good

main.py

@@ -31,7 +31,7 @@ X_train, X_test, y_train, y_test = train_test_split(
 date_train, date_test = train_test_split(date_index, test_size=0.02, shuffle=False)
 
 # Step 4: Train model with gradient descent
-theta, losses = train_linear_regression_gd(X_train, y_train, lr=0.005, n_epochs=300)
+theta, losses = train_linear_regression_gd(X_train, y_train, lr=0.005, n_epochs=10000)
 predictions = predict_linear_regression(X_test, theta)
 
 # Step 5: Evaluate
@@ -47,7 +47,12 @@ residuals = y_test - predictions
 errors = abs(residuals)
 
 plot_predictions(date_test, y_test_rescaled, predictions_rescaled)
-plot_residuals(date_test, residuals)
+print("date_test shape:", date_test.shape)
+print("residuals shape:", residuals.shape)
+print(date_test[:5])
+print(date_test[-5:])
+
+plot_residuals(date_test.to_numpy(), residuals)
 plot_error_hist(errors)
 plot_scatter(y_test.reshape(-1, 1), predictions.reshape(-1, 1))
 plot_loss_curve(losses)

model_train.py

@@ -1,7 +1,7 @@
 import numpy as np
 
 
-def train_linear_regression_gd(X_train, y_train, lr=0.005, n_epochs=300):
+def train_linear_regression_gd(X_train, y_train, lr=0.01, n_epochs=1000):
     """
     Train linear regression model using batch gradient descent.
 
@@ -18,15 +18,15 @@ def train_linear_regression_gd(X_train, y_train, lr=0.005, n_epochs=300):
         predictions = X_b @ theta
         errors = predictions - y_train
         gradients = (2 / m) * X_b.T @ errors
-        gradients = np.clip(gradients, -1, 1)  # gradient clipping to prevent divergence
+        gradients = np.clip(gradients, -1, 1)  # gradient clipping
         theta -= lr * gradients
         loss = np.mean(errors ** 2)
         losses.append(loss)
 
-        # Debug print every 10 epochs
-        if epoch % 10 == 0:
+        if epoch % 500 == 0:
             print(f"Epoch {epoch}: Loss = {loss:.6f}")
 
+    print(f"Final Loss: {losses[-1]:.6f}")
     return theta, losses
 
 

visualize.py

@@ -20,6 +20,7 @@ def plot_residuals(date_test, residuals):
     plt.figure(figsize=(10, 4))
     plt.plot(date_test, residuals, label="Residuals")
     plt.hlines(0, 0, len(residuals), colors="red")
+    plt.xlim(date_test.min(), date_test.max())
     plt.title("Residuals Over Time")
     plt.xlabel("Date")
     plt.ylabel("Error")