1. Know Your Data.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "6HIwlbmDJjc7"
   },
   "source": [
    "Download the questions and Annotation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 782
    },
    "colab_type": "code",
    "id": "M_qmlxmKzjV4",
    "outputId": "e0275e79-7825-43d8-f7e6-cd0876b52786"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/content\n",
      "--2019-06-15 12:11:35--  https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/Questions_Val_mscoco.zip\n",
      "Resolving s3.amazonaws.com (s3.amazonaws.com)... 52.216.236.237\n",
      "Connecting to s3.amazonaws.com (s3.amazonaws.com)|52.216.236.237|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 10594497 (10M) [application/zip]\n",
      "Saving to: ‘Questions_Val_mscoco.zip.1’\n",
      "\n",
      "Questions_Val_mscoc 100%[===================>]  10.10M  1.92MB/s    in 9.8s    \n",
      "\n",
      "2019-06-15 12:11:46 (1.03 MB/s) - ‘Questions_Val_mscoco.zip.1’ saved [10594497/10594497]\n",
      "\n",
      "--2019-06-15 12:11:48--  https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/Annotations_Val_mscoco.zip\n",
      "Resolving s3.amazonaws.com (s3.amazonaws.com)... 52.216.108.213\n",
      "Connecting to s3.amazonaws.com (s3.amazonaws.com)|52.216.108.213|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 6031604 (5.8M) [application/zip]\n",
      "Saving to: ‘Annotations_Val_mscoco.zip.1’\n",
      "\n",
      "Annotations_Val_msc 100%[===================>]   5.75M  1.42MB/s    in 6.8s    \n",
      "\n",
      "2019-06-15 12:11:56 (872 KB/s) - ‘Annotations_Val_mscoco.zip.1’ saved [6031604/6031604]\n",
      "\n",
      "--2019-06-15 12:11:58--  https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/Questions_Train_mscoco.zip\n",
      "Resolving s3.amazonaws.com (s3.amazonaws.com)... 52.216.137.134\n",
      "Connecting to s3.amazonaws.com (s3.amazonaws.com)|52.216.137.134|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 21985607 (21M) [application/zip]\n",
      "Saving to: ‘Questions_Train_mscoco.zip.1’\n",
      "\n",
      "Questions_Train_msc 100%[===================>]  20.97M  10.7MB/s    in 2.0s    \n",
      "\n",
      "2019-06-15 12:12:01 (10.7 MB/s) - ‘Questions_Train_mscoco.zip.1’ saved [21985607/21985607]\n",
      "\n",
      "--2019-06-15 12:12:03--  https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/Annotations_Train_mscoco.zip\n",
      "Resolving s3.amazonaws.com (s3.amazonaws.com)... 54.231.81.83\n",
      "Connecting to s3.amazonaws.com (s3.amazonaws.com)|54.231.81.83|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 12167843 (12M) [application/zip]\n",
      "Saving to: ‘Annotations_Train_mscoco.zip’\n",
      "\n",
      "Annotations_Train_m 100%[===================>]  11.60M  6.77MB/s    in 1.7s    \n",
      "\n",
      "2019-06-15 12:12:06 (6.77 MB/s) - ‘Annotations_Train_mscoco.zip’ saved [12167843/12167843]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%cd /content\n",
    "!wget https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/Questions_Val_mscoco.zip\n",
    "!wget https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/Annotations_Val_mscoco.zip\n",
    "!wget https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/Questions_Train_mscoco.zip\n",
    "!wget https://s3.amazonaws.com/cvmlp/vqa/mscoco/vqa/Annotations_Train_mscoco.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 187
    },
    "colab_type": "code",
    "id": "-8jACScI0GZZ",
    "outputId": "0636c4db-93bc-4130-fab5-0f5722a93b18"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Archive:  Annotations_Val_mscoco.zip\n",
      "  inflating: mscoco_val2014_annotations.json  \n",
      "Archive:  Questions_Val_mscoco.zip\n",
      "  inflating: OpenEnded_mscoco_val2014_questions.json  \n",
      "  inflating: MultipleChoice_mscoco_val2014_questions.json  \n",
      "Archive:  Annotations_Train_mscoco.zip\n",
      "  inflating: mscoco_train2014_annotations.json  \n",
      "Archive:  Questions_Train_mscoco.zip\n",
      "  inflating: OpenEnded_mscoco_train2014_questions.json  \n",
      "  inflating: MultipleChoice_mscoco_train2014_questions.json  \n"
     ]
    }
   ],
   "source": [
    "# Unzip questions and answers of training and validation data!\n",
    "!unzip Annotations_Val_mscoco.zip\n",
    "!unzip Questions_Val_mscoco.zip\n",
    "\n",
    "!unzip Annotations_Train_mscoco.zip\n",
    "!unzip Questions_Train_mscoco.zip"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "KexDqOWABgo1"
   },
   "source": [
    "## Get To know Your Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "7oBubrrc71Fz"
   },
   "outputs": [],
   "source": [
    "# Load json question file.\n",
    "\n",
    "import json\n",
    "from pprint import pprint\n",
    "\n",
    "with open('OpenEnded_mscoco_val2014_questions.json') as file:\n",
    "    open_question_json_val = json.load(file)\n",
    "    \n",
    "with open('OpenEnded_mscoco_train2014_questions.json') as file:\n",
    "    open_question_json_train = json.load(file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "open_question_json_train\n",
    "# pprint(open_question_json_train)\n",
    "'''\n",
    "The output looks like this:\n",
    "{'data_subtype': 'train2014',\n",
    " 'data_type': 'mscoco',\n",
    " 'info': {'contributor': 'VQA Team',\n",
    "  'date_created': '2015-10-02 19:28:08',\n",
    "  'description': 'This is v1.0 of the VQA dataset.',\n",
    "  'url': 'http://visualqa.org',\n",
    "  'version': '1.0',\n",
    "  'year': 2015},\n",
    " 'license': {'name': 'Creative Commons Attribution 4.0 International License',\n",
    "  'url': 'http://creativecommons.org/licenses/by/4.0/'},\n",
    " 'questions': [{'image_id': 487025,\n",
