PPolySim OS

Neural Network Playground

Watch a neural network learn in real time. As backpropagation adjusts its weights, the decision boundary bends to separate the two classes.

Neural Network PlaygroundLive

Controls

Presets

A tiny multilayer perceptron trained by real backpropagation. The background shows its learned decision boundary; dots are the training data, colored by true class.

▶ Run in Python

Data Inspector

Epoch0
Loss0.0000
Architecture2→8→1

Governing equation

Reading this result: With 8 hidden units the net bends the boundary into curves — more units capture finer shapes like the spiral, but too many can overfit.

Runs locally in your browser — free forever. Scale to the cloud when reality gets heavy.

or unlock everything with Pro →
★ Sign in to save this setup
Save your tuned setup, or drop this simulation into your own site, docs, or course page.

How it works

This is a genuine two-layer perceptron trained by gradient descent and backpropagation, right in your browser. The colored background is the network output at every point; the dots are training data. Change the dataset, hidden-layer size, and learning rate to see how each shapes learning.

Ask the AI about this model

The math, the assumptions, real-world uses, or a code translation — explained for this exact simulation.

More CS / AI simulations

Frequently asked questions

Is this neural network playground tool really free?
Yes. Neural Network Playground runs entirely in your browser using your device's own compute, so local use is free forever. You only pay Compute Tokens if you scale a job to the cloud.
Do I need to install anything?
No. Everything runs client-side in a modern browser — no downloads, no license, no account required to start.
Can I save or share my simulation?
Create a free account to save projects, and use a shareable embed or minted DOI to publish a live, interactive version anywhere.
How accurate are the results?
The solver uses established numerical methods, but results are for research and educational purposes and should be validated against experiment or professional review before you rely on them.