<!-- 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. -->
legal_bert_small_summarized_defined
This model is a fine-tuned version of nlpaueb/legal-bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.8897
- Accuracy: 0.835
- Precision: 0.5
- Recall: 0.1515
- F1: 0.2326
- D-index: 1.5181
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: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1600
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | D-index |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 200 | 0.4467 | 0.835 | 0.0 | 0.0 | 0.0 | 1.4607 |
| No log | 2.0 | 400 | 0.4909 | 0.835 | 0.0 | 0.0 | 0.0 | 1.4607 |
| 0.5409 | 3.0 | 600 | 0.4941 | 0.83 | 0.4545 | 0.1515 | 0.2273 | 1.5113 |
| 0.5409 | 4.0 | 800 | 0.5612 | 0.84 | 0.6 | 0.0909 | 0.1579 | 1.5021 |
| 0.4849 | 5.0 | 1000 | 0.6301 | 0.84 | 0.5714 | 0.1212 | 0.2 | 1.5135 |
| 0.4849 | 6.0 | 1200 | 0.8969 | 0.84 | 0.6 | 0.0909 | 0.1579 | 1.5021 |
| 0.4849 | 7.0 | 1400 | 1.3171 | 0.82 | 0.3636 | 0.1212 | 0.1818 | 1.4865 |
| 0.2104 | 8.0 | 1600 | 1.6653 | 0.775 | 0.2692 | 0.2121 | 0.2373 | 1.4593 |
| 0.2104 | 9.0 | 1800 | 1.7041 | 0.795 | 0.3182 | 0.2121 | 0.2545 | 1.4866 |
| 0.0314 | 10.0 | 2000 | 1.7495 | 0.815 | 0.3571 | 0.1515 | 0.2128 | 1.4911 |
| 0.0314 | 11.0 | 2200 | 1.7627 | 0.815 | 0.3571 | 0.1515 | 0.2128 | 1.4911 |
| 0.0314 | 12.0 | 2400 | 1.7892 | 0.825 | 0.375 | 0.0909 | 0.1463 | 1.4819 |
| 0.0067 | 13.0 | 2600 | 1.8211 | 0.83 | 0.4444 | 0.1212 | 0.1905 | 1.5000 |
| 0.0067 | 14.0 | 2800 | 1.8567 | 0.83 | 0.4444 | 0.1212 | 0.1905 | 1.5000 |
| 0.0 | 15.0 | 3000 | 1.8817 | 0.83 | 0.4444 | 0.1212 | 0.1905 | 1.5000 |
| 0.0 | 16.0 | 3200 | 1.8590 | 0.825 | 0.4167 | 0.1515 | 0.2222 | 1.5046 |
| 0.0 | 17.0 | 3400 | 1.8619 | 0.835 | 0.5 | 0.1515 | 0.2326 | 1.5181 |
| 0.0014 | 18.0 | 3600 | 1.8744 | 0.835 | 0.5 | 0.1515 | 0.2326 | 1.5181 |
| 0.0014 | 19.0 | 3800 | 1.8849 | 0.835 | 0.5 | 0.1515 | 0.2326 | 1.5181 |
| 0.0 | 20.0 | 4000 | 1.8897 | 0.835 | 0.5 | 0.1515 | 0.2326 | 1.5181 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
