dqa.tasks.time_series.SplitByIntervals#
- class dqa.tasks.time_series.SplitByIntervals(timestamp_name: str, interval_dataset: str, interval_start: str, interval_end: str, interval_class: str | None = None, class_key: str = 'Class', **kwargs)#
Splits time series up using a list of intervals.
In the main dataset one or multiple time series with one common timestamp row are given. In another dataset, a table of intervals is given. This contains one data row for the interval start, interval end and optionally an interval class, respectively. Then the time series are split into multiple measurements by the timestamps according to the intervals. The input is specified with the input_dataset, input_name parameters. The intervals must be sorted and non-overlapping. The SortIntervals task can be used to sort and clean intervals if needed.
- Parameters:
timestamp_name (str) – The name of the timestamp data row in the input dataset.
interval_dataset (str) – The dataset containing the intervals.
interval_start (str) – Name of the data row containing the interval start timestamps.
interval_end (str) – Name of the data row containing the interval end timestamps.
interval_class (str, default=None) – The class index assigned to the metadata of the output. If None, then no class index will be assigned.
class_key (str, default='Class') – The name that is used to register the interval class in the metadata of the resulting measurements.
Methods
finish()Can perform actions that are required to clean up after the task has finished, e.g. close network connections etc.
in_out_default
input_output_dataset
input_output_machine
input_output_mode
input_output_name
log
modify_data_row
modify_dataset
modify_dataset_dict
modify_machine
modify_measurement
set_logging_level
transfer_metadata