Datasets
tsseg packages helpers to access curated datasets used in the test suite
and benchmarks. You can load available datasets via
tsseg.data.datasets.
from tsseg.data.datasets import load_mocap
X, y = load_mocap(trial=0)
Each loader returns a tuple (X, y):
X– a 2D array with shape(n_timepoints, n_channels).y– an array of integer state labels aligned withX.
When a dataset ships with annotated change points, you can convert labels to
change-point indices using tsseg.algorithms.utils.extract_cps().
Creating custom datasets
You can wrap your own time series as lightweight dataset objects by exposing a
callable returning (X, y, metadata). For testing you may reuse the synthetic
fixtures in tests/algorithms/conftest.py:
from tsseg.tests.algorithms.conftest import synthetic_series
series = synthetic_series()
X = series["multivariate"]["X"]
y = series["multivariate"]["y"]
For large datasets, prefer storing them in an external location and provide a lazy loader to avoid shipping them inside the package.