T5-small_finetuned_billsum_subset_model_bs8_lr5e-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: 2.1527
- Rouge1: 0.1976
- Rouge2: 0.0998
- Rougel: 0.1691
- Rougelsum: 0.1689
- Gen Len: 19.0
- Bleu: 0.0009
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- 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 | 124 | 2.2466 | 0.1991 | 0.0966 | 0.1693 | 0.169 | 19.0 | 0.001 |
| No log | 2.0 | 248 | 2.1993 | 0.1944 | 0.0946 | 0.1654 | 0.1652 | 19.0 | 0.0009 |
| No log | 3.0 | 372 | 2.1687 | 0.1944 | 0.0961 | 0.1675 | 0.1674 | 19.0 | 0.0009 |
| No log | 4.0 | 496 | 2.1527 | 0.1976 | 0.0998 | 0.1691 | 0.1689 | 19.0 | 0.0009 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3