add debugging print statements and temp remove json.loads

app.py

@@ -97,7 +97,9 @@ def model_call(question, document, answer):
         prompt=NEW_FORMAT
     )
     print("RESPONSE FROM CLIENT:", response)
-    generated_text = json.loads(response.choices[0].text)
+    generated_text = response.choices[0].text
+    print("GENERATED TEXT", generated_text)
+    print("type of GENERATED TEXT", type(generated_text))
     reasoning = generated_text["REASONING"][0]
     score = generated_text["SCORE"]
     # inputs = tokenizer(NEW_FORMAT, return_tensors="pt")