Skip to main content

Table 2 Base CNN model summary

From: Autonomous detection of nail disorders using a hybrid capsule CNN: a novel deep learning approach for early diagnosis

Layer type

Output shape

Parameters

Input

(128, 128, 3)

0

Conv2D

(None, 128, 128, 32)

896

Batch_Norm

(None, 128, 128, 32)

128

Max-Pool2D

(None, 64, 64, 32)

0

Conv2d_1

(None, 64, 64, 64)

18,496

Batch_Norm_1

(None, 64, 64, 64)

256

Max-Pool2D_1

(None, 32, 32, 64)

0

Conv2d_2

(None, 32, 32, 128)

73,856

Batch_Norm_2

(None, 32, 32, 128)

512

Max-Pool2D_2

(None, 16, 16, 128)

0

Conv2d_3

(None, 16, 16, 256)

295,168

Batch_Norm_3

(None, 16, 16, 256)

1024

Max-Pool2D_3

(None, 8, 8, 256)

0

Flatten

(None, 16384)

0

Dense

(None, 512)

8,388,736

Dropout

(None, 512)

0

Dense_1

(None, 6)

3078

Total Parameters: 8,787,112

Trainable Parameters: 8,787,112

Non-Trainable Parameters: 0