workshop-rag-eval-qdrant-ragas/notebook/naive_rag_eval_qdrant_ragas-with-dspy.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "io4-9iTZrfgp"
   },
   "source": [
    "# **Evaluating a RAG Application using Qdrant and RAGAS using DSPy**\n",
    "\n",
    "**Introduction**\n",
    "\n",
    "Creating a foundation for a Retrieval-Augmented Generation (RAG) pipeline is often straightforward. However, optimizing and tweaking it for production and ensuring top-notch response quality presents formidable hurdles. Given the multitude of tools and parameters available, identifying the ideal configuration can feel overwhelming.\n",
    "\n",
    "In this notebook, we delve into the evaluation, visualization, and analysis of RAG systems using a blend of open-source libraries. \n",
    "Our toolkit includes [RAGAS](https://docs.ragas.io/) for evaluation and Qdrant for context augmentation, serving as a vector database to construct the RAG pipeline using [DSPy](https://github.com/stanfordnlp/dspy).\n",
    "\n",
    "Throughout this article, we'll leverage data sourced from [Qdrant](https://qdrant.tech/)'s documentation and articles to create an assistant capable of answering queries about Qdrant. Additionally, we'll explore methods for generating synthetic test data provided by RAGAS.\n",
    "\n",
    "❗This notebook requires **OpenAI Key**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tbY_IXyvvSoz"
   },
   "source": [
    "### **1. Import relevant packages**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "_A1rBXAowQ22"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import openai\n",
    "import pandas as pd\n",
    "import qdrant_client\n",
    "from getpass import getpass\n",
    "from dotenv import load_dotenv\n",
    "from tqdm.notebook import tqdm\n",
    "from datasets import load_dataset\n",
    "from typing import Optional, List, Tuple\n",
    "from qdrant_client import QdrantClient, models\n",
    "from qdrant_client.http.models import PointStruct\n",
    "from langchain_core.language_models import BaseChatModel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "load_dotenv()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QSUcsgP-xLQr"
   },
   "source": [
    "### **2. Setup your openai key**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "q1uaUw9VxKSa",
    "outputId": "6be05076-6548-4bc7-fede-fe254024ba30"
   },
   "outputs": [],
   "source": [
    "if not (openai_api_key := os.environ.get(\"OPENAI_API_KEY\")):\n",
    "    openai_api_key = getpass(\"🔑 Enter your OpenAI API key: \")\n",
    "openai.api_key = openai_api_key\n",
    "os.environ[\"OPENAI_API_KEY\"] = openai_api_key"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vbjLlIPYxYgU"
   },
   "source": [
    "### **3.  Retrieve the documents / dataset to be used**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 237,
     "referenced_widgets": [
      "582420861a1e451ca96a6dc69846c257",
      "99788ffbf8d74daf8930873d8405c08a",
      "4c1d3cbfefad4f5aa6b9f08f0422957c",
      "fcdcbbf40e3e44c9ad96986d44ab6fce",
      "611c5f880a834188a7b272c03dd3f991",
      "b264a13b05a947cd8d8aad15a313f5db",
      "bf73dde5eb1d40f5a607128605932bd1",
      "d1db86f2576443f5835f133056568b39",
      "549e4961bb7c47c8a148e45da7b52f7e",
      "3071076a711c4db19459b6142fcfc8d0",
      "ee0f3a23c1924342b1501facdba8902f",
      "100d7c563ce846c8a18915559e66239d",
      "124305eadc9b4effabfe59052ffc4813",
      "d2c3d4000c594a429ea12b7bcca280d1",
      "7cedf7369e9f4927893cddbeffd1b145",
      "cc6cc585fdf94f92bca17d83bcc63de4",
      "818b70c8a7b2429a90c135c37c25a241",
      "46ad56ca51004315a7ce5077c8b2d8c8",
      "75e31b30ef79414ab96c0ec406f1f4b9",
      "039d179c5821401c9d0b5caa056d805c",
      "e6ababdab8fc48679bd2f8163ef61724",
      "4beae8d8f5cc48a2bb7e9b7bebfb8e6f",
      "1ebdac4f164442049cecd2592e118f2d",
      "b322ee25201747b2aac1ef9cb4f51818",
      "6cbc14541229488ab452cd547778c88c",
      "1d22fe0dfd73421e9bb5881320f8b4fb",
      "a3e3bfcf40ae464c929bff322b14d45f",
      "79618b906a264ec0b5b2238a448d818e",
      "6cd56494d8ff44709188731fd3d71b4d",
      "888a4a15cae1407fb5912c4aa9823e33",
      "00ad85e3dae347e09d272d3537b881fa",
      "dc04d32cfc724cb6ac3dd926041badce",
      "4596d5f1c4e4433ea000a6a8b8cca946"
     ]
    },
    "id": "hZl3eaEC1O1m",
    "outputId": "d86f3374-3763-495e-aaf5-e77593364e67"
   },
   "outputs": [],
   "source": [
    "\n",
    "dataset = load_dataset(\"atitaarora/qdrant_doc\", split=\"train\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vlbBeHMD41I7"
   },
   "source": [
    "### **4. Process dataset as langchain document for further processing**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 66,
     "referenced_widgets": [
      "64fb95e3aaa7409dbd55c1474b108fef",
      "17acda449aa544218c3db7c03c7789c9",
      "26cc261df479483b92af2650a2db0fef",
      "14947bce04ee442996215a1cbc009cbe",
      "72b32c83b30a49bba9eb99c612ccfe2b",
      "dd2274085a1d415c9ab4dbf90813af2c",
      "9a24f140106141d6b7c7a85c5a2bcbdd",
      "97b1448064c64d829544c8d4cf4522de",
      "0e1729146be54c5395cf5e2dcd34e156",
      "0812081f8c2048c6ba57fa709acd657e",
      "0c7a6ae7d8c54e39b0f436a96525bbc4"
     ]
    },
    "id": "TXpn1xZF4_y6",
    "outputId": "f8cf5d92-e833-48a4-b558-5d28b82ac2c6"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6a285f22459a4885bc2f29ae2f477b21",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/240 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "240"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from langchain.text_splitter import RecursiveCharacterTextSplitter\n",
    "from langchain.docstore.document import Document as LangchainDocument\n",
    "\n",
    "## Dataset to langchain document\n",
    "langchain_docs = [\n",
    "    LangchainDocument(page_content=doc[\"text\"], metadata={\"source\": doc[\"source\"]})\n",
    "    for doc in tqdm(dataset)\n",
    "]\n",
    "\n",
    "len(langchain_docs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "6UJcYaJRDzxQ"
   },
   "source": [
    "### **5. Test Evaluation Dataset Generation using Ragas**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from ragas.testset.generator import TestsetGenerator\n",
    "from ragas.testset.evolutions import simple, reasoning, multi_context\n",
    "from langchain_openai import ChatOpenAI, OpenAIEmbeddings\n",
    "\n",
    "# generator with openai models\n",
    "generator_llm = ChatOpenAI(model=\"gpt-3.5-turbo-16k\")\n",
    "critic_llm = ChatOpenAI(model=\"gpt-4\")\n",
    "embeddings = OpenAIEmbeddings()\n",
    "\n",
    "generator = TestsetGenerator.from_langchain(generator_llm , critic_llm , embeddings)\n",
    "\n",
    "# generate testset\n",
    "testset = generator.generate_with_langchain_docs(langchain_docs[:10], test_size=10, \n",
    "                                                 distributions={simple: 0.5, reasoning: 0.25, multi_context: 0.25})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MA0zLkDbGzgV"
   },
   "source": [
    "### **5a. Evaluation dataset to Pandas for better visualization**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 537
    },
    "id": "YUe7Z6haGwJu",
    "outputId": "7dbbdf99-3891-4e55-999c-3270fb2757e0"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question</th>\n",
       "      <th>contexts</th>\n",
       "      <th>ground_truth</th>\n",
       "      <th>evolution_type</th>\n",
       "      <th>metadata</th>\n",
       "      <th>episode_done</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>How do you load data into a Qdrant collection?</td>\n",
       "      <td>[---\\n\\ntitle: Quickstart\\n\\nweight: 11\\n\\nali...</td>\n",
       "      <td>To load data into a Qdrant collection, you can...</td>\n",
       "      <td>simple</td>\n",
       "      <td>[{'source': 'documentation/quick-start.md'}]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>What is the purpose of the query vector in the...</td>\n",
       "      <td>[;\\n\\n\\n\\nvar operationInfo = await client.Ups...</td>\n",
       "      <td>nan</td>\n",
       "      <td>simple</td>\n",
       "      <td>[{'source': 'documentation/quick-start.md'}, {...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>How does the multi-language model help in cate...</td>\n",
       "      <td>[---\\n\\ndraft: false\\n\\ntitle: E-commerce prod...</td>\n",
       "      <td>nan</td>\n",
       "      <td>simple</td>\n",
       "      <td>[{'source': 'demo/demo-3.md'}]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>What is the purpose of the Food Discovery Demo...</td>\n",
       "      <td>[---\\n\\ndraft: false\\n\\ntitle: Food Discovery\\...</td>\n",
       "      <td>This demo uses data from Delivery Service. Use...</td>\n",
       "      <td>simple</td>\n",
       "      <td>[{'source': 'demo/demo-2.md'}]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>What is the Basic RAG pipeline and how does it...</td>\n",
       "      <td>[ for Skin Conditions](https://colab.research....</td>\n",
       "      <td>Basic RAG pipeline is a pipeline that incorpor...</td>\n",
       "      <td>simple</td>\n",
       "      <td>[{'source': 'documentation/examples.md'}]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>What does the \"limit\" parameter do in the quer...</td>\n",
       "      <td>[;\\n\\n\\n\\nvar operationInfo = await client.Ups...</td>\n",
       "      <td>nan</td>\n",
       "      <td>reasoning</td>\n",
       "      <td>[{'source': 'documentation/quick-start.md'}, {...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>How does the Qdrant Startup Search demo perfor...</td>\n",
       "      <td>[---\\n\\ndraft: false\\n\\ntitle: Startup Search\\...</td>\n",
       "      <td>This demo uses short descriptions of startups ...</td>\n",
       "      <td>reasoning</td>\n",
       "      <td>[{'source': 'demo/demo-1.md'}]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>How does the Qdrant Startup Search demo use sh...</td>\n",
       "      <td>[---\\n\\ndraft: false\\n\\ntitle: Startup Search\\...</td>\n",
       "      <td>This demo uses short descriptions of startups ...</td>\n",
       "      <td>multi_context</td>\n",
       "      <td>[{'source': 'demo/demo-1.md'}]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>What community resources are available for bui...</td>\n",
       "      <td>[---\\n\\ntitle: Community links\\n\\nweight: 42\\n...</td>\n",
       "      <td>The community resources available for building...</td>\n",
       "      <td>multi_context</td>\n",
       "      <td>[{'source': 'documentation/community-links.md'}]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>How does the multi-language model help categor...</td>\n",
       "      <td>[---\\n\\ndraft: false\\n\\ntitle: E-commerce prod...</td>\n",
       "      <td>nan</td>\n",
       "      <td>multi_context</td>\n",
       "      <td>[{'source': 'demo/demo-3.md'}]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            question  \\\n",
       "0     How do you load data into a Qdrant collection?   \n",
       "1  What is the purpose of the query vector in the...   \n",
       "2  How does the multi-language model help in cate...   \n",
       "3  What is the purpose of the Food Discovery Demo...   \n",
       "4  What is the Basic RAG pipeline and how does it...   \n",
       "5  What does the \"limit\" parameter do in the quer...   \n",
       "6  How does the Qdrant Startup Search demo perfor...   \n",
       "7  How does the Qdrant Startup Search demo use sh...   \n",
       "8  What community resources are available for bui...   \n",
       "9  How does the multi-language model help categor...   \n",
       "\n",
       "                                            contexts  \\\n",
       "0  [---\\n\\ntitle: Quickstart\\n\\nweight: 11\\n\\nali...   \n",
       "1  [;\\n\\n\\n\\nvar operationInfo = await client.Ups...   \n",
       "2  [---\\n\\ndraft: false\\n\\ntitle: E-commerce prod...   \n",
       "3  [---\\n\\ndraft: false\\n\\ntitle: Food Discovery\\...   \n",
       "4  [ for Skin Conditions](https://colab.research....   \n",
       "5  [;\\n\\n\\n\\nvar operationInfo = await client.Ups...   \n",
       "6  [---\\n\\ndraft: false\\n\\ntitle: Startup Search\\...   \n",
       "7  [---\\n\\ndraft: false\\n\\ntitle: Startup Search\\...   \n",
       "8  [---\\n\\ntitle: Community links\\n\\nweight: 42\\n...   \n",
       "9  [---\\n\\ndraft: false\\n\\ntitle: E-commerce prod...   \n",
       "\n",
       "                                        ground_truth evolution_type  \\\n",
       "0  To load data into a Qdrant collection, you can...         simple   \n",
       "1                                                nan         simple   \n",
       "2                                                nan         simple   \n",
       "3  This demo uses data from Delivery Service. Use...         simple   \n",
       "4  Basic RAG pipeline is a pipeline that incorpor...         simple   \n",
       "5                                                nan      reasoning   \n",
       "6  This demo uses short descriptions of startups ...      reasoning   \n",
       "7  This demo uses short descriptions of startups ...  multi_context   \n",
       "8  The community resources available for building...  multi_context   \n",
       "9                                                nan  multi_context   \n",
       "\n",
       "                                            metadata  episode_done  \n",
       "0       [{'source': 'documentation/quick-start.md'}]          True  \n",
       "1  [{'source': 'documentation/quick-start.md'}, {...          True  \n",
       "2                     [{'source': 'demo/demo-3.md'}]          True  \n",
       "3                     [{'source': 'demo/demo-2.md'}]          True  \n",
       "4          [{'source': 'documentation/examples.md'}]          True  \n",
       "5  [{'source': 'documentation/quick-start.md'}, {...          True  \n",
       "6                     [{'source': 'demo/demo-1.md'}]          True  \n",
       "7                     [{'source': 'demo/demo-1.md'}]          True  \n",
       "8   [{'source': 'documentation/community-links.md'}]          True  \n",
       "9                     [{'source': 'demo/demo-3.md'}]          True  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = testset.to_pandas()\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WsLXXncpHRz9"
   },
   "source": [
    "### **5b. Export the evaluation dataset to csv (if needed)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "p9xfsy1BHZv-"
   },
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "filepath = Path('sample_question_ragas_101.csv')\n",
    "filepath.parent.mkdir(parents=True, exist_ok=True)\n",
    "df.to_csv(filepath)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WDXhb870F_Du"
   },
   "source": [
    "### **6. Document chunk processing**\n",
    "\n",
    "Processing each document with desired **TEXT_SPLITTER_ALGO , CHUNK_SIZE , CHUNK_OVERLAP** etc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "OLWqwoszGI3z"
   },
   "outputs": [],
   "source": [
    "text_splitter = RecursiveCharacterTextSplitter(\n",
    "    chunk_size=512,\n",
    "    chunk_overlap=50,\n",
    "    add_start_index=True,\n",
    "    separators=[\"\\n\\n\", \"\\n\", \".\", \" \", \"\"],\n",
    ")\n",
    "\n",
    "docs_processed = []\n",
    "for doc in langchain_docs:\n",
    "    docs_processed += text_splitter.split_documents([doc])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9JSfRcxOHzL7"
   },
   "source": [
    "### **7. Setting up Qdrant**\n",
    "\n",
    "Set up the qdrant client and then create a collection so that our documents can be stored."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "Ymq23aPCIA7j"
   },
   "outputs": [],
   "source": [
    "\n",
    "##Uncomment to initialise qdrant client in memory\n",
    "#client = qdrant_client.QdrantClient(\n",
    "#    location=\":memory:\",\n",
    "#)\n",
    "\n",
    "##Uncomment below to connect to Qdrant Cloud\n",
    "client = qdrant_client.QdrantClient(\n",
    "    os.environ.get(\"QDRANT_URL\"),\n",
    "    api_key=os.environ.get(\"QDRANT_API_KEY\"),\n",
    ")\n",
    "\n",
    "## Uncomment below to connect to local Qdrant\n",
    "#client = qdrant_client.QdrantClient(\"http://localhost:6333\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "xLLsfOsgJRJ3"
   },
   "outputs": [],
   "source": [
    "## Collection name that will be used throughtout in the notebook\n",
    "COLLECTION_NAME = \"qdrant-docs-ragas\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "FNduakyeInzn",
    "outputId": "650fedf0-f27e-4875-f18c-b06853520cfb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CollectionsResponse(collections=[CollectionDescription(name='qdrant_docs_arize_hybrid'), CollectionDescription(name='experiment_512_64_all-MiniLM-L6-v2'), CollectionDescription(name='qdrant-docs-ragas-haystack-mxbai'), CollectionDescription(name='experiment_1024_128_bge-small-en'), CollectionDescription(name='qdrant_docs_arize_dense'), CollectionDescription(name='qdrant-docs-ragas'), CollectionDescription(name='email'), CollectionDescription(name='qdrant-docs-ragas-haystack-mxbai-livestream'), CollectionDescription(name='qdrant-docs-semantic-chunked'), CollectionDescription(name='hybrid_experiment_512_64_bge-small-en'), CollectionDescription(name='experiment_512_64_bge-small-en')])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## General Collection level operations\n",
    "\n",
    "## Get information about existing collections\n",
    "client.get_collections()\n",
    "\n",
    "## Get information about specific collection\n",
    "#collection_info = client.get_collection(COLLECTION_NAME)\n",
    "#print(collection_info)\n",
    "\n",
    "## Deleting collection , if need be\n",
    "#client.delete_collection(COLLECTION_NAME)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 494
    },
    "id": "u3ZeplTaLUhD",
    "outputId": "fd910f09-a726-47db-81d3-32753bbf7411"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[32m2024-07-04 14:02:35.403\u001b[0m | \u001b[33m\u001b[1mWARNING \u001b[0m | \u001b[36mfastembed.embedding\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m7\u001b[0m - \u001b[33m\u001b[1mDefaultEmbedding, FlagEmbedding, JinaEmbedding are deprecated.Use from fastembed import TextEmbedding instead.\u001b[0m\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>model</th>\n",
       "      <th>dim</th>\n",
       "      <th>description</th>\n",
       "      <th>size_in_GB</th>\n",
       "      <th>sources</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BAAI/bge-base-en</td>\n",
       "      <td>768</td>\n",
       "      <td>Base English model</td>\n",
       "      <td>0.420</td>\n",
       "      <td>{'url': 'https://storage.googleapis.com/qdrant...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BAAI/bge-base-en-v1.5</td>\n",
       "      <td>768</td>\n",
       "      <td>Base English model, v1.5</td>\n",
       "      <td>0.210</td>\n",
       "      <td>{'url': 'https://storage.googleapis.com/qdrant...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BAAI/bge-large-en-v1.5</td>\n",
       "      <td>1024</td>\n",
       "      <td>Large English model, v1.5</td>\n",
       "      <td>1.200</td>\n",
       "      <td>{'hf': 'qdrant/bge-large-en-v1.5-onnx'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BAAI/bge-small-en</td>\n",
       "      <td>384</td>\n",
       "      <td>Fast English model</td>\n",
       "      <td>0.130</td>\n",
       "      <td>{'url': 'https://storage.googleapis.com/qdrant...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BAAI/bge-small-en-v1.5</td>\n",
       "      <td>384</td>\n",
       "      <td>Fast and Default English model</td>\n",
       "      <td>0.067</td>\n",
       "      <td>{'hf': 'qdrant/bge-small-en-v1.5-onnx-q'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>BAAI/bge-small-zh-v1.5</td>\n",
       "      <td>512</td>\n",
       "      <td>Fast and recommended Chinese model</td>\n",
       "      <td>0.090</td>\n",
       "      <td>{'url': 'https://storage.googleapis.com/qdrant...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>sentence-transformers/all-MiniLM-L6-v2</td>\n",
       "      <td>384</td>\n",
       "      <td>Sentence Transformer model, MiniLM-L6-v2</td>\n",
       "      <td>0.090</td>\n",
       "      <td>{'url': 'https://storage.googleapis.com/qdrant...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>sentence-transformers/paraphrase-multilingual-...</td>\n",
       "      <td>384</td>\n",
       "      <td>Sentence Transformer model, paraphrase-multili...</td>\n",
       "      <td>0.220</td>\n",
       "      <td>{'hf': 'qdrant/paraphrase-multilingual-MiniLM-...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>nomic-ai/nomic-embed-text-v1</td>\n",
       "      <td>768</td>\n",
       "      <td>8192 context length english model</td>\n",
       "      <td>0.520</td>\n",
       "      <td>{'hf': 'nomic-ai/nomic-embed-text-v1'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>nomic-ai/nomic-embed-text-v1.5</td>\n",
       "      <td>768</td>\n",
       "      <td>8192 context length english model</td>\n",
       "      <td>0.520</td>\n",
       "      <td>{'hf': 'nomic-ai/nomic-embed-text-v1.5'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>thenlper/gte-large</td>\n",
       "      <td>1024</td>\n",
       "      <td>Large general text embeddings model</td>\n",
       "      <td>1.200</td>\n",
       "      <td>{'hf': 'qdrant/gte-large-onnx'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>mixedbread-ai/mxbai-embed-large-v1</td>\n",
       "      <td>1024</td>\n",
       "      <td>MixedBread Base sentence embedding model, does...</td>\n",
       "      <td>0.640</td>\n",
       "      <td>{'hf': 'mixedbread-ai/mxbai-embed-large-v1'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>intfloat/multilingual-e5-large</td>\n",
       "      <td>1024</td>\n",
       "      <td>Multilingual model, e5-large. Recommend using ...</td>\n",
       "      <td>2.240</td>\n",
       "      <td>{'url': 'https://storage.googleapis.com/qdrant...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>sentence-transformers/paraphrase-multilingual-...</td>\n",
       "      <td>768</td>\n",
       "      <td>Sentence-transformers model for tasks like clu...</td>\n",
       "      <td>1.000</td>\n",
       "      <td>{'hf': 'xenova/paraphrase-multilingual-mpnet-b...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>jinaai/jina-embeddings-v2-base-en</td>\n",
       "      <td>768</td>\n",
       "      <td>English embedding model supporting 8192 sequen...</td>\n",
       "      <td>0.520</td>\n",
       "      <td>{'hf': 'xenova/jina-embeddings-v2-base-en'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>jinaai/jina-embeddings-v2-small-en</td>\n",
       "      <td>512</td>\n",
       "      <td>English embedding model supporting 8192 sequen...</td>\n",
       "      <td>0.120</td>\n",
       "      <td>{'hf': 'xenova/jina-embeddings-v2-small-en'}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                model   dim  \\\n",
       "0                                    BAAI/bge-base-en   768   \n",
       "1                               BAAI/bge-base-en-v1.5   768   \n",
       "2                              BAAI/bge-large-en-v1.5  1024   \n",
       "3                                   BAAI/bge-small-en   384   \n",
       "4                              BAAI/bge-small-en-v1.5   384   \n",
       "5                              BAAI/bge-small-zh-v1.5   512   \n",
       "6              sentence-transformers/all-MiniLM-L6-v2   384   \n",
       "7   sentence-transformers/paraphrase-multilingual-...   384   \n",
       "8                        nomic-ai/nomic-embed-text-v1   768   \n",
       "9                      nomic-ai/nomic-embed-text-v1.5   768   \n",
       "10                                 thenlper/gte-large  1024   \n",
       "11                 mixedbread-ai/mxbai-embed-large-v1  1024   \n",
       "12                     intfloat/multilingual-e5-large  1024   \n",
       "13  sentence-transformers/paraphrase-multilingual-...   768   \n",
       "14                  jinaai/jina-embeddings-v2-base-en   768   \n",
       "15                 jinaai/jina-embeddings-v2-small-en   512   \n",
       "\n",
       "                                          description  size_in_GB  \\\n",
       "0                                  Base English model       0.420   \n",
       "1                            Base English model, v1.5       0.210   \n",
       "2                           Large English model, v1.5       1.200   \n",
       "3                                  Fast English model       0.130   \n",
       "4                      Fast and Default English model       0.067   \n",
       "5                  Fast and recommended Chinese model       0.090   \n",
       "6            Sentence Transformer model, MiniLM-L6-v2       0.090   \n",
       "7   Sentence Transformer model, paraphrase-multili...       0.220   \n",
       "8                   8192 context length english model       0.520   \n",
       "9                   8192 context length english model       0.520   \n",
       "10                Large general text embeddings model       1.200   \n",
       "11  MixedBread Base sentence embedding model, does...       0.640   \n",
       "12  Multilingual model, e5-large. Recommend using ...       2.240   \n",
       "13  Sentence-transformers model for tasks like clu...       1.000   \n",
       "14  English embedding model supporting 8192 sequen...       0.520   \n",
       "15  English embedding model supporting 8192 sequen...       0.120   \n",
       "\n",
       "                                              sources  \n",
       "0   {'url': 'https://storage.googleapis.com/qdrant...  \n",
       "1   {'url': 'https://storage.googleapis.com/qdrant...  \n",
       "2             {'hf': 'qdrant/bge-large-en-v1.5-onnx'}  \n",
       "3   {'url': 'https://storage.googleapis.com/qdrant...  \n",
       "4           {'hf': 'qdrant/bge-small-en-v1.5-onnx-q'}  \n",
       "5   {'url': 'https://storage.googleapis.com/qdrant...  \n",
       "6   {'url': 'https://storage.googleapis.com/qdrant...  \n",
       "7   {'hf': 'qdrant/paraphrase-multilingual-MiniLM-...  \n",
       "8              {'hf': 'nomic-ai/nomic-embed-text-v1'}  \n",
       "9            {'hf': 'nomic-ai/nomic-embed-text-v1.5'}  \n",
       "10                    {'hf': 'qdrant/gte-large-onnx'}  \n",
       "11       {'hf': 'mixedbread-ai/mxbai-embed-large-v1'}  \n",
       "12  {'url': 'https://storage.googleapis.com/qdrant...  \n",
       "13  {'hf': 'xenova/paraphrase-multilingual-mpnet-b...  \n",
       "14        {'hf': 'xenova/jina-embeddings-v2-base-en'}  \n",
       "15       {'hf': 'xenova/jina-embeddings-v2-small-en'}  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Declaring the intended Embedding Model with Fastembed\n",
    "from fastembed.embedding import TextEmbedding\n",
    "\n",
    "pd.DataFrame(TextEmbedding.list_supported_models())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 355,
     "referenced_widgets": [
      "5349d3555ff942b0ac7f169462262428",
      "2ba7118900894716b3bf13ade4fdf8df",
      "51c0415940ac4b63824bf337117f12b2",
      "f9eae2e334e943038c03218e25875699",
      "84323279be93459a9922ec18e82ab181",
      "e3448574566a44a7af6eaf656aaf04a2",
      "8f12bf7b629347f29c6c5dd18299a9f7",
      "1956ebc39de944c9ab28e11c642d7407",
      "99a56eb9c2fe44749b622021cb8d8452",
      "b3ae8239ded842349db8a7b29c2a4834",
      "eb899fde8d554b65b15f08a054307e19",
      "b33229c70a5843dea4d11d863614b973",
      "616f2b95148142b3a09aaba104876760",
      "244321954b7d4d52bb1a3e7f29c7269a",
      "121d7781e13448e8b0f4ac3c810f2ae4",
      "eb54968170da46cca97be9fd4dff583d",
      "b854513cc82644f8a030ab9aabdf5519",
      "cee19513901a4118ae27d404016e19cd",
      "1641c25d0efe4cefb96fb2ef1898b582",
      "068502547b1f4897ab49fb82c8c88b01",
      "75b76c5043724a98b18a0b26904b36e6",
      "75adbf20c7304c0c87a37490b8077868",
      "7ec86c1af6ad4972a7c546d2e184bab2",
      "a2208de652394c83bd4dfa9bbc65d65b",
      "2f2bb2c795584fbb957338a7fc0ca201",
      "e3cebc635c054448a036f98068b17069",
      "19438ce279ae4d109f2985e648a92c29",
      "09390a4fa70748be98f0eb3ffc73aca9",
      "da51b9894bd549faaee8fb5b6f44c82b",
      "9014fbd6fdd5459cb4a2a2ccfef0410c",
      "d736f19ec8e34bb88be9f2e72455686e",
      "81d0e002e1a64549a4f054f543163f41",
      "c84d1d76f4b94e3488a7de6d09e825ef",
      "bb7e1b80459446bca5d84d7b774ebac1",
      "a1a0b3759a4d49adb3ee8e825a433525",
      "6404ef0323d44235860341cae362f080",
      "e21a783b09294ae29d2173f2663c9e7f",
      "8675a654291647728c40423dd8b99e4b",
      "e6e27abc259846d2b67a1301756f6a96",
      "33f03cd63ab845fabceb1fd62c2c170f",
      "4ce78d87759b4924af8a1075dbf6ddce",
      "5bb74dd29f304ff89bff4899d19b5b7a",
      "adb4ce9c86b94e91a8a7c8c7e09c7079",
      "7acdba6d2c1645239c1aae7267020be5",
      "999f768d715140999ae3d2bc35c23c2d",
      "5e4335350b7d49f2b14e3db86c063a83",
      "cc3bb7178c5641c697890e138cf25c34",
      "0ad979ea1b7347bdbf666fc457d8359c",
      "48192520dcc44b41ae26f0f4b418ea20",
      "465a1be16a3948639b030652d3707de5",
      "d5fa8b84bccb41daaa5849580fac1a3d",
      "f5fc6313d683461daacf952f8bf34392",
      "951a3bec46164bcf8e70f17ae17608bd",
      "cf98e5b8b48a44209597fe59072e4b7d",
      "0bd9b38743bb43c9b9ed26b500be200c",
      "a9fcba2e3a03432893b231f374a91a17",
      "b0a6d22905154691bbddfc3c9fbffeb9",
      "75f6cb50d6dc4c5194b844333fcd3e91",
      "2b9df743e9e743d39e4b81d6d9d3e101",
      "b0e1d711ce974476802d52c31b65bf81",
      "af4f51dee2244015a2fa3b3eb4de13c6",
      "494d8de002504275a5e079c832b74880",
      "ec665d7d3d17417581ab400ba1f94196",
      "c398d6da2b584ec4a979f71c5af66fd9",
      "4055e023504d4be1b2a664b47708bd29",
      "5b5fc2c030584237b3b3eabf0b76d9ef",
      "c450110d6ffc4ecd9b4c4edb547f9c07",
      "6cb0a2aed78d4a99b8daed682e801523",
      "917985f2bae24070b00147c660ad0e61",
      "01119d8277364563a441ef40db68acf6",
      "d4a27e4f5f9f4f27a674f6f3d300d7cb",
      "e46b524d949843b78f580db4195b914a",
      "05161fd8e6064845a28e1b44d6e89a73",
      "42a0528e6dbe492093228804f946f4cd",
      "bf1026966eaf442ea80d6ccd3224e924",
      "22d970d039eb491ab9e97099477e8734",
      "4ff810d3f45947579c552dbae9c45172",
      "b34c10d5c5214553a7b52e60ff78aabf",
      "34a594530ea04e2d927967699e5dc7cc",
      "30f018e451b54b169a31a967662bde72",
      "cef90c22af034904b35183a2f779bebd",
      "53b2a5f70c554476bde0e2514a25218c",
      "78055f624a0740d9b00dd15cef3dd038",
      "1ec3904b32a24a2a9a9b70763b870c95",
      "25cb85fe706c4e16abbae238e6c90a65",
      "b77e8cf9674a442db0b08c64d3ff2c83",
      "55e2854160e04bf292df38bbee17ffdd",
      "6ee161b4f1aa4112a207c4ee2492668f",
      "3c065d9c08fd4367863e8e35ac476e81",
      "9657423bd9074d51b484d46c8d18c42b",
      "85938c37605a4f63afbb15c0d5a2d0c4",
      "757d8e61bfb242719502565bacbd3a9b",
      "529e93880868406ba7f8b1afc2c7c278",
      "be3e0cbefef944498cf34d22aaedaa9d",
      "070cc007ec7c4c4483996f61b2d6f7ce",
      "63bc6a0124504cb9a15daeaca71b001c",
      "02281ac9625e4e4b991f888a70680cea",
      "98f8dbd20d1e4e5ea1f78f72542650a5",
      "98913edde9554127a1effb444dc3605d",
      "b408f04b92864babac1e09d7610a5474",
      "db697a98ea974914ad510ccec6c4132a",
      "d3c7b570c4384eaaaf598026dc335b48",
      "35ac5a07c29f4597bed1f55dd1202778",
      "95fb56f7042a47f8afa55dcb31bdd244",
      "3301faa89b814001aaff9d492990a1bc",
      "727bff98c3a944e98d475e1dd9ae56b9",
      "07d4537cd49e469e9d45ec51d859e3a8",
      "e911aae60d4941829562e205e9a99530",
      "cce81b158f2345619e2fec221770f9a2",
      "00a353f3c5744ff3b76c1e0a382ba6e7"
     ]
    },
    "id": "ya4x32y2MCDG",
    "outputId": "9e7486a5-6316-4844-c58e-a4eccf562e2c"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "472a9ce49e3b48e8b7c152d22096d35c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Fetching 7 files:   0%|          | 0/7 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cb1780944c0546c59b7b068f9e133100",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer_config.json:   0%|          | 0.00/1.43k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4bb5587a474042eaba8996cdd5a99dc8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "special_tokens_map.json:   0%|          | 0.00/695 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7d316611cc93473eb9824e31229125e0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       ".gitattributes:   0%|          | 0.00/1.52k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8c9c7ae608174d319ab3af31b3fd6cb8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "vocab.txt:   0%|          | 0.00/232k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "d415534bd12e4117b1ca44880bdcbe71",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "config.json:   0%|          | 0.00/650 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b048d018716c46489c5fede8664a2317",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model.onnx:   0%|          | 0.00/90.4M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e4d6b15aa7d64b1a89fd6211f7f65f1d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer.json:   0%|          | 0.00/712k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'sentence-transformers/all-MiniLM-L6-v2'"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##Initilising embedding model\n",
    "## Using Default Model - BAAI/bge-small-en-v1.5\n",
    "#embedding_model = TextEmbedding()\n",
    "\n",
    "## For custom model supported by Fastembed\n",
    "#embedding_model = TextEmbedding(model_name=\"BAAI/bge-small-en\", max_length=512)\n",
    "embedding_model = TextEmbedding(model_name=\"sentence-transformers/all-MiniLM-L6-v2\", max_length=384)\n",
    "\n",
    "embedding_model.model_name"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zZMhjl1NJz_X"
   },
   "source": [
    "### **8. Processing chunks into text to be processed by [FASTEMBED](https://qdrant.github.io/fastembed/)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "d2_E_VKqIvu7",
    "outputId": "0f64c35e-0891-44c4-d94a-3c99f36b3f16"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "content :  4431\n",
      "metadata :  4431\n"
     ]
    }
   ],
   "source": [
    "docs_contents = []\n",
    "docs_metadatas = []\n",
    "\n",
    "for doc in docs_processed:\n",
    "    if hasattr(doc, 'page_content') and hasattr(doc, 'metadata'):\n",
    "        docs_contents.append(doc.page_content)\n",
    "        docs_metadatas.append(doc.metadata)\n",
    "    else:\n",
    "        # Handle the case where attributes are missing\n",
    "        print(\"Warning: Some documents do not have 'page_content' or 'metadata' attributes.\")\n",
    "\n",
    "print(\"content : \",len(docs_contents))\n",
    "print(\"metadata : \",len(docs_metadatas))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nw_yWHWGK0vX"
   },
   "source": [
    "### **9. Adding document chunks into Qdrant Collection**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "RwdEyzvOK5mZ",
    "outputId": "fbb785d7-96b0-4975-bc35-137103a4bf31"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████████████| 77.7M/77.7M [00:19<00:00, 4.08MiB/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['ab2e436dc7a54f88a36decf9a767b205',\n",
       " '1c338ae0388845e281aadd127f62d827',\n",
       " '93e5d106eedd49bead2c7594966047f0',\n",
       " '143eeed777ac4f319008c60c0ba58cdd',\n",
       " '518f7d4f88c247298ff3feb4080f625d',\n",
       " 'e2fb0f43a3af41038a6285046f8125e6',\n",
       " 'ccf97ddfa98e43d4aa524e96144e4fc4',\n",
       " '921461da8a614c90bcae411ec407fa0b',\n",
       " '110b058859b54952b7a25e19f4601b39',\n",
       " 'b17f85b8e75b479caa3e8f30a70906b9',\n",
       " '42ec6d70abaf4d158eb5b291c614801c',\n",
       " '550ef9a7adf642a4aca08b82ddcf957f',\n",
