Table 4 Experimental results of optimizer comparison

From: Convolutional neural network based on transfer learning for discriminating the fermentation degree of black tea

Network

Optimizer

FLOPs

Params

Accuracy

Precision

Recall

F1

FPS

ResNet18

RMSProp

1.824G

11.178M

0.9265

0.91639

0.86295

0.88362

73.75

SGD

0.8877

0.82844

0.7937

0.80936

73.83

AdamW

0.9425

0.9272

0.8881

0.9064

74.60

Back to article page