An interactive web application that recognizes handwritten digits (0-9) using a convolutional neural network (CNN) trained on the MNIST dataset.

Important Notice: This application only works locally due to Google's restrictions on cross-origin requests for the MNIST dataset.

• ✍️ Interactive drawing canvas
• 🔍 Real-time digit classification
• 📊 Confidence percentage display
• 🧠 Pre-trained CNN model (80% accuracy)
• 🎨 Modern UI with neural network visualization

1. Download Node.JS nodejs.org/en/download
2. Clone the repository:

   git clone https://github.com/arturr0/HWR.git
   cd hwr

3. Install http-server:

   npm install -g http-server

4. Run:

   node server.js

5. Run in browser http://localhost:3000