T5-small_finetuned_billsum_subset_model_bs32_lr2e-05
This model is a fine-tuned version of t5-small on the billsum dataset. It achieves the following results on the evaluation set:
- Loss: 1.9763
- Rouge1: 0.1887
- Rouge2: 0.0967
- Rougel: 0.1659
- Rougelsum: 0.1657
- Gen Len: 19.0
- Bleu: 0.0008
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: 32
- eval_batch_size: 32
- 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 | Bleu |
|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 31 | 1.9837 | 0.1873 | 0.0945 | 0.1635 | 0.1633 | 19.0 | 0.0007 |
| No log | 2.0 | 62 | 1.9812 | 0.1884 | 0.0955 | 0.1652 | 0.1648 | 19.0 | 0.0007 |
| No log | 3.0 | 93 | 1.9785 | 0.1866 | 0.0936 | 0.1636 | 0.1634 | 19.0 | 0.0007 |
| No log | 4.0 | 124 | 1.9763 | 0.1887 | 0.0967 | 0.1659 | 0.1657 | 19.0 | 0.0008 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3