    "   'question': 'What shape is the bench seat?',\n",
    "   'question_id': 4870250},\n",
    "  {'image_id': 487025,\n",
    "   'question': 'Is there a shadow?',\n",
    "   'question_id': 4870251},\n",
    "  {'image_id': 487025,\n",
    "   'question': 'Is this one bench or multiple benches?',\n",
    "   'question_id': 4870252},\n",
    "  {'image_id': 78077,\n",
    "   'question': 'Is this a modern train?',\n",
    "   'question_id': 780770},\n",
    "  {'image_id': 78077,\n",
    "   'question': 'What color is the stripe on the train?',\n",
    "   'question_id': 780771},\n",
    "  {'image_id': 78077,\n",
    "   'question': 'What is on the other side of the train?',\n",
    "   'question_id': 780772},\n",
    "  {'image_id': 501867,\n",
    "   'question': 'Is the bus driver on any kind of antidepressant medication?',\n",
    "   'question_id': 5018672},\n",
    "  {'image_id': 501867,\n",
    "   'question': 'Is the bus moving?',\n",
    "   'question_id': 5018670},\n",
    "  {'image_id': 501867,\n",
    "   'question': 'What color is the bus?',\n",
    "   'question_id': 5018671},\n",
    "   ...\n",
    "   {'image_id': 246636,\n",
    "   'question': \"What color is the closest women's shirt?\",\n",
    "   'question_id': 2466361},\n",
    "  {'image_id': 246636,\n",
    "   'question': 'Are these canoes overcrowded?',\n",
    "   'question_id': 2466362},\n",
    "  {'image_id': 405541,\n",
    "   'question': 'What breed of cat is in the photo?',\n",
    "   'question_id': 4055410},\n",
    "  ...],\n",
    " 'task_type': 'Open-Ended'}\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "jtXIyxE3-PPj"
   },
   "outputs": [],
   "source": [
    "# Load json answers file\n",
    "\n",
    "with open('mscoco_val2014_annotations.json') as file:\n",
    "    annotation_json_val = json.load(file)\n",
    "    \n",
    "with open('mscoco_train2014_annotations.json') as file:\n",
    "    annotation_json_train = json.load(file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "annotation_json_train\n",
    "#pprint(annotation_json_train)\n",
    "'''\n",
    "The output looks like this:\n",
    "{'annotations': [{'answer_type': 'other',\n",
    "   'answers': [{'answer': 'oval', 'answer_confidence': 'yes', 'answer_id': 1},\n",
    "    {'answer': 'semi circle', 'answer_confidence': 'yes', 'answer_id': 2},\n",
    "    {'answer': 'curved', 'answer_confidence': 'yes', 'answer_id': 3},\n",
    "    {'answer': 'curved', 'answer_confidence': 'yes', 'answer_id': 4},\n",
    "    {'answer': 'double curve', 'answer_confidence': 'yes', 'answer_id': 5},\n",
    "    {'answer': 'banana', 'answer_confidence': 'maybe', 'answer_id': 6},\n",
    "    {'answer': 'curved', 'answer_confidence': 'yes', 'answer_id': 7},\n",
    "    {'answer': 'wavy', 'answer_confidence': 'yes', 'answer_id': 8},\n",
    "    {'answer': 'twisting', 'answer_confidence': 'no', 'answer_id': 9},\n",
    "    {'answer': 'curved', 'answer_confidence': 'maybe', 'answer_id': 10}],\n",
    "   'image_id': 487025,\n",
    "   'multiple_choice_answer': 'curved',\n",
    "   'question_id': 4870250,\n",
    "   'question_type': 'what'},\n",
    "  {'answer_type': 'yes/no',\n",
    "   'answers': [{'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 1},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 2},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 3},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 4},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 5},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 6},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 7},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 8},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 9},\n",
    "    {'answer': 'yes', 'answer_confidence': 'yes', 'answer_id': 10}],\n",
    "   'image_id': 487025,\n",
    "   'multiple_choice_answer': 'yes',\n",
    "   'question_id': 4870251,\n",
    "   'question_type': 'is there a'},\n",
    "   ...\n",
    "   {'answer_type': 'other',\n",
    "   'answers': [{'answer': 'siamese',\n",
    "     'answer_confidence': 'yes',\n",
    "     'answer_id': 1},\n",
    "    {'answer': 'siamese', 'answer_confidence': 'yes', 'answer_id': 2},\n",
    "    {'answer': 'siamese', 'answer_confidence': 'yes', 'answer_id': 3},\n",
    "    {'answer': 'siamese', 'answer_confidence': 'yes', 'answer_id': 4},\n",
    "    {'answer': 'not sure', 'answer_confidence': 'no', 'answer_id': 5},\n",
    "    {'answer': 'siamese', 'answer_confidence': 'no', 'answer_id': 6},\n",
    "    {'answer': 'siamese', 'answer_confidence': 'yes', 'answer_id': 7},\n",
    "    {'answer': 'persian', 'answer_confidence': 'yes', 'answer_id': 8},\n",
    "    {'answer': 'siamese', 'answer_confidence': 'maybe', 'answer_id': 9},\n",
    "    {'answer': 'tortoiseshell', 'answer_confidence': 'no', 'answer_id': 10}],\n",
    "   'image_id': 405541,\n",
    "   'multiple_choice_answer': 'siamese',\n",
    "   'question_id': 4055410,\n",
    "   'question_type': 'what'},\n",
    "  ...],\n",
    " 'data_subtype': 'train2014',\n",
    " 'data_type': 'mscoco',\n",
    " 'info': {'contributor': 'VQA Team',\n",
    "  'date_created': '2015-10-02 19:28:08',\n",
    "  'description': 'This is v1.0 of the VQA dataset.',\n",
    "  'url': 'http://visualqa.org',\n",
    "  'version': '1.0',\n",
    "  'year': 2015},\n",
    " 'license': {'name': 'Creative Commons Attribution 4.0 International License',\n",
    "  'url': 'http://creativecommons.org/licenses/by/4.0/'}}\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "6upDbW19_VTe",
    "outputId": "c56059c1-9c8a-4e3d-85b9-69142b8de3eb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Is this a modern train?'"
