Numerical Methods Python Packages

mpmath

Python library for arbitrary-precision floating-point arithmetic

kanjize

Kanjize(カンジャイズ): Easy converter between Kanji-Number and Integer

pyroots

Pure python solvers for single-variable equations.

findiff

Python package for numerical derivatives and partial differential equations in any number of dimensions.

py-bobyqa

Python-based Derivative-Free Optimization with Bound Constraints

scinumtools3

Define, validate, and report scientific parameters with units and provenance across C++, Python, CLI, CMake, and REST.

convinterp

High-order convolution interpolation, derivatives and integrals on uniform grids in any dimension, for Python with a Rust core. Port of ConvolutionInterpolations.jl.

metal-linalg

QR, symmetric eigendecomposition and SVD for batches of matrices on Apple Silicon GPUs, for MLX and PyTorch. 2x to 10x gain in performance. Metal kernels per regime, routed to the GPU or the CPU by policies measured per Mac. C++, Python (pip: metal-linalg, metal-linalg-torch), Swift and C.

dfo-ls

Python-based Derivative-Free Optimizer for Least-Squares

scientific-computing-system-2-0

Scientific computing platform on NumPy/SciPy/pandas/matplotlib: linear algebra, stats, optimization, signals, ML, information theory, chaos, Bayes, metaheuristics, geometry and RL.

desolver

A Python library for solving Initial Value Problems using various numerical integration methods.

tff-nightly

High-performance TensorFlow library for quantitative finance.

ngstrefftz

Add-On to NGSolve for Trefftz methods

pathsim

A Python native dynamical system simulation framework in the block diagram paradigm.

acoustotreams

A Python package for acoustic-wave scattering based on the T-matrix method

qnm

Python package for computing Kerr quasinormal mode frequencies, separation constants, and spherical-spheroidal mixing coefficients

matplobblib

The library provides a set of tools and functions for various subject areas, including Analysis and Data Structures (AISD), Probability Theory and Mathematical Statistics (TViMS), Machine Learning (ML), and Numerical Methods (NM).

qsvt-pennylane

Python toolkit for Quantum Singular Value Transformation (QSVT), including polynomial constructions, matrix function workflows, and reproducible tools for research in quantum algorithms and numerical linear algebra.

num-dual

Generalized (hyper-) dual numbers in rust

openballistics

High-performance, lightweight, header-only C++ library with native Python bindings for projectile trajectory prediction and firing-solution solving

scientific-computing-system

Pure-Python computational science with zero runtime dependencies: linear algebra, ODE/PDE solvers, quadrature, Monte Carlo, statistics, signal processing, quantum simulation, uncertainty quantification, sensitivity analysis, dimensional analysis, and reproducible workflows.

mathematicskit

See how the algorithms of computational mathematics work: a Python toolkit for learning and teaching numerical methods, dynamical systems, and discrete math across 18 domains, built on numpy/scipy

pyroot

Python library for root-finding in one dimension

metal-linalg-torch

QR, symmetric eigendecomposition and SVD for batches of matrices on Apple Silicon GPUs, for MLX and PyTorch. 2x to 10x gain in performance. Metal kernels per regime, routed to the GPU or the CPU by policies measured per Mac. C++, Python (pip: metal-linalg, metal-linalg-torch), Swift and C.