Skip to main content

Table 11 Classifier performance with A1 ∪ A2 ∪ A3 fusion strategy

From: Fusing pre-trained convolutional neural networks features for multi-differentiated subtypes of liver cancer on histopathological images

Classifier

Strategies

A1 ∪ A2 ∪ A3

Stacking

Test acc/validation acc (unit: %)

63.04/94.67

Images of precision, recall and F1 values

View full size image

Bagging

Test acc/validation acc (unit: %)

51.45/78.33

Images of precision, recall and F1 values

View full size image

Gradient-boosting

Test acc/validation acc (unit: %)

57.97/88.33

Images of precision, recall and F1 values

View full size image