t5-small-scientific_lay_summarisation
This model is a fine-tuned version of t5-small on the scientific_lay_summarisation dataset. It achieves the following results on the evaluation set:
- Loss: 3.0503
- Rouge1: 0.0546
- Rouge2: 0.0154
- Rougel: 0.0461
- Rougelsum: 0.0462
- 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: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 272 | 3.1627 | 0.048 | 0.0123 | 0.0402 | 0.0402 | 19.0 |
| 3.6506 | 2.0 | 544 | 3.0881 | 0.0524 | 0.0143 | 0.0441 | 0.0442 | 19.0 |
| 3.6506 | 3.0 | 816 | 3.0586 | 0.0543 | 0.0155 | 0.0461 | 0.0462 | 19.0 |
| 3.2737 | 4.0 | 1088 | 3.0503 | 0.0546 | 0.0154 | 0.0461 | 0.0462 | 19.0 |
Framework versions
- Transformers 4.27.2
- Pytorch 2.1.0+cu121
- Datasets 2.11.0
- Tokenizers 0.13.3