|
--- |
|
tags: |
|
- simplification |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: NASES-clara-med |
|
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. --> |
|
|
|
# NASES-clara-med |
|
|
|
This model is a fine-tuned version of [ELiRF/NASES](https://huggingface.co/ELiRF/NASES) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.2763 |
|
- Rouge1: 42.9986 |
|
- Rouge2: 25.2365 |
|
- Rougel: 37.0782 |
|
- Rougelsum: 37.278 |
|
|
|
## 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: 5.6e-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: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:| |
|
| No log | 1.0 | 190 | 2.1704 | 42.0923 | 24.011 | 36.2317 | 36.3819 | |
|
| No log | 2.0 | 380 | 2.1260 | 42.3364 | 24.6464 | 36.8836 | 37.0023 | |
|
| 1.9093 | 3.0 | 570 | 2.1582 | 43.7464 | 26.0481 | 38.1321 | 38.2575 | |
|
| 1.9093 | 4.0 | 760 | 2.2436 | 43.1348 | 25.6313 | 37.5276 | 37.688 | |
|
| 0.7294 | 5.0 | 950 | 2.3852 | 43.9276 | 26.2853 | 38.1775 | 38.3529 | |
|
| 0.7294 | 6.0 | 1140 | 2.5096 | 42.6241 | 25.1825 | 36.9084 | 37.1236 | |
|
| 0.7294 | 7.0 | 1330 | 2.5986 | 43.4603 | 25.7703 | 37.762 | 38.0026 | |
|
| 0.2438 | 8.0 | 1520 | 2.6878 | 42.483 | 24.6796 | 36.7012 | 36.9424 | |
|
| 0.2438 | 9.0 | 1710 | 2.7096 | 43.3953 | 25.6418 | 37.4906 | 37.8048 | |
|
| 0.1422 | 10.0 | 1900 | 2.7879 | 43.1926 | 25.3773 | 37.2548 | 37.4858 | |
|
| 0.1422 | 11.0 | 2090 | 2.8629 | 43.7788 | 25.7912 | 37.6712 | 37.8664 | |
|
| 0.1422 | 12.0 | 2280 | 2.9139 | 43.5132 | 25.6003 | 37.5426 | 37.7154 | |
|
| 0.0911 | 13.0 | 2470 | 2.9267 | 43.2335 | 25.5807 | 37.4857 | 37.6547 | |
|
| 0.0911 | 14.0 | 2660 | 2.9826 | 42.4726 | 24.6801 | 36.8142 | 36.9149 | |
|
| 0.0704 | 15.0 | 2850 | 2.9834 | 42.7464 | 25.0051 | 37.0043 | 37.188 | |
|
| 0.0704 | 16.0 | 3040 | 3.0423 | 42.7331 | 25.1076 | 36.8757 | 37.1165 | |
|
| 0.0704 | 17.0 | 3230 | 3.0602 | 43.5046 | 25.9845 | 37.9281 | 38.0868 | |
|
| 0.0529 | 18.0 | 3420 | 3.0882 | 42.7186 | 25.0104 | 36.943 | 37.1559 | |
|
| 0.0529 | 19.0 | 3610 | 3.0713 | 43.0051 | 25.3356 | 37.0809 | 37.2836 | |
|
| 0.0383 | 20.0 | 3800 | 3.1547 | 43.2239 | 25.3545 | 37.2722 | 37.4304 | |
|
| 0.0383 | 21.0 | 3990 | 3.1408 | 43.2171 | 25.266 | 37.1733 | 37.4219 | |
|
| 0.0383 | 22.0 | 4180 | 3.1739 | 43.1094 | 25.2674 | 37.3491 | 37.5596 | |
|
| 0.0252 | 23.0 | 4370 | 3.2036 | 43.0451 | 25.3833 | 37.2896 | 37.469 | |
|
| 0.0252 | 24.0 | 4560 | 3.2291 | 43.2983 | 25.5308 | 37.6024 | 37.7772 | |
|
| 0.0173 | 25.0 | 4750 | 3.2607 | 43.0005 | 25.0403 | 37.2126 | 37.367 | |
|
| 0.0173 | 26.0 | 4940 | 3.2498 | 42.869 | 24.9531 | 37.0616 | 37.2307 | |
|
| 0.0173 | 27.0 | 5130 | 3.3016 | 43.1913 | 25.1199 | 37.2238 | 37.4256 | |
|
| 0.0135 | 28.0 | 5320 | 3.2813 | 43.1867 | 25.2193 | 37.2014 | 37.4029 | |
|
| 0.0135 | 29.0 | 5510 | 3.2757 | 42.9765 | 25.2217 | 37.0312 | 37.2317 | |
|
| 0.0113 | 30.0 | 5700 | 3.2763 | 42.9986 | 25.2365 | 37.0782 | 37.278 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.25.1 |
|
- Pytorch 1.13.0 |
|
- Datasets 2.8.0 |
|
- Tokenizers 0.12.1 |
|
|