Welcome to skforecast

About The Project¶

Skforecast is a Python library for time series forecasting using machine learning models. It works with any regressor compatible with the scikit-learn API, including popular options like LightGBM, XGBoost, CatBoost, Keras, and many others.

Why use skforecast?

The fields of statistics and machine learning have developed many excellent regression algorithms that can be useful for forecasting, but applying them effectively to time series analysis can still be a challenge. To address this issue, the skforecast library provides a comprehensive set of tools for training, validation and prediction in a variety of scenarios commonly encountered when working with time series. The library is built using the widely used scikit-learn API, making it easy to integrate into existing workflows. With skforecast, users have access to a wide range of functionalities such as feature engineering, model selection, hyperparameter tuning and many others. This allows users to focus on the essential aspects of their projects and leave the intricacies of time series analysis to skforecast. In addition, skforecast is developed according to the following priorities:

Fast and robust prototyping.
Validation and backtesting methods to have a realistic assessment of model performance.
Models must be deployed in production.
Models must be interpretable.

Share Your Thoughts with Us

Thank you for choosing skforecast! We value your suggestions, bug reports and recommendations as they help us identify areas for improvement and ensure that skforecast meets the needs of the community. Please consider sharing your experiences, reporting bugs, making suggestions or even contributing to the codebase on GitHub. Together, let's make time series forecasting more accessible and accurate for everyone.

Installation & Dependencies¶

To install the basic version of skforecast with its core dependencies, run:

1	`pip install skforecast`

If you want to learn more about the installation process, dependencies and optional features, please refer to the Installation Guide.

Forecasters¶

A Forecaster object in the skforecast library is a comprehensive container that provides essential functionality and methods for training a forecasting model and generating predictions for future points in time.

The skforecast library offers a variety of forecaster types, each tailored to specific requirements such as single or multiple time series, direct or recursive strategies, or custom predictors. Regardless of the specific forecaster type, all instances share the same API.

Forecaster	Single series	Multiple series	Recursive strategy	Direct strategy	Probabilistic prediction	Time series differentiation	Exogenous features	Custom features
ForecasterAutoreg	✔️		✔️		✔️	✔️	✔️
ForecasterAutoregCustom	✔️		✔️		✔️	✔️	✔️	✔️
ForecasterAutoregDirect	✔️			✔️	✔️		✔️
ForecasterMultiSeries		✔️	✔️		✔️	✔️	✔️
ForecasterMultiSeriesCustom		✔️	✔️		✔️	✔️	✔️	✔️
ForecasterMultiVariate		✔️		✔️	✔️		✔️
ForecasterRNN		✔️		✔️
ForecasterSarimax	✔️		✔️		✔️	✔️	✔️

Features¶

Create Forecasters from any regressor that follows the scikit-learn API
Include exogenous variables as predictors
Include custom predictors (rolling mean, rolling variance ...)
Multiple backtesting methods for model validation
Grid search, random search and Bayesian search to find optimal lags (predictors) and best hyperparameters
Prediction interval estimated by bootstrapping and quantile regression
Include custom metrics for model validation and grid search
Get predictor importance
Forecaster in production

Examples and tutorials¶

English¶

Skforecast: time series forecasting with Machine Learning

ARIMA and SARIMAX models

Forecasting with gradient boosting: XGBoost, LightGBM and CatBoost

Forecasting with XGBoost

Global Forecasting Models: Multi-series forecasting

Probabilistic forecasting models

Forecasting with Deep Learning

Modelling time series trend with tree based models

Interpretable forecasting models

Intermittent demand forecasting

Forecasting time series with missing values

Stacking ensemble of machine learning models to improve forecasting

Forecasting energy demand with machine learning

Global Forecasting Models: Comparative Analysis of Single and Multi-Series Forecasting Modeling

