n2p2 - A Neural Network Potential Package. GPL-3.0 C++
  • TensorMol (πŸ₯ˆ12 Β· ⭐ 270 Β· πŸ’€) - Tensorflow + Molecules = TensorMol. GPL-3.0 single-paper
  • SIMPLE-NN (πŸ₯ˆ11 Β· ⭐ 47 Β· πŸ’€) - SIMPLE-NN(SNU Interatomic Machine-learning PotentiaL packagE version Neural Network). GPL-3.0
  • Allegro (πŸ₯‰10 Β· ⭐ 300 Β· πŸ’€) - Allegro is an open-source code for building highly scalable and accurate equivariant deep learning interatomic.. MIT
  • SchNet (πŸ₯‰9 Β· ⭐ 210 Β· πŸ’€) - SchNet - a deep learning architecture for quantum chemistry. MIT
  • GemNet (πŸ₯‰9 Β· ⭐ 170 Β· πŸ’€) - GemNet model in PyTorch, as proposed in GemNet: Universal Directional Graph Neural Networks for Molecules (NeurIPS.. Custom
  • ACE.jl (πŸ₯‰9 Β· ⭐ 66 Β· πŸ’€) - Parameterisation of Equivariant Properties of Particle Systems. Custom Julia
  • NNsforMD (πŸ₯‰9 Β· ⭐ 10 Β· πŸ’€) - Neural network class for molecular dynamics to predict potential energy, forces and non-adiabatic couplings. MIT
  • AIMNet (πŸ₯‰8 Β· ⭐ 92 Β· πŸ’€) - Atoms In Molecules Neural Network Potential. MIT single-paper
  • SNAP (πŸ₯‰8 Β· ⭐ 36 Β· πŸ’€) - Repository for spectral neighbor analysis potential (SNAP) model development. BSD-3
  • Atomistic Adversarial Attacks (πŸ₯‰8 Β· ⭐ 29 Β· πŸ’€) - Code for performing adversarial attacks on atomistic systems using NN potentials. MIT probabilistic
  • PhysNet (πŸ₯‰7 Β· ⭐ 89 Β· πŸ’€) - Code for training PhysNet models. MIT electrostatics
  • SIMPLE-NN v2 (πŸ₯‰7 Β· ⭐ 38 Β· πŸ’€) - GPL-3.0
  • calorine (πŸ₯‰7 Β· ⭐ 12 Β· πŸ’€) - A Python package for constructing and sampling neuroevolution potential models. https://doi.org/10.21105/joss.06264. Custom
  • MLIP-3 (πŸ₯‰6 Β· ⭐ 25 Β· πŸ’€) - MLIP-3: Active learning on atomic environments with Moment Tensor Potentials (MTP). BSD-2 C++
  • testing-framework (πŸ₯‰6 Β· ⭐ 11 Β· πŸ’€) - The purpose of this repository is to aid the testing of a large number of interatomic potentials for a variety of.. Unlicensed benchmarking
  • PANNA (πŸ₯‰6 Β· ⭐ 9 Β· πŸ’€) - A package to train and validate all-to-all connected network models for BP[1] and modified-BP[2] type local atomic.. MIT benchmarking
  • Alchemical learning (πŸ₯‰5 Β· ⭐ 2 Β· πŸ’€) - Code for the Modeling high-entropy transition metal alloys with alchemical compression article. BSD-3
  • glp (πŸ₯‰4 Β· ⭐ 17) - tools for graph-based machine-learning potentials in jax. MIT
  • NequIP-JAX (πŸ₯‰4 Β· ⭐ 16 Β· πŸ’€) - JAX implementation of the NequIP interatomic potential. Unlicensed
  • TensorPotential (πŸ₯‰4 Β· ⭐ 7 Β· πŸ’€) - Tensorpotential is a TensorFlow based tool for development, fitting ML interatomic potentials from electronic.. Custom
  • ACE Workflows (πŸ₯‰4 Β· πŸ’€) - Workflow Examples for ACE Models. Unlicensed Julia workflows
  • PeriodicPotentials (πŸ₯‰4 Β· πŸ’€) - A Periodic table app that displays potentials based on the selected elements. MIT community-resource viz JavaScript
  • MEGNetSparse (πŸ₯‰3 Β· ⭐ 1 Β· πŸ’€) - A library imlementing a graph neural network with sparse representation from Code for Kazeev, N., Al-Maeeni, A.R.,.. MIT material-defect
  • SingleNN (πŸ₯‰2 Β· ⭐ 8 Β· πŸ’€) - An efficient package for training and executing neural-network interatomic potentials. Unlicensed C++
  • RuNNer (πŸ₯‰2) - The RuNNer Neural Network Energy Representation is a Fortran-based framework for the construction of Behler-.. GPL-3.0 Fortran
  • mlp (πŸ₯‰1 Β· ⭐ 1 Β· πŸ’€) - Proper orthogonal descriptors for efficient and accurate interatomic potentials... Unlicensed Julia

Language Models

Back to top

Projects that use (large) language models (LMs, LLMs) or natural language procesing (NLP) techniques for atomistic ML.

paper-qa (πŸ₯‡26 Β· ⭐ 3.8K) - LLM Chain for answering questions from documents with citations. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 17 Β· πŸ”€ 350 Β· πŸ“¦ 64 Β· πŸ“‹ 140 - 48% open Β· ⏱️ 28.06.2024):

     git clone https://github.com/whitead/paper-qa
    
  • PyPi (πŸ“₯ 4.4K / month):

     pip install paper-qa
    
ChemCrow (πŸ₯‡16 Β· ⭐ 520) - Chemcrow. MIT
  • GitHub (πŸ‘¨β€πŸ’» 3 Β· πŸ”€ 69 Β· πŸ“¦ 4 Β· πŸ“‹ 16 - 12% open Β· ⏱️ 27.03.2024):

     git clone https://github.com/ur-whitelab/chemcrow-public
    
  • PyPi (πŸ“₯ 540 / month):

     pip install chemcrow
    
ChemNLP project (πŸ₯ˆ14 Β· ⭐ 140) - ChemNLP project. MIT datasets
  • GitHub (πŸ‘¨β€πŸ’» 26 Β· πŸ”€ 46 Β· πŸ“‹ 250 - 44% open Β· ⏱️ 01.04.2024):

     git clone https://github.com/OpenBioML/chemnlp
    
  • PyPi (πŸ“₯ 46 / month):

     pip install chemnlp
    
gptchem (πŸ₯ˆ13 Β· ⭐ 220 Β· πŸ’€) - Use GPT-3 to solve chemistry problems. MIT
  • GitHub (πŸ‘¨β€πŸ’» 4 Β· πŸ”€ 39 Β· πŸ“‹ 21 - 90% open Β· ⏱️ 04.10.2023):

     git clone https://github.com/kjappelbaum/gptchem
    
  • PyPi (πŸ“₯ 50 / month):

     pip install gptchem
    
LLaMP (πŸ₯ˆ12 Β· ⭐ 45) - A web app and Python API for multi-modal RAG framework to ground LLMs on high-fidelity materials informatics. An.. BSD-3 materials-discovery cheminformatics generative MD language-models Python general-tool
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 4 Β· πŸ“‹ 25 - 32% open Β· ⏱️ 03.07.2024):

     git clone https://github.com/chiang-yuan/llamp
    
MoLFormer (πŸ₯‰9 Β· ⭐ 220 Β· πŸ’€) - Repository for MolFormer. Apache-2 transformer pre-trained drug-discovery
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 39 Β· πŸ“‹ 18 - 44% open Β· ⏱️ 16.10.2023):

