read evals

src/leaderboard/read_evals.py

@@ -159,8 +159,6 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
     model_result_filepaths = []
     print(results_path)
     for root, _, files in os.walk(results_path):
-        print(root)
-        print(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
@@ -173,12 +171,15 @@ 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(model_result_filepaths)
     eval_results = {}
     for model_result_filepath in model_result_filepaths:
+        print(f'creation of result {model_result_filepath}')
         # Creation of result
         eval_result = EvalResult.init_from_json_file(model_result_filepath)
+        print(eval_result)
         eval_result.update_with_request_file(requests_path)
+        print(eval_result)
 
         # Store results of same eval together
         eval_name = eval_result.eval_name
@@ -191,8 +192,10 @@ def get_raw_eval_results(results_path: str, requests_path: str) -> list[EvalResu
     for v in eval_results.values():
         try:
             v.to_dict() # we test if the dict version is complete
+            print(f'to dict value {str(v)}')
             results.append(v)
         except KeyError:  # not all eval values present
+            print('error key')
             continue
 
     return results