Enhancing Photovoltaic Module Fault Diagnosis with Unmanned Aerial Vehicles and Deep Learning-Based Image Analysis

Table 10

Training (a), validation (b), and test (c) accuracies of Bayes-based classifiers for various pretrained networks.

(a) Training accuracy of Bayes-based classifiers for various pretrained networks

Classifier	Training accuracy (%)
	AlexNet	DenseNet-201	GoogleNet	ResNet-50	VGG16	VGG19
Bayes net	96.38	96.58	93.25	96.15	92.18	94.04
Naïve Bayes	93.01	91.15	89.24	90.63	84.48	90.35
Naïve Bayes multinomial text	16.66	16.66	16.66	16.66	16.66	16.66
Naïve Bayes updateable	93.01	91.15	89.24	90.63	84.48	90.35

(b) Validation accuracy of Bayes-based classifiers for various pretrained networks

Classifier	Validation accuracy (%)
	AlexNet	DenseNet-201	GoogleNet	ResNet-50	VGG16	VGG19
Bayes net	93.01	93.13	89.88	92.77	88.53	90.31
Naïve Bayes	92.50	91.15	88.49	90.07	83.92	89.80
Naïve Bayes multinomial text	16.66	16.66	16.66	16.66	16.66	16.66
Naïve Bayes updateable	92.50	91.15	88.49	90.07	83.92	89.80

(c) Test accuracy of Bayes-based classifiers for various pretrained networks

Classifier	Test accuracy (%)
	AlexNet	DenseNet-201	GoogleNet	ResNet-50	VGG16	VGG19
Bayes net	94.60	94.13	89.33	93.65	87.30	89.84
Naïve Bayes	93.01	91.42	88.09	89.84	83.96	88.09
Naïve Bayes multinomial text	16.66	16.66	20.00	16.66	16.66	16.66
Naïve Bayes updateable	93.01	91.42	89.19	89.84	83.96	88.09

Boldface entries represent the highest values obtained.