     git clone https://github.com/IBM/molformer
    
MolSkill (πŸ₯‰9 Β· ⭐ 99 Β· πŸ’€) - Extracting medicinal chemistry intuition via preference machine learning. MIT drug-discovery recommender
  • GitHub (πŸ‘¨β€πŸ’» 4 Β· πŸ”€ 8 Β· πŸ“‹ 5 - 40% open Β· ⏱️ 31.10.2023):

     git clone https://github.com/microsoft/molskill
    
  • Conda (πŸ“₯ 260 Β· ⏱️ 18.06.2023):

     conda install -c msr-ai4science molskill
    
chemlift (πŸ₯‰7 Β· ⭐ 30 Β· πŸ’€) - Language-interfaced fine-tuning for chemistry. MIT
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 3 Β· πŸ“‹ 18 - 61% open Β· ⏱️ 14.10.2023):

     git clone https://github.com/lamalab-org/chemlift
    
LLM-Prop (πŸ₯‰6 Β· ⭐ 24) - A repository for the LLM-Prop implementation. MIT
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 5 Β· πŸ“‹ 2 - 50% open Β· ⏱️ 26.04.2024):

     git clone https://github.com/vertaix/LLM-Prop
    
SciBot (πŸ₯‰5 Β· ⭐ 28) - SciBot is a simple demo of building a domain-specific chatbot for science. Unlicensed
  • GitHub (πŸ”€ 8 Β· πŸ“¦ 1 Β· ⏱️ 19.04.2024):

     git clone https://github.com/CFN-softbio/SciBot
    
BERT-PSIE-TC (πŸ₯‰5 Β· ⭐ 10 Β· πŸ’€) - A dataset of Curie temperatures automatically extracted from scientific literature with the use of the BERT-PSIE.. MIT magnetism
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 3 Β· ⏱️ 18.08.2023):

     git clone https://github.com/StefanoSanvitoGroup/BERT-PSIE-TC
    
MAPI_LLM (πŸ₯‰5 Β· ⭐ 8) - A LLM application developed during the LLM March MADNESS Hackathon https://doi.org/10.1039/D3DD00113J. MIT dataset
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 2 Β· ⏱️ 11.04.2024):

     git clone https://github.com/maykcaldas/MAPI_LLM
    
Show 5 hidden projects...
  • ChemDataExtractor (πŸ₯‡16 Β· ⭐ 290 Β· πŸ’€) - Automatically extract chemical information from scientific documents. MIT literature-data
  • mat2vec (πŸ₯ˆ12 Β· ⭐ 620 Β· πŸ’€) - Supplementary Materials for Tshitoyan et al. Unsupervised word embeddings capture latent knowledge from materials.. MIT rep-learn
  • nlcc (πŸ₯‰11 Β· ⭐ 44 Β· πŸ’€) - Natural language computational chemistry command line interface. MIT single-paper
  • ChemDataWriter (πŸ₯‰4 Β· ⭐ 14 Β· πŸ’€) - ChemDataWriter is a transformer-based library for automatically generating research books in the chemistry area. MIT literature-data
  • CatBERTa (πŸ₯‰3 Β· ⭐ 19) - Large Language Model for Catalyst Property Prediction. Unlicensed transformer catalysis

Materials Discovery

Back to top

Projects that implement materials discovery methods using atomistic ML.

aviary (πŸ₯‡13 Β· ⭐ 44) - The Wren sits on its Roost in the Aviary. MIT
  • GitHub (πŸ‘¨β€πŸ’» 4 Β· πŸ”€ 10 Β· πŸ“‹ 26 - 7% open Β· ⏱️ 11.07.2024):

     git clone https://github.com/CompRhys/aviary
    
Materials Discovery: GNoME (πŸ₯ˆ10 Β· ⭐ 840 Β· πŸ’€) - Graph Networks for Materials Science (GNoME) and dataset of 381,000 novel stable materials. Apache-2 rep-learn datasets
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 130 Β· πŸ“‹ 20 - 80% open Β· ⏱️ 02.12.2023):

     git clone https://github.com/google-deepmind/materials_discovery
    
CSPML (crystal structure prediction with machine learning-based element substitution) (πŸ₯‰5 Β· ⭐ 16 Β· πŸ“ˆ) - Original implementation of CSPML. Unlicensed structure-prediction
  • GitHub (πŸ”€ 8 Β· πŸ“‹ 3 - 66% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/minoru938/cspml
    
Show 6 hidden projects...
  • BOSS (πŸ₯ˆ7 Β· ⭐ 20 Β· πŸ’€) - Bayesian Optimization Structure Search (BOSS). Unlicensed probabilistic
  • AGOX (πŸ₯ˆ6 Β· ⭐ 13 Β· πŸ’€) - AGOX is a package for global optimization of atomic system using e.g. the energy calculated from density functional.. GPL-3.0 structure-optimization
  • Computational Autonomy for Materials Discovery (CAMD) (πŸ₯ˆ6 Β· ⭐ 1 Β· πŸ’€) - Agent-based sequential learning software for materials discovery. Apache-2
  • closed-loop-acceleration-benchmarks (πŸ₯‰4 Β· πŸ’€) - Data and scripts in support of the publication By how much can closed-loop frameworks accelerate computational.. MIT materials-discovery active-learning single-paper
  • SPINNER (πŸ₯‰3 Β· ⭐ 11 Β· πŸ’€) - SPINNER (Structure Prediction of Inorganic crystals using Neural Network potentials with Evolutionary and Random.. GPL-3.0 C++ structure-prediction
  • sl_discovery (πŸ₯‰3 Β· ⭐ 5 Β· πŸ’€) - Data processing and models related to Quantifying the performance of machine learning models in materials discovery. Apache-2 materials-discovery single-paper

Mathematical tools

Back to top

Projects that implement mathematical objects used in atomistic machine learning.

KFAC-JAX (πŸ₯‡19 Β· ⭐ 210 Β· πŸ“‰) - Second Order Optimization and Curvature Estimation with K-FAC in JAX. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 13 Β· πŸ”€ 16 Β· πŸ“¦ 10 Β· πŸ“‹ 13 - 30% open Β· ⏱️ 24.06.2024):

     git clone https://github.com/deepmind/kfac-jax
    
  • PyPi (πŸ“₯ 880 / month):

     pip install kfac-jax
    
gpax (πŸ₯‡17 Β· ⭐ 190) - Gaussian Processes for Experimental Sciences. MIT probabilistic active-learning
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 23 Β· πŸ“¦ 1 Β· πŸ“‹ 39 - 17% open Β· ⏱️ 21.05.2024):

     git clone https://github.com/ziatdinovmax/gpax
    
  • PyPi (πŸ“₯ 270 / month):

     pip install gpax
    
SpheriCart (πŸ₯ˆ15 Β· ⭐ 58) - Multi-language library for the calculation of spherical harmonics in Cartesian coordinates. MIT
  • GitHub (πŸ‘¨β€πŸ’» 10 Β· πŸ”€ 10 Β· πŸ“₯ 60 Β· πŸ“¦ 3 Β· πŸ“‹ 28 - 60% open Β· ⏱️ 12.06.2024):

     git clone https://github.com/lab-cosmo/sphericart
    
  • PyPi (πŸ“₯ 100 / month):

     pip install sphericart
    
Polynomials4ML.jl (πŸ₯ˆ15 Β· ⭐ 12) - Polynomials for ML: fast evaluation, batching, differentiation. MIT Julia
  • GitHub (πŸ‘¨β€πŸ’» 10 Β· πŸ”€ 5 Β· πŸ“‹ 51 - 33% open Β· ⏱️ 22.06.2024):

