EvaluationSubsetReport

class openstef_beam.evaluation.EvaluationSubsetReport(**data: Any) → None

Bases: BaseModel

Container for evaluation results on a specific data subset.

Bundles filtering criteria, evaluation subset data, and computed metrics for a particular slice of the evaluation dataset. Enables persistence and retrieval of evaluation results for analysis.

Parameters:

data (Any)

filtering: TypeAliasType

subset: ForecastDataset

metrics: list[SubsetMetric]

to_parquet(path: Path)

Save the subset report to parquet files in the specified directory.

Parameters:

• path (Path) – Directory where to save the report data.

• path

classmethod read_parquet(path: Path) → Self

Load a subset report from parquet files in the specified directory.

Parameters:

• path (Path) – Directory containing the saved report data.

• path

Returns:

Loaded EvaluationSubsetReport instance.

Return type:

Self

get_global_metric() → SubsetMetric | None

Returns the SubsetMetric with window=’global’, or None if not found.

Return type:

SubsetMetric | None

get_windowed_metrics() → list[SubsetMetric]

Returns all SubsetMetrics with window != ‘global’.

Return type:

list[SubsetMetric]

get_measurements() → Series

Extract measurements Series from the report for the given target.

Return type:

Series

Returns:

Ground truth measurements as a pandas Series.

get_quantile_predictions() → DataFrame

Extract forecasted quantiles from the report.

Return type:

DataFrame

Returns:

Dataset containing forecasted quantile predictions.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'protected_namespaces': (), 'ser_json_inf_nan': 'null'}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].