results
This model is a fine-tuned version of sshleifer/distilbart-xsum-12-3 on the News-summary-Kaggle dataset. It achieves the following results on the evaluation set:
- Loss: 3.1426
- Rouge1: 51.2701
- Rouge2: 28.3575
- Rougel: 37.9263
- Rougelsum: 45.8934
- Gen Len: 75.777
Model description
This model use pre-trained model: sshleifer/distilbart-xsum-12-3 fined tuned on the datasets: News-summary-Kaggle. Our aims is to build model can summerize news efficiently.
Intended uses & limitations
More information needed
Training and evaluation data
News-summary. Link: https://www.kaggle.com/datasets/sunnysai12345/news-summary
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 5
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 3.4812 | 1.0 | 425 | 3.3209 | 47.7226 | 26.3282 | 35.5063 | 42.5426 | 66.523 |
| 3.2269 | 2.0 | 850 | 3.1838 | 50.4271 | 27.7047 | 37.2638 | 45.1897 | 77.115 |
| 2.9504 | 3.0 | 1275 | 3.1401 | 50.6362 | 28.2773 | 37.6 | 45.4901 | 74.992 |
| 2.8014 | 4.0 | 1700 | 3.1346 | 51.2942 | 28.4684 | 38.0877 | 46.0386 | 74.299 |
| 2.71 | 5.0 | 2125 | 3.1426 | 51.2701 | 28.3575 | 37.9263 | 45.8934 | 75.777 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.2
- Datasets 2.1.0
- Tokenizers 0.15.2
- Downloads last month
- 104
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for LA1512/fine-tuned-distilbart-xsum-12-3-news-summary
Base model
sshleifer/distilbart-xsum-12-3