BreezeML

Quiet Mode

Machine learning, without the maze.

BreezeML is a production-aware workflow layer for students, analysts, and AI agents. Train, compare, explain, export, and deploy scikit-learn models without the boilerplate.

pip install breezemlGitHubJourney Mode
v1.9.0
latest PyPI release
55
commits in the main repo
2
GitHub contributors
3
forks
10
open issues
27
test files plus CI on Python 3.9-3.13

Four vows, carved in stone

Vow 1

Four dependencies. Always.

The core installs with only scikit-learn, pandas, numpy, and joblib. A CI test fails the build if anyone adds a fifth. No dependency hell, ever.

Vow 2

Zero lock-in.

export() hands you a standalone scikit-learn script that reproduces your exact pipeline, with no BreezeML import. Graduate the moment you outgrow it.

Vow 3

It teaches you.

explain_decisions=True narrates every pipeline choice in plain English, and card() writes an honest model card with auto-detected caveats.

Vow 4

AI agents can use it.

breezeml-mcp is a built-in Model Context Protocol server, so Claude and other agents train, compare, explain, and deploy models with sound statistical defaults.

The Four Breaths

A layered path. Walk only as far as you need; every layer is optional.

Breath 1 First model

Dawn field

A complete model in three lines.

fitpredictautofrom_csv
from breezeml import datasets, fit, predict

model = fit(datasets.iris(), "species")
predict(model, new_df)

Breath 2 Understand and choose

Lantern court

Twenty-two models compete, and every choice stays visible.

comparequick_tuneexplain_decisionscard
model, report = breezeml.auto(df, "target", explain_decisions=True)
breezeml.classifiers.compare(df, "target")   # leaderboard of 22 models
breezeml.card(model, "MODEL_CARD.md")

Breath 3 Automate and ship

Bridge to production

Search, export, deploy, and watch for drift.

automlexportdeploydrifttimeseries
breezeml.automl(df, "target", time_budget=60)
breezeml.export(model, "train.py")   # pure sklearn, zero lock-in
breezeml.deploy(model, "api/")       # FastAPI + Docker
breezeml.drift.check(model, new_df)

Breath 4 Beyond

Mountain temple

Advanced tools, an agent guide, and a quiet garden.

conformalcausalfairnessauditmcpzen
breezeml.conformal.conformal_regressor(model, calib, "y")
breezeml.causal.estimate_ate(df, "treat", "outcome")
breezeml.fairness.report(model, df, sensitive="group")
breezeml.zen()

Where to go next

ExamplesBenchmarksDocsReleases