bert-sentiment-digikala-augmented-WithTokens
This model is a fine-tuned version of HooshvareLab/bert-fa-base-uncased-sentiment-digikala on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8273
- Accuracy: 0.8213
- F1: 0.8201
- Precision: 0.8211
- Recall: 0.8197
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.5772 | 1.0 | 986 | 0.4229 | 0.8065 | 0.8057 | 0.8086 | 0.8049 |
| 0.3363 | 2.0 | 1972 | 0.4512 | 0.8168 | 0.8163 | 0.8187 | 0.8155 |
| 0.2009 | 3.0 | 2958 | 0.5756 | 0.8122 | 0.8097 | 0.8140 | 0.8103 |
| 0.1264 | 4.0 | 3944 | 0.8273 | 0.8213 | 0.8201 | 0.8211 | 0.8197 |
| 0.0762 | 5.0 | 4930 | 1.0302 | 0.8139 | 0.8130 | 0.8151 | 0.8123 |
| 0.0464 | 6.0 | 5916 | 1.2564 | 0.8116 | 0.8097 | 0.8098 | 0.8096 |
Framework versions
- Transformers 4.55.2
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4
- Downloads last month
- 1