# skforecast/skforecast

Python library for time series forecasting using scikit-learn compatible models, statistical methods, and foundation models

Repository: https://github.com/skforecast/skforecast
Canonical: https://ross.abutalabs.com/products/skforecast
Homepage: https://skforecast.org
Language: Python
License: BSD-3-Clause
License Family: permissive
Topics: forecasting, scikit-learn, machine-learning, python, data-science, exogenous-predictors, time-series, multi-series-forecasting, multi-step-forecasting, backtesting-forecasters, arima, sarimax, lightgbm, probabilistic-forecasting, xgboost, catboost, deep-learning, foundation-models, prediction-intervals, time-series-forecasting
Last push: 2026-08-26T12:09:52+00:00

## Health v2 (maintenance only)
Score: 95/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 87, longevity 100
- inputs: {"age_days": 2030, "days_push": 7, "days_rel": 9, "gap_med": 44, "n_releases_24m": 14}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1527, forks 198 (observed 2026-08-28T04:04:58.482730+00:00)

## What it is
Skforecast is a Python library for time series forecasting that turns any scikit-learn compatible estimator (LightGBM, XGBoost, CatBoost, Keras, etc.) into a forecaster, alongside statistical methods like ARIMA/SARIMAX and foundation models. It provides multi-series and multi-step forecasting, exogenous variable support, backtesting, and probabilistic forecasting with prediction intervals.

## Use cases
- forecast future values of a time series with machine learning models
- predict sales or demand across multiple related time series
- backtest forecasting models with walk-forward validation
- estimate prediction intervals for probabilistic forecasts
- include exogenous variables like weather or holidays in forecasts
- forecast multiple steps ahead with recursive or direct strategies
- compare statistical models like SARIMAX against gradient boosting forecasters

## When to choose
- you want scikit-learn compatible regressors for time series forecasting
- you need backtesting, hyperparameter tuning, and prediction intervals in one library
- you forecast many related series with exogenous predictors
- you want a well-documented, actively maintained Python forecasting library

## When to avoid
- you need deep learning sequence models like N-BEATS or TFT as first-class citizens
- you want automated forecasting with zero configuration like Prophet or Nixtla AutoML
- your project is not in Python

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, data-science, benchmarking
- domain: machine-learning, data-science, time-series
- platform: python
- tags: forecasting, time-series-forecasting, scikit-learn, backtesting, probabilistic-forecasting, exogenous-variables, multi-step-forecasting, arima, sarimax, lightgbm, xgboost, catboost, prediction-intervals

## Member repositories
- skforecast/skforecast (main) score 95

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:58.482730+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:31:34.631087+00:00, confidence not recorded.
  - readme: https://github.com/skforecast/skforecast (fetched 2026-08-28T04:04:58.482730+00:00, sha 905ce7f89257)
  - homepage: https://skforecast.org (fetched 2026-08-29T11:34:07.921518+00:00, sha 36c6c3c2e4f9)
  - registry_pypi: https://pypi.org/pypi/skforecast/json (fetched 2026-08-29T11:34:07.930556+00:00, sha 03f754410cac)
- Data as of 2026-08-30T08:39:29.467469+00:00.