Update app.py

app.py

@@ -7,23 +7,23 @@ from langchain_community.vectorstores import Chroma
 from langchain_openai import ChatOpenAI, OpenAIEmbeddings
 from dotenv import load_dotenv
 
-# 加載環境變量
+# Load environment variables
 load_dotenv()
 os.environ["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY")
 
-# 驗證 OpenAI API Key
+# Validate OpenAI API Key
 api_key = os.getenv('OPENAI_API_KEY')
 if not api_key:
-    raise ValueError("請設置 'OPENAI_API_KEY' 環境變數")
+    raise ValueError("Please set the 'OPENAI_API_KEY' environment variable")
 
 # OpenAI API key
 openai_api_key = api_key
 
-# 將聊天歷史轉換為適合 LangChain 的二元組格式
+# Transform chat history for LangChain format
 def transform_history_for_langchain(history):
-    return [(chat[0], chat[1]) for chat in history if chat[0]]  # 使用整數索引來訪問元組中的元素
+    return [(chat[0], chat[1]) for chat in history if chat[0]]
 
-# 將 Gradio 的歷史紀錄轉換為 OpenAI 格式
+# Transform chat history for OpenAI format
 def transform_history_for_openai(history):
     new_history = []
     for chat in history:
@@ -33,7 +33,7 @@ def transform_history_for_openai(history):
             new_history.append({"role": "assistant", "content": chat[1]})
     return new_history
 
-# 載入和處理文件的函數
+# Load and process documents function
 def load_and_process_documents(folder_path):
     documents = []
     for file in os.listdir(folder_path):
@@ -58,24 +58,24 @@ def load_and_process_documents(folder_path):
     )
     return vectordb
 
-# 初始化向量數據庫為全局變量
+# Initialize vector database as a global variable
 if 'vectordb' not in globals():
     vectordb = load_and_process_documents("./")
 
-# 定義查詢處理函數
+# Define query handling function for RAG
 def handle_query(user_message, temperature, chat_history):
     try:
         if not user_message:
-            return chat_history  # 返回不變的聊天記錄
+            return chat_history  # Return unchanged chat history
 
-        # 使用 LangChain 的 ConversationalRetrievalChain 處理查詢
+        # Use LangChain's ConversationalRetrievalChain to handle the query
         preface = """
-指令: 全部以繁體中文呈現，200字以內。
-除了與文件相關內容可回答之外，與文件內容不相關的問題都必須回答：這問題很深奧,需要請示JohnLiao大神...
+Instruction: Answer in Traditional Chinese, within 200 characters.
+If the question is unrelated to the documents, respond with: 此事無可奉告,話說這件事須請教海虔王...
 """
-        query = f"{preface} 查詢內容：{user_message}"
+        query = f"{preface} Query content: {user_message}"
 
-        # 提取之前的回答作為上下文，並轉換成 LangChain 支持的格式
+        # Extract previous answers as context, converting them to LangChain format
         previous_answers = transform_history_for_langchain(chat_history)
 
         pdf_qa = ConversationalRetrievalChain.from_llm(
@@ -85,54 +85,54 @@ def handle_query(user_message, temperature, chat_history):
             verbose=False
         )
 
-        # 調用模型進行查詢
+        # Invoke the model to handle the query
         result = pdf_qa.invoke({"question": query, "chat_history": previous_answers})
 
-        # 確保 'answer' 在結果中
+        # Ensure 'answer' is present in the result
         if "answer" not in result:
-            return chat_history + [("系統", "抱歉，出現了一個錯誤。")]
+            return chat_history + [("System", "Sorry, an error occurred.")]
 
-        # 更新對話歷史中的 AI 回應
-        chat_history[-1] = (user_message, result["answer"])  # 更新最後一個記錄，配對用戶輸入和 AI 回應
+        # Update the AI response in chat history
+        chat_history[-1] = (user_message, result["answer"])  # Update the last record, pairing user input with AI response
 
         return chat_history
 
     except Exception as e:
-        return chat_history + [("系統", f"出現錯誤: {str(e)}")]
+        return chat_history + [("System", f"An error occurred: {str(e)}")]
 
-# 使用 Gradio 的 Blocks API 創建自訂聊天介面
+# Create a custom chat interface using Gradio Blocks API
 with gr.Blocks() as demo:
-    gr.Markdown("<h1 style='text-align: center;'>AI 小助教</h1>")
+    gr.Markdown("<h1 style='text-align: center;'>AI Assistant for AI Forum</h1>")
 
     chatbot = gr.Chatbot()
     state = gr.State([])
 
     with gr.Row():
         with gr.Column(scale=0.85):
-            txt = gr.Textbox(show_label=False, placeholder="請輸入您的問題...")
+            txt = gr.Textbox(show_label=False, placeholder="Please enter your question...")
         with gr.Column(scale=0.15, min_width=0):
-            submit_btn = gr.Button("提問")
+            submit_btn = gr.Button("Ask")
 
-    # 用戶輸入後立即顯示提問文字，不添加回應部分，並清空輸入框
+    # Immediately show user input without response part, and clear input box
     def user_input(user_message, history):
-        history.append((user_message, ""))  # 顯示提問文字，回應部分為空字符串
-        return history, "", history  # 返回清空的輸入框以及更新的聊天歷史
+        history.append((user_message, ""))  # Show user message, response part as empty string
+        return history, "", history  # Return cleared input box and updated chat history
 
-    # 處理 AI 回應，更新回應部分
+    # Handle AI response, update response part
     def bot_response(history):
-        user_message = history[-1][0]  # 獲取最新的用戶輸入
-        history = handle_query(user_message, 0.7, history)  # 調用處理函數
-        return history, history  # 返回更新後的聊天記錄
+        user_message = history[-1][0]  # Get the latest user input
+        history = handle_query(user_message, 0.7, history)  # Call the query handler
+        return history, history  # Return updated chat history
 
-    # 先顯示提問文字，然後處理 AI 回應，並清空輸入框
+    # First show user message, then handle AI response, clear input box
     submit_btn.click(user_input, [txt, state], [chatbot, txt, state], queue=False).then(
         bot_response, state, [chatbot, state]
     )
 
-    # 支援按 "Enter" 提交問題，立即顯示提問文字並清空輸入框
+    # Support pressing "Enter" to submit question, immediately show user input, clear input box
     txt.submit(user_input, [txt, state], [chatbot, txt, state], queue=False).then(
         bot_response, state, [chatbot, state]
     )
 
-# 啟動 Gradio 應用
-demo.launch()
+# Launch Gradio app
+demo.launch()