python-adaptive/adaptive

Health v2 · maintenance only

Adoption not part of the score

1227 stars · 63 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Adaptive is an open-source Python library for parallel active learning of mathematical functions. It intelligently selects the most informative points in a parameter space instead of evaluating a dense grid, with support for cluster execution and live plotting.

Use cases

• adaptively sample expensive mathematical functions in parallel
• plot functions with fewer evaluation points than a dense grid
• run adaptive function evaluations on a computing cluster
• visualize live plots of function convergence in Jupyter
• reduce simulation cost by sampling only interesting regions of parameter space

When to choose

• each function evaluation takes roughly 50ms or more
• you need efficient sampling of 1D-ND functions without dense grids
• you want parallel evaluation across workers or a cluster with live plots

When to avoid

• function evaluations are very fast (well under ~50ms) where sampling overhead dominates
• you need general supervised machine learning on datasets rather than function sampling

Facets

library · maturity active

machine-learning data-science math benchmarking machine-learning mathematics data-science python cross-platform active-learning adaptive-sampling parallel-computing live-plots function-evaluation algorithms

Member repositories

For agents

markdown · JSON · MCP: product_card(name="python-adaptive/adaptive")