GEO Bot Prototype

A chatbot application for querying geological data using LLMs (Llama 3.2, Deepseek 1.5, OpenAI) via Ollama.

🚀 Installation Guide

✅ System Requirements

• Operating System: Windows 10/11
• Python Version: 3.13
• Recommended Environment: Virtual Environment (venv)

---

📂 Data Folder Structure

The application uses a Data folder to store:

• 📝 Uploaded files → uploaded_document.txt
• 📝 Processed content → processed_content.txt
• 💬 Chat history → ChatSessions/
• 📩 Feedback data → feedback_dataset.json
• 📊 Expected query responses → expected_query_responses.xlsx
• 🔍 Query logs → query_responses.xlsx

---

🛠️ Installation Steps

1⃣ Clone the Repository

git clone https://github.com/hlin-0420/GEO-bot-prototype.git
cd GEO-bot-prototype

2⃣ Set Up a Virtual Environment

python -m venv venv
# Windows
venv\Scripts\activate  
# macOS/Linux
source venv/bin/activate

3⃣ Install Dependencies

pip install -r requirements.txt

4⃣ Install Ollama

• Download and install Ollama from Ollama Official Site.

5⃣ Download Required Ollama Models

This project uses Llama 3.2, Deepseek 1.5, and OpenAI models. Install them using:

ollama pull llama3.2:latest
ollama pull deepseek1.5

⚠️ Note: OpenAI models are not available via ollama pull.
Instead, configure OpenAI API by setting an environment variable:

export OPENAI_API_KEY="your-api-key-here"  # macOS/Linux
set OPENAI_API_KEY="your-api-key-here"  # Windows

6⃣ Start the Chatbot

Run the application:

python offline-app.py

• Wait until the local host link appears.
• Open the link in your browser:

  http://127.0.0.1:5000/
  

✅ Before running, make sure Ollama is open to enable the chatbot's connection with the model.

Optional 7⃣ Install and Build Documentation

Sphinx generates the documentation through the following steps:

1. Install Sphinx and its required themes:

pip install sphinx sphinx-rtd-theme

2. Navigate to the docs folder from your project root folder:

cd docs

Based on your OS, build the documentation with the instructions:

• For Windows (Powershell or CMD)

.\make.bat html

• For macOS / Linux

make html

3. Move the built documentation file to the Flask templates directory.

---

💡 Troubleshooting

❌ Ollama Models Not Found?

Ensure Ollama is running:

• Open Ollama from the Windows search bar (or Terminal for macOS/Linux).
• Verify the installed models using:

  ollama list

  If no models appear, re-run:

  ollama pull llama3.2:latest
  ollama pull deepseek1.5
  ollama pull tinyllama:latest
  ollama pull gemma3:1b

❌ Flask App Not Starting?

• Run the command:

  python offline-app.py

• If you see ModuleNotFoundError, try:

  pip install -r requirements.txt

❌ 404 Page Not Found?

• Ensure Flask is running and open:

  http://127.0.0.1:5000/
  

---

✅ Expected Output

Upon successful execution, you should see:

 * Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)

Copy and paste this link into your browser.

---

🎯 Next Steps

• 📚 Read the [Documentation] by viewing the option 📖 Documentation from the menu page.
• 🛠️ Customize models in configurations.
• 🚀 Extend functionality with additional APIs.

---