     git clone https://github.com/ACEsuit/Polynomials4ML.jl
    
GElib (πŸ₯ˆ9 Β· ⭐ 19) - C++/CUDA library for SO(3) equivariant operations. MPL-2.0 C++
  • GitHub (πŸ‘¨β€πŸ’» 4 Β· πŸ”€ 3 Β· πŸ“‹ 5 - 40% open Β· ⏱️ 26.04.2024):

     git clone https://github.com/risi-kondor/GElib
    
COSMO Toolbox (πŸ₯‰6 Β· ⭐ 6) - Assorted libraries and utilities for atomistic simulation analysis. Unlicensed C++
  • GitHub (πŸ‘¨β€πŸ’» 9 Β· πŸ”€ 5 Β· ⏱️ 19.03.2024):

     git clone https://github.com/lab-cosmo/toolbox
    
Show 5 hidden projects...
  • lie-nn (πŸ₯ˆ9 Β· ⭐ 26 Β· πŸ’€) - Tools for building equivariant polynomials on reductive Lie groups. MIT rep-learn
  • cnine (πŸ₯‰6 Β· ⭐ 4) - Cnine tensor library. Unlicensed C++
  • EquivariantOperators.jl (πŸ₯‰5 Β· ⭐ 18 Β· πŸ’€) - MIT Julia
  • torch_spex (πŸ₯‰3 Β· ⭐ 3) - Spherical expansions in PyTorch. Unlicensed
  • Wigner Kernels (πŸ₯‰2 Β· ⭐ 1 Β· πŸ’€) - Collection of programs to benchmark Wigner kernels. Unlicensed benchmarking

Molecular Dynamics

Back to top

Projects that simplify the integration of molecular dynamics and atomistic machine learning.

JAX-MD (πŸ₯‡23 Β· ⭐ 1.1K Β· πŸ“‰) - Differentiable, Hardware Accelerated, Molecular Dynamics. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 31 Β· πŸ”€ 180 Β· πŸ“¦ 52 Β· πŸ“‹ 150 - 45% open Β· ⏱️ 17.04.2024):

     git clone https://github.com/jax-md/jax-md
    
  • PyPi (πŸ“₯ 2.5K / month):

     pip install jax-md
    
mlcolvar (πŸ₯ˆ19 Β· ⭐ 88) - A unified framework for machine learning collective variables for enhanced sampling simulations. MIT enhanced-sampling
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 21 Β· πŸ“¦ 2 Β· πŸ“‹ 70 - 18% open Β· ⏱️ 14.06.2024):

     git clone https://github.com/luigibonati/mlcolvar
    
  • PyPi (πŸ“₯ 310 / month):

     pip install mlcolvar
    
openmm-torch (πŸ₯ˆ17 Β· ⭐ 170) - OpenMM plugin to define forces with neural networks. Custom ML-IAP C++
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 24 Β· πŸ“‹ 89 - 28% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/openmm/openmm-torch
    
  • Conda (πŸ“₯ 350K Β· ⏱️ 03.06.2024):

     conda install -c conda-forge openmm-torch
    
FitSNAP (πŸ₯ˆ17 Β· ⭐ 140) - Software for generating SNAP machine-learning interatomic potentials. GPL-2.0
  • GitHub (πŸ‘¨β€πŸ’» 24 Β· πŸ”€ 46 Β· πŸ“₯ 8 Β· πŸ“¦ 2 Β· πŸ“‹ 65 - 13% open Β· ⏱️ 27.06.2024):

     git clone https://github.com/FitSNAP/FitSNAP
    
  • Conda (πŸ“₯ 7K Β· ⏱️ 16.06.2023):

     conda install -c conda-forge fitsnap3
    
OpenMM-ML (πŸ₯‰14 Β· ⭐ 76) - High level API for using machine learning models in OpenMM simulations. MIT ML-IAP
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 19 Β· πŸ“‹ 51 - 39% open Β· ⏱️ 06.06.2024):

     git clone https://github.com/openmm/openmm-ml
    
  • Conda (πŸ“₯ 4.1K Β· ⏱️ 07.06.2024):

     conda install -c conda-forge openmm-ml
    
pair_nequip (πŸ₯‰11 Β· ⭐ 37) - LAMMPS pair style for NequIP. MIT ML-IAP rep-learn
  • GitHub (πŸ‘¨β€πŸ’» 3 Β· πŸ”€ 13 Β· πŸ“‹ 28 - 28% open Β· ⏱️ 05.06.2024):

     git clone https://github.com/mir-group/pair_nequip
    
pair_allegro (πŸ₯‰9 Β· ⭐ 33) - LAMMPS pair style for Allegro deep learning interatomic potentials with parallelization support. MIT ML-IAP rep-learn
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 7 Β· πŸ“‹ 26 - 34% open Β· ⏱️ 05.06.2024):

     git clone https://github.com/mir-group/pair_allegro
    
PACE (πŸ₯‰9 Β· ⭐ 21 Β· πŸ’€) - The LAMMPS ML-IAP `pair_style pace`, aka Atomic Cluster Expansion (ACE), aka ML-PACE,.. Custom
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 10 Β· ⏱️ 27.11.2023):

     git clone https://github.com/ICAMS/lammps-user-pace
    
SOMD (πŸ₯‰6 Β· ⭐ 11) - Molecular dynamics package designed for the SIESTA DFT code. AGPL-3.0 ML-IAP active-learning
  • GitHub (πŸ”€ 2 Β· ⏱️ 03.07.2024):

     git clone https://github.com/initqp/somd
    
Show 1 hidden projects...

Reinforcement Learning

Back to top

Projects that focus on reinforcement learning for atomistic ML.

Show 2 hidden projects...
  • ReLeaSE (πŸ₯‡11 Β· ⭐ 340 Β· πŸ’€) - Deep Reinforcement Learning for de-novo Drug Design. MIT drug-discovery
  • CatGym (πŸ₯‰6 Β· ⭐ 11 Β· πŸ’€) - Surface segregation using Deep Reinforcement Learning. GPL

Representation Engineering

Back to top

Projects that offer implementations of representations aka descriptors, fingerprints of atomistic systems, and models built with them, aka feature engineering.

cdk (πŸ₯‡23 Β· ⭐ 480) - The Chemistry Development Kit. LGPL-2.1 cheminformatics Java
  • GitHub (πŸ‘¨β€πŸ’» 160 Β· πŸ”€ 150 Β· πŸ“₯ 21K Β· πŸ“‹ 280 - 10% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/cdk/cdk
    
  • Maven:

     <dependency>
     	<groupId>org.openscience.cdk</groupId>
     	<artifactId>cdk-bundle</artifactId>
     	<version>[VERSION]</version>
     </dependency>
    
