legal-camembert
This model is a fine-tuned version of camembert/camembert-large on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4965
- Accuracy: 0.8778
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: 64
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6857 | 2.93 | 1000 | 0.6172 | 0.8513 |
| 0.618 | 5.87 | 2000 | 0.5735 | 0.8610 |
| 0.5831 | 8.8 | 3000 | 0.5496 | 0.8666 |
| 0.5558 | 11.73 | 4000 | 0.5279 | 0.8692 |
| 0.5436 | 14.66 | 5000 | 0.5230 | 0.8707 |
| 0.5321 | 17.6 | 6000 | 0.5024 | 0.8743 |
| 0.5182 | 20.53 | 7000 | 0.5089 | 0.8740 |
| 0.5069 | 23.46 | 8000 | 0.4983 | 0.8776 |
| 0.5077 | 26.39 | 9000 | 0.4914 | 0.8775 |
| 0.4965 | 29.33 | 10000 | 0.4965 | 0.8778 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1
- Datasets 2.13.1
- Tokenizers 0.13.3