Skip to content

mAP - torchmetrics

Category: metric.detection.2d
Version: 1.0.0
UUID: a7251cdb40cc40ca6155782a6d4e26d505bf5b79300e2e227ed17cabdf2ef780
Execution context: leip_af
Help: Computes precision and recall for different object sizes, and per class. Refer here

Value Parameters

Name Synonyms Type Values
Whether to enable metrics per-class metric.apply_per_class scalar bool
IOU thresholds for evaluation metric.iou_thresholds choice [0.5, 0.55, 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95] (0.5 : .05 : 1.0)
[0.5, 0.75, 0.95] (0.5 : .25 : 1.0)
[0.5] (0.5)

This component fits into

Name UUID Synonyms
Default Evaluation c24a20... metrics.evaluation.algorithm
Training accumulator fc94f7... metrics.train.algorithm