About Spiral Neural Network Playground
This playground opens on the two-arm spiral, the classic hard case for small networks. It starts with 3 hidden layers of 8 Tanh neurons, the X1 and X2 inputs plus sin(πX1) and sin(πX2), and 150 epochs. It is free, needs no sign-up and runs in your browser, so nothing you enter is uploaded.
A straight line cannot separate the arms, and one small hidden layer rarely can either, which is why the spiral is used to show the value of depth and of useful input features. The sine features give the network curved building blocks to start from. The decision boundary redraws every few epochs, so you can watch it wind outwards as the loss falls.
Results vary, because training starts from random weights. A run can stall at a boundary that is part of the way there. Raise the epochs, add neurons, lower the learning rate or press Train again to compare.
How to use Spiral Neural Network Playground
- Choose a dataset: Circle, XOR, Two Gaussians or Spiral. The points are drawn on the plot straight away.
- Set the network: hidden layers (1 to 4), neurons per layer (1 to 16), the activation function (ReLU, Tanh, Sigmoid or Linear) and which input features it receives.
- Set the training: learning rate, L2 regularization, batch size, noise and the number of epochs (10 to 200).
- Press Train model. The status shows the epoch, the loss lines grow, and the boundary and the network diagram refresh as the network learns. Choose how often the boundary redraws, or leave it at every 2 epochs. Press Stop to end early.
- Read the result: the coloured background is the decision boundary, the diagram shows each link coloured by the sign of its weight and sized by its size, and the accuracy line gives training and test accuracy.
- Change one setting and train again to compare. Reset clears the plot and the curves.
Features
- Four datasets: circle, XOR, two Gaussians and spiral, 200 points each, with a noise slider.
- Seven selectable input features: X1, X2, X1 squared, X2 squared, X1 times X2, sin(πX1) and sin(πX2), where the sine features really are sin(π × X), so one full period spans the plot.
- One to four hidden layers of 1 to 16 neurons, with ReLU, Tanh, Sigmoid or Linear activation.
- Learning rate from 0.003 to 1, L2 regularization from none to 0.1, batch size 4 to 64 and 10 to 200 epochs.
- Live training loss and test loss curves, with the latest values in the heading.
- Training and test accuracy after the run, on a held-out fifth of the points.
- A decision boundary painted behind the data and redrawn every 1, 2, 5 or 10 epochs while the network trains, without freezing the page. With reduced motion switched on in your system, it updates once every 10 epochs instead.
- A network diagram: links are blue for positive weights and orange for negative ones, thicker for larger weights, and neurons are sized by their bias.
- A scatter plot that resizes with its container and has a text description for screen readers.
- Four starting points: the circle, XOR, spiral and two-Gaussians pages each open the tool with their own dataset, features and network.
- Stop and Reset buttons.
Worked examples
Spiral, X1, X2, sin(πX1), sin(πX2), 3 layers of 8 Tanh neurons, 150 epochs gives The loss usually falls steadily and the boundary starts to follow the arms. The exact accuracy depends on the random start.
No fixed accuracy is promised: the weights start random.
Spiral, X1 and X2 only, 1 layer of 4 neurons gives The boundary stays a smooth curve and cannot follow the arms, so accuracy stays well below the deeper network.
Privacy: how it works
- There is nothing to upload: the datasets are generated in the page, and the network is trained by TensorFlow.js running on your device. TensorFlow.js ships with this site and is loaded from it when you first press Train. Your settings are not sent anywhere or stored.
Tips
- Change one setting per run. If you change three, you will not know which one mattered.
- A learning rate of 1 can make the loss jump around, and 0.003 can be too slow for 60 epochs. The default of 0.03 is a reasonable start.
- Compare the two loss lines. Training loss well below test loss is the sign of overfitting.
- Tanh and Sigmoid give smooth boundaries, while ReLU gives boundaries made of straight pieces.
- For the spiral, add the sin features, use more layers and neurons, and raise the epochs.
Frequently asked questions
Is it really free?
Yes. There is no price, no trial and no credit card. Ulkaa Tool does the work with your own device, so there is no server bill to pass on to you.
Do I need an account?
No. There is no sign-up and no login. Open the page and use it. An Ulkaa account is not needed for anything on this tool.
Does it run in my browser?
Yes. Your settings are processed by code running in your browser tab, on your own device.
Is my data uploaded?
No. Your settings are never sent to Ulkaa or to anyone else. You can check this yourself by watching the Network tab in your browser developer tools while you use the tool.
Why is the spiral so hard for a neural network?
The two classes are interleaved, so separating them needs a boundary that winds around the centre. That takes several layers or good input features, and many epochs.
Which settings help most on the spiral?
More hidden layers and neurons, the sin(πX1) and sin(πX2) features, a learning rate near 0.03 or lower, and 150 to 200 epochs.
More TensorFlow Neural Network Playground pages
- TensorFlow Neural Network PlaygroundTrain a small neural network on a 2D dataset and watch the decision boundary and the loss curves change.
- Neural Network Playground OnlineSee what layers, neurons and activations do to a small network.
- TensorFlow Playground OnlineA TensorFlow.js playground that trains in your browser.
- Two Gaussians Classifier PlaygroundThe easy case: one neuron, a straight boundary, two blobs.