|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: Yepes_0.0001_0404_ES6 |
|
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. --> |
|
|
|
# Yepes_0.0001_0404_ES6 |
|
|
|
This model is a fine-tuned version of [microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext](https://huggingface.co/microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1207 |
|
- Precision: 0.4902 |
|
- Recall: 0.3743 |
|
- F1: 0.4244 |
|
- Accuracy: 0.9769 |
|
|
|
## 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: 0.0001 |
|
- 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 |
|
- training_steps: 2000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.4338 | 0.43 | 25 | 0.1979 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.2051 | 0.86 | 50 | 0.1923 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1601 | 1.29 | 75 | 0.1618 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1742 | 1.72 | 100 | 0.1400 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1506 | 2.16 | 125 | 0.1462 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1507 | 2.59 | 150 | 0.1516 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1566 | 3.02 | 175 | 0.1382 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1467 | 3.45 | 200 | 0.1360 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1492 | 3.88 | 225 | 0.1400 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1543 | 4.31 | 250 | 0.1364 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1435 | 4.74 | 275 | 0.1384 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1369 | 5.17 | 300 | 0.1282 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1284 | 5.6 | 325 | 0.1337 | 0.2381 | 0.1198 | 0.1594 | 0.9704 | |
|
| 0.1235 | 6.03 | 350 | 0.1215 | 0.0 | 0.0 | 0.0 | 0.9705 | |
|
| 0.1165 | 6.47 | 375 | 0.1337 | 0.3613 | 0.1677 | 0.2290 | 0.9739 | |
|
| 0.1184 | 6.9 | 400 | 0.1228 | 0.2303 | 0.1228 | 0.1602 | 0.9718 | |
|
| 0.1076 | 7.33 | 425 | 0.1174 | 0.2646 | 0.3263 | 0.2922 | 0.9671 | |
|
| 0.0964 | 7.76 | 450 | 0.1094 | 0.3972 | 0.2545 | 0.3102 | 0.9751 | |
|
| 0.0902 | 8.19 | 475 | 0.1217 | 0.4264 | 0.2515 | 0.3164 | 0.9742 | |
|
| 0.0891 | 8.62 | 500 | 0.1075 | 0.3746 | 0.3263 | 0.3488 | 0.9736 | |
|
| 0.0813 | 9.05 | 525 | 0.1295 | 0.4354 | 0.2725 | 0.3352 | 0.9738 | |
|
| 0.078 | 9.48 | 550 | 0.1067 | 0.375 | 0.3413 | 0.3574 | 0.9742 | |
|
| 0.0751 | 9.91 | 575 | 0.1042 | 0.4905 | 0.3084 | 0.3787 | 0.9765 | |
|
| 0.0683 | 10.34 | 600 | 0.1028 | 0.4672 | 0.3413 | 0.3945 | 0.9761 | |
|
| 0.0687 | 10.78 | 625 | 0.1070 | 0.4975 | 0.2994 | 0.3738 | 0.9762 | |
|
| 0.0664 | 11.21 | 650 | 0.1225 | 0.3256 | 0.3383 | 0.3319 | 0.9703 | |
|
| 0.0565 | 11.64 | 675 | 0.1000 | 0.4487 | 0.3144 | 0.3697 | 0.9767 | |
|
| 0.0555 | 12.07 | 700 | 0.1033 | 0.4463 | 0.3234 | 0.375 | 0.9757 | |
|
| 0.045 | 12.5 | 725 | 0.1150 | 0.4237 | 0.3323 | 0.3725 | 0.9746 | |
|
| 0.0514 | 12.93 | 750 | 0.1126 | 0.6 | 0.3503 | 0.4423 | 0.9774 | |
|
| 0.0387 | 13.36 | 775 | 0.1409 | 0.3986 | 0.3473 | 0.3712 | 0.9742 | |
|
| 0.0419 | 13.79 | 800 | 0.1096 | 0.4336 | 0.4401 | 0.4368 | 0.9723 | |
|
| 0.0349 | 14.22 | 825 | 0.1207 | 0.4902 | 0.3743 | 0.4244 | 0.9769 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.27.4 |
|
- Pytorch 2.0.0+cu118 |
|
- Datasets 2.11.0 |
|
- Tokenizers 0.13.2 |
|
|
