Edge Impulse Experts / Adiuvo_BrainChip_VWW Public
Target: BrainChip AKD1000 or AKD1500 Clone this project

Test data

Set thresholds
Model variants
Test 'Unoptimized (float32)' model Test 'Quantized (akida)' model
Set the 'expected outcome' for each sample to the desired outcome to automatically score the impulse.
Sample name Expected outcome Accuracy Result
COCO_train2014_000000080391 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000080279 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000080231 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000080084 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000080180 person 0% 1 non-person
Show classification Show in data acquisition
COCO_train2014_000000079968 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000079912 person 0% 1 non-person
Show classification Show in data acquisition
COCO_train2014_000000079680 person 0% 1 non-person
Show classification Show in data acquisition
COCO_train2014_000000079806 person 0% 1 uncertain
Show classification Show in data acquisition
COCO_train2014_000000079641 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000079517 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000079503 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000079459 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000079456 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000079313 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000079242 person 0% 1 non-person
Show classification Show in data acquisition
COCO_train2014_000000079058 person 0% 1 uncertain
Show classification Show in data acquisition
COCO_train2014_000000078963 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000078861 person 0% 1 uncertain
Show classification Show in data acquisition
COCO_train2014_000000078684 person 100% 1 person
Show classification Show in data acquisition
COCO_train2014_000000078553 person 0% 1 uncertain
Show classification Show in data acquisition

Model testing output

Results

Model version:
Accuracy
80.55%

Confusion matrix

non-personpersonuncertain
non-person81.2%8.5%10.3%
person11.5%79.9%8.6%
f1 score0.840.85

Feature explorer

Error while loading features: {"response":{"readyState":0,"status":0,"statusText":"error"},"errorThrown":""}
Help & documentation Get help on the forum Send feedback Contact sales
Join us
Terms of Service Privacy policy