Update app.py

app.py

@@ -3,10 +3,8 @@ import gradio as gr
 import numpy as np
 import re
 
-# 初始化 OpenAI 客戶端
-client = OpenAI(api_key="OPENAIAPIKEY")
+client = OpenAI(api_key="你的OPENAI_API_KEY")
 
-# ============ 1. 定義主要專業領域 ============ #
 PROFESSIONS = {
     "程式設計": "你是一位資深程式設計師，回答必須專業、詳細，附上程式範例與步驟。",
     "行銷": "你是一位行銷專家，回答必須專業、詳細，提供可執行行銷策略與步驟。",
@@ -16,27 +14,24 @@ PROFESSIONS = {
     "設計": "你是一位設計師，回答必須專業、詳細，提供設計步驟與案例。"
 }
 
-# ============ 2. Embedding 工具 (UTF-8 安全 + 防呆) ============ #
+# Embedding 函數
 def get_embedding(text, model="text-embedding-3-small"):
+    if not isinstance(text, str):
+        text = str(text)
+    text = text.encode("utf-8", "ignore").decode("utf-8")
     try:
-        if not isinstance(text, str):
-            text = str(text)
-        text = text.encode("utf-8", "ignore").decode("utf-8")
         emb = client.embeddings.create(input=text, model=model)
         return np.array(emb.data[0].embedding)
-    except Exception as e:
-        print(f"[Embedding 錯誤] {e}")
-        return np.zeros(1536)  # fallback 避免程式 crash
+    except:
+        return np.zeros(1536)
 
-profession_embeddings = {}  # 延遲初始化
+profession_embeddings = {}
 
 def ensure_embeddings():
     global profession_embeddings
     if not profession_embeddings:
-        for field in PROFESSIONS.keys():
-            profession_embeddings[field] = get_embedding(field)
+        profession_embeddings = {field: get_embedding(field) for field in PROFESSIONS.keys()}
 
-# ============ 3. NLP 判斷職業 (自動比對) ============ #
 def detect_profession(user_input: str) -> str:
     ensure_embeddings()
     if not user_input.strip():
@@ -45,39 +40,26 @@ def detect_profession(user_input: str) -> str:
     detail_emb = get_embedding(user_input)
     scores = {}
     for field, emb in profession_embeddings.items():
-        if np.linalg.norm(detail_emb) == 0 or np.linalg.norm(emb) == 0:
+        if np.linalg.norm(detail_emb)==0 or np.linalg.norm(emb)==0:
             scores[field] = -1
         else:
             scores[field] = np.dot(detail_emb, emb) / (np.linalg.norm(detail_emb) * np.linalg.norm(emb))
-
     best_field = max(scores, key=scores.get)
     return PROFESSIONS[best_field]
 
-# ============ 4. AI Agent 回答邏輯 ============ #
+# AI 回答
 def professional_agent(user_input, state):
-    """
-    state 結構:
-    {
-        "chat_history": [(user, assistant), ...],
-        "profession_prompt": None or str
-    }
-    """
-
-    # 如果尚未設定專業 → 自動偵測
     if state.get("profession_prompt") is None:
         profession_prompt = detect_profession(user_input)
         state["profession_prompt"] = profession_prompt
-
-        # 嘗試把「問題部分」抽出來
         question = re.sub(r"我是.*?(，|,)", "", user_input)
         if not question.strip():
-            answer = f"✅ 已設定你的專業領域。請提出問題。"
+            answer = f"✅ 已設定你的專業領域，請提出問題。"
             state["chat_history"].append((user_input, answer))
             return answer, state["chat_history"], state
         else:
-            user_input = question  # 問題部分直接交給模型
+            user_input = question
 
-    # 正常對話
     messages = [{"role": "system", "content": state["profession_prompt"]}]
     for h in state["chat_history"]:
         messages.append({"role": "user", "content": h[0]})
@@ -94,13 +76,14 @@ def professional_agent(user_input, state):
     except Exception as e:
         answer = f"⚠️ 發生錯誤: {str(e)}"
 
+    # 確保每一筆都是 (user, assistant) tuple
     state["chat_history"].append((user_input, answer))
     if len(state["chat_history"]) > 10:
         state["chat_history"] = state["chat_history"][-10:]
 
     return answer, state["chat_history"], state
 
-# ============ 5. Gradio 介面 ============ #
+# Gradio 介面
 with gr.Blocks() as demo:
     gr.Markdown("## 🧑‍💼 全職業專業 AI 顧問 (NLP 智能判斷)")
     gr.Markdown("👉 第一次輸入時，可以同時輸入職業 + 問題，例如：`我是專業會計師，我想知道該如何進行台灣的稅務分析`")