       " '71927d1fbd7641ccb2adc7e3375ebe54',\n",
       " '4fe7527f5f4645ed81411ca8cdcdb307',\n",
       " '1f10e9c5e5844dadba647d8e013472c2',\n",
       " '592cbf2996ef4fbbb54d2c4611fcac2f',\n",
       " '7da36de6c7764c8d95a5c785892542cd',\n",
       " '39140fc65d90450d832f4e09cf03bb35',\n",
       " '393cc799b25d49b0b626dee11311683e',\n",
       " 'c5793ca6d2924d1897302a588245375a',\n",
       " '6fb10d5042bb4b82976741da3c6ef4ef',\n",
       " '4255d7cf11e44bbc98fed0916518f733',\n",
       " 'fdc04a68d34e401ebfefe0ab3880c63a',\n",
       " '3bb21818fb8243f49c481df2f049b746',\n",
       " '4ccdcdbac8e24504b17b67ae61178e17',\n",
       " '81145214a1a24c6484bf4c592652387b',\n",
       " 'f06920a51183488c8d367f0417c1ba5c',\n",
       " '8836fa38f3934d8cb154eb68deff6f64',\n",
       " '6f06a84bf5c041899a5d5588f0529f2e',\n",
       " 'e16dca7c39064195886b12ee5fdeade7',\n",
       " '522cb8d1a78c4778b310f9babb1c236d',\n",
       " 'cdc6f109a1a34c2bb74df9af0b957888',\n",
       " '56071e71feb44ed4b7733b477ba43c38',\n",
       " '7c1ff694aeb64e758d1755df211dcb7d',\n",
       " 'c803acf38e5a4baaa649a769ccd9f279',\n",
       " 'd70bae389dca4682a1151b2ffbfe6396',\n",
       " 'd5ccec90e80340729c6213552b973dfc',\n",
       " '30e30bb554794b93ad46fd133addf99f',\n",
       " 'e2520676be5248c0b0c8a525e7f0b2af',\n",
       " '7e6be5521c6d4c1cb9deecd4b6f53100',\n",
       " '86460ae2a47748e69179468131a0537a',\n",
       " 'bed50c11c5194362b98c8dd280a4b484',\n",
       " '476a2d38510c4f0aa11a448338067ca2',\n",
       " '480b9a19d4f0407a9ee48b6a7677d4fc',\n",
       " 'fcf2c4dbbeba4a59bb82e5352e78ab62',\n",
       " 'd630d3c54c1e428fac36b630acce2f9f',\n",
       " '30bc02bb31e049c095e94ace633df0ec',\n",
       " '009ac65af86c433c8350aed83514452a',\n",
       " '591b120c31814fa8be77fbbbca24f9f2',\n",
       " 'c834db52e27b43dda35cb74f6cd0b8f1',\n",
       " '823b98f5701c4ee2baeafbfa359d37da',\n",
       " '34c0acef47284d8496568f2c22a14053',\n",
       " '975d52206d654501a71dc7c52ed474ef',\n",
       " '7fb16234f4d84a35a33ed69aa34d4e42',\n",
       " '0f1bc47a0d8a4bebaabd974d3f664b2a',\n",
       " '3c11d8c29a5f4bd58a54a6894687cdd7',\n",
       " 'd3f6609dc94c417fb0191c2804bcc8b3',\n",
       " 'be6e94ec5b4042acaf4796c7839a81e4',\n",
       " '938f0492bf614244ae26cd400587793c',\n",
       " '2ee965ec0f974a81be99bb261e16a26c',\n",
       " '17d72f27d6264792824deccd76f265b4',\n",
       " 'f5985bd37f10499389f069967be33699',\n",
       " '7defb852be254d1f91170f5744bfa021',\n",
       " '65d7eac647104ab99386c402092db906',\n",
       " 'a94e323d7cd14a9691c6916fb9595269',\n",
       " '4bbbc3ac4c5e4b35b78f02459ee066d3',\n",
       " '329f6462752c44d9977163e140c83b35',\n",
       " '3f39b6a90fdb4a1bb5241ab81b0e72ca',\n",
       " '9652b5a089b6407f83a7a72f61524e85',\n",
       " '05987f9a68bf4d8695e203e058935420',\n",
       " 'd3d855f3a8ee42aca3acfca027588a3e',\n",
       " '45143c6153a84e58bd0acc4fdd2b7b05',\n",
       " '05fa0843f6c14cb69e9786a823000e03',\n",
       " 'f89e80204a1e4082a6af644a290e8b8f',\n",
       " '4ec9eb04285f457bbd3af039dcaa592c',\n",
       " '744f4646d8904f8492fe654052f0a3c8',\n",
       " '003f4cfee05c47b39e37bdc92eef40e7',\n",
       " 'd22ca647b71c42e3a69a03e986f9e792',\n",
       " 'b0994f8692674ea1bd83eaa434c70ff2',\n",
       " '8888bbe87eab40e0a2d74ccf91db0646',\n",
       " '90c8c77966044251b59d1739d650f196',\n",
       " '47f64475765c4957a1b6dbdf23a658bb',\n",
       " '55330ee0b3454965b919560451a2218f',\n",
       " '38f5994f3c564e27b5fbaf53de778278',\n",
       " '70d3d0970d584670954d4875b4572106',\n",
       " '3674c31d9f8842fea8257cae65db8751',\n",
       " '63660c7761434dd99956c10c54bd0746',\n",
       " 'd6952384f35a4943ad1bb70f932d38d8',\n",
       " 'aafc70ea166d43f8953533e53fbbc88d',\n",
       " '75cabd453ae24952aa914272acab90d9',\n",
       " '94f03d7364ac4826b8ba2571a22b91bc',\n",
       " '37cd4ba1d36f423f9f10f4d56320df60',\n",
       " '764640d526eb4a49ac96ede48a7c508f',\n",
       " '5e35e236f0c6401bb8b5ed8c44e1106f',\n",
       " '4e989b3a7c264ad4a1059bf66b800da7',\n",
       " '95ff2976d94448f3a8ec7a2292e063b2',\n",
       " '9b1ae40eb5cc49b6b06e06a20649b0f6',\n",
       " '82ea12c4952643c28c5092ab17f64522',\n",
       " '993e38ec8d2642c79b1dcbf2310c7e89',\n",
       " 'f9cf094e258c48c99312b72a0638d430',\n",
       " '269fb1d086d94f448188eba58e5d7023',\n",
       " 'b9ce749c60224f86ad3a94c943e37f95',\n",
       " 'a17e18d57b2f4580a413162a53630234',\n",
       " '0fb882dfdff246e4bc77a00c1f7f38a3',\n",
       " 'd4bad0db547a445ea7ed272a7a439ec3',\n",
       " '920f823122cd4ec6afeebd0b64c88405',\n",
       " '54c8ea776d4f42edacb11cefd97d1c3a',\n",
       " '61ef9464f7404489aa1c928cb579e8fe',\n",
       " '10202b6cd2dc4009acd7a12186423854',\n",
       " '5e1b9a62444b425fbbfaba29e17754d2',\n",
       " '29cf25be43174ffdaaf206edbd502e71',\n",
       " '72fe8bf098bf4713ac377b5775d0d894',\n",
       " '9a956bf264a54a439c30b397c6495418',\n",
       " 'a1d56829907342daa597701ea500c9b6',\n",
       " 'd189b531c9f241fd956e32d8c64818c5',\n",
       " '9a7a3b6b11e1404c9a58c37c90d5084e',\n",
       " 'b29827f1e0c448699008b604aabcc478',\n",
       " '8a2818c8faf948dabe8f75ad3c648059',\n",
       " 'ad6f65c1283b41f6972355bc8957c861',\n",
       " '897dc1e5b7c24b168c52d744604a766b',\n",
       " 'b7dcd1d6ad444e3d9d287c3489c6736e',\n",
       " '6ae48d0df3bf4e7d83c4adb411fe5069',\n",
       " 'ee4b7d0c3d974577a2c213a7d4e8fda2',\n",
       " '58e1ce227c6041a0be4000657ee51b5b',\n",
       " '15b09f6f498a494aab998e2f2da34760',\n",
       " '768257f807994774b49128ba37d83ee4',\n",
       " '01c4e9ab898246cbbddedfb966ab2a5b',\n",
       " '7d5b5d3b7c5f46b2979f285f61e187a9',\n",
       " 'af04738204cd44de84599ce6faca3edf',\n",
       " 'a0474024f3104f37bf76fb7265c689d3',\n",
       " 'cd1422350b3045d59514aa84e535153e',\n",
       " 'c3c3ebb97dce4f6aa6e8e0e4deeca03b',\n",
       " '16165246df6e4a01a84c76117845594f',\n",
       " 'f4431f8c0b1c49de86a59a5673df8f10',\n",
       " 'd1f7f8bab96e4beca079c1f0c28ceb51',\n",
       " '52c42645e63e4a2899c694e6ebf4494a',\n",
       " '7192970af6584f0aa5b4493f08d38f67',\n",
       " '798072d8985644089c043b73a68d31f2',\n",
       " 'e5312e826f474883894ea7c51e4735bd',\n",
       " 'ea9799064e074a818a0fa16b26ac1cf5',\n",
       " '05fc239b8d444925b2601675a8079ce9',\n",
       " 'bc463235458d4c8f929a825e2e7da4e1',\n",
       " '0f26549584474bba84f31799111bf44f',\n",
       " '8672e21a8160470e930c04edbe758ec6',\n",
       " '21f79943ff9a452aaee078ca5a4fdae1',\n",
       " 'a163c16f043e4ddba7fb5c1af293776e',\n",
       " '97d6729ff61646979ec9b778c7404d3d',\n",
       " '887c16ddbb8b4ccb8b1b149d2f254f19',\n",
       " 'f0e1f427fbf94da5830aa0499eadd00d',\n",
       " 'e367822e8ff1405d9894807061b39b30',\n",
       " '7b5bc184417a4031b958c28c17129850',\n",
       " 'afd5950d9c4c4461a770882f990b95f2',\n",
       " '7afd9d5a0054420cab036b5d9cdd1cc1',\n",
       " '6708a28d5f7d4b1a8b40c16afe1be853',\n",
       " '0b84e404c296493ea586f07494084786',\n",
       " 'd51ec4c5ac8f496d81fc3360b18ea144',\n",
       " '36375db11eba4797926f5bf6050fa0f9',\n",
       " '459694dd65c14e4b975ba1296944ecae',\n",
       " 'ac60c926de2d43baa103d6b03e8da182',\n",
       " 'ed000186843f4794b1fb393335c26f6b',\n",
       " '16287f92ed734bf480fb6e8e0d103aa5',\n",
       " '98ec3a3652c141598c428b9b1f43abe1',\n",
       " '0a5b62ed89ae4252b4a3c64055f6b344',\n",
       " 'f7f172809a2640f6be43c1fcea443e38',\n",
       " 'a3a1aa7509ce43f2a3f9028a79651b6c',\n",
       " '3c6d9b65ac96442793355d1c55ba1d0a',\n",
       " '37800b392700415f809e899a165263d9',\n",
       " '0ab0308cd369418fbbe076f65dd1090a',\n",
       " '35be30b4fa5445d682f35c0e896322a1',\n",
       " '051d5e41935e4b8dab20ec5841d0dd0c',\n",
       " 'ea9d4f940dfa459db5d3c0d8d79069b9',\n",
       " '89fce74af0d5439e825e71f32df01c11',\n",
       " '63891e13c3d34ce6bc2193832f913ce6',\n",
       " 'a1b47052c13f41cfa835bfab3889d2fa',\n",
       " '722939505a4d4bf0b1b4b35a4aacde76',\n",
       " '33ff9021037d472ab8154a4f795d4919',\n",
       " '517cf6a3715c4144914a4b9769e1c719',\n",
       " 'c439234354d14dfdb44379dc840654a7',\n",
       " 'b300e88ba4b84911809d2e8331f49551',\n",
       " 'dfd97e245bdb479684269f55e1da0ee7',\n",
       " '1953cc38b70044c0b68e2215cf3822d9',\n",
       " '29707b36ccb645dc8f5256c0489859f6',\n",
       " 'f7f72f1ab28c4cc6af45b342948405dc',\n",
       " '34e0d195cbba4c20b6653276e8af34e7',\n",
       " 'f7238d0d265247ec945a17e9789af37f',\n",
       " '880304ff5f7646bdb758235735164417',\n",
       " '8acd63c95d6b46de9dda158756f3d5e1',\n",
       " 'b788eff0efc64cb29bc829509f4a7ad1',\n",
       " 'f8b54e6a765a460fa5d556148c7734fd',\n",
       " 'b7df0dd23ac44ea1ab921a1d2c748287',\n",
       " '56a69c57a79d4732b52c374363dfdee7',\n",
       " '3367c7cf32484bc8b2d2d65e4aa926ad',\n",
       " 'b15c39ff063a47f3947f0cf888cfd388',\n",
       " '38c5a8b6aac749448a493934fed98560',\n",
       " 'f1ebf225717549568bdb2f41de8a37d6',\n",
       " '3b279b36a14c427c8db88eb75e8a23a7',\n",
       " 'ea74fbd93a8249e4b0dba5cbb6f49115',\n",
       " '7e1cc17a8cdb471284545025fb4de359',\n",
       " 'e5b12226aa374ada994cbd4ea794c3b5',\n",
       " 'af8b5de556f941ed96993b14118b1c32',\n",
       " 'a177eb9fa99b4b428e84478b7fe900e3',\n",
       " 'dc9926a7114f4a21af30478a266f6adf',\n",
       " 'cfff51e696f84d31829b0728d51c1f84',\n",
       " 'b66bab81976942c4a5f9ea0c63f1655e',\n",
       " '15544d45852c4e45bfa580b2153d9aab',\n",
       " '8a14cfb19e174cc1912c1a70a208d834',\n",
       " '1bf65fb5c46c4cd58599a86912d1333e',\n",
       " '08bb6f7377d3468891bd36350643db5e',\n",
       " 'fab39e8bc7914eeb84d552b2bce78aa5',\n",
       " '385ab152ed164854a7eb5b5ba93229a3',\n",
       " 'fa854dd53402420189cf9556bd76a45b',\n",
       " '84329a0ff3514c6db4aad932150cc1d3',\n",
       " '641cb6e463be453ea1110b610c276239',\n",
       " '1f8ebb1ae14a48efade0abae73e49bff',\n",
       " 'd2bf03bfdb334178a20f54cf5fe11486',\n",
       " 'cd5040762b014cc0925815ade88cf231',\n",
       " 'b8f689315260477f859996b1a5f830b4',\n",
       " '235e791f5495476f9b3131407452c301',\n",
       " '3503f3f9a60945718e05f1b289393985',\n",
       " '57cf1a18add641b7a8db079311623495',\n",
       " '7325ab44074a4a978d3a8c774c028995',\n",
       " '2fe805be576d49e0b103bc98c25c1dbd',\n",
       " 'f085570531af406ab0100139890b6733',\n",
       " 'd12c6c129fce4fa696cb65af0c79c4fb',\n",
       " '06d0701a8f144a15a8936dfc2257a2dd',\n",
       " 'a34a7d7f480c4ecda084f5711e0e3658',\n",
       " 'd49cd030169c41a7b112efd67a505254',\n",
       " '7fbbaed8441847448127c28abff4ff17',\n",
       " 'c2feacc530f04ac09567ef9772d128de',\n",
       " '07db3ea5a2e84e9989a86d3525a25aeb',\n",
       " 'c0e16cd21c0a4a3391f19dd7dce9b451',\n",
       " '84f73c448c294d4c805439f1a859a046',\n",
       " '6de3e0f350f844ff8f69d97c08ff45b7',\n",
       " 'b63e0ae4f2ac4e0285f7c8ead013b0bd',\n",
       " '7056d5f1b13e41b9a29f2fd31b393fc4',\n",
       " 'd3e1890246294b9ba8c0bb0600186062',\n",
       " '39cd79cb29604abd92523da6b92e6b73',\n",
       " '2ded00d19ba54b6c95cfaf6044b86aaf',\n",
       " '2145d73ce4b54394bec602f49f82356a',\n",
       " '9cb9749c76a045fcb20d6b217be72c12',\n",
       " 'c7a37ffb1b10476b8609338a1f079ddf',\n",
       " 'ed8b1bf72e2a4e02aa9da83d8d021a76',\n",
       " 'd2c52f4afb9e45d68004288fed1677b3',\n",
       " '48074a1fe2c84276a6397d79b7e372e7',\n",
       " 'c841d9d05f2d4257a08323578a27af7a',\n",
       " 'e123c44a14574e3da395f42e5e5c1842',\n",
       " 'c490d7f16b744373ac85d11fd0a52467',\n",
       " 'bd4ee8b842cd451fa5fe5a341beba538',\n",
       " '884a9a33d8aa413facbc16e99e6a2d8c',\n",
       " '644ab26119cd4a2baaef8e7d4562c79d',\n",
       " '5e98ba0d831847f58afd3648200dce49',\n",
       " '9e62bdec6fb94131b8bc204501f1a825',\n",
       " '14e0c69655344cde9f96ab1d18442c3f',\n",
       " '8da79692620c465fa093448859b80b3c',\n",
       " '4727f27261c04b938d40784ed9b61afd',\n",
       " '6b7ab22e03e34e0d8cfcb28148d3d9d8',\n",
       " '9ed517be02cf4981806474a7d5d3a95b',\n",
       " 'f537530ba6ed4856af23d3ddcdd20830',\n",
       " '16731042f67246c6a985528087acac3e',\n",
       " 'cedfdcadcb7d404db8ee2af7a678e053',\n",
       " '51fcd2bd1c4a4f4da8b49ecf48c2df31',\n",
       " '9979bc269e72489c8275b4cf90908f33',\n",
       " '041087a448a14341a2141ae7d6b8e1e2',\n",
       " '85e387ac226b452dbd6fb91ceda4b3ad',\n",
       " '5840ffed4bed4e0183bac2e0ec24a8d7',\n",
       " '46ac2eebf6854e16a618096d200829a1',\n",
       " '2d47f57135f44364aa3f67a759e238b5',\n",
       " '4201985f41de47a4ae790b21f672ab37',\n",
       " '707ecadba8b24454b4ab386680438d8f',\n",
       " '7b5d39848321408ab404598fdb7acd3e',\n",
       " '024ac44f8c0b4619a34aa7559d568f8f',\n",
       " 'fba3b63687184e239c9789afcc368561',\n",
       " 'ab97a0d8f9b649aa8b5e950d0baa8433',\n",
       " '0710246160df4f03a90d6601434489b7',\n",
       " '518f3096247d4e8d9af2510fa5ae0884',\n",
       " 'e2b7393823ae44d1950dec8286487c78',\n",
       " '8e365d5630c84a0a93f629dff11d900b',\n",
       " '5a5864af763b45969862a7151fd23256',\n",
       " '724c667486a749c68adc413b2dba1270',\n",
       " 'aec57eb787ba48b69578cdc3d154cdba',\n",
       " '26eecb9333ae434daec23482b9df52af',\n",
       " '00c2666b2f344a399fb8076d1cdce215',\n",
       " '2dbfdf38d2804c54a8cd438bd1f84510',\n",
       " '15a16b55d5d6488aa4fd8c8d0960c307',\n",
       " '731a8eea2bae4d8fa1cd79574b595d1a',\n",
       " '11e24c9f6bf54ed8811246b033d85fab',\n",
       " 'd4bd94b000534aaa82862cb0a18cdf30',\n",
       " '533d0a044f85452fb0acd45fc5496f9a',\n",
       " 'e19c79265c4c4d7183c0428e6a60ba47',\n",
       " '42ddd8a2abb341cba516e0fb2c418b40',\n",
       " 'd530dd6efe9c4623a465d72421af59d0',\n",
       " '04efa0081e9a4687a1a6d8810cd9cc67',\n",
       " '4596e81102b044529a58d40ec50f304c',\n",
       " '2ec969d734294f2ebaa8ff65f154ef16',\n",
       " '7e260c9dfe2a4e548e4ed3b74fa00119',\n",
       " '63b9ab0ebdd34fcf955de90b69495f9f',\n",
       " 'd219552730de4189a9ba442c8ad8cf49',\n",
       " 'cd12a065fe2b4ab4bb60aed4b7d879e5',\n",
       " 'b86d234dbf984285a3ccd737af56bf5c',\n",
       " '5fa2105ad36f40699beaffe5e0970010',\n",
       " '284a7e482ef24045bd6b9951c26a1157',\n",
       " '3b643206f1314b0e97f1434e9b6a250a',\n",
       " 'c1b472b5d0874d798c3002a0ad90f16f',\n",
       " '7a409b02e76a4d0ea077cdbbf79c1b68',\n",
       " '4e38cf4fedcf4bb1af9a668f360a439f',\n",
       " '272b894b75684b41aa2b7bdfaf40d4bf',\n",
       " 'bdc87c26ee7d432a87dabc2e28f53b79',\n",
       " '448e012435cd4f24ae7f47b9feca7fe7',\n",
       " '4838052cf2cf4f108669434c93c9cc7e',\n",
       " '965f41b0a89d417794f9057ad57db266',\n",
       " '880c2c66cf5d469b93841958304a7d11',\n",
       " 'c2c5138c85d743b39941c66bcd9da328',\n",
       " '3fe4807bf1564f4fa0722775f7602658',\n",
       " '49dee4e4083d4710aac2a787fd8dd996',\n",
       " 'ee49f853419d47f2acc4c3cf53b4012b',\n",
       " 'a91c5bf8cfe047c89b3babcd4ef0c8af',\n",
       " '250dea70bb3144ceb5f41979d9909e3c',\n",
       " '366e2da41afa4b87a33439bc25ba5185',\n",
       " '8309110926494ab2bd54ef519d9be1a8',\n",
       " '0c6d6b3c85374a639be101b31c5a283a',\n",
       " '33b9473f26ef42e7ab5de8b65b38b82a',\n",
       " 'b95ec69ab32b4ae6a02ece9fbdb0e064',\n",
       " 'b6f9b35cb01a4192a3da44fbd45c5f88',\n",
       " 'e77ed28c805342769b108475e2ceff5d',\n",
       " '7fb630d3de75485298f22b13ed82a0a9',\n",
       " 'e10cfed1f6d248fe8fe9ddb04ba2e9e4',\n",
       " 'd3b6fb6b0bd04d78aa8967960e52afed',\n",
       " '8b55a46cba0e46e885de1df6feed02dc',\n",
       " 'f3461cd8df5441df8c3abf6e5d1a023e',\n",
       " '106c15552207448b8a63a10961e0389f',\n",
       " '9a025d6cd2ff4ca8a99ae0caf30517ea',\n",
       " '71d14077a781482bb5963994015f9b40',\n",
       " 'ad5210d3f12d4f63962b470ae7145792',\n",
       " '996f32ad22954ca5bedb45cf10846d8c',\n",
       " 'fcb08a17aaac44a289c998801fcd3325',\n",
       " '11b2c69de1614f7aaff692517a11eedf',\n",
       " '0488484de1764d42a0aa090d5977b2f0',\n",
       " '1cfce0ddba554e8c923309aba6ee4410',\n",
       " '4e6731f6ad934da4810fbaf256fbf1e3',\n",
       " '7e02104d74eb4fecbd3f3bc1f27dfcf1',\n",
       " '3f0925882d07435a82c2cdd0328dcb0c',\n",
       " '10d7c77b71d34060b83e8a3b535470e5',\n",
       " 'f796ea6b2dda4fb7b4f943c582e1f41d',\n",
       " '1f738cbdcf0d4c3495b10a8646af502a',\n",
       " 'ce742d5bf8314a6a8048c84a9edd3325',\n",
       " '76a2bcaa37224dc2b1eefb963abfd93e',\n",
       " 'c86939ff28ba4fd4a663d4ec480349ec',\n",
       " 'ebaa969147f54d05abc446f97d26db7a',\n",
       " '66e25ec0aa004ed58f5b4f9318ccc74a',\n",
       " '0b26545e2d244cc1a0fb157b57642f44',\n",
       " '358cb4f6963f4e17b2eb804a234b84cb',\n",
       " '265b49dc7b714290a1da0101ddc0d00c',\n",
       " 'cdbd3d67037644079da7c68142b9fb44',\n",
       " '87b259d11bd24f87ad772ea69a2fe2f5',\n",
       " '5cca1af15e2d411b8e54fe9cfc7b7850',\n",
       " 'ed9c6437e5114275abbd84dd0f083a03',\n",
       " 'a905e59bec934e7b9ea2cbc15b2fc9c2',\n",
       " '4ee77e996af04c4f9caa9b2a30427a27',\n",
       " 'bd88704052764b649a644166ca56adc8',\n",
       " '5302267b6b9248f9a55aa91acc8da2f9',\n",
       " 'c685158d914a4dc5b12b8f492a52a2bc',\n",
       " '5a842e5c0b124719a117ff1a5a593523',\n",
       " 'eaf1f2eea12348a6ae4f4c7a8650c61e',\n",
       " '295a0ebc5456419da808d6358cd67e58',\n",
       " 'c8cdb9dbb2134a758726699423dc4d92',\n",
       " '57944809119c483e9d3ba4487f019df4',\n",
       " '1d28ed6f50184f55b4c28609fc3c938b',\n",
       " '7435ef84b013482f978eaccf24db2575',\n",
       " 'f32004d1bf374870a5864c6978ac6b23',\n",
       " '35145db1f5b14e6db2acb96c258cf3fe',\n",
       " 'e4f974f952704fef8b2ac3c8da0c59ee',\n",
       " '2e0302b0ead84ca8914cea0e773e116b',\n",
       " 'fceacc11e8884d1cb4f168aa8485d693',\n",
       " '3c99545323ed4d26aa89abe82cc6a6d8',\n",
       " '03d6eefcbd8f44bbacf0efeaad04afa1',\n",
       " 'c6f0ea08b6d3464aa5a57fa217da7f4c',\n",
       " 'ef3a8d0051974b2fbe9aff58863125ff',\n",
       " '25821d19f834407bbd54a93ae5d20b62',\n",
       " '05658fd549f6421283b6a19196c7248c',\n",
       " '9cff7480b43a4c429e035bd5cdb4dbb7',\n",
       " '67897e579f364a85930e2be3dc26c53c',\n",
       " 'fb290be82bc84c0b927b7cd5e7797523',\n",
       " 'c1c6de1a7059456e8fa2ff526774b074',\n",
       " '805bba844a3748cdb5c8240333b8d7ac',\n",
       " 'cf862e68342d45338aa6d41ce5d6e919',\n",
       " '0cb69fb144bd46cd90b4c445e6b4c23a',\n",
       " '877bfa2bab1542d3998b6ff442a59240',\n",
       " '049dcb33c29e4852baa9a668d6997ae4',\n",
       " 'ced42bbdb28a40849dbdc9d17cc51346',\n",
       " '769e63f4334c43f1a3dc0f711fe3c4e0',\n",
       " 'ac683648828e47699383ea35d5a70d65',\n",
       " 'cf326d3d51064527bbb0a4c181eb1905',\n",
       " '0b036698f9774b069733ef309c486657',\n",
       " '1a4ecd291c794e27a3e501b3c12f0a65',\n",
       " '3c8780e91fb6474ca9c33d749cde3951',\n",
       " '01f0b31f0272418a9dabfce7dc912ef5',\n",
       " '081a7d67808e412587ecc1d14adb56f8',\n",
       " '0d737c1a2d504180a58809aaefc6a3a6',\n",
       " '3895f2a6583d4b0581af1da5385835c0',\n",
       " '2f4f7701683c4fa8af01bcfce24ba479',\n",
       " 'cc4b50fc178b407eb253c50d068b9165',\n",
       " 'a8d42fecf7f040fab08708813b2a25e8',\n",
       " '5bb4dace2e2147e0b120e8da3b067e1a',\n",
       " '8fb07680a99f48b59acd102bb12757b9',\n",
       " 'bf660c958f0448429f39c9957c806f21',\n",
       " '39d42a01407848eea0ebfcc9a8c4c199',\n",
       " '199820ff26374edfa57cb7eb333b374e',\n",
       " '0e20287769a9452da83a6bca194ccda6',\n",
       " 'b586d35c68d8465d8670e5c662e830f3',\n",
       " '8f82e79f4244478f98c4c3cd2d59f626',\n",
       " 'df4f74058b6a482ead717a50af2751bd',\n",
       " 'af1bbe5a5c654e9b963a52dc8bdbd261',\n",
       " '6617f0decff34676adbf10a2d211fac0',\n",
       " '6a7228067eac4737a29381256bf16460',\n",
       " 'ffa045a0825f43f3b699c334cdf92019',\n",
       " '8029ac7a8e0e4ce1b0045bad05cdaf8e',\n",
       " '0a6f4b073b5e40a7a2bfd2120d316147',\n",
       " '4a2efe42d2b14604b8504fd38dc608b2',\n",
       " 'fbeeb820162642ff86eb505afe477c95',\n",
       " 'd391db4adf3b46e98ce7490a5589cee3',\n",
       " '4c6562a2f28d405e981bdecbfb6e6672',\n",
       " 'f43367f693734e57b2babb954595f0e0',\n",
       " '08615fe8f7014ab98ff4cd4bf4a22709',\n",
       " 'bb332c5e262d4de3aa7b612041c399d7',\n",
       " '9837803f02554592b95c0f691c39c5e9',\n",
       " '4d7730e1b3134809b705e44538b5fea0',\n",
       " '554694484dd34ac29cb825a6cf48b748',\n",
       " '78636c5811d04953bd08cacbd82be24b',\n",
       " 'fa533efc864b40dab0eac0f121bae07e',\n",
       " 'ed956b6a8e3e452da93a27f49c7545ce',\n",
       " '7432210c61b44df9afe8bdf890e2ef7b',\n",
       " '7bb04e0177b7468a936a2e3106ff6ea6',\n",
       " 'd3a15603118e47f9b30a20dd4776048a',\n",
       " 'e7af96c06c58471c98d3ce4dac0ccb18',\n",
       " '7f9bf93da8514ca29e95cfa729080644',\n",
       " '80fe51a956be43ec9e241ed28f39e2cf',\n",
       " '3c10b173b2ff4e3586da400b45a9ee83',\n",
       " 'cd97125bd2be4ecc86d7d00778ec04cc',\n",
       " 'e3996e0144ca4dd994a837f140238a0d',\n",
       " 'e8c1f6a9ef484becb1e5628157434bab',\n",
       " '4ebd9d5b8b64435e9c691202ed557854',\n",
       " '8e57964aec474489ab775956750f4920',\n",
       " '423b9bcbf9af4a088cd85db7ea11f809',\n",
       " '67584aac9fca40cb8e38ba9eb534975f',\n",
       " '094b5bb049984a78978d915cba54aa8a',\n",
       " 'fdee6d2a0be44d05b8ea807c22cfc62f',\n",
       " 'a9222504ae02479dbde855e0a126fc43',\n",
       " '012f60c5222f4c31a5a30908a94a605f',\n",
       " '6260e3c5619e4cb78ee7ed92b911a667',\n",
       " '3abdaf371edb4ad2bd21c48022cf6b5b',\n",
       " 'bb721781bc684b459ae52c6fd155fac5',\n",
       " '64ce6ff35e2244cf9f1758c9b0bcea4b',\n",
       " 'c5c6d5d40b3c40e8945be06121acdd1d',\n",
       " 'c005a06f88a04f2a8915df6203cbf165',\n",
       " 'eab5c6f97afc428bbcd9a2f8ead92739',\n",
       " '04893e64459c4c6986c8695c750910b5',\n",
       " '6351c0be941a4f049c7e5d875f6caa5c',\n",
       " 'e1a3bae5c2354c9da0f74aec44264ffd',\n",
       " '2c8317933d914839b1132fd189f7c56b',\n",
       " '0c02269772b248e9bc0eaa93087679db',\n",
       " 'c845f82e85b04aa58f4f8ade566d1c0d',\n",
       " '9da3a69b698f4860a72831ac1ae46359',\n",
       " '030b1e5a0b90495298cf2484e5a50e70',\n",
       " '52c7f3075ebc482f81d828d5558285b0',\n",
       " '44d5c2938d40446bb3c7a4d5a10f8dc3',\n",
       " 'f91ba9670872425d9818aac8af78c265',\n",
       " 'c2fd812106bd48d29ca84ae49a52863a',\n",
       " 'f05a5a816f684e2b8d392d7d4ec09447',\n",
       " 'ca22b154dbcf4a3eab4ce1a084b5c01f',\n",
       " 'e2bec4899d08474eba22a673e2b19db2',\n",
       " 'ef60c38a9b2240dbbf7e45debb4c63ef',\n",
       " 'd1ef41c6d2a34507a628f7c028b99d29',\n",
       " 'a3b3e4383aa6475cb06da2e477084737',\n",
       " '2e213ee790534d5882b4485184957097',\n",
       " 'a401416dc09443b28a17f883990ce366',\n",
       " '55fb28da5d4e4f9cb56e26d42607aff8',\n",
       " '04fa77a08c4d45aa9bd01c39ba76ac73',\n",
       " '51ce01d36a214b29becc6de0ed77d2f8',\n",
       " 'a8a453a8a53d4c198111b48d08def77c',\n",
       " 'aad6b3e12cb348739be7c75cb5b6fcbe',\n",
       " '1e8bea11b09841e495356ee3aeb54dea',\n",
       " 'd6e12d09278342268e0ebaac63aabc26',\n",
       " '1adfe373c8ff4335b3a482f96e5510c5',\n",
       " '6b1bd87e164d48aba8dcd01705ed432a',\n",
       " '3d75a0d9a8124c958473f6d96aeb6b39',\n",
       " '1ad1b5002af9464c96b6a1d0c441e1c0',\n",
       " 'a14012f63328404c988c89586fba6ee4',\n",
       " '0343678b37ba4891b140c0fc1c93a312',\n",
       " '9d968cc3d8e749df806db39ae38369c2',\n",
       " '55264a426c2e49abb82850a8d67cb320',\n",
       " '2d02bbc0a5bd4f7e9200b0953aa1c832',\n",
       " 'cb053573fbaf4ac09d0247b44d5c9cdf',\n",
       " 'acbaaa1575014cb58c20f01ec5497081',\n",
       " '559bc1b3b0f64abd8ae6021ea9931b13',\n",
       " '3705c945dc7d46b5956af8a87319c7e1',\n",
       " '9f77eef74e8d447385280b103c6c1090',\n",
       " 'fa04e0ae946e41c4a5b576f82e2b3a13',\n",
       " '22421ad725ff472a928b01439af5e6f9',\n",
       " '8afe626c897f4c9b8594a79395b41294',\n",
       " 'c86621527f3845ada1f15f5c759e1427',\n",
       " 'fd8d215a669a4474a39ba96a042e2691',\n",
       " '617c55018474477f87fd528ff6cb6b7e',\n",
       " 'ed4008913a76432099cfddd0e2e4a194',\n",
       " '57f1ce97f4524fd9988f454c0e01d360',\n",
       " '9d1bfeb7fadf469d9ef924db2138f5cf',\n",
       " '7b71d8b17f2340fc92eaffb5d9978cca',\n",
       " 'fd22a8e92bfd4edcbed134862a8d0982',\n",
       " '5565d25e06124b2a907bc3bebbaca53d',\n",
       " 'b4b6fd22b9634693a97bc5229b002024',\n",
       " '26ff702e0b624ad6b971130e538f76e9',\n",
       " '0c09e9ebe95142bb91296071cd937841',\n",
       " 'c1c11749993f4448aaefcb35c0261c9e',\n",
       " '038698a767e54e67a965bfca683bc965',\n",
       " 'b34e5653f3f9426fbd51eff422556077',\n",
       " '2783a7e1572e40aba377c5f7be232069',\n",
       " '5f55dc1af9d442a8aef28dd1dba318f9',\n",
       " '952ff8bbe9934c8eb297001a879f7a53',\n",
       " '63bfb07dd07d4f87bf8cac7cf93b86eb',\n",
       " '96dd75bfe31848a49d4caa8314f0c2b4',\n",
       " '7595c334b1894adabee033d5f4f595ba',\n",
       " '21d2d369a0664196a7c510a4cff67c9f',\n",
       " '93e093909ff6421c84ac4b2ab1c609e3',\n",
       " '219ed06d1fb34e7ebee076a29fb99ea4',\n",
       " '875210eda42143aea7bf02e92a8aa73a',\n",
       " '76f5724160cd47d78ac0b0f5909f2aac',\n",
       " '9b3cda1c88f544618e586012e758eef8',\n",
       " '1e1c72944a6c4d2e8e20f113f67d6955',\n",
       " 'c72f5788849d4ec2ba126d990e3e7e9b',\n",
       " '9f7d3ee84f9a4c5e93ec8f72963b387b',\n",
       " '49a7475e048c410eb43a248e05b45880',\n",
       " '90d07f12d1fb4e90b2cdcdfd6d45e465',\n",
       " 'a21f34dc77d24f4bb796a7eca210d2de',\n",
       " '7ff9168a1a6747cc850d553b39036b94',\n",
       " '004c6dd8da394690bb6ec95ef91c923c',\n",
       " 'c46bb5cbf74943a3b3253f0d59343f5c',\n",
       " '125d6f55d4af4ca79bc16f5442294cec',\n",
       " 'c270f40650eb4af4a0c4cee31d5d9dda',\n",
       " '02468a42458b4c7b9e7e03e216802c02',\n",
       " 'e20a3a2b5e6f45c1afc956fc5736a51d',\n",
       " '3c7345c41bd24e719845e625d01754c4',\n",
       " 'e18704e193ab4524ac72bc83e2104720',\n",
       " 'daa3982ac1bd482180f1628c66be0071',\n",
       " 'c680aed4173f4aceb3c2374eb78b81c9',\n",
       " '5a40d124f4e74394ac42f149adc901f5',\n",
       " '67a0d715945b4f7695e04fc2bc5b767a',\n",
       " 'cb64499c6a8041ecafe5839b39927d02',\n",
       " 'f459e697fb614721952c591bd0e4bed2',\n",
       " 'd0767128d1994d839a8a0173fe99e8b9',\n",
       " '415a5ff02441421892ebe660eebec90e',\n",
       " '952698da4da3480c921e2e5a64853b01',\n",
       " 'a3589e5070c044d29232ba860df2bf7e',\n",
       " 'c075c603f0e84a54be3b0b0feb008916',\n",
       " 'f9e95999e57d494f800a21fc3c8a89c0',\n",
       " '8c7505fa9b04495faaf7826a60d158cc',\n",
       " 'ecb18a982d6f489dab2284135f1e1f32',\n",
       " '249cbbebd102453cb7f30cd2c3182d81',\n",
       " '42f2f8c39ff945c2b62b735458efb63b',\n",
       " '4b718d358cd44742b9c9a2b303e14630',\n",
       " 'aab1297a5dd24e1fae8536add69de78f',\n",
       " '375eafc16cfe4a51a478e3feb4a51a2f',\n",
       " '0e16362e5561495b943a93824f571084',\n",
       " '627c770cf00d4c02aa0527a2a9483e2c',\n",
       " 'f2ffdc309e414422bdd51a7177ac96e7',\n",
       " 'f83c7ccf65024ecf8223123c48874374',\n",
       " 'a4e819f84287456c94619ef1bb46001b',\n",
       " 'ca0db85f5e904570a9199842c62b2a0a',\n",
       " 'ea967520ea8b4aa1b6387eff93a57222',\n",
       " '468fe918fd1b4b4dbc13f49e0b10eb81',\n",
       " 'da3eb23c504b46af83bce01bfed9669f',\n",
       " '92a59bd1ee2d4d3a839f3ebfab91c647',\n",
       " '52b91d47a7a0464bb4df2d3f43905626',\n",
       " '9fc76c52135b42fa8197cb30356b114f',\n",
       " 'fe7330ce5fa44da59c493789fee8959c',\n",
       " '52a00cdffa71499d8e2962f300a4c683',\n",
       " '09a3c1cb8ec741cc98890eab8085da0f',\n",
       " 'c71ecf0579dd41b39122a97f6ad31a6e',\n",
       " '2361da7ded0348ed89052ef3a64fe09b',\n",
       " '0c899e64cf1b44aa99a28b622d2d8301',\n",
       " '3e763b7a04d6405f8f25c0df4d0c138b',\n",
       " '2943d83a290543e2a3d6b006399a7332',\n",
       " 'e89985068fb344079876454f1f62de4b',\n",
       " 'c023e219e94641638641a894a8dc7a4f',\n",
       " '2d80e998bf2b415aaeb5702e61e77d3c',\n",
       " 'af59df2116f34bd6a60395731afa48d6',\n",
       " 'f739c08f29484a798a77a2bab39a7c3b',\n",
       " '65ce65ed9d1c4dd9bc63018d1270333c',\n",
       " '76ba0d79bfa64c239d6b0f428d15f9df',\n",
       " '1550ccc9d2b94765853e44485057f79f',\n",
       " '9345d3a565c94f7788996554d42eb69b',\n",
       " '109a670e2b6c4026892838cc981cec67',\n",
       " 'a08c02adf26e43fcbf4698740486573e',\n",
       " 'cd8502a266624d4db27414a27ff8039d',\n",
       " '32a8fae55dc649268da627032af58b4f',\n",
       " '0fe7f309a8754c79b0b347a1c93c0903',\n",
       " 'fa5092e5752140ed909c7d655996d76c',\n",
       " 'd5616e2f4cff43f6aea9f9618a8dbca2',\n",
       " '09b1e3ff7cb8447a95cbd2b61e0985dd',\n",
       " 'eb4b5166bbf4490ca51ccfde56a1857a',\n",
       " '34fa81ad22f14f69a2fdc74b44cacea6',\n",
       " 'fbfbbbd24a134bfe92c95db2d3982d32',\n",
       " 'f7790167c1304491ba1b56ab1621ecef',\n",
       " 'cba01b4f801747bbb04b48e1e39539b9',\n",
       " '23ccfedea6dd4596a6342ea8e3e86f8f',\n",
       " 'aad31846891647d68feda48f34edac92',\n",
       " '241854e3c5774c088ef6d34f92912858',\n",
       " 'b05e6fb1e22f4ffb89463a59deb48466',\n",
       " '817cb854d81b4a72a66ed6220bdcac50',\n",
       " 'd9fa2a919a42418eb02292fbd5c3c2f8',\n",
       " 'c33d118e4a274563843d2267dd9443ef',\n",
       " 'a04395ffb3e74e558a111d499c161a03',\n",
       " 'c2c7b568b29248179efe9476ebed0364',\n",
       " '72cb056915034469bd8537049576d32a',\n",
       " '449613b0d67f4d45804d2c9433d1da67',\n",
       " 'd7408f99651e430cb2d6994ef26190bd',\n",
       " '89d4a730fc3940e3b42c6bf977896972',\n",
       " 'e022750d93db4c2a804bf0e9614f2ad7',\n",
       " '878c7cf80ceb44268285f549cdc225be',\n",
       " 'bbc7e9c13ee94bcc88e393df8d054240',\n",
       " '08903fe2a05744da9168583f2379ab5d',\n",
       " '1cf4380ab8944938914d35c4a1a174eb',\n",
       " '5d415db77c3e47c6945ca5168c789964',\n",
       " '82faabf704ff4baf8f6ca45dff3b136e',\n",
       " '4b2bbb65fa974d6681152da8fad54c25',\n",
       " 'fd8b44fcf8b24585bb0a2f667c4c6ad3',\n",
       " '8b9fc96569054dbf9380a3316a874e52',\n",
       " '93c3c92f7fef494f8a812ad93bafeefe',\n",
       " '4476819094f24a1c855dd7c73514ceed',\n",
       " 'bd5b744059574c629d27c62d501e4fd8',\n",
       " 'b22738c875ee4321bd724bb1e2b4481e',\n",
       " '2ea72c3430424055ba50a1815ddcae4a',\n",
       " 'd8cddbee4a9f42ea812bf4e7fdec4c84',\n",
       " 'b46d8d48ecff4cc5a14741e09e50c104',\n",
       " '693fecf5952a42fa8cfb821a9a2e40ef',\n",
       " 'ca330f9f226344989c07ab0b4d2107c5',\n",
       " 'e1a4fb111ff143a0a2fc58245899df4e',\n",
       " '161a4cac091c4bfabb5d9fb2b82f0d3e',\n",
       " '30141708e5504e3fb2bc8585997923b7',\n",
       " '10f2d82dc2a540ebabd3f581d24fcef2',\n",
       " '3779a11e1b984b149bdddce985caed8b',\n",
       " '23c5c5a2a07140cdba17a466f02800b9',\n",
       " 'c81b5c2e6faa41c9921e1a3b2bfce67f',\n",
       " '4276573024094bfeba17513c1d330954',\n",
       " '29f1777185104a228fe33941a9d204bc',\n",
       " '44a14d97f68b4b66b63d018b7e265482',\n",
       " '8a9ba2f8dfac4a579310aed8dc94875a',\n",
       " 'f4db803620a54acda9517e10b3fd4307',\n",
       " '69190687ef7a4cbd8d8cb7f02b156088',\n",
       " 'd5635dbfe732456684abe791b24a1e76',\n",
       " '2b367d71c0b7432898c14f9ff1850378',\n",
       " '2fe0322d5b854fa9928ab5bc135dd7d2',\n",
       " '941c4c4b8c30432e8e423b0670cc067d',\n",
       " '43e772ca132f4a2c83b84ae12fe2d339',\n",
       " '126d123b98e648d788b0bbd6d6bf3091',\n",
       " 'd67c67b36d3840749957f213f0d602b6',\n",
       " '98382c9e3845400181bfb3a956135031',\n",
       " 'e74801c129c14887a3edfdece125383c',\n",
       " '7f31adcb0e654a9283b95c2f21f66fe0',\n",
       " '4ccef251c11341ceb7a11ecd2470bdeb',\n",
       " '3b80a41727824b648d3e6626766dddee',\n",
       " '150467d54e314b409ecfb6e0b7007bb2',\n",
       " '3c8c7e1811ad46d4bceb10fe4522bea8',\n",
       " 'e026d55c0de346e0b5d79a685f638b41',\n",
       " '830da0dbbf0844c3a5fb06990d56a833',\n",
       " '17663074d95b44329cdb1e5e43655532',\n",
       " '26261c56edb548c8917142360b29ff5a',\n",
       " 'bb63022f16fb4636bad6da4b262ca20d',\n",
       " '849afb7703324c61a04eff62850debf7',\n",
       " '0c482794cf24492ebfffb4b763784e82',\n",
       " 'f9a953115f244c13a1aa180d4d6c4b43',\n",
       " '7e95350e73994773ac5a1de019db7dd5',\n",
       " 'b83d9d5dde0d487bae20231caf999aff',\n",
       " 'b957147737a44a10ad4f22645c3adc30',\n",
       " '6d9a8138800747f1bccc69bff15446bd',\n",
       " '1564c307252747f1be0be98e8cfa6020',\n",
       " '3c34c25ea0c549c7be7a16781ab3457e',\n",
       " '336ccd983e9a4246bf6a08d96a514c8a',\n",
       " '462eace606354f0ab9876ea06063251c',\n",
       " '421d75b10f5c48b18ab49db6f49c586c',\n",
       " '4a460b5ed6194215ba62c5b02a2f53ee',\n",
       " '60056a770c2148698d1a497462bac0f9',\n",
       " 'a9c9dcaa63794d7da53407605e0ce350',\n",
       " '1d196ade9fa1424aada94f8002a3df0d',\n",
       " 'c5a373f360c54a728768b5640af6c031',\n",
       " 'a52cc9b194234479ae3adcfe946615cb',\n",
       " '85185ea180814d1daf3451d9606ca2b1',\n",
       " 'be52e0a1dea0455aa0f7517766acfe34',\n",
       " 'c3f77fd2520b4b74855b9d3be2cef909',\n",
       " '70964a0c46a54db5b8c359da4f171401',\n",
       " '815d8297157448b087da0d3d61866806',\n",
       " '9c20e5c57d2e46b0b7206845944947bf',\n",
       " '7241a152bbe44de694a9ce2b4883ae68',\n",
       " '9ea78917733e4a678094ad27883ceea3',\n",
       " '943ee268b8404415af5b2db49997d05f',\n",
       " '533803ececc848fea2345115ca33b306',\n",
       " '52a790c32ba14140852c2885e357f1b1',\n",
       " '953b8612ac864f0aa74b809fbef5b185',\n",
       " 'b13511809a7647d9ad9c4911edd1f289',\n",
       " '4bf17a6b800946dc9dd78531d4c3ff0a',\n",
       " '0dd13b017f1d42ca8663e1f973efd9b5',\n",
       " '4542d7dce06b40b3a564033d2aa12ca7',\n",
       " '1e2e758ca0c44af3b8733e94dbe0f5eb',\n",
       " '6618c316e271496db3f0d4ddf5d26e49',\n",
       " 'a3805cdcaf184a1b8a61bbce5f027bb2',\n",
       " 'd4b609afada145a692c2d6cde0d55a74',\n",
       " '8aafb971776a459c875a8612776647d6',\n",
       " 'a42e441843204224903e15b1786503ae',\n",
       " 'd721ba7ffb3340eaa68626ff322a9c9b',\n",
       " '55fa4c1a057f4ce29bb213f02b75185f',\n",
       " '1c15bde571d44c6db419e0d99e60789d',\n",
       " 'e0af70ea09c949998d9a4a7a1b2dbcfd',\n",
       " '8d4e4a2461fe48e8b3e58a4fb5bfa62c',\n",
       " '28357f7399c94893ade78c4db2916e6f',\n",
       " '9da34748ea13460b81d35dd5be3e6e3f',\n",
       " 'bbfff2d930e0447aa084b1fa8d7b0bfc',\n",
       " 'cac37cf114964daeab20a8cf5d6d514a',\n",
       " '90160991331245b9a48bf3db11c39783',\n",
       " '3fa9310241dc4ae6bf3e3fad9aa44214',\n",
       " 'd16dec1a6f0d4428ada23282729de392',\n",
       " 'ec7631b834b14968a52b2f2b6905958d',\n",
       " '5664f8070e9249d8829f147e11b59504',\n",
       " 'b2a6262e41f847de8165cf34d56791f4',\n",
       " '57955c01151d432b83ead70e33bf2273',\n",
       " 'ecab9ec6e7454b5893ed645c562851ef',\n",
       " '982bc650b7d943eb975cd6d1456d078a',\n",
       " 'c4c751f45101444d8a0e20e753d6907d',\n",
       " '0efed10cd2e048bea91288b51114e557',\n",
       " 'e81401629d47421b81e08622d1def64d',\n",
       " 'df2f610f2ab640c8863c49f97f556eb5',\n",
       " '9b005a1c70574350b545ac8eb0c8c22b',\n",
       " '36bfb28241614167b8e11ec54b12cb4a',\n",
       " 'a7beab81995148b6a97b715665d9625f',\n",
       " 'ca11f5e24e9d4feabedbf9e75536ca89',\n",
       " 'b16d8d744d9447d88a69798ea92a06c8',\n",
       " 'c0f864002bde4e14bd3c4de2c65e492b',\n",
       " '432a3dc363964c97a93b2767031ea948',\n",
       " '8fc8671755614552a45a99770d2a0c54',\n",
       " 'f5724682a9d64179ad7f78244ae56993',\n",
       " 'c8900cbd17fd450aaa9c5c8f2ef5049e',\n",
       " 'fdd0deab6bf7463d9f2eb3f58678d185',\n",
       " '56f0c0ed9e5e44d9a168cc5894c64869',\n",
       " 'cafb63f4b3fa49089d8f543be1943d26',\n",
       " 'd96cdff5bb944a9ab6fbbb8743b6c3ad',\n",
       " 'cd8c04f7215f454cb58883a06227e736',\n",
       " 'aa374e7b2afa4a928c79f293b27fd8d7',\n",
       " 'aa7b81b29da645dca9a509c0c2534fb1',\n",
       " '2f1427355c4841e9b63c82294fff989a',\n",
       " '6cce3c66ca7b4ca08d94f148142d35d3',\n",
       " '903d6c9aa9c74c04b2266fbee961ab1f',\n",
       " 'f1a8b810db5b4a93b6d4bd6902e88caa',\n",
       " '4233b65a8146460a90348246a28377d8',\n",