DScribe (πŸ₯‡22 Β· ⭐ 380) - DScribe is a python package for creating machine learning descriptors for atomistic systems. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 18 Β· πŸ”€ 85 Β· πŸ“¦ 190 Β· πŸ“‹ 98 - 7% open Β· ⏱️ 28.05.2024):

     git clone https://github.com/SINGROUP/dscribe
    
  • PyPi (πŸ“₯ 300K / month):

     pip install dscribe
    
  • Conda (πŸ“₯ 110K Β· ⏱️ 28.05.2024):

     conda install -c conda-forge dscribe
    
MODNet (πŸ₯‡16 Β· ⭐ 71) - MODNet: a framework for machine learning materials properties. MIT pre-trained small-data transfer-learning
  • GitHub (πŸ‘¨β€πŸ’» 8 Β· πŸ”€ 31 Β· πŸ“¦ 6 Β· πŸ“‹ 42 - 35% open Β· ⏱️ 24.06.2024):

     git clone https://github.com/ppdebreuck/modnet
    
SISSO (πŸ₯ˆ13 Β· ⭐ 220 Β· πŸ’€) - A data-driven method combining symbolic regression and compressed sensing for accurate & interpretable models. Apache-2 Fortran
  • GitHub (πŸ‘¨β€πŸ’» 3 Β· πŸ”€ 74 Β· πŸ“‹ 61 - 6% open Β· ⏱️ 12.09.2023):

     git clone https://github.com/rouyang2017/SISSO
    
Librascal (πŸ₯ˆ13 Β· ⭐ 79 Β· πŸ’€) - A scalable and versatile library to generate representations for atomic-scale learning. LGPL-2.1
  • GitHub (πŸ‘¨β€πŸ’» 30 Β· πŸ”€ 19 Β· πŸ“‹ 230 - 43% open Β· ⏱️ 30.11.2023):

     git clone https://github.com/lab-cosmo/librascal
    
GlassPy (πŸ₯ˆ13 Β· ⭐ 26) - Python module for scientists working with glass materials. GPL-3.0
  • GitHub (πŸ”€ 7 Β· πŸ“¦ 5 Β· πŸ“‹ 6 - 16% open Β· ⏱️ 21.01.2024):

     git clone https://github.com/drcassar/glasspy
    
  • PyPi (πŸ“₯ 150 / month):

     pip install glasspy
    
Rascaline (πŸ₯ˆ12 Β· ⭐ 44) - Computing representations for atomistic machine learning. BSD-3 Rust C++
  • GitHub (πŸ‘¨β€πŸ’» 14 Β· πŸ”€ 13 Β· πŸ“‹ 65 - 47% open Β· ⏱️ 02.07.2024):

     git clone https://github.com/Luthaf/rascaline
    
NICE (πŸ₯‰6 Β· ⭐ 12) - NICE (N-body Iteratively Contracted Equivariants) is a set of tools designed for the calculation of invariant and.. MIT
  • GitHub (πŸ‘¨β€πŸ’» 4 Β· πŸ”€ 3 Β· πŸ“‹ 3 - 66% open Β· ⏱️ 15.04.2024):

     git clone https://github.com/lab-cosmo/nice
    
Show 15 hidden projects...
  • CatLearn (πŸ₯‡16 Β· ⭐ 98 Β· πŸ’€) - GPL-3.0 surface-science
  • cmlkit (πŸ₯ˆ10 Β· ⭐ 34 Β· πŸ’€) - tools for machine learning in condensed matter physics and quantum chemistry. MIT benchmarking
  • BenchML (πŸ₯‰9 Β· ⭐ 15 Β· πŸ’€) - ML benchmarking and pipeling framework. Apache-2 benchmarking
  • CBFV (πŸ₯‰8 Β· ⭐ 21 Β· πŸ’€) - Tool to quickly create a composition-based feature vector. Unlicensed
  • SkipAtom (πŸ₯‰7 Β· ⭐ 23 Β· πŸ’€) - Distributed representations of atoms, inspired by the Skip-gram model. MIT
  • milad (πŸ₯‰6 Β· ⭐ 29 Β· πŸ’€) - Moment Invariants Local Atomic Descriptor. GPL-3.0 generative
  • SA-GPR (πŸ₯‰6 Β· ⭐ 18 Β· πŸ’€) - Public repository for symmetry-adapted Gaussian Process Regression (SA-GPR). LGPL-3.0 C-lang
  • fplib (πŸ₯‰6 Β· ⭐ 7 Β· πŸ’€) - a fingerprint library. MIT C-lang single-paper
  • SOAPxx (πŸ₯‰6 Β· ⭐ 7 Β· πŸ’€) - A SOAP implementation. GPL-2.0 C++
  • soap_turbo (πŸ₯‰6 Β· ⭐ 4 Β· πŸ’€) - soap_turbo comprises a series of libraries to be used in combination with QUIP/GAP and TurboGAP. Custom Fortran
  • pyLODE (πŸ₯‰6 Β· ⭐ 2 Β· πŸ’€) - Pythonic implementation of LOng Distance Equivariants. Apache-2 electrostatics
  • SISSO++ (πŸ₯‰5 Β· ⭐ 3 Β· πŸ’€) - C++ Implementation of SISSO with python bindings. Apache-2 C++
  • magnetism-prediction (πŸ₯‰4 Β· ⭐ 1 Β· πŸ’€) - DFT-aided Machine Learning Search for Magnetism in Fe-based Bimetallic Chalcogenides. Apache-2 magnetism single-paper
  • ML-for-CurieTemp-Predictions (πŸ₯‰3 Β· ⭐ 1 Β· πŸ’€) - Machine Learning Predictions of High-Curie-Temperature Materials. MIT single-paper magnetism
  • AMP (πŸ₯‰2) - Amp is an open-source package designed to easily bring machine-learning to atomistic calculations. Unlicensed

Representation Learning

Back to top

General models that learn a representations aka embeddings of atomistic systems, such as message-passing neural networks (MPNN).

Deep Graph Library (DGL) (πŸ₯‡38 Β· ⭐ 13K) - Python package built to ease deep learning on graph, on top of existing DL frameworks. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 290 Β· πŸ”€ 2.9K Β· πŸ“¦ 270 Β· πŸ“‹ 2.7K - 14% open Β· ⏱️ 11.07.2024):

     git clone https://github.com/dmlc/dgl
    
  • PyPi (πŸ“₯ 130K / month):

     pip install dgl
    
  • Conda (πŸ“₯ 360K Β· ⏱️ 28.06.2024):

     conda install -c dglteam dgl
    
PyG Models (πŸ₯‡35 Β· ⭐ 21K) - Representation learning models implemented in PyTorch Geometric. MIT general-ml
  • GitHub (πŸ‘¨β€πŸ’» 500 Β· πŸ”€ 3.5K Β· πŸ“¦ 6.1K Β· πŸ“‹ 3.5K - 24% open Β· ⏱️ 08.07.2024):

     git clone https://github.com/pyg-team/pytorch_geometric
    
e3nn (πŸ₯‡26 Β· ⭐ 900) - A modular framework for neural networks with Euclidean symmetry. MIT
  • GitHub (πŸ‘¨β€πŸ’» 31 Β· πŸ”€ 130 Β· πŸ“¦ 270 Β· πŸ“‹ 150 - 11% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/e3nn/e3nn
    
  • PyPi (πŸ“₯ 400K / month):

     pip install e3nn
    
  • Conda (πŸ“₯ 18K Β· ⏱️ 18.06.2023):

     conda install -c conda-forge e3nn
    
SchNetPack (πŸ₯‡26 Β· ⭐ 750) - SchNetPack - Deep Neural Networks for Atomistic Systems. MIT
  • GitHub (πŸ‘¨β€πŸ’» 35 Β· πŸ”€ 200 Β· πŸ“¦ 83 Β· πŸ“‹ 240 - 1% open Β· ⏱️ 09.07.2024):

     git clone https://github.com/atomistic-machine-learning/schnetpack
    
  • PyPi (πŸ“₯ 840 / month):

     pip install schnetpack
    
MatGL (Materials Graph Library) (πŸ₯‡22 Β· ⭐ 230) - Graph deep learning library for materials. BSD-3
  • GitHub (πŸ‘¨β€πŸ’» 17 Β· πŸ”€ 57 Β· πŸ“¦ 41 Β· πŸ“‹ 80 - 3% open Β· ⏱️ 08.07.2024):

     git clone https://github.com/materialsvirtuallab/matgl
    
  • PyPi (πŸ“₯ 680 / month):

