L2MXception: an improved Xception network for classification of peach diseases

Table 1 Results and parameters based on seven original models

Network	Batch size	Epoch	Learning rate	Training accuracy (%)	Validation accuracy (%)
AlexNet	64	60	0.001	72.02	70.55
ResNet50	64	60	0.001	68.28	65.23
Xception	64	60	0.001	67.86	65.37
SENet154	64	60	0.001	53.00	56.63
DenseNet169	32	60	0.001	90.49	89.32
HRNet-w48	64	60	0.001	89.06	80.91
MobileNetV3	64	60	0.001	57.63	57.60

ISSN: 1746-4811