       " '67976cd7b0544ab99b3a9f203640ccb6',\n",
       " 'b4849f58e40344d6bdd01083447abf24',\n",
       " '427321953f824f9098ef503cd3c638b5',\n",
       " '11486551bf7d4f30b5abca1ad54d6ea0',\n",
       " '1ae607b90cfa40d89c143190d41abade',\n",
       " '37eba91b057d4f22a11aeb392b4f6555',\n",
       " 'ef60a658386d4c0b84b91b46af9f998f',\n",
       " '3ba13214ccd14cfba98f4372aeb76663',\n",
       " 'd9daa814899e49e9a090fe616ec2ecb1',\n",
       " 'ad04762a784f4e1496055529bbba116c',\n",
       " '93b4f242e4eb4aa7aef6e729a37f4c21',\n",
       " 'ebb12c2f9afd4f1f9a70da721e359eb5',\n",
       " '2d6f4182c1b54f1d9943a19e9ba03d79',\n",
       " '916863d091554dc59098985ae8079c1f',\n",
       " 'f725f3664bdf452e9da42fb529cfcba8',\n",
       " 'ca0c722085ee42bc98786a904449bd40',\n",
       " 'd5ed6c0df1f74f30864c4e79fd41564a',\n",
       " '22886a9d1a1f42078e1348322dab56b3',\n",
       " 'baeb1d4529284daabf7056c01d6e4ad0',\n",
       " '6808507a09ee42d6ae38b028769e8de4',\n",
       " '15c9cba5ee8c499692cac77219ec6e10',\n",
       " '310793a6e9ea42228bc4620efa7f2f06',\n",
       " '64729de4d5244013b408f0d192f68be9',\n",
       " 'dc01acdc7a6449bfb3bdb934b2c22103',\n",
       " 'f8a9396e066c4042bc7e68869f8708e0',\n",
       " 'cceec3919b214e8c80a535885cd7721d',\n",
       " '40e9bb78cf6a4ba69ff8ee38c3bc89ec',\n",
       " '5443989562164d86a03463c6893e0224',\n",
       " '95423404a338477883c828b6d5a6f70a',\n",
       " '1bfdb3d09fb948bea3d3056d1c99b5f4',\n",
       " 'b0333631e0524ac8aa85746729788a2f',\n",
       " '546465fa34c14310933523190a3ec2d2',\n",
       " '87cf276efd624228800af71390debb24',\n",
       " '7ecc42a1062f4fd4a4c89509e52a3b59',\n",
       " '7a7a8bd0ba4e460bade51be712a7876e',\n",
       " 'a944ef03833a44178068be253f060c0f',\n",
       " 'c93e3c79a52840a3a04dc17dbcf3beb3',\n",
       " '192b3bf635fc4db3a7ca79884180196b',\n",
       " 'a951d5e3f86d4b8d809c26682ea339d1',\n",
       " '29817c8f8e084342ae0a4beb7bf3c0b9',\n",
       " '390228a2bc9f406f8569ba102fd1ab80',\n",
       " '45565d21a95a4b5db8bcfa0f59965234',\n",
       " '9a81516969904851b518a7ba9b8008a5',\n",
       " '87f63861bdf146598434077a599c0865',\n",
       " '3d3fcbeae9214727aee953fad4db1355',\n",
       " '80d26614f7e44f49babd681cc8e7c478',\n",
       " '496b880e75f544beaa7995479b965722',\n",
       " '48a108a146f14ab9aaba948a944bc07e',\n",
       " 'fb9a8da9d39241b692e2d012c2db1b6e',\n",
       " '3f65ee6c5924493dbe80870503f8ab42',\n",
       " 'fd9f257745784068a0501e9c5ce6259f',\n",
       " '8a47a9d870584ee5a265a1ab91e3965d',\n",
       " 'e9c46a5188414fb7899ad2bcfac1257e',\n",
       " 'c75a3f4edaad47e58a323bf1c5d8968b',\n",
       " '628b2fccfa284cfe9fa04203d45b307a',\n",
       " '9e5f0e4241b0456a8495020db3261f5b',\n",
       " '277809310fce4df28b96687d2c3a90c7',\n",
       " '005dc32559044009beeb6c9ed27fc945',\n",
       " 'f7b756cb4da14244bcd57bb3e714bb86',\n",
       " '1f1897c0ef4541caa1d2f8af355b9fa4',\n",
       " '7a052e5a35a74871b29ab8a93aac2ec7',\n",
       " '704285ba2ba64d7193eaf5d2aa65dd37',\n",
       " 'e72ace4a68684380856392752aa13066',\n",
       " 'c7aebb9243234d81bc670d45b15c55a5',\n",
       " 'e1f5082eef9942df8da015f624333496',\n",
       " 'd74de40f3e4448bc8a4170713057d716',\n",
       " 'f9b1def33e8f45c6b77aa7fc3aa0289b',\n",
       " 'f358cef1055b44d6bccd13ceb231f1a7',\n",
       " '7b96e49879954827bfc3d872936f9a4e',\n",
       " '766b7f4cdaec4c56a86f1b4cccb561b3',\n",
       " 'c6c59249059840e2a10b7edb7cc6d12f',\n",
       " '549312c501f14626be1e9cef345206b0',\n",
       " '2271264835f24e88ae50676f3a263dc8',\n",
       " '9a9c9906e8054744ad4198f908efa752',\n",
       " '587b7b2099b44e0f95324d4e081ade9f',\n",
       " 'bd0ad57e08a14373a36779ff451b7555',\n",
       " '3cc884d73e724f8db7ddb59bb0c1fa03',\n",
       " '8b5eebc4b02b4143abdf9f6f5905725a',\n",
       " '623ec4864f4742dab5ddef360264d686',\n",
       " 'a91cf1d6e5a046b48938c213fb3c06f7',\n",
       " '2effba1f56da40d497a45270487b73e0',\n",
       " '5d154a9baa1a43628029c03c4630894d',\n",
       " '3bd7df530f034509bd94fbd097e2a34d',\n",
       " '07136bc2ae46449fb1f132217afbeee8',\n",
       " 'b62d2be18847448a8769e4bff1054452',\n",
       " 'f3f469ad4c2d4e7ba174bbca9a490022',\n",
       " 'a5b6b7daada94ddfa87b42ce0c52959b',\n",
       " 'dcf7b5dad60f44028e0f4b150a8ccd8f',\n",
       " 'f1f15c559cab42b3a0c145e7d4edb963',\n",
       " '7497c87c35db4d438c794f95ea9e7d78',\n",
       " '9f160e608d7d432a938c3b10af62e410',\n",
       " '50713e650d5f4de5b352b8fb466e8b03',\n",
       " '18ef296e3b464e959fbc36e0e17cff73',\n",
       " '08d728d9f9134669af736833edc0214d',\n",
       " '85abe112feae4594bd340c622b3f2d67',\n",
       " '5e9d30e30f4849e99c5b964c07a13d3f',\n",
       " '27dd2dc9703845279e7115de0ea43c13',\n",
       " 'bbb2ba30ced44818a7d125f601bd2a65',\n",
       " 'e64c589abee8454a863dda771ecae6f0',\n",
       " 'de835d3a687b4fe9bd76a03c7bae7800',\n",
       " '8febe3f467d34d9ebac1b0c1854a663e',\n",
       " 'a4e10a3db1014e5c803e6483e38c8aba',\n",
       " 'e36e25a53fb44d2792415561cf77c491',\n",
       " '0aba76e28d92417ea4e906531fee1c29',\n",
       " '522f81bf053548e0bbd8a62f27721484',\n",
       " 'f2d4629d7b1e407a9565ffaee1a61ef7',\n",
       " '057fa3fa1e73448baa12efaa2d4ce4a9',\n",
       " '8bfe700a8a1244069f625010a5f3d85d',\n",
       " 'ed9a4cfffd48458b8478b17490de9a1b',\n",
       " 'c8f155065983473ba024fc8d182aadf7',\n",
       " 'f55deef69ace40568dd6e76dcfcc7832',\n",
       " 'b47e10ce9b1949949139266db0deba1f',\n",
       " 'd8358177eb6e4a1fa34459f0e27259cb',\n",
       " '876029a74793499e9fc0cd45b03e7a9f',\n",
       " 'e083e470b9754988bd3eea07746f5717',\n",
       " '716ab83b938b4239b6f4ae0a1ee66286',\n",
       " '143b0c9b1e77427fb8e61327d42c33fc',\n",
       " '06bb825621c74b9a98468827b919276d',\n",
       " 'af77df2c49174252b01a32a8b429fbda',\n",
       " 'd202dbc5c2a049b69c6e1a853a10132e',\n",
       " '4fd1076f612a4f318d783a8e07ce2819',\n",
       " '4dda538a56b440498803639b33ac70a6',\n",
       " '9925a4e777544a8ea0455b7d40df7e88',\n",
       " '7f385d170b894ff1b0594176ee99b76b',\n",
       " '09289a7228644abaa36553f3ddf526f4',\n",
       " 'ba10ae3738ed43738783877a3eb842b6',\n",
       " 'f3de2eb9d38543eb93da4d6e3c6abdb3',\n",
       " '1411216b7065472c8c7f6592a41d730b',\n",
       " '615c1fbfd3fa49179680c5e78c5512b8',\n",
       " 'd90ddb874bc44fa59e8d1c0249a52cae',\n",
       " '5d6fb0ccfa994e0490a71bfe66d8fa52',\n",
       " 'deafb8e708cb46fea12f4dbf0bcd7470',\n",
       " '6e1009a4bd4a46d78955a04ff7b08496',\n",
       " '5228a640e5614797bd28a705440f0aa6',\n",
       " 'e66d1265164e43eb97b68427d3a6dc16',\n",
       " '60f3864cbd60420ba6e6ab17ce9eadc9',\n",
       " '93be41ce5ca34258856df567b0f3d4ae',\n",
       " 'af2ee299550d45e3aeda2eb011f335df',\n",
       " 'f19f9954ff484372908385d7ab98a982',\n",
       " '3ab5b682ba93497eaf89acb2d137665e',\n",
       " 'f89a802aab5641d2a42cf1b3d561913f',\n",
       " 'cac83c1c17d644d2980f88ca03172618',\n",
       " 'd7fedc9c2c2f4715b6b76ae258864025',\n",
       " 'a680df040e0d4d0380319ea25f1b2aa4',\n",
       " '5c90f316307d4a3baa525e3ec16f4cde',\n",
       " '50f2a160eecc4c698acea8326d77a667',\n",
       " 'd640199ab7684c8aa8e15037da3a5b76',\n",
       " '6dbbad8184114b009c8227779f36fa2c',\n",
       " 'f85cd0a759a94789a3815bb2f75d8552',\n",
       " 'd94ab3733fba4855bbabea3d1e3830ea',\n",
       " '2cb38d07678644e3b3b7ba5b871415a2',\n",
       " '1455c59d08d449f6aaa864f29f7bc07d',\n",
       " '1bc59f36d7504df5be1d14f9432818b9',\n",
       " 'c8536e41357c4be1839a0e825fbdb65d',\n",
       " '8be5af3f304f454688adbc99fd93f4f4',\n",
       " '7354e5364ec545ed98724649ba3dcc25',\n",
       " '77ced35c95ef40f5a4d6da737cec856a',\n",
       " '6d92d464c07b450aa222f1dd03c0897d',\n",
       " 'fd62032c524047bba8fea296146ea567',\n",
       " '0e3a18618a7d457bb21457885aafc398',\n",
       " 'eb45ba05199d40b5a81d029d7b47461f',\n",
       " '319e7b33225d4eb3972868e8c646ffe2',\n",
       " '136f4127628342868471c1ce1812d2c8',\n",
       " 'b378646c1a854a43b71ccd257b8b485d',\n",
       " '1a3e9fdc42734d0a9b8335603e4dfadf',\n",
       " '6991ed8f57bd451da3f6da92c33b2ec7',\n",
       " '58ba406c58dc499b8798e0d171df341b',\n",
       " '6e637e81a8124fdaacc623e87de765b5',\n",
       " '16fb1ab725044c8a83906664efb42ca8',\n",
       " 'fa1d529855d749d6b63b4b50666ed6cb',\n",
       " '98712dfbad2c4cf8a18dba3b89f4a4c5',\n",
       " '653e7ffeb40f4827aea10b4536fa9cb9',\n",
       " 'ac8e52ef82bd436f893d22a5f78c60e9',\n",
       " '173ddb61cdcf4bcaa17134de8cc36935',\n",
       " '9d30a7c4d64d456498e8e7836de1e3f0',\n",
       " '0b640ed961044e59a74de1cb600c41a3',\n",
       " 'ee0e9150e3904cfe96288bc16e1b530d',\n",
       " 'bef3615b58da4f43b61993cc74b17f69',\n",
       " 'a753491462b84b4788be67648b96cf5b',\n",
       " '30ff764ec15341bc9e4bc63a0c7a0f00',\n",
       " '3544293fee9e4b0981f6a4dc8706239d',\n",
       " 'c9ac3e68c52c4ea88ae27778777b179d',\n",
       " 'bcfd2f717520438aba2582c9d0aa9846',\n",
       " 'e8ea858f8964499fac41268eae0552b8',\n",
       " '337c25c09aa24cc78d1f883151fe84bc',\n",
       " '40664ee2fd81488b8c15e7c7b24d224a',\n",
       " 'b8d0b69a67834d6bb697039fa684f5cf',\n",
       " '6cc92ba42c2e4a4fbf1f14a37d11952d',\n",
       " '0706106b1e824d0790e8c5f21bc4b9c6',\n",
       " 'fbb2104b3ed5476abe769b8e9079d0a4',\n",
       " 'b6fbe9dc8b4d4a3481824a890aed11cf',\n",
       " '79f5b21422f44fabbfa4243e2d9da694',\n",
       " '483350d3ca0f4df9bf05ce68b630cc3a',\n",
       " '8ae622c201914caca4addc9d03e195df',\n",
       " 'dbfc60c4324b4f0486935e6eed50747f',\n",
       " 'e6eee69f4b4840efbaeeedbb7eb6e007',\n",
       " '7d9d3c0edc0c41e3b959874bcdad5e75',\n",
       " '8a303738d17c4d00b27f519840996322',\n",
       " '3cf17b56d1584625ba254f28248f88dc',\n",
       " '79554673e3b2440493e2efc035f15bb4',\n",
       " '2a715d607c7f4089a50977531b85f1dc',\n",
       " '36c6be80f6554513bb39cc368d4a150e',\n",
       " '76f34318eecd44f7aeb8001e8ac45cd1',\n",
       " '9a73384599244fc19adddfa1e96a4103',\n",
       " 'a9299311e92847ed988716ea2bad2592',\n",
       " '36bc1c953cf54fae9b350156894b9536',\n",
       " '74c05dcec4a14d048c0158201e5f916b',\n",
       " 'bf0b9a9d8e31494399391c3f3e53ad6b',\n",
       " '43b8b9088f3949c3b01ea2513880e6ec',\n",
       " '15f453eba2164272af541e47c17934db',\n",
       " 'cb8ed3c6dceb4f63b2d7faf0cd987b00',\n",
       " '6d4d5966e10a4b8abeb4fa7c53302f7f',\n",
       " '599befa02ab8429d85be248fe59c7371',\n",
       " '1d8c8e70e7d74e608eada6688bd3775d',\n",
       " 'f9142cabe4404552985be0c5d451dccc',\n",
       " '8c771f9369ad4ec3b5e0ef253bc58a36',\n",
       " 'e205d61fc84949a989b678a25023ca4d',\n",
       " 'b8ffdcbb9a2c4d8d9ab8d54bc46ad47d',\n",
       " 'f9ab7b4dc31d49b7a2aeee0a1e4060da',\n",
       " '4df63818af02497bb7428a8796a17c08',\n",
       " '75409f6a6856406387c68a19c79895f0',\n",
       " '52fac16b7b964728b6b20df9fdb4f9a9',\n",
       " '74ad857a2c734267ac72a3fba8a2ba60',\n",
       " '1e986537e8e84c99891637da44ef1db1',\n",
       " 'ad09da78a2744845b889057fe93346c1',\n",
       " '3088ff6997044204aeeb35ae3c6e9a82',\n",
       " 'dd8df31f930d49d3a5608c1c2c86a67b',\n",
       " '4779ebaa6a884fa083b9583355331931',\n",
       " '83fb5b155eb0491993f6335eecfd5158',\n",
       " '9e7279703134477699db89c06aea57d9',\n",
       " '4d2fe45e15044eb29212091f14577e10',\n",
       " '99664d519cde4a0d859246ed151cfd0d',\n",
       " '249306d7bf864559979a878fbc1dd8fb',\n",
       " '5a5fe28fe0024d09b44b219b4b333178',\n",
       " '71e5eb7590cd42d7b83644aa9b2eb86c',\n",
       " '1e51a394229b4cd4939e4bd204521f03',\n",
       " 'd5a341ceb2ab46fda7c1026961af0bf5',\n",
       " 'f85b5a3ce9df45759747f51bbbf6b336',\n",
       " '6fade7c942bf4b1a9836e7ca0002a94b',\n",
       " 'ec439b2e255f404e98445ee002d03ffb',\n",
       " 'd8fc896cbc3f4191be6fda867042f757',\n",
       " 'c040495bb88b4f70b5d9c98bc47bd4c5',\n",
       " '2840d8f813854435a1939a6b5942720a',\n",
       " '63b329be4cab4be29895d1e5c12cf009',\n",
       " '35fb9b43ccd1475b89dc368e1a9cfa26',\n",
       " '76e98c8a0cb8476aa31b05c0d6edf27e',\n",
       " 'd57166008b224828ad412d6f95ccdc02',\n",
       " '50eb9b98893743bf82a9b974b3b34de9',\n",
       " '716c226d36a4440089f9cc12afae2652',\n",
       " '67a3778063114d40b2bc24eb6fb051c9',\n",
       " ...]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "client.add(collection_name=COLLECTION_NAME, metadata=docs_metadatas, documents=docs_contents)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "OFOxzL3PN8xp",
    "outputId": "db23acef-08e9-4241-f869-981889c09b77"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CountResult(count=4431)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Ensuring we have expected number of document chunks\n",
    "client.count(collection_name=COLLECTION_NAME)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0Hc5zeXVWJN_"
   },
   "source": [
    "### **10. Searching for the document**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "DjKP_kYkV5Ck",
    "outputId": "8c37f64d-c8b3-4ee8-d23e-b1ad916ba9b5"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████████████████████████| 77.7M/77.7M [00:07<00:00, 9.94MiB/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Id:  8b1d6591-1ab3-44d1-b50c-83c23aeb9b21\n",
      "Document :  Here, we show how the accuracy of binary quantization is quite good across different dimensions -- for both the models. \n",
      "\n",
      "\n",
      "\n",
      "## Enhanced Performance and Efficiency with Binary Quantization\n",
      "Score :  0.93278515\n",
      "Source :  articles/binary-quantization-openai.md\n",
      "Id:  d712ca08-cc4b-41f0-a592-7c0a06b8945c\n",
      "Document :  Demetrios:\n",
      "\n",
      "I see, okay, so basically this is the quantization. And now let's dive into the binary quantization and how it works.\n",
      "\n",
      "\n",
      "\n",
      "Andrey Vasnetsov:\n",
      "Score :  0.9211245\n",
      "Source :  blog/binary-quantization-andrey-vasnetsov-vector-space-talk-001.md\n"
     ]
    }
   ],
   "source": [
    "## Searching for document chunks similar to query for context\n",
    "search_result = client.query(collection_name=COLLECTION_NAME, query_text=\"what is binary quantization\",limit=2)\n",
    "context = [\"document:\"+r.document+\",source:\"+r.metadata['source'] for r in search_result]\n",
    "for res in search_result:\n",
    "    print(\"Id: \", res.id)\n",
    "    print(\"Document : \" , res.document)\n",
    "    print(\"Score : \" , res.score)\n",
    "    print(\"Source : \" , res.metadata['source'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vKSaosV1WaEH"
   },
   "source": [
    "### **11. Inititalize OpenAI**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "Fnne48JLWe-H"
   },
   "outputs": [],
   "source": [
    "from openai import OpenAI\n",
    "openai_client = OpenAI(api_key=openai_api_key)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "JhqBR3t7WGf0"
   },
   "source": [
    "### **12. Querying without the context**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "h-X69HiXWQzH"
   },
   "outputs": [],
   "source": [
    "def query_without_context(query):\n",
    "\n",
    "    prompt_start = (\n",
    "        \"\"\" You're assisting a user who has a question based on the documentation.\n",
    "        Your goal is to provide a clear and concise response that addresses their query while referencing relevant information \n",
    "        from the documentation.\n",
    "        Remember to: \n",
    "        Understand the user's question thoroughly.\n",
    "        If the user's query is general (e.g., \"hi,\" \"good morning\"), \n",
    "        greet them normally and avoid using the context from the documentation.\n",
    "        If the user's query is specific and related to the documentation, locate and extract the pertinent information.\n",
    "        Craft a response that directly addresses the user's query and provides accurate information\n",
    "        referring the relevant source and page from the 'source' field of fetched context from the documentation to support your answer.\n",
    "        Use a friendly and professional tone in your response.\n",
    "        If you cannot find the answer in the provided context, do not pretend to know it.\n",
    "        Instead, respond with \"I don't know\".\\n \"\"\"\n",
    "    )\n",
    "\n",
    "    prompt_end = ( f\"\\n\\nQuestion: {query}\\nAnswer:\" )\n",
    "\n",
    "    prompt = ( prompt_start + \"\\n\\n---\\n\\n\" +  prompt_end )\n",
    "\n",
    "    res = openai_client.completions.create(\n",
    "        model=\"gpt-3.5-turbo-instruct\",\n",
    "        prompt=prompt,\n",
    "        temperature=0,\n",
    "        max_tokens=636,\n",
    "        top_p=1,\n",
    "        frequency_penalty=0,\n",
    "        presence_penalty=0,\n",
    "        stop=None\n",
    "    )\n",
    "    return (res.choices[0].text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "NJo9cCF6Wkzj",
    "outputId": "ec0b1b16-5ca7-4e24-91d1-d3d4933d8487"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
      "To disable this warning, you can either:\n",
      "\t- Avoid using `tokenizers` before the fork if possible\n",
      "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Binary quantization is a process of converting numerical data into binary form, where each value is represented by a sequence of 0s and 1s. This technique is commonly used in machine learning and data compression to reduce the size of data while preserving its essential information. You can learn more about binary quantization and its applications on page 12 of the \"Data Compression\" section in the documentation.\n"
     ]
    }
   ],
   "source": [
    "print (query_without_context(query = \"what is binary quantization?\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "U0zSfcp9WlYO",
    "outputId": "196854b9-f88d-4b70-bc1e-bfb42abe0cad"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " To reproduce the benchmark for Open Source vector databases, you can follow the steps outlined in the documentation on page 5 of the 'Benchmarking' section. This section provides detailed instructions on how to set up and run the benchmark, as well as how to interpret the results. Additionally, you can refer to the 'Benchmarking' section on page 4 for more information on the purpose and methodology of the benchmark. If you encounter any issues or have further questions, please refer to the 'Troubleshooting' section on page 6 or reach out to the community for assistance.\n"
     ]
    }
   ],
   "source": [
    "print (query_without_context(query = \"How can one reproduce the benchmark for Open Source vector databases?\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fKioSnZIWqbi"
   },
   "source": [
    "### **13. Querying with Context from Qdrant**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "unnDrJjPWxDQ"
   },
   "outputs": [],
   "source": [
    "def query_with_context(query,limit):\n",
    "\n",
    "    ## Fetch context from Qdrant\n",
    "    search_result = client.query(collection_name=COLLECTION_NAME, query_text=query, limit=limit)\n",
    "\n",
    "    contexts = [\n",
    "        \"document:\"+r.document+\",source:\"+r.metadata['source'] for r in search_result\n",
    "    ]\n",
    "    prompt_start = (\n",
    "        \"\"\" You're assisting a user who has a question based on the documentation.\n",
    "        Your goal is to provide a clear and concise response that addresses their query while referencing relevant information \n",
    "        from the documentation.\n",
    "        Remember to: \n",
    "        Understand the user's question thoroughly.\n",
    "        If the user's query is general (e.g., \"hi,\" \"good morning\"), \n",
    "        greet them normally and avoid using the context from the documentation.\n",
    "        If the user's query is specific and related to the documentation, locate and extract the pertinent information.\n",
    "        Craft a response that directly addresses the user's query and provides accurate information\n",
    "        referring the relevant source and page from the 'source' field of fetched context from the documentation to support your answer.\n",
    "        Use a friendly and professional tone in your response.\n",
    "        If you cannot find the answer in the provided context, do not pretend to know it.\n",
    "        Instead, respond with \"I don't know\".\n",
    "        \n",
    "        Context:\\n\"\"\"\n",
    "    )\n",
    "\n",
    "    prompt_end = (\n",
    "        f\"\\n\\nQuestion: {query}\\nAnswer:\"\n",
    "    )\n",
    "\n",
    "    prompt = (\n",
    "        prompt_start + \"\\n\\n---\\n\\n\".join(contexts) +\n",
    "        prompt_end\n",
    "    )\n",
    "\n",
    "    res = openai_client.completions.create(\n",
    "        model=\"gpt-3.5-turbo-instruct\",\n",
    "        prompt=prompt,\n",
    "        temperature=0,\n",
    "        max_tokens=636,\n",
    "        top_p=1,\n",
    "        frequency_penalty=0,\n",
    "        presence_penalty=0,\n",
    "        stop=None\n",
    "    )\n",
    "\n",
    "    return (contexts , res.choices[0].text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "question1 = \"what is quantization?\"\n",
    "question2 = \"How can one reproduce the benchmark for Open Source vector databases?\"\n",
    "RETRIEVAL_WINDOW_SIZE_3 = 3\n",
    "RETRIEVAL_WINDOW_SIZE_5 = 5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### **Let's try our RAG pipeline with `Question 1` and `Question 2` along with `3 and 5 as number of retrieved documents` to see the impact of size of `CONTEXT`**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Quantization is a technique that uses quantum computing to improve the search process in artificial neural networks (ANNs). It involves converting traditional float32 vectors into qbit vectors and creating quantum entanglement between them. This allows for highly efficient vector search algorithms to be developed. This information can be found in the article \"Quantum Quantization and Entanglement\" in the 'articles' section of the documentation.\n"
     ]
    }
   ],
   "source": [
    "## Question 1 ; Context Size - 3\n",
    "context1 , rag_response1 = query_with_context(question1,RETRIEVAL_WINDOW_SIZE_3)\n",
    "#print(context)\n",
    "print(rag_response1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Quantization is an optional feature in Qdrant that enables efficient storage and search of high-dimensional vectors. It compresses data while preserving relative distances between vectors, making vector search algorithms more efficient. This is explained in more detail in the documentation on Quantum Quantization and Entanglement, specifically in the section on Quantization.\n"
     ]
    }
   ],
   "source": [
    "## Question 1 ; Context Size - 5\n",
    "context2 , rag_response2 = query_with_context(question1,RETRIEVAL_WINDOW_SIZE_5)\n",
    "#print(context)\n",
    "print(rag_response2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6WjCWzN7W7AE",
    "outputId": "6173e8be-15a1-4099-e7ba-abe36f4e2ae6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " To reproduce the benchmark for Open Source vector databases, you can refer to the source code available on Github and follow the instructions in the README.md file. The benchmarks are fully open-sourced and contributions are welcome. This information can be found in the 'How to reproduce the benchmark?' section of the 'benchmarks/benchmark-faq.md' page.\n"
     ]
    }
   ],
   "source": [
    "## Question 2 ; Context Size - 3\n",
    "context2_1 , rag_response2_1 = query_with_context(question2,RETRIEVAL_WINDOW_SIZE_3)\n",
    "#print(context)\n",
    "print(rag_response2_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " To reproduce the benchmark for Open Source vector databases, you can refer to the source code available on Github and follow the instructions in the README.md file. This information can be found in the document titled \"The Emergence and Relevance of Vector Databases\" in the blog/series-A-funding-round.md file.\n"
     ]
    }
   ],
   "source": [
    "## Question 2 ; Context Size - 5\n",
    "context2_1 , rag_response2_1 = query_with_context(question2,RETRIEVAL_WINDOW_SIZE_5)\n",
    "#print(context)\n",
    "print(rag_response2_1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**As noticed, the quality of response does get worse with larger context length.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fKioSnZIWqbi"
   },
   "source": [
    "### **13.1 Querying with Context from Qdrant with DSPy**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Additionally Install DSPy\n",
    "#!pip install \"dspy-ai[qdrant]\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/atitaarora/.zshenv:2: command too long: /Users/atitaarora/qdrant/workspace/qdrant-rag-eval/workshop-rag-eval-qdrant-ragas/ragas/bin:/opt/homebrew/bin:/opt/homebrew/sbin:/usr/local/bin:/System/Cryptexes/App/usr/bin:/usr/bin:/bin:/usr/sbin:/sbin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/local/bin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/bin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/appleinternal/bin:/Library/Apple/usr/bin:/Users/atitaarora/.cargo/bin:/Applications/iTerm.app/Contents/Resources/utilities=/Users/atitaarora/qdrant/workspace/qdrant-rag-eval/workshop-rag-eval-qdrant-ragas/ragas/bin:/opt/homebrew/bin:/opt/homebrew/sbin:/usr/local/bin:/System/Cryptexes/App/usr/bin:/usr/bin:/bin:/usr/sbin:/sbin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/local/bin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/bin:/var/run/com.apple.security.cryptexd/codex.system/bootstrap/usr/appleinternal/bin:/Library/Apple/usr/bin:/Users/atitaarora/.cargo/bin:/Applications/iTerm.app/Contents/Resources/utilities+,,/Users/atitaarora/Library/Python/3.10/bin\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
      "To disable this warning, you can either:\n",
      "\t- Avoid using `tokenizers` before the fork if possible\n",
      "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dspy-ai==2.4.9\n"
     ]
    }
   ],
   "source": [
    "#Verify the version on DSPy\n",
    "!pip freeze | grep dspy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "import dspy\n",
    "from dspy.retrieve.qdrant_rm import QdrantRM\n",
    "\n",
    "turbo = dspy.OpenAI(model=\"gpt-3.5-turbo\")\n",
    "qdrant_retriever_model = QdrantRM(COLLECTION_NAME, client, k=3)\n",
    "\n",
    "dspy.settings.configure(lm=turbo, rm=qdrant_retriever_model)\n",
    "\n",
    "retrieve = dspy.Retrieve(k=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top 3 passages for question: How can one reproduce the benchmark for Open Source vector databases? \n",
      "\n",
      "1] ## The Emergence and Relevance of Vector Databases \n",
      "\n",
      "2] That makes the comparison unfair. That’s why we purely focused on testing the Open Source vector databases, so everybody may reproduce the benchmarks easily.\n",
      "\n",
      "\n",
      "\n",
      "This is not the final list, and we’ll continue benchmarking as many different engines as possible.\n",
      "\n",
      "\n",
      "\n",
      "## How to reproduce the benchmark?\n",
      "\n",
      "\n",
      "\n",
      "The source code is available on [Github](https://github.com/qdrant/vector-db-benchmark) and has a `README.md` file describing the process of running the benchmark for a specific engine. \n",
      "\n",
      "3] However, we tried our best, kept scrolling the docs up and down, experimented with combinations of different configurations, and gave all of them an equal chance to stand out. If you believe you can do it better than us, our **benchmarks are fully [open-sourced](https://github.com/qdrant/vector-db-benchmark), and contributions are welcome**!\n",
      "\n",
      "\n",
      "\n",
      "\n",
      "\n",
      "## What do we measure?\n",
      "\n",
      "\n",
      "\n",
      "There are several factors considered while deciding on which database to use. \n",
      "\n"
     ]
    }
   ],
   "source": [
    "topK_passages = retrieve(question2).passages\n",
    "\n",
    "print(f\"Top {retrieve.k} passages for question: {question2}\", \"\\n\")\n",
    "for idx, passage in enumerate(topK_passages):\n",
    "    print(f\"{idx+1}]\", passage, \"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "import dspy\n",
    "from dspy.retrieve.qdrant_rm import QdrantRM\n",
    "\n",
    "turbo = dspy.OpenAI(model=\"gpt-3.5-turbo\")\n",
    "qdrant_retriever_model = QdrantRM(COLLECTION_NAME, client, k=3)\n",
    "dspy.settings.configure(lm=turbo, rm=qdrant_retriever_model)\n",
    "\n",
    "def query_with_DSPy_context(query,limit):\n",
    "\n",
    "    ## Fetch context from Qdrant using DsPy\n",
    "    search_result_retrieve = dspy.Retrieve(k=limit)\n",
    "    search_result = search_result_retrieve(query).passages\n",
    "\n",
    "    prompt_start = (\n",
    "        \"\"\" You're assisting a user who has a question based on the documentation.\n",
    "        Your goal is to provide a clear and concise response that addresses their query while referencing relevant information \n",
    "        from the documentation.\n",
    "        Remember to: \n",
    "        Understand the user's question thoroughly.\n",
    "        Think about the question in a step by step approach\n",
    "        If the user's query is general (e.g., \"hi,\" \"good morning\"), \n",
    "        greet them normally and avoid using the context from the documentation.\n",
    "        If the user's query is specific and related to the documentation, locate and extract the pertinent information.\n",
    "        Craft a response that directly addresses the user's query and provides accurate information\n",
    "        referring the relevant source and page from the 'source' field of fetched context from the documentation to support your answer.\n",
    "        Use a friendly and professional tone in your response.\n",
    "        If you cannot find the answer in the provided context, do not pretend to know it.\n",
    "        Instead, respond with \"I don't know\".\n",
    "        \n",
    "        Context:\\n\"\"\"\n",
    "    )\n",
    "\n",
    "    prompt_end = (\n",
    "        f\"\\n\\nQuestion: {query}\\nAnswer:\"\n",
    "    )\n",
    "\n",
    "    prompt = (\n",
    "        prompt_start + \"\\n\\n---\\n\\n\".join(search_result) +\n",
    "        prompt_end\n",
    "    )\n",
    "\n",
    "    res = openai_client.completions.create(\n",
    "        model=\"gpt-3.5-turbo-instruct\",\n",
    "        prompt=prompt,\n",
    "        temperature=0,\n",
    "        max_tokens=636,\n",
    "        top_p=1,\n",
    "        frequency_penalty=0,\n",
    "        presence_penalty=0,\n",
    "        stop=None\n",
    "    )\n",
    "\n",
    "    return (search_result , res.choices[0].text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " To reproduce the benchmark for Open Source vector databases, you can follow the instructions provided in the README.md file on the Github repository for the benchmark. The source code is also available on Github, and contributions are welcome. The benchmarks measure several factors, which are described in detail in the documentation.\n"
     ]
    }
   ],
   "source": [
    "question2 = \"How can one reproduce the benchmark for Open Source vector databases?\"\n",
    "print(query_with_DSPy_context(question2,RETRIEVAL_WINDOW_SIZE_3)[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " To reproduce the benchmark for Open Source vector databases, you can follow the steps outlined in the `README.md` file on the [Github](https://github.com/qdrant/vector-db-benchmark) page. This includes downloading the source code and following the instructions provided for running the benchmark for a specific engine. This process allows for easy reproduction of the benchmarks for various engines.\n"
     ]
    }
   ],
   "source": [
    "print(query_with_DSPy_context(question2,2)[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "iLkWOGItZX0S"
   },
   "source": [
    "### **14. Prepare the evaluation dataset to evaluate our RAG system**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 113,
     "referenced_widgets": [
      "1725ae2cf0c94aa6ba6e19ced1901e2c",
      "2b95ee150f6d4487909dff3c231cb429",
      "028943c4019943f4b57c383d573efb3c",
      "80b1dff28f5f4b30827a9c8189fefc30",
      "dee469d9cfd44d9181c313da9bc195aa",
      "f818582473954544bfd497fe7f19a640",
      "f57fb5f982594ae98c3bb71ce414e760",
      "fe20966df20042ef8dc66c133f261079",
      "6dc07cc1c9c6463a8146c798772269af",
      "d4f0f13ed6104f51baf7ed6293638662",
      "28d59a9ce5884e498c4b7763c79f1a65",
      "4d555cff165a40fb8cad76acf7c2712d",
      "1a0827041a0b46f3bc050631b29bc066",
      "1ed740b281d341ee867d782f53097ed3",
      "1dad3b7b797e4da994566242893f5d4e",
      "aabdc3a15dc54497a14a7132882082d8",
      "ccc3ce4362ea4cbbaa606fe28b74cadd",
      "18e47dede65141d7a11b1c0a583d5666",
      "413c186f31fd4f0dbd6e5956c9d31d93",
      "e3d7cae3f6244ad9990cd984bc2e4855",
      "f7169503c1b049149bab032496d378b6",
      "41606d866ebb4172989d79d1eea7139f",
      "7bd6c2a1e3534f88aa92443875e18f43",
      "16c72bd3bc614955ae48760ce44e67e3",
      "b0b5c0f5bed0478e9a82df8a6ea5d201",
      "ac83ceeb91bd492490de27f7a5810501",
      "6371a34a2a02480cbc60328fdab8cc5d",
      "a8375dc193b8470dad9c73cffb87731c",
      "58290b9bd8824604aac3bc5e9d7c7f20",
      "d015e097e6094b6891e15fdb04d34d25",
      "b6e7dd16ab144afeb9f3b8ba49d4aea4",
      "115ad0595f8d4ac7852a5983f5f1b2b5",
      "564a657b517a4dcbb0b24515bfea4e51"
     ]
    },
    "id": "Hvpes48gj4TC",
    "outputId": "ca6e7115-1d06-4266-840c-3c5ecc8fbe95"
   },
   "outputs": [],
   "source": [
    "from datasets import Dataset\n",
    "\n",
    "# RAGAS Expect ['question', 'answer', 'contexts', 'ground_truths'] format\n",
    "'''\n",
    "{\n",
    "    \"question\": ['What is quantization?', ...],\n",
    "    \"answer\": [], ## answer\n",
    "    \"contexts\": [], ## context\n",
    "    \"ground_truth\": ## answer expected\n",
    "}\n",
    "'''\n",
    "def create_eval_dataset(dataset, eval_size,retrieval_window_size):\n",
    "    questions = []\n",
    "    answers = []\n",
    "    contexts = []\n",
    "    ground_truths = []\n",
    "    \n",
    "    # Iterate over the first 10 entries\n",
    "    for i in range(eval_size):\n",
    "        entry = dataset[i]\n",
    "        question = entry['question']\n",
    "        answer = entry['answer']\n",
    "        questions.append(question)\n",
    "        ground_truths.append(answer)\n",
    "        context , rag_response = query_with_context(question,retrieval_window_size)\n",
    "        contexts.append(context)\n",
    "        answers.append(rag_response)\n",
    "    \n",
    "    rag_response_data = {\n",
    "        \"question\": questions,\n",
    "        \"answer\": answers,\n",
    "        \"contexts\": contexts,\n",
    "        \"ground_truth\": ground_truths\n",
    "    }\n",
    "    \n",
    "    return rag_response_data\n",
    "\n",
    "\n",
    "def create_eval_dataset_for_dspy(dataset, eval_size,retrieval_window_size):\n",
    "    questions = []\n",
    "    answers = []\n",
    "    contexts = []\n",
    "    ground_truths = []\n",
    "    \n",
    "    # Iterate over the first 10 entries\n",
    "    for i in range(eval_size):\n",
    "        entry = dataset[i]\n",
    "        question = entry['question']\n",
    "        answer = entry['answer']\n",
    "        questions.append(question)\n",
    "        ground_truths.append(answer)\n",
    "        context , rag_response = query_with_DSPy_context(question,retrieval_window_size)\n",
    "        contexts.append(context)\n",
    "        answers.append(rag_response)\n",
    "    \n",
    "    rag_response_data = {\n",
    "        \"question\": questions,\n",
    "        \"answer\": answers,\n",
