Dmitry / fruit_detection Public
Set the 'expected outcome' for each sample to the desired outcome to automatically score the impulse.

Apply filters

Model testing output

Results

Model version:
Precision score (legacy) 58.6%
mAP 0.45
mAP@[IoU=50] 0.64
mAP@[IoU=75] 0.55
mAP@[area=small] 0.00
mAP@[area=medium] 0.43
mAP@[area=large] 0.66
Recall@[max_detections=1] 0.37
Recall@[max_detections=10] 0.48
Recall@[max_detections=100] 0.48
Recall@[area=small] 0.00
Recall@[area=medium] 0.44
Recall@[area=large] 0.70
Accuracy (legacy) 45.45%
Loading...
Loading...