downgrade gradio to working version + small nits

app.py

@@ -13,9 +13,9 @@ from huggingface_hub import HfApi
 
 # InfoStrings
 from scorer import question_scorer
-from content import format_warning, format_log, TITLE, INTRODUCTION_TEXT, CITATION_BUTTON_LABEL, CITATION_BUTTON_TEXT, model_hyperlink
+from content import format_error, format_warning, format_log, TITLE, INTRODUCTION_TEXT, CITATION_BUTTON_LABEL, CITATION_BUTTON_TEXT, model_hyperlink
 
-TOKEN = os.environ.get("TOKEN", None)
+TOKEN = os.environ.get("WTOKEN", None)
 
 OWNER="gaia-benchmark"
 DATA_DATASET = f"{OWNER}/GAIA"
@@ -109,7 +109,10 @@ def add_new_eval(
                     raise Exception("No model_answer key in the file provided")
                 answer = task["model_answer"]
                 task_id = task["task_id"]
-                level = int(gold_results[val_or_test][task_id]["Level"])
+                try:
+                    level = int(gold_results[val_or_test][task_id]["Level"])
+                except KeyError:
+                    return format_error(f"{task_id} not found in split {val_or_test}. Are you sure you submitted the correct file?")
 
                 score = question_scorer(task['model_answer'], gold_results[val_or_test][task_id]["Final answer"])
                 

content.py

@@ -30,6 +30,9 @@ CITATION_BUTTON_TEXT = r"""@misc{gaia, # TODO
 }"""
 
 
+def format_error(msg):
+    return f"<p style='color: red; font-size: 20px; text-align: center;'>{msg}</p>"
+
 def format_warning(msg):
     return f"<p style='color: orange; font-size: 20px; text-align: center;'>{msg}</p>"
 

requirements.txt

@@ -1,5 +1,5 @@
 datasets==2.14.5
-gradio==4.1.1
+gradio==3.43.2
 huggingface-hub==0.18.0
 numpy==1.24.2
 APScheduler==3.10.1