Ran for all models

model_selection/test_models.py

@@ -250,31 +250,30 @@ if __name__ == "__main__":
             # Scores df
             scores_df = pd.DataFrame(index=models.keys(), columns=scorings.keys())
             # Create a figure for all models in this group-method
-            fig, axes = plt.subplots(len(models), 2, figsize=(12, 8 * len(models)))
+            fig, axes = plt.subplots(len(models), 2, figsize=(8, 8 * len(models)))
             if len(models) == 1:  # Adjustment if there's only one model (axes indexing issue)
                 axes = [axes]
             # Evaluate each model
             for model_idx, (model_name, model) in enumerate(models.items()):
                 # ----------- TEMPORAL -------------
-                if model_name == "DT" or model_name == "RF":
-                    # Train the model (it was just initialized above)
-                    model.fit(X_train, y_train)
-                    if hasattr(model, "decision_function"):
-                        y_score = model.decision_function(X_test)
-                    else:
-                        y_score = model.predict_proba(X_test)[:, 1]  # Use probability of positive class
-                    # Calculate ROC curve and ROC area for each class
-                    fpr, tpr, _ = roc_curve(y_test, y_score, pos_label=model.classes_[1])
-                    roc_display = RocCurveDisplay(fpr=fpr, tpr=tpr).plot(ax=axes[model_idx][0]) 
-                    # Calculate precision-recall curve
-                    precision, recall, _ = precision_recall_curve(y_test, y_score, pos_label=model.classes_[1])
-                    pr_display = PrecisionRecallDisplay(precision=precision, recall=recall).plot(ax=axes[model_idx][1])
-                    axes[model_idx][0].set_title(f'ROC Curve for {model_name}')
-                    axes[model_idx][1].set_title(f'PR Curve for {model_name}')
-                    # Evaluate at each of the scores of interest
-                    for score_name, scorer in scorings.items():
-                        score_value = scorer(model, X_test, y_test)
-                        scores_df.at[model_name, score_name] = score_value
+                # Train the model (it was just initialized above)
+                model.fit(X_train, y_train)
+                if hasattr(model, "decision_function"):
+                    y_score = model.decision_function(X_test)
+                else:
+                    y_score = model.predict_proba(X_test)[:, 1]  # Use probability of positive class
+                # Calculate ROC curve and ROC area for each class
+                fpr, tpr, _ = roc_curve(y_test, y_score, pos_label=model.classes_[1])
+                roc_display = RocCurveDisplay(fpr=fpr, tpr=tpr).plot(ax=axes[model_idx][0]) 
+                # Calculate precision-recall curve
+                precision, recall, _ = precision_recall_curve(y_test, y_score, pos_label=model.classes_[1])
+                pr_display = PrecisionRecallDisplay(precision=precision, recall=recall).plot(ax=axes[model_idx][1])
+                axes[model_idx][0].set_title(f'ROC Curve for {model_name}')
+                axes[model_idx][1].set_title(f'PR Curve for {model_name}')
+                # Evaluate at each of the scores of interest
+                for score_name, scorer in scorings.items():
+                    score_value = scorer(model, X_test, y_test)
+                    scores_df.at[model_name, score_name] = score_value
             # Adjust layout and save/show figure
             plt.tight_layout()
             plt.savefig(f'./test_results/roc_pr_curves/{group}_{method_names[j]}.svg', format='svg', dpi=500)