chhaythean/Camtour-On-Mistral-Ai

CamTour Chatbot (v0.2)

A conversational Khmer Tourism Assistant built using a LoRA fine-tuned version of Mistral-7B-Instruct-v0.3, optimized to answer questions about Cambodia’s attractions, culture, food, and travel tips.

This model was developed by Chhaythean LY as part of a hands-on AI research project exploring fine-tuning large language models for localized conversational tasks.

---

🧠 Model Details

• Base Model: mistralai/Mistral-7B-Instruct-v0.3
• Fine-tuning Method: LoRA (Low-Rank Adaptation) via PEFT
• Framework: Hugging Face Transformers + PEFT
• Developed by: Chhaythean LY
• Model Type: Chatbot / Conversational Agent
• Language(s): English (Cambodian context)
• License: ITC Academic License
• Model Version: v0.2
• Pipeline: text-generation

---

💬 Model Purpose

CamTour Chatbot is designed to:

• Provide tourism-related information about Cambodia (e.g., places, food, hotels, activities).
• Engage in friendly, conversational dialogue.
• Serve as a local language adaptation use case of open LLMs for Cambodian contexts.

---

🚀 How to Use

You can quickly try the model with transformers:

from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

model_id = "chhaytheanLY/camtour-chatbot"
pipe = pipeline("text-generation", model=model_id)

prompt = "Tell me about the best places to visit in Siem Reap."
output = pipe(prompt, max_new_tokens=120, temperature=0.7)
print(output[0]["generated_text"])