NVIDIA / cuEquivariance
cuEquivariance is a math library that is a collective of low-level primitives and tensor ops to accelerate widely-used models, like DiffDock, MACE, Allegro and NEQUIP, based on equivariant neural networks.
☆199Updated last week
Alternatives and similar repositories for cuEquivariance:
Users that are interested in cuEquivariance are comparing it to the libraries listed below
- jax library for E3 Equivariant Neural Networks☆202Updated 2 months ago
- Build neural networks for machine learning force fields with JAX☆111Updated last month
- E3x is a JAX library for constructing efficient E(3)-equivariant deep learning architectures built on top of Flax.☆103Updated 4 months ago
- Official implementation of All Atom Diffusion Transformers☆158Updated last week
- [ICLR 2024] EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations☆255Updated last month
- Boltzmann Generators and Normalizing Flows in PyTorch☆160Updated last year
- Neural Network Force Field based on PyTorch☆267Updated last month
- [ICLR 2024 Spotlight] Official Implementation of "Enabling Efficient Equivariant Operations in the Fourier Basis via Gaunt Tensor Product…☆58Updated 5 months ago
- repository and website for tutorials on 3d Euclidean equivariant neural networks☆71Updated 4 years ago
- A collection of QM data for training potential functions☆166Updated last month
- Pytorch differentiable molecular dynamics☆175Updated 2 years ago
- [TMLR 2023] Training and simulating MD with ML force fields☆112Updated 5 months ago
- Equivariant machine learning interatomic potentials in JAX.☆72Updated last year
- scalable molecular simulation☆130Updated 2 weeks ago
- Training neural network potentials☆390Updated this week
- Higher order equivariant graph neural networks for 3D point clouds☆36Updated last year
- train and use graph-based ML models of potential energy surfaces☆79Updated this week
- GemNet model in PyTorch, as proposed in "GemNet: Universal Directional Graph Neural Networks for Molecules" (NeurIPS 2021)☆194Updated last year
- sGDML - Reference implementation of the Symmetric Gradient Domain Machine Learning model☆147Updated last year
- OpenMM plugin to define forces with neural networks☆193Updated last month
- A Library for Gaussian Processes in Chemistry☆223Updated 5 months ago
- An evaluation framework for machine learning models simulating high-throughput materials discovery.☆138Updated this week
- OpenEquivariance: a fast, open-source GPU JIT kernel generator for the Clebsch-Gordon Tensor Product.☆46Updated this week
- A plugin to use Nvidia GPU in PySCF package☆186Updated this week
- [ICLR 2023 Spotlight] Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs☆233Updated last month
- Higher-order equivariant neural networks for charge density prediction in materials☆53Updated last month
- LAMMPS pair style for Allegro deep learning interatomic potentials with parallelization support☆41Updated 5 months ago
- [NeurIPS 2024] Official implementation of the Efficiently Scaled Attention Interatomic Potential☆45Updated 3 weeks ago
- Workflow for creating and analyzing the Open Catalyst Dataset☆106Updated last month
- Reference implementation of "SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects"☆71Updated 2 years ago