GroupbyCategoryForecaster

class GroupbyCategoryForecaster(forecasters, transformer=None, fallback_forecaster=None)[source]

Choosing a global forecaster based on category or cluster of time series.

Programmatic generalization of “cluster then apply forecaster” approach, or the Syntetos/Boylan heuristic to apply forecaster by categories smooth, erratic, intermittent, lumpy.

Applies a series-to-primitives transformer on a given time series. Series are grouped by the generated value from the transformer, and the corresponding forecaster is used to predict the time series.

Different from TransformSelectForecaster, this compositor passes all timeseries of a given category to the forecaster, instead of passing only one at a time.

Parameters:

forecastersdict[sktime forecasters]

dict of forecasters with the key corresponding to categories generated by the given transformer and the value corresponding to a sktime forecaster.

transformersktime transformer or clusterer, default = ADICVTransformer()

A series-to-primitives sktime transformer that generates a value which can be used to quantify a choice of forecaster for the time series.

If a clusterer is used, it must support cluster assignment, i.e, have the capability:predict tag.

Note: To ensure correct functionality, the transformer must store the generated category in the first column of the returned values when the transform() or fit_transform() functions are called.

fallback_forecastersktime forecaster | None, Optional

A fallback forecaster that will be used if the category generated by the transformer does not match any of the given forecasters.

Attributes:

cutoff: Cut-off = “present time” state of forecaster.
fh: Forecasting horizon that was passed.
is_fitted: Whether fit has been called.
state: State of the estimator.
steps_

Raises:

AssertionError: If a valid transformer (an instance of BaseTransformer)
is not passed or if valid forecasters (instances of BaseForecaster) are not given.

Examples

This example showcases how the GroupbyCategoryForecaster can be utilized to select appropriate forecasters on the basis of the time series category determined by the ADICVTransformer:

>>> from sktime.forecasting.compose import GroupbyCategoryForecaster
>>> from sktime.forecasting.croston import Croston
>>> from sktime.forecasting.trend import PolynomialTrendForecaster
>>> from sktime.forecasting.naive import NaiveForecaster
>>> from sktime.transformations.adi_cv import ADICVTransformer

Importing the methods which can generate data of specific categories depending on their variance and average demand intervals.

>>> from sktime.transformations.tests.test_adi_cv import (
...     _generate_erratic_series)

The forecaster is defined which accepts a dictionary of forecasters, a transformer and optionally a fallback_forecaster

>>> group_forecaster = GroupbyCategoryForecaster(
...     forecasters =
...         {"smooth": NaiveForecaster(),
...         "erratic": PolynomialTrendForecaster(),
...         "intermittent": Croston()},
...     transformer=ADICVTransformer(features=["class"]))

>>> generated_data = _generate_erratic_series()

The fit function firstly passes the data through the given transformer to generate a given category. This category can be seen by the variable self.category_.

>>> group_forecaster = group_forecaster.fit(generated_data, fh=50)
>>> #print(f"The chosen category is: {group_forecaster.category}")

>>> # Print out the predicted value over the given forecasting horizon!
>>> # print(group_forecaster.predict(fh=50, X=None))

Methods

`check_is_fitted`([method_name])	Check if the estimator has been fitted.
`clone`()	Obtain a clone of the object with same hyper-parameters and config.
`clone_tags`(estimator[, tag_names])	Clone tags from another object as dynamic override.
`create_test_instance`([parameter_set])	Construct an instance of the class, using first test parameter set.
`create_test_instances_and_names`([parameter_set])	Create list of all test instances and a list of names for them.
`fit`(y[, X, fh])	Fit forecaster to training data.
`fit_predict`(y[, X, fh, X_pred])	Fit and forecast time series at future horizon.
`get_class_tag`(tag_name[, tag_value_default])	Get class tag value from class, with tag level inheritance from parents.
`get_class_tags`()	Get class tags from class, with tag level inheritance from parent classes.
`get_config`()	Get config flags for self.
`get_fitted_params`([deep])	Get fitted parameters.
`get_param_defaults`()	Get object's parameter defaults.
`get_param_names`([sort])	Get object's parameter names.
`get_params`([deep])	Get a dict of parameters values for this object.
`get_pretrained_params`([deep])	Get pretrained parameters of this estimator.
`get_tag`(tag_name[, tag_value_default, ...])	Get tag value from instance, with tag level inheritance and overrides.
`get_tags`()	Get tags from instance, with tag level inheritance and overrides.
`get_test_params`([parameter_set])	Return testing parameter settings for the estimator.
`is_composite`()	Check if the object is composed of other BaseObjects.
`load_from_path`(serial)	Load object from file location.
`load_from_serial`(serial)	Load object from serialized memory container.
`predict`([fh, X])	Forecast time series at future horizon.
`predict_interval`([fh, X, coverage])	Compute/return prediction interval forecasts.
`predict_proba`([fh, X, marginal])	Compute/return fully probabilistic forecasts.
`predict_quantiles`([fh, X, alpha])	Compute/return quantile forecasts.
`predict_residuals`([y, X])	Return residuals of time series forecasts.
`predict_var`([fh, X, cov])	Compute/return variance forecasts.
`pretrain`(y[, X, fh])	Pre-train forecaster on panel (global) data.
`reset`()	Reset the object to a clean post-init state.
`save`([path, serialization_format])	Save serialized self to bytes-like object or to (.zip) file.
`score`(y[, X, fh])	Scores forecast against ground truth, using MAPE (non-symmetric).
`set_config`(**config_dict)	Set config flags to given values.
`set_params`(**params)	Set the parameters of this object.
`set_random_state`([random_state, deep, ...])	Set random_state pseudo-random seed parameters for self.
`set_tags`(**tag_dict)	Set instance level tag overrides to given values.
`update`(y[, X, update_params])	Update cutoff value and, optionally, fitted parameters.
`update_predict`(y[, cv, X, update_params, ...])	Make predictions and update model iteratively over the test set.
`update_predict_single`([y, fh, X, update_params])	Update model with new data and make forecasts.