Mitigating the impact of covid on forecasting models

Forecasting web traffic with machine learning and Python

Bitcoin price prediction with Python

Español¶

Skforecast: forecasting series temporales con Machine Learning

Modelos ARIMA y SARIMAX

Forecasting con gradient boosting: XGBoost, LightGBM y CatBoost

Forecasting con XGBoost

Global Forecasting Models: Multi-series forecasting

Modelos de forecasting globales: Análisis comparativo de modelos de una y múltiples series

Forecasting probabilístico

Forecasting con Deep Learning

Forecasting de series incompletas con valores faltantes

Predicción demanda intermitente

Modelar series temporales con tendencia utilizando modelos de árboles

Interpretabilidad en modelos de forecasting

Forecasting de la demanda eléctrica

Forecasting de las visitas a una página web

Reducir el impacto del Covid en modelos de forecasting

Predicción del precio de Bitcoin con Python

Workshop predicción de series temporales con machine learning Universidad de Deusto / Deustuko Unibertsitatea

How to contribute¶

Primarily, skforecast development consists of adding and creating new Forecasters, new validation strategies, or improving the performance of the current code. However, there are many other ways to contribute:

Submit a bug report or feature request on GitHub Issues.
Contribute a Jupyter notebook to our examples.
Write unit or integration tests for our project.
Answer questions on our issues, Stack Overflow, and elsewhere.
Translate our documentation into another language.
Write a blog post, tweet, or share our project with others.

For more information on how to contribute to skforecast, see our Contribution Guide.

Visit our authors section to meet all the contributors to skforecast.

Citation¶

If you use skforecast for a scientific publication, we would appreciate citations to the published software.

Zenodo

Amat Rodrigo, Joaquin, & Escobar Ortiz, Javier. (2024). skforecast (v0.13.0). Zenodo. https://doi.org/10.5281/zenodo.8382788

APA:

Amat Rodrigo, J., & Escobar Ortiz, J. (2024). skforecast (Version 0.13.0) [Computer software]. https://doi.org/10.5281/zenodo.8382788

BibTeX:

@software{skforecast,
author = {Amat Rodrigo, Joaquin and Escobar Ortiz, Javier},
title = {skforecast},
version = {0.13.0},
month = {8},
year = {2024},
license = {BSD-3-Clause},
url = {https://skforecast.org/},
doi = {10.5281/zenodo.8382788}
}

Publications citing skforecast¶

Sanan, O., Sperling, J., Greene, D., & Greer, R. (2024, April). Forecasting Weather and Energy Demand for Optimization of Renewable Energy and Energy Storage Systems for Water Desalination. In 2024 IEEE Conference on Technologies for Sustainability (SusTech) (pp. 175-182). IEEE. https://doi.org/10.1109/SusTech60925.2024.10553570
Bojer, A. K., Biru, B. H., Al-Quraishi, A. M. F., Debelee, T. G., Negera, W. G., Woldesillasie, F. F., & Esubalew, S. Z. (2024). Machine learning and remote sensing based time series analysis for drought risk prediction in Borena Zone, Southwest Ethiopia. Journal of Arid Environments, 222, 105160. https://doi.org/10.1016/j.jaridenv.2024.105160
V. Negri, A. Mingotti, R. Tinarelli and L. Peretto, "Comparison Between the Machine Learning and the Statistical Approach to the Forecasting of Voltage, Current, and Frequency," 2023 IEEE 13th International Workshop on Applied Measurements for Power Systems (AMPS), Bern, Switzerland, 2023, pp. 01-06, doi: 10.1109/AMPS59207.2023.10297192. https://doi.org/10.1109/AMPS59207.2023.10297192
Marcillo Vera, F., Rosado, R., Zambrano, P., Velastegui, J., Morales, G., Lagla, L., & Herrera, A. (2024). Forecasting con Python, caso de estudio: visitas a las redes sociales en Ecuador con machine learning. CONECTIVIDAD, 5(2), 15-29.

