latest-changes.md

📈 Trending Up

Projects that have a higher project-quality score compared to the last update. There might be a variety of reasons, such as increased downloads or code activity.

• DeepChem (🥇34 · ⭐ 6.7K · 📈) - Democratizing Deep-Learning for Drug Discovery, Quantum Chemistry, Materials Science and Biology. MIT
• MatGL (Materials Graph Library) (🥇28 · ⭐ 540 · 📈) - Graph deep learning library for materials. BSD-3 ML-IAP pretrained multifidelity
• DScribe (🥇25 · ⭐ 460 · 💤) - DScribe is a python package for creating machine learning descriptors for atomistic systems. Apache-2
• MatterSim (🥈24 · ⭐ 530 · 📈) - MatterSim: A deep learning atomistic model across elements, temperatures and pressures. MIT ML-IAP active-learning multimodal phase-transition pretrained
• MUSE (🥉7 · ⭐ 7 · 📈) - A python package for fast building amorphous solids and liquid mixtures from @materialsproject computed structures and.. MIT ML-IAP Defects & Disorder

📉 Trending Down

Projects that have a lower project-quality score compared to the last update. There might be a variety of reasons such as decreased downloads or code activity.

• FAIR Chemistry datasets (🥇31 · ⭐ 2.1K · 📉) - Datasets OC20, OC22, etc. Formerly known as Open Catalyst Project. MIT catalysis
• fairchem (🥇31 · ⭐ 2.1K · 📉) - FAIR Chemistrys library of machine learning methods for chemistry. Formerly known as Open Catalyst Project. MIT pretrained UIP rep-learn catalysis
• FAIRChem EquiformerV2 models (🥇31 · ⭐ 2.1K · 📉) - FAIRChem implementation of Equiformer V2 (eqV2) models. MIT pretrained UIP rep-learn catalysis
• FAIRChem eSEN models (🥇31 · ⭐ 2.1K · 📉) - FAIRChem implementation of Smooth Energy Network (eSEN) models arXiv:2502.12147. MIT pretrained UIP rep-learn catalysis
• Meta Open Materials 2024 (OMat24) Dataset (🥇30 · ⭐ 2.1K · 📉) - Contains over 100 million Density Functional Theory calculations focused on structural and compositional diversity. CC-BY-4.0