Back to models
Transformer

TemporianTransformer

API referenceSource
MultivariateUnequal-length series

Applies a Temporian function to the input time series.

This transformer applies a Temporian [1] function to the input time series.

The conversion from sktime’s internal representation to Temporian’s EventSet [2] and back is handled automatically by the transformer.

Quickstart

python
from sktime.transformations.temporian import TemporianTransformer

estimator = TemporianTransformer(function, compile=False)

Tags

Capabilities

  • Multivariate
  • Unequal-length series
  • Inverse transform: Not supported
  • Missing values: Not supported
  • Removes missing values: Not supported
  • Equalizes series length: Not supported

Properties

Input typescitype:transform-input
Series
Output typescitype:transform-output
Series
Label typescitype:transform-labels
None
Fit is emptyfit_is_empty
Yes
Keeps the time indextransform-returns-same-time-index
No
Requires Xrequires_X
Yes
Requires yrequires_y
No
X and y need the same indexX-y-must-have-same-index
No

Parameters(2)

functionCallable[[temporian.EventSet], temporian.EventSet]
Temporian function to apply to the input time series. The function must receive and return a single Temporian EventSet, and can apply an arbitrary number of Temporian operators to its input.
compilebool, default=False

If True, the function will be compiled using Temporian’s @tp.compile [3] decorator, which can lead to significant speedups by optimizing the graph of operations.

Examples

>>> from sktime.datasets import load_airline
>>> from sktime.transformations.temporian import TemporianTransformer
>>> import temporian as tp
>>> 
>>> def function (evset):
... return evset. simple_moving_average (tp. duration. days (3 * 365))
>>> transformer = TemporianTransformer (function = function)
>>> X = load_airline ()
>>> X_averaged = transformer. fit_transform (X)

References