Untitled

mail@pastecode.io avatar
unknown
plain_text
20 days ago
2.8 kB
1
Indexable
Never
import numpy as np

class NeuralNetwork:
    def __init__(self, input_size, hidden_size, output_size):
        self.input_size = input_size
        self.hidden_size = hidden_size
        self.output_size = output_size
        
        # Initialize weights and biases
        self.W1 = np.random.randn(self.input_size, self.hidden_size)
        self.b1 = np.zeros((1, self.hidden_size))
        self.W2 = np.random.randn(self.hidden_size, self.output_size)
        self.b2 = np.zeros((1, self.output_size))

    def forward(self, X):
        # Forward propagation
        self.z1 = np.dot(X, self.W1) + self.b1
        self.a1 = self.sigmoid(self.z1)
        self.z2 = np.dot(self.a1, self.W2) + self.b2
        return self.z2

    def backward(self, X, y, learning_rate=0.01):
        # Forward propagation
        predictions = self.forward(X)

        # Backpropagation
        N = len(X)
        K = self.output_size
        for i in range(N):
            delta3 = predictions[i] - y[i]
            dW2 = np.outer(self.a1[i], delta3)
            db2 = delta3
            delta2 = np.dot(delta3, self.W2.T) * (self.a1[i] * (1 - self.a1[i]))
            dW1 = np.outer(X[i], delta2)
            db1 = delta2

            # Update weights and biases
            self.W1 -= learning_rate * dW1
            self.b1 -= learning_rate * db1
            self.W2 -= learning_rate * dW2
            self.b2 -= learning_rate * db2


    def sigmoid(self, x):
        return 1 / (1 + np.exp(-x))

    def custom_cost_function(self, X, y):
        # Compute custom cost function
        predictions = self.forward(X)
        N = len(X)
        K = self.output_size
        cost = 0.0
        for i in range(N):
            for k in range(K):
                cost += 0.5 * (predictions[i][k] - y[i][k]) ** 2
        return cost
    
    def accuracy(self, X, y):
        # Make predictions
        predictions = np.argmax(self.forward(X), axis=1)
        # Compare predictions with true labels
        correct = np.sum(predictions == y)
        # Calculate accuracy
        accuracy = correct / len(y)
        return accuracy

# Example usage:
# Create an instance of the NeuralNetwork class
input_size = 3
hidden_size = 4
output_size = 2
model = NeuralNetwork(input_size, hidden_size, output_size)

# Example data
X = np.random.randn(5, input_size)  # 5 samples, 3 features
y = np.random.randint(0, 2, size=(5, output_size))  # 5 samples, 2 output values

# Training loop
num_epochs = 1000
learning_rate = 0.01
for epoch in range(num_epochs):
    # Forward pass and backward pass
    model.backward(X, y, learning_rate)
    
    # Print the loss
    if epoch % 100 == 0:
        loss = model.custom_cost_function(X, y)
        print(f'Epoch {epoch}, Loss: {loss}')

accuracy = model.accuracy(X, y)
print(f'Test Accuracy: {accuracy}')
Leave a Comment