my_awesome_billsum_model
This model is a fine-tuned version of t5-small on the billsum dataset. It achieves the following results on the evaluation set:
- Loss: 2.4802
- Rouge1: 0.1434
- Rouge2: 0.0522
- Rougel: 0.118
- Rougelsum: 0.118
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 62 | 2.7812 | 0.13 | 0.037 | 0.1067 | 0.1068 | 19.0 |
| No log | 2.0 | 124 | 2.5608 | 0.1355 | 0.0449 | 0.1132 | 0.1131 | 19.0 |
| No log | 3.0 | 186 | 2.4977 | 0.1417 | 0.0515 | 0.118 | 0.1179 | 19.0 |
| No log | 4.0 | 248 | 2.4802 | 0.1434 | 0.0522 | 0.118 | 0.118 | 19.0 |
Framework versions
- Transformers 4.30.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3