debug

app.py

@@ -360,11 +360,27 @@ def init_leaderboard_mib_subgraph(dataframe, track):
     
     # First, create our display name mapping
     # This is like creating a translation dictionary between internal names and display names
+    model_name_mapping = {
+        "qwen2_5": "Qwen-2.5",
+        "gpt2": "GPT-2",
+        "gemma2": "Gemma-2",
+        "llama3": "LLaMA-3"
+    }
+
+    benchmark_mapping = {
+        "ioi": "IOI",
+        "mcqa": "MCQA",
+        "arithmetic_addition": "Arithmetic Addition",
+        "arithmetic_subtraction": "Arithmetic Subtraction",
+        "arc_easy": "ARC Easy",
+        "arc_challenge": "ARC Challenge"
+    }
+
     display_mapping = {}
     for task in TasksMib_Subgraph:
         for model in task.value.models:
             field_name = f"{task.value.benchmark}_{model}"
-            display_name = f"{task.value.benchmark}({model})"
+            display_name = f"{benchmark_mapping[task.value.benchmark]}({model_name_mapping[model]})"
             display_mapping[field_name] = display_name
 
     # Now when creating benchmark groups, we'll use display names
@@ -402,8 +418,6 @@ def init_leaderboard_mib_subgraph(dataframe, track):
     # Important: We need to rename our DataFrame columns to match display names
     
     renamed_df = dataframe.rename(columns=display_mapping)
-
-    renamed_df = dataframe
     all_columns = dataframe.columns.tolist()
 
     # Original code