debug

app.py

@@ -413,13 +413,13 @@ def init_leaderboard_mib_subgraph(dataframe, track):
     # Create selection groups
     benchmark_selections = {
         # For each benchmark, store which columns should be shown
-        benchmark: [col for col in result_columns if col.startswith(f"{benchmark}(")]
+        benchmark: [col for col in result_columns if col.startswith(f"{benchmark}_")]
         for benchmark in benchmarks
     }
     
     model_selections = {
         # For each model, store which columns should be shown
-        model: [col for col in result_columns if col.endswith(f"({model})")]
+        model: [col for col in result_columns if col.startswith(f"_{model}")]
         for model in models
     }
     

src/display/utils.py

@@ -132,7 +132,7 @@ auto_eval_column_dict_mib_subgraph.append(
 for task in TasksMib_Subgraph:
     for model in task.value.models:
         # Create a valid field name by using underscores
-        field_name = f"{task.value.benchmark}-{model}"
+        field_name = f"{task.value.benchmark}_{model}"
         # The display name can be more descriptive
         display_name = f"{task.value.benchmark}({model})"