|
--- |
|
license: apache-2.0 |
|
base_model: facebook/bart-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: fine-tune-bart |
|
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. --> |
|
|
|
# fine-tune-bart |
|
|
|
This model is a fine-tuned version of [facebook/bart-base](https://huggingface.co/facebook/bart-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8951 |
|
- Rouge1: 0.3436 |
|
- Rouge2: 0.1406 |
|
- Rougel: 0.3117 |
|
- Rougelsum: 0.3108 |
|
- Gen Len: 15.43 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 4 |
|
- 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 | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:| |
|
| No log | 1.0 | 301 | 0.7910 | 0.2441 | 0.0841 | 0.2149 | 0.2154 | 14.55 | |
|
| 1.8181 | 2.0 | 602 | 0.7323 | 0.256 | 0.0926 | 0.2294 | 0.2291 | 13.25 | |
|
| 1.8181 | 3.0 | 903 | 0.7217 | 0.2794 | 0.1079 | 0.2491 | 0.2465 | 14.48 | |
|
| 0.6902 | 4.0 | 1204 | 0.7233 | 0.3095 | 0.1209 | 0.2782 | 0.277 | 14.38 | |
|
| 0.5826 | 5.0 | 1505 | 0.7241 | 0.2985 | 0.1239 | 0.2628 | 0.2633 | 14.68 | |
|
| 0.5826 | 6.0 | 1806 | 0.7184 | 0.3312 | 0.1309 | 0.2968 | 0.2978 | 15.53 | |
|
| 0.4967 | 7.0 | 2107 | 0.7332 | 0.3127 | 0.1324 | 0.2856 | 0.2857 | 14.86 | |
|
| 0.4967 | 8.0 | 2408 | 0.7419 | 0.3379 | 0.1391 | 0.3027 | 0.3035 | 14.7 | |
|
| 0.429 | 9.0 | 2709 | 0.7580 | 0.3473 | 0.1417 | 0.318 | 0.3178 | 14.65 | |
|
| 0.3799 | 10.0 | 3010 | 0.7505 | 0.338 | 0.1406 | 0.3057 | 0.3033 | 15.18 | |
|
| 0.3799 | 11.0 | 3311 | 0.7783 | 0.3444 | 0.1341 | 0.3139 | 0.3126 | 15.12 | |
|
| 0.341 | 12.0 | 3612 | 0.7893 | 0.3231 | 0.1294 | 0.2991 | 0.2993 | 14.97 | |
|
| 0.341 | 13.0 | 3913 | 0.7957 | 0.347 | 0.1376 | 0.3105 | 0.3101 | 15.3 | |
|
| 0.299 | 14.0 | 4214 | 0.8134 | 0.3275 | 0.1367 | 0.3023 | 0.3012 | 14.84 | |
|
| 0.263 | 15.0 | 4515 | 0.8191 | 0.3125 | 0.1364 | 0.2873 | 0.2875 | 15.17 | |
|
| 0.263 | 16.0 | 4816 | 0.8196 | 0.3276 | 0.1334 | 0.3011 | 0.2996 | 15.32 | |
|
| 0.2394 | 17.0 | 5117 | 0.8389 | 0.3168 | 0.1244 | 0.2856 | 0.2881 | 15.07 | |
|
| 0.2394 | 18.0 | 5418 | 0.8502 | 0.3398 | 0.1328 | 0.3123 | 0.3112 | 15.06 | |
|
| 0.2157 | 19.0 | 5719 | 0.8584 | 0.3257 | 0.1197 | 0.2937 | 0.2936 | 15.36 | |
|
| 0.1957 | 20.0 | 6020 | 0.8633 | 0.3325 | 0.1295 | 0.2986 | 0.2994 | 15.4 | |
|
| 0.1957 | 21.0 | 6321 | 0.8620 | 0.3254 | 0.1208 | 0.2952 | 0.2949 | 15.28 | |
|
| 0.181 | 22.0 | 6622 | 0.8762 | 0.3395 | 0.1306 | 0.3054 | 0.3045 | 15.27 | |
|
| 0.181 | 23.0 | 6923 | 0.8775 | 0.3419 | 0.14 | 0.3137 | 0.3126 | 15.24 | |
|
| 0.1622 | 24.0 | 7224 | 0.8780 | 0.3397 | 0.1311 | 0.3069 | 0.3063 | 15.15 | |
|
| 0.1613 | 25.0 | 7525 | 0.8859 | 0.3231 | 0.1225 | 0.2887 | 0.288 | 15.14 | |
|
| 0.1613 | 26.0 | 7826 | 0.8905 | 0.3289 | 0.1284 | 0.2953 | 0.2941 | 15.23 | |
|
| 0.1463 | 27.0 | 8127 | 0.8883 | 0.3358 | 0.1303 | 0.3002 | 0.2988 | 15.19 | |
|
| 0.1463 | 28.0 | 8428 | 0.8933 | 0.3414 | 0.139 | 0.3113 | 0.3098 | 15.5 | |
|
| 0.1444 | 29.0 | 8729 | 0.8949 | 0.3449 | 0.1369 | 0.311 | 0.31 | 15.43 | |
|
| 0.135 | 30.0 | 9030 | 0.8951 | 0.3436 | 0.1406 | 0.3117 | 0.3108 | 15.43 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
