unit8co/darts

A python library for user-friendly forecasting and anomaly detection on time series.

74
/ 100
Verified

This tool helps forecasters and analysts predict future trends and spot unusual patterns in data that changes over time, such as sales figures, stock prices, or sensor readings. You provide historical time-series data, and it outputs predictions for future values or highlights data points that deviate significantly from expected behavior. It's designed for data scientists, business analysts, and operations managers who need reliable forecasts and anomaly detection without deep machine learning expertise.

9,248 stars. Used by 2 other packages. Actively maintained with 16 commits in the last 30 days. Available on PyPI.

Use this if you need to forecast future values or identify anomalies in time-series data using a wide range of models, from traditional statistics to advanced deep learning, with a consistent and user-friendly approach.

Not ideal if your data is not sequential or time-dependent, or if you only need very basic, non-statistical summaries of your data.

time-series-forecasting anomaly-detection predictive-analytics operations-monitoring business-intelligence
Maintenance 17 / 25
Adoption 12 / 25
Maturity 25 / 25
Community 20 / 25

How are scores calculated?

Stars

9,248

Forks

989

Language

Python

License

Apache-2.0

Last pushed

Mar 07, 2026

Commits (30d)

16

Dependencies

16

Reverse dependents

2

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/unit8co/darts"

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