/pyforesight

Enhanced Automatic time series forecasting using ARIMA family models

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

pyforesight

Automatic time series forecasting using ARIMA family of models. Implements the basic Auto-ARIMA algorithm in python to handle any type of series for forecasting. Automatically chooses the best parameters for model based on the model score.

Usage:

forecast <data_file_path_on_disk> <date_column_name> <time_series_column_name> <forecast_steps> ["FREQUENCY=<date_frequency>,SEASONAL_PERIOD=<seasonal_period>"]

FREQUENCY and SEASONAL_PERIOD are optional parameters.

Output after model training on the Air Passengers dataset Airlines Dataset example