metadata
license: mit
base_model: w11wo/indonesian-roberta-base-sentiment-classifier
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: test-reward-model
results: []
test-reward-model
This model is a fine-tuned version of w11wo/indonesian-roberta-base-sentiment-classifier on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2784
- Accuracy: 0.8817
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: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.7179 | 0.67 | 50 | 0.6866 | 0.6237 |
| 0.6866 | 1.33 | 100 | 0.6661 | 0.7742 |
| 0.6546 | 2.0 | 150 | 0.6039 | 0.8280 |
| 0.5421 | 2.67 | 200 | 0.4624 | 0.8172 |
| 0.3965 | 3.33 | 250 | 0.3958 | 0.8280 |
| 0.3244 | 4.0 | 300 | 0.3502 | 0.8495 |
| 0.251 | 4.67 | 350 | 0.4012 | 0.8602 |
| 0.1579 | 5.33 | 400 | 0.3184 | 0.8602 |
| 0.135 | 6.0 | 450 | 0.3141 | 0.8710 |
| 0.1114 | 6.67 | 500 | 0.3474 | 0.8495 |
| 0.0929 | 7.33 | 550 | 0.2931 | 0.8495 |
| 0.0829 | 8.0 | 600 | 0.2757 | 0.8710 |
| 0.0834 | 8.67 | 650 | 0.2889 | 0.8817 |
| 0.057 | 9.33 | 700 | 0.2810 | 0.8925 |
| 0.0503 | 10.0 | 750 | 0.2800 | 0.8817 |
| 0.062 | 10.67 | 800 | 0.2806 | 0.8817 |
| 0.0303 | 11.33 | 850 | 0.2971 | 0.8817 |
| 0.0246 | 12.0 | 900 | 0.2784 | 0.8817 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.14.5
- Tokenizers 0.15.2