    "        \"contexts\": contexts,\n",
    "        \"ground_truth\": ground_truths\n",
    "    }\n",
    "    \n",
    "    return rag_response_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "id": "gQnaLMIcv3X-"
   },
   "outputs": [],
   "source": [
    "## Define the Config for generating the Eval dataset as below :\n",
    "\n",
    "## The dataset used to evaluate RAG using RAGAS\n",
    "## loading the eval dataset from HF\n",
    "qdrant_qna_dataset = load_dataset(\"atitaarora/qdrant_doc_qna\", split=\"train\")\n",
    "\n",
    "EVAL_SIZE = 10\n",
    "\n",
    "#for Exp 1\n",
    "RETRIEVAL_SIZE_2 = 2\n",
    "\n",
    "#for Exp 2\n",
    "RETRIEVAL_SIZE_3 = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'question': ['What is vaccum optimizer ?',\n",
       "  'Tell me about ‘always_ram’ parameter?',\n",
       "  'What is difference between scalar and product quantization?',\n",
       "  'What is ‘best_score’ strategy?',\n",
       "  'How does oversampling helps?',\n",
       "  'What is the purpose of ‘CreatePayloadIndexAsync’?',\n",
       "  'What is the purpose of ef_construct in HNSW ?',\n",
       "  'How do you use ‘ordering’ parameter?',\n",
       "  'What is significance of ‘on_disk_payload’ setting?',\n",
       "  'What is the impact of ‘write_consistency_factor’ ?'],\n",
       " 'answer': [' The vacuum optimizer is a tool used to optimize the performance of a database by removing unused or redundant data. It works by identifying and removing empty or outdated data, freeing up space and improving the overall efficiency of the database. This can help improve the speed and reliability of queries and reduce the amount of storage needed for the database. You can learn more about the vacuum optimizer and how to use it in the documentation on optimization with web workers, specifically in the section on optimizing database performance (https://www.example.com/optimization/web-workers#database-performance).',\n",
       "  ' The `always_ram` parameter is used to determine whether quantized vectors should be kept cached in RAM or not. By default, quantized vectors are loaded in the same way as the original vectors. However, in some cases, it may be beneficial to keep quantized vectors in RAM to improve search speed. To enable this, you can set the `always_ram` parameter to `true`. This information can be found in the documentation under the `quantization_config` section, specifically under the `scalar` subsection.',\n",
       "  ' Scalar quantization is a method of converting floating-point numbers to integers, which can result in a loss of accuracy. On the other hand, product quantization is a method that divides the vector space into smaller subspaces and quantizes each subspace separately, which can also result in a loss of accuracy. However, product quantization is recommended for high-dimensional vectors as it tends to have a smaller error compared to scalar quantization. For more information on how to optimize quantization parameters, please refer to the \"Quantization Tips\" section in the documentation.',\n",
       "  \" The 'best_score' strategy is a method used to measure the performance of a candidate by comparing it to every example and selecting the best positive and best negative scores. The final score is then chosen using a step formula that takes into account both the best positive and best negative scores. This strategy is impacted by the amount of examples and its performance is linearly affected by it. For more information, please refer to the documentation on the 'best_score' strategy on page 3.\",\n",
       "  \" Oversampling helps by equalizing the representation of classes in the training dataset, thus enabling more fair and accurate modeling of real-world scenarios. This is explained in the documentation under the 'Context' section, which states that oversampling helps balance speed versus accuracy by producing more accurate results. You can find this information on page 1 of the documentation.\",\n",
       "  \" The 'CreatePayloadIndexAsync' function is used to create a new payload index in the specified collection. This allows for efficient storage and retrieval of data in the collection. It takes in parameters such as the collection name, group ID, and payload schema type to customize the index according to the user's needs. This information can be found in the documentation under the 'Payload Indexing' section, specifically on page 3 of the 'Qdrant Client' document.\",\n",
       "  ' The ef_construct parameter in HNSW is used to control the size of the dynamic list of nearest neighbors during the construction of the graph. This parameter is used to balance the search speed and accuracy. A higher value of ef_construct will result in a more accurate search but will also increase the search time. This parameter can be set using the HnswConfigDiff struct in the VectorParamsDiff struct. For more information, please refer to the documentation on HNSW configuration parameters on page 3.',\n",
       "  \" To use the 'ordering' parameter, you need to include it in the PUT request for the collection points. This parameter allows you to specify the order in which the points will be returned. You can set the value to 'strong' to return the points with the strongest vectors first, or 'weak' to return the points with the weakest vectors first. This information can be found in the 'source' field of the fetched context from the documentation, specifically in the PUT request example.\",\n",
       "  \" The 'on_disk_payload' setting is used to define where to store payload data. If set to 'true', the payload will be stored on disk only. This can be useful for limiting the RAM usage in case of large payload. This information can be found in the 'vectors' section of the configuration file, under the 'on_disk_payload' parameter. For more details, please refer to the 'quantization' section of the documentation, specifically the 'setting-up-quantization-in-qdrant' page.\",\n",
       "  \" The 'write_consistency_factor' determines the number of replicas that must acknowledge a write operation before responding to the client. This value can be increased to make write operations more tolerant to network partitions in the cluster, but it will also require a higher number of replicas to be active in order to perform write operations. This information can be found in the documentation under the 'write_consistency_factor' section on page 2.\"],\n",
       " 'contexts': [['optimizersConfig: new OptimizersConfigDiff { DefaultSegmentNumber = 2 }\\n\\n);\\n\\n```',\n",
       "   '#### Optimization with Web Workers'],\n",
       "  ['\"compression\": \"x32\",\\n\\n                    \"always_ram\": true\\n\\n                }\\n\\n            },\\n\\n            \"on_disk\": true\\n\\n        }\\n\\n    },\\n\\n    \"hnsw_config\": {\\n\\n        \"ef_construct\": 123\\n\\n    },\\n\\n    \"quantization_config\": {\\n\\n        \"scalar\": {\\n\\n            \"type\": \"int8\",\\n\\n            \"quantile\": 0.8,\\n\\n            \"always_ram\": false\\n\\n        }\\n\\n    }\\n\\n}\\n\\n```\\n\\n\\n\\n```bash\\n\\ncurl -X PATCH http://localhost:6333/collections/test_collection1 \\\\',\n",
       "   'It might be worth tuning this parameter if you experience a significant decrease in search quality.\\n\\n\\n\\n`always_ram` - whether to keep quantized vectors always cached in RAM or not. By default, quantized vectors are loaded in the same way as the original vectors.\\n\\nHowever, in some setups you might want to keep quantized vectors in RAM to speed up the search process.\\n\\n\\n\\nIn this case, you can set `always_ram` to `true` to store quantized vectors in RAM.\\n\\n\\n\\n### Setting up Binary Quantization'],\n",
       "  ['But there are some tradeoffs. Product quantization distance calculations are not SIMD-friendly, so it is slower than scalar quantization.\\n\\nAlso, product quantization has a loss of accuracy, so it is recommended to use it only for high-dimensional vectors.\\n\\n\\n\\nPlease refer to the [Quantization Tips](#quantization-tips) section for more information on how to optimize the quantization parameters for your use case.\\n\\n\\n\\n## How to choose the right quantization method',\n",
       "   'The main drawback of scalar quantization is the loss of accuracy. The `float32 -> uint8` conversion introduces an error that can lead to a slight decrease in search quality.\\n\\nHowever, this error is usually negligible, and tends to be less significant for high-dimensional vectors.\\n\\nIn our experiments, we found that the error introduced by scalar quantization is usually less than 1%. \\n\\n\\n\\nHowever, this value depends on the data and the quantization parameters.'],\n",
       "  ['#   [ScoredPoint(id=3, version=0, score=0.09, \\n\\n#                payload=None, vector=[0.0, 0.0, 0.0, 0.1]),\\n\\n#    ScoredPoint(id=1, version=0, score=0.10049876, \\n\\n#                payload=None, vector=[0.0, 0.1, 0.0, 0.0])]\\n\\n# ]\\n\\n\\n\\n```',\n",
       "   'The way it works is that each candidate is measured against every example, then we select the best positive and best negative scores. The final score is chosen with this step formula:\\n\\n\\n\\n```rust\\n\\nlet score = if best_positive_score > best_negative_score {\\n\\n    best_positive_score;\\n\\n} else {\\n\\n    -(best_negative_score * best_negative_score);\\n\\n};\\n\\n```\\n\\n\\n\\n<aside role=\"alert\">\\n\\nThe performance of <code>best_score</code> strategy will be linearly impacted by the amount of examples.\\n\\n</aside>'],\n",
       "  ['oversampling helps equalize the representation of classes in the training dataset, thus enabling more fair and accurate modeling of real-world scenarios.',\n",
       "   'These two parameters are how you are going to balance speed versus accuracy. The larger the size of your oversample, the more items you need to read from disk and the more elements you have to search with the relatively slower full vector index. On the other hand, doing this will produce more accurate results.'],\n",
       "  ['```java\\n\\nimport io.qdrant.client.QdrantClient;\\n\\nimport io.qdrant.client.QdrantGrpcClient;\\n\\nimport io.qdrant.client.grpc.Collections.PayloadSchemaType;\\n\\n\\n\\nQdrantClient client =\\n\\n    new QdrantClient(QdrantGrpcClient.newBuilder(\"localhost\", 6334, false).build());\\n\\n\\n\\nclient\\n\\n    .createPayloadIndexAsync(\\n\\n        \"{collection_name}\", \"group_id\", PayloadSchsemaType.Keyword, null, null, null, null)\\n\\n    .get();\\n\\n```\\n\\n\\n\\n```csharp\\n\\nusing Qdrant.Client;\\n\\n\\n\\nvar client = new QdrantClient(\"localhost\", 6334);',\n",
       "   'client.createPayloadIndex(\"{collection_name}\", {\\n\\n  field_name: \"name_of_the_field_to_index\",\\n\\n  field_schema: {\\n\\n    type: \"text\",\\n\\n    tokenizer: \"word\",\\n\\n    min_token_len: 2,\\n\\n    max_token_len: 15,\\n\\n    lowercase: true,\\n\\n  },\\n\\n});\\n\\n```\\n\\n\\n\\n```rust\\n\\nuse qdrant_client::{\\n\\n    client::QdrantClient,\\n\\n    qdrant::{\\n\\n        payload_index_params::IndexParams, FieldType, PayloadIndexParams, TextIndexParams,\\n\\n        TokenizerType,\\n\\n    },\\n\\n};'],\n",
       "  ['(\"my_vector\".into()),\\n\\n                        VectorParamsDiff {\\n\\n                            hnsw_config: Some(HnswConfigDiff {\\n\\n                                m: Some(32),\\n\\n                                ef_construct: Some(123),\\n\\n                                ..Default::default()\\n\\n                            }),\\n\\n                            ..Default::default()\\n\\n                        },\\n\\n                    )]),\\n\\n                },\\n\\n            )),\\n\\n        }),',\n",
       "   'hnsw_ef: Some(128),\\n\\n            exact: Some(false),\\n\\n            ..Default::default()\\n\\n        }),\\n\\n        limit: 3,\\n\\n        ..Default::default()\\n\\n    })\\n\\n    .await?;\\n\\n```\\n\\n\\n\\n```java\\n\\nimport java.util.List;\\n\\n\\n\\nimport io.qdrant.client.QdrantClient;\\n\\nimport io.qdrant.client.QdrantGrpcClient;\\n\\nimport io.qdrant.client.grpc.Points.SearchParams;\\n\\nimport io.qdrant.client.grpc.Points.SearchPoints;\\n\\n\\n\\nQdrantClient client ='],\n",
       "  ['```http\\n\\nPUT /collections/{collection_name}/points?ordering=strong\\n\\n{\\n\\n    \"batch\": {\\n\\n        \"ids\": [1, 2, 3],\\n\\n        \"payloads\": [\\n\\n            {\"color\": \"red\"},\\n\\n            {\"color\": \"green\"},\\n\\n            {\"color\": \"blue\"}\\n\\n        ],\\n\\n        \"vectors\": [\\n\\n            [0.9, 0.1, 0.1],\\n\\n            [0.1, 0.9, 0.1],\\n\\n            [0.1, 0.1, 0.9]\\n\\n        ]\\n\\n    }\\n\\n}\\n\\n```\\n\\n\\n\\n```python\\n\\nclient.upsert(\\n\\n    collection_name=\"{collection_name}\",\\n\\n    points=models.Batch(\\n\\n        ids=[1, 2, 3],',\n",
       "   'limit: 10,\\n\\n        ..Default::default()\\n\\n    })\\n\\n    .await?;\\n\\n```\\n\\n\\n\\n```java\\n\\nimport java.util.List;\\n\\n\\n\\nimport static io.qdrant.client.PointIdFactory.id;\\n\\n\\n\\nimport io.qdrant.client.grpc.Points.RecommendPoints;\\n\\n\\n\\nclient\\n\\n    .recommendAsync(\\n\\n        RecommendPoints.newBuilder()\\n\\n            .setCollectionName(\"{collection_name}\")\\n\\n            .addAllPositive(List.of(id(100), id(231)))\\n\\n            .addAllNegative(List.of(id(718)))\\n\\n            .setUsing(\"image\")\\n\\n            .setLimit(10)'],\n",
       "  ['* `on_disk_payload` - defines where to store payload data. If `true` - payload will be stored on disk only. Might be useful for limiting the RAM usage in case of large payload.\\n\\n* `quantization_config` - see [quantization](../../guides/quantization/#setting-up-quantization-in-qdrant) for details.\\n\\n\\n\\nDefault parameters for the optional collection parameters are defined in [configuration file](https://github.com/qdrant/qdrant/blob/master/config/config.yaml).',\n",
       "   '\"vectors\": {\\n\\n                    \"size\": 384,\\n\\n                    \"distance\": \"Cosine\"\\n\\n                },\\n\\n                \"shard_number\": 1,\\n\\n                \"replication_factor\": 1,\\n\\n                \"write_consistency_factor\": 1,\\n\\n                \"on_disk_payload\": false\\n\\n            },\\n\\n            \"hnsw_config\": {\\n\\n                \"m\": 16,\\n\\n                \"ef_construct\": 100,\\n\\n                \"full_scan_threshold\": 10000,\\n\\n                \"max_indexing_threads\": 0\\n\\n            },'],\n",
       "  ['### Write consistency factor\\n\\n\\n\\nThe `write_consistency_factor` represents the number of replicas that must acknowledge a write operation before responding to the client. It is set to one by default.\\n\\nIt can be configured at the collection\\'s creation time.\\n\\n\\n\\n```http\\n\\nPUT /collections/{collection_name}\\n\\n{\\n\\n    \"vectors\": {\\n\\n        \"size\": 300,\\n\\n        \"distance\": \"Cosine\"\\n\\n    },\\n\\n    \"shard_number\": 6,\\n\\n    \"replication_factor\": 2,\\n\\n    \"write_consistency_factor\": 2,\\n\\n}\\n\\n```\\n\\n\\n\\n```python',\n",
       "   '- `write_consistency_factor` - defines the number of replicas that must acknowledge a write operation before responding to the client. Increasing this value will make write operations tolerant to network partitions in the cluster, but will require a higher number of replicas to be active to perform write operations.']],\n",
       " 'ground_truth': ['The Vacuum Optimizer is a feature in the Qdrant database designed to manage and improve system performance by removing obsolete data. In Qdrant, when records are deleted, they are not immediately removed from the database. Instead, they are marked as deleted and subsequently ignored in future queries. This approach helps minimize disk access but results in accumulated deleted records that can occupy valuable memory and degrade performance over time.\\nThe Vacuum Optimizer addresses this issue by systematically clearing out these marked-as-deleted records once they reach a certain threshold. This threshold is configurable and includes conditions such as the minimum fraction of deleted vectors in a segment and the minimum number of vectors present in a segment. ',\n",
       "  \"In the context of vector quantization in Qdrant, the 'always_ram' parameter determines whether quantized vectors should be kept always cached in RAM or not. By default, quantized vectors are loaded in the same manner as the original vectors. Setting 'always_ram' to true ensures that the quantized vectors are consistently cached in RAM, providing faster access times.\",\n",
       "  'While both methods aim to reduce the memory footprint and potentially speed up operations, scalar quantization offers faster processing with SIMD-friendly operations and minimal accuracy loss, making it suitable for scenarios where speed is critical. Product quantization achieves higher compression rates and is better suited for handling very high-dimensional vectors where memory space is more of a concern than computation speed.',\n",
       "  'The `best_score` strategy is based on the idea of finding similar vectors by identifying those that are closer to a positive example while avoiding those that are closer to a negative one. This strategy measures each candidate against every example and selects the best positive and best negative scores. The final score is determined using a specific formula: if the best positive score is greater than the best negative score, the final score is set as the best positive score. Otherwise, the final score is calculated as the negative of the square of the best negative score. It is important to note that the performance of the `best_score` strategy is linearly impacted by the number of examples provided.',\n",
       "  'Defines how many extra vectors should be pre-selected using quantized index, and then re-scored using original vectors. For example, if oversampling is 2.4 and limit is 100, then 240 vectors will be pre-selected using quantized index, and then top-100 will be returned after re-scoring. Oversampling is useful if you want to tune the tradeoff between search speed and search quality in the query time.',\n",
       "  'CreatePayloadIndexAsync is a method in the Qdrant library that enables the creation of a payload index in Qdrant. A payload index is a data structure designed to store supplemental information about the data stored in Qdrant. This method requires a Qdrant instance and a payload index name as input parameters.',\n",
       "  'In HNSW algorithm the ef_construct parameter is the number of neighbours to consider during the index building. The larger the value, the higher the precision, but the longer the indexing time. The default values of this parameters 100',\n",
       "  'Write ordering can be specified for any write request to serialize it through a single “leader” node, which ensures that all write operations (issued with the same ordering) are performed and observed sequentially. It is of 3 types weak , medium and strong and is used in python with additional param ordering=models.WriteOrdering.STRONG to upsert request.',\n",
       "  \"The `on_disk_payload` setting in the storage configuration determines whether a point's payload will be stored in memory or read from disk every time it is requested. When set to `true`, the point's payload will not be stored in memory, saving RAM but slightly increasing the response time as the data needs to be retrieved from disk. It is important to note that payload values involved in filtering and indexed values will still remain in RAM for efficient access. This setting allows for a balance between RAM usage and response time in handling data storage and retrieval processes.\",\n",
       "  'The `write_consistency_factor` parameter in a distributed deployment using Qdrant defines the number of replicas that must acknowledge a write operation before responding to the client. By increasing this value, the write operations become more tolerant to network partitions within the cluster. However, this also means that a higher number of replicas need to be active in order to perform write operations successfully.']}"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Generating Eval dataset for the given RAG pipeline\n",
    "## Note this is the dataset needed for evaluation hence has to be recreated everytime changes to RAG config is made\n",
    "\n",
    "rag_eval_dataset_dspy_512_2 = create_eval_dataset_for_dspy(qdrant_qna_dataset,EVAL_SIZE,RETRIEVAL_SIZE_2)\n",
    "rag_eval_dataset_dspy_512_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ba379a0d712a47d896debe0a9526a61a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Creating CSV from Arrow format:   0%|          | 0/1 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "18501"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rag_response_dataset_dspy_512_2 = Dataset.from_dict(rag_eval_dataset_dspy_512_2)\n",
    "rag_response_dataset_dspy_512_2.to_csv('rag_response_with_dspy_512_2.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'question': ['What is vaccum optimizer ?',\n",
       "  'Tell me about ‘always_ram’ parameter?',\n",
       "  'What is difference between scalar and product quantization?',\n",
       "  'What is ‘best_score’ strategy?',\n",
       "  'How does oversampling helps?',\n",
       "  'What is the purpose of ‘CreatePayloadIndexAsync’?',\n",
       "  'What is the purpose of ef_construct in HNSW ?',\n",
       "  'How do you use ‘ordering’ parameter?',\n",
       "  'What is significance of ‘on_disk_payload’ setting?',\n",
       "  'What is the impact of ‘write_consistency_factor’ ?'],\n",
       " 'answer': [' The vacuum optimizer is a feature that helps to optimize the performance of the database by removing unused or deleted data from the database. This is done by setting a threshold for the amount of deleted data that can be present in the database before it is removed. This information can be found in the \"optimizer_config\" section of the documentation, specifically under the \"deleted_threshold\" key. For more information, please refer to page 5 of the \\'source\\' field of fetched context from the documentation.',\n",
       "  ' The `always_ram` parameter is used to determine whether quantized vectors should be kept cached in RAM or not. By default, quantized vectors are loaded in the same way as the original vectors. However, in some setups, it may be beneficial to keep quantized vectors in RAM to speed up the search process. To enable this, you can set `always_ram` to `true` in the `quantization_config` section of your collection configuration. This information can be found in the documentation under the `always_ram` field in the `quantization_config` section.',\n",
       "  \" Scalar quantization is a method of compressing data by converting floating-point numbers to integers, which can lead to a slight decrease in search quality due to the loss of accuracy. On the other hand, product quantization is a more efficient method of compression that can compress data up to 64 times the original size, but it also has a loss of accuracy. It is recommended to use product quantization for high-dimensional vectors, as the error introduced is usually negligible. For more information on how to optimize quantization parameters, please refer to the 'Quantization Tips' section of the documentation.\",\n",
       "  ' The ‘best_score’ strategy is a new strategy introduced in v1.6. It is based on the idea that the best way to find similar vectors is to find the ones that are closer to a positive example, while avoiding the ones that are closer to a negative one. This strategy is set implicitly, but can also be explicitly defined by setting \"strategy\": \"best_score\" in the recommendation request. It is linearly impacted by the amount of examples and is considered the default strategy.',\n",
       "  ' Oversampling is a technique used to balance the representation of classes in the training dataset, which leads to more fair and accurate modeling of real-world scenarios. This is achieved by retrieving more vectors than needed from quantized storage and assigning more precise scores during rescoring, resulting in the selection of only the most relevant vectors for the user. This information can be found in the documentation under the \"Oversampling\" section on page 5.',\n",
       "  \" The purpose of 'CreatePayloadIndexAsync' is to create a new payload index for a specified collection in the Qdrant database. This allows for faster and more efficient searching and retrieval of data within the collection. This information can be found in the documentation under the 'Creating a Payload Index' section on page 3.\",\n",
       "  ' The ef_construct parameter in HNSW is used to control the trade-off between the index build time and the index quality. It is used to set the size of the dynamic list of neighbors that is used during the construction of the HNSW index. This parameter can be set in the HnswConfigDiff object and is used to optimize the index for different datasets. For more information, please refer to the documentation on HNSW configuration parameters at https://qdrant.gitbook.io/qdrant/configuration/hnsw-configuration.',\n",
       "  ' To use the ‘ordering’ parameter, you need to include it in the URL of your request. For example, if you want to order your results by strength, you would include ‘ordering=strong’ in the URL. This will sort your results in descending order of strength. This information can be found in the documentation under the ‘PUT /collections/{collection_name}/points’ section.',\n",
       "  \" The 'on_disk_payload' setting is used to define where to store payload data. If set to 'true', the payload will be stored on disk only. This can be useful for limiting the RAM usage in case of large payload, as it will not be loaded into RAM at service startup. This type of storage may require more space, but it can improve performance when dealing with large payload values such as abstracts of text or images. For more information, please refer to the 'on_disk_payload' section in the [configuration file](https://github.com/qdrant/qdrant/blob/master/config/config.yaml).\",\n",
       "  \" The 'write_consistency_factor' determines the number of replicas that must acknowledge a write operation before responding to the client. This value can be increased to make write operations more tolerant to network partitions in the cluster, but it will also require a higher number of replicas to be active in order to perform write operations. This information can be found in the documentation under the 'write_consistency_factor' section of the 'collections' endpoint.\"],\n",
       " 'contexts': [['optimizersConfig: new OptimizersConfigDiff { DefaultSegmentNumber = 2 }\\n\\n);\\n\\n```',\n",
       "   '#### Optimization with Web Workers',\n",
       "   '},\\n\\n            \"optimizer_config\": {\\n\\n                \"deleted_threshold\": 0.2,\\n\\n                \"vacuum_min_vector_number\": 1000,\\n\\n                \"default_segment_number\": 0,\\n\\n                \"max_segment_size\": null,\\n\\n                \"memmap_threshold\": null,\\n\\n                \"indexing_threshold\": 20000,\\n\\n                \"flush_interval_sec\": 5,\\n\\n                \"max_optimization_threads\": 1\\n\\n            },\\n\\n            \"wal_config\": {\\n\\n                \"wal_capacity_mb\": 32,'],\n",
       "  ['\"compression\": \"x32\",\\n\\n                    \"always_ram\": true\\n\\n                }\\n\\n            },\\n\\n            \"on_disk\": true\\n\\n        }\\n\\n    },\\n\\n    \"hnsw_config\": {\\n\\n        \"ef_construct\": 123\\n\\n    },\\n\\n    \"quantization_config\": {\\n\\n        \"scalar\": {\\n\\n            \"type\": \"int8\",\\n\\n            \"quantile\": 0.8,\\n\\n            \"always_ram\": false\\n\\n        }\\n\\n    }\\n\\n}\\n\\n```\\n\\n\\n\\n```bash\\n\\ncurl -X PATCH http://localhost:6333/collections/test_collection1 \\\\',\n",
       "   'It might be worth tuning this parameter if you experience a significant decrease in search quality.\\n\\n\\n\\n`always_ram` - whether to keep quantized vectors always cached in RAM or not. By default, quantized vectors are loaded in the same way as the original vectors.\\n\\nHowever, in some setups you might want to keep quantized vectors in RAM to speed up the search process.\\n\\n\\n\\nIn this case, you can set `always_ram` to `true` to store quantized vectors in RAM.\\n\\n\\n\\n### Setting up Binary Quantization',\n",
       "   '\"vectors\": {\\n\\n                    \"size\": 384,\\n\\n                    \"distance\": \"Cosine\"\\n\\n                },\\n\\n                \"shard_number\": 1,\\n\\n                \"replication_factor\": 1,\\n\\n                \"write_consistency_factor\": 1,\\n\\n                \"on_disk_payload\": false\\n\\n            },\\n\\n            \"hnsw_config\": {\\n\\n                \"m\": 16,\\n\\n                \"ef_construct\": 100,\\n\\n                \"full_scan_threshold\": 10000,\\n\\n                \"max_indexing_threads\": 0\\n\\n            },'],\n",
       "  ['But there are some tradeoffs. Product quantization distance calculations are not SIMD-friendly, so it is slower than scalar quantization.\\n\\nAlso, product quantization has a loss of accuracy, so it is recommended to use it only for high-dimensional vectors.\\n\\n\\n\\nPlease refer to the [Quantization Tips](#quantization-tips) section for more information on how to optimize the quantization parameters for your use case.\\n\\n\\n\\n## How to choose the right quantization method',\n",
       "   'The main drawback of scalar quantization is the loss of accuracy. The `float32 -> uint8` conversion introduces an error that can lead to a slight decrease in search quality.\\n\\nHowever, this error is usually negligible, and tends to be less significant for high-dimensional vectors.\\n\\nIn our experiments, we found that the error introduced by scalar quantization is usually less than 1%. \\n\\n\\n\\nHowever, this value depends on the data and the quantization parameters.',\n",
       "   '. For example, on one of the previous releases we have scalar quantization and product quantization, which can compress up to 64 times the size of the vector. And we only keep in fast storage these compressed vectors.'],\n",
       "  ['#   [ScoredPoint(id=3, version=0, score=0.09, \\n\\n#                payload=None, vector=[0.0, 0.0, 0.0, 0.1]),\\n\\n#    ScoredPoint(id=1, version=0, score=0.10049876, \\n\\n#                payload=None, vector=[0.0, 0.1, 0.0, 0.0])]\\n\\n# ]\\n\\n\\n\\n```',\n",
       "   'The way it works is that each candidate is measured against every example, then we select the best positive and best negative scores. The final score is chosen with this step formula:\\n\\n\\n\\n```rust\\n\\nlet score = if best_positive_score > best_negative_score {\\n\\n    best_positive_score;\\n\\n} else {\\n\\n    -(best_negative_score * best_negative_score);\\n\\n};\\n\\n```\\n\\n\\n\\n<aside role=\"alert\">\\n\\nThe performance of <code>best_score</code> strategy will be linearly impacted by the amount of examples.\\n\\n</aside>',\n",
       "   'This is the default strategy that\\'s going to be set implicitly, but you can explicitly define it by setting `\"strategy\": \"average_vector\"` in the recommendation request.\\n\\n\\n\\n### Best score strategy\\n\\n\\n\\n*Available as of v1.6.0*\\n\\n\\n\\nA new strategy introduced in v1.6, is called `best_score`. It is based on the idea that the best way to find similar vectors is to find the ones that are closer to a positive example, while avoiding the ones that are closer to a negative one.'],\n",
       "  ['oversampling helps equalize the representation of classes in the training dataset, thus enabling more fair and accurate modeling of real-world scenarios.',\n",
       "   'These two parameters are how you are going to balance speed versus accuracy. The larger the size of your oversample, the more items you need to read from disk and the more elements you have to search with the relatively slower full vector index. On the other hand, doing this will produce more accurate results.',\n",
       "   'Yeah, so oversampling is a special technique we use to control precision of the search in real time, in query time. And the thing is, we can internally retrieve from quantized storage a bit more vectors than we actually need. And when we do rescoring with original vectors, we assign more precise score. And therefore from this overselection, we can pick only those vectors which are actually good for the user'],\n",
       "  ['```java\\n\\nimport io.qdrant.client.QdrantClient;\\n\\nimport io.qdrant.client.QdrantGrpcClient;\\n\\nimport io.qdrant.client.grpc.Collections.PayloadSchemaType;\\n\\n\\n\\nQdrantClient client =\\n\\n    new QdrantClient(QdrantGrpcClient.newBuilder(\"localhost\", 6334, false).build());\\n\\n\\n\\nclient\\n\\n    .createPayloadIndexAsync(\\n\\n        \"{collection_name}\", \"group_id\", PayloadSchsemaType.Keyword, null, null, null, null)\\n\\n    .get();\\n\\n```\\n\\n\\n\\n```csharp\\n\\nusing Qdrant.Client;\\n\\n\\n\\nvar client = new QdrantClient(\"localhost\", 6334);',\n",
       "   'client.createPayloadIndex(\"{collection_name}\", {\\n\\n  field_name: \"name_of_the_field_to_index\",\\n\\n  field_schema: {\\n\\n    type: \"text\",\\n\\n    tokenizer: \"word\",\\n\\n    min_token_len: 2,\\n\\n    max_token_len: 15,\\n\\n    lowercase: true,\\n\\n  },\\n\\n});\\n\\n```\\n\\n\\n\\n```rust\\n\\nuse qdrant_client::{\\n\\n    client::QdrantClient,\\n\\n    qdrant::{\\n\\n        payload_index_params::IndexParams, FieldType, PayloadIndexParams, TextIndexParams,\\n\\n        TokenizerType,\\n\\n    },\\n\\n};',\n",
       "   'await client.CreatePayloadIndexAsync(collectionName: \"{collection_name}\", fieldName: \"group_id\");\\n\\n```\\n\\n\\n\\n## Limitations\\n\\n\\n\\nOne downside to this approach is that global requests (without the `group_id` filter) will be slower since they will necessitate scanning all groups to identify the nearest neighbors.'],\n",
       "  ['(\"my_vector\".into()),\\n\\n                        VectorParamsDiff {\\n\\n                            hnsw_config: Some(HnswConfigDiff {\\n\\n                                m: Some(32),\\n\\n                                ef_construct: Some(123),\\n\\n                                ..Default::default()\\n\\n                            }),\\n\\n                            ..Default::default()\\n\\n                        },\\n\\n                    )]),\\n\\n                },\\n\\n            )),\\n\\n        }),',\n",
       "   'hnsw_ef: Some(128),\\n\\n            exact: Some(false),\\n\\n            ..Default::default()\\n\\n        }),\\n\\n        limit: 3,\\n\\n        ..Default::default()\\n\\n    })\\n\\n    .await?;\\n\\n```\\n\\n\\n\\n```java\\n\\nimport java.util.List;\\n\\n\\n\\nimport io.qdrant.client.QdrantClient;\\n\\nimport io.qdrant.client.QdrantGrpcClient;\\n\\nimport io.qdrant.client.grpc.Points.SearchParams;\\n\\nimport io.qdrant.client.grpc.Points.SearchPoints;\\n\\n\\n\\nQdrantClient client =',\n",
       "   '\"ef_construct\": 123\\n\\n    },\\n\\n    \"quantization_config\": {\\n\\n        \"scalar\": {\\n\\n            \"type\": \"int8\",\\n\\n            \"quantile\": 0.8,\\n\\n            \"always_ram\": false\\n\\n        }\\n\\n    }\\n\\n}\\'\\n\\n```\\n\\n\\n\\n```python\\n\\nclient.update_collection(\\n\\n    collection_name=\"{collection_name}\",\\n\\n    vectors_config={\\n\\n        \"my_vector\": models.VectorParamsDiff(\\n\\n            hnsw_config=models.HnswConfigDiff(\\n\\n                m=32,\\n\\n                ef_construct=123,\\n\\n            ),'],\n",
