Skip to main content

Table 1 Yolov8 object detection model validation metrics

From: Enhancing fall risk assessment: instrumenting vision with deep learning during walks

Epoch

Precision

Recall

mAP50

Box loss

Cls loss

DFL loss

62

0.89

0.92

0.92

1.06

0.58

1.02

63

0.88

0.93

0.92

1.06

0.57

1.03

64

0.90

0.93

0.93

1.06

0.57

1.03

65

0.89

0.92

0.92

1.07

0.57

1.03

66

0.88

0.91

0.92

1.06

0.57

1.02

  1. Typical for best epoch indicated in bold
  2. DFL loss Distribution focal loss, Cls loss Class loss, Box loss Bounding box loss
Back to article page

Journal of NeuroEngineering and Rehabilitation

ISSN: 1743-0003

Contact us