Pick And Place - Model testing

Edge Impulse Experts / Pick And Place Public

Target: Raspberry Pi 4 Clone this project

This lists all test data. You can manage this data through Data acquisition.

Test data

Set the 'expected outcome' for each sample to the desired outcome to automatically score the impulse.

Sample name	Expected outcome	Precision score
pnp_91	penguin, penguin, penguin, penguin, pig, pig	100%	Show classification Show in data acquisition
pnp_71	penguin, penguin, penguin, pig, pig, pig, pig	100%	Show classification Show in data acquisition
pnp_99	pig, pig, pig, pig	100%	Show classification Show in data acquisition
pnp_83	penguin, penguin	100%	Show classification Show in data acquisition
pnp_74	penguin, penguin	100%	Show classification Show in data acquisition
pnp_75	penguin, pig, pig, pig, pig	100%	Show classification Show in data acquisition
pnp_79	penguin, penguin, penguin, penguin, penguin	100%	Show classification Show in data acquisition
pnp_62	penguin, penguin, penguin, penguin, penguin, pig, pig	100%	Show classification Show in data acquisition
pnp_64	penguin, penguin, penguin, penguin	93%	Show classification Show in data acquisition
pnp_69	pig, pig, pig, pig	100%	Show classification Show in data acquisition
pnp_42	pig	100%	Show classification Show in data acquisition
pnp_41	pig	100%	Show classification Show in data acquisition
pnp_36	pig, pig	100%	Show classification Show in data acquisition
pnp_1	penguin, penguin, penguin, penguin, penguin, pig, pig, pig, pig	100%	Show classification Show in data acquisition
pnp_43	penguin, penguin, penguin, penguin, penguin	100%	Show classification Show in data acquisition
pnp_14	pig, pig	100%	Show classification Show in data acquisition
pnp_21	penguin, penguin, penguin, penguin, pig, pig, pig, pig	100%	Show classification Show in data acquisition
pnp_32	penguin, pig, pig, pig	100%	Show classification Show in data acquisition
pnp_16	penguin, penguin, penguin, penguin, penguin	100%	Show classification Show in data acquisition
pnp_101	penguin	100%	Show classification Show in data acquisition

Model testing output

Results

Accuracy

100.00%

Metrics for Object detection

Metric	Value
mAP	0.64
mAP@[IoU=50]	0.98
mAP@[IoU=75]	0.80
mAP@[area=small]	0.58
mAP@[area=medium]	0.65
mAP@[area=large]	-1.00
Recall@[max_detections=1]	0.24
Recall@[max_detections=10]	0.71
Recall@[max_detections=100]	0.71
Recall@[area=small]	0.62
Recall@[area=medium]	0.72
Recall@[area=large]	-1.00

Feature explorer

These users will be notified over email when this job finishes.

	Name
	Naveen

Test data

Add data

Model testing output

Results

Accuracy

Metrics for Object detection

Confusion matrix

Feature explorer

Test data

Model variants

Add data

Model testing output

Results

Accuracy

Metrics for Object detection

Confusion matrix

Feature explorer

Settings