Edge Impulse Experts / NVIDIA_TAO_traffic_analysis Public
Target: Nvidia Jetson Orin Nano Clone this project

Test data

Set confidence thresholds
Model variants
Test 'Unoptimized (float32)' model Test 'Quantized (int8)' model
Set the 'expected outcome' for each sample to the desired outcome to automatically score the impulse.
Sample name Expected outcome Precision score Result
small_carvideo2.s861 car, car, car 55%
Show classification Show in data acquisition
small_carvideo2.s901 car, car 100%
Show classification Show in data acquisition
small_carvideo2.s851 car, car 95%
Show classification Show in data acquisition
small_carvideo2.s801 car, car, car, car 89%
Show classification Show in data acquisition
small_carvideo2.s751 car, car, car, car 100%
Show classification Show in data acquisition
small_carvideo2.s701 car, car, car, car 70%
Show classification Show in data acquisition
small_carvideo2.s601 car, car, car 93%
Show classification Show in data acquisition
small_carvideo2.s551 car, car, car, car, car, car 76%
Show classification Show in data acquisition
small_carvideo2.s501 car, car, car, car 96%
Show classification Show in data acquisition
small_carvideo2.s451 car, car, car, car 100%
Show classification Show in data acquisition
small_carvideo2.s401 car, car, car 94%
Show classification Show in data acquisition
small_carvideo2.s351 car, car, car, car, car 76%
Show classification Show in data acquisition
small_carvideo2.s301 car, car, car, car 88%
Show classification Show in data acquisition
small_carvideo2.s251 car, car, car 96%
Show classification Show in data acquisition
small_carvideo2.s201 car, car, car 100%
Show classification Show in data acquisition
small_carvideo2.s151 car, car, car, car 94%
Show classification Show in data acquisition
small_carvideo2.s101 car, car, car, car 96%
Show classification Show in data acquisition
small_carvideo2.s51 car, car, car 94%
Show classification Show in data acquisition
small_carvideo2.s1 car, car, car, car, car 91%
Show classification Show in data acquisition
small_carvideo1.s701 car, car, car 94%
Show classification Show in data acquisition
small_carvideo1.s601 car, car, car, car 90%
Show classification Show in data acquisition
small_carvideo1.s551 car, car, car 93%
Show classification Show in data acquisition
small_carvideo1.s501 car, car, car 93%
Show classification Show in data acquisition
small_carvideo1.s151 car, car 83%
Show classification Show in data acquisition
small_carvideo1.s101 car, car 95%
Show classification Show in data acquisition
small_carvideo1.s51 car, car 85%
Show classification Show in data acquisition
small_carvideo1.s1 car, car 89%
Show classification Show in data acquisition

Model testing output

Results

Model version:
Accuracy
85.19%

Metrics for Object detection

Metric Value
mAP 0.29
mAP@[IoU=50] 0.80
mAP@[IoU=75] 0.11
mAP@[area=small] 0.24
mAP@[area=medium] 0.32
mAP@[area=large] -1.00
Recall@[max_detections=1] 0.13
Recall@[max_detections=10] 0.37
Recall@[max_detections=100] 0.37
Recall@[area=small] 0.29
Recall@[area=medium] 0.41
Recall@[area=large] -1.00

Confusion matrix

Loading...

Feature explorer

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