Build a Network.
Drag neurons onto the canvas, connect them into layers, and watch data flow through your creation.
After this lesson you'll know
- How neurons connect to form layers
- The difference between input, hidden, and output layers
- Why architecture matters for what a network can learn
- What happens when data flows through a network
The Concept
Layers are the architecture of intelligence.
A single neuron can make simple decisions. But stack neurons into layers — input, hidden, output — and suddenly the network can recognize faces, translate languages, and write code. The architecture (how many layers, how they connect) determines what the network can learn.
Order the Layers
Arrange the three layer types in the order data flows through them
1Hidden Layer — finds patterns and intermediate features
2Output Layer — makes the final prediction or decision
3Input Layer — receives raw data (pixels, text, numbers)
Build It
Drag, drop, connect, train.
Neuron Palette
Input Neuron
Hidden Neuron
Output Neuron
Challenge: Cats vs Dogs Classifier
Build a network that could classify images of cats and dogs.
Place at least 2 input neurons
Add 2+ hidden neurons
Place 2 output neurons (cat/dog)
Connect all layers
Run training to see data flow
Knowledge Check
Test your understanding.
Match Network Components
Tap one on the left, then its match on the right
Neural networks are layers of neurons connected together. Input neurons receive data. Hidden neurons find patterns. Output neurons make decisions. The magic is in the connections — each one has a weight that gets adjusted during training.