fix

src/leaderboard/read_evals.py

@@ -205,12 +205,12 @@ class EvalResult:
 
         baselines = {task.value.benchmark: task.value.baseline*100 for task in Tasks}
 
-        average_old = sum([v for task, v in self.results.items() if v is not None and task in all_tasks_wo_polqa]) / len(all_tasks_wo_polqa)
+        # average_old = sum([v for task, v in self.results.items() if v is not None and task in all_tasks_wo_polqa]) / len(all_tasks_wo_polqa)
 
         average = sum([(self.results.get(task,0) - baselines.get(task, 0)) / (100 - baselines.get(task, 0)) * 100 for task in all_tasks]) / len(all_tasks)
-        average_g = sum([(self.results.get(task,0) - baselines.get(task, 0)) / (100 - baselines.get(task, 0)) * 100 for task in g_tasks]) / len(g_tasks)
-        average_mc = sum([(self.results.get(task,0) - baselines.get(task, 0)) / (100 - baselines.get(task, 0)) * 100 for task in mc_tasks]) / len(mc_tasks)
-        average_rag = sum([(self.results.get(task,0) - baselines.get(task, 0)) / (100 - baselines.get(task, 0)) * 100 for task in rag_tasks]) / len(rag_tasks)
+        # average_g = sum([(self.results.get(task,0) - baselines.get(task, 0)) / (100 - baselines.get(task, 0)) * 100 for task in g_tasks]) / len(g_tasks)
+        # average_mc = sum([(self.results.get(task,0) - baselines.get(task, 0)) / (100 - baselines.get(task, 0)) * 100 for task in mc_tasks]) / len(mc_tasks)
+        # average_rag = sum([(self.results.get(task,0) - baselines.get(task, 0)) / (100 - baselines.get(task, 0)) * 100 for task in rag_tasks]) / len(rag_tasks)
 
         data_dict = {}
         # data_dict = {