<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
electramed-small-BC4CHEMD-ner
This model is a fine-tuned version of giacomomiolo/electramed_small_scivocab on the bc4chemd dataset. It achieves the following results on the evaluation set:
- Loss: 0.0655
- Precision: 0.7716
- Recall: 0.6761
- F1: 0.7207
- Accuracy: 0.9771
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.0882 | 1.0 | 1918 | 0.1058 | 0.6615 | 0.3942 | 0.4940 | 0.9635 |
| 0.0555 | 2.0 | 3836 | 0.0820 | 0.7085 | 0.5133 | 0.5954 | 0.9689 |
| 0.0631 | 3.0 | 5754 | 0.0769 | 0.6892 | 0.5743 | 0.6266 | 0.9699 |
| 0.0907 | 4.0 | 7672 | 0.0682 | 0.7623 | 0.5923 | 0.6666 | 0.9740 |
| 0.0313 | 5.0 | 9590 | 0.0675 | 0.7643 | 0.6223 | 0.6860 | 0.9749 |
| 0.0306 | 6.0 | 11508 | 0.0662 | 0.7654 | 0.6398 | 0.6970 | 0.9754 |
| 0.0292 | 7.0 | 13426 | 0.0656 | 0.7694 | 0.6552 | 0.7077 | 0.9763 |
| 0.1025 | 8.0 | 15344 | 0.0658 | 0.7742 | 0.6687 | 0.7176 | 0.9769 |
| 0.0394 | 9.0 | 17262 | 0.0662 | 0.7741 | 0.6731 | 0.7201 | 0.9770 |
| 0.0378 | 10.0 | 19180 | 0.0655 | 0.7716 | 0.6761 | 0.7207 | 0.9771 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
