Back to Dashboard
Neural Network Simulator
3-4-4-2 Network • SIGMOID
Idle
Epoch 0/100
Loss 0.500
Accuracy 75.0%
Train
Reset
Network Type
Single Layer Perceptron
Multi-Layer Network
Architecture
Input Nodes: 3
Hidden Layers: 2
Nodes per Hidden Layer: 4
Output Nodes: 2
Activation Function
Sigmoid
ReLU
Tanh
Leaky ReLU
Linear
Training
Learning Rate: 0.010
Max Epochs: 100
Training Speed: 200ms
Multi-Layer Neural Network
3-4-4-2 • SIGMOID
Input Layer
Hidden Layer 1
Hidden Layer 2
Output Layer
Activation Function: SIGMOID
x
y
0
-3
-2
-1
1
2
3
0.2
0.4
0.6
0.8
1.0
sigmoid
f(x) = 1/(1+e^(-x))
Range: (0,1)