Linear Perceptron Learning Simulation

Explore the perceptron algorithm with modern visuals

Instructions

Stage 1: Click on the canvas to add points. Toggle between Class 1 (red) and Class 2 (blue) using the radio buttons. Click "Start" to initialize the weight vector and "Step" to execute the learning algorithm step by step.

Stage 2: Create different datasets by adjusting the separation between points and observe how this influences convergence.

Stage 3: Adjust the decay rate and automation speed using the sliders to observe their effects on the learning rate and convergence.

Stage 4: Use the "Show Best" button to toggle between showing the current boundary and the best boundary found so far (with the fewest errors).

0.05
500
Iterations: 0
Current Boundary
Best Boundary