🐍 Python Code Generator

A production-ready Streamlit application that generates Python code using Together AI's language models.

🌟 Features

• Multiple AI Models: Choose from Llama 3.1, CodeLlama, and Mixtral models
• Advanced Controls: Fine-tune temperature and token limits
• Example Prompts: 8 pre-built prompts for quick testing
• Code Download: One-click download of generated code
• Professional UI: Clean, responsive interface with syntax highlighting
• Session Management: Maintains state between interactions

🚀 Quick Start

Prerequisites

1. Together AI API Key - Get it here
2. Python 3.8+ installed
3. Git (for deployment)

Local Development

1. Clone or download this repository

2. Install dependencies

   pip install -r requirements.txt

3. Set up environment variable

   # Windows (PowerShell)
   $env:TOGETHER_API_KEY="your_api_key_here"
   
   # Windows (CMD)
   set TOGETHER_API_KEY=your_api_key_here
   
   # Linux/Mac
   export TOGETHER_API_KEY=your_api_key_here

4. Run the application

   streamlit run app.py

5. Open in browser - Navigate to http://localhost:8501

📦 Deployment to Streamlit Cloud

Step 1: Upload to GitHub

1. Create a new GitHub repository
2. Upload the following files:
   • app.py
   • requirements.txt
   • README.md
   • .gitignore

Step 2: Deploy on Streamlit Cloud

1. Go to share.streamlit.io
2. Sign in with GitHub
3. Click "New app"
4. Configure:
   • Repository: Your GitHub repository
   • Branch: main
   • Main file: app.py
5. Click "Advanced settings"
6. In Secrets, add:

   TOGETHER_API_KEY = "your_api_key_here"

7. Click "Deploy!"

Your app will be live in 2-3 minutes! 🎉

🎯 Usage

1. Select a Model - Choose from the sidebar (Llama, CodeLlama, or Mixtral)
2. Adjust Settings (Optional) - Fine-tune temperature and max tokens
3. Enter a Prompt - Describe the code you want to generate
4. Generate Code - Click the generate button
5. Download - Save the generated code to a .py file

Example Prompts

• "Write a function to reverse a string"
• "Create a class for a binary search tree with insert and search methods"
• "Write a script to read a CSV file and calculate statistics"
• "Create a decorator for timing function execution"
• "Write a function to validate email addresses using regex"

🔧 Configuration

Available Models

Model	ID	Best For
Llama 3.1 8B	meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo	Fast responses
Llama 3.1 70B	meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo	Complex tasks
CodeLlama 34B	codellama/CodeLlama-34b-Instruct-hf	Code-specialized
Mixtral 8x7B	mistralai/Mixtral-8x7B-Instruct-v0.1	Balanced performance

Parameters

• Temperature (0.0-1.0): Controls creativity
  • Lower = More deterministic
  • Higher = More creative
• Max Tokens (500-4000): Controls response length

📁 Project Structure

Value App/
├── app.py              # Main Streamlit application
├── requirements.txt    # Python dependencies
├── README.md          # Documentation (this file)
├── .gitignore         # Git ignore rules
└── Research/          # Documentation and reference materials

🔒 Security

• ✅ API keys via environment variables
• ✅ No secrets in repository
• ✅ Streamlit secrets management
• ✅ .gitignore configured

🧪 Testing

Manual Testing

1. Test basic code generation
2. Try different models
3. Adjust parameters
4. Test example prompts
5. Verify download functionality

Test Prompts

See the Research/ folder for comprehensive test prompts across various categories:

• Basic functions
• Data structures
• Algorithms
• File operations
• Web development
• OOP concepts

🆘 Troubleshooting

Common Issues

"API key not found"

• Check environment variable is set correctly
• For Streamlit Cloud: Verify secrets are configured
• Ensure key name is exactly TOGETHER_API_KEY

"Module not found"

• Run pip install -r requirements.txt
• Check Python version is 3.8+

App won't load

• Check console for error messages
• Verify all files are in the same directory
• Check internet connection (for API calls)

Code generation fails

• Verify API key is valid
• Check Together AI service status
• Try a simpler prompt first

📚 Resources

• Streamlit Documentation
• Together AI Documentation
• Python Documentation

📄 License

This project is created for educational purposes as part of CAP 4767 Data Mining course.

🎓 Assignment Information

Course: CAP 4767 - Data Mining
Module: 5
Topic: AI Code Generation with Together AI

Requirements Met

• ✅ Streamlit application deployed
• ✅ Together AI integration
• ✅ Clean user interface
• ✅ API key management via secrets
• ✅ Complete documentation
• ✅ GitHub repository

🙏 Acknowledgments

• Built with Streamlit
• Powered by Together AI
• Models: Meta Llama, CodeLlama, Mixtral

---

Made with ❤️ for CAP 4767

Last Updated: October 2025