Skip to main content

Table 8 Optimized parameters for hyper-tuning of the customized CNN

From: An effective correlation-based data modeling framework for automatic diabetes prediction using machine and deep learning techniques

Hyperparameter

Assigned value

Model

Customized CNN

Input

Numerical features

Input size

768x8 (original)

768x32 (data modeling)

Input labels

2

(Diabetic and Normal)

Learning rate

0.0001

Number of epochs

80

Batch Size

8

Optimizer

ADAM

Loss function

Cross entropy

Training time

1024 sec

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us