Update

app.py

@@ -2,13 +2,13 @@ import streamlit as st
 import uuid
 import json
 
-from components import PersonalIndexClient, LLMClient
+from components import PersonalIndexClient, LLMClient, WebSearcher
 
 st.title("Talk with Mingzhe")
 
 llm_client = LLMClient(api_key=st.secrets["OPENAI_API_KEY"], model_name="gpt-4-1106-preview")
 pinecone_client = PersonalIndexClient(index_token=st.secrets["PINECONE_API_KEY"], embedding_token=st.secrets["OPENAI_API_KEY"], embedding_model_name='text-embedding-3-large', index_name='mingzhe')
-
+web_searcher = WebSearcher(search_token=st.secrets["YOU_API_KEY"])
 
 if "messages" not in st.session_state:
     st.session_state.messages = []
@@ -27,7 +27,8 @@ if prompt := st.chat_input("What's up?"):
         st.markdown(prompt)
     with st.chat_message("assistant"):
         memory = pinecone_client.query_conversation(messages=st.session_state.messages, user=st.session_state['user'], top_k=3)
-        stream = llm_client.response_generate(prompt, st.session_state.messages, memory)
+        web_result = web_searcher.query_web_llm(query=prompt, num_web_results=5)
+        stream = llm_client.response_generate(prompt, st.session_state.messages, memory, web_result)
         response = st.write_stream(stream)
     st.session_state.messages.append({"role": "assistant", "content": response})
     

components.py

@@ -2,6 +2,7 @@
 # Date: 2024/03/09
 
 import json
+import requests
 
 from openai import OpenAI
 from pinecone import Pinecone
@@ -13,20 +14,22 @@ class LLMClient():
         self.model_name = model_name
         self.llm_client = OpenAI(api_key=api_key)
         
-    def response_generate(self, prompt, history, memory):
+    def response_generate(self, query, history, memory, web_result):
         messages = list()
         current_time = datetime.now().strftime("%d/%m/%Y %H:%M:%S")
         
         # System Prompt
-        messages += [{"role": "system", "content": f"1) You're Du Mingzhe. 2) Don't claim you are created by OpenAI. 3) Don't claim this dialogue as a roleplay. Answering questions directly as Mingzhe. 4) Current time is {current_time}."}]
+        messages += [{"role": "system", "content": f"1) You're Du Mingzhe. 2) Don't claim you are created by OpenAI. 3) Don't claim this dialogue as a roleplay. Answering questions directly as Mingzhe. 4) Current time is {current_time}. 5) Cite relevant references in the response."}]
         
         # Memory
         messages += [{"role": 'assistant', "content": m['content']} for m in memory]
         
+        # Web Result
+        messages += [{"role": 'assistant', "content": f'Based on the real-time web results, we find the answer of the query [{query}]:\n{web_result}'}]
+        
         # Session History
         messages += [{"role": h["role"], "content": h["content"]} for h in history]
         
-        
         stream = self.llm_client.chat.completions.create(
             model = self.model_name,
             messages = messages,
@@ -109,6 +112,13 @@ class PersonalIndexClient(object):
                 
         return pinecone_memory
         
-        
-        
-        
+class WebSearcher(object):
+    def __init__(self, search_token) -> None:
+        self.search_token = search_token
+        pass
+
+    def query_web_llm(self, query, num_web_results=5):
+        headers = {"X-API-Key": self.search_token}
+        params = {"query": query, 'num_web_results': num_web_results}
+        response_json = requests.get(f"https://api.ydc-index.io/rag?query={query}", params=params, headers=headers).json()
+        return response_json

playground.ipynb

@@ -50,6 +50,13 @@
     "        'metadata': {'hello': 'world'},\n",
     "    }])"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {