pycaret/pycaret

Health v2 · maintenance only

Flags: prerelease_only no_license

Adoption not part of the score

9834 stars · 1845 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

PyCaret is an open-source, low-code AutoML library for Python that wraps scikit-learn to automate training, tuning, and comparison of models for classification, regression, clustering, and anomaly detection. Version 4.0 expands it into a self-hosted ML platform with a FastAPI control plane and React UI for experiments, model registry, deployments, and monitoring.

Use cases

• train and compare machine learning models with a few lines of code
• automate model selection and hyperparameter tuning
• run classification, regression, clustering, and anomaly detection experiments
• deploy and monitor ML models from a self-hosted web UI
• track experiments and manage a model registry
• detect data drift in production ML models

When to choose

• you want low-code AutoML without writing boilerplate scikit-learn code
• you need a self-hosted alternative to managed ML platforms that runs locally via docker compose
• you want experiment tracking, model registry, and deployment tooling in one package
• you are a data scientist prototyping models quickly in Python notebooks

When to avoid

• you need fine-grained manual control over every training step - the abstraction may hide details
• you require a battle-tested production platform - the 4.0 control plane is explicitly work in progress
• you need distributed training on large-scale data beyond scikit-learn's single-node limits
• you depend on the stable 3.x API - that line is frozen with no further commits

Facets

library · maturity active

machine-learning data-science workflow-automation monitoring api-framework web-framework machine-learning data-science artificial-intelligence developer-tools self-hosted python self-hosted cross-platform automl low-code mlops scikit-learn model-training model-deployment fastapi react-ui model-registry drift-monitoring docker web-server

Member repositories

For agents

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