Information Gain

Information gain lab — parent entropy, categorical feature splits, weighted child entropy, and IG = H(parent) − H(children). See how a decision tree picks splits; compare features side by side.

• Foundations
• Vector Math
• Matrix Multiply
• Inverse Matrix
• Eigenvalues & Eigenvectors
• SVD & Orthogonality
• Derivatives

Ngerk Visualizer — interactive ML simulations in the browser.