# CMA-ES/pycma

Python implementation of CMA-ES

Repository: https://github.com/CMA-ES/pycma
Canonical: https://ross.abutalabs.com/products/pycma
Language: Jupyter Notebook
License: NOASSERTION
License Family: other
Last push: 2026-08-05T11:06:26+00:00

## Health v2 (maintenance only)
Score: 84/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 96, release rhythm 60, longevity 100
- inputs: {"age_days": 3632, "days_push": 28, "days_rel": 189, "gap_med": 56.5, "n_releases_24m": 7}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1350, forks 201 (observed 2026-08-28T04:04:27.952191+00:00)

## What it is
pycma is a Python implementation of the CMA-ES (Covariance Matrix Adaptation Evolution Strategy) derivative-free optimization algorithm, along with related numerical optimization tools. It handles difficult non-convex, ill-conditioned, multi-modal, noisy problems in continuous and mixed-integer search spaces, including bound, linear, and nonlinear constraints.

## Use cases
- optimize a noisy black-box objective function without gradients
- solve non-convex multi-modal continuous optimization problems
- handle bound and nonlinear constraints in evolutionary optimization
- optimize functions with mixed-integer variables
- tune hyperparameters of a simulation or model
- use an ask-and-tell interface for interactive optimization loops

## When to choose
- your objective function is non-convex, rugged, noisy, or ill-conditioned
- gradients are unavailable or expensive to compute
- you need constraint handling or mixed-integer support in continuous optimization
- you want a well-cited, reference implementation of CMA-ES in Python

## When to avoid
- your problem is smooth and convex where gradient-based solvers are faster
- you need discrete/combinatorial optimization rather than continuous search spaces
- you need massively parallel large-scale optimization where CMA-ES covariance updates are too costly

## Facets
- artifact type: library
- maturity: stable
- function: math, simulation
- domain: machine-learning, mathematics
- platform: python, cross-platform
- tags: cma-es, evolution-strategy, derivative-free-optimization, black-box-optimization, numerical-optimization, global-optimization, algorithms

## Member repositories
- CMA-ES/pycma (main) score 84

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:27.952191+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 2026-08-30T04:42:14.939497+00:00, confidence not recorded.
  - readme: https://github.com/CMA-ES/pycma (fetched 2026-08-28T04:04:27.952191+00:00, sha c6c5609c9295)
- Data as of 2026-08-30T08:39:29.467469+00:00.