     pip install m3gnet
    
NVIDIA Deep Learning Examples for Tensor Cores (πŸ₯ˆ21 Β· ⭐ 13K) - State-of-the-Art Deep Learning scripts organized by models - easy to train and deploy with reproducible accuracy and.. Custom educational drug-discovery
  • GitHub (πŸ‘¨β€πŸ’» 120 Β· πŸ”€ 3K Β· πŸ“‹ 820 - 30% open Β· ⏱️ 04.04.2024):

     git clone https://github.com/NVIDIA/DeepLearningExamples
    
DIG: Dive into Graphs (πŸ₯ˆ20 Β· ⭐ 1.8K) - A library for graph deep learning research. GPL-3.0
  • GitHub (πŸ‘¨β€πŸ’» 50 Β· πŸ”€ 280 Β· πŸ“‹ 200 - 14% open Β· ⏱️ 04.02.2024):

     git clone https://github.com/divelab/DIG
    
  • PyPi (πŸ“₯ 560 / month):

     pip install dive-into-graphs
    
ALIGNN (πŸ₯ˆ20 Β· ⭐ 200) - Atomistic Line Graph Neural Network https://scholar.google.com/citations?user=9Q-tNnwAAAAJ&hl=en. Custom
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 77 Β· πŸ“¦ 12 Β· πŸ“‹ 54 - 55% open Β· ⏱️ 28.06.2024):

     git clone https://github.com/usnistgov/alignn
    
  • PyPi (πŸ“₯ 830 / month):

     pip install alignn
    
kgcnn (πŸ₯ˆ20 Β· ⭐ 100) - Graph convolutions in Keras with TensorFlow, PyTorch or Jax. MIT
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 28 Β· πŸ“¦ 18 Β· πŸ“‹ 86 - 13% open Β· ⏱️ 06.05.2024):

     git clone https://github.com/aimat-lab/gcnn_keras
    
  • PyPi (πŸ“₯ 350 / month):

     pip install kgcnn
    
Uni-Mol (πŸ₯ˆ18 Β· ⭐ 610 Β· πŸ“ˆ) - Official Repository for the Uni-Mol Series Methods. MIT pre-trained
  • GitHub (πŸ‘¨β€πŸ’» 16 Β· πŸ”€ 110 Β· πŸ“₯ 14K Β· πŸ“‹ 140 - 39% open Β· ⏱️ 08.07.2024):

     git clone https://github.com/dptech-corp/Uni-Mol
    
e3nn-jax (πŸ₯ˆ17 Β· ⭐ 170) - jax library for E3 Equivariant Neural Networks. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 18 Β· πŸ“¦ 36 Β· ⏱️ 24.06.2024):

     git clone https://github.com/e3nn/e3nn-jax
    
  • PyPi (πŸ“₯ 1.7K / month):

     pip install e3nn-jax
    
matsciml (πŸ₯ˆ17 Β· ⭐ 130) - Open MatSci ML Toolkit is a framework for prototyping and scaling out deep learning models for materials discovery.. MIT workflows benchmarking
  • GitHub (πŸ‘¨β€πŸ’» 11 Β· πŸ”€ 17 Β· πŸ“‹ 49 - 30% open Β· ⏱️ 10.07.2024):

     git clone https://github.com/IntelLabs/matsciml
    
escnn (πŸ₯ˆ14 Β· ⭐ 330 Β· πŸ’€) - Equivariant Steerable CNNs Library for Pytorch https://quva-lab.github.io/escnn/. Custom
  • GitHub (πŸ‘¨β€πŸ’» 10 Β· πŸ”€ 42 Β· πŸ“‹ 62 - 41% open Β· ⏱️ 17.10.2023):

     git clone https://github.com/QUVA-Lab/escnn
    
  • PyPi (πŸ“₯ 250 / month):

     pip install escnn
    
hippynn (πŸ₯ˆ11 Β· ⭐ 60) - python library for atomistic machine learning. Custom workflows
  • GitHub (πŸ‘¨β€πŸ’» 13 Β· πŸ”€ 22 Β· πŸ“¦ 1 Β· πŸ“‹ 14 - 42% open Β· ⏱️ 24.06.2024):

     git clone https://github.com/lanl/hippynn
    
EquiformerV2 (πŸ₯ˆ9 Β· ⭐ 160) - [ICLR24] EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations. MIT
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 23 Β· ⏱️ 08.05.2024):

     git clone https://github.com/atomicarchitects/equiformer_v2
    
graphite (πŸ₯ˆ9 Β· ⭐ 51) - A repository for implementing graph network models based on atomic structures. MIT
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 9 Β· πŸ“¦ 11 Β· πŸ“‹ 3 - 66% open Β· ⏱️ 13.05.2024):

     git clone https://github.com/llnl/graphite
    
ai4material_design (πŸ₯ˆ9 Β· ⭐ 5 Β· πŸ’€) - Code for Kazeev, N., Al-Maeeni, A.R., Romanov, I. et al. Sparse representation for machine learning the properties of.. Apache-2 pre-trained material-defect
  • GitHub (πŸ‘¨β€πŸ’» 11 Β· πŸ”€ 3 Β· ⏱️ 21.11.2023):

     git clone https://github.com/HSE-LAMBDA/ai4material_design
    
DeeperGATGNN (πŸ₯‰8 Β· ⭐ 44) - Scalable graph neural networks for materials property prediction. MIT
  • GitHub (πŸ‘¨β€πŸ’» 3 Β· πŸ”€ 8 Β· ⏱️ 19.01.2024):

     git clone https://github.com/usccolumbia/deeperGATGNN
    
AdsorbML (πŸ₯‰8 Β· ⭐ 33 Β· πŸ’€) - MIT surface-science single-paper
  • GitHub (πŸ‘¨β€πŸ’» 5 Β· πŸ”€ 4 Β· πŸ“‹ 4 - 75% open Β· ⏱️ 31.07.2023):

     git clone https://github.com/Open-Catalyst-Project/AdsorbML
    
ML4pXRDs (πŸ₯‰7 Β· πŸ’€) - Contains code to train neural networks based on simulated powder XRDs from synthetic crystals. MIT XRD single-paper
  • GitHub (πŸ“₯ 2 Β· ⏱️ 14.07.2023):

