Neural Network Sandbox

Configure input features, add hidden layers, modify neuron densities, and select activations. Backpropagation computes gradients live to shape the classification space.

Input Features

x₁
x₂
x₁²
x₂²
x₁x₂
sin(x₁)
sin(x₂)

Network Topology

Output & Boundary

Epoch
0
Train Loss
–
Test Loss
–
Accuracy
–%

Loss History Log