Table 4 The parameters used in the pre-trained CNN models.

From: Automated identification of autism spectrum disorder from facial images using explainable deep learning models

Parameter

Value

Dense layer

128,64

Global max pooling layer size

\(3 \times 3\)

Output classification layer

Softmax

Activation function

Sigmoid

Optimizer

ADAM

Learning Rate

0.0001

Number of epochs

100

Batch Size

32

Back to article page