Add back propagation code

code/back_propagation.py

@@ -0,0 +1,73 @@
+import numpy as np
+
+
+# Initialize the weights and bias
+def initialize_parameters():
+    weights = np.array([0.0])
+    bias = np.array([4.0])
+    return weights, bias
+
+
+# Define the mean squared error loss function
+def mse_loss(y_true, y_pred):
+    return np.mean((y_true - y_pred) ** 2)
+
+
+# Define the derivative of the mean squared error loss function
+def mse_loss_derivative(y_true, y_pred):
+    return -2 * (y_true - y_pred)
+
+
+# Perform forward propagation
+def forward_propagation(x, weights, bias):
+    return x * weights + bias
+
+
+# Perform backward propagation and update parameters
+def backward_propagation(x, y_true, y_pred, weights, bias, learning_rate):
+    dW = np.sum(mse_loss_derivative(y_true, y_pred) * x)
+    dB = np.sum(mse_loss_derivative(y_true, y_pred))
+
+    weights -= learning_rate * dW
+    bias -= learning_rate * dB
+
+    return weights, bias
+
+
+# Train the model using backpropagation
+def train(x_train, y_train, epochs, learning_rate):
+    weights, bias = initialize_parameters()
+
+    for epoch in range(1, epochs + 1):
+        y_pred = forward_propagation(x_train, weights, bias)
+        loss = mse_loss(y_train, y_pred)
+
+        weights, bias = backward_propagation(
+            x_train, y_train, y_pred, weights, bias, learning_rate
+        )
+
+        if epoch % 10 == 0:
+            print(f"Epoch {epoch}, Loss: {loss}, Weights: {weights}, Bias: {bias}")
+
+    return weights, bias
+
+
+# Test data
+x_data = np.array([[-1], [0], [1], [2], [3]])
+y_data = np.array([-3, -1, 1, 3, 5])
+
+# Hyperparameters
+learning_rate = 0.001
+epochs = 1000
+
+# Train the model
+weights, bias = train(x_data, y_data, epochs, learning_rate)
+
+print(f"Trained Weights: {weights}")
+print(f"Trained Bias: {bias}")
+
+# Predict new output
+x_test = np.array([[5], [6]])
+y_test = forward_propagation(x_test, weights, bias)
+
+print(f"Predictions: {y_test}")

docs/_toc.yml

@@ -24,7 +24,6 @@ parts:
       - file: tutorials/analyzing_ai_wiki_page
       - file: tutorials/hello_world_machine_learning
       - file: tutorials/house_price_prediction
-      - file: tutorials/iris_classification
       - file: tutorials/number_classification
       - file: tutorials/openai_api