pmdarima
alkaline-mlPython equivalent of R's auto.arima, which searches ARIMA and seasonal ARIMA orders for a time series automatically. Around that search it adds stationarity and seasonality tests, differencing, Box-Cox and Fourier transformers, pipelines and time-series cross-validation, with a scikit-learn-style interface over statsmodels.
Openness
5 high confidence- license
- MIT(OSI)
- source
- public(github.com/alkaline-ml/pmdarima)
- core features withheld
- no — volunteer project
MIT-licensed and built from the public repository, maintained by volunteers with nothing for sale. Releases have slowed; the last, 2.1.1, shipped in November 2025.
- https://pypi.org/pypi/pmdarima/json recorded 2026-09-27
PyPI record for pmdarima 2.1.1, uploaded 2025-11-17, license_expression MIT, Repository URL github.com/alkaline-ml/pmdarima.
- https://raw.githubusercontent.com/alkaline-ml/pmdarima/master/LICENSE recorded 2026-09-27
LICENSE is the "MIT License", "Copyright (c) 2017 Taylor G Smith".
- https://raw.githubusercontent.com/alkaline-ml/pmdarima/master/README.md recorded 2026-09-27
README describes the library and its install and usage; no paid tier, hosted service or commercial license mentioned.
- https://ungh.cc/repos/alkaline-ml/pmdarima recorded 2026-09-27
Repository record for alkaline-ml/pmdarima, default branch master, last pushed 2025-11-17.
- https://ungh.cc/repos/alkaline-ml/pmdarima/files/master recorded 2026-09-27
Repository tree of 240 paths with no ee, enterprise, commercial or pro directory.
Adoption
4 high confidenceMeasured on monthly PyPI downloads of the pmdarima package.
- https://pypistats.org/api/packages/pmdarima/recent recorded 2026-09-27
last_month 1,622,889 downloads of pmdarima
Capability
2 high confidencepmdarima fits one kind of forecasting model and automates choosing its orders, as R's auto.arima does. That search stays inside the ARIMA family, so it sits with single-family forecasters like Prophet rather than with toolkits that compare many kinds of model.
- https://alkaline-ml.com/pmdarima/modules/classes.html recorded 2026-09-27
API reference: pmdarima.arima (ARIMA, AutoARIMA, ADFTest, KPSSTest, PPTest, CHTest, OCSBTest, auto_arima), preprocessing transformers, pipeline and model_selection modules.
- https://raw.githubusercontent.com/alkaline-ml/pmdarima/master/README.md recorded 2026-09-27
README: "The equivalent of R's auto.arima functionality"; "A collection of statistical tests of stationarity and seasonality"; example "model = pm.auto_arima(train, seasonal=True, m=12)".
Verified 2026-09-27