From: Improving reference prioritisation with PICO recognition
Layer | Size | Source | |
---|---|---|---|
1a | Word embedding | 200 | [21], not updated |
1b | Character embedding | 28 | trained from random initialisation |
1c | Character-based word representation | 2 ×28 | biLSTM applied to 1b |
1d | Combined embedding | 256 | concatenation of 1a and 1c |
2 | Recurrent layer | 2 ×128 | biLSTM over 1d |
3 | Linear layer | 41 | affine projection of 2 |
 | CRF output | 1 | most likely sequence of tags |