      ]
     },
     "execution_count": 6,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "open_question_json_train[\"questions\"][3][\"question\"] # fourth question"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Kbsj8c8OBo-N"
   },
   "source": [
    "## Process the Questions and Answers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "IAVFLdPwEXSV"
   },
   "outputs": [],
   "source": [
    "# Preprocess openended_question_json file\n",
    "keys_to_remove = [\"data_subtype\", \"data_type\", \"info\", \"license\", \"task_type\"]\n",
    "for key in keys_to_remove:\n",
    "    open_question_json_val.pop(key, None)\n",
    "    open_question_json_train.pop(key, None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "NQoTnv01H6AD"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "df_question_val = pd.DataFrame.from_dict(open_question_json_val[\"questions\"], orient = \"columns\")\n",
    "df_question_train = pd.DataFrame.from_dict(open_question_json_train[\"questions\"], orient = \"columns\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "colab_type": "code",
    "id": "Q4Fsl8gaIN-7",
    "outputId": "6495a476-7c7b-4100-890d-d9e9b5f7bafc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   image_id                                question  question_id\n",
      "0    350623              What is the table made of?      3506232\n",
      "1    350623       Is the food napping on the table?      3506230\n",
      "2    350623  What has been upcycled to make lights?      3506231\n",
      "3      8647                Is this an Spanish town?        86472\n",
      "4      8647      Are there shadows on the sidewalk?        86470\n",
      "   image_id                                question  question_id\n",
      "0    487025           What shape is the bench seat?      4870250\n",
      "1    487025                      Is there a shadow?      4870251\n",
      "2    487025  Is this one bench or multiple benches?      4870252\n",
      "3     78077                 Is this a modern train?       780770\n",
      "4     78077  What color is the stripe on the train?       780771\n"
     ]
    }
   ],
   "source": [
    "print(df_question_val.head(5) )\n",
    "print(df_question_train.head(5) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "csQ0yrt0IQ46",
    "outputId": "a0ad7efd-6628-423c-f539-fed8fd63aa79"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'curved'"
      ]
     },
     "execution_count": 10,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "annotation_json_train[\"annotations\"][0][\"multiple_choice_answer\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "GaYAcUAsJSZF"
   },
   "outputs": [],
   "source": [
    "#Remove keys from answers json file.\n",
    "keys_to_remove = [\"data_subtype\", \"data_type\", \"info\", \"license\"]\n",
    "for key in keys_to_remove:\n",
    "    annotation_json_val.pop(key, None)\n",
    "    annotation_json_train.pop(key, None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "PFSgCAhlJmUy"
   },
   "outputs": [],
   "source": [
    "df_annotation_val = pd.DataFrame.from_dict(annotation_json_val[\"annotations\"], orient = \"columns\")\n",
    "df_annotation_train = pd.DataFrame.from_dict(annotation_json_train[\"annotations\"], orient = \"columns\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 2193
    },
    "colab_type": "code",
    "id": "DY6BdKR_gANe",
    "outputId": "9bd61b21-cbc2-4703-cf74-ab4fd0d74785"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       answer_type  ...           question_type\n",
      "0            other  ...             what is the\n",
      "1           yes/no  ...                  is the\n",
      "2            other  ...                    what\n",
      "3           yes/no  ...              is this an\n",
      "4           yes/no  ...               are there\n",
      "5            other  ...          what is in the\n",
      "6           yes/no  ...                   is it\n",
      "7            other  ...                 what is\n",
      "8           number  ...                how many\n",
      "9           yes/no  ...                 is this\n",
      "10          yes/no  ...              is there a\n",
      "11          yes/no  ...           is the person\n",
      "12          yes/no  ...               is this a\n",
      "13           other  ...                    what\n",
      "14           other  ...       what color is the\n",
      "15           other  ...       what color is the\n",
      "16          yes/no  ...                  is the\n",
      "17           other  ...                     how\n",
      "18           other  ...             what is the\n",
      "19          yes/no  ...                      do\n",
      "20           other  ...                 what is\n",
      "21          yes/no  ...                  is the\n",
      "22          number  ...  how many people are in\n",
      "23           other  ...                     why\n",
      "24          number  ...                how many\n",
      "25          yes/no  ...                 are the\n",
      "26          number  ...                how many\n",
      "27           other  ...       none of the above\n",
      "28           other  ...                how many\n",
      "29          number  ...                how many\n",
      "...            ...  ...                     ...\n",
