# sumpy **Repository Path**: mirrors_inducer/sumpy ## Basic Information - **Project Name**: sumpy - **Description**: Symbolic code generators for multipole and local expansions and translations - **Primary Language**: Unknown - **License**: Not specified - **Default Branch**: main - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 0 - **Forks**: 0 - **Created**: 2020-09-25 - **Last Updated**: 2026-11-09 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README sumpy: n-body kernels and translation operators =============================================== .. image:: https://github.com/inducer/sumpy/actions/workflows/ci.yml/badge.svg :alt: Github Build Status :target: https://github.com/inducer/sumpy/actions/workflows/ci.yml .. image:: https://badge.fury.io/py/sumpy.svg :alt: Python Package Index Release Page :target: https://pypi.org/project/sumpy/ .. image:: https://zenodo.org/badge/1856097.svg :alt: Zenodo DOI for latest release :target: https://zenodo.org/badge/latestdoi/1856097 sumpy is mainly a 'scaffolding' package for Fast Multipole and quadrature methods. If you're building one of those and need code generation for the required multipole and local expansions, come right on in. Together with ``boxtree``, there is a full, symbolically kernel-independent FMM implementation here. It relies on * `boxtree `__ for FMM tree building * `loopy `__ for fast array operations * `pytest `__ for automated testing and, indirectly, * `PyOpenCL `__ as computational infrastructure Resources: * `documentation `__ * `source code via git `__ * `benchmarks `__