|
--- |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- rouge |
|
model-index: |
|
- name: pegasus-legalease |
|
results: [] |
|
datasets: |
|
- hheiden/us-congress-117-bills |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# pegasus-legalease |
|
|
|
This model was trained from scratch on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 4.2373 |
|
- Rouge1: 0.4847 |
|
- Rouge2: 0.3225 |
|
- Rougel: 0.4194 |
|
- Rougelsum: 0.4177 |
|
- Gen Len: 43.02 |
|
|
|
## 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: 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: 5 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:| |
|
| 5.2738 | 1.0 | 125 | 4.6004 | 0.4363 | 0.2769 | 0.375 | 0.3743 | 37.24 | |
|
| 4.8164 | 2.0 | 250 | 4.4350 | 0.464 | 0.3085 | 0.405 | 0.4038 | 40.3 | |
|
| 4.8494 | 3.0 | 375 | 4.3372 | 0.473 | 0.3153 | 0.412 | 0.41 | 41.2 | |
|
| 4.6062 | 4.0 | 500 | 4.2669 | 0.4791 | 0.3196 | 0.4159 | 0.4141 | 43.03 | |
|
| 4.5682 | 5.0 | 625 | 4.2373 | 0.4847 | 0.3225 | 0.4194 | 0.4177 | 43.02 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.1 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
