bbc_news_summary_model
This model is a fine-tuned version of t5-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3238
- Rouge1: 0.2026
- Rouge2: 0.145
- Rougel: 0.1843
- Rougelsum: 0.1844
- Gen Len: 18.9865
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 223 | 0.4221 | 0.189 | 0.1251 | 0.1664 | 0.1666 | 19.0 |
| No log | 2.0 | 446 | 0.3585 | 0.1931 | 0.1321 | 0.1724 | 0.1727 | 19.0 |
| 0.7174 | 3.0 | 669 | 0.3370 | 0.1983 | 0.1391 | 0.1784 | 0.1787 | 18.9865 |
| 0.7174 | 4.0 | 892 | 0.3269 | 0.1994 | 0.1408 | 0.1803 | 0.1803 | 18.9865 |
| 0.4039 | 5.0 | 1115 | 0.3238 | 0.2026 | 0.145 | 0.1843 | 0.1844 | 18.9865 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3