load_gun_point_segmentation

load_gun_point_segmentation()

Load the GunPoint time series segmentation problem and returns X.

We group TS of the UCR GunPoint dataset by class label and concatenate all TS to create segments with repeating temporal patterns and characteristics. The location at which different classes were concatenated are marked as change points.

We resample the resulting TS to control the TS resolution. The window sizes for these datasets are hand-selected to capture temporal patterns but are approximate and limited to the values [10,20,50,100] to avoid over-fitting.

Returns:

Xpd.Series

Single time series for segmentation

period_lengthint

The annotated period length by a human expert

change_pointsnumpy array

The change points annotated within the dataset

Examples

>>> from aeon.datasets import load_gun_point_segmentation
>>> X, period_length, change_points = load_gun_point_segmentation()