Table 1 Basic analysis of SPACCC corpus documents
From: Combining word embeddings to extract chemical and drug entities in biomedical literature
Train | Dev | Test | |
|---|---|---|---|
Number of documents | 500 | 250 | 250 |
Avg sentences | 25.14 | 25.85 | 25.69 |
No. tokens | 202,901 | 96,869 | 100,963 |
No. unique tokens | 18,623 | 12,170 | 12,442 |