README.md

# 🧠 AgriQA TinyLlama LoRA Adapter

This repository contains a [LoRA](https://arxiv.org/abs/2106.09685) adapter fine-tuned on the [AgriQA](https://huggingface.co/datasets/shchoi83/agriQA) dataset using the [TinyLlama/TinyLlama-1.1B-Chat](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat) base model.

---

## 🔧 Model Details

- **Base Model**: [`TinyLlama/TinyLlama-1.1B-Chat`](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat)
- **Adapter Type**: LoRA (Low-Rank Adaptation)
- **Adapter Size**: ~4.5MB
- **Dataset**: [`shchoi83/agriQA`](https://huggingface.co/datasets/shchoi83/agriQA)
- **Language**: English
- **Task**: Instruction-tuned Question Answering in Agriculture domain
- **Trained by**: [@theone049](https://huggingface.co/theone049)

---

## 📌 Usage

To use this adapter, load it on top of the base model:

```python
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel, PeftConfig

# Load base model
base_model = AutoModelForCausalLM.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat")
tokenizer = AutoTokenizer.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat")

# Load adapter
model = PeftModel.from_pretrained(base_model, "theone049/agriqa-tinyllama-lora-adapter")

# Run inference
prompt = """### Instruction:
Answer the agricultural question.

### Input:
What is the ideal pH range for growing rice?

### Response:"""

inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))