|
--- |
|
license: apache-2.0 |
|
base_model: facebook/bart-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: bart-base-summarize-finetuned |
|
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. --> |
|
|
|
# bart-base-summarize-finetuned |
|
|
|
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: 0.3408 |
|
- Rouge1: 79.6622 |
|
- Rouge2: 77.9282 |
|
- Rougel: 79.6654 |
|
- Rougelsum: 79.6384 |
|
- Gen Len: 7.8821 |
|
|
|
## 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: 64 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 20 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:| |
|
| No log | 1.0 | 62 | 0.3856 | 67.6564 | 65.4045 | 67.6202 | 67.6206 | 6.6825 | |
|
| No log | 2.0 | 124 | 0.3529 | 70.23 | 68.4349 | 70.2289 | 70.1265 | 6.5756 | |
|
| No log | 3.0 | 186 | 0.3303 | 75.4875 | 73.3149 | 75.3918 | 75.3835 | 7.9808 | |
|
| No log | 4.0 | 248 | 0.3165 | 76.17 | 74.0354 | 76.2341 | 76.1363 | 7.4435 | |
|
| No log | 5.0 | 310 | 0.3094 | 76.9425 | 75.0561 | 76.9582 | 76.8794 | 7.9567 | |
|
| No log | 6.0 | 372 | 0.3130 | 78.1808 | 76.2533 | 78.1846 | 78.1377 | 7.9062 | |
|
| No log | 7.0 | 434 | 0.3081 | 78.5859 | 76.7258 | 78.6782 | 78.5825 | 7.6946 | |
|
| No log | 8.0 | 496 | 0.3195 | 78.8452 | 76.85 | 78.8076 | 78.7562 | 8.1663 | |
|
| 0.3758 | 9.0 | 558 | 0.3103 | 78.9204 | 77.2131 | 78.9671 | 78.9562 | 8.1341 | |
|
| 0.3758 | 10.0 | 620 | 0.3091 | 78.7793 | 76.8877 | 78.7503 | 78.7031 | 7.7319 | |
|
| 0.3758 | 11.0 | 682 | 0.3173 | 79.1693 | 77.4324 | 79.2141 | 79.1671 | 7.8881 | |
|
| 0.3758 | 12.0 | 744 | 0.3192 | 79.3653 | 77.6962 | 79.4379 | 79.3547 | 7.7339 | |
|
| 0.3758 | 13.0 | 806 | 0.3246 | 79.041 | 77.1587 | 79.1201 | 79.0828 | 7.8438 | |
|
| 0.3758 | 14.0 | 868 | 0.3312 | 79.4605 | 77.7629 | 79.5227 | 79.4425 | 7.8014 | |
|
| 0.3758 | 15.0 | 930 | 0.3300 | 79.7724 | 78.167 | 79.8187 | 79.799 | 7.8609 | |
|
| 0.3758 | 16.0 | 992 | 0.3409 | 79.4618 | 77.694 | 79.4758 | 79.4325 | 7.8296 | |
|
| 0.14 | 17.0 | 1054 | 0.3436 | 79.1169 | 77.3095 | 79.1082 | 79.092 | 8.0302 | |
|
| 0.14 | 18.0 | 1116 | 0.3440 | 78.9896 | 77.2319 | 78.984 | 78.9472 | 7.9325 | |
|
| 0.14 | 19.0 | 1178 | 0.3399 | 79.531 | 77.8083 | 79.5489 | 79.5005 | 7.871 | |
|
| 0.14 | 20.0 | 1240 | 0.3408 | 79.6622 | 77.9282 | 79.6654 | 79.6384 | 7.8821 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.41.1 |
|
- Pytorch 1.13.1+cu117 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
