{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using deprecated Together LLM. Please use langchain_together instead.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import os \n",
    "from dotenv import load_dotenv\n",
    "\n",
    "from langchain_community.document_loaders.csv_loader import CSVLoader\n",
    "from langchain_community.embeddings import HuggingFaceEmbeddings\n",
    "from langchain_community.vectorstores import Vectara\n",
    "from langchain_core.output_parsers import StrOutputParser\n",
    "from langchain_core.prompts import ChatPromptTemplate\n",
    "from langchain_core.runnables import RunnablePassthrough\n",
    "\n",
    "try:\n",
    "    from langchain_together import Together\n",
    "except:\n",
    "    print('Using deprecated Together LLM. Please use langchain_together instead.')\n",
    "    from langchain_community.llms import Together\n",
    "\n",
    "MODEL_NAME = \"mistralai/Mixtral-8x7B-Instruct-v0.1\"\n",
    "\n",
    "load_dotenv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "vectara_customer_id = os.environ['VECTARA_CUSTOMER_ID']\n",
    "vectara_corpus_id = os.environ['VECTARA_CORPUS_ID']\n",
    "vectara_api_key = os.environ['VECTARA_API_KEY']\n",
    "\n",
    "embeddings = HuggingFaceEmbeddings(model_name=\"intfloat/multilingual-e5-large\")\n",
    "\n",
    "vectara = Vectara(vectara_customer_id=vectara_customer_id,\n",
    "                      vectara_corpus_id=vectara_corpus_id,\n",
    "                      vectara_api_key=vectara_api_key)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/salgadev/code/DocVerifyRAG/.venv/lib/python3.11/site-packages/langchain_core/_api/deprecation.py:117: LangChainDeprecationWarning: The class `langchain_community.llms.together.Together` was deprecated in langchain-community 0.0.12 and will be removed in 0.2. An updated version of the class exists in the langchain-together package and should be used instead. To use it run `pip install -U langchain-together` and import as `from langchain_together import Together`.\n",
      "  warn_deprecated(\n"
     ]
    }
   ],
   "source": [
    "model = Together(model=MODEL_NAME,\n",
    "        temperature=0.7,\n",
    "        max_tokens=256,\n",
    "        top_k=50,\n",
    "        together_api_key=os.environ[\"TOGETHER_API_KEY\"]\n",
    "        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_sources(documents):\n",
    "    return documents[:-1]\n",
    "\n",
    "def get_summary(documents):\n",
    "    return documents[-1].page_content"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain.prompts import PromptTemplate\n",
    "\n",
    "summary_config = {\"is_enabled\": True, \"max_results\": 3, \"response_lang\": \"eng\"}\n",
    "retriever = vectara.as_retriever(\n",
    "    search_kwargs={\"k\": 3, \"summary_config\": summary_config}\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "wrong_disc_meta = 'electrical_doc.pdf, Electrical wiring scheme and specifications for a generator room, S - Sanitaer'\n",
    "good_meta = 'ISB-020-U3-W-S-01-B18003-001-020.pdf, Schieber / Hawle / Schieber 4000 + Handrad 7800 DN100 Schutzraum, S - Sanitaer'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "template = \"\"\"\n",
    "passage: You are a helpful assistant that understands BIM building documents and engineering disciplines.\n",
    "passage: You will analyze BIM document metadata composed of filename, description, and discipline.\n",
    "passage: The metadata is written in German.\n",
    "passage: metadata: {metadata}\n",
    "query: Does the filename match other filenames within the same discipline?\n",
    "query: Does the description match the engineering discipline?\n",
    "query: How different is the metadata to your curated information?\n",
    "query: Highligh any discrepancies and comment on wether or not the metadata is anomalous.\n",
    "\"\"\"\n",
    "\n",
    "prompt = PromptTemplate(template=template, input_variables=[\"metadata\"])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\npassage: You are a helpful assistant that understands BIM building documents and engineering disciplines.\\npassage: You will analyze BIM document metadata composed of filename, description, and discipline.\\npassage: The metadata is written in German.\\npassage: metadata: ISB-020-U3-W-S-01-B18003-001-020.pdf, Schieber / Hawle / Schieber 4000 + Handrad 7800 DN100 Schutzraum, S - Sanitaer\\nquery: Does the filename match other filenames within the same discipline?\\nquery: Does the description match the engineering discipline?\\nquery: How different is the metadata to your curated information?\\nquery: Highligh any discrepancies and comment on wether or not the metadata is anomalous.\\n'"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "formatted_prompt = prompt.format(metadata=good_meta)\n",
    "formatted_prompt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Based on the provided BIM document metadata in German, the filename \"ISB-020-U3-W-S-01-B18003-001-020.pdf\" belongs to the discipline S - Sanitaer [2]. Comparing it to other filenames within the same discipline, there are similar filenames like \"ISB-020-U3-W-S-01-B17012-011-000\" and \"ISB-020-U3-W-S-01-B19009-001-020\" [3]. The description \"Schieber / Hawle / Schieber 4000 + Handrad 7800 DN100 Schutzraum\" corresponds to the engineering discipline S - Sanitaer [2]. The metadata displays a specific naming convention and content related to sanitary engineering, aligning with the discipline indicated [2]. No significant discrepancies were found in the metadata analyzed, suggesting that the provided metadata is consistent and not anomalous within the context of BIM building documents and engineering disciplines.'"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans = (retriever | get_summary).invoke(formatted_prompt)\n",
    "ans"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# metadata matches!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\npassage: You are a helpful assistant that understands BIM building documents and engineering disciplines.\\npassage: You will analyze BIM document metadata composed of filename, description, and discipline.\\npassage: The metadata is written in German.\\npassage: metadata: electrical_doc.pdf, Electrical wiring scheme and specifications for a generator room, S - Sanitaer\\nquery: Does the filename match other filenames within the same discipline?\\nquery: Does the description match the engineering discipline?\\nquery: How different is the metadata to your curated information?\\nquery: Highligh any discrepancies and comment on wether or not the metadata is anomalous.\\n'"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "formatted_prompt = prompt.format(metadata=wrong_disc_meta)\n",
    "formatted_prompt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Based on the provided search results, the filename \"electrical_doc.pdf\" matches other filenames within the same discipline of E - Elektroanlagen [7]. However, the description \"Electrical wiring scheme and specifications for a generator room\" aligns more with the discipline of electrical engineering rather than \"S - Sanitaer\" [7]. The metadata presents discrepancies as the description does not directly correspond to the discipline mentioned, indicating a mismatch [7]. This inconsistency suggests that the metadata could be considered anomalous due to the mismatch between the description and the specified discipline [7]. The metadata exhibits a clear discrepancy between the content of the file and the discipline it is categorized under, raising a question about the accuracy of the metadata classification.'"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans = (retriever | get_summary).invoke(formatted_prompt)\n",
    "ans"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Anomaly detected!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "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": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}