      "121482       other  ...                    what\n",
      "121483      yes/no  ...                 are the\n",
      "121484       other  ...                   which\n",
      "121485       other  ...           what sport is\n",
      "121486       other  ...                   which\n",
      "121487       other  ...                    what\n",
      "121488      number  ...                how many\n",
      "121489       other  ...      what color are the\n",
      "121490      number  ...                how many\n",
      "121491      yes/no  ...                does the\n",
      "121492       other  ...             what is the\n",
      "121493      yes/no  ...                 is this\n",
      "121494       other  ...       what color is the\n",
      "121495      yes/no  ...               is this a\n",
      "121496       other  ...       what color is the\n",
      "121497       other  ...            where is the\n",
      "121498      number  ...                how many\n",
      "121499      number  ...             what is the\n",
      "121500      number  ...                how many\n",
      "121501      number  ...                how many\n",
      "121502       other  ...             what is the\n",
      "121503       other  ...                 what is\n",
      "121504       other  ...                    what\n",
      "121505      yes/no  ...              is the man\n",
      "121506      yes/no  ...                     has\n",
      "121507       other  ...                 what is\n",
      "121508       other  ...            what kind of\n",
      "121509      number  ...               what time\n",
      "121510      yes/no  ...                does the\n",
      "121511      yes/no  ...                is there\n",
      "\n",
      "[121512 rows x 6 columns]\n",
      "       answer_type  ...        question_type\n",
      "0            other  ...                 what\n",
      "1           yes/no  ...           is there a\n",
      "2            other  ...              is this\n",
      "3           yes/no  ...            is this a\n",
      "4            other  ...    what color is the\n",
      "5            other  ...       what is on the\n",
      "6           yes/no  ...               is the\n",
      "7           yes/no  ...               is the\n",
      "8            other  ...    what color is the\n",
      "9           yes/no  ...            are these\n",
      "10           other  ...              what is\n",
      "11          yes/no  ...            is this a\n",
      "12          yes/no  ...               is the\n",
      "13           other  ...       what is in the\n",
      "14           other  ...         where is the\n",
      "15           other  ...   what color are the\n",
      "16           other  ...                 what\n",
      "17           other  ...                 what\n",
      "18           other  ...        where are the\n",
      "19           other  ...       what animal is\n",
      "20           other  ...    none of the above\n",
      "21           other  ...         what kind of\n",
      "22           other  ...          what is the\n",
      "23          yes/no  ...              is this\n",
      "24           other  ...                 what\n",
      "25          number  ...    none of the above\n",
      "26          yes/no  ...        are there any\n",
      "27           other  ...    what color is the\n",
      "28           other  ...   what color are the\n",
      "29           other  ...    what color is the\n",
      "...            ...  ...                  ...\n",
      "248319      yes/no  ...           is the man\n",
      "248320       other  ...      what is the man\n",
      "248321      number  ...  how many people are\n",
      "248322       other  ...    what color is the\n",
      "248323      yes/no  ...              are the\n",
      "248324       other  ...          what is the\n",
      "248325      yes/no  ...            is this a\n",
      "248326       other  ...         what kind of\n",
      "248327      yes/no  ...        are there any\n",
      "248328      yes/no  ...               is the\n",
      "248329       other  ...         what kind of\n",
      "248330       other  ...          what is the\n",
      "248331      number  ...                  how\n",
      "248332      yes/no  ...               is the\n",
      "248333      yes/no  ...             is there\n",
      "248334      yes/no  ...               is the\n",
      "248335      yes/no  ...               is the\n",
      "248336       other  ...    what color is the\n",
      "248337      number  ...             how many\n",
      "248338       other  ...             what are\n",
      "248339       other  ...          what is the\n",
      "248340       other  ...    what color is the\n",
      "248341       other  ...    what color is the\n",
      "248342       other  ...                 what\n",
      "248343      number  ...             how many\n",
      "248344       other  ...       what is on the\n",
      "248345      yes/no  ...                   is\n",
      "248346      yes/no  ...            is this a\n",
      "248347       other  ...                which\n",
      "248348       other  ...         where is the\n",
      "\n",
      "[248349 rows x 6 columns]\n"
     ]
    }
   ],
   "source": [
    "print(df_annotation_val)\n",
