- minor.

src/4_run_evaluation.py

@@ -307,7 +307,7 @@ def enumerate_tasks(self, max_instance_count: int):
         input_format = "both"
         image_format = "full_page"
 
-        tasks = eval.load_task_names()
+        tasks = self.load_task_names()
 
         results = {}
         self.driver.get(TURKLE_URL)
@@ -347,7 +347,7 @@ def enumerate_tasks(self, max_instance_count: int):
                 input_names = [col.replace('Answer.', '') for col in df.columns if col.startswith('Answer.')]
                 inputs = Evaluation.extract_input_values_from_url(url, input_names)
 
-                print(" --> inputs: {}".format(inputs))
+                print(" --> inputs: {}".format([x.name for x in inputs]))
 
                 answers_map = self.retrieve_gold_labels(
                     task_name, row_number, [x.name for x in inputs]

src/tests.py

@@ -21,9 +21,8 @@ def test_evaluation():
     evaluation.enumerate_tasks(20)
 
 if __name__ == "__main__":
-
-    test_actions()
     test_evaluation()
+    test_actions()
 
     # TODO: test that we can apply the gold labels on the tasks