tutorials/04_FAQ_style_QA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Utilizing existing FAQs for Question Answering\n",
    "\n",
    "While *extractive Question Answering* works on pure texts and is therefore more generalizable, there's also a common alternative that utilizes existing FAQ data.\n",
    "\n",
    "**Pros**:\n",
    "\n",
    "- Very fast at inference time\n",
    "- Utilize existing FAQ data\n",
    "- Quite good control over answers\n",
    "\n",
    "**Cons**:\n",
    "\n",
    "- Generalizability: We can only answer questions that are similar to existing ones in FAQ\n",
    "\n",
    "In some use cases, a combination of extractive QA and FAQ-style can also be an interesting option."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "### Prepare environment\n",
    "\n",
    "#### Colab: Enable the GPU runtime\n",
    "Make sure you enable the GPU runtime to experience decent speed in this tutorial.\n",
    "**Runtime -> Change Runtime type -> Hardware accelerator -> GPU**\n",
    "\n",
    "<img src=\"https://github.com/deepset-ai/haystack-tutorials/raw/main/tutorials/img/colab_gpu_runtime.jpg\">\n",
    "\n",
    "You can double check whether the GPU runtime is enabled with the following command:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    },
    "vscode": {
     "languageId": "shellscript"
    }
   },
   "outputs": [],
   "source": [
    "%%bash\n",
    "\n",
    "nvidia-smi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To start, install the latest release of Haystack with `pip`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "shellscript"
    }
   },
   "outputs": [],
   "source": [
    "%%bash\n",
    "\n",
    "pip install --upgrade pip\n",
    "pip install git+https://github.com/deepset-ai/haystack.git#egg=farm-haystack[colab]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Logging\n",
    "\n",
    "We configure how logging messages should be displayed and which log level should be used before importing Haystack.\n",
    "Example log message:\n",
    "INFO - haystack.utils.preprocessing -  Converting data/tutorial1/218_Olenna_Tyrell.txt\n",
    "Default log level in basicConfig is WARNING so the explicit parameter is not necessary but can be changed easily:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import logging\n",
    "\n",
    "logging.basicConfig(format=\"%(levelname)s - %(name)s -  %(message)s\", level=logging.WARNING)\n",
    "logging.getLogger(\"haystack\").setLevel(logging.INFO)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Start an Elasticsearch server\n",
    "You can start Elasticsearch on your local machine instance using Docker. If Docker is not readily available in your environment (eg., in Colab notebooks), then you can manually download and execute Elasticsearch from source."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Recommended: Start Elasticsearch using Docker via the Haystack utility function\n",
    "from haystack.utils import launch_es\n",
    "\n",
    "launch_es()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Start an Elasticsearch server in Colab\n",
    "\n",
    "If Docker is not readily available in your environment (e.g. in Colab notebooks), then you can manually download and execute Elasticsearch from source."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "shellscript"
    }
   },
   "outputs": [],
   "source": [
    "%%bash\n",
    "\n",
    "wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-7.9.2-linux-x86_64.tar.gz -q\n",
    "tar -xzf elasticsearch-7.9.2-linux-x86_64.tar.gz\n",
    "chown -R daemon:daemon elasticsearch-7.9.2\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "shellscript"
    }
   },
   "outputs": [],
   "source": [
    "%%bash --bg\n",
    "\n",
    "sudo -u daemon -- elasticsearch-7.9.2/bin/elasticsearch"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "### Init the DocumentStore\n",
    "In contrast to Tutorial 1 (Build your first QA system), we:\n",
    "\n",
    "* specify the name of our `embedding_field` in Elasticsearch where we'll store the embedding of our question and that is used later for calculating our similarity to the incoming user question\n",
    "* set `excluded_meta_data=[\"question_emb\"]` so that we don't return the huge embedding vectors in our search results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import time\n",
    "\n",
    "from haystack.document_stores import ElasticsearchDocumentStore\n",
    "\n",
    "# Wait 30 seconds only to be sure Elasticsearch is ready before continuing\n",
