You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

🧮 Qwen Math Assistant

A powerful AI-powered mathematical problem solver built with Qwen language model and deployed on Hugging Face Spaces.

🚀 Features

Step-by-step Solutions: Get detailed explanations for mathematical problems
Wide Range of Topics: Algebra, calculus, geometry, trigonometry, and more
Interactive Interface: Clean and intuitive Gradio-powered chat interface
Real-time Processing: Fast response times with optimized model loading
Example Problems: Quick-start with pre-loaded example questions

🎯 Supported Math Topics

✅ Algebra: Linear equations, quadratic equations, polynomial operations
✅ Calculus: Derivatives, integrals, limits
✅ Geometry: Area, volume, perimeter calculations
✅ Trigonometry: Sin, cos, tan functions and identities
✅ Statistics: Basic statistical calculations
✅ Number Theory: Prime numbers, factorization

🛠️ How to Use

Load the Model: Click the "🚀 Load Model" button to initialize the AI assistant
Ask Questions: Type your math problem in the text box
Get Solutions: Receive step-by-step explanations and answers
Try Examples: Click on example problems to get started quickly

Example Queries:

"Solve: 2x + 5 = 15"
"What is the derivative of x² + 3x - 2?"
"Factor: x² - 9"
"Calculate the area of a circle with radius 5"

🏗️ Technical Details

Model Architecture

Base Model: Qwen2.5-0.5B (optimized for Hugging Face Spaces)
Framework: Transformers, PyTorch
Interface: Gradio
Deployment: Hugging Face Spaces

Performance Optimizations

Efficient memory usage with low_cpu_mem_usage=True
Dynamic device mapping for GPU/CPU compatibility
Optimized tokenization with proper padding handling
Response caching for better performance

📋 Requirements

Python 3.8+
PyTorch 2.0+
Transformers 4.35+
Gradio 4.0+
PEFT 0.6+

🚀 Deployment

Local Setup

git clone <your-repo>
cd qwen-math-assistant
pip install -r requirements.txt
python app.py

Hugging Face Spaces

Create a new Space on Hugging Face
Upload app.py and requirements.txt
Set the SDK to "gradio"
The Space will automatically build and deploy

🔧 Configuration

The application includes several configurable parameters:

Max Response Length: Adjustable slider (50-500 tokens)
Temperature: Set to 0.7 for balanced creativity/accuracy
Top-p Sampling: 0.9 for diverse but relevant responses

📝 License

This project is open source and available under the MIT License.

🤝 Contributing

Contributions are welcome! Please feel free to submit issues or pull requests.

📞 Support

For questions or support, please open an issue in the repository or contact the maintainers.

Note: This application uses the Qwen language model. Please ensure you comply with the model's usage terms and conditions.

Downloads last month: -

simmonsc
/

qwen-lm