load_arrow_head(split=None, return_X_y=True, return_type='numpy3d')[source]

Load the ArrowHead univariate time series classification problem.

split: None or one of “TRAIN”, “TEST”, default=None

Whether to load the train or test instances of the problem. By default it loads both train and test instances into a single array.

return_X_y: bool, default=True

If True, returns (features, target) separately instead of as single data structure.

return_type: string, default=”numpy3d”

Data structure to use for time series, should be either “numpy2d” or “numpy3d”.


shape (n_cases, 1, 251) (if return_type=”numpy3d”) or shape (n_cases, 251) (return_type=”numpy2d”), where n_cases where n_cases is either 36 (split = “train”), 175 (split=”test”) or 211.

y: np.ndarray

1D array of length 36, 175 or 211, only returned if return_X_y is True The class labels for each time series instance in X If return_X_y is False, y is appended to X instead.

ValueError is raised if the data cannot be stored in the requested return_type.


Dimensionality: univariate Series length: 251 Train cases: 36 Test cases: 175 Number of classes: 3 Details: http://timeseriesclassification.com/description.php?Dataset=ArrowHead


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