<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
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: 2.5420
- Rouge1: 0.1438
- Rouge2: 0.0513
- Rougel: 0.1175
- Rougelsum: 0.1174
- 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 | 62 | 2.8357 | 0.1287 | 0.0392 | 0.1069 | 0.1069 | 19.0 |
No log | 2.0 | 124 | 2.6209 | 0.1361 | 0.0479 | 0.1127 | 0.1127 | 19.0 |
No log | 3.0 | 186 | 2.5600 | 0.1412 | 0.052 | 0.1171 | 0.117 | 19.0 |
No log | 4.0 | 248 | 2.5420 | 0.1438 | 0.0513 | 0.1175 | 0.1174 | 19.0 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3