finbert-indonesia-finansial
This model is a fine-tuned version of mdhugol/indonesia-bert-sentiment-classification on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2178
- Accuracy: 0.8083
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: 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 |
|---|---|---|---|---|
| No log | 1.0 | 83 | 0.4804 | 0.7745 |
| No log | 2.0 | 166 | 0.4861 | 0.7929 |
| No log | 3.0 | 249 | 0.6559 | 0.7991 |
| No log | 4.0 | 332 | 0.9976 | 0.7791 |
| No log | 5.0 | 415 | 1.0642 | 0.8021 |
| No log | 6.0 | 498 | 1.1542 | 0.8006 |
| 0.2558 | 7.0 | 581 | 1.1414 | 0.8083 |
| 0.2558 | 8.0 | 664 | 1.1794 | 0.8113 |
| 0.2558 | 9.0 | 747 | 1.2184 | 0.8052 |
| 0.2558 | 10.0 | 830 | 1.2178 | 0.8083 |
Framework versions
- Transformers 4.56.2
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1