Network

Hidden width H: 5 Activation Output

Softmax (classification) Linear (regression)

Learning rate η: 0.10 Dataset

Mode

Stats

Layers2→5→5→2

Parameters0

Sample—

Loss—

Accuracy—

Epoch0

|∇|—

Click the canvas to add a labelled point (left = class 0 red, right-click = class 1 blue). Backward pass shows δ pulses right→left; edges glow by |∂L/∂w|.