{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "faec3801",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "openai.api_key = os.environ[\"OPENAI_API_KEY\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a1e3860e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import openai\n",
    "from langchain.llms import OpenAI\n",
    "from langchain.document_loaders import TextLoader\n",
    "from langchain.embeddings import OpenAIEmbeddings\n",
    "from langchain.text_splitter import RecursiveCharacterTextSplitter\n",
    "from langchain.vectorstores import Chroma\n",
    "import gradio as gr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2131f179",
   "metadata": {},
   "outputs": [],
   "source": [
    "loader = TextLoader(\"Machine Learning Introduction A Com.txt\", encoding='utf-8')\n",
    "document = loader.load()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "02ec4e57",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(document)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "89310192",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12177"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(document[0].page_content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8e4c3e72",
   "metadata": {},
   "outputs": [],
   "source": [
    "text_splitter = RecursiveCharacterTextSplitter(\n",
    "    chunk_size = 100,\n",
    "    chunk_overlap  = 20,\n",
    "    length_function = len,\n",
    "    add_start_index = True,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c552df57",
   "metadata": {},
   "outputs": [],
   "source": [
    "texts = text_splitter.split_documents(document)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1e109910",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "181"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(texts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4f84f608",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "52"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(texts[0].page_content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0c550bb5",
   "metadata": {},
   "outputs": [],
   "source": [
    "db = Chroma.from_documents(texts, OpenAIEmbeddings())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "461925e5",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "query = \"recommend machine learning courses\"\n",
    "docs = db.similarity_search(query)\n",
    "\n",
    "print(f\"Query: {query}\\n\")\n",
    "for index,answer in enumerate(docs):\n",
    "    print(f\"{index +1}: {answer.page_content}\")\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c8fb01e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def answer_question(fn_query):\n",
    "    docs = db.similarity_search(fn_query)\n",
    "\n",
    "#     for answer in docs:\n",
    "#         print(answer.page_content)\n",
    "#         print()\n",
    "\n",
    "    return \"\\n\".join([doc.page_content for doc in docs])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eb3b7131",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Create a Gradio interface\n",
    "interface = gr.Interface(\n",
    "    fn=answer_question,\n",
    "    inputs=\"text\",\n",
    "    outputs=\"text\",\n",
    ")\n",
    "\n",
    "interface.launch()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}