    "time.sleep(30)\n",
    "\n",
    "# Get the host where Elasticsearch is running, default to localhost\n",
    "host = os.environ.get(\"ELASTICSEARCH_HOST\", \"localhost\")\n",
    "\n",
    "document_store = ElasticsearchDocumentStore(\n",
    "    host=host,\n",
    "    username=\"\",\n",
    "    password=\"\",\n",
    "    index=\"document\",\n",
    "    embedding_field=\"question_emb\",\n",
    "    embedding_dim=384,\n",
    "    excluded_meta_data=[\"question_emb\"],\n",
    "    similarity=\"cosine\",\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "### Create a Retriever using embeddings\n",
    "Instead of retrieving via Elasticsearch's plain BM25, we want to use vector similarity of the questions (user question vs. FAQ ones).\n",
    "We can use the `EmbeddingRetriever` for this purpose and specify a model that we use for the embeddings."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from haystack.nodes import EmbeddingRetriever\n",
    "\n",
    "retriever = EmbeddingRetriever(\n",
    "    document_store=document_store,\n",
    "    embedding_model=\"sentence-transformers/all-MiniLM-L6-v2\",\n",
    "    use_gpu=True,\n",
    "    scale_score=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "### Prepare & Index FAQ data\n",
    "We create a pandas dataframe containing some FAQ data (i.e curated pairs of question + answer) and index those in elasticsearch.\n",
    "Here: We download some question-answer pairs related to COVID-19"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "from haystack.utils import fetch_archive_from_http\n",
    "\n",
    "\n",
    "# Download\n",
    "doc_dir = \"data/tutorial4\"\n",
    "s3_url = \"https://s3.eu-central-1.amazonaws.com/deepset.ai-farm-qa/datasets/documents/small_faq_covid.csv.zip\"\n",
    "fetch_archive_from_http(url=s3_url, output_dir=doc_dir)\n",
    "\n",
    "# Get dataframe with columns \"question\", \"answer\" and some custom metadata\n",
    "df = pd.read_csv(f\"{doc_dir}/small_faq_covid.csv\")\n",
    "# Minimal cleaning\n",
    "df.fillna(value=\"\", inplace=True)\n",
    "df[\"question\"] = df[\"question\"].apply(lambda x: x.strip())\n",
    "print(df.head())\n",
    "\n",
    "# Get embeddings for our questions from the FAQs\n",
    "questions = list(df[\"question\"].values)\n",
    "df[\"question_emb\"] = retriever.embed_queries(queries=questions).tolist()\n",
    "df = df.rename(columns={\"question\": \"content\"})\n",
    "\n",
    "# Convert Dataframe to list of dicts and index them in our DocumentStore\n",
    "docs_to_index = df.to_dict(orient=\"records\")\n",
    "document_store.write_documents(docs_to_index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "### Ask questions\n",
    "Initialize a Pipeline (this time without a reader) and ask questions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from haystack.pipelines import FAQPipeline\n",
    "\n",
    "pipe = FAQPipeline(retriever=retriever)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from haystack.utils import print_answers\n",
    "\n",
    "prediction = pipe.run(query=\"How is the virus spreading?\", params={\"Retriever\": {\"top_k\": 10}})\n",
    "print_answers(prediction, details=\"medium\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "## About us\n",
    "\n",
    "This [Haystack](https://github.com/deepset-ai/haystack/) notebook was made with love by [deepset](https://deepset.ai/) in Berlin, Germany\n",
    "\n",
    "We bring NLP to the industry via open source!  \n",
    "Our focus: Industry specific language models & large scale QA systems.  \n",
    "  \n",
    "Some of our other work: \n",
    "- [German BERT](https://deepset.ai/german-bert)\n",
    "- [GermanQuAD and GermanDPR](https://deepset.ai/germanquad)\n",
    "- [FARM](https://github.com/deepset-ai/FARM)\n",
    "\n",
    "Get in touch:\n",
    "[Twitter](https://twitter.com/deepset_ai) | [LinkedIn](https://www.linkedin.com/company/deepset-ai/) | [Discord](https://haystack.deepset.ai/community/join) | [GitHub Discussions](https://github.com/deepset-ai/haystack/discussions) | [Website](https://deepset.ai)\n",
    "\n",
    "By the way: [we're hiring!](https://www.deepset.ai/jobs)"
   ]
  }
 ],
 "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": "3.10.6 (main, Aug 11 2022, 13:36:31) [Clang 13.1.6 (clang-1316.0.21.2.5)]"
  },
  "vscode": {
   "interpreter": {
    "hash": "bda33b16be7e844498c7c2d368d72665b4f1d165582b9547ed22a0249a29ca2e"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}