metadata
license: mit
base_model: ayameRushia/bert-base-indonesian-1.5G-sentiment-analysis-smsa
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: best_bert_model_fold_5
results: []
best_bert_model_fold_5
This model is a fine-tuned version of ayameRushia/bert-base-indonesian-1.5G-sentiment-analysis-smsa on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0532
- Accuracy: 0.8586
- Precision: 0.8342
- Recall: 0.8005
- F1: 0.8134
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 |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 252 | 0.5880 | 0.8048 | 0.8281 | 0.6620 | 0.6694 |
| 0.5304 | 2.0 | 504 | 0.8974 | 0.7928 | 0.7497 | 0.7582 | 0.7525 |
| 0.5304 | 3.0 | 756 | 1.0145 | 0.7928 | 0.7483 | 0.7455 | 0.7418 |
| 0.1872 | 4.0 | 1008 | 1.0229 | 0.8227 | 0.7856 | 0.7561 | 0.7678 |
| 0.1872 | 5.0 | 1260 | 1.0532 | 0.8586 | 0.8342 | 0.8005 | 0.8134 |
| 0.037 | 6.0 | 1512 | 1.2624 | 0.8347 | 0.7927 | 0.7997 | 0.7957 |
| 0.037 | 7.0 | 1764 | 1.3287 | 0.8227 | 0.7806 | 0.7951 | 0.7870 |
| 0.0076 | 8.0 | 2016 | 1.3369 | 0.8347 | 0.8064 | 0.7747 | 0.7863 |
| 0.0076 | 9.0 | 2268 | 1.3292 | 0.8406 | 0.8093 | 0.7883 | 0.7974 |
| 0.0002 | 10.0 | 2520 | 1.3507 | 0.8347 | 0.7970 | 0.7844 | 0.7901 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.19.1