Upload app.py

app.py

@@ -4,6 +4,8 @@
 import os
 from dotenv import load_dotenv
 load_dotenv()
+import gradio as gr
+from openai import OpenAI
 from langchain_community.utils import user_agent
 from langchain_groq import ChatGroq
 from langchain.text_splitter import RecursiveCharacterTextSplitter
@@ -13,12 +15,10 @@ from langchain.chains import RetrievalQA
 from langchain_community.document_loaders import WebBaseLoader, TextLoader
 from langchain.prompts import PromptTemplate
 from langchain.schema import Document
-import gradio as gr
-from openai import OpenAI
 import resend
+import requests
 import re
 import time
-import requests
 
 def load_documents(sources):
     documents = []
@@ -58,7 +58,7 @@ sources = [
 ]
 
 documents = load_documents(sources)
-print(f"\n成功載入 {len(documents)} 個網址或檔案")
+print(f"\n成功載入 {len(documents)} 個檔案")
 
 text_splitter = RecursiveCharacterTextSplitter(
     chunk_size=512,
@@ -72,7 +72,7 @@ split_docs = text_splitter.split_documents(documents)
 print(f"分割後的文件數量：{len(split_docs)}")
 
 embeddings = HuggingFaceEmbeddings(model_name="intfloat/multilingual-e5-large")
-print(f"\n成功初始化 Microsoft ���入型")
+print(f"\n成功初始化 Microsoft 嵌入模型")
 
 print(f"\n開始建立向量資料庫")
 vectorstore = Chroma.from_documents(split_docs, embeddings, persist_directory="./Knowledge-base")
@@ -207,7 +207,11 @@ def fetch_law_summary(tax_law, keywords):
     url = "https://ttc.mof.gov.tw/Api/GetData"
     headers = {
         "Content-Type": "application/x-www-form-urlencoded; charset=UTF-8",
-        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/130.0.0.0 Safari/537.36"
+        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/130.0.0.0 Safari/537.36",
+        "accept": "application/json, text/javascript, */*; q=0.01",
+        "accept-encoding": "gzip, deflate, br, zstd",
+        "accept-language": "zh-TW,zh;q=0.9,en-US;q=0.8,en;q=0.7",
+        "referer": "https://ttc.mof.gov.tw/"
     }
     gr.Info("檢索法令彙編函釋中......")
     version_payload = {
@@ -263,7 +267,7 @@ def fetch_law_summary(tax_law, keywords):
         except Exception as e:
             print(f"檢索關鍵字 '{keyword}' 的法令彙編函釋時發生錯誤：{str(e)}")
     if all_results:
-        summary = f"<h3>相關法令彙編函釋檢索結果（{latest_version}）：</h3>"
+        summary = f"<h3>相關法令彙編函釋檢索結果（{tax_law} {latest_version}）：</h3>"
         unique_results = {}
         for result in all_results:
             tax_sn = result.get('TaxSN', '')
@@ -288,7 +292,7 @@ def llm_openai_api(query, answer):
     user_prompt = f"""
     「題目：{query}
     答案：{answer}」
-    請詳細分析答案內容後，依據與題目相關性最高的稅目名稱及最多3個重點關鍵字回應我，問題與答案中的稅目名稱列入TaxName，關鍵字列入KeyWord，只須根據格式回應，不要寫其他的。
+    請詳細分析答案內容後，依據與題目相關性最高的稅目名稱及最多3個重點關鍵字回應我，提供的3個重點關鍵字不能與稅目名稱相同，問題與答案中的稅目名稱列入TaxName，關鍵字列入KeyWord，只須根據格式回應，不要寫其他的。
 
     # 回應字典格式範例：
     {{"TaxName": "地價稅", "KeyWord": "宿舍用地,醫護人員"}}
@@ -323,8 +327,8 @@ def handle_interaction(query, api_key, state):
             "html": f"<strong>查詢內容：<br>{query}</strong>",
         }
         try:
-            email = resend.Emails.send(params)
-            print(f"Email sent successfully. ID：{email.id}")
+            email_response = resend.Emails.send(params)
+            print(f"Email sent successfully. Response：{email_response}")
         except Exception as e:
             print(f"Failed to send email：{str(e)}")
         api_key = os.getenv("YOUR_API_KEY")
@@ -448,7 +452,7 @@ custom_css = """
 with gr.Blocks(theme=gr.themes.Soft(), css=custom_css) as iface:
     gr.Markdown("""
     # 地方稅知識庫系統 - 財政部財政資訊中心
-    > ### **※ RAG-based 系統部署：江信宗，以地方稅極少知識資料示範，僅供參考，準確資訊請依地方稅稽徵機關回覆為準，LLM：Llama-3.1-70B。**
+    > ### **※ RAG-based KM 以地方稅極少知識資料作示範，僅供參考，準確資訊請依地方稅稽徵機關回覆為準。系統部署：江信宗，LLM：Llama-3.1-70B。**
     """, elem_classes="center-text")
     with gr.Row():
         query_input = gr.Textbox(label="輸入您的問題，系統將基於學習到的知識資料提供相關答案。", placeholder="請輸入您的問題（支援同時輸入多個問題，例如：問題1？問題2？）", autofocus=True, scale=3, max_lines=5, elem_classes="query-input")