This project is an interactive JavaScript visualization of a perceptron neural network that learns to classify numbers as even or odd. It provides a live, browser-based interface with visualization powered by D3.js.

• 🔍 Input a number and test if it's classified as even or odd
• 🧠 Visual graph of the neural network structure (input → neuron → output)
• ⚙️ Train automatically until reaching a target accuracy
• ⏹️ Stop training at any time
• 📊 Live display of weights, bias, epoch count, and accuracy

This project uses a single-layer perceptron — the simplest form of a neural network — to classify numbers.

The perceptron calculates the output y from input x using:

y = f(w * x + b) 

Where:

• x is the input (in this case, x = n % 2)
• w is the weight
• b is the bias
• f(z) is the activation function:

  f(z) = { 1, if z >= 0.5 0, otherwise } 

The output y will be:

• 1 for odd numbers
• 0 for even numbers

Each epoch updates weights using the perceptron learning rule:

error = expected - predicted w = w + learningRate * error * x b = b + learningRate * error 

This update process is repeated until the accuracy reaches a user-defined threshold.

D3.js is used to render a simple visual of the perceptron network:

• Input Node
• Single Neuron
• Output Node

These are connected with lines and display live updates of current weights and bias.

1. Open index.html in your browser.
2. Enter a minimum accuracy (default: 90%).
3. Click Train Perceptron.
4. The network will train until the desired accuracy is reached.
5. Use Test Number to verify the classification of any integer.
6. Click Stop Training to manually halt training.

• HTML5
• JavaScript (Vanilla)
• D3.js v7

No build tools or dependencies required — just open the HTML file in your browser.

This project is meant as a learning tool to demonstrate:

• How perceptrons work
• Binary classification
• Live training feedback
• Basic D3.js visualizations

MIT — Free to use, modify, and share.