       "  ['```http\\n\\nPUT /collections/{collection_name}/points?ordering=strong\\n\\n{\\n\\n    \"batch\": {\\n\\n        \"ids\": [1, 2, 3],\\n\\n        \"payloads\": [\\n\\n            {\"color\": \"red\"},\\n\\n            {\"color\": \"green\"},\\n\\n            {\"color\": \"blue\"}\\n\\n        ],\\n\\n        \"vectors\": [\\n\\n            [0.9, 0.1, 0.1],\\n\\n            [0.1, 0.9, 0.1],\\n\\n            [0.1, 0.1, 0.9]\\n\\n        ]\\n\\n    }\\n\\n}\\n\\n```\\n\\n\\n\\n```python\\n\\nclient.upsert(\\n\\n    collection_name=\"{collection_name}\",\\n\\n    points=models.Batch(\\n\\n        ids=[1, 2, 3],',\n",
       "   'limit: 10,\\n\\n        ..Default::default()\\n\\n    })\\n\\n    .await?;\\n\\n```\\n\\n\\n\\n```java\\n\\nimport java.util.List;\\n\\n\\n\\nimport static io.qdrant.client.PointIdFactory.id;\\n\\n\\n\\nimport io.qdrant.client.grpc.Points.RecommendPoints;\\n\\n\\n\\nclient\\n\\n    .recommendAsync(\\n\\n        RecommendPoints.newBuilder()\\n\\n            .setCollectionName(\"{collection_name}\")\\n\\n            .addAllPositive(List.of(id(100), id(231)))\\n\\n            .addAllNegative(List.of(id(718)))\\n\\n            .setUsing(\"image\")\\n\\n            .setLimit(10)',\n",
       "   '\"must\": [\\n\\n            {\\n\\n                \"key\": \"diet[].food\",\\n\\n                  \"match\": {\\n\\n                    \"value\": \"meat\"\\n\\n                }\\n\\n            },\\n\\n            {\\n\\n                \"key\": \"diet[].likes\",\\n\\n                  \"match\": {\\n\\n                    \"value\": true\\n\\n                }\\n\\n            }\\n\\n        ]\\n\\n    }\\n\\n}\\n\\n```\\n\\n\\n\\n```python\\n\\nclient.scroll(\\n\\n    collection_name=\"{collection_name}\",\\n\\n    scroll_filter=models.Filter(\\n\\n        must=[\\n\\n            models.FieldCondition('],\n",
       "  ['* `on_disk_payload` - defines where to store payload data. If `true` - payload will be stored on disk only. Might be useful for limiting the RAM usage in case of large payload.\\n\\n* `quantization_config` - see [quantization](../../guides/quantization/#setting-up-quantization-in-qdrant) for details.\\n\\n\\n\\nDefault parameters for the optional collection parameters are defined in [configuration file](https://github.com/qdrant/qdrant/blob/master/config/config.yaml).',\n",
       "   '\"vectors\": {\\n\\n                    \"size\": 384,\\n\\n                    \"distance\": \"Cosine\"\\n\\n                },\\n\\n                \"shard_number\": 1,\\n\\n                \"replication_factor\": 1,\\n\\n                \"write_consistency_factor\": 1,\\n\\n                \"on_disk_payload\": false\\n\\n            },\\n\\n            \"hnsw_config\": {\\n\\n                \"m\": 16,\\n\\n                \"ef_construct\": 100,\\n\\n                \"full_scan_threshold\": 10000,\\n\\n                \"max_indexing_threads\": 0\\n\\n            },',\n",
       "   'The payload data is loaded into RAM at service startup while disk and [RocksDB](https://rocksdb.org/) are used for persistence only.\\n\\nThis type of storage works quite fast, but it may require a lot of space to keep all the data in RAM, especially if the payload has large values attached - abstracts of text or even images.\\n\\n\\n\\nIn the case of large payload values, it might be better to use OnDisk payload storage.'],\n",
       "  ['### Write consistency factor\\n\\n\\n\\nThe `write_consistency_factor` represents the number of replicas that must acknowledge a write operation before responding to the client. It is set to one by default.\\n\\nIt can be configured at the collection\\'s creation time.\\n\\n\\n\\n```http\\n\\nPUT /collections/{collection_name}\\n\\n{\\n\\n    \"vectors\": {\\n\\n        \"size\": 300,\\n\\n        \"distance\": \"Cosine\"\\n\\n    },\\n\\n    \"shard_number\": 6,\\n\\n    \"replication_factor\": 2,\\n\\n    \"write_consistency_factor\": 2,\\n\\n}\\n\\n```\\n\\n\\n\\n```python',\n",
       "   '- `write_consistency_factor` - defines the number of replicas that must acknowledge a write operation before responding to the client. Increasing this value will make write operations tolerant to network partitions in the cluster, but will require a higher number of replicas to be active to perform write operations.',\n",
       "   '\"vectors\": {\\n\\n                    \"size\": 384,\\n\\n                    \"distance\": \"Cosine\"\\n\\n                },\\n\\n                \"shard_number\": 1,\\n\\n                \"replication_factor\": 1,\\n\\n                \"write_consistency_factor\": 1,\\n\\n                \"on_disk_payload\": false\\n\\n            },\\n\\n            \"hnsw_config\": {\\n\\n                \"m\": 16,\\n\\n                \"ef_construct\": 100,\\n\\n                \"full_scan_threshold\": 10000,\\n\\n                \"max_indexing_threads\": 0\\n\\n            },']],\n",
       " 'ground_truth': ['The Vacuum Optimizer is a feature in the Qdrant database designed to manage and improve system performance by removing obsolete data. In Qdrant, when records are deleted, they are not immediately removed from the database. Instead, they are marked as deleted and subsequently ignored in future queries. This approach helps minimize disk access but results in accumulated deleted records that can occupy valuable memory and degrade performance over time.\\nThe Vacuum Optimizer addresses this issue by systematically clearing out these marked-as-deleted records once they reach a certain threshold. This threshold is configurable and includes conditions such as the minimum fraction of deleted vectors in a segment and the minimum number of vectors present in a segment. ',\n",
       "  \"In the context of vector quantization in Qdrant, the 'always_ram' parameter determines whether quantized vectors should be kept always cached in RAM or not. By default, quantized vectors are loaded in the same manner as the original vectors. Setting 'always_ram' to true ensures that the quantized vectors are consistently cached in RAM, providing faster access times.\",\n",
       "  'While both methods aim to reduce the memory footprint and potentially speed up operations, scalar quantization offers faster processing with SIMD-friendly operations and minimal accuracy loss, making it suitable for scenarios where speed is critical. Product quantization achieves higher compression rates and is better suited for handling very high-dimensional vectors where memory space is more of a concern than computation speed.',\n",
       "  'The `best_score` strategy is based on the idea of finding similar vectors by identifying those that are closer to a positive example while avoiding those that are closer to a negative one. This strategy measures each candidate against every example and selects the best positive and best negative scores. The final score is determined using a specific formula: if the best positive score is greater than the best negative score, the final score is set as the best positive score. Otherwise, the final score is calculated as the negative of the square of the best negative score. It is important to note that the performance of the `best_score` strategy is linearly impacted by the number of examples provided.',\n",
       "  'Defines how many extra vectors should be pre-selected using quantized index, and then re-scored using original vectors. For example, if oversampling is 2.4 and limit is 100, then 240 vectors will be pre-selected using quantized index, and then top-100 will be returned after re-scoring. Oversampling is useful if you want to tune the tradeoff between search speed and search quality in the query time.',\n",
       "  'CreatePayloadIndexAsync is a method in the Qdrant library that enables the creation of a payload index in Qdrant. A payload index is a data structure designed to store supplemental information about the data stored in Qdrant. This method requires a Qdrant instance and a payload index name as input parameters.',\n",
       "  'In HNSW algorithm the ef_construct parameter is the number of neighbours to consider during the index building. The larger the value, the higher the precision, but the longer the indexing time. The default values of this parameters 100',\n",
       "  'Write ordering can be specified for any write request to serialize it through a single “leader” node, which ensures that all write operations (issued with the same ordering) are performed and observed sequentially. It is of 3 types weak , medium and strong and is used in python with additional param ordering=models.WriteOrdering.STRONG to upsert request.',\n",
       "  \"The `on_disk_payload` setting in the storage configuration determines whether a point's payload will be stored in memory or read from disk every time it is requested. When set to `true`, the point's payload will not be stored in memory, saving RAM but slightly increasing the response time as the data needs to be retrieved from disk. It is important to note that payload values involved in filtering and indexed values will still remain in RAM for efficient access. This setting allows for a balance between RAM usage and response time in handling data storage and retrieval processes.\",\n",
       "  'The `write_consistency_factor` parameter in a distributed deployment using Qdrant defines the number of replicas that must acknowledge a write operation before responding to the client. By increasing this value, the write operations become more tolerant to network partitions within the cluster. However, this also means that a higher number of replicas need to be active in order to perform write operations successfully.']}"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Generating Eval dataset for the given RAG pipeline\n",
    "## Note this is the dataset needed for evaluation hence has to be recreated everytime changes to RAG config is made\n",
    "\n",
    "rag_eval_dataset_dspy_512_3 = create_eval_dataset_for_dspy(qdrant_qna_dataset,EVAL_SIZE,RETRIEVAL_SIZE_3)\n",
    "rag_eval_dataset_dspy_512_3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fe2100be4d614995ba2c702edca3122e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Creating CSV from Arrow format:   0%|          | 0/1 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "23037"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rag_response_dataset_dspy_512_3 = Dataset.from_dict(rag_eval_dataset_dspy_512_3)\n",
    "rag_response_dataset_dspy_512_3.to_csv('rag_response_with_dspy_512_3.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "te1iegPpW__p"
   },
   "source": [
    "### **15. Lets evaluate , how good is our RAG !!**\n",
    "\n",
    "![image.png](data:image/png;base64,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)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "id": "qPzkNNgKat3R"
   },
   "outputs": [],
   "source": [
    "from ragas.metrics import (\n",
    "    faithfulness,\n",
    "    answer_relevancy,\n",
    "    context_recall,\n",
    "    context_precision,\n",
    "    context_relevancy,\n",
    "    context_entity_recall,\n",
    "    answer_similarity,\n",
    "    answer_correctness\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Method to encapsulate ragas evaluate method with all the 8 metrics\n",
    "def evaluate_with_ragas(rag_response_dataset_df):\n",
    "    result = evaluate(\n",
    "        rag_response_dataset_df,\n",
    "        metrics=[\n",
    "                faithfulness,\n",
    "                answer_relevancy,\n",
    "                context_recall,\n",
    "                context_precision,\n",
    "                context_relevancy,\n",
    "                context_entity_recall,\n",
    "                answer_similarity,\n",
    "                answer_correctness\n",
    "        ],\n",
    "       raise_exceptions=False \n",
    "    )\n",
    "    return result"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **16. Evaluation of RAG with Retrieval Window = 2**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question</th>\n",
       "      <th>answer</th>\n",
       "      <th>contexts</th>\n",
       "      <th>ground_truth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>What is vaccum optimizer ?</td>\n",
       "      <td>The vacuum optimizer is a tool used to optimi...</td>\n",
       "      <td>['optimizersConfig: new OptimizersConfigDiff {...</td>\n",
       "      <td>The Vacuum Optimizer is a feature in the Qdran...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Tell me about ‘always_ram’ parameter?</td>\n",
       "      <td>The `always_ram` parameter is used to determi...</td>\n",
       "      <td>['\"compression\": \"x32\",\\n\\n                   ...</td>\n",
       "      <td>In the context of vector quantization in Qdran...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>What is difference between scalar and product ...</td>\n",
       "      <td>Scalar quantization is a method of converting...</td>\n",
       "      <td>['But there are some tradeoffs. Product quanti...</td>\n",
       "      <td>While both methods aim to reduce the memory fo...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>What is ‘best_score’ strategy?</td>\n",
       "      <td>The 'best_score' strategy is a method used to...</td>\n",
       "      <td>['#   [ScoredPoint(id=3, version=0, score=0.09...</td>\n",
       "      <td>The `best_score` strategy is based on the idea...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>How does oversampling helps?</td>\n",
       "      <td>Oversampling helps by equalizing the represen...</td>\n",
       "      <td>['oversampling helps equalize the representati...</td>\n",
       "      <td>Defines how many extra vectors should be pre-s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>What is the purpose of ‘CreatePayloadIndexAsync’?</td>\n",
       "      <td>The 'CreatePayloadIndexAsync' function is use...</td>\n",
       "      <td>['```java\\n\\nimport io.qdrant.client.QdrantCli...</td>\n",
       "      <td>CreatePayloadIndexAsync is a method in the Qdr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>What is the purpose of ef_construct in HNSW ?</td>\n",
       "      <td>The ef_construct parameter in HNSW is used to...</td>\n",
       "      <td>['(\"my_vector\".into()),\\n\\n                   ...</td>\n",
       "      <td>In HNSW algorithm the ef_construct parameter i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>How do you use ‘ordering’ parameter?</td>\n",
       "      <td>To use the 'ordering' parameter, you need to ...</td>\n",
       "      <td>['```http\\n\\nPUT /collections/{collection_name...</td>\n",
       "      <td>Write ordering can be specified for any write ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>What is significance of ‘on_disk_payload’ sett...</td>\n",
       "      <td>The 'on_disk_payload' setting is used to defi...</td>\n",
       "      <td>['* `on_disk_payload` - defines where to store...</td>\n",
       "      <td>The `on_disk_payload` setting in the storage c...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>What is the impact of ‘write_consistency_facto...</td>\n",
       "      <td>The 'write_consistency_factor' determines the...</td>\n",
       "      <td>['### Write consistency factor\\n\\n\\n\\nThe `wri...</td>\n",
       "      <td>The `write_consistency_factor` parameter in a ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            question  \\\n",
       "0                         What is vaccum optimizer ?   \n",
       "1              Tell me about ‘always_ram’ parameter?   \n",
       "2  What is difference between scalar and product ...   \n",
       "3                     What is ‘best_score’ strategy?   \n",
       "4                       How does oversampling helps?   \n",
       "5  What is the purpose of ‘CreatePayloadIndexAsync’?   \n",
       "6      What is the purpose of ef_construct in HNSW ?   \n",
       "7               How do you use ‘ordering’ parameter?   \n",
       "8  What is significance of ‘on_disk_payload’ sett...   \n",
       "9  What is the impact of ‘write_consistency_facto...   \n",
       "\n",
       "                                              answer  \\\n",
       "0   The vacuum optimizer is a tool used to optimi...   \n",
       "1   The `always_ram` parameter is used to determi...   \n",
       "2   Scalar quantization is a method of converting...   \n",
       "3   The 'best_score' strategy is a method used to...   \n",
       "4   Oversampling helps by equalizing the represen...   \n",
       "5   The 'CreatePayloadIndexAsync' function is use...   \n",
       "6   The ef_construct parameter in HNSW is used to...   \n",
       "7   To use the 'ordering' parameter, you need to ...   \n",
       "8   The 'on_disk_payload' setting is used to defi...   \n",
       "9   The 'write_consistency_factor' determines the...   \n",
       "\n",
       "                                            contexts  \\\n",
       "0  ['optimizersConfig: new OptimizersConfigDiff {...   \n",
       "1  ['\"compression\": \"x32\",\\n\\n                   ...   \n",
       "2  ['But there are some tradeoffs. Product quanti...   \n",
       "3  ['#   [ScoredPoint(id=3, version=0, score=0.09...   \n",
       "4  ['oversampling helps equalize the representati...   \n",
       "5  ['```java\\n\\nimport io.qdrant.client.QdrantCli...   \n",
       "6  ['(\"my_vector\".into()),\\n\\n                   ...   \n",
       "7  ['```http\\n\\nPUT /collections/{collection_name...   \n",
       "8  ['* `on_disk_payload` - defines where to store...   \n",
       "9  ['### Write consistency factor\\n\\n\\n\\nThe `wri...   \n",
       "\n",
       "                                        ground_truth  \n",
       "0  The Vacuum Optimizer is a feature in the Qdran...  \n",
       "1  In the context of vector quantization in Qdran...  \n",
       "2  While both methods aim to reduce the memory fo...  \n",
       "3  The `best_score` strategy is based on the idea...  \n",
       "4  Defines how many extra vectors should be pre-s...  \n",
       "5  CreatePayloadIndexAsync is a method in the Qdr...  \n",
       "6  In HNSW algorithm the ef_construct parameter i...  \n",
       "7  Write ordering can be specified for any write ...  \n",
       "8  The `on_disk_payload` setting in the storage c...  \n",
       "9  The `write_consistency_factor` parameter in a ...  "
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Evaluating Experiment 1 - With Retrieval Window = 2\n",
    "\n",
    "from ragas import evaluate\n",
    "eval_dataset_with_dspy_512_2 = pd.read_csv('rag_response_with_dspy_512_2.csv')\n",
    "eval_dataset_with_dspy_512_2.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "23a021100f0b4f8abbad47f7a99818fe",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating:   0%|          | 0/80 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result_512_with_dspy_2 = evaluate(\n",
    "        rag_response_dataset_dspy_512_2,\n",
    "        metrics=[\n",
    "                faithfulness,\n",
    "                answer_relevancy,\n",
    "                context_recall,\n",
    "                context_precision,\n",
    "                context_relevancy,\n",
    "                context_entity_recall,\n",
    "                answer_similarity,\n",
    "                answer_correctness\n",
    "        ],\n",
    "       raise_exceptions=False \n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "ZOmHi8jkyqTE",
    "outputId": "aa34fa70-ad96-4232-e83c-027fe43f6e81",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/atitaarora/qdrant/workspace/qdrant-rag-eval/workshop-rag-eval-qdrant-ragas/ragas/lib/python3.10/site-packages/datasets/table.py:1395: FutureWarning: promote has been superseded by promote_options='default'.\n",
      "  block_group = [InMemoryTable(cls._concat_blocks(list(block_group), axis=axis))]\n",
      "/Users/atitaarora/qdrant/workspace/qdrant-rag-eval/workshop-rag-eval-qdrant-ragas/ragas/lib/python3.10/site-packages/datasets/table.py:1421: FutureWarning: promote has been superseded by promote_options='default'.\n",
      "  table = cls._concat_blocks(blocks, axis=0)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question</th>\n",
       "      <th>answer</th>\n",
       "      <th>contexts</th>\n",
       "      <th>ground_truth</th>\n",
       "      <th>faithfulness</th>\n",
       "      <th>answer_relevancy</th>\n",
       "      <th>context_recall</th>\n",
       "      <th>context_precision</th>\n",
       "      <th>context_relevancy</th>\n",
       "      <th>context_entity_recall</th>\n",
       "      <th>answer_similarity</th>\n",
       "      <th>answer_correctness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>What is vaccum optimizer ?</td>\n",
       "      <td>The vacuum optimizer is a tool used to optimi...</td>\n",
       "      <td>[optimizersConfig: new OptimizersConfigDiff { ...</td>\n",
       "      <td>The Vacuum Optimizer is a feature in the Qdran...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.957428</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.898055</td>\n",
       "      <td>0.360877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Tell me about ‘always_ram’ parameter?</td>\n",
       "      <td>The `always_ram` parameter is used to determi...</td>\n",
       "      <td>[\"compression\": \"x32\",\\n\\n                    ...</td>\n",
       "      <td>In the context of vector quantization in Qdran...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.914733</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.148148</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.935302</td>\n",
       "      <td>0.733826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>What is difference between scalar and product ...</td>\n",
       "      <td>Scalar quantization is a method of converting...</td>\n",
       "      <td>[But there are some tradeoffs. Product quantiz...</td>\n",
       "      <td>While both methods aim to reduce the memory fo...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.920023</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.230769</td>\n",
       "      <td>0.921613</td>\n",
       "      <td>0.680403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>What is ‘best_score’ strategy?</td>\n",
       "      <td>The 'best_score' strategy is a method used to...</td>\n",
       "      <td>[#   [ScoredPoint(id=3, version=0, score=0.09,...</td>\n",
       "      <td>The `best_score` strategy is based on the idea...</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.977437</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.951858</td>\n",
       "      <td>0.647055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>How does oversampling helps?</td>\n",
       "      <td>Oversampling helps by equalizing the represen...</td>\n",
       "      <td>[oversampling helps equalize the representatio...</td>\n",
       "      <td>Defines how many extra vectors should be pre-s...</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.983625</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.807052</td>\n",
       "      <td>0.535096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>What is the purpose of ‘CreatePayloadIndexAsync’?</td>\n",
       "      <td>The 'CreatePayloadIndexAsync' function is use...</td>\n",
       "      <td>[```java\\n\\nimport io.qdrant.client.QdrantClie...</td>\n",
       "      <td>CreatePayloadIndexAsync is a method in the Qdr...</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.893574</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.027778</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.904508</td>\n",
       "      <td>0.559438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>What is the purpose of ef_construct in HNSW ?</td>\n",
       "      <td>The ef_construct parameter in HNSW is used to...</td>\n",
       "      <td>[(\"my_vector\".into()),\\n\\n                    ...</td>\n",
       "      <td>In HNSW algorithm the ef_construct parameter i...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.955200</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.027778</td>\n",
       "      <td>0.125000</td>\n",
       "      <td>0.906901</td>\n",
       "      <td>0.772177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>How do you use ‘ordering’ parameter?</td>\n",
       "      <td>To use the 'ordering' parameter, you need to ...</td>\n",
       "      <td>[```http\\n\\nPUT /collections/{collection_name}...</td>\n",
       "      <td>Write ordering can be specified for any write ...</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.901608</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.802468</td>\n",
       "      <td>0.800620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>What is significance of ‘on_disk_payload’ sett...</td>\n",
       "      <td>The 'on_disk_payload' setting is used to defi...</td>\n",
       "      <td>[* `on_disk_payload` - defines where to store ...</td>\n",
       "      <td>The `on_disk_payload` setting in the storage c...</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.959801</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.150000</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.906335</td>\n",
       "      <td>0.726584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>What is the impact of ‘write_consistency_facto...</td>\n",
       "      <td>The 'write_consistency_factor' determines the...</td>\n",
       "      <td>[### Write consistency factor\\n\\n\\n\\nThe `writ...</td>\n",
       "      <td>The `write_consistency_factor` parameter in a ...</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.943122</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.157895</td>\n",
       "      <td>0.357143</td>\n",
       "      <td>0.933992</td>\n",
       "      <td>0.608448</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            question  \\\n",
       "0                         What is vaccum optimizer ?   \n",
       "1              Tell me about ‘always_ram’ parameter?   \n",
       "2  What is difference between scalar and product ...   \n",
       "3                     What is ‘best_score’ strategy?   \n",
       "4                       How does oversampling helps?   \n",
       "5  What is the purpose of ‘CreatePayloadIndexAsync’?   \n",
       "6      What is the purpose of ef_construct in HNSW ?   \n",
       "7               How do you use ‘ordering’ parameter?   \n",
       "8  What is significance of ‘on_disk_payload’ sett...   \n",
       "9  What is the impact of ‘write_consistency_facto...   \n",
       "\n",
       "                                              answer  \\\n",
       "0   The vacuum optimizer is a tool used to optimi...   \n",
       "1   The `always_ram` parameter is used to determi...   \n",
       "2   Scalar quantization is a method of converting...   \n",
       "3   The 'best_score' strategy is a method used to...   \n",
       "4   Oversampling helps by equalizing the represen...   \n",
       "5   The 'CreatePayloadIndexAsync' function is use...   \n",
       "6   The ef_construct parameter in HNSW is used to...   \n",
       "7   To use the 'ordering' parameter, you need to ...   \n",
       "8   The 'on_disk_payload' setting is used to defi...   \n",
       "9   The 'write_consistency_factor' determines the...   \n",
       "\n",
       "                                            contexts  \\\n",
       "0  [optimizersConfig: new OptimizersConfigDiff { ...   \n",
       "1  [\"compression\": \"x32\",\\n\\n                    ...   \n",
       "2  [But there are some tradeoffs. Product quantiz...   \n",
       "3  [#   [ScoredPoint(id=3, version=0, score=0.09,...   \n",
       "4  [oversampling helps equalize the representatio...   \n",
       "5  [```java\\n\\nimport io.qdrant.client.QdrantClie...   \n",
       "6  [(\"my_vector\".into()),\\n\\n                    ...   \n",
       "7  [```http\\n\\nPUT /collections/{collection_name}...   \n",
       "8  [* `on_disk_payload` - defines where to store ...   \n",
       "9  [### Write consistency factor\\n\\n\\n\\nThe `writ...   \n",
       "\n",
       "                                        ground_truth  faithfulness  \\\n",
       "0  The Vacuum Optimizer is a feature in the Qdran...      1.000000   \n",
       "1  In the context of vector quantization in Qdran...      1.000000   \n",
       "2  While both methods aim to reduce the memory fo...      1.000000   \n",
       "3  The `best_score` strategy is based on the idea...      0.750000   \n",
       "4  Defines how many extra vectors should be pre-s...      0.600000   \n",
       "5  CreatePayloadIndexAsync is a method in the Qdr...      0.666667   \n",
       "6  In HNSW algorithm the ef_construct parameter i...      1.000000   \n",
       "7  Write ordering can be specified for any write ...      0.666667   \n",
       "8  The `on_disk_payload` setting in the storage c...      0.600000   \n",
       "9  The `write_consistency_factor` parameter in a ...      0.750000   \n",
       "\n",
       "   answer_relevancy  context_recall  context_precision  context_relevancy  \\\n",
       "0          0.957428        0.000000                1.0           0.250000   \n",
       "1          0.914733        0.666667                1.0           0.148148   \n",
       "2          0.920023        1.000000                1.0           0.200000   \n",
       "3          0.977437        0.000000                1.0           0.166667   \n",
       "4          0.983625        0.500000                0.5           0.250000   \n",
       "5          0.893574        1.000000                1.0           0.027778   \n",
       "6          0.955200        0.333333                1.0           0.027778   \n",
       "7          0.901608        1.000000                1.0           0.020000   \n",
       "8          0.959801        1.000000                1.0           0.150000   \n",
       "9          0.943122        1.000000                1.0           0.157895   \n",
       "\n",
       "   context_entity_recall  answer_similarity  answer_correctness  \n",
       "0               0.000000           0.898055            0.360877  \n",
       "1               0.142857           0.935302            0.733826  \n",
       "2               0.230769           0.921613            0.680403  \n",
       "3               0.000000           0.951858            0.647055  \n",
       "4               0.142857           0.807052            0.535096  \n",
       "5               0.000000           0.904508            0.559438  \n",
       "6               0.125000           0.906901            0.772177  \n",
       "7               0.000000           0.802468            0.800620  \n",
       "8               0.083333           0.906335            0.726584  \n",
       "9               0.357143           0.933992            0.608448  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluation_result_df_512_with_dspy_2 = result_512_with_dspy_2.to_pandas()\n",
    "evaluation_result_df_512_with_dspy_2.iloc[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 177,
     "referenced_widgets": [
      "ed4d43fadb1f48319ea790b56cb4d343",
      "3b9465a8398c45f498a414ac46c80355",
      "b457ff179f38448fa55a384e8c395de1",
      "8ec2eae08ad643c3b03609c66554ed4c",
      "a2278e779f0f4c0783d0c1602d3dbb4f",
      "64511930cae14bfeb384d0fbbcdc7548",
      "cb8287f82d0043c49ba3da16b4f387c5",
      "0f0ecde0157841e0a2bd661f59fea733",
      "ad70be7e9f16404985c2f13f60b7447f",
      "16b940f370074141b23b3f8b1f38dbe2",
      "e22891a28b8747ef9c860ed54c3a73b8",
      "fbf413ef0b274ccea2130db147f06e25",
      "25db4e94fc61492a9752600158fae7b4",
      "30fcf2693b154033881fee05b0d17db6",
      "a3353c3b1828438591573c03a106d466",
      "3124ef41463846a999c1fa01f1530b1a",
      "bb7a42f693b0461994ccb49c9720f8dd",
      "1db283a5206741baa2fbd547e759b000",
      "326aae4816574318b01b999e18e10f37",
      "e94a8581d14842bda30438961ad294e1",
      "a20883beb87c499db4700d935d56957b",
      "77aab4a259754f709ddbd2ceadacce4d",
      "a634801d0aad4e21b2bfa8f92624c087",
      "26a051f4728a4c8c88afc75097bd648a",
      "a7c3dd5bae484d959e7955746e121800",
      "e747142671b64db4a4d5f6f307ebb801",
      "acc1bb1a2bfb4a6eb953f1b7535ac8bf",
      "0faab5cb87f84d6fa29016e1775a376d",
      "01c3ae35884e45b18d062cb49ff6ff00",
      "6365cb72b0e740aba2bb1b933aedc844",
      "7c48f92a57064802b2273072d5261605",
      "156da40cbd834a518cddb08cbef3198d",
      "2805c5fbcfab4313b321f7ab125d9f2b",
      "550aeb314ffb42ee98b2ca2fd44c7bf0",
      "6679acc33bd2411e9dc261f1ef2f03e1",
      "da47e1bf65a345b7b7a6d2bcdccf4dc2",
      "1c37f78ea68840aaa9c1fe77541ae81e",
      "9d3466be165d4323984c45fda1003ba2",
      "cb75a19ae09d496c857a982e9654a0cb",
      "daf4e8c673564fed9e03f7fddaa81ad3",
      "81730b126e424a7c97ab157af8ffcd3a",
      "8119b99d4ac247bb8a759ad1a8a0e66c",
      "174745d013354898837589a21fe5fb8e",
      "f226de79c38b4ed4bae675ec213f18be",
      "345208d4d61d4e378db846912706f1a1",
      "fe30399d1aeb44d3a6d69091b606b80f",
      "f6fd6fcdc4f349fba497027d7fcf8a9f",
      "b13dc56442bc437f97bd9f1165337b38",
      "4c0d4444761346ee8cccb42ea17e6ed6",
      "fe594f22be574d3b8c2ca187ac9c7bfb",
      "1eec371016434a6b8c4ac8b83bd77b4f",
      "805f4b56fe77487581c9c605a4411b4c",
      "00963b69a2524a8a91f42e46a6ec8d0b",
      "3ef373eb9dae44f89a794c4cd8584ad4",
      "78577744100e484fb8ee6c02473ed0b2"
     ]
    },
    "id": "iPi4vr3ad5m7",
    "outputId": "8f24b120-2dbe-4045-edb1-4d146a259994"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ed4d43fadb1f48319ea790b56cb4d343",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading readme:   0%|          | 0.00/43.0 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fbf413ef0b274ccea2130db147f06e25",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data:   0%|          | 0.00/125k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a634801d0aad4e21b2bfa8f92624c087",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Generating train split: 0 examples [00:00, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "550aeb314ffb42ee98b2ca2fd44c7bf0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Filter:   0%|          | 0/81 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "345208d4d61d4e378db846912706f1a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Map:   0%|          | 0/21 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# loading the eval dataset from HF\n",
    "qdrant_qa = load_dataset(\"atitaarora/qdrant_docs_qna_ragas\")\n",
    "\n",
    "#Converting contexts string to list of contexts for running further evals\n",
    "qdrant_qa_eval = load_dataset(\"atitaarora/qdrant_docs_qna_ragas\", split=\"train\")\n",
    "column_to_filter = \"contexts\"\n",
    "# Filter rows with null values in the specified column\n",
    "filtered_dataset = qdrant_qa_eval.filter(lambda example, column_name=column_to_filter: example[column_name] is not None)\n",
    "#len(filtered_dataset)\n",
    "\n",
    "filtered_dataset = filtered_dataset.map(lambda x: {\"contexts\":[x[\"contexts\"]]})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 49,
     "referenced_widgets": [
      "fdd8cf7194de46d2b2c7440077e673f8",
      "3c0415f3a6d944ebbddaa4fc04a22f22",
      "1faed09d17f74721a5462f42ad61caf1",
      "edc22a2adeb34550b24791cb89762383",
      "85a871757b9b45eda68a11812e39e534",
      "bd64efaf240e42eb9b84dec9987a3c4f",
      "fa1cfd25cadd4a42ab990a336eea095c",
      "f93019d15bf64bac93bad2c3c3c53714",
      "f642894d3c574bdf8e19fbe42cf027d5",
      "7fbbcb02feef491fb2f61b252c296758",
      "4f8b9fb466694e1aacf8cf8da4c56618"
     ]
    },
    "id": "Ei23twt-eK_K",
    "outputId": "0bbe41fa-02e1-403d-d45a-a76de1fb9daf"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fdd8cf7194de46d2b2c7440077e673f8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating:   0%|          | 0/105 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from ragas import evaluate\n",
