Table 16 The training time and model parameters are compared between the suggested methods and other conventional DL models.

From: An interpretable framework for gastric cancer classification using multi-channel attention mechanisms and transfer learning approach on histopathology images

Framework/Model

Size (MB)

Time(s)

VGG-1684

512

7060

Xception86

79.6

4015

InceptionResNet-V189

30.8

3260

AlexNet136

217

1331

Inception-V387

83.4

5340

DenseNet-12185

27.1

2860

ResNet-5083

90

4772

ResNeXt-5088

88

4564

BoTNet-5098

72.1

4772

ViT92

31.2

1502

CoaT96

21.0

3120

DeiT94

21.1

2566

CaiT93

460

6956

LeViT97

65.8

2943

T2T-ViT95

16.01

2792

gMLP143

73.2

6396

MLP-Mixer144

225

11284

ResMLP145

169

8943

MCAM

613

4212

Back to article page