AraBERT_token_classification__AraEval24_augmented_no_trun
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: 0.9821
- Precision: 0.0476
- Recall: 0.0160
- F1: 0.0239
- Accuracy: 0.8462
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 |
---|---|---|---|---|---|---|---|
0.5623 | 1.0 | 5967 | 0.8493 | 0.0426 | 0.0027 | 0.0051 | 0.8574 |
0.4227 | 2.0 | 11934 | 0.8246 | 0.0298 | 0.0035 | 0.0062 | 0.8540 |
0.3563 | 3.0 | 17901 | 0.8213 | 0.0684 | 0.0040 | 0.0075 | 0.8600 |
0.3119 | 4.0 | 23868 | 0.8518 | 0.0329 | 0.0027 | 0.0050 | 0.8581 |
0.2855 | 5.0 | 29835 | 0.8638 | 0.0525 | 0.0098 | 0.0165 | 0.8523 |
0.2662 | 6.0 | 35802 | 0.9211 | 0.0645 | 0.0092 | 0.0160 | 0.8548 |
0.2314 | 7.0 | 41769 | 0.9358 | 0.0493 | 0.0111 | 0.0182 | 0.8502 |
0.2281 | 8.0 | 47736 | 0.9458 | 0.0459 | 0.0151 | 0.0227 | 0.8459 |
0.2003 | 9.0 | 53703 | 0.9553 | 0.0496 | 0.0153 | 0.0234 | 0.8473 |
0.2115 | 10.0 | 59670 | 0.9821 | 0.0476 | 0.0160 | 0.0239 | 0.8462 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.12.1
- Datasets 2.13.2
- Tokenizers 0.13.3
- Downloads last month
- 2