{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bde2a402",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "from langchain_community.vectorstores.azuresearch import AzureSearch\n",
    "from langchain_openai import AzureOpenAIEmbeddings, OpenAIEmbeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d6b66606",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Option 1: use an OpenAI account\n",
    "# openai_api_key: str = \"PLACEHOLDER FOR YOUR API KEY\"\n",
    "# openai_api_version: str = \"2023-05-15\"\n",
    "# model: str = \"text-embedding-ada-002\"\n",
    "\n",
    "import openai\n",
    "\n",
    "API_SECRET_KEY = \"sk-r0WeYOdkMjzYdnSxEcC8B931Aa904e4bBaCcAc2a57D803F1\"\n",
    "BASE_URL = \"https://svip.xty.app/v1\"\n",
    "os.environ[\"OPENAI_API_KEY\"] = API_SECRET_KEY\n",
    "os.environ[\"OPENAI_API_BASE\"] = BASE_URL\n",
    "openai.api_key = os.environ['OPENAI_API_KEY']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d272bf45",
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain.embeddings import HuggingFaceEmbeddings\n",
    "from langchain.vectorstores import Chroma\n",
    "from langchain.text_splitter import RecursiveCharacterTextSplitter\n",
    "from langchain.document_loaders import TextLoader\n",
    "from langchain.chains import RetrievalQA\n",
    "from langchain.llms import OpenAI\n",
    "import os\n",
    "\n",
    "# 设置 OpenAI API 密钥\n",
    "# os.environ[\"OPENAI_API_KEY\"] = \"your_openai_api_key\"\n",
    "\n",
    "# 加载文档\n",
    "loader = TextLoader(\"term.txt\")\n",
    "documents = loader.load()\n",
    "\n",
    "# 分割文档\n",
    "text_splitter = RecursiveCharacterTextSplitter(chunk_size=300, chunk_overlap=100)\n",
    "texts = text_splitter.split_documents(documents)\n",
    "\n",
    "# 使用 HuggingFace 的 sentence-transformers/all-mpnet-base-v2 作为 embedding 模型\n",
    "embeddings = HuggingFaceEmbeddings(model_name=\"sentence-transformers/all-mpnet-base-v2\")\n",
    "\n",
    "# 创建 Chroma 数据库\n",
    "db = Chroma.from_documents(texts, embeddings)\n",
    "\n",
    "# 使用 similarity_search_with_relevance_scores 进行检索\n",
    "query = \"特别说明：针对有些专家学者对《意见》中的两处疑问，编者专门联系《意见》起草者\"\n",
    "results = db.similarity_search_with_relevance_scores(query, k=5)\n",
    " \n",
    "# 打印检索结果\n",
    "for doc, score in results:\n",
    "    print(f\"Content: {doc.page_content}\")\n",
    "    print(f\"Relevance Score: {score}\")\n",
    "    print(\"-----\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b5eee4c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 使用 GPT-4 进行生成\n",
    "llm = OpenAI(model_name=\"gpt-4\")\n",
    "retriever = db.as_retriever(search_type=\"similarity\", search_kwargs={\"k\": 5})\n",
    "qa_chain = RetrievalQA.from_chain_type(llm=llm, chain_type=\"stuff\", retriever=retriever)\n",
    "\n",
    "# 提问并获取回答\n",
    "query = \"your_query\"\n",
    "result = qa_chain.run(query)\n",
    "print(f\"Answer: {result}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "77d1ffee",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}