metadata
library_name: transformers
license: agpl-3.0
base_model: vinai/phobert-base-v2
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: phobert-v2-3class_v1
results: []
phobert-v2-3class_v1
This model is a fine-tuned version of vinai/phobert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2110
- Accuracy: 0.9526
- Precision Macro: 0.8907
- Recall Macro: 0.8637
- F1 Macro: 0.8762
- F1 Weighted: 0.9519
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: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision Macro | Recall Macro | F1 Macro | F1 Weighted |
|---|---|---|---|---|---|---|---|---|
| 0.5276 | 1.0 | 90 | 0.2313 | 0.9330 | 0.9562 | 0.7135 | 0.7472 | 0.9219 |
| 0.2071 | 2.0 | 180 | 0.1934 | 0.9488 | 0.8663 | 0.8697 | 0.8679 | 0.9489 |
| 0.1535 | 3.0 | 270 | 0.1780 | 0.9520 | 0.8910 | 0.8427 | 0.8634 | 0.9505 |
| 0.133 | 4.0 | 360 | 0.1885 | 0.9507 | 0.9063 | 0.8376 | 0.8654 | 0.9488 |
| 0.1051 | 5.0 | 450 | 0.1948 | 0.9488 | 0.8749 | 0.8611 | 0.8677 | 0.9484 |
| 0.1016 | 6.0 | 540 | 0.2034 | 0.9520 | 0.9061 | 0.8509 | 0.8743 | 0.9506 |
| 0.0805 | 7.0 | 630 | 0.2120 | 0.9501 | 0.8674 | 0.8700 | 0.8687 | 0.9502 |
| 0.074 | 8.0 | 720 | 0.2037 | 0.9564 | 0.9200 | 0.8625 | 0.8869 | 0.9551 |
| 0.0616 | 9.0 | 810 | 0.2101 | 0.9526 | 0.8907 | 0.8637 | 0.8762 | 0.9519 |
| 0.0612 | 10.0 | 900 | 0.2110 | 0.9526 | 0.8907 | 0.8637 | 0.8762 | 0.9519 |
Framework versions
- Transformers 4.55.0
- Pytorch 2.7.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4