Zurück zu den Modellen
Regressor

TimeSeriesForestRegressor

API-ReferenzQuellcode

Time series forest regressor.

A time series forest is an ensemble of decision trees built on random intervals.

Overview: For input data with n series of length m, for each tree:

  • sample sqrt(m) intervals,

  • find mean, std and slope for each interval, concatenate to form new data set,

  • build decision tree on new data set.

Ensemble the trees with averaged probability estimates.

This implementation deviates from the original in minor ways. It samples intervals with replacement and does not use the splitting criteria tiny refinement described in [1]. This is an intentionally stripped down, non configurable version for use as a HIVE-COTE component.

Schnellstart

python
from sktime.regression.interval_based import TimeSeriesForestRegressor

estimator = TimeSeriesForestRegressor(min_interval=3, n_estimators=200, n_jobs=1, random_state=None)

Parameter(4)

n_estimatorsint, default=200
Number of estimators.
min_intervalint, default=3
Minimum width of an interval.
n_jobsint, default=1

The number of jobs to run in parallel for both fit and predict. -1 means using all processors.

random_stateint, default=None

Beispiele

>>> from sktime.regression.interval_based import TimeSeriesForestRegressor
>>> from sktime.datasets import load_unit_test
>>> X_train, y_train = load_unit_test (split = "train")
>>> X_test, y_test = load_unit_test (split = "test")
>>> regressor = TimeSeriesForestRegressor (n_estimators = 150)
>>> regressor. fit (X_train, y_train) TimeSeriesForestRegressor(n_estimators=150)
>>> y_pred = regressor. predict (X_test)

Referenzen

[1]

H.Deng, G.Runger, E.Tuv and M.Vladimir, “A time series forest for classification and feature extraction”, Information Sciences, 239, 2013