fixRAG with deepseek

app.py

@@ -1,23 +1,32 @@
 import gradio as gr
-import random
-import os
 from huggingface_hub import InferenceClient
 
-
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
 def respond(message, history):
-    messages = [{"role": "system","content": "You are a freindly chatbot!"}]
-    if history: 
-        message.extend(history)   # will keep conversation going
-    message.append({"role": "user", "content": message})
+    messages = [{"role": "system", "content": "You are a friendly chatbot!"}]
+    
+    # 添加历史消息
+    for user_msg, bot_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": bot_msg})
+    
+    # 添加当前消息
+    messages.append({"role": "user", "content": message})
+    
+    # 调用API
     response = client.chat_completion(
-        messages
-        max_tokens=100 #limit characters in response to 100
+        messages,
+        max_tokens=100  # 限制响应为100个token
     )
+    
+    return response.choices[0].message.content.strip()
 
-    return response["choices"][0]["message"]["content"].strip()
-
-chatbot = gr.ChatInterface(respond, type = "messages")
+chatbot = gr.ChatInterface(
+    respond,
+    title="Zephyr-7b Chatbot",
+    description="A chatbot powered by Zephyr-7b-beta"
+)
 
+if __name__ == "__main__":
     chatbot.launch()