AraBERT_token_classification_AraEval24_18_labels_mlm1k_augmented
This model is a fine-tuned version of aubmindlab/bert-base-arabert on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9374
- Precision: 0.0475
- Recall: 0.0165
- F1: 0.0245
- Accuracy: 0.8620
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.5534 | 1.0 | 7396 | 0.7643 | 0.0031 | 0.0002 | 0.0003 | 0.8712 |
0.4414 | 2.0 | 14792 | 0.7713 | 0.0159 | 0.0018 | 0.0032 | 0.8638 |
0.3961 | 3.0 | 22188 | 0.7715 | 0.0137 | 0.0014 | 0.0026 | 0.8684 |
0.3484 | 4.0 | 29584 | 0.7929 | 0.0421 | 0.0065 | 0.0113 | 0.8661 |
0.3131 | 5.0 | 36980 | 0.8180 | 0.04 | 0.0107 | 0.0169 | 0.8578 |
0.2899 | 6.0 | 44376 | 0.8650 | 0.0448 | 0.0098 | 0.0161 | 0.8625 |
0.2682 | 7.0 | 51772 | 0.8725 | 0.0556 | 0.0186 | 0.0279 | 0.8551 |
0.2433 | 8.0 | 59168 | 0.8841 | 0.0521 | 0.0146 | 0.0228 | 0.8603 |
0.2384 | 9.0 | 66564 | 0.9149 | 0.0502 | 0.0155 | 0.0237 | 0.8635 |
0.2094 | 10.0 | 73960 | 0.9374 | 0.0475 | 0.0165 | 0.0245 | 0.8620 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.12.1
- Datasets 2.13.2
- Tokenizers 0.13.3
- Downloads last month
- 1