#5454th of 105 in ML frameworks and librariesFor data and documents, data analysis
darts
Python library for time series forecasting and anomaly detection
- Stars
- 9.6k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
9,243 on 1 Apr 20269,559 on 21 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Darts is a Python library that gives forecasting models, from ARIMA and exponential smoothing to deep networks such as N-BEATS, a common scikit-learn-style fit() and predict() interface. It also covers backtesting, probabilistic and conformal forecasts, covariates, and a darts.ad module of anomaly scorers and detectors, including PyOD wrappers. Deep learning models are built on PyTorch Lightning.
Who it is for: Engineers building time series forecasting or anomaly detection in Python
Strengths
- One fit()/predict() interface across statistical, ML and deep learning forecasting models
- Built-in backtesting, probabilistic and conformal forecasts, and past, future and static covariates
- Anomaly detection module with scorers, detectors, PyOD and forecast-model wrappers
- Global models train on multiple series; TimeSeries converts to pandas, polars, numpy and others
Weaknesses
- Requires Python 3.11 or newer
- Deep learning models depend on PyTorch Lightning, which adds a heavy install
- Local statistical models such as ARIMA are fitted on one series at a time
- Not LLM-related; fits this list only as a general ML library
What it needs
- GPU optional
- Docker
- Needs PyTorch Lightning, scikit-learn, pandas
- Models: ARIMA, VARIMA, ExponentialSmoothing, N-BEATS, TCN
Also in ML frameworks and libraries
See all 105| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 92 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |