#This model is part of a test for creating multilingual BioMedical NER systems. Not intended for proffesional use now.
This model is a fine-tuned version of bert-base-multilingual-cased on the CRAFT+BC4CHEMD+BioNLP09 datasets concatenated. It achieves the following results on the evaluation set:
More information needed
More information needed
More information needed
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0658 | 1.0 | 6128 | 0.0751 | 0.9795 | 0.9795 | 0.9795 | 0.9758 |
| 0.0406 | 2.0 | 12256 | 0.0753 | 0.9827 | 0.9815 | 0.9821 | 0.9786 |
| 0.0182 | 3.0 | 18384 | 0.0934 | 0.9834 | 0.9825 | 0.9829 | 0.9796 |
| 0.011 | 4.0 | 24512 | 0.1027 | 0.9830 | 0.9832 | 0.9831 | 0.9799 |
pytorch_model.bin 709 MBtraining_args.bin 3.1 kB