<!-- 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_gen_1
This model is a fine-tuned version of wiorz/bert_legal_test_sm on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.4297
- Accuracy: 0.7992
- Precision: 0.4576
- Recall: 0.2687
- F1: 0.3386
- D-index: 1.5225
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 | 0.99 | 65 | 0.4457 | 0.8002 | 0.4554 | 0.2289 | 0.3046 | 1.5100 |
| No log | 1.99 | 131 | 0.4289 | 0.7973 | 0.4444 | 0.2388 | 0.3107 | 1.5095 |
| No log | 3.0 | 197 | 0.5157 | 0.7555 | 0.4034 | 0.5821 | 0.4766 | 1.5683 |
| No log | 4.0 | 263 | 0.6436 | 0.7983 | 0.4433 | 0.2139 | 0.2886 | 1.5022 |
| No log | 4.99 | 328 | 0.6772 | 0.8021 | 0.4598 | 0.1990 | 0.2778 | 1.5021 |
| No log | 5.99 | 394 | 0.7292 | 0.8078 | 0.4964 | 0.3383 | 0.4024 | 1.5578 |
| No log | 7.0 | 460 | 0.9566 | 0.8021 | 0.4755 | 0.3383 | 0.3953 | 1.5501 |
| 0.2346 | 8.0 | 526 | 1.0280 | 0.8002 | 0.4651 | 0.2985 | 0.3636 | 1.5340 |
| 0.2346 | 8.99 | 591 | 1.0350 | 0.7840 | 0.4330 | 0.4179 | 0.4253 | 1.5526 |
| 0.2346 | 9.99 | 657 | 1.2664 | 0.8002 | 0.4444 | 0.1791 | 0.2553 | 1.4925 |
| 0.2346 | 11.0 | 723 | 1.2846 | 0.7812 | 0.4040 | 0.3035 | 0.3466 | 1.5098 |
| 0.2346 | 12.0 | 789 | 1.2157 | 0.7897 | 0.4351 | 0.3333 | 0.3775 | 1.5317 |
| 0.2346 | 12.99 | 854 | 1.3208 | 0.8030 | 0.4688 | 0.2239 | 0.3030 | 1.5121 |
| 0.2346 | 13.99 | 920 | 1.3100 | 0.7783 | 0.4101 | 0.3632 | 0.3852 | 1.5263 |
| 0.2346 | 15.0 | 986 | 1.2587 | 0.8154 | 0.5347 | 0.2687 | 0.3576 | 1.5444 |
| 0.0277 | 16.0 | 1052 | 1.3552 | 0.7878 | 0.4304 | 0.3383 | 0.3788 | 1.5308 |
| 0.0277 | 16.99 | 1117 | 1.3783 | 0.8059 | 0.4872 | 0.2836 | 0.3585 | 1.5366 |
| 0.0277 | 17.99 | 1183 | 1.4071 | 0.7907 | 0.4336 | 0.3085 | 0.3605 | 1.5245 |
| 0.0277 | 19.0 | 1249 | 1.4283 | 0.8011 | 0.4655 | 0.2687 | 0.3407 | 1.5251 |
| 0.0277 | 19.77 | 1300 | 1.4297 | 0.7992 | 0.4576 | 0.2687 | 0.3386 | 1.5225 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
