Update src/leaderboard/read_evals.py

src/leaderboard/read_evals.py

@@ -198,10 +198,11 @@ def get_request_file_for_model(requests_path, model_name, precision):
 def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResult]:
     """From the path of the results folder root, extract all needed info for results"""
     model_result_filepaths = []
-    print(f"Checking results in: {results_path}")
-    print(f"Model result filepaths found: {model_result_filepaths}")
     
-    for root, _, files in os.walk(results_path):
+    for root, dirs, files in os.walk(results_path):
+        print(f"Current directory: {root}")
+        print(f"Subdirectories: {dirs}")
+        print(f"Files: {files}")
         # We should only have json files in model results
         if len(files) == 0 or any([not f.endswith(".json") for f in files]):
             continue
@@ -214,15 +215,15 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
 
         for file in files:
             model_result_filepaths.append(os.path.join(root, file))
+            
+    print(f"Model result filepaths found: {model_result_filepaths}")
 
     eval_results = {}
     for model_result_filepath in model_result_filepaths:
         print(f"Processing result file: {model_result_filepath}")
         # Creation of result
-        print(f"Eval result for {model_result_filepath}: {eval_result}")
         eval_result = EvalResult.init_from_json_file(model_result_filepath)
         eval_result.update_with_request_file(requests_path)
-        print(f"Updated eval result with request file: {eval_result}")
 
         # Store results of same eval together
         eval_name = eval_result.eval_name
@@ -230,9 +231,15 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
             eval_results[eval_name].results.update({k: v for k, v in eval_result.results.items() if v is not None})
         else:
             eval_results[eval_name] = eval_result
+
+    # デバッグ: 最終的なデータ確認
+    if eval_result:
+        data_dict = eval_result.to_dict()
+        print(f"Final leaderboard entry: {data_dict}")
+    else:
+        print("No valid eval result was processed.")
     
     data_dict = eval_result.to_dict()
-    print(f"Final leaderboard entry: {data_dict}")
 
     results = []
     for v in eval_results.values():