<!-- 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. -->
Alglegal_text_classifier_10_class
This model is a fine-tuned version of hatemestinbejaia/AlgLegal3_bert-base-arabertv2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3054
- F1: 0.9329
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| No log | 1.0 | 84 | 0.5610 | 0.8658 |
| No log | 2.0 | 168 | 0.3273 | 0.9262 |
| No log | 3.0 | 252 | 0.3204 | 0.9128 |
| No log | 4.0 | 336 | 0.2442 | 0.9329 |
| No log | 5.0 | 420 | 0.2321 | 0.9463 |
| 0.3554 | 6.0 | 504 | 0.3061 | 0.9128 |
| 0.3554 | 7.0 | 588 | 0.2701 | 0.9396 |
| 0.3554 | 8.0 | 672 | 0.2575 | 0.9329 |
| 0.3554 | 9.0 | 756 | 0.2605 | 0.9329 |
| 0.3554 | 10.0 | 840 | 0.2894 | 0.9396 |
| 0.3554 | 11.0 | 924 | 0.2898 | 0.9329 |
| 0.0665 | 12.0 | 1008 | 0.3025 | 0.9329 |
| 0.0665 | 13.0 | 1092 | 0.3115 | 0.9329 |
| 0.0665 | 14.0 | 1176 | 0.2995 | 0.9329 |
| 0.0665 | 15.0 | 1260 | 0.3054 | 0.9329 |
| 0.0665 | 16.0 | 1344 | 0.3054 | 0.9329 |
| 0.0665 | 17.0 | 1428 | 0.3079 | 0.9329 |
| 0.0435 | 18.0 | 1512 | 0.3051 | 0.9329 |
| 0.0435 | 19.0 | 1596 | 0.3038 | 0.9329 |
| 0.0435 | 20.0 | 1680 | 0.3054 | 0.9329 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1
