scikit-optimize/scikit-optimize

Health v2 · maintenance only

Flags: archived

Adoption not part of the score

2828 stars · 559 forks observed · 2026-08-28

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

Scikit-Optimize (skopt) is a Python library for sequential model-based optimization of expensive and noisy black-box functions, built on NumPy, SciPy, and scikit-learn. It offers a scipy.optimize-like interface with methods such as gp_minimize and an ask/tell Optimizer class, plus plotting tools for visualizing objectives.

Use cases

• tune hyperparameters of a machine learning model
• minimize an expensive noisy black-box function
• run bayesian optimization with gaussian processes
• find optimal parameters for a simulation
• optimize a function where gradients are unavailable
• run an ask/tell optimization loop across processes

When to choose

• you need gradient-free optimization of costly or noisy objective functions
• you want a scikit-learn/scipy-friendly API for hyperparameter search
• you want built-in visualization of the optimized objective

When to avoid

• you need gradient-based optimization (use scipy.optimize instead)
• you need large-scale or highly parallel hyperparameter search at scale
• you need a project under heavy active development

Facets

library · maturity maintenance

machine-learning math data-science machine-learning data-science python cross-platform bayesian-optimization hyperparameter-tuning black-box-optimization scikit-learn scipy sequential-model-based-optimization algorithms

Member repositories

For agents

markdown · JSON · MCP: product_card(name="scikit-optimize/scikit-optimize")