pyvale.valid.validation module¶

class PointValData(val_points: dict[str, numpy.ndarray] = <factory>, epistemic_intervals: dict[str, numpy.ndarray | None] = <factory>, val_label_to_ind: dict[tuple[str, str], int] = <factory>, ind_to_val_label: dict[tuple[str, int], str] = <factory>)[source]¶

Bases: object

val_points¶

shape=(n_sensors,n_epistemic,n_aleatory) EXP:shape=(n_sensors,n_epistemic,n_steady_repeats)

Type:: SIM

epistemic_intervals¶: shape=(n_sensors,2), where 2 = (low,high)

val_label_to_ind¶

ind_to_val_label¶: Use these to index into the above numpy arrays

__init__(val_points=<factory>, epistemic_intervals=<factory>, val_label_to_ind=<factory>, ind_to_val_label=<factory>)¶

extract_val_data_by_key(exp_data, epistemic_intervals, sensor_keys, steady_slice=None)[source]¶

extract_val_data_by_slice(exp_data, epistemic_intervals, sensor_keys, steady_slice=None)[source]¶

class ImageValData(val_images: numpy.ndarray, epistemic_intervals: numpy.ndarray | None = None)[source]¶

Bases: object

val_images¶

epistemic_intervals¶

__init__(val_images, epistemic_intervals=None)¶

load_val_data(load_file)[source]¶

gen_val_data(nominal_data, aleatory_gen, epistemic_gen)[source]¶

calc_limit_cdfs_point(val_data)[source]¶

calc_mavm_point(exp_data, sim_data)[source]¶

vectorised_ecdf(data, axis)[source]¶