#iSmile Academy: https://www.youtube.com/watch?v=PGyUlMe2dPg&t=181s
import numpy as np

# Define the sigmoid activation function and its derivative
def sigmoid(x):
    return 1 / (1 + np.exp(-x))

def sigmoid_derivative(x):
    return x * (1 - x)

# Define the neural network class
class NeuralNetwork:
    def __init__(self, input_size, hidden_size, output_size): 
    # In the above code there are two _ (under scores) before and after init
        # Initialize weights with random values
        self.weights_input_hidden = np.random.uniform(-1, 1, (input_size, hidden_size))
        self.weights_hidden_output = np.random.uniform(-1, 1, (hidden_size, output_size))

    def forward(self, inputs):
        # Forward propagation
        self.hidden_input = np.dot(inputs, self.weights_input_hidden)
        self.hidden_output = sigmoid(self.hidden_input)
        self.output_input = np.dot(self.hidden_output, self.weights_hidden_output)
        self.predicted_output = sigmoid(self.output_input)
        return self.predicted_output

    def backward(self, inputs, target, learning_rate):
        # Backpropagation
        error = target - self.predicted_output
        delta_output = error * sigmoid_derivative(self.predicted_output)

        error_hidden = delta_output.dot(self.weights_hidden_output.T)
        delta_hidden = error_hidden * sigmoid_derivative(self.hidden_output)

        # Update weights
        self.weights_hidden_output += np.outer(self.hidden_output, delta_output) * learning_rate
        self.weights_input_hidden += np.outer(inputs, delta_hidden) * learning_rate

    def train(self, training_data, targets, epochs, learning_rate):
        for epoch in range(epochs):
            for i in range(len(training_data)):
                inputs = training_data[i]
                target = targets[i]
                self.forward(inputs)
                self.backward(inputs, target, learning_rate)

    def predict(self, inputs):
        return self.forward(inputs)

# Define XOR dataset
training_data = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
targets = np.array([[0], [1], [1], [0]])

# Create and train the neural network
input_size = 2
hidden_size = 4
output_size = 1
learning_rate = 0.1
epochs = 10000

nn = NeuralNetwork(input_size, hidden_size, output_size)
nn.train(training_data, targets, epochs, learning_rate)

# Test the trained network
for i in range(len(training_data)):
    inputs = training_data[i]
    prediction = nn.predict(inputs)
    print(f"Input: {inputs}, Predicted Output: {prediction}")