    "print(df_annotation_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "colab_type": "code",
    "id": "Wdkti4ILJwMx",
    "outputId": "cff62230-65df-49bb-c353-14f9ba47cd1f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  multiple_choice_answer  question_id\n",
      "0                   wood      3506232\n",
      "1                     no      3506230\n",
      "2                kettles      3506231\n",
      "3                     no        86472\n",
      "4                    yes        86470\n",
      "  multiple_choice_answer  question_id\n",
      "0                 curved      4870250\n",
      "1                    yes      4870251\n",
      "2                      1      4870252\n",
      "3                     no       780770\n",
      "4                  white       780771\n"
     ]
    }
   ],
   "source": [
    "df_annotation_val = df_annotation_val.drop(columns = [\"answer_type\", \"answers\", \"image_id\", \"question_type\"])\n",
    "df_annotation_train = df_annotation_train.drop(columns = [\"answer_type\", \"answers\", \"image_id\", \"question_type\"])\n",
    "print( df_annotation_val.head(5) )\n",
    "print( df_annotation_train.head(5) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "PPrDk2pHKmVs"
   },
   "outputs": [],
   "source": [
    "data_val_qa = pd.merge(df_question_val, df_annotation_val, how = \"outer\", on = \"question_id\")\n",
    "data_train_qa = pd.merge(df_question_train, df_annotation_train, how = \"outer\", on = \"question_id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "P_VEtd7iMrVb"
   },
   "outputs": [],
   "source": [
    "data_val_qa = data_val_qa.drop( columns = [\"question_id\"])\n",
    "data_train_qa = data_train_qa.drop( columns = [\"question_id\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "colab_type": "code",
    "id": "5cqw7z4HM3zC",
    "outputId": "2f949bf1-9215-417a-bffd-095481b6b1d3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   image_id                                question multiple_choice_answer\n",
      "0    350623              What is the table made of?                   wood\n",
      "1    350623       Is the food napping on the table?                     no\n",
      "2    350623  What has been upcycled to make lights?                kettles\n",
      "3      8647                Is this an Spanish town?                     no\n",
      "4      8647      Are there shadows on the sidewalk?                    yes\n",
      "   image_id                                question multiple_choice_answer\n",
      "0    487025           What shape is the bench seat?                 curved\n",
      "1    487025                      Is there a shadow?                    yes\n",
      "2    487025  Is this one bench or multiple benches?                      1\n",
      "3     78077                 Is this a modern train?                     no\n",
      "4     78077  What color is the stripe on the train?                  white\n"
     ]
    }
   ],
   "source": [
    "print( data_val_qa.head() )\n",
    "print(data_train_qa.head() )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 153
    },
    "colab_type": "code",
    "id": "Oi7OkLDQL-Go",
    "outputId": "6e28712e-3b5c-405e-c16b-ffe2f0c4c8b0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "image_id                  121512\n",
      "question                  121512\n",
      "multiple_choice_answer    121512\n",
      "dtype: int64\n",
      "image_id                  248349\n",
      "question                  248349\n",
      "multiple_choice_answer    248349\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print( data_val_qa.count() )\n",
    "print( data_train_qa.count() )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "0ANQr53nGr_S"
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "file = open(\"Training Data QA.pickle\", \"rb\")\n",
    "data_train_qa = pickle.load(file)\n",
    "file.close()\n",
    "\n",
    "file = open(\"Validation Data QA.pickle\", \"rb\")\n",
    "data_val_qa = pickle.load(file)\n",
    "file.close()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "MsXkXwM1F3-L"
   },
   "outputs": [],
   "source": [
    "# Before saving the file, lets pre-process the questions.\n",
    "# Find all the punctuations.\n",
    "import re\n",
    "\n",
    "punctuation = []\n",
    "length = int(len(data_train_qa))\n",
    "for i in range(length):\n",
    "    punctuation.append(re.findall(r\"[^a-zA-Z0-9 ]\", data_train_qa[\"question\"][i]))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "54kKJnY3IwjB",
    "outputId": "e5369a0e-211c-43e6-dd7c-798bb69f38f8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{',', \"'\", '?'}\n"
     ]
    }
   ],
   "source": [
    "# Get all unique punctuation from Top 200 questions\n",
    "myset = set()\n",
    "for i in range(200):\n",
    "    for j in range(len(punctuation[i])):\n",
    "        myset.add(punctuation[i][j])\n",
    "print(myset)\n",
    "  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "w2BYE58NJNYi"
   },
   "source": [
    "So, the punctuation we are going to foucs is on \" ' '\", while for the rest we are going to simply remove it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 187
    },
    "colab_type": "code",
    "id": "6_7sCfElILp5",
    "outputId": "16292763-b363-43df-82c2-c9bc5f2b3577"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[19, \"What animal is on the man's lap?\"],\n",
