ZeMA dataset API
An API for accessing the data in the ZeMA remaining-useful life dataset
- class zema_emc_annotated.dataset.ExtractionDataType(value)[source]
Identifiers of data types in ZeMA dataset
- class zema_emc_annotated.dataset.ZeMASamples(n_samples: int = 1, size_scaler: int = 1, normalize: bool = False, idx_start: int = 0)[source]
Extracts requested number of samples of values with associated uncertainties
The underlying dataset is the annotated “Sensor data set of one electromechanical cylinder at ZeMA testbed (ZeMA DAQ and Smart-Up Unit)” by Dorst et al. [Dorst2021].
- Parameters
n_samples (int, optional) – number of samples each containing the first
size_scaler
readings from each of the eleven sensors for one of the cycles with associated uncertainties, defaults to 1 and must be between 1 and 4766 - idx_startsize_scaler (int, optional) – number of sensor readings from each of the individual sensors per sample/cycle, defaults to 1 and should be between 1 and 2000, as there are only 2000 readings per cycle, higher values will be clipped to 2000
normalize (bool, optional) – if
True
, then values are centered around zero and values and uncertainties are scaled to values’ unit std, defaults toFalse
idx_start (int, optional) – index of first sample to be extracted, defaults to 0 and must be between 0 and 4765
- uncertain_values
The collection of samples of values with associated uncertainties, will be of shape (n_samples, 11 x size_scaler)
- Type