    "\n",
    "result_baseline = evaluate(\n",
    "    filtered_dataset,\n",
    "    metrics=[\n",
    "        answer_correctness,\n",
    "        answer_relevancy,\n",
    "        faithfulness,\n",
    "        context_recall,\n",
    "        context_precision\n",
    "    ],\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "SyK7fT4mePtx",
    "outputId": "52f2c433-8294-48f1-ffef-83a3ab27942b"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.google.colaboratory.intrinsic+json": {
       "summary": "{\n  \"name\": \"evaluation_result_df\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"question\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"What is the purpose of the locking feature in Qdrant administration tools?\",\n          \"How is metric learning utilized in addressing the challenge of detecting anomalies in coffee beans?\",\n          \"How does Qdrant optimize memory and search speed for sparse vectors?\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ground_truth\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"The locking feature in Qdrant administration tools serves the purpose of restricting the possible operations on a Qdrant process. It allows users to control the behavior of a Qdrant instance at runtime without manually changing its configuration. When a lock is applied, specific operations such as creating new collections or adding new data to the existing storage can be disabled. It is important to note that the locking configuration is not persistent, so it needs to be reapplied after a restart. Additionally, the locking feature applies to a single node only, requiring the lock to be set on all desired nodes in a distributed deployment setup. The locking API in Qdrant enables users to manage and control the access and operations on the Qdrant instance effectively.\",\n          \"Metric learning is utilized in addressing the challenge of detecting anomalies in coffee beans by encoding images in an n-dimensional vector space and using learned similarities to label images during the inference process. This approach involves representing the images in a vector space where similar images are closer together in the space. By doing so, the model can effectively classify and detect anomalies in coffee beans based on the similarities learned during training. The KNN (K-Nearest Neighbors) classification method is commonly used in this approach as it simplifies the process of determining the class label of an image based on the labels of its nearest neighbors in the vector space. This method allows for the adaptation to new types of defects and changing shooting conditions, making the model more robust and accurate in detecting anomalies in coffee beans.\",\n          \"Qdrant optimizes memory and search speed for sparse vectors by utilizing an inverted index structure to store vectors for each non-zero dimension. This approach allows Qdrant to efficiently represent sparse vectors, which are characterized by a high proportion of zeroes. By only storing information about non-zero dimensions, Qdrant reduces the memory footprint required to store sparse vectors and also speeds up search operations by focusing only on relevant dimensions during indexing and querying processes. This optimization ensures that Qdrant can handle sparse vectors effectively while maintaining efficient memory usage and search performance.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"answer\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"The locking feature in Qdrant administration tools serves the purpose of restricting the possible operations on a Qdrant process. It allows users to control the behavior of a Qdrant instance at runtime without manually changing its configuration. When a lock is applied, specific operations such as creating new collections or adding new data to the existing storage can be disabled. It is important to note that the locking configuration is not persistent, so it needs to be reapplied after a restart. Additionally, the locking feature applies to a single node only, requiring the lock to be set on all desired nodes in a distributed deployment setup. The locking API in Qdrant enables users to manage and control the access and operations on the Qdrant instance effectively.\",\n          \"Metric learning is utilized in addressing the challenge of detecting anomalies in coffee beans by encoding images in an n-dimensional vector space and using learned similarities to label images during the inference process. This approach involves representing the images in a vector space where similar images are closer together in the space. By doing so, the model can effectively classify and detect anomalies in coffee beans based on the similarities learned during training. The KNN (K-Nearest Neighbors) classification method is commonly used in this approach as it simplifies the process of determining the class label of an image based on the labels of its nearest neighbors in the vector space. This method allows for the adaptation to new types of defects and changing shooting conditions, making the model more robust and accurate in detecting anomalies in coffee beans.\",\n          \"Qdrant optimizes memory and search speed for sparse vectors by utilizing an inverted index structure to store vectors for each non-zero dimension. This approach allows Qdrant to efficiently represent sparse vectors, which are characterized by a high proportion of zeroes. By only storing information about non-zero dimensions, Qdrant reduces the memory footprint required to store sparse vectors and also speeds up search operations by focusing only on relevant dimensions during indexing and querying processes. This optimization ensures that Qdrant can handle sparse vectors effectively while maintaining efficient memory usage and search performance.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"source\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 10,\n        \"samples\": [\n          \"benchmarks/benchmark-faq.md\",\n          \"documentation/guides/configuration.md\",\n          \"documentation/concepts/optimizer.md\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"contexts\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"answer_correctness\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.0,\n        \"min\": 1.0,\n        \"max\": 1.0,\n        \"num_unique_values\": 1,\n        \"samples\": [\n          1.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"answer_relevancy\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.021810012539079885,\n        \"min\": 0.9242368723190696,\n        \"max\": 1.0000000000000007,\n        \"num_unique_values\": 20,\n        \"samples\": [\n          1.0000000000000007\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"faithfulness\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.07273929674533079,\n        \"min\": 0.6666666666666666,\n        \"max\": 1.0,\n        \"num_unique_values\": 2,\n        \"samples\": [\n          0.6666666666666666\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"context_recall\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.2073388964700497,\n        \"min\": 0.16666666666666666,\n        \"max\": 1.0,\n        \"num_unique_values\": 3,\n        \"samples\": [\n          0.5\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"context_precision\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4.550560269027492e-16,\n        \"min\": 0.9999999999,\n        \"max\": 0.9999999999,\n        \"num_unique_values\": 1,\n        \"samples\": [\n          0.9999999999\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}",
       "type": "dataframe",
       "variable_name": "evaluation_result_df"
      },
      "text/html": [
       "\n",
       "  <div id=\"df-19f53f6a-0bc3-42ce-951c-7acdca857ae2\" class=\"colab-df-container\">\n",
       "    <div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question</th>\n",
       "      <th>ground_truth</th>\n",
       "      <th>answer</th>\n",
       "      <th>source</th>\n",
       "      <th>contexts</th>\n",
       "      <th>answer_correctness</th>\n",
       "      <th>answer_relevancy</th>\n",
       "      <th>faithfulness</th>\n",
       "      <th>context_recall</th>\n",
       "      <th>context_precision</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>What is the purpose of the locking feature in ...</td>\n",
       "      <td>The locking feature in Qdrant administration t...</td>\n",
       "      <td>The locking feature in Qdrant administration t...</td>\n",
       "      <td>documentation/guides/administration.md</td>\n",
       "      <td>[title: Administration\\n\\nweight: 10\\n\\naliase...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>What is the significance of the `on_disk_paylo...</td>\n",
       "      <td>The `on_disk_payload` setting in the storage c...</td>\n",
       "      <td>The `on_disk_payload` setting in the storage c...</td>\n",
       "      <td>documentation/guides/configuration.md</td>\n",
       "      <td>[storage:\\n\\n  # Where to store all the data\\n...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.953128</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>What is the significance of the parameter `max...</td>\n",
       "      <td>The `max_segment_size_kb` parameter in the con...</td>\n",
       "      <td>The `max_segment_size_kb` parameter in the con...</td>\n",
       "      <td>documentation/guides/configuration.md</td>\n",
       "      <td>[default_segment_number: 0\\n\\n\\n\\n    # Do not...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.979010</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>What is the significance of the `indexing_thre...</td>\n",
       "      <td>The `indexing_threshold_kb` parameter specifie...</td>\n",
       "      <td>The `indexing_threshold_kb` parameter specifie...</td>\n",
       "      <td>documentation/guides/configuration.md</td>\n",
       "      <td>[memmap_threshold_kb: null\\n\\n\\n\\n    # Maximu...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.969266</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>How can you optimize Qdrant for minimizing lat...</td>\n",
       "      <td>To optimize Qdrant for minimizing latency in s...</td>\n",
       "      <td>To optimize Qdrant for minimizing latency in s...</td>\n",
       "      <td>documentation/guides/optimize.md</td>\n",
       "      <td>[- `exact` - if set to `true`, will perform ex...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>How can Qdrant be configured to optimize throu...</td>\n",
       "      <td>To optimize throughput in Qdrant and utilize m...</td>\n",
       "      <td>To optimize throughput in Qdrant and utilize m...</td>\n",
       "      <td>documentation/guides/optimize.md</td>\n",
       "      <td>[.get();\\n\\n```\\n\\n\\n\\n```csharp\\n\\nusing Qdra...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.924237</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>How can one enable distributed deployment mode...</td>\n",
       "      <td>To enable distributed deployment mode in Qdran...</td>\n",
       "      <td>To enable distributed deployment mode in Qdran...</td>\n",
       "      <td>documentation/guides/distributed_deployment.md</td>\n",
       "      <td>[---\\n\\ntitle: Distributed Deployment\\n\\nweigh...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>What is the recommended approach for determini...</td>\n",
       "      <td>When creating a collection using Qdrant, it is...</td>\n",
       "      <td>When creating a collection using Qdrant, it is...</td>\n",
       "      <td>documentation/guides/distributed_deployment.md</td>\n",
       "      <td>[using Qdrant.Client.Grpc;\\n\\n\\n\\nvar client =...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.967004</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>How is the discovery score computed in the con...</td>\n",
       "      <td>In the context provided, the discovery score i...</td>\n",
       "      <td>In the context provided, the discovery score i...</td>\n",
       "      <td>documentation/concepts/explore.md</td>\n",
       "      <td>[-1, &amp;\\quad s(v^+) &lt; s(v^-)\\n\\n\\end{cases}\\n\\n...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.968043</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>How does Qdrant optimize storage at the segmen...</td>\n",
       "      <td>Qdrant optimizes storage at the segment level ...</td>\n",
       "      <td>Qdrant optimizes storage at the segment level ...</td>\n",
       "      <td>documentation/concepts/optimizer.md</td>\n",
       "      <td>[---\\n\\ntitle: Optimizer\\n\\nweight: 70\\n\\nalia...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>What is the purpose of the Vacuum Optimizer in...</td>\n",
       "      <td>The Vacuum Optimizer in the Qdrant database sy...</td>\n",
       "      <td>The Vacuum Optimizer in the Qdrant database sy...</td>\n",
       "      <td>documentation/concepts/optimizer.md</td>\n",
       "      <td>[Like many other databases, Qdrant does not de...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>What is the purpose of shard replication in Qd...</td>\n",
       "      <td>Shard replication in Qdrant increases the reli...</td>\n",
       "      <td>Shard replication in Qdrant increases the reli...</td>\n",
       "      <td>None</td>\n",
       "      <td>[[' method is the simplest available. It simpl...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.938783</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>What are the steps to set up product quantizat...</td>\n",
       "      <td>To set up product quantization in QdrantClient...</td>\n",
       "      <td>To set up product quantization in QdrantClient...</td>\n",
       "      <td>None</td>\n",
       "      <td>[['`\\n\\n\\n\\n`always_ram` - whether to keep qua...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.962514</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>What are the different write ordering options ...</td>\n",
       "      <td>Qdrant provides three different write ordering...</td>\n",
       "      <td>Qdrant provides three different write ordering...</td>\n",
       "      <td>None</td>\n",
       "      <td>[[' hnsw_ef: Some(128),\\n\\n exact: Some(false)...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.983073</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>How can Qdrant be optimized for low memory foo...</td>\n",
       "      <td>To optimize Qdrant for low memory footprint an...</td>\n",
       "      <td>To optimize Qdrant for low memory footprint an...</td>\n",
       "      <td>None</td>\n",
       "      <td>[['---\\n\\ntitle: Optimize Resources\\n\\nweight:...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.985083</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>How does Qdrant optimize memory and search spe...</td>\n",
       "      <td>Qdrant optimizes memory and search speed for s...</td>\n",
       "      <td>Qdrant optimizes memory and search speed for s...</td>\n",
       "      <td>documentation/concepts/indexing.md</td>\n",
       "      <td>[### Key Features of Sparse Vector Index\\n\\n- ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>How does Qdrant address the limitations faced ...</td>\n",
       "      <td>Qdrant addresses the limitations faced by the ...</td>\n",
       "      <td>Qdrant addresses the limitations faced by the ...</td>\n",
       "      <td>documentation/concepts/indexing.md</td>\n",
       "      <td>[In the case of weak filters, you can use the ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.959215</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>How is metric learning utilized in addressing ...</td>\n",
       "      <td>Metric learning is utilized in addressing the ...</td>\n",
       "      <td>Metric learning is utilized in addressing the ...</td>\n",
       "      <td>articles/detecting-coffee-anomalies.md</td>\n",
       "      <td>[They have collected and labeled more than **3...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.982490</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>How can one reproduce the benchmark for Open S...</td>\n",
       "      <td>To reproduce the benchmark for Open Source vec...</td>\n",
       "      <td>To reproduce the benchmark for Open Source vec...</td>\n",
       "      <td>benchmarks/benchmark-faq.md</td>\n",
       "      <td>[That makes the comparison unfair. That’s why ...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>What is the purpose of the `ordering` paramete...</td>\n",
       "      <td>The `ordering` parameter in Qdrant is used wit...</td>\n",
       "      <td>The `ordering` parameter in Qdrant is used wit...</td>\n",
       "      <td>documentation/guides/distributed_deployment.md</td>\n",
       "      <td>[- Write `ordering` param, can be used with up...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.981956</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "    <div class=\"colab-df-buttons\">\n",
       "\n",
       "  <div class=\"colab-df-container\">\n",
       "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-19f53f6a-0bc3-42ce-951c-7acdca857ae2')\"\n",
       "            title=\"Convert this dataframe to an interactive table.\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
       "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
       "  </svg>\n",
       "    </button>\n",
       "\n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    .colab-df-buttons div {\n",
       "      margin-bottom: 4px;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "    <script>\n",
       "      const buttonEl =\n",
       "        document.querySelector('#df-19f53f6a-0bc3-42ce-951c-7acdca857ae2 button.colab-df-convert');\n",
       "      buttonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "      async function convertToInteractive(key) {\n",
       "        const element = document.querySelector('#df-19f53f6a-0bc3-42ce-951c-7acdca857ae2');\n",
       "        const dataTable =\n",
       "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                    [key], {});\n",
       "        if (!dataTable) return;\n",
       "\n",
       "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "          + ' to learn more about interactive tables.';\n",
       "        element.innerHTML = '';\n",
       "        dataTable['output_type'] = 'display_data';\n",
       "        await google.colab.output.renderOutput(dataTable, element);\n",
       "        const docLink = document.createElement('div');\n",
       "        docLink.innerHTML = docLinkHtml;\n",
       "        element.appendChild(docLink);\n",
       "      }\n",
       "    </script>\n",
       "  </div>\n",
       "\n",
       "\n",
       "<div id=\"df-01b94519-6532-4c56-8299-55804fea1ae1\">\n",
       "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-01b94519-6532-4c56-8299-55804fea1ae1')\"\n",
       "            title=\"Suggest charts\"\n",
       "            style=\"display:none;\">\n",
       "\n",
       "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "     width=\"24px\">\n",
       "    <g>\n",
       "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
       "    </g>\n",
       "</svg>\n",
       "  </button>\n",
       "\n",
       "<style>\n",
       "  .colab-df-quickchart {\n",
       "      --bg-color: #E8F0FE;\n",
       "      --fill-color: #1967D2;\n",
       "      --hover-bg-color: #E2EBFA;\n",
       "      --hover-fill-color: #174EA6;\n",
       "      --disabled-fill-color: #AAA;\n",
       "      --disabled-bg-color: #DDD;\n",
       "  }\n",
       "\n",
       "  [theme=dark] .colab-df-quickchart {\n",
       "      --bg-color: #3B4455;\n",
       "      --fill-color: #D2E3FC;\n",
       "      --hover-bg-color: #434B5C;\n",
       "      --hover-fill-color: #FFFFFF;\n",
       "      --disabled-bg-color: #3B4455;\n",
       "      --disabled-fill-color: #666;\n",
       "  }\n",
       "\n",
       "  .colab-df-quickchart {\n",
       "    background-color: var(--bg-color);\n",
       "    border: none;\n",
       "    border-radius: 50%;\n",
       "    cursor: pointer;\n",
       "    display: none;\n",
       "    fill: var(--fill-color);\n",
       "    height: 32px;\n",
       "    padding: 0;\n",
       "    width: 32px;\n",
       "  }\n",
       "\n",
       "  .colab-df-quickchart:hover {\n",
       "    background-color: var(--hover-bg-color);\n",
       "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "    fill: var(--button-hover-fill-color);\n",
       "  }\n",
       "\n",
       "  .colab-df-quickchart-complete:disabled,\n",
       "  .colab-df-quickchart-complete:disabled:hover {\n",
       "    background-color: var(--disabled-bg-color);\n",
       "    fill: var(--disabled-fill-color);\n",
       "    box-shadow: none;\n",
       "  }\n",
       "\n",
       "  .colab-df-spinner {\n",
       "    border: 2px solid var(--fill-color);\n",
       "    border-color: transparent;\n",
       "    border-bottom-color: var(--fill-color);\n",
       "    animation:\n",
       "      spin 1s steps(1) infinite;\n",
       "  }\n",
       "\n",
       "  @keyframes spin {\n",
       "    0% {\n",
       "      border-color: transparent;\n",
       "      border-bottom-color: var(--fill-color);\n",
       "      border-left-color: var(--fill-color);\n",
       "    }\n",
       "    20% {\n",
       "      border-color: transparent;\n",
       "      border-left-color: var(--fill-color);\n",
       "      border-top-color: var(--fill-color);\n",
       "    }\n",
       "    30% {\n",
       "      border-color: transparent;\n",
       "      border-left-color: var(--fill-color);\n",
       "      border-top-color: var(--fill-color);\n",
       "      border-right-color: var(--fill-color);\n",
       "    }\n",
       "    40% {\n",
       "      border-color: transparent;\n",
       "      border-right-color: var(--fill-color);\n",
       "      border-top-color: var(--fill-color);\n",
       "    }\n",
       "    60% {\n",
       "      border-color: transparent;\n",
       "      border-right-color: var(--fill-color);\n",
       "    }\n",
       "    80% {\n",
       "      border-color: transparent;\n",
       "      border-right-color: var(--fill-color);\n",
       "      border-bottom-color: var(--fill-color);\n",
       "    }\n",
       "    90% {\n",
       "      border-color: transparent;\n",
       "      border-bottom-color: var(--fill-color);\n",
       "    }\n",
       "  }\n",
       "</style>\n",
       "\n",
       "  <script>\n",
       "    async function quickchart(key) {\n",
       "      const quickchartButtonEl =\n",
       "        document.querySelector('#' + key + ' button');\n",
       "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
       "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
       "      try {\n",
       "        const charts = await google.colab.kernel.invokeFunction(\n",
       "            'suggestCharts', [key], {});\n",
       "      } catch (error) {\n",
       "        console.error('Error during call to suggestCharts:', error);\n",
       "      }\n",
       "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
       "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
       "    }\n",
       "    (() => {\n",
       "      let quickchartButtonEl =\n",
       "        document.querySelector('#df-01b94519-6532-4c56-8299-55804fea1ae1 button');\n",
       "      quickchartButtonEl.style.display =\n",
       "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "    })();\n",
       "  </script>\n",
       "</div>\n",
       "    </div>\n",
       "  </div>\n"
      ],
      "text/plain": [
       "                                             question  \\\n",
       "0   What is the purpose of the locking feature in ...   \n",
       "1   What is the significance of the `on_disk_paylo...   \n",
       "2   What is the significance of the parameter `max...   \n",
       "3   What is the significance of the `indexing_thre...   \n",
       "4   How can you optimize Qdrant for minimizing lat...   \n",
       "5   How can Qdrant be configured to optimize throu...   \n",
       "6   How can one enable distributed deployment mode...   \n",
       "7   What is the recommended approach for determini...   \n",
       "8   How is the discovery score computed in the con...   \n",
       "9   How does Qdrant optimize storage at the segmen...   \n",
       "10  What is the purpose of the Vacuum Optimizer in...   \n",
       "11  What is the purpose of shard replication in Qd...   \n",
       "12  What are the steps to set up product quantizat...   \n",
       "13  What are the different write ordering options ...   \n",
       "14  How can Qdrant be optimized for low memory foo...   \n",
       "15  How does Qdrant optimize memory and search spe...   \n",
       "16  How does Qdrant address the limitations faced ...   \n",
       "17  How is metric learning utilized in addressing ...   \n",
       "18  How can one reproduce the benchmark for Open S...   \n",
       "19  What is the purpose of the `ordering` paramete...   \n",
       "\n",
       "                                         ground_truth  \\\n",
       "0   The locking feature in Qdrant administration t...   \n",
       "1   The `on_disk_payload` setting in the storage c...   \n",
       "2   The `max_segment_size_kb` parameter in the con...   \n",
       "3   The `indexing_threshold_kb` parameter specifie...   \n",
       "4   To optimize Qdrant for minimizing latency in s...   \n",
       "5   To optimize throughput in Qdrant and utilize m...   \n",
       "6   To enable distributed deployment mode in Qdran...   \n",
       "7   When creating a collection using Qdrant, it is...   \n",
       "8   In the context provided, the discovery score i...   \n",
       "9   Qdrant optimizes storage at the segment level ...   \n",
       "10  The Vacuum Optimizer in the Qdrant database sy...   \n",
       "11  Shard replication in Qdrant increases the reli...   \n",
       "12  To set up product quantization in QdrantClient...   \n",
       "13  Qdrant provides three different write ordering...   \n",
       "14  To optimize Qdrant for low memory footprint an...   \n",
       "15  Qdrant optimizes memory and search speed for s...   \n",
       "16  Qdrant addresses the limitations faced by the ...   \n",
       "17  Metric learning is utilized in addressing the ...   \n",
       "18  To reproduce the benchmark for Open Source vec...   \n",
       "19  The `ordering` parameter in Qdrant is used wit...   \n",
       "\n",
       "                                               answer  \\\n",
       "0   The locking feature in Qdrant administration t...   \n",
       "1   The `on_disk_payload` setting in the storage c...   \n",
       "2   The `max_segment_size_kb` parameter in the con...   \n",
       "3   The `indexing_threshold_kb` parameter specifie...   \n",
       "4   To optimize Qdrant for minimizing latency in s...   \n",
       "5   To optimize throughput in Qdrant and utilize m...   \n",
       "6   To enable distributed deployment mode in Qdran...   \n",
       "7   When creating a collection using Qdrant, it is...   \n",
       "8   In the context provided, the discovery score i...   \n",
       "9   Qdrant optimizes storage at the segment level ...   \n",
       "10  The Vacuum Optimizer in the Qdrant database sy...   \n",
       "11  Shard replication in Qdrant increases the reli...   \n",
       "12  To set up product quantization in QdrantClient...   \n",
       "13  Qdrant provides three different write ordering...   \n",
       "14  To optimize Qdrant for low memory footprint an...   \n",
       "15  Qdrant optimizes memory and search speed for s...   \n",
       "16  Qdrant addresses the limitations faced by the ...   \n",
       "17  Metric learning is utilized in addressing the ...   \n",
       "18  To reproduce the benchmark for Open Source vec...   \n",
       "19  The `ordering` parameter in Qdrant is used wit...   \n",
       "\n",
       "                                            source  \\\n",
       "0           documentation/guides/administration.md   \n",
       "1            documentation/guides/configuration.md   \n",
       "2            documentation/guides/configuration.md   \n",
       "3            documentation/guides/configuration.md   \n",
       "4                 documentation/guides/optimize.md   \n",
       "5                 documentation/guides/optimize.md   \n",
       "6   documentation/guides/distributed_deployment.md   \n",
       "7   documentation/guides/distributed_deployment.md   \n",
       "8                documentation/concepts/explore.md   \n",
       "9              documentation/concepts/optimizer.md   \n",
       "10             documentation/concepts/optimizer.md   \n",
       "11                                            None   \n",
       "12                                            None   \n",
       "13                                            None   \n",
       "14                                            None   \n",
       "15              documentation/concepts/indexing.md   \n",
       "16              documentation/concepts/indexing.md   \n",
       "17          articles/detecting-coffee-anomalies.md   \n",
       "18                     benchmarks/benchmark-faq.md   \n",
       "19  documentation/guides/distributed_deployment.md   \n",
       "\n",
       "                                             contexts  answer_correctness  \\\n",
       "0   [title: Administration\\n\\nweight: 10\\n\\naliase...                 1.0   \n",
       "1   [storage:\\n\\n  # Where to store all the data\\n...                 1.0   \n",
       "2   [default_segment_number: 0\\n\\n\\n\\n    # Do not...                 1.0   \n",
       "3   [memmap_threshold_kb: null\\n\\n\\n\\n    # Maximu...                 1.0   \n",
       "4   [- `exact` - if set to `true`, will perform ex...                 1.0   \n",
       "5   [.get();\\n\\n```\\n\\n\\n\\n```csharp\\n\\nusing Qdra...                 1.0   \n",
       "6   [---\\n\\ntitle: Distributed Deployment\\n\\nweigh...                 1.0   \n",
       "7   [using Qdrant.Client.Grpc;\\n\\n\\n\\nvar client =...                 1.0   \n",
       "8   [-1, &\\quad s(v^+) < s(v^-)\\n\\n\\end{cases}\\n\\n...                 1.0   \n",
       "9   [---\\n\\ntitle: Optimizer\\n\\nweight: 70\\n\\nalia...                 1.0   \n",
       "10  [Like many other databases, Qdrant does not de...                 1.0   \n",
       "11  [[' method is the simplest available. It simpl...                 1.0   \n",
       "12  [['`\\n\\n\\n\\n`always_ram` - whether to keep qua...                 1.0   \n",
       "13  [[' hnsw_ef: Some(128),\\n\\n exact: Some(false)...                 1.0   \n",
       "14  [['---\\n\\ntitle: Optimize Resources\\n\\nweight:...                 1.0   \n",
       "15  [### Key Features of Sparse Vector Index\\n\\n- ...                 1.0   \n",
       "16  [In the case of weak filters, you can use the ...                 1.0   \n",
       "17  [They have collected and labeled more than **3...                 1.0   \n",
       "18  [That makes the comparison unfair. That’s why ...                 1.0   \n",
       "19  [- Write `ordering` param, can be used with up...                 1.0   \n",
       "\n",
       "    answer_relevancy  faithfulness  context_recall  context_precision  \n",
       "0           1.000000      1.000000        0.500000                1.0  \n",
       "1           0.953128      1.000000        1.000000                1.0  \n",
       "2           0.979010      1.000000        1.000000                1.0  \n",
       "3           0.969266      1.000000        1.000000                1.0  \n",
       "4           1.000000      1.000000        1.000000                1.0  \n",
       "5           0.924237      1.000000        1.000000                1.0  \n",
       "6           1.000000      1.000000        1.000000                1.0  \n",
       "7           0.967004      1.000000        1.000000                1.0  \n",
       "8           0.968043      1.000000        1.000000                1.0  \n",
       "9           1.000000      1.000000        1.000000                1.0  \n",
       "10          1.000000      1.000000        0.166667                1.0  \n",
       "11          0.938783      1.000000        1.000000                1.0  \n",
       "12          0.962514      1.000000        1.000000                1.0  \n",
       "13          0.983073      1.000000        1.000000                1.0  \n",
       "14          0.985083      1.000000        1.000000                1.0  \n",
       "15          1.000000      1.000000        1.000000                1.0  \n",
       "16          0.959215      0.666667        1.000000                1.0  \n",
       "17          0.982490      1.000000        1.000000                1.0  \n",
       "18          1.000000      1.000000        1.000000                1.0  \n",
       "19          0.981956      1.000000        1.000000                1.0  "
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluation_result_df_baseline = result_baseline.to_pandas()\n",
    "evaluation_result_df_baseline.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **17. Evaluation of RAG with Retrieval Window = 3**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question</th>\n",
       "      <th>answer</th>\n",
       "      <th>contexts</th>\n",
       "      <th>ground_truth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>What is vaccum optimizer ?</td>\n",
       "      <td>The vacuum optimizer is a feature that helps ...</td>\n",
       "      <td>['optimizersConfig: new OptimizersConfigDiff {...</td>\n",
       "      <td>The Vacuum Optimizer is a feature in the Qdran...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Tell me about ‘always_ram’ parameter?</td>\n",
       "      <td>The `always_ram` parameter is used to determi...</td>\n",
       "      <td>['\"compression\": \"x32\",\\n\\n                   ...</td>\n",
       "      <td>In the context of vector quantization in Qdran...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>What is difference between scalar and product ...</td>\n",
       "      <td>Scalar quantization is a method of compressin...</td>\n",
       "      <td>['But there are some tradeoffs. Product quanti...</td>\n",
       "      <td>While both methods aim to reduce the memory fo...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>What is ‘best_score’ strategy?</td>\n",
       "      <td>The ‘best_score’ strategy is a new strategy i...</td>\n",
       "      <td>['#   [ScoredPoint(id=3, version=0, score=0.09...</td>\n",
       "      <td>The `best_score` strategy is based on the idea...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>How does oversampling helps?</td>\n",
       "      <td>Oversampling is a technique used to balance t...</td>\n",
       "      <td>['oversampling helps equalize the representati...</td>\n",
       "      <td>Defines how many extra vectors should be pre-s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>What is the purpose of ‘CreatePayloadIndexAsync’?</td>\n",
       "      <td>The purpose of 'CreatePayloadIndexAsync' is t...</td>\n",
       "      <td>['```java\\n\\nimport io.qdrant.client.QdrantCli...</td>\n",
       "      <td>CreatePayloadIndexAsync is a method in the Qdr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>What is the purpose of ef_construct in HNSW ?</td>\n",
       "      <td>The ef_construct parameter in HNSW is used to...</td>\n",
       "      <td>['(\"my_vector\".into()),\\n\\n                   ...</td>\n",
       "      <td>In HNSW algorithm the ef_construct parameter i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>How do you use ‘ordering’ parameter?</td>\n",
