Table 4 TextCNN-Attention parameter setting of the first layer model.
From: Emerging infectious disease surveillance using a hierarchical diagnosis model and the Knox algorithm
Model | Parameters | Explanations | Range | Step length |
|---|---|---|---|---|
TextCNN- | Learning rate = 0.001 | Learning rate | [0.0005,0.001,0.005,0.01,0.05,0.1 | |
Attention | Conv1D = 5 | Multiple attention | 0.3–0.7 | 0.1 |
multi_heads = 8 | Dimension of each head | 200–800 | 80 | |
head_dim = 16 | Neuronal inactivation | 5–15 | 5 | |
dropout = 0.5 | Number of samples for one training | |||
batch size = 520 | Early stop setting | |||
early_stopping_rounds = 10 |