       " [20, \"What's in the man's right hand?\"],\n",
       " [33, \"What color is this person's shirt?\"],\n",
       " [35, \"What color is the man's pants?\"],\n",
       " [36, \"Why is the woman's head blurry?\"],\n",
       " [40, \"What color is the man's shirt?\"],\n",
       " [54, \"What is the theme of the kites' designs?\"],\n",
       " [59, \"Can you see the dog's teeth?\"],\n",
       " [65, \"What color is the man's shirt?\"],\n",
       " [78, \"What is on the man's head?\"]]"
      ]
     },
     "execution_count": 28,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "question_punc = []\n",
    "length = int(len(data_train_qa))\n",
    "for i in range(length):  \n",
    "    punctuation = re.findall(r\"[']\", data_train_qa[\"question\"][i])\n",
    "    if punctuation:\n",
    "        question_punc.append([i, data_train_qa[\"question\"][i]])\n",
    "    \n",
    "question_punc[0:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "9Zgnc1KASbZK"
   },
   "outputs": [],
   "source": [
    "# what's -> What is, we are going to use wikipedia data.\n",
    "import requests\n",
    "from bs4 import BeautifulSoup\n",
    "\n",
    "website_url = requests.get(\"https://en.wikipedia.org/wiki/Wikipedia:List_of_English_contractions\").text\n",
    "soup = BeautifulSoup(website_url, 'lxml')\n",
    "\n",
    "# we need \"wikitable sortable\" file from the html script.\n",
    "table = soup.find('table', {'class' : 'wikitable sortable'})\n",
    "\n",
    "#columns = soup.find_all('td')\n",
    "contraction_list = []\n",
    "meaning_list = []\n",
    "for row in table.find_all('tr')[1:]:  # iterate over the rows, starting from \n",
    "                                      # the second (first one is the header row)\n",
    "    contraction = row.find_all('td')[0]  #  the Symbol col is the first <td> in every row \n",
    "    meaning = row.find_all('td')[1]  #  the Symbol col is the first <td> in every row\n",
    "        \n",
    "    contraction_list.append(contraction.text)\n",
    "    meaning_list.append(meaning.text)\n",
    "\n",
    "# Preprocess the words\n",
    "\n",
    "#splitword = []\n",
    "#print(len(contraction_list))\n",
    "#print(len(meaning_list))\n",
    "for word in contraction_list:\n",
    "  #print(word)\n",
    "  if (' ' in word):\n",
    "    position_to_add = contraction_list.index(word)\n",
    "    meaning_word = meaning_list[position_to_add]\n",
    "    \n",
    "    word_to_process = re.sub(r'[(),]', \"\", word)\n",
    "    splitword = word_to_process.split()\n",
    "    length_split_word = len(splitword)\n",
    "    #print(splitword)\n",
    "    \n",
    "    contraction_list.remove(word)\n",
    "    meaning_list.pop(position_to_add)\n",
    "    \n",
    "    for i in range(len(splitword)):\n",
    "        contraction_list.insert((position_to_add + i), splitword[i] )\n",
    "        meaning_list.insert((position_to_add + i), meaning_word)\n",
    "    break\n",
    "    #print(splitword)\n",
    "\n",
    "# pop out last two element\n",
    "del contraction_list[-2:]\n",
    "del meaning_list[-2:]\n",
    "\n",
    "# Now, our contraction list is well defined, we need to process meaning list.\n",
    "\n",
    "meaningfull_list = []\n",
    "def process_word(word):\n",
    "  \n",
    "    word = re.sub(r\"[0-9]\", \"\", word)\n",
    "    word = re.sub(r\"\\n\", \"\", word)\n",
    "    word = word.replace('[','').replace(']','')\n",
    "    word = word.replace(' (colloquial)', '')\n",
    "    word = word.replace(' (archaic)', '')\n",
    "    meaningfull_list.append(word)\n",
    "  \n",
    "    return meaningfull_list\n",
    "#[re.sub(r'[0-9]', '', word) if ('[]' in word) else x for x in meaning_list]\n",
    "\n",
    "for word in meaning_list:\n",
    "    meaningfull_list = process_word(word)\n",
    "    \n",
    "# Now that we have process the words, lets split these too!\n",
    "\n",
    "for word in meaningfull_list:\n",
    "    if (' / ' in word or \", \" in word ):\n",
    "        position_to_split = meaningfull_list.index(word)\n",
    "\n",
    "        word_to_process = re.sub(r' / ', \",\", word)\n",
    "        splitword = word_to_process.split(',')\n",
    "        length_split_word = len(splitword)\n",
    "\n",
    "        meaningfull_list.remove(word)\n",
    "        meaningfull_list.insert((position_to_split ), splitword[0] ) # For simplicity lets add only 1st word.\n",
    "\n",
    "# FOr \"What are/whate were\" -> to \"what are\"\n",
    "meaningfull_list_2 = []\n",
    "for word in meaningfull_list:\n",
    "  \n",
    "  for alpha in word:\n",
    "    if(\"/\" in alpha):\n",
    "        word = word.split(\"/\")[0]\n",
    "    meaningfull_list_2.append(word)  \n",
    "    \n",
    "another_list = []\n",
    "for words in meaningfull_list_2:\n",
    "    split_word = words.split(\" \")\n",
    "  \n",
    "    if(\"has\" in split_word):\n",
    "        split_word = [word.replace(\"has\", \"is\") for word in split_word]   \n",
    "    another_list.append(\" \".join(split_word))\n",
    "    \n",
    "#print(len(another_list) )\n",
    "#print(len(meaningfull_list_2) )\n",
    "dictionary = {k: v for k,v in zip(contraction_list, another_list)}\n",
    "del dictionary['']\n",
