nicetoolbox.evaluation.metrics.results_schema

Result containers and schemas for evaluation metrics.

Classes

AggregatedResult	Result container for final summary metrics.
BatchResult	Results container during batch evaluations.
FrameResult	Result container for final frame based metrics
ResultFileGroup
ResultGrid

class nicetoolbox.evaluation.metrics.results_schema.AggregatedResult(value: float, metric_type: str, metric_name: str, component: str, algorithm: str, person: str, camera: str)

Result container for final summary metrics. No further processing needed.

class nicetoolbox.evaluation.metrics.results_schema.BatchResult(results_tensor: torch.Tensor, results_description: List[str], meta_chunk: ChunkWorkItem, meta_frames: List[FrameInfo])

Results container during batch evaluations. Further processing needed.

class nicetoolbox.evaluation.metrics.results_schema.FrameResult(value: ndarray, metric_name: str, metric_type: str, session: str, sequence: str, component: str, algorithm: str, person: str, camera: str, frame: int, axis3_description: List[str])

Result container for final frame based metrics

class nicetoolbox.evaluation.metrics.results_schema.ResultFileGroup(session: str, sequence: str, component: str, algorithm: str, metric_type: str, grids: List[nicetoolbox.evaluation.metrics.results_schema.ResultGrid] = <factory>)

class nicetoolbox.evaluation.metrics.results_schema.ResultGrid(metric_name: str, values: numpy.ndarray, description: Dict[str, Any])