Skip to main content

Table 2 Parameter values of the proposed model

From: Automatic multi-disease classification on retinal images using multilevel glowworm swarm convolutional neural network

Parameters

Range

Input size

256 × 256 × 3

Output size

5

Epoch

50

Optimizer

Adam

Learning rate

0.0010340233 (using glowworm swarm optimization)

Loss function

Categorical_Crossentropy

Convolutional kernel size

3

Pooling kernel size

2

Activation function for activation layer

“relu”

Batch size

64

Activation function

Softmax