Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

py-pde

py-pde.readthedocs.io

  • Free plan
  • Windows
  • macOS
  • Linux
  • Self-hosted

py-pde

Free

Open source

What the maker publishes

Audience
The package is aimed at researchers and students investigating PDE behavior and boundary conditions. source
Backends
Core computations can use numba, jax, or torch backends, with optional dependencies needed for some backends. source
Backends
Core computations can use numba, jax, or torch backends, with compiled backends potentially requiring startup compilation. source
Boundary conditions
Differential operators support flexible boundary conditions, including Dirichlet, Neumann, Robin, periodic, and anti-periodic conditions. source
Features
It supports nonlinear PDEs, complex boundary conditions, and collections of scalar, vector, or tensor fields. source
Fields and equations
It supports scalar, vector, and rank-2 tensor fields, and PDEs can be expressed with the PDE class or implemented by subclassing PDEBase. source
Geometries
Supported grids include Cartesian, polar symmetric, spherical symmetric, and cylindrical symmetric grids. source
Installation
The package can be installed with pip or conda-forge, and the project also lists an Arch Linux package. source
Integrations
Optional packages include h5py for hierarchical data storage, pandas for tabular data, napari for interactive image display, and mpi4py or numba-mpi for parallel processing. source
Integrations
Documented optional packages include JAX, PyTorch, MPI, napari, pandas, h5py, ipywidgets, and py-modelrunner. source
Intended users
The project says it is aimed at researchers and students exploring PDE behavior and boundary conditions. source
License
The repository identifies the project as MIT licensed. source
Limitations
The package is intended for simple geometries with fixed discretization, and time-independent and integro-differential systems are not fully supported. source
Limits
The package is suited to simple geometries with fixed discretization and does not fully support time-independent systems or integro-differential equations. source
Numerical method
It uses the method of lines with finite-difference approximations on fixed grids. source
Parallel processing
The package supports MPI-based parallel simulations using its ExplicitMPISolver and an installed MPI implementation. source
Platforms
The package is developed for Python 3.12, tested through Python 3.14, and documented for Linux, Windows, and macOS. source
Platform support
The documentation says py-pde has been tested on Linux, Windows, and macOS with Python versions up to 3.14. source
Publication
The documentation asks users to cite David Zwicker’s 2020 Journal of Open Source Software paper about py-pde. source
Purpose
py-pde is a Python package for simulating partial differential equations using finite differences. source
Purpose
py-pde provides methods and classes for simulating partial differential equations and stochastic differential equations. source
Results and visualization
Trackers can report progress, plot simulation states, detect steady states, and store intermediate states in memory or files. source
Security
The documentation warns that interpreting arbitrary expressions uses exec() and should not be used where malicious input may occur. source
Security note
The documentation warns that arbitrary expressions are interpreted using exec() and advises against using them where malicious input could occur. source

Checked against the maker's pages on 12 Oct 2026.