update existing qas.json

app.py

@@ -24,10 +24,15 @@ def num_tokens_from_messages(messages, model="gpt-3.5-turbo"):
         encoding = tiktoken.get_encoding("cl100k_base")
     if model == "gpt-3.5-turbo":  # note: future models may deviate from this
         num_tokens = 0
+        len_values = 0
         for message in messages:
             num_tokens += 4  # every message follows <im_start>{role/name}\n{content}<im_end>\n
             for key, value in message.items():
-                num_tokens += len(encoding.encode(value))
+                try:
+                    num_tokens += len(encoding.encode(value))
+                except:
+                    num_tokens += int(num_tokens/len_values*len(value)) # linear estimation
+                len_values += len(value)
                 if key == "name":  # if there's a name, the role is omitted
                     num_tokens += -1  # role is always required and always 1 token
         num_tokens += 2  # every reply is primed with <im_start>assistant
@@ -95,6 +100,7 @@ def ask(query, max_attempt_times=3):
                 idx = openai.api_keys.index(openai.api_key)
                 idx = (idx + 1) % len(openai.api_keys)
                 openai.api_key = openai.api_keys[idx]
+                attempt_times -= 0.7
                 print(f"switch api_key")
             print(f"retry in {attempt_times*10} seconds...")
             time.sleep(attempt_times*10)