Table 3 Performance results on BreakHis with ResNet50.

From: Applying YOLOv6 as an ensemble federated learning framework to classify breast cancer pathology images

Epochs

Ac

L

VA

VL

1

81.48

44.43

84.50

41.34

10

85.23

36.22

78.89

42.45

20

87.55

31.34

87.02

34.34

30

89.23

26.45

85.79

33.20

40

88.91

27.45

87.54

34.21

50

90.91

21.36

84.69

39.67

60

93.29

18.24

84.25

36.16

70

91.55

21.56

89.47

32.68

80

93.88

15.77

88.79

30.12

90

94.03

15.11

90.89

25.24

100

94.89

11.89

91.21

26.45

Back to article page