cnn_news_summary_model_new_version
This model is a fine-tuned version of t5-small on the cnn_dailymail dataset. It achieves the following results on the evaluation set:
- Loss: 1.6112
- Rouge1: 0.2187
- Rouge2: 0.0955
- Rougel: 0.1843
- Rougelsum: 0.1845
- Gen Len: 19.0
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: 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: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 431 | 1.6288 | 0.2186 | 0.0945 | 0.1837 | 0.1838 | 19.0 |
| 1.9164 | 2.0 | 862 | 1.6144 | 0.2195 | 0.0954 | 0.1845 | 0.1846 | 19.0 |
| 1.8282 | 3.0 | 1293 | 1.6112 | 0.2187 | 0.0955 | 0.1843 | 0.1845 | 19.0 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3