Fix

pycaret/internal/tabular.py

@@ -9875,30 +9875,32 @@ def get_leaderboard():
     """
     generates leaderboard for all models run in current run.
     """
-    model_container = get_config('master_model_container')
-    result_container = get_config('create_model_container')
-    
+    model_container = get_config("master_model_container")
+    result_container = get_config("create_model_container")
+
     result_container_mean = []
-
-    mc = 0
 
     for i in range(len(result_container)):
         model_results = result_container[i]
         mean_scores = model_results[-2:-1]
-        model_name = _get_model_name(model_container[mc]) 
-        mean_scores['Model Name'] = model_name
-        mean_scores['Model'] = model_container[mc]
-        mean_scores.set_index('Model Name', drop=True, inplace=True)
-        mean_scores = mean_scores[['Model', 'Accuracy', 'AUC', 'Recall', 'Prec.', 'F1', 'Kappa', 'MCC']]
+        model_name = _get_model_name(model_container[i])
+        mean_scores["Index"] = i
+        mean_scores["Model Name"] = model_name
+        mean_scores["Model"] = str(model_container[i])
+        rearranged_columns = list(mean_scores.columns)
+        rearranged_columns.remove("Model")
+        rearranged_columns.remove("Model Name")
+        rearranged_columns = ["Model Name", "Model"] + rearranged_columns
+        mean_scores = mean_scores[rearranged_columns]
         result_container_mean.append(mean_scores)
-        mc += 1
-
+
     results = pd.concat(result_container_mean)
-    
-    results.reset_index(inplace=True)
-    
+
+    results.set_index("Index", inplace=True, drop=True)
+
     return results
-
+
+
 def _choose_better(
     models_and_results: list,
     compare_dimension: str,