Update app.py

app.py

@@ -34,8 +34,8 @@ class SearchResult(BaseModel):
     score: float
 
 # 搜尋＋rerank API
-@app.post("/search", response_model=List[SearchResult])
-def search(input: QueryInput):
+@app.post("/chat", response_model=List[SearchResult])
+def search_and_generate(input: QueryInput):
     reranked = retrieve_and_rerank(input.query, model_name, collection_name, cross_encoder_model, score_threshold=0.5, search_top_k=20, rerank_top_k=input.top_k)
     
     # 如果沒有找到相關答案，則返回 404 錯誤
@@ -51,22 +51,55 @@ def search(input: QueryInput):
         raise HTTPException(status_code=404, detail="無法生成回答，請檢查輸入或模型設定。")
     return answer
 
-demo = gr.ChatInterface(
-    answer,
+# demo = gr.ChatInterface(
+#     answer,
+#     additional_inputs=[
+#         gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+#         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+#         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+#         gr.Slider(
+#             minimum=0.1,
+#             maximum=1.0,
+#             value=0.95,
+#             step=0.05,
+#             label="Top-p (nucleus sampling)",
+#         ),
+#     ],
+# )
+
+
+# if __name__ == "__main__":
+#     demo.launch()
+
+import gradio as gr
+
+# def respond(message, history, system_message, max_tokens, temperature, top_p):
+#     try:
+#         llm.temperature = temperature
+#         llm.max_output_tokens = max_tokens
+
+#         search_results = hybrid_search(message)
+
+#         rerank_response = rerank_chunks_with_llm(message, search_results, llm, top_n=3)
+#         reranked_indices = [int(i.strip()) - 1 for i in rerank_response.content.split(",") if i.strip().isdigit()]
+#         reranked_docs = [search_results[i] for i in reranked_indices]
+
+#         answer = generate_answer_with_rag(message, reranked_docs, llm)
+#         return answer.content
+
+#     except Exception as e:
+#         return f"[錯誤] {str(e)}"
+
+chat_interface = gr.ChatInterface(
+    fn=search_and_generate,
+    title="Chatbot問答系統",
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
+        gr.Textbox(value="你是個樂於助人的AI助手。", label="System Message"),
+        gr.Slider(1, 2048, value=512, step=1, label="Max tokens"),
+        gr.Slider(0.1, 2.0, value=0.7, step=0.1, label="Temperature"),
+        gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-p"),
     ],
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    chat_interface.launch()