🧠 Perceptron Step-by-Step Interactive Demo
🎯 Learning Objectives
By the end of this demo, you will be able to:
- Understand how a single neuron makes decisions – See the magic behind binary classification
- Learn about weights, bias, and activation functions – The building blocks of neural networks
- See how learning happens through weight updates – Watch the perceptron learn from mistakes
- Discover the limitations of linear decision boundaries – Why we need more complex networks
- Visualize the learning process step by step – See the training in action
- Interactively add data points – Observe how new data influences the dataset.
+-------------------------------------------------------------+
| PERCEPTRON FLOW |
+-------------------------------------------------------------+
| |
| INPUT DATA WEIGHTS SUMMATION ACTIVATION OUTPUT
| | | | | |
| [x1, x2] --> [w1, w2] --> S(xi*wi) --> Step Function --> 0 or 1
| |
| +---------------------------------------------------------+ |
| | TRAINING LOOP | |
| | +---------+ +---------+ +---------+ | |
| | | Predict | --> | Compare | --> | Update | | |
| | | Output | | with | | Weights | | |
| | +---------+ | Target | +---------+ | |
| | +---------+ | |
| +---------------------------------------------------------+ |
+-------------------------------------------------------------+
Welcome! This interactive demo lets you
learn how a perceptron works step by step.
- Adjust the parameters below to see how learning rate, epochs, noise, and data size affect learning.
- Read the explanations, run each step, and observe the output and plots.
- New in Step 1: Click on the plot to add your own data points!
- All code is beginner-friendly and richly commented, with analogies to help you understand.
Tip: Try changing the sliders and see how the perceptron adapts!
Learning Rate:
0.10
Epochs:
20
Samples per Class:
100
Noise:
0.00