metadata
library_name: peft
license: mit
base_model: vinai/bartpho-syllable
tags:
- base_model:adapter:vinai/bartpho-syllable
- lora
- transformers
metrics:
- sacrebleu
model-index:
- name: bartpho-vietnamese-correction-bmd1905
results: []
datasets:
- bmd1905/vi-error-correction-v2
bartpho-vietnamese-correction-bmd1905
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: 0.3466
- Sacrebleu: 26.5168
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- 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: linear
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Sacrebleu |
|---|---|---|---|---|
| 0.5948 | 1.0 | 2657 | 0.4292 | 24.7435 |
| 0.5053 | 2.0 | 5314 | 0.3612 | 26.2000 |
| 0.4854 | 3.0 | 7971 | 0.3466 | 26.5168 |
Framework versions
- PEFT 0.17.1
- Transformers 4.57.1
- Pytorch 2.8.0+cu126
- Datasets 4.4.1
- Tokenizers 0.22.1