Downloads · 30 days
24
2% of all-time downloads
alexd063/bio-clinicalbert-finetuned-medmentions-v1
bio-clinicalbert-finetuned-medmentions-v1 is a token classification model from alexd063. Use it when you need labels on individual words, such as names. It is set up for transformers. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
Downloads · 30 days
24
2% of all-time downloads
All-time downloads
1K
Public
Parameters
108M
20.3 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors431 MB · 100%
From the Hugging Face model README
This model is a fine-tuned version of emilyalsentzer/Bio_ClinicalBERT on the Ben10x/MedMentions-MTI881-NER dataset. 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 |
|---|---|---|---|---|---|---|---|
| 2.0211 | 1.0 | 1463 | 1.2990 | 0.181 | 0.441 | 0.256 | 0.609 |
| 1.0855 | 2.0 | 2926 | 1.0143 | 0.23 | 0.519 | 0.318 | 0.656 |
| 0.7216 | 3.0 | 4389 | 1.0351 | 0.256 | 0.537 | 0.346 | 0.668 |
| 0.4696 | 4.0 | 5852 | 0.9641 | 0.292 | 0.601 | 0.393 | 0.698 |
| 0.3200 | 5.0 | 7315 | 1.0366 | 0.342 | 0.639 | 0.445 | 0.746 |
| 0.2432 | 6.0 | 8778 | 1.1099 | 0.356 | 0.678 | 0.467 | 0.76 |
| 0.1669 | 7.0 | 10241 | 1.2021 | 0.379 | 0.681 | 0.487 | 0.774 |
| 0.1375 | 8.0 | 11704 | 1.3615 | 0.422 | 0.69 | 0.524 | 0.804 |
| 0.1036 | 9.0 | 13167 | 1.4416 | 0.439 | 0.686 | 0.535 | 0.812 |
| 0.0831 | 10.0 | 14630 | 1.5576 | 0.451 | 0.706 | 0.55 | 0.82 |
| 0.0604 | 11.0 | 16093 | 1.7574 | 0.492 | 0.718 | 0.584 | 0.842 |
| 0.0438 | 12.0 | 17556 | 1.7740 | 0.504 | 0.712 | 0.59 | 0.845 |
| 0.0334 | 13.0 | 19019 | 1.9544 | 0.521 | 0.713 | 0.602 | 0.851 |
| 0.0252 | 14.0 | 20482 | 2.1627 | 0.541 | 0.719 | 0.617 | 0.859 |
| 0.0214 | 15.0 | 21945 | 2.2473 | 0.56 | 0.72 | 0.63 | 0.866 |
| 0.0160 | 16.0 | 23408 | 2.3725 | 0.569 | 0.716 | 0.634 | 0.869 |
| 0.0137 | 17.0 | 24871 | 2.4949 | 0.587 | 0.711 | 0.643 | 0.873 |
| 0.0118 | 18.0 | 26334 | 2.5750 | 0.594 | 0.709 | 0.646 | 0.875 |
| 0.0109 | 19.0 | 27797 | 2.5828 | 0.596 | 0.712 | 0.649 | 0.876 |
| 0.0099 | 20.0 | 29260 | 2.5898 | 0.595 | 0.711 | 0.648 | 0.876 |