Table 2 Hyper-parameter settings
From: A neural joint model for entity and relation extraction from biomedical text
| Type | Hyper-parameter |
|---|---|
| Training | α=0.03,λ=10−8 |
| Embedding | dim(emb(w i ))=200 |
| \(dim(emb(p_{i}), emb(d_{i})~\text {or}~emb(l^{e}_{i}))=25\) | |
| CNN | dim(emb(c))=25,C=3 |
| dim(r w )=25 | |
| Bi-LSTM-RNN (Entity) | \(dim(\overrightarrow {h}_{i})\) or \(dim(\overleftarrow {h}_{i})=100\) |
| \(dim(h^{e}_{i})=100\) | |
| Bi-LSTM-RNN (Relation) | dim(↑ h a ,↑ h b ,↓ h a or ↓ h b )=100 |
| dim(h r)=100 |