dfm/emcee

Health v2 · maintenance only

Adoption not part of the score

1597 stars · 440 forks observed · 2026-08-28

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

emcee is a stable, well-tested Python implementation of the affine-invariant ensemble sampler for Markov chain Monte Carlo (MCMC) proposed by Goodman & Weare (2010). It is widely used for probabilistic data analysis, particularly in astrophysics.

Use cases

• sample posterior distributions with MCMC in Python
• fit Bayesian models to data
• run ensemble sampling for high-dimensional parameter spaces
• do probabilistic data analysis for astronomy research
• estimate uncertainty on model parameters via MCMC
• replace hand-rolled Metropolis-Hastings samplers

When to choose

• you need a mature, well-tested MCMC sampler with minimal setup
• your likelihood is expensive and parallelizable across walkers
• you want the affine-invariant ensemble sampler without tuning covariance matrices
• you're doing Bayesian inference in astronomy or the sciences

When to avoid

• you need gradient-based samplers like Hamiltonian Monte Carlo or NUTS
• you need a full probabilistic programming framework with model syntax
• your posterior is extremely high-dimensional or multimodal where ensemble sampling struggles
• you need GPU-accelerated sampling

Facets

library · maturity stable

machine-learning data-science math data-science astronomy python cross-platform mcmc ensemble-sampler bayesian-inference probabilistic-programming monte-carlo algorithms

Member repositories

For agents

markdown · JSON · MCP: product_card(name="dfm/emcee")