       "      <td>To use the ‘ordering’ parameter, you need to ...</td>\n",
       "      <td>['```http\\n\\nPUT /collections/{collection_name...</td>\n",
       "      <td>Write ordering can be specified for any write ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>What is significance of ‘on_disk_payload’ sett...</td>\n",
       "      <td>The 'on_disk_payload' setting is used to defi...</td>\n",
       "      <td>['* `on_disk_payload` - defines where to store...</td>\n",
       "      <td>The `on_disk_payload` setting in the storage c...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>What is the impact of ‘write_consistency_facto...</td>\n",
       "      <td>The 'write_consistency_factor' determines the...</td>\n",
       "      <td>['### Write consistency factor\\n\\n\\n\\nThe `wri...</td>\n",
       "      <td>The `write_consistency_factor` parameter in a ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            question  \\\n",
       "0                         What is vaccum optimizer ?   \n",
       "1              Tell me about ‘always_ram’ parameter?   \n",
       "2  What is difference between scalar and product ...   \n",
       "3                     What is ‘best_score’ strategy?   \n",
       "4                       How does oversampling helps?   \n",
       "5  What is the purpose of ‘CreatePayloadIndexAsync’?   \n",
       "6      What is the purpose of ef_construct in HNSW ?   \n",
       "7               How do you use ‘ordering’ parameter?   \n",
       "8  What is significance of ‘on_disk_payload’ sett...   \n",
       "9  What is the impact of ‘write_consistency_facto...   \n",
       "\n",
       "                                              answer  \\\n",
       "0   The vacuum optimizer is a feature that helps ...   \n",
       "1   The `always_ram` parameter is used to determi...   \n",
       "2   Scalar quantization is a method of compressin...   \n",
       "3   The ‘best_score’ strategy is a new strategy i...   \n",
       "4   Oversampling is a technique used to balance t...   \n",
       "5   The purpose of 'CreatePayloadIndexAsync' is t...   \n",
       "6   The ef_construct parameter in HNSW is used to...   \n",
       "7   To use the ‘ordering’ parameter, you need to ...   \n",
       "8   The 'on_disk_payload' setting is used to defi...   \n",
       "9   The 'write_consistency_factor' determines the...   \n",
       "\n",
       "                                            contexts  \\\n",
       "0  ['optimizersConfig: new OptimizersConfigDiff {...   \n",
       "1  ['\"compression\": \"x32\",\\n\\n                   ...   \n",
       "2  ['But there are some tradeoffs. Product quanti...   \n",
       "3  ['#   [ScoredPoint(id=3, version=0, score=0.09...   \n",
       "4  ['oversampling helps equalize the representati...   \n",
       "5  ['```java\\n\\nimport io.qdrant.client.QdrantCli...   \n",
       "6  ['(\"my_vector\".into()),\\n\\n                   ...   \n",
       "7  ['```http\\n\\nPUT /collections/{collection_name...   \n",
       "8  ['* `on_disk_payload` - defines where to store...   \n",
       "9  ['### Write consistency factor\\n\\n\\n\\nThe `wri...   \n",
       "\n",
       "                                        ground_truth  \n",
       "0  The Vacuum Optimizer is a feature in the Qdran...  \n",
       "1  In the context of vector quantization in Qdran...  \n",
       "2  While both methods aim to reduce the memory fo...  \n",
       "3  The `best_score` strategy is based on the idea...  \n",
       "4  Defines how many extra vectors should be pre-s...  \n",
       "5  CreatePayloadIndexAsync is a method in the Qdr...  \n",
       "6  In HNSW algorithm the ef_construct parameter i...  \n",
       "7  Write ordering can be specified for any write ...  \n",
       "8  The `on_disk_payload` setting in the storage c...  \n",
       "9  The `write_consistency_factor` parameter in a ...  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Evaluating Experiment 3 - With Retrieval Window = 3\n",
    "\n",
    "from ragas import evaluate\n",
    "eval_dataset_with_dspy_512_3 = pd.read_csv('rag_response_with_dspy_512_3.csv')\n",
    "eval_dataset_with_dspy_512_3.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 86,
     "referenced_widgets": [
      "d57d4563682645da8ac4c2551157fc05",
      "02f1ae27b6a2480280c4e6e7edda92bc",
      "94d8dd42b15c4bd79a4c3260347a9844",
      "7b069b3aa893463e8fc8ebd4858d7056",
      "be4d621a67d941d587dcf2967f10df6a",
      "297edd80c3a747e79242334c25d23d73",
      "09767b978f6743f6a54e0080b38d1052",
      "5f4e3650bb474f2595d073277db33837",
      "ae80fcc82fbe45cb9c924abc696dc1d7",
      "21352eeac90a4bb5b326ad0260c089a3",
      "a5251a2de0e947239eea948b04e01d6e"
     ]
    },
    "id": "j1A-aebaco0r",
    "outputId": "66f443b8-1cd7-4e99-fc30-989fd25f89c5"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6c990e9ec59e41d0a14f2fad4744f0bd",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating:   0%|          | 0/80 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result_512_with_dspy_3 = evaluate(\n",
    "        rag_response_dataset_dspy_512_3,\n",
    "        metrics=[\n",
    "                faithfulness,\n",
    "                answer_relevancy,\n",
    "                context_recall,\n",
    "                context_precision,\n",
    "                context_relevancy,\n",
    "                context_entity_recall,\n",
    "                answer_similarity,\n",
    "                answer_correctness\n",
    "        ],\n",
    "       raise_exceptions=False \n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/atitaarora/qdrant/workspace/qdrant-rag-eval/workshop-rag-eval-qdrant-ragas/ragas/lib/python3.10/site-packages/datasets/table.py:1395: FutureWarning: promote has been superseded by promote_options='default'.\n",
      "  block_group = [InMemoryTable(cls._concat_blocks(list(block_group), axis=axis))]\n",
      "/Users/atitaarora/qdrant/workspace/qdrant-rag-eval/workshop-rag-eval-qdrant-ragas/ragas/lib/python3.10/site-packages/datasets/table.py:1421: FutureWarning: promote has been superseded by promote_options='default'.\n",
      "  table = cls._concat_blocks(blocks, axis=0)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question</th>\n",
       "      <th>answer</th>\n",
       "      <th>contexts</th>\n",
       "      <th>ground_truth</th>\n",
       "      <th>faithfulness</th>\n",
       "      <th>answer_relevancy</th>\n",
       "      <th>context_recall</th>\n",
       "      <th>context_precision</th>\n",
       "      <th>context_relevancy</th>\n",
       "      <th>context_entity_recall</th>\n",
       "      <th>answer_similarity</th>\n",
       "      <th>answer_correctness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>What is vaccum optimizer ?</td>\n",
       "      <td>The vacuum optimizer is a feature that helps ...</td>\n",
       "      <td>[optimizersConfig: new OptimizersConfigDiff { ...</td>\n",
       "      <td>The Vacuum Optimizer is a feature in the Qdran...</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.941122</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.058824</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.914186</td>\n",
       "      <td>0.378524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Tell me about ‘always_ram’ parameter?</td>\n",
       "      <td>The `always_ram` parameter is used to determi...</td>\n",
       "      <td>[\"compression\": \"x32\",\\n\\n                    ...</td>\n",
       "      <td>In the context of vector quantization in Qdran...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.876962</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.023810</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.935454</td>\n",
       "      <td>0.733863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>What is difference between scalar and product ...</td>\n",
       "      <td>Scalar quantization is a method of compressin...</td>\n",
       "      <td>[But there are some tradeoffs. Product quantiz...</td>\n",
       "      <td>While both methods aim to reduce the memory fo...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.920023</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.153846</td>\n",
       "      <td>0.230769</td>\n",
       "      <td>0.921691</td>\n",
       "      <td>0.639514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>What is ‘best_score’ strategy?</td>\n",
       "      <td>The ‘best_score’ strategy is a new strategy i...</td>\n",
       "      <td>[#   [ScoredPoint(id=3, version=0, score=0.09,...</td>\n",
       "      <td>The `best_score` strategy is based on the idea...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.988289</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.347826</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.922464</td>\n",
       "      <td>0.692154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>How does oversampling helps?</td>\n",
       "      <td>Oversampling is a technique used to balance t...</td>\n",
       "      <td>[oversampling helps equalize the representatio...</td>\n",
       "      <td>Defines how many extra vectors should be pre-s...</td>\n",
       "      <td>0.833333</td>\n",
       "      <td>0.937466</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.583333</td>\n",
       "      <td>0.125000</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>0.876570</td>\n",
       "      <td>0.594143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>What is the purpose of ‘CreatePayloadIndexAsync’?</td>\n",
       "      <td>The purpose of 'CreatePayloadIndexAsync' is t...</td>\n",
       "      <td>[```java\\n\\nimport io.qdrant.client.QdrantClie...</td>\n",
       "      <td>CreatePayloadIndexAsync is a method in the Qdr...</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.992006</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.050000</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0.929798</td>\n",
       "      <td>0.446735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>What is the purpose of ef_construct in HNSW ?</td>\n",
       "      <td>The ef_construct parameter in HNSW is used to...</td>\n",
       "      <td>[(\"my_vector\".into()),\\n\\n                    ...</td>\n",
       "      <td>In HNSW algorithm the ef_construct parameter i...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.955200</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.017857</td>\n",
       "      <td>0.125000</td>\n",
       "      <td>0.905071</td>\n",
       "      <td>0.676284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>How do you use ‘ordering’ parameter?</td>\n",
       "      <td>To use the ‘ordering’ parameter, you need to ...</td>\n",
       "      <td>[```http\\n\\nPUT /collections/{collection_name}...</td>\n",
       "      <td>Write ordering can be specified for any write ...</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.957948</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.013699</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.776974</td>\n",
       "      <td>0.694218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>What is significance of ‘on_disk_payload’ sett...</td>\n",
       "      <td>The 'on_disk_payload' setting is used to defi...</td>\n",
       "      <td>[* `on_disk_payload` - defines where to store ...</td>\n",
       "      <td>The `on_disk_payload` setting in the storage c...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.967465</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.260870</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.917227</td>\n",
       "      <td>0.690845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>What is the impact of ‘write_consistency_facto...</td>\n",
       "      <td>The 'write_consistency_factor' determines the...</td>\n",
       "      <td>[### Write consistency factor\\n\\n\\n\\nThe `writ...</td>\n",
       "      <td>The `write_consistency_factor` parameter in a ...</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.943129</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.058824</td>\n",
       "      <td>0.538462</td>\n",
       "      <td>0.929972</td>\n",
       "      <td>0.607493</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            question  \\\n",
       "0                         What is vaccum optimizer ?   \n",
       "1              Tell me about ‘always_ram’ parameter?   \n",
       "2  What is difference between scalar and product ...   \n",
       "3                     What is ‘best_score’ strategy?   \n",
       "4                       How does oversampling helps?   \n",
       "5  What is the purpose of ‘CreatePayloadIndexAsync’?   \n",
       "6      What is the purpose of ef_construct in HNSW ?   \n",
       "7               How do you use ‘ordering’ parameter?   \n",
       "8  What is significance of ‘on_disk_payload’ sett...   \n",
       "9  What is the impact of ‘write_consistency_facto...   \n",
       "\n",
       "                                              answer  \\\n",
       "0   The vacuum optimizer is a feature that helps ...   \n",
       "1   The `always_ram` parameter is used to determi...   \n",
       "2   Scalar quantization is a method of compressin...   \n",
       "3   The ‘best_score’ strategy is a new strategy i...   \n",
       "4   Oversampling is a technique used to balance t...   \n",
       "5   The purpose of 'CreatePayloadIndexAsync' is t...   \n",
       "6   The ef_construct parameter in HNSW is used to...   \n",
       "7   To use the ‘ordering’ parameter, you need to ...   \n",
       "8   The 'on_disk_payload' setting is used to defi...   \n",
       "9   The 'write_consistency_factor' determines the...   \n",
       "\n",
       "                                            contexts  \\\n",
       "0  [optimizersConfig: new OptimizersConfigDiff { ...   \n",
       "1  [\"compression\": \"x32\",\\n\\n                    ...   \n",
       "2  [But there are some tradeoffs. Product quantiz...   \n",
       "3  [#   [ScoredPoint(id=3, version=0, score=0.09,...   \n",
       "4  [oversampling helps equalize the representatio...   \n",
       "5  [```java\\n\\nimport io.qdrant.client.QdrantClie...   \n",
       "6  [(\"my_vector\".into()),\\n\\n                    ...   \n",
       "7  [```http\\n\\nPUT /collections/{collection_name}...   \n",
       "8  [* `on_disk_payload` - defines where to store ...   \n",
       "9  [### Write consistency factor\\n\\n\\n\\nThe `writ...   \n",
       "\n",
       "                                        ground_truth  faithfulness  \\\n",
       "0  The Vacuum Optimizer is a feature in the Qdran...      0.800000   \n",
       "1  In the context of vector quantization in Qdran...      1.000000   \n",
       "2  While both methods aim to reduce the memory fo...      1.000000   \n",
       "3  The `best_score` strategy is based on the idea...      1.000000   \n",
       "4  Defines how many extra vectors should be pre-s...      0.833333   \n",
       "5  CreatePayloadIndexAsync is a method in the Qdr...      0.666667   \n",
       "6  In HNSW algorithm the ef_construct parameter i...      1.000000   \n",
       "7  Write ordering can be specified for any write ...      0.750000   \n",
       "8  The `on_disk_payload` setting in the storage c...      1.000000   \n",
       "9  The `write_consistency_factor` parameter in a ...      0.750000   \n",
       "\n",
       "   answer_relevancy  context_recall  context_precision  context_relevancy  \\\n",
       "0          0.941122        1.000000           1.000000           0.058824   \n",
       "1          0.876962        0.666667           1.000000           0.023810   \n",
       "2          0.920023        1.000000           1.000000           0.153846   \n",
       "3          0.988289        1.000000           1.000000           0.347826   \n",
       "4          0.937466        1.000000           0.583333           0.125000   \n",
       "5          0.992006        1.000000           1.000000           0.050000   \n",
       "6          0.955200        0.333333           1.000000           0.017857   \n",
       "7          0.957948        1.000000           1.000000           0.013699   \n",
       "8          0.967465        1.000000           1.000000           0.260870   \n",
       "9          0.943129        1.000000           1.000000           0.058824   \n",
       "\n",
       "   context_entity_recall  answer_similarity  answer_correctness  \n",
       "0               0.000000           0.914186            0.378524  \n",
       "1               0.142857           0.935454            0.733863  \n",
       "2               0.230769           0.921691            0.639514  \n",
       "3               0.000000           0.922464            0.692154  \n",
       "4               0.428571           0.876570            0.594143  \n",
       "5               0.142857           0.929798            0.446735  \n",
       "6               0.125000           0.905071            0.676284  \n",
       "7               0.000000           0.776974            0.694218  \n",
       "8               0.083333           0.917227            0.690845  \n",
       "9               0.538462           0.929972            0.607493  "
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluation_result_df_512_with_dspy_3 = result_512_with_dspy_3.to_pandas()\n",
    "evaluation_result_df_512_with_dspy_3.iloc[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **19. Compare results with the previous evaluation iteration**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Result comparison\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def compare_dataframes(df1, df2):\n",
    "    # Extract the specified columns from both dataframes\n",
    "    headers = [ 'faithfulness',\n",
    "                'answer_relevancy',\n",
    "                'context_recall',\n",
    "                'context_precision',\n",
    "                'context_relevancy',\n",
    "                'context_entity_recall',\n",
    "                'answer_similarity',\n",
    "                'answer_correctness'\n",
    "                ]\n",
    "    \n",
    "    # Create subplots\n",
    "    fig, axes = plt.subplots(nrows=1, ncols=len(headers), figsize=(20, 4))\n",
    "    \n",
    "    # Plot bar plots for each header\n",
    "    for i, header in enumerate(headers):\n",
    "        ax = axes[i]\n",
    "        ax.bar(['RW_2','RW_3'], [df1[header].mean(), df2[header].mean()])\n",
    "        ax.set_title(header)\n",
    "        ax.set_ylabel('Mean Value')\n",
    "    \n",
    "    # Adjust layout\n",
    "    plt.tight_layout()\n",
    "    \n",
    "    # Show plot\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x400 with 8 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Example result comparison:\n",
    "compare_dataframes(evaluation_result_df_512_with_dspy_2, evaluation_result_df_512_with_dspy_3)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### **20. Finally let us quickly glance the critical aspects of the responses**\n",
    "This is designed to assess submissions based on predefined aspects such as harmlessness , maliciousness, coherence , correctness and conciseness of the response."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import Dataset \n",
    "from ragas.metrics.critique import harmfulness\n",
    "from ragas.metrics.critique import maliciousness\n",
    "from ragas.metrics.critique import coherence\n",
    "from ragas.metrics.critique import correctness\n",
    "from ragas.metrics.critique import conciseness\n",
    "from ragas import evaluate\n",
    "\n",
    "def show_aspect_critic(dataset):\n",
    "    return evaluate(dataset,metrics=[harmfulness, maliciousness, coherence, correctness, conciseness,])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "02599a83503345dc98b72862fc7e423a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Evaluating:   0%|          | 0/50 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/atitaarora/qdrant/workspace/qdrant-rag-eval/workshop-rag-eval-qdrant-ragas/ragas/lib/python3.10/site-packages/datasets/table.py:1395: FutureWarning: promote has been superseded by promote_options='default'.\n",
      "  block_group = [InMemoryTable(cls._concat_blocks(list(block_group), axis=axis))]\n",
      "/Users/atitaarora/qdrant/workspace/qdrant-rag-eval/workshop-rag-eval-qdrant-ragas/ragas/lib/python3.10/site-packages/datasets/table.py:1421: FutureWarning: promote has been superseded by promote_options='default'.\n",
      "  table = cls._concat_blocks(blocks, axis=0)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question</th>\n",
       "      <th>answer</th>\n",
       "      <th>contexts</th>\n",
       "      <th>ground_truth</th>\n",
       "      <th>harmfulness</th>\n",
       "      <th>maliciousness</th>\n",
       "      <th>coherence</th>\n",
       "      <th>correctness</th>\n",
       "      <th>conciseness</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>What is vaccum optimizer ?</td>\n",
       "      <td>The vacuum optimizer is a feature that helps ...</td>\n",
       "      <td>[optimizersConfig: new OptimizersConfigDiff { ...</td>\n",
       "      <td>The Vacuum Optimizer is a feature in the Qdran...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Tell me about ‘always_ram’ parameter?</td>\n",
       "      <td>The `always_ram` parameter is used to determi...</td>\n",
       "      <td>[\"compression\": \"x32\",\\n\\n                    ...</td>\n",
       "      <td>In the context of vector quantization in Qdran...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>What is difference between scalar and product ...</td>\n",
       "      <td>Scalar quantization is a method of compressin...</td>\n",
       "      <td>[But there are some tradeoffs. Product quantiz...</td>\n",
       "      <td>While both methods aim to reduce the memory fo...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>What is ‘best_score’ strategy?</td>\n",
       "      <td>The ‘best_score’ strategy is a new strategy i...</td>\n",
       "      <td>[#   [ScoredPoint(id=3, version=0, score=0.09,...</td>\n",
       "      <td>The `best_score` strategy is based on the idea...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>How does oversampling helps?</td>\n",
       "      <td>Oversampling is a technique used to balance t...</td>\n",
       "      <td>[oversampling helps equalize the representatio...</td>\n",
       "      <td>Defines how many extra vectors should be pre-s...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>What is the purpose of ‘CreatePayloadIndexAsync’?</td>\n",
       "      <td>The purpose of 'CreatePayloadIndexAsync' is t...</td>\n",
       "      <td>[```java\\n\\nimport io.qdrant.client.QdrantClie...</td>\n",
       "      <td>CreatePayloadIndexAsync is a method in the Qdr...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>What is the purpose of ef_construct in HNSW ?</td>\n",
       "      <td>The ef_construct parameter in HNSW is used to...</td>\n",
       "      <td>[(\"my_vector\".into()),\\n\\n                    ...</td>\n",
       "      <td>In HNSW algorithm the ef_construct parameter i...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>How do you use ‘ordering’ parameter?</td>\n",
       "      <td>To use the ‘ordering’ parameter, you need to ...</td>\n",
       "      <td>[```http\\n\\nPUT /collections/{collection_name}...</td>\n",
       "      <td>Write ordering can be specified for any write ...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>What is significance of ‘on_disk_payload’ sett...</td>\n",
       "      <td>The 'on_disk_payload' setting is used to defi...</td>\n",
       "      <td>[* `on_disk_payload` - defines where to store ...</td>\n",
       "      <td>The `on_disk_payload` setting in the storage c...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>What is the impact of ‘write_consistency_facto...</td>\n",
       "      <td>The 'write_consistency_factor' determines the...</td>\n",
       "      <td>[### Write consistency factor\\n\\n\\n\\nThe `writ...</td>\n",
       "      <td>The `write_consistency_factor` parameter in a ...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            question  \\\n",
       "0                         What is vaccum optimizer ?   \n",
       "1              Tell me about ‘always_ram’ parameter?   \n",
       "2  What is difference between scalar and product ...   \n",
       "3                     What is ‘best_score’ strategy?   \n",
       "4                       How does oversampling helps?   \n",
       "5  What is the purpose of ‘CreatePayloadIndexAsync’?   \n",
       "6      What is the purpose of ef_construct in HNSW ?   \n",
       "7               How do you use ‘ordering’ parameter?   \n",
       "8  What is significance of ‘on_disk_payload’ sett...   \n",
       "9  What is the impact of ‘write_consistency_facto...   \n",
       "\n",
       "                                              answer  \\\n",
       "0   The vacuum optimizer is a feature that helps ...   \n",
       "1   The `always_ram` parameter is used to determi...   \n",
       "2   Scalar quantization is a method of compressin...   \n",
       "3   The ‘best_score’ strategy is a new strategy i...   \n",
       "4   Oversampling is a technique used to balance t...   \n",
       "5   The purpose of 'CreatePayloadIndexAsync' is t...   \n",
       "6   The ef_construct parameter in HNSW is used to...   \n",
       "7   To use the ‘ordering’ parameter, you need to ...   \n",
       "8   The 'on_disk_payload' setting is used to defi...   \n",
       "9   The 'write_consistency_factor' determines the...   \n",
       "\n",
       "                                            contexts  \\\n",
       "0  [optimizersConfig: new OptimizersConfigDiff { ...   \n",
       "1  [\"compression\": \"x32\",\\n\\n                    ...   \n",
       "2  [But there are some tradeoffs. Product quantiz...   \n",
       "3  [#   [ScoredPoint(id=3, version=0, score=0.09,...   \n",
       "4  [oversampling helps equalize the representatio...   \n",
       "5  [```java\\n\\nimport io.qdrant.client.QdrantClie...   \n",
       "6  [(\"my_vector\".into()),\\n\\n                    ...   \n",
       "7  [```http\\n\\nPUT /collections/{collection_name}...   \n",
       "8  [* `on_disk_payload` - defines where to store ...   \n",
       "9  [### Write consistency factor\\n\\n\\n\\nThe `writ...   \n",
       "\n",
       "                                        ground_truth  harmfulness  \\\n",
       "0  The Vacuum Optimizer is a feature in the Qdran...            0   \n",
       "1  In the context of vector quantization in Qdran...            0   \n",
       "2  While both methods aim to reduce the memory fo...            0   \n",
       "3  The `best_score` strategy is based on the idea...            0   \n",
       "4  Defines how many extra vectors should be pre-s...            0   \n",
       "5  CreatePayloadIndexAsync is a method in the Qdr...            0   \n",
       "6  In HNSW algorithm the ef_construct parameter i...            0   \n",
       "7  Write ordering can be specified for any write ...            0   \n",
       "8  The `on_disk_payload` setting in the storage c...            0   \n",
       "9  The `write_consistency_factor` parameter in a ...            0   \n",
       "\n",
       "   maliciousness  coherence  correctness  conciseness  \n",
       "0              0          1            1            1  \n",
       "1              0          1            1            1  \n",
       "2              0          1            1            1  \n",
       "3              0          1            1            1  \n",
       "4              0          1            1            1  \n",
       "5              0          1            1            1  \n",
       "6              0          1            1            1  \n",
       "7              0          1            1            1  \n",
       "8              0          1            1            1  \n",
       "9              0          1            1            1  "
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#show_aspect_critic(rag_response_dataset_dspy_512_2).to_pandas()\n",
    "show_aspect_critic(rag_response_dataset_dspy_512_3).to_pandas()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "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": "3.10.7"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "00963b69a2524a8a91f42e46a6ec8d0b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "00a353f3c5744ff3b76c1e0a382ba6e7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "00ad85e3dae347e09d272d3537b881fa": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "01119d8277364563a441ef40db68acf6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_22d970d039eb491ab9e97099477e8734",
      "placeholder": "​",
      "style": "IPY_MODEL_4ff810d3f45947579c552dbae9c45172",
      "value": " 1.24k/1.24k [00:00&lt;00:00, 5.53kB/s]"
     }
    },
    "01c3ae35884e45b18d062cb49ff6ff00": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "02281ac9625e4e4b991f888a70680cea": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "028943c4019943f4b57c383d573efb3c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_fe20966df20042ef8dc66c133f261079",
      "max": 28,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_6dc07cc1c9c6463a8146c798772269af",
      "value": 28
     }
    },
    "02f1ae27b6a2480280c4e6e7edda92bc": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_297edd80c3a747e79242334c25d23d73",
      "placeholder": "​",
      "style": "IPY_MODEL_09767b978f6743f6a54e0080b38d1052",
      "value": "Evaluating: 100%"
     }
    },
    "039d179c5821401c9d0b5caa056d805c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "05161fd8e6064845a28e1b44d6e89a73": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "068502547b1f4897ab49fb82c8c88b01": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "070cc007ec7c4c4483996f61b2d6f7ce": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "07d4537cd49e469e9d45ec51d859e3a8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "0812081f8c2048c6ba57fa709acd657e": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "09390a4fa70748be98f0eb3ffc73aca9": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "09767b978f6743f6a54e0080b38d1052": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "0ad979ea1b7347bdbf666fc457d8359c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_cf98e5b8b48a44209597fe59072e4b7d",
      "placeholder": "​",
      "style": "IPY_MODEL_0bd9b38743bb43c9b9ed26b500be200c",
      "value": " 695/695 [00:00&lt;00:00, 2.95kB/s]"
     }
    },
    "0bd9b38743bb43c9b9ed26b500be200c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "0c7a6ae7d8c54e39b0f436a96525bbc4": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "0e1729146be54c5395cf5e2dcd34e156": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "0ee834d61c9f4098b1bf91d3c77ba350": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_7c70d712ce514dbdaeaaeb7ae75da1c2",
       "IPY_MODEL_75cea6987d0a461ead31d2018a209ac0",
       "IPY_MODEL_9e468980647c4c6198e8dcb7d51e33b1"
      ],
      "layout": "IPY_MODEL_7745b86a86164b4782ca5f2b0ba349d2"
     }
    },
    "0f0ecde0157841e0a2bd661f59fea733": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "0faab5cb87f84d6fa29016e1775a376d": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "100d7c563ce846c8a18915559e66239d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_124305eadc9b4effabfe59052ffc4813",
       "IPY_MODEL_d2c3d4000c594a429ea12b7bcca280d1",
       "IPY_MODEL_7cedf7369e9f4927893cddbeffd1b145"
      ],
      "layout": "IPY_MODEL_cc6cc585fdf94f92bca17d83bcc63de4"
     }
    },
    "115ad0595f8d4ac7852a5983f5f1b2b5": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "121d7781e13448e8b0f4ac3c810f2ae4": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_75b76c5043724a98b18a0b26904b36e6",
      "placeholder": "​",
      "style": "IPY_MODEL_75adbf20c7304c0c87a37490b8077868",
      "value": " 1.52k/1.52k [00:00&lt;00:00, 6.48kB/s]"
     }
    },
    "124305eadc9b4effabfe59052ffc4813": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_818b70c8a7b2429a90c135c37c25a241",
      "placeholder": "​",
      "style": "IPY_MODEL_46ad56ca51004315a7ce5077c8b2d8c8",
      "value": "Downloading data: 100%"
     }
    },
    "129a7656b1854c51970c275d7431d2b8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "14947bce04ee442996215a1cbc009cbe": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_0812081f8c2048c6ba57fa709acd657e",
      "placeholder": "​",
      "style": "IPY_MODEL_0c7a6ae7d8c54e39b0f436a96525bbc4",
      "value": " 240/240 [00:00&lt;00:00, 3501.02it/s]"
     }
    },
    "156da40cbd834a518cddb08cbef3198d": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1641c25d0efe4cefb96fb2ef1898b582": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "16b940f370074141b23b3f8b1f38dbe2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "16c72bd3bc614955ae48760ce44e67e3": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_a8375dc193b8470dad9c73cffb87731c",
      "placeholder": "​",
      "style": "IPY_MODEL_58290b9bd8824604aac3bc5e9d7c7f20",
      "value": "Generating train split: "
     }
    },
    "1725ae2cf0c94aa6ba6e19ced1901e2c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_2b95ee150f6d4487909dff3c231cb429",
       "IPY_MODEL_028943c4019943f4b57c383d573efb3c",
       "IPY_MODEL_80b1dff28f5f4b30827a9c8189fefc30"
      ],
      "layout": "IPY_MODEL_dee469d9cfd44d9181c313da9bc195aa"
     }
    },
    "174745d013354898837589a21fe5fb8e": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "17acda449aa544218c3db7c03c7789c9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_dd2274085a1d415c9ab4dbf90813af2c",
      "placeholder": "​",
      "style": "IPY_MODEL_9a24f140106141d6b7c7a85c5a2bcbdd",
      "value": "100%"
     }
    },
    "18e47dede65141d7a11b1c0a583d5666": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "19438ce279ae4d109f2985e648a92c29": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1956ebc39de944c9ab28e11c642d7407": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1a0827041a0b46f3bc050631b29bc066": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_ccc3ce4362ea4cbbaa606fe28b74cadd",
      "placeholder": "​",
      "style": "IPY_MODEL_18e47dede65141d7a11b1c0a583d5666",
      "value": "Downloading data: 100%"
     }
    },
    "1c37f78ea68840aaa9c1fe77541ae81e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_174745d013354898837589a21fe5fb8e",
      "placeholder": "​",
      "style": "IPY_MODEL_f226de79c38b4ed4bae675ec213f18be",
      "value": " 81/81 [00:00&lt;00:00, 1796.08 examples/s]"
     }
    },
    "1d22fe0dfd73421e9bb5881320f8b4fb": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_dc04d32cfc724cb6ac3dd926041badce",
      "placeholder": "​",
      "style": "IPY_MODEL_4596d5f1c4e4433ea000a6a8b8cca946",
      "value": " 240/0 [00:00&lt;00:00, 1138.15 examples/s]"
     }
    },
    "1dad3b7b797e4da994566242893f5d4e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f7169503c1b049149bab032496d378b6",
      "placeholder": "​",
      "style": "IPY_MODEL_41606d866ebb4172989d79d1eea7139f",
      "value": " 80.7k/80.7k [00:00&lt;00:00, 644kB/s]"
     }
    },
    "1db283a5206741baa2fbd547e759b000": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "1e5a2b140acf441482218a1dcde84706": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "1ebdac4f164442049cecd2592e118f2d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_b322ee25201747b2aac1ef9cb4f51818",
       "IPY_MODEL_6cbc14541229488ab452cd547778c88c",
       "IPY_MODEL_1d22fe0dfd73421e9bb5881320f8b4fb"
      ],
      "layout": "IPY_MODEL_a3e3bfcf40ae464c929bff322b14d45f"
     }
    },
