Table 5 Comparison of the performance of state-of-the-art CNN architectures.

From: An explainable hybrid feature aggregation network with residual inception positional encoding attention and EfficientNet for cassava leaf disease classification

Architecture

Number of

Parameters (M)

Accuracy

(in %)

Precision

(in %)

Recall

(in %)

F1-score

(in %)

Inference Time

(per sample)

Xception

22.8

77.0

62.0

54.0

57.72

0.237

VGG16

138

77.56

61.89

59.25

60.56

0.155

AlexNet

57

77.97

61.20

55.69

58.30

0.612

DenseNet

7.3

79.25

67.34

55.45

60.82

0.625

Resnet50

24.11

84.25

74.0

69.30.

71.61

0.213

EfficientNet

4.37

88.09

88.98

88.60

88.79

0.378

Proposed Network

49

93.06

88.18

87.18

87.62

0.088

Back to article page