metadata
license: mit
base_model: indobenchmark/indobart
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: bdc2024-indobartv2
results: []
bdc2024-indobartv2
This model is a fine-tuned version of indobenchmark/indobart on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1750
- Accuracy: 0.9432
- Balanced Accuracy: 0.8553
- Precision: 0.9451
- Recall: 0.9432
- F1: 0.9424
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Balanced Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 271 | 0.7994 | 0.7664 | 0.4357 | 0.7110 | 0.7664 | 0.7293 |
| 0.8301 | 2.0 | 542 | 0.5663 | 0.8231 | 0.5167 | 0.8054 | 0.8231 | 0.7946 |
| 0.8301 | 3.0 | 813 | 0.3837 | 0.8690 | 0.6027 | 0.8607 | 0.8690 | 0.8564 |
| 0.4329 | 4.0 | 1084 | 0.2614 | 0.9192 | 0.7725 | 0.9192 | 0.9192 | 0.9161 |
| 0.4329 | 5.0 | 1355 | 0.2037 | 0.9345 | 0.8442 | 0.9345 | 0.9345 | 0.9330 |
| 0.2228 | 6.0 | 1626 | 0.1750 | 0.9432 | 0.8553 | 0.9451 | 0.9432 | 0.9424 |
Framework versions
- Transformers 4.33.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.13.3