load_human_activity_segmentation_datasets

load_human_activity_segmentation_datasets(extract_path: PathLike | None = None, return_metadata: bool = False) → tuple[list[ndarray], list[ndarray]] | tuple[list[ndarray], list[ndarray], list[tuple[str, str, int, int, ndarray]]]

Load the Human Activity Segmentation Challenge data sets.

This function loads the Human Activity Segmentation challenge data sets into memory, downloading from GitHub (https://github.com/patrickzib/human_activity_segmentation_challenge) [1] if the data is not available at the specified extract_path. The data sets were used in the discovery challenge held at ECML/PKDD and AALTD 2023. They contain 250 annotated TS with 1-15 segments, capturing a total of 15 students performing 6 distinct motion sequences. TS are sampled at 50 Hz, multivariate and consist of measurements from 9 out 12 smartphone sensors: triaxial accelerometer, gyroscope, magnetometer as well as latitude, longitude, and speed. Annotations include information about the challenge split (public / private), groups and subjects, as well as activity transition offsets (the change points) and activity labels.

If you do not specify extract_path, it will set the path to aeon/datasets/local_data. If the problem is not present in extract_path, it will attempt to download the data.

Parameters:

extract_pathstr, default=None

The path to look for the data. If no path is provided, the function looks in aeon/datasets/local_data/. If a path is given, it can be an absolute, e.g., C:/Temp/ or relative, e.g. Temp/ or ./Temp/, path to an existing CSV-file.

return_metadataboolean, default = False

If True, returns a tuple (X, y, metadata).

Returns:

X: list of np.ndarray

The list of multivariate (2d) time series with variable shape (n_instances, 9).

y: list of np.ndarray

The list of change points for every time series.

metadata: optional

The list of tuples containing data set names, splits, groups, subjects, and activities information.

Raises:

URLError or HTTPError

If the GitHub repository is not accessible.

References

[1]

Arik Ermshaus, Patrick Schäfer, Anthony Bagnall, Thomas Guyet, Georgiana Ifrim, Vincent Lemaire, Ulf Leser, Colin Leverger, Simon Malinowski: Human Activity Segmentation Challenge @ ECML/PKDD’23. AALTD@ECML, 2023, DOI:10.1007/978-3-031-49896-1_1.

Examples

>>> from aeon.datasets import load_human_activity_segmentation_datasets
>>> X, y = load_human_activity_segmentation_datasets()
... )