save plot figure

test/test_model.py

@@ -3,10 +3,30 @@ import pandas as pd
 from sklearn.linear_model import LinearRegression
 from sklearn.metrics import mean_absolute_error, r2_score
 import re  # For using regular expressions
+import matplotlib.pyplot as plt
+import datetime
 
 train_path = "../Datasets_all/A2_dataset_80.csv"
 test_path = "../Datasets_all/A2_dataset_20.csv"
 
+def save_prediction_plot(y_test, y_test_pred_baseline, baseline_test_r2):
+    # Visualize baseline predictions
+    fig, axes = plt.subplots(figsize=(5, 5))
+
+    # Actual vs Predicted
+    axes.scatter(y_test, y_test_pred_baseline, alpha=0.5)
+    axes.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], 'r--', lw=2)
+    axes.set_xlabel('Actual AimoScore')
+    axes.set_ylabel('Predicted AimoScore')
+    axes.set_title(f'Baseline: Actual vs Predicted (R²={baseline_test_r2:.4f})')
+    axes.grid(True, alpha=0.3)
+
+    # Save the figure
+    timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")   # e.g., 20260130_143210
+    fig_path = f"baseline_actual_vs_predicted_{timestamp}.png"
+    plt.savefig(fig_path, dpi=300, bbox_inches='tight')
+    print(f"Figure saved to {fig_path}")
+
 def extract_missing_feature(error_message):
     # Use regex to find feature names in the ValueError message
     match = re.search(r"Feature names unseen at fit time:\s*-\s*(.+)", error_message)
@@ -39,8 +59,8 @@ def load_and_evaluate_model(model_path):
     X_test = test_df[features_cols]
     y_test = test_df[target_col]
 
-    #define y_pred
-    y_pred = 0
+    #define y_pred and r2
+    y_pred, r2 = 0, 0
 
     # Continue to predict until no ValueErrors occur
     while True:
@@ -81,6 +101,10 @@ def load_and_evaluate_model(model_path):
         # Save predictions to CSV
         test_df["Predicted_AimoScore"] = y_pred
         test_df.to_csv("predicted_test.csv", index=False)
+    else:
+        print("no predictions!!!")
+
+    save_prediction_plot(y_test, y_pred, r2)
 
 
 if __name__ == "__main__":