    "1ec3904b32a24a2a9a9b70763b870c95": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "1ed740b281d341ee867d782f53097ed3": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_413c186f31fd4f0dbd6e5956c9d31d93",
      "max": 80673,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_e3d7cae3f6244ad9990cd984bc2e4855",
      "value": 80673
     }
    },
    "1eec371016434a6b8c4ac8b83bd77b4f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "1faed09d17f74721a5462f42ad61caf1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f93019d15bf64bac93bad2c3c3c53714",
      "max": 105,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_f642894d3c574bdf8e19fbe42cf027d5",
      "value": 105
     }
    },
    "21352eeac90a4bb5b326ad0260c089a3": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "22d970d039eb491ab9e97099477e8734": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "244321954b7d4d52bb1a3e7f29c7269a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_1641c25d0efe4cefb96fb2ef1898b582",
      "max": 1519,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_068502547b1f4897ab49fb82c8c88b01",
      "value": 1519
     }
    },
    "25cb85fe706c4e16abbae238e6c90a65": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "25db4e94fc61492a9752600158fae7b4": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_bb7a42f693b0461994ccb49c9720f8dd",
      "placeholder": "​",
      "style": "IPY_MODEL_1db283a5206741baa2fbd547e759b000",
      "value": "Downloading data: 100%"
     }
    },
    "264c067001aa407cbbc6fee3e36428fc": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "26a051f4728a4c8c88afc75097bd648a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_0faab5cb87f84d6fa29016e1775a376d",
      "placeholder": "​",
      "style": "IPY_MODEL_01c3ae35884e45b18d062cb49ff6ff00",
      "value": "Generating train split: "
     }
    },
    "26cc261df479483b92af2650a2db0fef": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_97b1448064c64d829544c8d4cf4522de",
      "max": 240,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_0e1729146be54c5395cf5e2dcd34e156",
      "value": 240
     }
    },
    "2805c5fbcfab4313b321f7ab125d9f2b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "28d59a9ce5884e498c4b7763c79f1a65": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "297edd80c3a747e79242334c25d23d73": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "2b95ee150f6d4487909dff3c231cb429": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f818582473954544bfd497fe7f19a640",
      "placeholder": "​",
      "style": "IPY_MODEL_f57fb5f982594ae98c3bb71ce414e760",
      "value": "Downloading readme: 100%"
     }
    },
    "2b9df743e9e743d39e4b81d6d9d3e101": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_4055e023504d4be1b2a664b47708bd29",
      "placeholder": "​",
      "style": "IPY_MODEL_5b5fc2c030584237b3b3eabf0b76d9ef",
      "value": " 1.27k/1.27k [00:00&lt;00:00, 5.92kB/s]"
     }
    },
    "2ba7118900894716b3bf13ade4fdf8df": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e3448574566a44a7af6eaf656aaf04a2",
      "placeholder": "​",
      "style": "IPY_MODEL_8f12bf7b629347f29c6c5dd18299a9f7",
      "value": "Fetching 9 files: 100%"
     }
    },
    "2f2bb2c795584fbb957338a7fc0ca201": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_9014fbd6fdd5459cb4a2a2ccfef0410c",
      "max": 28,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_d736f19ec8e34bb88be9f2e72455686e",
      "value": 28
     }
    },
    "3071076a711c4db19459b6142fcfc8d0": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "30f018e451b54b169a31a967662bde72": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_25cb85fe706c4e16abbae238e6c90a65",
      "max": 711396,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_b77e8cf9674a442db0b08c64d3ff2c83",
      "value": 711396
     }
    },
    "30fcf2693b154033881fee05b0d17db6": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_326aae4816574318b01b999e18e10f37",
      "max": 124978,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_e94a8581d14842bda30438961ad294e1",
      "value": 124978
     }
    },
    "3124ef41463846a999c1fa01f1530b1a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "326aae4816574318b01b999e18e10f37": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "3301faa89b814001aaff9d492990a1bc": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "33f03cd63ab845fabceb1fd62c2c170f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "345208d4d61d4e378db846912706f1a1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_fe30399d1aeb44d3a6d69091b606b80f",
       "IPY_MODEL_f6fd6fcdc4f349fba497027d7fcf8a9f",
       "IPY_MODEL_b13dc56442bc437f97bd9f1165337b38"
      ],
      "layout": "IPY_MODEL_4c0d4444761346ee8cccb42ea17e6ed6"
     }
    },
    "34a594530ea04e2d927967699e5dc7cc": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_78055f624a0740d9b00dd15cef3dd038",
      "placeholder": "​",
      "style": "IPY_MODEL_1ec3904b32a24a2a9a9b70763b870c95",
      "value": "tokenizer.json: 100%"
     }
    },
    "35ac5a07c29f4597bed1f55dd1202778": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_cce81b158f2345619e2fec221770f9a2",
      "placeholder": "​",
      "style": "IPY_MODEL_00a353f3c5744ff3b76c1e0a382ba6e7",
      "value": " 66.5M/66.5M [00:00&lt;00:00, 110MB/s]"
     }
    },
    "3b9465a8398c45f498a414ac46c80355": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_64511930cae14bfeb384d0fbbcdc7548",
      "placeholder": "​",
      "style": "IPY_MODEL_cb8287f82d0043c49ba3da16b4f387c5",
      "value": "Downloading readme: 100%"
     }
    },
    "3c0415f3a6d944ebbddaa4fc04a22f22": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_bd64efaf240e42eb9b84dec9987a3c4f",
      "placeholder": "​",
      "style": "IPY_MODEL_fa1cfd25cadd4a42ab990a336eea095c",
      "value": "Evaluating: 100%"
     }
    },
    "3c065d9c08fd4367863e8e35ac476e81": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_9657423bd9074d51b484d46c8d18c42b",
       "IPY_MODEL_85938c37605a4f63afbb15c0d5a2d0c4",
       "IPY_MODEL_757d8e61bfb242719502565bacbd3a9b"
      ],
      "layout": "IPY_MODEL_529e93880868406ba7f8b1afc2c7c278"
     }
    },
    "3ef373eb9dae44f89a794c4cd8584ad4": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4055e023504d4be1b2a664b47708bd29": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "413c186f31fd4f0dbd6e5956c9d31d93": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "41606d866ebb4172989d79d1eea7139f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "42a0528e6dbe492093228804f946f4cd": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4596d5f1c4e4433ea000a6a8b8cca946": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "465a1be16a3948639b030652d3707de5": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "46ad56ca51004315a7ce5077c8b2d8c8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "48192520dcc44b41ae26f0f4b418ea20": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "494d8de002504275a5e079c832b74880": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "4beae8d8f5cc48a2bb7e9b7bebfb8e6f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "4c0d4444761346ee8cccb42ea17e6ed6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4c1d3cbfefad4f5aa6b9f08f0422957c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_d1db86f2576443f5835f133056568b39",
      "max": 43,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_549e4961bb7c47c8a148e45da7b52f7e",
      "value": 43
     }
    },
    "4ce78d87759b4924af8a1075dbf6ddce": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "4d555cff165a40fb8cad76acf7c2712d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_1a0827041a0b46f3bc050631b29bc066",
       "IPY_MODEL_1ed740b281d341ee867d782f53097ed3",
       "IPY_MODEL_1dad3b7b797e4da994566242893f5d4e"
      ],
      "layout": "IPY_MODEL_aabdc3a15dc54497a14a7132882082d8"
     }
    },
    "4f8b9fb466694e1aacf8cf8da4c56618": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "4ff810d3f45947579c552dbae9c45172": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "51c0415940ac4b63824bf337117f12b2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_1956ebc39de944c9ab28e11c642d7407",
      "max": 9,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_99a56eb9c2fe44749b622021cb8d8452",
      "value": 9
     }
    },
    "529e93880868406ba7f8b1afc2c7c278": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5349d3555ff942b0ac7f169462262428": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_2ba7118900894716b3bf13ade4fdf8df",
       "IPY_MODEL_51c0415940ac4b63824bf337117f12b2",
       "IPY_MODEL_f9eae2e334e943038c03218e25875699"
      ],
      "layout": "IPY_MODEL_84323279be93459a9922ec18e82ab181"
     }
    },
    "53b2a5f70c554476bde0e2514a25218c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "549e4961bb7c47c8a148e45da7b52f7e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "550aeb314ffb42ee98b2ca2fd44c7bf0": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_6679acc33bd2411e9dc261f1ef2f03e1",
       "IPY_MODEL_da47e1bf65a345b7b7a6d2bcdccf4dc2",
       "IPY_MODEL_1c37f78ea68840aaa9c1fe77541ae81e"
      ],
      "layout": "IPY_MODEL_9d3466be165d4323984c45fda1003ba2"
     }
    },
    "55e2854160e04bf292df38bbee17ffdd": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "564a657b517a4dcbb0b24515bfea4e51": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "582420861a1e451ca96a6dc69846c257": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_99788ffbf8d74daf8930873d8405c08a",
       "IPY_MODEL_4c1d3cbfefad4f5aa6b9f08f0422957c",
       "IPY_MODEL_fcdcbbf40e3e44c9ad96986d44ab6fce"
      ],
      "layout": "IPY_MODEL_611c5f880a834188a7b272c03dd3f991"
     }
    },
    "58290b9bd8824604aac3bc5e9d7c7f20": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "5b5fc2c030584237b3b3eabf0b76d9ef": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "5bb74dd29f304ff89bff4899d19b5b7a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "5e4335350b7d49f2b14e3db86c063a83": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_465a1be16a3948639b030652d3707de5",
      "placeholder": "​",
      "style": "IPY_MODEL_d5fa8b84bccb41daaa5849580fac1a3d",
      "value": "special_tokens_map.json: 100%"
     }
    },
    "5f4e3650bb474f2595d073277db33837": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "611c5f880a834188a7b272c03dd3f991": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "616f2b95148142b3a09aaba104876760": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_b854513cc82644f8a030ab9aabdf5519",
      "placeholder": "​",
      "style": "IPY_MODEL_cee19513901a4118ae27d404016e19cd",
      "value": ".gitattributes: 100%"
     }
    },
    "6365cb72b0e740aba2bb1b933aedc844": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "6371a34a2a02480cbc60328fdab8cc5d": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "63bc6a0124504cb9a15daeaca71b001c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "6404ef0323d44235860341cae362f080": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_4ce78d87759b4924af8a1075dbf6ddce",
      "max": 706,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_5bb74dd29f304ff89bff4899d19b5b7a",
      "value": 706
     }
    },
    "64511930cae14bfeb384d0fbbcdc7548": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "64fb95e3aaa7409dbd55c1474b108fef": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_17acda449aa544218c3db7c03c7789c9",
       "IPY_MODEL_26cc261df479483b92af2650a2db0fef",
       "IPY_MODEL_14947bce04ee442996215a1cbc009cbe"
      ],
      "layout": "IPY_MODEL_72b32c83b30a49bba9eb99c612ccfe2b"
     }
    },
    "6679acc33bd2411e9dc261f1ef2f03e1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_cb75a19ae09d496c857a982e9654a0cb",
      "placeholder": "​",
      "style": "IPY_MODEL_daf4e8c673564fed9e03f7fddaa81ad3",
      "value": "Filter: 100%"
     }
    },
    "6cb0a2aed78d4a99b8daed682e801523": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e46b524d949843b78f580db4195b914a",
      "placeholder": "​",
      "style": "IPY_MODEL_05161fd8e6064845a28e1b44d6e89a73",
      "value": "tokenizer_config.json: 100%"
     }
    },
    "6cbc14541229488ab452cd547778c88c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_888a4a15cae1407fb5912c4aa9823e33",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_00ad85e3dae347e09d272d3537b881fa",
      "value": 1
     }
    },
    "6cd56494d8ff44709188731fd3d71b4d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "6dc07cc1c9c6463a8146c798772269af": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "6ee161b4f1aa4112a207c4ee2492668f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "727bff98c3a944e98d475e1dd9ae56b9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "72b32c83b30a49bba9eb99c612ccfe2b": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "757d8e61bfb242719502565bacbd3a9b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_98f8dbd20d1e4e5ea1f78f72542650a5",
      "placeholder": "​",
      "style": "IPY_MODEL_98913edde9554127a1effb444dc3605d",
      "value": " 232k/232k [00:00&lt;00:00, 1.41MB/s]"
     }
    },
    "75adbf20c7304c0c87a37490b8077868": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "75b76c5043724a98b18a0b26904b36e6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "75cea6987d0a461ead31d2018a209ac0": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_7a3fe5b2865e414d958f18ff472ce0ae",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_d96721297828477faa852736039c0cf9",
      "value": 1
     }
    },
    "75e31b30ef79414ab96c0ec406f1f4b9": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "75f6cb50d6dc4c5194b844333fcd3e91": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_ec665d7d3d17417581ab400ba1f94196",
      "max": 1272,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_c398d6da2b584ec4a979f71c5af66fd9",
      "value": 1272
     }
    },
    "7745b86a86164b4782ca5f2b0ba349d2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "77aab4a259754f709ddbd2ceadacce4d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "78055f624a0740d9b00dd15cef3dd038": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "78577744100e484fb8ee6c02473ed0b2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "79618b906a264ec0b5b2238a448d818e": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7a3fe5b2865e414d958f18ff472ce0ae": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "7acdba6d2c1645239c1aae7267020be5": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "7b069b3aa893463e8fc8ebd4858d7056": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_21352eeac90a4bb5b326ad0260c089a3",
      "placeholder": "​",
      "style": "IPY_MODEL_a5251a2de0e947239eea948b04e01d6e",
      "value": " 405/405 [04:03&lt;00:00,  1.64it/s]"
     }
    },
    "7bd6c2a1e3534f88aa92443875e18f43": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_16c72bd3bc614955ae48760ce44e67e3",
       "IPY_MODEL_b0b5c0f5bed0478e9a82df8a6ea5d201",
       "IPY_MODEL_ac83ceeb91bd492490de27f7a5810501"
      ],
      "layout": "IPY_MODEL_6371a34a2a02480cbc60328fdab8cc5d"
     }
    },
    "7c48f92a57064802b2273072d5261605": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "7c70d712ce514dbdaeaaeb7ae75da1c2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_1e5a2b140acf441482218a1dcde84706",
      "placeholder": "​",
      "style": "IPY_MODEL_264c067001aa407cbbc6fee3e36428fc",
      "value": "Creating CSV from Arrow format: 100%"
     }
    },
    "7cedf7369e9f4927893cddbeffd1b145": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e6ababdab8fc48679bd2f8163ef61724",
      "placeholder": "​",
      "style": "IPY_MODEL_4beae8d8f5cc48a2bb7e9b7bebfb8e6f",
      "value": " 1.78M/1.78M [00:00&lt;00:00, 3.21MB/s]"
     }
    },
    "7ec86c1af6ad4972a7c546d2e184bab2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_a2208de652394c83bd4dfa9bbc65d65b",
       "IPY_MODEL_2f2bb2c795584fbb957338a7fc0ca201",
       "IPY_MODEL_e3cebc635c054448a036f98068b17069"
      ],
      "layout": "IPY_MODEL_19438ce279ae4d109f2985e648a92c29"
     }
    },
    "7fbbcb02feef491fb2f61b252c296758": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "805f4b56fe77487581c9c605a4411b4c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "80b1dff28f5f4b30827a9c8189fefc30": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_d4f0f13ed6104f51baf7ed6293638662",
      "placeholder": "​",
      "style": "IPY_MODEL_28d59a9ce5884e498c4b7763c79f1a65",
      "value": " 28.0/28.0 [00:00&lt;00:00, 838B/s]"
     }
    },
    "8119b99d4ac247bb8a759ad1a8a0e66c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "81730b126e424a7c97ab157af8ffcd3a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "818b70c8a7b2429a90c135c37c25a241": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "81d0e002e1a64549a4f054f543163f41": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "84323279be93459a9922ec18e82ab181": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "85938c37605a4f63afbb15c0d5a2d0c4": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_63bc6a0124504cb9a15daeaca71b001c",
      "max": 231508,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_02281ac9625e4e4b991f888a70680cea",
      "value": 231508
     }
    },
    "85a871757b9b45eda68a11812e39e534": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "8675a654291647728c40423dd8b99e4b": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "888a4a15cae1407fb5912c4aa9823e33": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "8ec2eae08ad643c3b03609c66554ed4c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_16b940f370074141b23b3f8b1f38dbe2",
      "placeholder": "​",
      "style": "IPY_MODEL_e22891a28b8747ef9c860ed54c3a73b8",
      "value": " 43.0/43.0 [00:00&lt;00:00, 1.45kB/s]"
     }
    },
    "8f12bf7b629347f29c6c5dd18299a9f7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "9014fbd6fdd5459cb4a2a2ccfef0410c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "917985f2bae24070b00147c660ad0e61": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_42a0528e6dbe492093228804f946f4cd",
      "max": 1242,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_bf1026966eaf442ea80d6ccd3224e924",
      "value": 1242
     }
    },
    "94d8dd42b15c4bd79a4c3260347a9844": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_5f4e3650bb474f2595d073277db33837",
      "max": 405,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_ae80fcc82fbe45cb9c924abc696dc1d7",
      "value": 405
     }
    },
    "951a3bec46164bcf8e70f17ae17608bd": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "95fb56f7042a47f8afa55dcb31bdd244": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "9657423bd9074d51b484d46c8d18c42b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_be3e0cbefef944498cf34d22aaedaa9d",
      "placeholder": "​",
      "style": "IPY_MODEL_070cc007ec7c4c4483996f61b2d6f7ce",
      "value": "vocab.txt: 100%"
     }
    },
    "97b1448064c64d829544c8d4cf4522de": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "98913edde9554127a1effb444dc3605d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "98f8dbd20d1e4e5ea1f78f72542650a5": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "99788ffbf8d74daf8930873d8405c08a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_b264a13b05a947cd8d8aad15a313f5db",
      "placeholder": "​",
      "style": "IPY_MODEL_bf73dde5eb1d40f5a607128605932bd1",
      "value": "Downloading readme: 100%"
     }
    },
    "999f768d715140999ae3d2bc35c23c2d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_5e4335350b7d49f2b14e3db86c063a83",
       "IPY_MODEL_cc3bb7178c5641c697890e138cf25c34",
       "IPY_MODEL_0ad979ea1b7347bdbf666fc457d8359c"
      ],
      "layout": "IPY_MODEL_48192520dcc44b41ae26f0f4b418ea20"
     }
    },
    "99a56eb9c2fe44749b622021cb8d8452": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "9a24f140106141d6b7c7a85c5a2bcbdd": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "9d3466be165d4323984c45fda1003ba2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "9e468980647c4c6198e8dcb7d51e33b1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_129a7656b1854c51970c275d7431d2b8",
      "placeholder": "​",
      "style": "IPY_MODEL_c8328854cc4149059d53df683ebf0b09",
      "value": " 1/1 [00:00&lt;00:00, 14.14ba/s]"
     }
    },
    "a1a0b3759a4d49adb3ee8e825a433525": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_e6e27abc259846d2b67a1301756f6a96",
      "placeholder": "​",
      "style": "IPY_MODEL_33f03cd63ab845fabceb1fd62c2c170f",
      "value": "config.json: 100%"
     }
    },
    "a20883beb87c499db4700d935d56957b": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "a2208de652394c83bd4dfa9bbc65d65b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_09390a4fa70748be98f0eb3ffc73aca9",
      "placeholder": "​",
      "style": "IPY_MODEL_da51b9894bd549faaee8fb5b6f44c82b",
      "value": "README.md: 100%"
     }
    },
    "a2278e779f0f4c0783d0c1602d3dbb4f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "a3353c3b1828438591573c03a106d466": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_a20883beb87c499db4700d935d56957b",
      "placeholder": "​",
      "style": "IPY_MODEL_77aab4a259754f709ddbd2ceadacce4d",
      "value": " 125k/125k [00:00&lt;00:00, 534kB/s]"
     }
    },
    "a3e3bfcf40ae464c929bff322b14d45f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "a5251a2de0e947239eea948b04e01d6e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "a634801d0aad4e21b2bfa8f92624c087": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_26a051f4728a4c8c88afc75097bd648a",
       "IPY_MODEL_a7c3dd5bae484d959e7955746e121800",
       "IPY_MODEL_e747142671b64db4a4d5f6f307ebb801"
      ],
      "layout": "IPY_MODEL_acc1bb1a2bfb4a6eb953f1b7535ac8bf"
     }
    },
    "a7c3dd5bae484d959e7955746e121800": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_6365cb72b0e740aba2bb1b933aedc844",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_7c48f92a57064802b2273072d5261605",
      "value": 1
     }
    },
    "a8375dc193b8470dad9c73cffb87731c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "a9fcba2e3a03432893b231f374a91a17": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_b0a6d22905154691bbddfc3c9fbffeb9",
       "IPY_MODEL_75f6cb50d6dc4c5194b844333fcd3e91",
       "IPY_MODEL_2b9df743e9e743d39e4b81d6d9d3e101"
      ],
      "layout": "IPY_MODEL_b0e1d711ce974476802d52c31b65bf81"
     }
    },
    "aabdc3a15dc54497a14a7132882082d8": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ac83ceeb91bd492490de27f7a5810501": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_115ad0595f8d4ac7852a5983f5f1b2b5",
      "placeholder": "​",
      "style": "IPY_MODEL_564a657b517a4dcbb0b24515bfea4e51",
      "value": " 81/0 [00:00&lt;00:00, 1619.72 examples/s]"
     }
    },
    "acc1bb1a2bfb4a6eb953f1b7535ac8bf": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ad70be7e9f16404985c2f13f60b7447f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "adb4ce9c86b94e91a8a7c8c7e09c7079": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ae80fcc82fbe45cb9c924abc696dc1d7": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "af4f51dee2244015a2fa3b3eb4de13c6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b0a6d22905154691bbddfc3c9fbffeb9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_af4f51dee2244015a2fa3b3eb4de13c6",
      "placeholder": "​",
      "style": "IPY_MODEL_494d8de002504275a5e079c832b74880",
      "value": "ort_config.json: 100%"
     }
    },
    "b0b5c0f5bed0478e9a82df8a6ea5d201": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_d015e097e6094b6891e15fdb04d34d25",
      "max": 1,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_b6e7dd16ab144afeb9f3b8ba49d4aea4",
      "value": 1
     }
    },
    "b0e1d711ce974476802d52c31b65bf81": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b13dc56442bc437f97bd9f1165337b38": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_3ef373eb9dae44f89a794c4cd8584ad4",
      "placeholder": "​",
      "style": "IPY_MODEL_78577744100e484fb8ee6c02473ed0b2",
      "value": " 21/21 [00:00&lt;00:00, 431.94 examples/s]"
     }
    },
    "b264a13b05a947cd8d8aad15a313f5db": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b322ee25201747b2aac1ef9cb4f51818": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_79618b906a264ec0b5b2238a448d818e",
      "placeholder": "​",
      "style": "IPY_MODEL_6cd56494d8ff44709188731fd3d71b4d",
      "value": "Generating train split: "
     }
    },
    "b33229c70a5843dea4d11d863614b973": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_616f2b95148142b3a09aaba104876760",
       "IPY_MODEL_244321954b7d4d52bb1a3e7f29c7269a",
       "IPY_MODEL_121d7781e13448e8b0f4ac3c810f2ae4"
      ],
      "layout": "IPY_MODEL_eb54968170da46cca97be9fd4dff583d"
     }
    },
    "b34c10d5c5214553a7b52e60ff78aabf": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_34a594530ea04e2d927967699e5dc7cc",
       "IPY_MODEL_30f018e451b54b169a31a967662bde72",
       "IPY_MODEL_cef90c22af034904b35183a2f779bebd"
      ],
      "layout": "IPY_MODEL_53b2a5f70c554476bde0e2514a25218c"
     }
    },
    "b3ae8239ded842349db8a7b29c2a4834": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "b408f04b92864babac1e09d7610a5474": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_db697a98ea974914ad510ccec6c4132a",
       "IPY_MODEL_d3c7b570c4384eaaaf598026dc335b48",
       "IPY_MODEL_35ac5a07c29f4597bed1f55dd1202778"
      ],
      "layout": "IPY_MODEL_95fb56f7042a47f8afa55dcb31bdd244"
     }
    },
    "b457ff179f38448fa55a384e8c395de1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_0f0ecde0157841e0a2bd661f59fea733",
      "max": 43,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_ad70be7e9f16404985c2f13f60b7447f",
      "value": 43
     }
    },
    "b6e7dd16ab144afeb9f3b8ba49d4aea4": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "b77e8cf9674a442db0b08c64d3ff2c83": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "b854513cc82644f8a030ab9aabdf5519": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "bb7a42f693b0461994ccb49c9720f8dd": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "bb7e1b80459446bca5d84d7b774ebac1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_a1a0b3759a4d49adb3ee8e825a433525",
       "IPY_MODEL_6404ef0323d44235860341cae362f080",
       "IPY_MODEL_e21a783b09294ae29d2173f2663c9e7f"
      ],
      "layout": "IPY_MODEL_8675a654291647728c40423dd8b99e4b"
     }
    },
    "bd64efaf240e42eb9b84dec9987a3c4f": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "be3e0cbefef944498cf34d22aaedaa9d": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "be4d621a67d941d587dcf2967f10df6a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "bf1026966eaf442ea80d6ccd3224e924": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "bf73dde5eb1d40f5a607128605932bd1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c398d6da2b584ec4a979f71c5af66fd9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "c450110d6ffc4ecd9b4c4edb547f9c07": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_6cb0a2aed78d4a99b8daed682e801523",
       "IPY_MODEL_917985f2bae24070b00147c660ad0e61",
       "IPY_MODEL_01119d8277364563a441ef40db68acf6"
      ],
      "layout": "IPY_MODEL_d4a27e4f5f9f4f27a674f6f3d300d7cb"
     }
    },
    "c8328854cc4149059d53df683ebf0b09": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "c84d1d76f4b94e3488a7de6d09e825ef": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "cb75a19ae09d496c857a982e9654a0cb": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "cb8287f82d0043c49ba3da16b4f387c5": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "cc3bb7178c5641c697890e138cf25c34": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_f5fc6313d683461daacf952f8bf34392",
      "max": 695,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_951a3bec46164bcf8e70f17ae17608bd",
      "value": 695
     }
    },
    "cc6cc585fdf94f92bca17d83bcc63de4": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ccc3ce4362ea4cbbaa606fe28b74cadd": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "cce81b158f2345619e2fec221770f9a2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "cee19513901a4118ae27d404016e19cd": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "cef90c22af034904b35183a2f779bebd": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_55e2854160e04bf292df38bbee17ffdd",
      "placeholder": "​",
      "style": "IPY_MODEL_6ee161b4f1aa4112a207c4ee2492668f",
      "value": " 711k/711k [00:00&lt;00:00, 1.71MB/s]"
     }
    },
    "cf98e5b8b48a44209597fe59072e4b7d": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "d015e097e6094b6891e15fdb04d34d25": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": "20px"
     }
    },
    "d1db86f2576443f5835f133056568b39": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "d2c3d4000c594a429ea12b7bcca280d1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_75e31b30ef79414ab96c0ec406f1f4b9",
      "max": 1777260,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_039d179c5821401c9d0b5caa056d805c",
      "value": 1777260
     }
    },
    "d3c7b570c4384eaaaf598026dc335b48": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_07d4537cd49e469e9d45ec51d859e3a8",
      "max": 66465124,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_e911aae60d4941829562e205e9a99530",
      "value": 66465124
     }
    },
    "d4a27e4f5f9f4f27a674f6f3d300d7cb": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "d4f0f13ed6104f51baf7ed6293638662": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "d57d4563682645da8ac4c2551157fc05": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_02f1ae27b6a2480280c4e6e7edda92bc",
       "IPY_MODEL_94d8dd42b15c4bd79a4c3260347a9844",
       "IPY_MODEL_7b069b3aa893463e8fc8ebd4858d7056"
      ],
      "layout": "IPY_MODEL_be4d621a67d941d587dcf2967f10df6a"
     }
    },
    "d5fa8b84bccb41daaa5849580fac1a3d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "d736f19ec8e34bb88be9f2e72455686e": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "d96721297828477faa852736039c0cf9": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "da47e1bf65a345b7b7a6d2bcdccf4dc2": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_81730b126e424a7c97ab157af8ffcd3a",
      "max": 81,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_8119b99d4ac247bb8a759ad1a8a0e66c",
      "value": 81
     }
    },
    "da51b9894bd549faaee8fb5b6f44c82b": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "daf4e8c673564fed9e03f7fddaa81ad3": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "db697a98ea974914ad510ccec6c4132a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_3301faa89b814001aaff9d492990a1bc",
      "placeholder": "​",
      "style": "IPY_MODEL_727bff98c3a944e98d475e1dd9ae56b9",
      "value": "model_optimized.onnx: 100%"
     }
    },
    "dc04d32cfc724cb6ac3dd926041badce": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "dd2274085a1d415c9ab4dbf90813af2c": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "dee469d9cfd44d9181c313da9bc195aa": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e21a783b09294ae29d2173f2663c9e7f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_adb4ce9c86b94e91a8a7c8c7e09c7079",
      "placeholder": "​",
      "style": "IPY_MODEL_7acdba6d2c1645239c1aae7267020be5",
      "value": " 706/706 [00:00&lt;00:00, 3.42kB/s]"
     }
    },
    "e22891a28b8747ef9c860ed54c3a73b8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "e3448574566a44a7af6eaf656aaf04a2": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e3cebc635c054448a036f98068b17069": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_81d0e002e1a64549a4f054f543163f41",
      "placeholder": "​",
      "style": "IPY_MODEL_c84d1d76f4b94e3488a7de6d09e825ef",
      "value": " 28.0/28.0 [00:00&lt;00:00, 137B/s]"
     }
    },
    "e3d7cae3f6244ad9990cd984bc2e4855": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "e46b524d949843b78f580db4195b914a": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e6ababdab8fc48679bd2f8163ef61724": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e6e27abc259846d2b67a1301756f6a96": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e747142671b64db4a4d5f6f307ebb801": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_156da40cbd834a518cddb08cbef3198d",
      "placeholder": "​",
      "style": "IPY_MODEL_2805c5fbcfab4313b321f7ab125d9f2b",
      "value": " 81/0 [00:00&lt;00:00, 1591.85 examples/s]"
     }
    },
    "e911aae60d4941829562e205e9a99530": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "e94a8581d14842bda30438961ad294e1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "eb54968170da46cca97be9fd4dff583d": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "eb899fde8d554b65b15f08a054307e19": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "ec665d7d3d17417581ab400ba1f94196": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "ed4d43fadb1f48319ea790b56cb4d343": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_3b9465a8398c45f498a414ac46c80355",
       "IPY_MODEL_b457ff179f38448fa55a384e8c395de1",
       "IPY_MODEL_8ec2eae08ad643c3b03609c66554ed4c"
      ],
      "layout": "IPY_MODEL_a2278e779f0f4c0783d0c1602d3dbb4f"
     }
    },
    "edc22a2adeb34550b24791cb89762383": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_7fbbcb02feef491fb2f61b252c296758",
      "placeholder": "​",
      "style": "IPY_MODEL_4f8b9fb466694e1aacf8cf8da4c56618",
      "value": " 105/105 [00:56&lt;00:00,  1.57it/s]"
     }
    },
    "ee0f3a23c1924342b1501facdba8902f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "f226de79c38b4ed4bae675ec213f18be": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "f57fb5f982594ae98c3bb71ce414e760": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "f5fc6313d683461daacf952f8bf34392": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f642894d3c574bdf8e19fbe42cf027d5": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "f6fd6fcdc4f349fba497027d7fcf8a9f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_805f4b56fe77487581c9c605a4411b4c",
      "max": 21,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_00963b69a2524a8a91f42e46a6ec8d0b",
      "value": 21
     }
    },
    "f7169503c1b049149bab032496d378b6": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f818582473954544bfd497fe7f19a640": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f93019d15bf64bac93bad2c3c3c53714": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "f9eae2e334e943038c03218e25875699": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_b3ae8239ded842349db8a7b29c2a4834",
      "placeholder": "​",
      "style": "IPY_MODEL_eb899fde8d554b65b15f08a054307e19",
      "value": " 9/9 [00:02&lt;00:00,  1.90it/s]"
     }
    },
    "fa1cfd25cadd4a42ab990a336eea095c": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "fbf413ef0b274ccea2130db147f06e25": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_25db4e94fc61492a9752600158fae7b4",
       "IPY_MODEL_30fcf2693b154033881fee05b0d17db6",
       "IPY_MODEL_a3353c3b1828438591573c03a106d466"
      ],
      "layout": "IPY_MODEL_3124ef41463846a999c1fa01f1530b1a"
     }
    },
    "fcdcbbf40e3e44c9ad96986d44ab6fce": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_3071076a711c4db19459b6142fcfc8d0",
      "placeholder": "​",
      "style": "IPY_MODEL_ee0f3a23c1924342b1501facdba8902f",
      "value": " 43.0/43.0 [00:00&lt;00:00, 2.01kB/s]"
     }
    },
    "fdd8cf7194de46d2b2c7440077e673f8": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_3c0415f3a6d944ebbddaa4fc04a22f22",
       "IPY_MODEL_1faed09d17f74721a5462f42ad61caf1",
       "IPY_MODEL_edc22a2adeb34550b24791cb89762383"
      ],
      "layout": "IPY_MODEL_85a871757b9b45eda68a11812e39e534"
     }
    },
    "fe20966df20042ef8dc66c133f261079": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "fe30399d1aeb44d3a6d69091b606b80f": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_fe594f22be574d3b8c2ca187ac9c7bfb",
      "placeholder": "​",
      "style": "IPY_MODEL_1eec371016434a6b8c4ac8b83bd77b4f",
      "value": "Map: 100%"
     }
    },
    "fe594f22be574d3b8c2ca187ac9c7bfb": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}