Table 3 Hyper parameters of RFAB
From: A risk factor attention-based model for cardiovascular disease prediction
Parameter | Description | Value |
|---|---|---|
\(d_w\) | Dimension of word embedding | 100 |
lr | Learning rate | le−3 |
B | Batch size | 10 |
dp | Each neuron’s deactivation rate | 0.5 |
dr | Decay rate for lr | 0.99 |
ds | Number of decay steps | 500 |
h | Each BiLSTM’s hidden unit quantity | 256 |
n | Number of epochs | 60 |