PaddleScience is SDK and library for developing AI-driven scientific computing applications based on PaddlePaddle.

Source code of 'Deep transfer operator learning for partial differential equations under conditional shift'.

Code for training and inferring acoustic wave propagation in 3D

Official repo for separable operator networks -- extreme-scale operator learning for parametric PDEs.

PyTorch implemention of the Position-induced Transformer for operator learning in partial differential equations

Folax (Finite Operator Learning with JAX) is a framework for solving and optimizing PDEs by integrating machine learning with numerical methods in computational mechanics.

An extension of Fourier Neural Operator to finite-dimensional input and/or output spaces.

Code for the paper "The Random Feature Model for Input-Output Maps between Banach Spaces" (SIREV SIGEST 2024, SISC 2021)

TANTE: Time-Adaptive Operator Learning via Neural Taylor Expansion

Official implementation of the paper "Neural Hamilton: Can A.I. Understand Hamiltonian Mechanics?"

Graph Feedforward Network (GFN) - a novel neural network layer for resolution-invariant machine learning

Nonlinear model reduction for operator learning

Benchmarking Surrogates for coupled ODE systems.

Code for the paper ``Error Bounds for Learning with Vector-Valued Random Features'' (NeurIPS 2023, Spotlight)

Hyperbolic Learning Rate Scheduler

A comparative analysis of DeepONet and FNO architectures, benchmarking their performance on Function-to-Function (Heat Equation) vs. Parameter-to-Function (Elastic Bar) PDE problems to motivate hybrid operator designs.

Open benchmark of FNO, conditional-diffusion (U-Net & DiT), and ensemble-UQ surrogates for two-phase porous-media flow (CO2 sequestration).

Learning the Electrical Impedance Tomography Inverse Map

Official code for "Spectral Convolutional Conditional Neural Processes" (NeurIPS 2025)

Implementation of several neural network (and neural operator) architectures and numerical methods for solving kinetic equations (Boltzmann, Fokker-Planck-Landau, etc.)