AraBERT_token_classification_merged
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: 1.4477
- Precision: 0.0264
- Recall: 0.0071
- F1: 0.0112
- Accuracy: 0.6557
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: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.0 | 254 | 1.4827 | 0.0 | 0.0 | 0.0 | 0.6786 |
1.2057 | 2.0 | 508 | 1.4545 | 0.0 | 0.0 | 0.0 | 0.6789 |
1.2057 | 3.0 | 762 | 1.4587 | 0.0022 | 0.0005 | 0.0008 | 0.6553 |
0.9359 | 4.0 | 1016 | 1.4667 | 0.0283 | 0.0041 | 0.0071 | 0.6686 |
0.9359 | 5.0 | 1270 | 1.4477 | 0.0264 | 0.0071 | 0.0112 | 0.6557 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.12.1
- Datasets 2.13.2
- Tokenizers 0.13.3
- Downloads last month
- 1