Practice: Neural Network Calculation

import math

# Build your own neuron and test it on different inputs

def sigmoid(z):
    return 1 / (1 + math.exp(-z))

def relu(z):
    return max(0.0, z)

class SimpleNeuron:
    def __init__(self, weights, bias, activation="relu"):
        self.weights = weights
        self.bias = bias
        self.activation = activation
    
    def forward(self, inputs):
        if len(inputs) != len(self.weights):
            raise ValueError("Input and weight count must match")
        
        z = sum(x * w for x, w in zip(inputs, self.weights)) + self.bias
        
        if self.activation == "relu":
            return relu(z)
        elif self.activation == "sigmoid":
            return sigmoid(z)
        else:
            return z
    
    def predict_class(self, inputs):
        output = self.forward(inputs)
        return 1 if output >= 0.5 else 0


# Create a neuron
neuron = SimpleNeuron(
    weights=[0.7, -0.4, 0.9],
    bias=-0.2,
    activation="sigmoid"
)

# Test with different inputs
test_cases = [
    ([1.0, 0.0, 1.0], "All positives"),
    ([0.0, 1.0, 0.0], "Middle input only"),
    ([0.5, 0.5, 0.5], "Equal inputs"),
    ([0.1, 0.9, 0.1], "Mostly negative-weighted"),
]

print("Neuron Output Test:")
print(f"Weights: {neuron.weights}, Bias: {neuron.bias}")
print()
for inputs, description in test_cases:
    output = neuron.forward(inputs)
    cls = neuron.predict_class(inputs)
    print(f"  {description}: inputs={inputs} -> output={output:.3f}, class={cls}")