mdsunivie / deeperwin
DeepErwin is a python 3.8+ package that implements and optimizes JAX 2.x wave function models for numerical solutions to the multi-electron Schrödinger equation. DeepErwin supports weight-sharing when optimizing wave functions for multiple nuclear geometries and the usage of pre-trained neural network weights to accelerate optimization.
☆51Updated 5 months ago
Related projects ⓘ
Alternatives and complementary repositories for deeperwin
- PySCF with auto-differentiation☆66Updated last week
- Supporting code for "Autoregressive neural-network wavefunctions for ab initio quantum chemistry".☆36Updated 2 years ago
- GradDFT is a JAX-based library enabling the differentiable design and experimentation of exchange-correlation functionals using machine l…☆76Updated 8 months ago
- Reference implementation of "Ab-Initio Potential Energy Surfaces by Pairing GNNs with Neural Wave Functions" (ICLR, 2022) and "Sampling-f…☆28Updated 4 months ago
- Equivariant machine learning interatomic potentials in JAX.☆61Updated last year
- Differentiable Quantum Chemistry (only Differentiable Density Functional Theory and Hartree Fock at the moment)☆106Updated 2 years ago
- Learning neural network potentials from experimental data via Differentiable Trajectory Reweighting☆30Updated 8 months ago
- A JAX library for Density Functional Theory.☆41Updated 6 months ago
- Arbitrary-order derivatives of popular electronic structure methods, such as Hartree-Fock and coupled cluster theory.☆62Updated 2 months ago
- CUDA implementations of MACE models☆10Updated 3 months ago
- ESI-DCAFM-TACO-VDSP Summer School on "Machine Learning for Materials Hard and Soft"☆34Updated 2 years ago
- ☆18Updated last year
- Space Group Informed Transformer for Crystalline Materials Generation☆54Updated 3 months ago
- Build neural networks for machine learning force fields with JAX☆91Updated last month
- Exchange correlation functionals translated from libxc to jax☆43Updated 10 months ago
- Official implementation of DeepDFT model☆60Updated last year
- A library combining solid quantum Monte Carlo and neural network.☆33Updated 5 months ago
- MESS: Modern Electronic Structure Simulations☆17Updated last month
- C++/CUDA library for SO(3) equivariant operations☆19Updated last month
- Reference implementation of "Ewald-based Long-Range Message Passing for Molecular Graphs" (ICML 2023)☆42Updated last year
- A fully autodifferentiable and variational HF☆41Updated 4 years ago
- jax library for E3 Equivariant Neural Networks☆183Updated last month
- MESS: Modern Electronic Structure Simulations☆18Updated last month
- Higher order equivariant graph neural networks for 3D point clouds☆33Updated last year
- Code repository for "Finding symmetry breaking order parameters with Euclidean Neural Networks"☆12Updated 3 years ago
- E3x is a JAX library for constructing efficient E(3)-equivariant deep learning architectures built on top of Flax.☆89Updated 3 weeks ago
- Reference implementation of "SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects"☆66Updated 2 years ago
- ☆11Updated 5 months ago
- Boltzmann Generators and Normalizing Flows in PyTorch☆150Updated 9 months ago
- This repo contains the datasets used for the paper The Design Space of E(3)-Equivariant Atom-Centered Interatomic Potentials.☆17Updated 2 years ago