Decision Tree Builder
CART grows a decision tree by repeatedly cutting the plane with the axis-aligned split that most reduces class impurity, measured by the Gini index or entropy. Load blob, checkerboard or diagonal data — or click to add points of three classes — set the maximum depth, the minimum samples per leaf and the criterion, then press Play and watch the splits appear one by one: the partition map shades every region by its majority class while the tree diagram below grows the matching nodes, each labelled with its sample count and impurity. A fixed 70/30 train/test split feeds the accuracy readouts, and the diagonal preset exposes the axis-aligned weakness: a staircase of cuts approximating one slanted line.
Runs 100% in your browser — models are trained and computed locally on your device.
Read the full guide to this tool
Notes
- Each split maximises the impurity decrease; Gini is 1 − Σpₖ² and entropy is −Σpₖ·log₂pₖ over the class shares pₖ in the node.
- Growth stops when a node is pure, the maximum depth is reached, or a split would leave fewer than the minimum samples in a leaf.
- Deeper trees push training accuracy toward 100% while test accuracy eventually falls — the tree memorises noise instead of structure.
- Splits are always axis-aligned, so an oblique class boundary needs a whole staircase of cuts that a linear model would capture with one line.
- Runs 100% in your browser — models are trained and computed locally on your device.
Related tools
Function Sketch Finder
Sketch a curve on a zoomable coordinate grid and search dozens of function families for the equations that fit it best.
Linear Regression Playground
Place points, watch the least-squares line follow, and see gradient descent crawl across the loss surface.
Overfitting Sandbox
Crank a polynomial’s degree on noisy data and watch the train/test error curves split apart.
Logistic Regression Playground
Paint two classes of points and watch logistic regression carve out its probability boundary by gradient descent.
Naive Bayes Classifier Demo
Score your own message with a word-count spam filter, or watch Gaussian naive Bayes classify 2D points.
Perceptron Learning Animation
Step through Rosenblatt’s 1958 learning rule and watch the decision boundary rotate with every mistake.