load_from_timeeval_csv_file¶
- load_from_timeeval_csv_file(path: Path) tuple[ndarray, ndarray][source]¶
Load a TimeEval-formatted CSV file into memory.
TimeEval datasets are stored in simple CSV files with the following format:
The first column contains the timestamps; usually with the header ‘timestamp’.
The last column contains the binary anomaly labels; usually with the header ‘is_anomaly’.
The columns in between contain the time series data.
This function does not parse the timestamp information because the data is expected to be equidistant and has no missing values.
- Parameters:
- pathPath
Full or relative path to the CSV file.
- Returns:
- Xnp.ndarray
The univariate (1d) or multivariate (2d) time series with shape (n_instances, n_channels). If univariate the shape is (n_instances,).
- ynp.ndarray
The binary anomaly labels with shape (n_instances,).