make_pipeline#
- make_pipeline(*steps)[source]#
Create a pipeline from estimators of any type.
- Parameters
- stepstuple of sktime estimators
in same order as used for pipeline construction
- Returns
- pipesktime pipeline containing steps, in order
always a descendant of BaseObject, precise object determined by scitype equivalent to result of step[0] * step[1] * … * step[-1]
Examples
>>> from sktime.datasets import load_airline >>> y = load_airline()
Example 1: forecaster pipeline >>> from sktime.datasets import load_airline >>> from sktime.forecasting.trend import PolynomialTrendForecaster >>> from sktime.pipeline import make_pipeline >>> from sktime.transformations.series.exponent import ExponentTransformer >>> y = load_airline() >>> pipe = make_pipeline(ExponentTransformer(), PolynomialTrendForecaster()) >>> type(pipe).__name__ ‘TransformedTargetForecaster’
Example 2: classifier pipeline >>> from sktime.classification.distance_based import KNeighborsTimeSeriesClassifier >>> from sktime.pipeline import make_pipeline >>> from sktime.transformations.series.exponent import ExponentTransformer >>> pipe = make_pipeline(ExponentTransformer(), KNeighborsTimeSeriesClassifier()) >>> type(pipe).__name__ ‘ClassifierPipeline’
Example 3: transformer pipeline >>> from sktime.pipeline import make_pipeline >>> from sktime.transformations.series.exponent import ExponentTransformer >>> pipe = make_pipeline(ExponentTransformer(), ExponentTransformer()) >>> type(pipe).__name__ ‘TransformerPipeline’