Datasets¶

This module is for loading time series data sets

seglearn.datasets.load_watch()[source]¶

Loads some of the 6-axis inertial sensor data from my smartwatch project. The sensor data was recorded as study subjects performed sets of 20 shoulder exercise repetitions while wearing a smartwatch. It is a multivariate time series.

The study can be found here: https://arxiv.org/abs/1802.01489

Returns

datadict

data[‘X’]list, length 140: inertial sensor data, each element with shape [n_samples, 6]

sampled at 50 Hz
data[‘y’]array, length 140: target vector (exercise type)
data[‘side’]array, length 140: the extremity side, 1 = right, 0 = left
data[‘subject’]array, length 140: the subject (participant) number
data[‘X_labels’]str list, length 6: ordered labels for the sensor data variables
data[‘y_labels’] :str list, length 7: ordered labels for the target (exercise type)

Examples

>>> from seglearn.datasets import load_watch
>>> data = load_watch()
>>> print(data.keys())