    "# Step by step output is available on : https://github.com/MohammadWasil/Quora-Insincere-Question-Classification/blob/master/Quora_Insincere_Questions.ipynb\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 2244
    },
    "colab_type": "code",
    "id": "2C7URFigWT_J",
    "outputId": "85990532-d645-4b9b-bb93-deb1b0ea1635"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{\"'cause\": 'because',\n",
       " \"'s\": 'is',\n",
       " \"'tis\": 'it is',\n",
       " \"'twas\": 'it was',\n",
       " \"Ha'ta\": 'even',\n",
       " \"I'd\": 'I had',\n",
       " \"I'll\": 'I shall',\n",
       " \"I'm\": 'I am',\n",
       " \"I'm'a\": 'I am about to',\n",
       " \"I'm'o\": 'I am going to',\n",
       " \"I've\": 'I have',\n",
       " \"S'e\": 'oh yeah',\n",
       " \"ain't\": 'am not',\n",
       " \"aren't\": 'are not',\n",
       " \"cain't\": 'cannot',\n",
       " \"can't\": 'cannot',\n",
       " \"could've\": 'could have',\n",
       " \"couldn't\": 'could not',\n",
       " \"couldn't've\": 'could not have',\n",
       " \"daren't\": 'dare not',\n",
       " \"daresn't\": 'dare not',\n",
       " \"dasn't\": 'dare not',\n",
       " \"didn't\": 'did not',\n",
       " \"doesn't\": 'does not',\n",
       " \"don't\": 'do not',\n",
       " \"e'er\": 'ever',\n",
       " \"everyone's\": 'everyone is',\n",
       " 'finna': 'fixing to',\n",
       " 'gimme': 'give me',\n",
       " \"giv'n\": 'given',\n",
       " \"gon't\": 'go not',\n",
       " 'gonna': 'going to',\n",
       " 'gotta': 'got to',\n",
       " \"hadn't\": 'had not',\n",
       " \"hasn't\": 'is not',\n",
       " \"haven't\": 'have not',\n",
       " \"he'd\": 'he had',\n",
       " \"he'll\": 'he shall',\n",
       " \"he's\": 'he is',\n",
       " \"he've\": 'he have',\n",
       " \"how'd\": 'how did',\n",
       " \"how'll\": 'how will',\n",
       " \"how're\": 'how are',\n",
       " \"how's\": 'how is',\n",
       " 'howdy': 'how do you do',\n",
       " \"isn't\": 'is not',\n",
       " \"it'd\": 'it would',\n",
       " \"it'll\": 'it shall',\n",
       " \"it's\": 'it is',\n",
       " \"let's\": 'let us',\n",
       " \"ma'am\": 'madam',\n",
       " \"may've\": 'may have',\n",
       " \"mayn't\": 'may not',\n",
       " \"might've\": 'might have',\n",
       " \"mightn't\": 'might not',\n",
       " \"must've\": 'must have',\n",
       " \"mustn't\": 'must not',\n",
       " \"mustn't've\": 'must not have',\n",
       " \"ne'er\": 'never',\n",
       " \"needn't\": 'need not',\n",
       " \"o'clock\": 'of the clock',\n",
       " \"o'er\": 'over',\n",
       " \"ol'\": 'old',\n",
       " \"oughtn't\": 'ought not',\n",
       " 'rarely': 'cannot',\n",
       " \"shalln't\": 'shall not',\n",
       " \"shan't\": 'shall not',\n",
       " \"she'd\": 'she had',\n",
       " \"she'll\": 'she shall',\n",
       " \"she's\": 'she is',\n",
       " \"should've\": 'should have',\n",
       " \"shouldn't\": 'should not',\n",
       " \"shouldn't've\": 'should not have',\n",
       " \"so're\": 'so are',\n",
       " \"somebody's\": 'somebody is',\n",
       " \"someone's\": 'someone is',\n",
       " \"something's\": 'something is',\n",
       " \"that'd\": 'that would',\n",
       " \"that'll\": 'that shall',\n",
       " \"that're\": 'that are',\n",
       " \"that's\": 'that is',\n",
       " \"there'd\": 'there had',\n",
       " \"there'll\": 'there shall',\n",
       " \"there're\": 'there are',\n",
       " \"there's\": 'there is',\n",
       " \"these're\": 'these are',\n",
       " \"they'd\": 'they had',\n",
       " \"they'll\": 'they shall',\n",
       " \"they're\": 'they are',\n",
       " \"they've\": 'they have',\n",
       " \"this's\": 'this is',\n",
       " \"those're\": 'those are',\n",
       " \"to've\": 'to have',\n",
       " \"wasn't\": 'was not',\n",
       " \"we'd\": 'we had',\n",
       " \"we'd've\": 'we would have',\n",
       " \"we'll\": 'we will',\n",
       " \"we're\": 'we are',\n",
       " \"we've\": 'we have',\n",
       " \"weren't\": 'were not',\n",
       " \"what'd\": 'what did',\n",
       " \"what'll\": 'what shall',\n",
       " \"what're\": 'what are',\n",
       " \"what's\": 'what is',\n",
       " \"what've\": 'what have',\n",
       " \"when's\": 'when is',\n",
       " \"where'd\": 'where did',\n",
       " \"where're\": 'where are',\n",
       " \"where's\": 'where is',\n",
       " \"where've\": 'where have',\n",
       " \"which's\": 'which is',\n",
       " \"who'd\": 'who would',\n",
       " \"who'd've\": 'who would have',\n",
       " \"who'll\": 'who shall',\n",
       " \"who're\": 'who are',\n",
       " \"who's\": 'who is',\n",
       " \"who've\": 'who have',\n",
       " \"whom'st\": 'whom hast',\n",
       " \"whom'st'd've\": 'whom hast had have',\n",
       " \"why'd\": 'why did',\n",
       " \"why're\": 'why are',\n",
       " \"why's\": 'why is',\n",
       " \"won't\": 'will not',\n",
       " \"would've\": 'would have',\n",
       " \"wouldn't\": 'would not',\n",
       " \"y'all\": 'you all',\n",
       " \"y'all'd've\": 'you all would have',\n",
       " \"you'd\": 'you had',\n",
       " \"you'll\": 'you shall',\n",
       " \"you're\": 'you are',\n",
       " \"you've\": 'you have'}"
      ]
     },
     "execution_count": 111,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dictionary "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "sTswnTCAPscn",
