Diagnosis of Retinal Diseases Based on Bayesian Optimization Deep Learning Network Using Optical Coherence Tomography Images

Table 4

Hyperparameters with tuned values.

Hyperparameters	VGG16		DenseNet201		InceptionV3		Xception
	Feature extractor	Fine-tuner	Feature extractor	Fine-tuner	Feature extractor	Fine-tuner	Feature extractor	Fine-tuner
Optimizer	Adam	Adam	Adam	Adam	RMSProp	RMSProp	Adam	Adam
Learning rate	0.0001	0.0001	0.00001	0.00001	0.00001	0.0001	0.001	0.0001
Activation	tanh	tanh	elu	relu	tanh	relu	elu	relu
No. of neurons	256	512	128	128	64	512	128	256
Batch size	32	32	32	32	32	32	32	32