Bert-Sentiment-Fa
This model is a fine-tuned version of dadashzadeh/roberta-sentiment-persian on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2509
- Accuracy: 0.8333
- F1: 0.8213
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 | Accuracy | F1 |
|---|---|---|---|---|---|
| No log | 1.0 | 270 | 0.4689 | 0.8208 | 0.8081 |
| 0.4419 | 2.0 | 540 | 0.7073 | 0.8042 | 0.7840 |
| 0.4419 | 3.0 | 810 | 0.7497 | 0.8042 | 0.7820 |
| 0.1823 | 4.0 | 1080 | 0.9258 | 0.8167 | 0.7981 |
| 0.1823 | 5.0 | 1350 | 1.0813 | 0.8042 | 0.7897 |
| 0.0742 | 6.0 | 1620 | 1.1488 | 0.8042 | 0.7866 |
| 0.0742 | 7.0 | 1890 | 1.2846 | 0.8167 | 0.8023 |
| 0.0311 | 8.0 | 2160 | 1.2308 | 0.8333 | 0.8193 |
| 0.0311 | 9.0 | 2430 | 1.2446 | 0.8333 | 0.8193 |
| 0.0108 | 10.0 | 2700 | 1.2509 | 0.8333 | 0.8213 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1