Edge Impulse Inc. / Tutorial: object detection Public
Target: Nvidia Jetson Orin Nano Clone this project

Test data

Set confidence thresholds
Model variants
No additional model variants available Test 'Unoptimized (float32)' model
Set the 'expected outcome' for each sample to the desired outcome to automatically score the impulse.
Sample name Expected outcome Precision score Result
sample.jpg.22l752sg coffee, lamp 94%
Show classification Show in data acquisition
sample.jpg.22l74u4f coffee, lamp 100%
Show classification Show in data acquisition
sample.jpg.22l74rsv coffee, lamp 100%
Show classification Show in data acquisition
sample.jpg.22l74kk8 coffee 89%
Show classification Show in data acquisition
sample.jpg.22l74iqv coffee 100%
Show classification Show in data acquisition
sample.jpg.22l74d6p coffee, lamp 94%
Show classification Show in data acquisition
sample.jpg.22l7479b coffee, lamp 100%
Show classification Show in data acquisition
sample.jpg.22l73stu coffee, lamp 94%
Show classification Show in data acquisition
sample.jpg.22l738q8 coffee 89%
Show classification Show in data acquisition
sample.jpg.22l737ca coffee 100%
Show classification Show in data acquisition
sample.jpg.22l7356n coffee 100%
Show classification Show in data acquisition
sample.jpg.22l72va3 lamp 89%
Show classification Show in data acquisition
sample.jpg.22l72pjd lamp 100%
Show classification Show in data acquisition

Model testing output

Results

Model version:
Accuracy
100.00%

Metrics for Object detection

Metric Value
mAP 0.68
mAP@[IoU=50] 1.00
mAP@[IoU=75] 0.82
mAP@[area=small] -1.00
mAP@[area=medium] 0.67
mAP@[area=large] 0.69
Recall@[max_detections=1] 0.72
Recall@[max_detections=10] 0.72
Recall@[max_detections=100] 0.72
Recall@[area=small] -1.00
Recall@[area=medium] 0.68
Recall@[area=large] 0.75

Confusion matrix

Loading...

Feature explorer

Loading...
Help & documentation Get help on the forum Send feedback Contact sales
Join us
Terms of Service Privacy policy