arabic2023_ner_model
This model is a fine-tuned version of distilbert-base-uncased on the wikiann dataset. It achieves the following results on the evaluation set:
- Loss: 0.3950
- Precision: 0.8255
- Recall: 0.8313
- F1: 0.8284
- Accuracy: 0.9048
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: 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: 2
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.1594 | 1.0 | 1250 | 0.4149 | 0.8145 | 0.8133 | 0.8139 | 0.8974 |
0.116 | 2.0 | 2500 | 0.3950 | 0.8255 | 0.8313 | 0.8284 | 0.9048 |
Framework versions
- Transformers 4.27.1
- Pytorch 2.0.1+cu118
- Datasets 2.9.0
- Tokenizers 0.13.3
- Downloads last month
- 2
Inference API (serverless) is not available, repository is disabled.
Dataset used to train Falah/arabic2023_ner_model
Evaluation results
- Precision on wikiannvalidation set self-reported0.826
- Recall on wikiannvalidation set self-reported0.831
- F1 on wikiannvalidation set self-reported0.828
- Accuracy on wikiannvalidation set self-reported0.905