online-ml/river

🌊 Online machine learning in Python

82
/ 100
Verified

When your data is constantly arriving in real-time, such as sensor readings, financial trades, or website clicks, River helps you build machine learning models that learn and adapt continuously. You feed in individual data points as they appear, and the system provides predictions or insights immediately, updating its understanding without needing to reprocess all past information. This is ideal for data scientists or machine learning engineers who need to deploy models that can react instantly to evolving data streams.

5,746 stars. Used by 5 other packages. Actively maintained with 61 commits in the last 30 days. Available on PyPI.

Use this if you need a machine learning model that can learn incrementally from a continuous stream of new data and adapt to changes over time without requiring full retraining.

Not ideal if your data arrives in large batches and remains static, as traditional batch machine learning methods are likely simpler and more efficient for those scenarios.

streaming-analytics real-time-predictions concept-drift live-data-processing adaptive-models
Maintenance 22 / 25
Adoption 15 / 25
Maturity 25 / 25
Community 20 / 25

How are scores calculated?

Stars

5,746

Forks

609

Language

Python

License

BSD-3-Clause

Last pushed

Mar 09, 2026

Commits (30d)

61

Dependencies

3

Reverse dependents

5

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/online-ml/river"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.