OUKHOUYA, H., KADIRI, H., EL HIMDI, K., & GUERBAZ, R. (2023). Forecasting International Stock Market Trends: XGBoost, LSTM, LSTM-XGBoost, and Backtesting XGBoost Models. Statistics, Optimization & Information Computing, 12(1), 200-209. https://doi.org/10.19139/soic-2310-5070-1822
DUDZIK, S., & Kowalczyk, B. (2023). Prognozowanie produkcji energii fotowoltaicznej z wykorzystaniem platformy NEXO i VRM Portal. Przeglad Elektrotechniczny, 2023(11). doi:10.15199/48.2023.11.41
Polo J, Martín-Chivelet N, Alonso-Abella M, Sanz-Saiz C, Cuenca J, de la Cruz M. Exploring the PV Power Forecasting at Building Façades Using Gradient Boosting Methods. Energies. 2023; 16(3):1495. https://doi.org/10.3390/en16031495
Popławski T, Dudzik S, Szeląg P. Forecasting of Energy Balance in Prosumer Micro-Installations Using Machine Learning Models. Energies. 2023; 16(18):6726. https://doi.org/10.3390/en16186726
Harrou F, Sun Y, Taghezouit B, Dairi A. Artificial Intelligence Techniques for Solar Irradiance and PV Modeling and Forecasting. Energies. 2023; 16(18):6731. https://doi.org/10.3390/en16186731
Amara-Ouali, Y., Goude, Y., Doumèche, N., Veyret, P., Thomas, A., Hebenstreit, D., ... & Phe-Neau, T. (2023). Forecasting Electric Vehicle Charging Station Occupancy: Smarter Mobility Data Challenge. arXiv preprint arXiv:2306.06142.
Emami, P., Sahu, A., & Graf, P. (2023). BuildingsBench: A Large-Scale Dataset of 900K Buildings and Benchmark for Short-Term Load Forecasting. arXiv preprint arXiv:2307.00142.
Dang, HA., Dao, VD. (2023). Building Power Demand Forecasting Using Machine Learning: Application for an Office Building in Danang. In: Nguyen, D.C., Vu, N.P., Long, B.T., Puta, H., Sattler, KU. (eds) Advances in Engineering Research and Application. ICERA 2022. Lecture Notes in Networks and Systems, vol 602. Springer, Cham. https://doi.org/10.1007/978-3-031-22200-9_32
Morate del Moral, Iván (2023). Predición de llamadas realizadas a un Call Center. Proyecto Fin de Carrera / Trabajo Fin de Grado, E.T.S.I. de Sistemas Informáticos (UPM), Madrid.
Lopez Vega, A., & Villanueva Vargas, R. A. (2022). Sistema para la automatización de procesos hospitalarios de control para pacientes para COVID-19 usando machine learning para el Centro de Salud San Fernando.
García Álvarez, J. D. (2022). Modelo predictivo de rentabilidad de criptomonedas para un futuro cercano.
Chilet Vera, Á. (2023). Elaboración de un algoritmo predictivo para la reposición de hipoclorito en los depósitos mediante técnicas de Machine Learning (Doctoral dissertation, Universitat Politècnica de València).
Bustinza Barrial, A. A., Bautista Abanto, A. M., Alva Alfaro, D. A., Villena Sotomayor, G. M., & Trujillo Sabrera, J. M. (2022). Predicción de los valores de la demanda máxima de energía eléctrica empleando técnicas de machine learning para la empresa Nexa Resources–Cajamarquilla.
Morgado, K. Desarrollo de una técnica de gestión de activos para transformadores de distribución basada en sistema de monitoreo (Doctoral dissertation, Universidad Nacional de Colombia).
Zafeiriou A., Chantzis G., Jonkaitis T., Fokaides P., Papadopoulos A., 2023, Smart Energy Strategy - A Comparative Study of Energy Consumption Forecasting Machine Learning Models, Chemical Engineering Transactions, 103, 691-696.

Donating¶

If you found skforecast useful, you can support us with a donation. Your contribution will help to continue developing and improving this project. Many thanks!

License¶

BSD-3-Clause License