Update bot/chatgpt/chat_gpt_bot.py

bot/chatgpt/chat_gpt_bot.py

@@ -26,9 +26,9 @@ class ChatGPTBot(Bot):
             new_query = Session.build_session_query(query, from_user_id)
             logger.debug("[OPEN_AI] session query={}".format(new_query))
 
-            # if context.get('stream'):
-            #     # reply in stream
-            #     return self.reply_text_stream(query, new_query, from_user_id)
+            if context.get('stream'):
+                # reply in stream
+                return self.reply_text_stream(query, new_query, from_user_id)
 
             reply_content = self.reply_text(new_query, from_user_id, 0)
             logger.debug("[OPEN_AI] new_query={}, user={}, reply_cont={}".format(new_query, from_user_id, reply_content))
@@ -42,18 +42,16 @@ class ChatGPTBot(Bot):
     def reply_text(self, query, user_id, retry_count=0):
         try:
             response = openai.ChatCompletion.create(
-                model="gpt-4",  # 对话模型的名称
+                model="chinese-llama-alpaca-2",  # 对话模型的名称
                 messages=query,
                 temperature=1,  # 值在[0,1]之间，越大表示回复越具有不确定性
-                max_tokens=600,  # 回复最大的字符数
-                top_p=1,
-                frequency_penalty=0,  # [-2,2]之间，该值越大则更倾向于产生不同的内容
-                presence_penalty=0,  # [-2,2]之间，该值越大则更倾向于产生不同的内容
+                max_tokens=2048,  # 回复最大的字符数
+                top_p=1
             )
             # res_content = response.choices[0]['text'].strip().replace('<|endoftext|>', '')
-            logger.info(response.choices[0]['message']['content'])
+            logger.info(response)
             # log.info("[OPEN_AI] reply={}".format(res_content))
-            return response.choices[0]['message']['content']
+            return response
         except openai.error.RateLimitError as e:
             # rate limit exception
             logger.warn(e)