FhPlexForecaster

class FhPlexForecaster(forecaster, fh_params=None, fh_lookup='relative', fh_contiguous=False)[source]

Uses different parameters by forecasting horizon element.

When provided with forecasting horizon [f1, f2, …, fn], will fit forecaster with fh=f1 and parameters fh_params[f1] to forecast f1, forecaster with fh=f2 and parameters fh_params[f2] to forecast f2, etc.

To use different estimators per horizon, combine FhPlexForecaster with one of MultiplexForecaster and MultiplexTransformer.

Parameters:

forecastersktime compatible forecaster

fh_paramsdict, list, callable, or str that eval-defines a callable

specifies forecaster to use per fh element

dict: keys = fh elements, values = param dict for forecaster
list: i-th entry is forecaster param dict for i-th fh element
callable: maps fh element to forecaster param dict
str: eval(fh_params) must define a lambda that maps fh element to param dict

param dict need not be complete, only overrides for forecaster params

fh_lookupstr, one of “relative” (default), “absolute”, or “as-is”

specifies fh elements used in dict or callable

if “relative”, fh will be coerced to relative ForecastingHorizon
if “absolute”, fh will be coerced to absolute ForecastingHorizon
if “as-is”, fh will be coerced to ForecastingHorizon (but not relative/absolute)

fh_contiguousbool, default=False

whether fh in inner loops are enforced to be contiguous

False: forecaster with fh_params[fN] is asked to forecast fN and only fN
True: forecaster with fh_params[fN] is asked to forecast 1, 2, …, fN and the output is then subset to the forecast of fN this is required if the forecaster can only forecast contiguous horizons

CAUTION: if using grid search inside, then True will cause the tuning metric to be evaluated on horizons 1, 2, …, fN, not just fN

Attributes:

forecasters_dict of sktime forecaster: keys are fh elements (coerced according to fh_lookup) entries are clones of forecaster used for fitting and forecasting

Examples

>>> from sktime.datasets import load_airline
>>> from sktime.forecasting.naive import NaiveForecaster
>>> from sktime.forecasting.compose import FhPlexForecaster

Simple example - same parameters per fh element >>> y = load_airline() >>> f = FhPlexForecaster(NaiveForecaster()) >>> f.fit(y, fh=[1, 2, 3]) FhPlexForecaster(…) >>> # get individual fitted forecasters >>> f.forecasters_ # doctest: +SKIP {1: NaiveForecaster(), 2: NaiveForecaster(), 3: NaiveForecaster()} >>> fitted_params = f.get_fitted_params() # or via get_fitted_params >>> y_pred = f.predict()

Simple example - different parameters per fh element >>> y = load_airline() >>> fh_params = [{}, {“strategy”: “last”}, {“strategy”: “mean”}] >>> f = FhPlexForecaster(NaiveForecaster(), fh_params=fh_params) >>> f.fit(y, fh=[1, 2, 3]) FhPlexForecaster(…) >>> # get individual fitted forecasters >>> f.forecasters_ # doctest: +SKIP {1: NaiveForecaster(), 2: NaiveForecaster(), 3: NaiveForecaster(strategy=’mean’)} >>> y_pred = f.predict()

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.