dataeval.shift.DriftMVDCOutput¶

class dataeval.shift.DriftMVDCOutput(results_data)¶

Results of the multivariate domain classifier drift detection.

Parameters:¶

results_data : polars.DataFrame¶

data()¶

Return a copy of the results data.

Return type:¶

polars.DataFrame

filter(period='all', metrics=None)¶

Filter results by period and optionally by metric.

Parameters:¶

period : str, default "all"¶

Filter by period: “all”, “reference”, or “analysis”.

metrics : str or None, default None¶

Metric name to filter by. Currently only “domain_classifier_auroc” is supported. If None, all metrics are included.

Returns:¶

Filtered results object.

Return type:¶

DriftMVDCOutput

Raises:¶

• ValueError – If metrics parameter is not a string or None.

• KeyError – If the requested metric is not available.

meta()¶

Metadata about the execution of the function or method for the Output class.

Return type:¶

ExecutionMetadata

property empty : bool¶

Check if the results are empty.

Return type:¶

bool

property plot_type : 'drift_mvdc'¶

Return the plot type identifier.

Return type:¶

Literal[‘drift_mvdc’]