|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: ner-2 |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# ner-2 |
|
|
|
This model is a fine-tuned version of [PlanTL-GOB-ES/bsc-bio-ehr-es-pharmaconer](https://huggingface.co/PlanTL-GOB-ES/bsc-bio-ehr-es-pharmaconer) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1618 |
|
- Precision: 0.7352 |
|
- Recall: 0.6436 |
|
- F1: 0.6863 |
|
- Accuracy: 0.9712 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| No log | 1.0 | 29 | 0.3028 | 0.0 | 0.0 | 0.0 | 0.9220 | |
|
| No log | 2.0 | 58 | 0.2800 | 0.0 | 0.0 | 0.0 | 0.9220 | |
|
| No log | 3.0 | 87 | 0.2136 | 0.2105 | 0.0277 | 0.0489 | 0.9302 | |
|
| No log | 4.0 | 116 | 0.1803 | 0.375 | 0.0727 | 0.1217 | 0.9391 | |
|
| No log | 5.0 | 145 | 0.1737 | 0.4923 | 0.2215 | 0.3055 | 0.9462 | |
|
| No log | 6.0 | 174 | 0.1354 | 0.6124 | 0.3772 | 0.4668 | 0.9584 | |
|
| No log | 7.0 | 203 | 0.1399 | 0.6062 | 0.4048 | 0.4855 | 0.9589 | |
|
| No log | 8.0 | 232 | 0.1444 | 0.6220 | 0.5294 | 0.5720 | 0.9623 | |
|
| No log | 9.0 | 261 | 0.1252 | 0.6439 | 0.6194 | 0.6314 | 0.9662 | |
|
| No log | 10.0 | 290 | 0.1757 | 0.7216 | 0.4394 | 0.5462 | 0.9604 | |
|
| No log | 11.0 | 319 | 0.1352 | 0.6707 | 0.5779 | 0.6208 | 0.9667 | |
|
| No log | 12.0 | 348 | 0.1276 | 0.6797 | 0.6021 | 0.6385 | 0.9677 | |
|
| No log | 13.0 | 377 | 0.1542 | 0.7328 | 0.5882 | 0.6526 | 0.9688 | |
|
| No log | 14.0 | 406 | 0.1418 | 0.7192 | 0.6471 | 0.6812 | 0.9712 | |
|
| No log | 15.0 | 435 | 0.1678 | 0.7162 | 0.5502 | 0.6223 | 0.9672 | |
|
| No log | 16.0 | 464 | 0.1559 | 0.7075 | 0.6194 | 0.6605 | 0.9689 | |
|
| No log | 17.0 | 493 | 0.1446 | 0.6568 | 0.6886 | 0.6723 | 0.9681 | |
|
| 0.079 | 18.0 | 522 | 0.1582 | 0.7348 | 0.5848 | 0.6513 | 0.9693 | |
|
| 0.079 | 19.0 | 551 | 0.1519 | 0.6977 | 0.6228 | 0.6581 | 0.9705 | |
|
| 0.079 | 20.0 | 580 | 0.1503 | 0.7251 | 0.6298 | 0.6741 | 0.9703 | |
|
| 0.079 | 21.0 | 609 | 0.1585 | 0.6834 | 0.6125 | 0.6460 | 0.9703 | |
|
| 0.079 | 22.0 | 638 | 0.1594 | 0.7126 | 0.6263 | 0.6667 | 0.9705 | |
|
| 0.079 | 23.0 | 667 | 0.1558 | 0.7008 | 0.6401 | 0.6691 | 0.9703 | |
|
| 0.079 | 24.0 | 696 | 0.1570 | 0.7273 | 0.6367 | 0.6790 | 0.9708 | |
|
| 0.079 | 25.0 | 725 | 0.1553 | 0.7022 | 0.6609 | 0.6809 | 0.9705 | |
|
| 0.079 | 26.0 | 754 | 0.1592 | 0.7148 | 0.6332 | 0.6716 | 0.9701 | |
|
| 0.079 | 27.0 | 783 | 0.1579 | 0.7170 | 0.6574 | 0.6859 | 0.9710 | |
|
| 0.079 | 28.0 | 812 | 0.1597 | 0.7148 | 0.6505 | 0.6812 | 0.9708 | |
|
| 0.079 | 29.0 | 841 | 0.1625 | 0.7309 | 0.6298 | 0.6766 | 0.9705 | |
|
| 0.079 | 30.0 | 870 | 0.1618 | 0.7352 | 0.6436 | 0.6863 | 0.9712 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.28.1 |
|
- Pytorch 2.0.0+cu118 |
|
- Datasets 2.12.0 |
|
- Tokenizers 0.13.3 |
|
|