GreykiteForecaster

Adapter for using Greykite forecasting models within sktime.

Quickstart

from sktime.forecasting.greykite import GreykiteForecaster

estimator = GreykiteForecaster(forecast_config: GreykiteForecaster.ForecastConfig | None=None, date_format: str | None=None, model_template: str='SILVERKITE', coverage: float=0.95)

Parameters(4)

forecast_configForecastConfig, optional

Configuration object for Greykite’s forecasting pipeline. If None, a default configuration is created.

date_formatstr, optional

Format of the timestamp in the data. If None, it is inferred.

model_templatestr, optional

Name of the model template to use (default: “SILVERKITE”).

coveragefloat, optional

Intended coverage of the prediction bands (0.0 to 1.0).

Examples

>>> from sktime.datasets import load_airline
>>> from sktime.forecasting.greykite import GreykiteForecaster
>>> from sktime.forecasting.base import ForecastingHorizon
>>> y = load_airline ()
>>> fh = ForecastingHorizon ([1, 2, 3 ])
>>> forecaster = GreykiteForecaster ()
>>> forecaster. fit (y = y, fh = fh)
>>> y_pred = forecaster. predict (fh = fh)

References

1. [1 ] https://linkedin.github.io/greykite/docs/1.0.0/html/pages/stepbystep/0400_configuration.html