     git clone https://github.com/aimat-lab/ML4pXRDs
    
Show 34 hidden projects...
  • dgl-lifesci (πŸ₯‡23 Β· ⭐ 700 Β· πŸ’€) - Python package for graph neural networks in chemistry and biology. Apache-2
  • benchmarking-gnns (πŸ₯ˆ14 Β· ⭐ 2.5K Β· πŸ’€) - Repository for benchmarking graph neural networks. MIT single-paper benchmarking
  • Crystal Graph Convolutional Neural Networks (CGCNN) (πŸ₯ˆ12 Β· ⭐ 610 Β· πŸ’€) - Crystal graph convolutional neural networks for predicting material properties. MIT
  • Neural fingerprint (nfp) (πŸ₯ˆ12 Β· ⭐ 57 Β· πŸ’€) - Keras layers for end-to-end learning with rdkit and pymatgen. Custom
  • Compositionally-Restricted Attention-Based Network (CrabNet) (πŸ₯ˆ12 Β· ⭐ 12 Β· πŸ’€) - Predict materials properties using only the composition information!. MIT
  • GDC (πŸ₯ˆ10 Β· ⭐ 260 Β· πŸ’€) - Graph Diffusion Convolution, as proposed in Diffusion Improves Graph Learning (NeurIPS 2019). MIT generative
  • SE(3)-Transformers (πŸ₯ˆ9 Β· ⭐ 480 Β· πŸ’€) - code for the SE3 Transformers paper: https://arxiv.org/abs/2006.10503. MIT single-paper transformer
  • molecularGNN_smiles (πŸ₯ˆ9 Β· ⭐ 280 Β· πŸ’€) - The code of a graph neural network (GNN) for molecules, which is based on learning representations of r-radius.. Apache-2
  • GATGNN: Global Attention Graph Neural Network (πŸ₯ˆ9 Β· ⭐ 66 Β· πŸ’€) - Pytorch Repository for our work: Graph convolutional neural networks with global attention for improved materials.. MIT
  • Equiformer (πŸ₯‰8 Β· ⭐ 180 Β· πŸ’€) - [ICLR23 Spotlight] Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs. MIT transformer
  • CGAT (πŸ₯‰8 Β· ⭐ 25 Β· πŸ’€) - Crystal graph attention neural networks for materials prediction. MIT
  • FAENet (πŸ₯‰8 Β· ⭐ 25 Β· πŸ’€) - MIT
  • UVVisML (πŸ₯‰8 Β· ⭐ 19 Β· πŸ’€) - Predict optical properties of molecules with machine learning. MIT optical-properties single-paper probabilistic
  • T-e3nn (πŸ₯‰8 Β· ⭐ 8 Β· πŸ’€) - Time-reversal Euclidean neural networks based on e3nn. MIT magnetism
  • DTNN (πŸ₯‰7 Β· ⭐ 76 Β· πŸ’€) - Deep Tensor Neural Network. MIT
  • Cormorant (πŸ₯‰7 Β· ⭐ 60 Β· πŸ’€) - Codebase for Cormorant Neural Networks. Custom
  • escnn_jax (πŸ₯‰7 Β· ⭐ 25 Β· πŸ’€) - Equivariant Steerable CNNs Library for Pytorch https://quva-lab.github.io/escnn/. Custom
  • tensorfieldnetworks (πŸ₯‰6 Β· ⭐ 150 Β· πŸ’€) - MIT
  • MACE-Layer (πŸ₯‰6 Β· ⭐ 32 Β· πŸ’€) - Higher order equivariant graph neural networks for 3D point clouds. MIT
  • charge_transfer_nnp (πŸ₯‰6 Β· ⭐ 27 Β· πŸ’€) - Graph neural network potential with charge transfer. MIT electrostatics
  • GLAMOUR (πŸ₯‰6 Β· ⭐ 18 Β· πŸ’€) - Graph Learning over Macromolecule Representations. MIT single-paper
  • Autobahn (πŸ₯‰5 Β· ⭐ 30 Β· πŸ’€) - Repository for Autobahn: Automorphism Based Graph Neural Networks. MIT
  • SCFNN (πŸ₯‰5 Β· ⭐ 15 Β· πŸ’€) - Self-consistent determination of long-range electrostatics in neural network potentials. MIT C++ electrostatics single-paper
  • CraTENet (πŸ₯‰5 Β· ⭐ 12 Β· πŸ’€) - An attention-based deep neural network for thermoelectric transport properties. MIT transport-phenomena
  • Per-Site CGCNN (πŸ₯‰5 Β· ⭐ 1 Β· πŸ’€) - Crystal graph convolutional neural networks for predicting material properties. MIT pre-trained single-paper
  • Per-site PAiNN (πŸ₯‰5 Β· ⭐ 1 Β· πŸ’€) - Fork of PaiNN for PerovskiteOrderingGCNNs. MIT probabilistic pre-trained single-paper
  • Atom2Vec (πŸ₯‰4 Β· ⭐ 31) - Atom2Vec: a simple way to describe atoms for machine learning. Unlicensed
  • FieldSchNet (πŸ₯‰4 Β· ⭐ 16 Β· πŸ’€) - MIT
  • Graph Transport Network (πŸ₯‰4 Β· ⭐ 16 Β· πŸ’€) - Graph transport network (GTN), as proposed in Scalable Optimal Transport in High Dimensions for Graph Distances,.. Custom transport-phenomena
  • gkx: Green-Kubo Method in JAX (πŸ₯‰4 Β· ⭐ 3) - Green-Kubo + JAX + MLPs = Anharmonic Thermal Conductivities Done Fast. MIT transport-phenomena
  • atom_by_atom (πŸ₯‰3 Β· ⭐ 6 Β· πŸ’€) - Atom-by-atom design of metal oxide catalysts for the oxygen evolution reaction with Machine Learning. Unlicensed surface-science single-paper
  • Element encoder (πŸ₯‰3 Β· ⭐ 6 Β· πŸ’€) - Autoencoder neural network to compress properties of atomic species into a vector representation. GPL-3.0 single-paper
  • Point Edge Transformer (πŸ₯‰2) - Smooth, exact rotational symmetrization for deep learning on point clouds. CC-BY-4.0
  • SphericalNet (πŸ₯‰1 Β· ⭐ 3 Β· πŸ’€) - Implementation of Clebsch-Gordan Networks (CGnet: https://arxiv.org/pdf/1806.09231.pdf) by GElib & cnine libraries in.. Unlicensed

Unsupervised Learning

Back to top

Projects that focus on unsupervised learning (USL) for atomistic ML, such as dimensionality reduction, clustering and visualization.

DADApy (πŸ₯‡18 Β· ⭐ 99) - Distance-based Analysis of DAta-manifolds in python. Apache-2
  • GitHub (πŸ‘¨β€πŸ’» 19 Β· πŸ”€ 16 Β· πŸ“¦ 5 Β· πŸ“‹ 31 - 19% open Β· ⏱️ 02.07.2024):

     git clone https://github.com/sissa-data-science/DADApy
    
  • PyPi (πŸ“₯ 240 / month):

     pip install dadapy
    
ASAP (πŸ₯ˆ12 Β· ⭐ 140) - ASAP is a package that can quickly analyze and visualize datasets of crystal or molecular structures. MIT
  • GitHub (πŸ‘¨β€πŸ’» 6 Β· πŸ”€ 28 Β· πŸ“¦ 6 Β· πŸ“‹ 25 - 24% open Β· ⏱️ 27.06.2024):

     git clone https://github.com/BingqingCheng/ASAP
    
Show 5 hidden projects...

Visualization

Back to top

Projects that focus on visualization (viz.) for atomistic ML.

pymatviz (πŸ₯‡20 Β· ⭐ 140 Β· πŸ“ˆ) - A toolkit for visualizations in materials informatics. MIT general-tool probabilistic
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 12 Β· πŸ“¦ 8 Β· πŸ“‹ 37 - 21% open Β· ⏱️ 11.07.2024):

     git clone https://github.com/janosh/pymatviz
    
  • PyPi (πŸ“₯ 1.6K / month):

     pip install pymatviz
    
Chemiscope (πŸ₯‰18 Β· ⭐ 120) - An interactive structure/property explorer for materials and molecules. BSD-3 JavaScript
  • GitHub (πŸ‘¨β€πŸ’» 22 Β· πŸ”€ 29 Β· πŸ“₯ 230 Β· πŸ“¦ 6 Β· πŸ“‹ 120 - 28% open Β· ⏱️ 01.07.2024):

     git clone https://github.com/lab-cosmo/chemiscope
    
  • npm (πŸ“₯ 12 / month):

     npm install chemiscope
    
ZnDraw (πŸ₯‰18 Β· ⭐ 27) - A powerful tool for visualizing, modifying, and analysing atomistic systems. EPL-2.0 MD generative JavaScript
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 1 Β· πŸ“¦ 3 Β· πŸ“‹ 240 - 24% open Β· ⏱️ 27.06.2024):

     git clone https://github.com/zincware/ZnDraw
    
  • PyPi (πŸ“₯ 1.3K / month):

     pip install zndraw
    

Wavefunction methods (ML-WFT)

Back to top

Projects and models that focus on quantities of wavefunction theory methods, such as Monte Carlo techniques like deep learning variational Monte Carlo (DL-VMC), quantum chemistry methods, etc.

