Table 5 Comparison of the attention mechanism and its impact on model performance.

From: Automatic pain classification in older patients with hip fracture based on multimodal information fusion

Optimization Algorithm

Precision

F1-Score

Loss value (Loss)

Training time (hours)

No attention mechanism

76.5%

73.3%

0.210

2.5

First-order attention mechanism

81.7%

000.001

81.5%

0.37 0

2.8

Sparse-based attention mechanisms

79.8%

77.5%

0.180

2.9

Back to article page