pegasus-legalease / README.md
etav22's picture
training completed[prod]: pegasus-custom-legaltrain
1189320 verified
---
license: mit
base_model: nsi319/legal-pegasus
tags:
- generated_from_trainer
model-index:
- name: pegasus-legalease
results: []
---
<!-- 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 is a fine-tuned version of [nsi319/legal-pegasus](https://huggingface.co/nsi319/legal-pegasus) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0662
## 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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| No log | 0.09 | 250 | 4.5867 |
| 4.446 | 0.18 | 500 | 1.5129 |
| 4.446 | 0.27 | 750 | 1.1850 |
| 1.4139 | 0.35 | 1000 | 1.1565 |
| 1.4139 | 0.44 | 1250 | 1.1347 |
| 1.2342 | 0.53 | 1500 | 1.1204 |
| 1.2342 | 0.62 | 1750 | 1.1115 |
| 1.168 | 0.71 | 2000 | 1.1046 |
| 1.168 | 0.8 | 2250 | 1.0977 |
| 1.1573 | 0.89 | 2500 | 1.0925 |
| 1.1573 | 0.98 | 2750 | 1.0871 |
| 1.177 | 1.06 | 3000 | 1.0816 |
| 1.177 | 1.15 | 3250 | 1.0797 |
| 1.1348 | 1.24 | 3500 | 1.0753 |
| 1.1348 | 1.33 | 3750 | 1.0738 |
| 1.1144 | 1.42 | 4000 | 1.0711 |
| 1.1144 | 1.51 | 4250 | 1.0686 |
| 1.1321 | 1.6 | 4500 | 1.0678 |
| 1.1321 | 1.68 | 4750 | 1.0662 |
### Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2