From: DECIMER 1.0: deep learning for chemical image recognition using transformers
Dataset 1 | Dataset 2 | Dataset 3 | ||||||
---|---|---|---|---|---|---|---|---|
Total dataset size | 39 million | 37 million | 37 million | |||||
Subset 1 | Subset 2 | Subset 3 | Subset 4 | Subset 5 | Subset 6 | Non augmented test set | Augmented test set | |
Train dataset size | 921,600 | 10,240,000 | 15,360,000 | 35,002,240 | 15,360,000 | 33,304,320 | 33,304,320 | 33,304,320 |
Test dataset size | 102,400 | 1,024,000 | 1,536,000 | 3,929,093 | 1,536,000 | 3,700,480 | 2,000,000 | 2,000,000 |