tum-pbs / PhiML
Intuitive scientific computing with dimension types for Jax, PyTorch, TensorFlow & NumPy
☆82Updated this week
Alternatives and similar repositories for PhiML:
Users that are interested in PhiML are comparing it to the libraries listed below
- [Neurips 2024] A benchmark suite for autoregressive neural emulation of PDEs. (≥46 PDEs in 1D, 2D, 3D; Differentiable Physics; Unrolled T…☆56Updated 2 months ago
- Code for the paper "Poseidon: Efficient Foundation Models for PDEs"☆121Updated last month
- Spectral Neural Operator☆75Updated last year
- Repo to the paper "Lie Point Symmetry Data Augmentation for Neural PDE Solvers"☆49Updated last year
- A JAX-based research framework for writing differentiable numerical simulators with arbitrary discretizations☆124Updated 4 months ago
- Code for the paper Universal Physics Transformers☆88Updated 3 months ago
- Convolutional Differential Operators for Physics-based Deep Learning Study☆24Updated 6 months ago