fine-tuned-indo-sentiment
This model is a fine-tuned version of Aardiiiiy/indobertweet-base-Indonesian-sentiment-analysis on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2057
- Accuracy: 0.9196
- Precision: 0.9237
- Recall: 0.9196
- F1: 0.9198
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
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.1916 | 1.0040 | 500 | 0.2213 | 0.9093 | 0.9182 | 0.9093 | 0.9096 |
| 0.2089 | 2.0080 | 1000 | 0.2447 | 0.9013 | 0.9114 | 0.9013 | 0.9015 |
| 0.1696 | 3.0120 | 1500 | 0.2057 | 0.9196 | 0.9237 | 0.9196 | 0.9198 |
| 0.2197 | 4.0161 | 2000 | 0.2222 | 0.9151 | 0.9192 | 0.9151 | 0.9155 |
| 0.3123 | 5.0201 | 2500 | 0.2106 | 0.9119 | 0.9171 | 0.9119 | 0.9122 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Tokenizers 0.21.1
- Downloads last month
- 11