add log

src/leaderboard/read_evals.py

@@ -161,6 +161,7 @@ class EvalResult:
 
 def get_request_file_for_model(requests_path, model_name, precision):
     """Selects the correct request file for a given model. Only keeps runs tagged as FINISHED"""
+    print("get_request_file_for_model", requests_path, model_name)
     request_files = os.path.join(
         requests_path,
         f"{model_name}_eval_request_*.json",

src/tools/plots.py

@@ -19,7 +19,8 @@ def create_scores_df(raw_data: list[EvalResult]) -> pd.DataFrame:
     """
     # Step 1: Ensure 'date' is in datetime format and sort the DataFrame by it
     results_df = pd.DataFrame(raw_data)
-    #results_df["date"] = pd.to_datetime(results_df["date"], format="mixed", utc=True)
+    print("results_df", results_df)
+    results_df["date"] = pd.to_datetime(results_df["date"], format="mixed", utc=True)
     results_df.sort_values(by="date", inplace=True)
 
     # Step 2: Initialize the scores dictionary