lingvo.tasks.car.waymo.waymo_metadata module

Metadata for Waymo dataset employed in evaluation.

class lingvo.tasks.car.waymo.waymo_metadata.WaymoMetadata

Bases: lingvo.tasks.car.evaluation_metadata.EvaluationMetadata

Metadata describing the Waymo dataset used for evaluation.

Some of these entries need to be revisited and set correctly for the final version of the Waymo dataset (e.g., MaximumDistance) based off an analysis of the dataset.

ClassNames()

Returns a list of human-interpretable strings.

DifficultyLevels()

Dictionary of difficulty level strings to int32 indices.

IoUThresholds()

Dictionary of IoU thresholds for every evaluated class.

The keys of the dictionary are used to compute EvalClassIndices().

EvalClassIndices()

List of int32 indices for the classes that should be evaled.

IgnoreClassIndices()

Dictionary of int32 indices for the classes that should be ignored.

NumberOfPrecisionRecallPoints()

Number of points on the precision-recall curve.

MaximumDistance()

Maximum empirically observed Euclidean distance in world coordinates.

DistanceBinWidth()

The width of each bin for Euclidean distance in world coordinates.

MaximumNumberOfPoints()

Maximum empirically observed number of points in bounding box.

NumberOfPointsBins()

Number of logarithmically space bins for examining points.

NumberOfRotationBins()

Number of linear spaced rotations to bin.

NumberOfCalibrationBins()

Number of linear spaced calibration bins.

MinHeight2D()

Minimum height of detections to be evaluated.