Open-source forecasting tool from Facebook for time series with multiple seasonality, trend changes and holiday effects, available for Python and R.
Prophet is an automatic forecasting procedure for time series data. It fits an additive model in which non-linear trends are combined with yearly, weekly and daily seasonality, plus the effects of holidays you specify. It is designed for series with strong seasonal patterns and several seasons of history.
The model is robust to missing values and to shifts in the trend, and it typically handles outliers well. Users can add custom seasonalities, extra regressors, holiday calendars by country, and saturating growth limits. Built-in tools cover cross-validation and forecast performance metrics.
Prophet is available on PyPI as `prophet` and on CRAN for R. The project is in maintenance mode as of v1.4.0, so only bug fixes and dependency updates are being accepted.
Install the Python package from PyPI
python -m pip install prophetOr install via conda-forge
conda install -c conda-forge prophetSummary generated by AI from the project's README. GitHub stats refresh daily.
A personal portfolio of Python and JavaScript packages for online time series prediction, covariance estimation, portfolio construction and black-box optimization, plus research notes.
AutoML library that trains and ensembles ML models for tabular, time series, and multimodal data with a few lines of Python.
Pretrained Transformer-based models from Amazon for zero-shot time series forecasting, including the Chronos-2 family.
Python SDK client for Nixtla's TimeGPT, a hosted pre-trained foundation model for time series forecasting and anomaly detection.
An R package that automates time series forecasting for financial data, with an AI agent, 25+ models and Azure-based parallel runs.
SciPy is an open-source Python library for mathematics, science, and engineering, built on NumPy arrays.