---
library_name: transformers
datasets:
- thainq107/vi-ntc-scv
language:
- vi
metrics:
- accuracy
base_model:
- vinai/phobert-base-v2
---

# 🇻🇳 PhoBERT Vietnamese Sentiment Analysis (with LoRA)

This model is a fine-tuned version of [`VinAI/phobert-base`](https://huggingface.co/VinAI/phobert-base) using LoRA (Low-Rank Adaptation) for **Vietnamese sentiment classification**.

It is trained on a Vietnamese dataset to classify sentiments into `positive`, `negative`.

---

## 🧠 Model Architecture

- Base: `PhoBERT` (`VinAI/phobert-base`)
- Fine-tuning technique: **[LoRA](https://arxiv.org/abs/2106.09685)** via PEFT
- Task: Sequence Classification (`TaskType.SEQ_CLS`)

### 🔧 LoRA Config
```python
LoraConfig(
    r=16,
    lora_alpha=32,
    target_modules=["query", "value"],
    lora_dropout=0.1,
    bias="none",
    task_type=TaskType.SEQ_CLS
)