metadata
license: mit
base_model: ayameRushia/roberta-base-indonesian-sentiment-analysis-smsa
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: best_berita_roberta_model_fold_2
results: []
best_berita_roberta_model_fold_2
This model is a fine-tuned version of ayameRushia/roberta-base-indonesian-sentiment-analysis-smsa on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1019
- Accuracy: 0.9884
- Precision: 0.9882
- Recall: 0.9889
- F1: 0.9884
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: 5e-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 | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.6027 | 1.0 | 601 | 0.4467 | 0.8943 | 0.9036 | 0.8947 | 0.8951 |
| 0.2296 | 2.0 | 1202 | 0.3961 | 0.9218 | 0.9282 | 0.9262 | 0.9213 |
| 0.1253 | 3.0 | 1803 | 0.2066 | 0.9651 | 0.9655 | 0.9670 | 0.9652 |
| 0.0869 | 4.0 | 2404 | 0.1721 | 0.9684 | 0.9691 | 0.9699 | 0.9688 |
| 0.0188 | 5.0 | 3005 | 0.1239 | 0.9842 | 0.9840 | 0.9850 | 0.9843 |
| 0.0049 | 6.0 | 3606 | 0.1186 | 0.9825 | 0.9823 | 0.9835 | 0.9826 |
| 0.0057 | 7.0 | 4207 | 0.1019 | 0.9884 | 0.9882 | 0.9889 | 0.9884 |
| 0.0043 | 8.0 | 4808 | 0.3787 | 0.9534 | 0.9549 | 0.9562 | 0.9534 |
| 0.0047 | 9.0 | 5409 | 0.2094 | 0.9759 | 0.9759 | 0.9772 | 0.9760 |
| 0.0007 | 10.0 | 6010 | 0.2093 | 0.9759 | 0.9759 | 0.9772 | 0.9760 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.1.2
- Datasets 2.20.0
- Tokenizers 0.19.1