my_awesome_billsum_model
This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.2277
- Rouge1: 0.1996
- Rouge2: 0.1042
- Rougel: 0.1706
- Rougelsum: 0.1704
- 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: 1
- eval_batch_size: 1
- 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 |
|---|---|---|---|---|---|---|---|---|
| 3.0704 | 1.0 | 989 | 2.3509 | 0.2003 | 0.1004 | 0.1693 | 0.1693 | 19.0 |
| 2.5236 | 2.0 | 1978 | 2.2727 | 0.1985 | 0.1032 | 0.169 | 0.1691 | 19.0 |
| 2.4492 | 3.0 | 2967 | 2.2363 | 0.1995 | 0.1025 | 0.1692 | 0.1692 | 19.0 |
| 2.394 | 4.0 | 3956 | 2.2277 | 0.1996 | 0.1042 | 0.1706 | 0.1704 | 19.0 |
Framework versions
- Transformers 4.43.3
- Pytorch 2.4.0+cu118
- Datasets 2.17.1
- Tokenizers 0.19.1