<!-- 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. -->
bert_legal_test_sm
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6233
- Accuracy: 0.6580
- Precision: 0.6683
- Recall: 0.6274
- F1: 0.6472
- D-index: 1.4589
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: 8
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | D-index |
|---|---|---|---|---|---|---|---|---|
| No log | 0.98 | 26 | 0.6905 | 0.5542 | 0.5418 | 0.7028 | 0.6119 | 1.2724 |
| No log | 2.0 | 53 | 0.6937 | 0.5024 | 0.5012 | 0.9623 | 0.6591 | 1.1745 |
| No log | 2.98 | 79 | 0.6505 | 0.6439 | 0.6894 | 0.5236 | 0.5952 | 1.4342 |
| No log | 4.0 | 106 | 0.6233 | 0.6580 | 0.6683 | 0.6274 | 0.6472 | 1.4589 |
| No log | 4.98 | 132 | 0.6369 | 0.6840 | 0.7053 | 0.6321 | 0.6667 | 1.5037 |
| No log | 6.0 | 159 | 0.9851 | 0.6085 | 0.7614 | 0.3160 | 0.4467 | 1.3714 |
| No log | 6.98 | 185 | 0.8765 | 0.6604 | 0.7537 | 0.4764 | 0.5838 | 1.4630 |
| No log | 8.0 | 212 | 0.9170 | 0.6745 | 0.7102 | 0.5896 | 0.6443 | 1.4875 |
| No log | 8.98 | 238 | 1.1931 | 0.6557 | 0.7324 | 0.4906 | 0.5876 | 1.4548 |
| No log | 10.0 | 265 | 1.0355 | 0.6840 | 0.7216 | 0.5991 | 0.6546 | 1.5037 |
| No log | 10.98 | 291 | 1.1690 | 0.6675 | 0.6878 | 0.6132 | 0.6484 | 1.4753 |
| No log | 12.0 | 318 | 1.1527 | 0.6651 | 0.64 | 0.7547 | 0.6926 | 1.4712 |
| No log | 12.98 | 344 | 1.2299 | 0.6675 | 0.6940 | 0.5991 | 0.6430 | 1.4753 |
| No log | 14.0 | 371 | 1.4807 | 0.6557 | 0.72 | 0.5094 | 0.5967 | 1.4548 |
| No log | 14.98 | 397 | 1.4303 | 0.6887 | 0.7083 | 0.6415 | 0.6733 | 1.5118 |
| No log | 16.0 | 424 | 1.5717 | 0.6792 | 0.6863 | 0.6604 | 0.6731 | 1.4956 |
| No log | 16.98 | 450 | 1.7842 | 0.6509 | 0.6975 | 0.5330 | 0.6043 | 1.4466 |
| No log | 18.0 | 477 | 1.6653 | 0.6698 | 0.6895 | 0.6179 | 0.6517 | 1.4794 |
| 0.2514 | 18.98 | 503 | 1.8285 | 0.6557 | 0.7143 | 0.5189 | 0.6011 | 1.4548 |
| 0.2514 | 19.62 | 520 | 1.8220 | 0.6486 | 0.7006 | 0.5189 | 0.5962 | 1.4425 |
Framework versions
- Transformers 4.27.4
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.2
