README.md

---
language:
- vi
- en
license: apache-2.0
tags:
- llm-judge
- training-checkpoint
- lora
- unsloth
---

# finetuned_11_12

Full training folder backup - Toàn bộ checkpoints và models.

## 📂 Cấu trúc Folder
```
train_
├── lora_adapters/           # LoRA adapters
├── README.md
├── zero_shot_metrics.json
└── zero_shot_results.csv
```

## 🚀 Sử Dụng

### 1️⃣ Clone Repo
```bash
git lfs install
git clone https://huggingface.co/ImNotTam/finetuned_11_12
cd finetuned_11_12
```

### 2️⃣ Load LoRA Adapters (Nhẹ nhất - khuyến nghị)
```python
from unsloth import FastLanguageModel

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name="ImNotTam/finetuned_11_12",
    subfolder="lora_adapters",
    max_seq_length=2048,
    dtype=None,
    load_in_4bit=True,
)

# Enable inference mode
FastLanguageModel.for_inference(model)

# Test
prompt = "Đánh giá response này..."
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

### 3️⃣ Load Final Model
```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "ImNotTam/finetuned_11_12",
    subfolder="final_model",
    device_map="auto",
    torch_dtype="auto"
)
tokenizer = AutoTokenizer.from_pretrained("ImNotTam/finetuned_11_12", subfolder="final_model")

# Inference
inputs = tokenizer("Your prompt", return_tensors="pt").to("cuda")
outputs = model.generate(**inputs)
print(tokenizer.decode(outputs[0]))
```

### 4️⃣ Resume Training từ Checkpoint
```python
from transformers import Trainer, TrainingArguments

# Load checkpoint muốn resume
model = AutoModelForCausalLM.from_pretrained(
    "ImNotTam/finetuned_11_12",
    subfolder="checkpoint-210",  # Chọn checkpoint
    device_map="auto"
)

# Continue training
trainer = Trainer(
    model=model,
    args=TrainingArguments(
        output_dir="./continue_training",
        # ... your training args
    ),
)
trainer.train(resume_from_checkpoint=True)
```

### 5️⃣ Fine-tune Tiếp từ LoRA Adapter
```python
from unsloth import FastLanguageModel
from trl import SFTTrainer

# Load LoRA adapter
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name="ImNotTam/finetuned_11_12",
    subfolder="lora_adapters",
    max_seq_length=2048,
    dtype=None,
    load_in_4bit=True,
)

# Add LoRA config để train tiếp
model = FastLanguageModel.get_peft_model(
    model,
    r=16,
    target_modules=["q_proj", "k_proj", "v_proj", "o_proj",
                    "gate_proj", "up_proj", "down_proj"],
    lora_alpha=16,
    lora_dropout=0,
    bias="none",
    use_gradient_checkpointing="unsloth",
)

# Train với data mới
trainer = SFTTrainer(
    model=model,
    tokenizer=tokenizer,
    train_dataset=your_new_dataset,
    # ... training args
)
trainer.train()
```

### 6️⃣ Xem Metrics và Results
```python
import json
import pandas as pd

# Load metrics
with open("zero_shot_metrics.json", "r") as f:
    metrics = json.load(f)
print("📊 Metrics:", metrics)

# Load results
results = pd.read_csv("zero_shot_results.csv")
print("\n📈 Results:")
print(results.head())
```

## 📋 Nội Dung Repo

| Folder/File | Mô tả | Kích thước |
|-------------|-------|------------|
| `lora_adapters/` | LoRA adapters (nhẹ) | ~50-100 MB |
| `final_model/` | Model merged đầy đủ | ~4-8 GB |
| `checkpoint-150/` | Training checkpoint | ~4-8 GB |
| `checkpoint-200/` | Training checkpoint | ~4-8 GB |
| `checkpoint-210/` | Training checkpoint | ~4-8 GB |
| `zero_shot_metrics.json` | Evaluation metrics | <1 MB |
| `zero_shot_results.csv` | Detailed results | <1 MB |

## 💡 Khuyến Nghị

- **Inference nhanh:** Dùng `lora_adapters/`
- **Production:** Dùng `final_model/`
- **Train tiếp:** Load `lora_adapters/` + add LoRA config
- **Resume training:** Load checkpoint cụ thể

## 📦 Requirements
```bash
pip install unsloth transformers torch trl
```

## 📄 License

Apache 2.0