# lululxvi/deepxde

A library for scientific machine learning and physics-informed learning

Repository: https://github.com/lululxvi/deepxde
Canonical: https://ross.abutalabs.com/products/deepxde
Homepage: https://deepxde.readthedocs.io
Language: Python
License: LGPL-2.1
License Family: copyleft
Topics: neural-network, deep-learning, scientific-machine-learning, pinn, multi-fidelity-data, operator, pytorch, physics-informed-learning, jax, deeponet, paddle, pde, tensorflow
Last push: 2026-08-18T01:17:55+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 48, longevity 100
- inputs: {"age_days": 2764, "days_push": 16, "days_rel": 271, "gap_med": 64, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4386, forks 988 (observed 2026-08-28T04:08:47.132552+00:00)

## What it is
DeepXDE is a Python library for scientific machine learning and physics-informed learning, built on top of PyTorch, TensorFlow, JAX, and PaddlePaddle. It implements algorithms such as physics-informed neural networks (PINNs) and DeepONet for solving forward and inverse ODEs, PDEs, and learning operators.

## Use cases
- solve forward and inverse PDEs with neural networks
- train physics-informed neural networks (PINNs)
- learn operators with DeepONet
- solve fractional and stochastic differential equations
- fit multi-fidelity physics data
- perform inverse design and topology optimization with hard constraints

## When to choose
- you need a mature, well-documented PINN library with multiple backend support
- you want to solve ODEs/PDEs or learn operators without writing training loops from scratch
- you need advanced PINN variants like gPINN, fPINN, or adaptive sampling

## When to avoid
- you need general-purpose deep learning outside scientific computing
- you require a permissively licensed library (it is LGPL-2.1)
- you need high-performance traditional numerical PDE solvers rather than neural approaches

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, simulation, math
- domain: machine-learning, deep-learning, simulation, mathematics
- platform: python, cross-platform
- tags: pinn, deeponet, physics-informed-neural-networks, pde-solver, operator-learning, scientific-computing, pytorch, tensorflow, jax, paddle, scientific-machine-learning

## Member repositories
- lululxvi/deepxde (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:47.132552+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:21:14.775133+00:00, confidence not recorded.
  - readme: https://github.com/lululxvi/deepxde (fetched 2026-08-28T04:08:47.132552+00:00, sha 26317d4c2fa8)
  - registry_pypi: https://pypi.org/pypi/deepxde/json (fetched 2026-08-29T09:09:54.876585+00:00, sha 8fcfba6138d6)
- Data as of 2026-08-30T08:39:29.467469+00:00.