DeepQMC (πŸ₯‡17 Β· ⭐ 330) - Deep learning quantum Monte Carlo for electrons in real space. MIT
  • GitHub (πŸ‘¨β€πŸ’» 13 Β· πŸ”€ 60 Β· πŸ“¦ 2 Β· πŸ“‹ 42 - 11% open Β· ⏱️ 23.02.2024):

     git clone https://github.com/deepqmc/deepqmc
    
  • PyPi (πŸ“₯ 78 / month):

     pip install deepqmc
    
FermiNet (πŸ₯ˆ14 Β· ⭐ 650) - An implementation of the Fermionic Neural Network for ab-initio electronic structure calculations. Apache-2 transformer
  • GitHub (πŸ‘¨β€πŸ’» 18 Β· πŸ”€ 110 Β· ⏱️ 04.06.2024):

     git clone https://github.com/deepmind/ferminet
    
DeepErwin (πŸ₯‰11 Β· ⭐ 46) - DeepErwin is a python 3.8+ package that implements and optimizes JAX 2.x wave function models for numerical solutions.. Custom
  • GitHub (πŸ‘¨β€πŸ’» 7 Β· πŸ”€ 6 Β· πŸ“₯ 6 Β· ⏱️ 07.06.2024):

     git clone https://github.com/mdsunivie/deeperwin
    
  • PyPi (πŸ“₯ 30 / month):

     pip install deeperwin
    
Show 1 hidden projects...
  • SchNOrb (πŸ₯‰5 Β· ⭐ 57 Β· πŸ’€) - Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions. MIT

Others

Back to top

pretrained-gnns (πŸ₯‡10 Β· ⭐ 940 Β· πŸ’€) - Strategies for Pre-training Graph Neural Networks. MIT pre-trained
  • GitHub (πŸ‘¨β€πŸ’» 2 Β· πŸ”€ 160 Β· πŸ“‹ 63 - 53% open Β· ⏱️ 29.07.2023):

     git clone https://github.com/snap-stanford/pretrain-gnns
    
Show 1 hidden projects...

Contribution

Contributions are encouraged and always welcome! If you like to add or update projects, choose one of the following ways:

  • Open an issue by selecting one of the provided categories from the issue page and fill in the requested information.
  • Modify the projects.yaml with your additions or changes, and submit a pull request. This can also be done directly via the Github UI.

If you like to contribute to or share suggestions regarding the project metadata collection or markdown generation, please refer to the best-of-generator repository. If you like to create your own best-of list, we recommend to follow this guide.

For more information on how to add or update projects, please read the contribution guidelines. By participating in this project, you agree to abide by its Code of Conduct.

License

CC0

best-of-atomistic-machine-learning's People

Contributors

blaiszik avatar irratzo avatar janosh avatar ml-evs avatar pythonfz avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar

best-of-atomistic-machine-learning's Issues

Improve labels II

Configuration Change:

Moved here from previous issue.

Problem 2). Classification by categories is not enough.

Change from categorical (one-dimensional) to tagging (multi-dimensional; here, tags are called 'labels') system. In this system, every list item has labels, and the category is only one of them, the 'main' or 'emphasized' one. Example from issue #10: Package equisolve would gain the labels math, mlp, general-tools, and the category general-tools is the category = emphasized label.

TODO:

  • Add labels: One for each category (duplicate structure).
  • Issue #10 add labels for equisolve
  • Issue #24 add labels for flare

Add project: ML-for-CurieTemp-Predictions

Project details:

  • Project Name: ML-for-CurieTemp-Predictions
  • Description: Machine Learning Predictions of High-Curie-Temperature Materials
  • Github URL: msg-byu/ML-for-CurieTemp-Predictions
  • Category: rep-eng
  • Labels: rep-eng, single-paper, magnetism
  • License: MIT
  • Package Managers: None

Additional context:

Original publication: http://arxiv.org/abs/2307.06879.

Add project: OpenMM ML projects

Community request, moved here from #10.

Project details:

  • Project Name: OpenMM-ML
  • Github URL: openmm/openmm-ml
  • Category: md
  • Labels: md, mliap
  • License: MIT
  • Package Managers: conda:conda-forge/openmm-ml

  • Project Name: SPICE
  • Github URL: openmm/spice-dataset
  • Category: datasets
  • Labels: datasets, mliap, md
  • License: MIT
  • Package Managers: None

  • Project Name: NNPOps
  • Github URL: openmm/NNPOps
  • Category: mliap
  • Labels: mliap, md, lang-cpp
  • License: MIT
  • Package Managers: conda:conda-forge/nnpops

Add project: MateriApps

Project details:

Additional context:

What’s MateriApps

A Portal Site of Materials Science Simulation for Computational Materials Science Researchers, Theoreticians, Experimentalists, and Computer Scientists

Add project: GT4SD

Project details:

  • Project Name: GT4SD
  • Github URL: GT4SD/gt4sd-core
  • Category: generative
  • Labels: generative, pre-trained, drug-discovery, rep-learn
  • License: MIT
  • Package Managers: pypi:gt4sd

Additional context:

Add project: mlp (Proper orthogonal descriptors, POD)

Project details:

  • Project Name: mlp
  • Description: "Proper orthogonal descriptors for efficient and accurate interatomic potentials. Paper."
  • Github URL: https://github.com/cesmix-mit/mlp
  • Category: mlp
  • Labels: lang-julia
  • License: None
  • Package Managers: None

Additional context:

POD descriptor benchmark, as referenced in the corresponding paper https://doi.org/10.1016/j.jcp.2023.112030.

Original publication not linked in README as of 2023-12-03. Repo description not descriptive. Replace with custom description: Paper title plus link to paper.

Add project: Materials repositories popular in AML research

Project details:

See below.

Additional context:

This is a tricky subject. Materials repositories. Should they be part of this list, or rather be in a separate "materials informatics" list or similar?

Yes: Other "static" datasets like QM9 are also listed here. Why? Because they are part of atomistic ML (AML) publications as much as the models on which they were trained. Same goes for publications that used materials repos like Materials Project, then.

No. This is a bigger undertaking. There are a lot of materials repositories. Which one should be added? Add only the homepagess as resources, or homepages and APIs as separate entries, or homepages and APIs and associated tools as separate entries?

A balance should be struck here. Adding all kinds of materials informatics and computational materials science tools in this AML list would dilute it. For example, pymatgen and spglib should not be listed here, but in such a dedicated list. So, strike a balance: Only add the minimal set of tools to access materials repo data via browser or via API. Two entries at most per repo as guideline.

List of projects to research

Note. Add only those that actually have a record of being used in published AML research.

  • Materials Project
  • AFLOWlib

List of projects to add to the list

Improve labels

Configuration Change:

Right now, labels have some issues.

Problem 1). Labels with emoji name don't show up in Explanations section.

Custom labels without an image but just a name (emoji or string) are not shown in README section Explanations. That makes some emoji labels confusing. For example, ⚑ for electrostatics can only be guessed at without an "Explanation" entry.

Possible solutions.

  • Labels with name emoji: replace with an image emoji (an actual image file to be added to config/images). That makes text less copy & paste-able.
  • Replace all custom labels with name string labels. Then they are all self-explanatory.
  • Read template docs again and see if I have not misunderstood. Otherwise, create a template issue.

