Detection Metrics Dashboard

Tracking detector performance vs Python GT and Hand-Annotated GT.

Last updated: 2026-01-10 16:56:52 UTC

vs Python GT

Overall score

0–1

–

vs prev:

Harmonic mean

–

vs prev:

Precision / Recall

Latest

–

P / R

FP / FN

Counts

–

false positives / negatives

Overall score

0–1

–

vs prev:

Harmonic mean

–

vs prev:

Precision / Recall

Latest

–

P / R

FP / FN

Counts

–

false positives / negatives

Overall score over time

Precision / Recall / F1 Both GTs per commit

False positives / negatives Both GTs per commit

Mean IoU vs combined error Both GTs - matched detections only

Per-commit metrics

Time (UTC)	Commit	Python GT						Hand-Annotated GT
Time (UTC)	Commit	Overall	P	R	F1	FP	FN	Overall	P	R	F1	FP	FN