Neural Network Architecture

Handwritten Digit 0/1 Recognition: Input(400) - Hidden1(25) - Hidden2(15) - Output(1)

Network Topology
Input (400) Hidden 1 (25) Hidden 2 (15) Output (1)
Interactive Controls
Connection Density 0.80
Activation Intensity 0.70
Propagation Speed 1.00
Ready: Click "Forward Pass" to watch the signal propagate left to right.
Parameter Statistics
Layer 1 Params
10,025
Layer 2 Params
390
Output Params
16
Total Params
10,431
Architecture Details
Layer 1: 400 x 25 + 25 = 10,025
Layer 2: 25 x 15 + 15 = 390
Output: 15 x 1 + 1 = 16
Total: 10,431 trainable parameters
Activation: ReLU (hidden) + Sigmoid (output)