Add project: EquiformerV2

Project details:

  • Project Name: EquiformerV2
  • Github URL: atomicarchitects/equiformer_v2
  • Category: rep-learn
  • License: None
  • Package Managers: None

Additional context:

The actual model is implemented in the ocp package, see README.

Add project: Artificial Intelligence for Science (AIRS)

Project details:

  • Project Name: Artificial Intelligence for Science (AIRS)
  • Github URL: https://github.com/divelab/AIRS
  • Category: General Tools or Representation Learning
  • License: GPL-3.0 license
  • Package Managers: None

  • Project Name: AI for Science Map
  • URL: https://www.air4.science/map
  • Category: Comunity resource
  • License: GPL-3.0 license
  • Description: Interactive visual treemap of the AI4Science research field, including atomistic machine learning, as of summer 2023, including papers, packages, learning resources.

TODO: add missing packages, resources from there here as well.


Add project: Citrine Informatics ERD projects

Project details:

  • Project Name: GlassPy
  • Github URL: drcassar/glasspy
  • Category: rep-eng
  • Labels: rep-eng glass, composition, materials
  • License: GPL-3.0
  • Package Managers: pypi:glasspy

Are compositional feature -- based models atomistic models? In this case, I decide for "Yes".


  • Project Name: SciGlass
  • Github URL: drcassar/SciGlass
  • Category: datasets
  • Labels: datasets, glass
  • License: MIT

  • Project Name: sl_discovery
  • Github URL: CitrineInformatics-ERD-public/sl_discovery
  • Category: materials-discovery
  • License: Apache-2.0
  • Package Managers: None
  • Labels: materials-discovery, single-paper

  • Project Name: Linear vs blackbox
  • Github URL: CitrineInformatics-ERD-public/linear-vs-blackbox
  • Category: xai
  • Labels: xai, single-paper, rep-eng
  • License: None

  • Project Name: closed-loop-acceleration-benchmarks
  • Github URL: aced-differentiate/closed-loop-acceleration-benchmarks
  • Category: materials-discovery
  • Labels: materials-discovery, active-learning, single-paper
  • License: MIT
  • Package Managers:

  • Project Name: EquivariantOperators.jl
  • Github URL: aced-differentiate/EquivariantOperators.jl
  • Category: math
  • Labels: math, symmetrized, lang-julia
  • License: MIT

  • Project Name: Closed-loop acceleration benchmarks
  • Github URL: aced-differentiate/closed-loop-acceleration-benchmarks
  • Category: materials-discovery
  • Labels: materials-discovery, active-learning, single-paper
  • License: MIT
  • Package Managers:


Add project: Add all LAMMPS ML potentials

Community request, moved here from #10.

MD category: Add LAMMPS & all the lammps-xxx repo integrating a given potential with lammps; as well as OpenMM and TorchMD, both usable with ML potentials.

Subtasks OpenMM, TorchMD moved to issues #77, #78.

Add project: MEGAN

Project details:

  • Project Name: MEGAN: Multi Explanation Graph Attention Student
  • Github URL: aimat-lab/graph_attention_student
  • Category: xai
  • Labels: xai, rep-learn
  • License: MIT
  • Package Managers: None

Additional context:

Original publication: https://arxiv.org/abs/2305.15961v1

(technically, the original publication is http://arxiv.org/abs/2211.13236, but the two papers link to repos that do not exist anymore or have been updated since. therefore, assume the newer publication is the closest one to current state of code.)

Add project: Materials Project Charge Densities & mp-pyrho

Project details:

Additional context:

URL for the MP CD datasets project entry:

https://next-gen.materialsproject.org/ml/charge_densities

The CDs are really part of MP API now. But still worthwhile to emphasize with a separate entry that they exist in the first place.

Update project: Flare

Update details:

Additional context:

The project is listed under General Tools. I think a better fit would be the Active Learning section since a core functionality is building force fields using AL.

Add project: paperswithcode

Project details:

This suggestion is not a single project, but several datasets and benchmark models aggregated via paperswithcode.

  • For some of the already existing list entries (e.g. QM9, OC20 dataset), maybe replace website with paperswithcode entry.
  • The paperswithcode tags could be entered as resource in respective category, if any. Collect here first, then assign.
  • Also add the linked models / packages.

Initial datasets finds.

Initial tags finds.

Add project: matsci.org

Project details:

  • Project Name: matsci.org
  • URL: https://matsci.org/
  • Description: "Computational materials science community, online forum."
  • Category: community-resource

Additional context:

Arguable whether this community resource should be put here, as it is more about general atomistic modeling and simulation, rather than atomistic machine learning. But on the other hand, there is no such dedicated AML community forum, so this is an okay place to put AML question. And AML is a sub- or site field of atomistic modeling will just merge into the general field as time moves on, so why even make a distinction.

Add project: matbench, MPContribs

Project details:

  • Project Name: MatBench
  • Github URL: materialsproject/matbench
  • Category: community
  • Labels: community, datasets, benchmarking
  • License: MIT
  • Package Managers: pypi:matbench

  • Project Name: MPContribs
  • Github URL: materialsproject/MPContribs
  • Category: datasets
  • Labels: datasets
  • License: MIT
  • Package Managers: pypi:mpcontribs-client

Add project: EDM

Project details:

  • Project Name: EDM
  • Description: E(3) Equivariant Diffusion Model for Molecule Generation in 3D.
  • Github URL: ehoogeboom/e3_diffusion_for_molecules
  • Category: generative
  • Labels: generative
  • License: MIT
  • Package Managers: None

Additional context:

Original publication: https://arxiv.org/abs/2203.17003v2

Anonymous community feedback

Configuration Change:

Anonymous community feedback 2023-06-13.

  • Disable license warnings. Which licenses are risky is a matter of opinion.
  • Add an explanation of the "combined project-quality score".
  • MD category: Add LAMMPS & all the lammps-xxx repo integrating a given potential with lammps; as well as OpenMM and TorchMD, both usable with ML potentials.

Improvement suggestions to the best-of-generator template.

  • Allow projects in multiple categories.
    • For now, the template only allows one category per project. However, one can implement this also by duplicating a project entry for different categories. That should not require a change to the template. The question is, then, where to draw the line of duplication, in order to stay fair to every package. Ultimately, this shows the deficit of the category (tree-based) vs. label (tag-based) classification schemes. Simplest would be to just do this case-by-case, whenever there is a package developer / community request for it. In that case:
    • Put equisolve in multiple categories: math, mlp, general-tools. Developer request.
  • If the way the project score is calculated raises concern, create an issue with improvement suggestions for .

Add project: wfl

Project details:

  • Project Name: wfl
  • Description: Workflow is a Python toolkit for building interatomic potential creation and atomistic simulation workflows.
  • Github URL: libAtoms/workflow
  • Category: mliap
  • Labels: workflows, htc
  • License: None
  • Package Managers: None

Additional context:

Original publication: https://arxiv.org/abs/2306.11421v1

Use manual description copied from README.md rather than automatic one, that one is too short.

Add project: TorchMD-NET

Community request, moved here from #10.

Project details:

  • Project Name: TorchMD-NET
  • Github URL: torchmd/torchmd-net
  • Category: md
  • Labels: md, mliap, rep-learn
  • License: MIT
  • Package Managers: None

Add project: Visual Graph Datasets

Project details:

  • Project Name: Visual Graph Datasets
  • Github URL: aimat-lab/visual_graph_datasets
  • Category: datasets
  • Tags: datasets, xai, rep-learn
  • License: MIT
  • Package Managers: None

Additional context:

Dataset used in original publication of #69 .

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    πŸ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. πŸ“ŠπŸ“ˆπŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❀️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.