Skip to main content

Table 1 Defining DCNN model parameter

From: Accuracy of advanced deep learning with tensorflow and keras for classifying teeth developmental stages in digital panoramic imaging

Layer (type)

Output shape

Parameter

conv2d (Conv2D)

(None, 90, 90, 64)

640

activation (Activation)

(None, 90, 90, 64)

0

conv2d_1 (Conv2D)

(None, 30, 30, 64)

36,928

activation_1 (Activation)

(None, 30, 30, 64)

0

max_pooling2d (MaxPooling2D)

(None, 15, 15, 64)

0

conv2d_2 (Conv2D)

(None, 15, 15, 64)

36,928

activation_2 (Activation)

(None, 15, 15, 64)

0

conv2d_3 (Conv2D)

(None, 5, 5, 64)

36,928

activation_3 (Activation)

(None, 5, 5, 64)

0

max_pooling2d_1 (MaxPooling2

(None, 2, 2, 64)

0

flatten (Flatten)

(None, 256)

0

dense (Dense)

(None, 30)

7710

activation_4 (Activation)

(None, 30)

0

dense_1 (Dense)

(None, 6)

186

activation_5 (Activation)

(None, 6)

0

  1. Total parameters: 119,320
  2. Trainable parameters: 119,320
  3. Non-trainable parameters: 0