metadata
license: mit
base_model: indobenchmark/indobert-large-p1
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: indobert-large-p1-twitter-indonesia-sarcastic
results: []
indobert-large-p1-twitter-indonesia-sarcastic
This model is a fine-tuned version of indobenchmark/indobert-large-p1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3207
- Accuracy: 0.8643
- F1: 0.7160
- Precision: 0.7480
- Recall: 0.6866
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: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 100.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.5836 | 1.0 | 59 | 0.4153 | 0.8060 | 0.5738 | 0.6364 | 0.5224 |
| 0.3766 | 2.0 | 118 | 0.3353 | 0.8433 | 0.5962 | 0.8378 | 0.4627 |
| 0.2476 | 3.0 | 177 | 0.3114 | 0.8619 | 0.6942 | 0.7778 | 0.6269 |
| 0.1356 | 4.0 | 236 | 0.3279 | 0.8694 | 0.7328 | 0.75 | 0.7164 |
| 0.0536 | 5.0 | 295 | 0.4265 | 0.8582 | 0.7164 | 0.7164 | 0.7164 |
| 0.0157 | 6.0 | 354 | 0.6448 | 0.8619 | 0.6667 | 0.8409 | 0.5522 |
| 0.0076 | 7.0 | 413 | 0.5739 | 0.8619 | 0.7218 | 0.7273 | 0.7164 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.1+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0