Model guides

Explore how classical machine-learning models actually work.

Choose a model to connect equations, parameters, training states, predictions, and evaluation evidence with the live browser-side visualizer.

Linear Regression Visualizer

Visualize how linear regression updates weights and bias, follows loss, and produces a prediction from real browser-side calculations.

Inspect logistic regression step by step: weighted sum, sigmoid probability, prediction, loss, gradients, and parameter updates.

Explore K-Means clustering with real centroid initialization, assignments, centroid updates, inertia, and iteration states in your browser.

Visualize K-Nearest Neighbors using genuine distances, selected neighbors, K, distance metrics, and prediction evidence.

Inspect how a decision tree chooses real feature splits, creates nodes, grows branches, and routes a sample to a prediction.

Visualize an SVM margin, support vectors, decision scores, kernels, C, gamma, and genuine optimization states.

Explore Random Forest predictions by inspecting real trees, feature sampling, bootstrap behavior, ensemble votes, and metrics.

Visualize Gradient Boosting stage by stage with real weak learners, learning rate, residual correction, predictions, and metrics.

Inspect AdaBoost weak learners, sample emphasis, estimator weights, learning rate, and ensemble classification evidence.

Explore Extra Trees with real randomized tree structures, feature settings, ensemble predictions, and held-out metrics.