Table 2 Test datasets compared to training datasets
From: STOUT V2.0: SMILES to IUPAC name conversion using transformer models
| Â | IUPAC name character split | IUPAC name word split | ||||
|---|---|---|---|---|---|---|
Train data size | 1 million | 10 million | 50 million | 1 million | 10 million | 50 million |
Test data size | 265,332 | 972,817 | 1,024,000 | 288,152 | 990,080 | 1,024,000 |