Downloads · 30 days
16
46% of all-time downloads
BARTPho_Syllable_ver2
qthuan2604/BARTPho_Syllable_ver2
BARTPho_Syllable_ver2 is a machine learning model from qthuan2604. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
All-time downloads
35
Public
Parameters
396M
3.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors1.6 GB · 100%
At a glance
- Library
- transformers
- License
- mit
- Model type
- mbart
- Access
- Public
- Created
- Nov 24, 2025
- Updated
- Nov 24, 2025
- SHA
- 46645a2d
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
BARTPho_Syllable_ver2
This model is a fine-tuned version of vinai/bartpho-syllable on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0529
- Cer: 0.3133
- Wer: 0.3747
- Char Acc: 0.6432
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: 0.0002
- train_batch_size: 48
- eval_batch_size: 53
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 192
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer | Wer | Char Acc |
|---|---|---|---|---|---|---|
| 0.4262 | 1.0 | 167 | 0.1963 | 0.4369 | 0.6588 | 0.3850 |
| 0.2054 | 2.0 | 334 | 0.0929 | 0.3479 | 0.4576 | 0.5507 |
| 0.1249 | 3.0 | 501 | 0.0716 | 0.3309 | 0.4192 | 0.6095 |
| 0.0943 | 4.0 | 668 | 0.0622 | 0.3232 | 0.4009 | 0.6261 |
| 0.0754 | 5.0 | 835 | 0.0555 | 0.3163 | 0.3846 | 0.6287 |
| 0.06 | 6.0 | 1002 | 0.0548 | 0.3159 | 0.3833 | 0.6423 |
| 0.051 | 7.0 | 1169 | 0.0540 | 0.3157 | 0.3809 | 0.6386 |
| 0.043 | 8.0 | 1336 | 0.0532 | 0.3138 | 0.3762 | 0.6442 |
| 0.0401 | 9.0 | 1503 | 0.0529 | 0.3133 | 0.3751 | 0.6450 |
| 0.0386 | 10.0 | 1670 | 0.0529 | 0.3133 | 0.3747 | 0.6432 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.9.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.1
- Library
- transformers
- Type
- mbart
- License
- mit
- Created
- Nov 24, 2025
- Updated
- Nov 24, 2025