    "outputId": "2c43929f-e2d5-4a5c-ec01-61078c464ebf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'2', '4', '8', '9', '3', '7', '1', '0', '6', '5'}\n"
     ]
    }
   ],
   "source": [
    "# We have numbers here, we are going to convert it to english names.\n",
    "numbers = []\n",
    "length = int(len(data_train_qa))\n",
    "for i in range(length):\n",
    "    no = re.findall(r\"[0-9]\", data_train_qa[\"question\"][i])\n",
    "    if no:\n",
    "        numbers.append(no)\n",
    "\n",
    "# Get all unique punctuation from Top 200 questions\n",
    "mysetnumber = set()\n",
    "for i in range(600):\n",
    "    for j in range(len(numbers[i])):\n",
    "        mysetnumber.add(numbers[i][j])\n",
    "print(mysetnumber)\n",
    "  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "KAYS4uKCwRkp"
   },
   "outputs": [],
   "source": [
    "number_dictionary = {\"1\" : \"one\", \"2\" : \"two\", \"3\": \"three\", \"4\" : \"four\", \"5\": \"five\", \"6\": \"six\", \"7\" : \"seven\", \"8\": \"eight\", \"9\":\"nine\", \"0\":\"zero\"}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ntH4NKsjxDOB"
   },
   "outputs": [],
   "source": [
    "# Review the questions:\n",
    "def mappingWords(questions,dictionary):\n",
    "    return \" \".join([dictionary.get(w,w) for w in questions.split()])\n",
    "\n",
    "def mappingNumber(questions, number_dictionary):\n",
    "    return \" \".join([number_dictionary.get(w,w) for w in questions.split()])\n",
    "\n",
    "def review_question(questions):\n",
    "    questions = questions.lower()\n",
    "    questions = mappingWords(questions, dictionary)\n",
    "    questions = mappingNumber(questions, number_dictionary)\n",
    "    questions = re.sub(r\"[^A-Za-z0-9 ]\", \"\", questions)\n",
    "    return questions\n",
    "\n",
    "train_question_list = []\n",
    "validation_question_list = []\n",
    "for i in range(len(data_train_qa[\"question\"])):\n",
    "    train_question_list.append(review_question(data_train_qa[\"question\"][i]))\n",
    "    \n",
    "for i in range(len(data_val_qa[\"question\"])):\n",
    "    validation_question_list.append(review_question(data_val_qa[\"question\"][i]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "fI-WsSv323zY"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "df_train_questions = pd.DataFrame(train_question_list, columns = [\"questions\"])\n",
    "df_val_questions = pd.DataFrame(validation_question_list, columns = [\"questions\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "iYHKQ3Tz3llP"
   },
   "outputs": [],
   "source": [
    "validation = pd.concat([data_val_qa, df_val_questions], axis = 1)\n",
    "training   = pd.concat([data_train_qa, df_train_questions], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "yWtv4dfI6GAb"
   },
   "outputs": [],
   "source": [
    "data_val_qa = validation.drop( columns = [\"question\"])\n",
    "data_train_qa = training.drop( columns = [\"question\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "iBVI0Dmf6jhl"
   },
   "outputs": [],
   "source": [
    "data_train_qa = data_train_qa.reindex(columns = [\"image_id\", \"questions\", \"multiple_choice_answer\"])\n",
    "data_val_qa = data_val_qa.reindex(columns = [\"image_id\", \"questions\", \"multiple_choice_answer\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "id": "1DtX6IoU7VGk",
    "outputId": "a8972cb3-06b6-4bbd-fb17-b8007e8a83c2"
   },
   "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>image_id</th>\n",
       "      <th>questions</th>\n",
       "      <th>multiple_choice_answer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>350623</td>\n",
       "      <td>what is the table made of</td>\n",
       "      <td>wood</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>350623</td>\n",
       "      <td>is the food napping on the table</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>350623</td>\n",
       "      <td>what has been upcycled to make lights</td>\n",
       "      <td>kettles</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8647</td>\n",
       "      <td>is this an spanish town</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8647</td>\n",
       "      <td>are there shadows on the sidewalk</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   image_id                              questions multiple_choice_answer\n",
       "0    350623              what is the table made of                   wood\n",
       "1    350623       is the food napping on the table                     no\n",
       "2    350623  what has been upcycled to make lights                kettles\n",
       "3      8647                is this an spanish town                     no\n",
       "4      8647      are there shadows on the sidewalk                    yes"
      ]
     },
     "execution_count": 133,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train_qa.head(5)\n",
    "data_val_qa.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "zquh0H7aMU0o"
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "open_pickle_val = open('Validation Data QA.pickle', 'wb')\n",
    "pickle.dump(data_val_qa, open_pickle_val)\n",
    "open_pickle_val.close()\n",
    "\n",
    "open_pickle_train = open('Training Data QA.pickle', 'wb')\n",
    "pickle.dump(data_train_qa, open_pickle_train)\n",
    "open_pickle_train.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "Ziqw9jpqibVi",
    "outputId": "3edc86fa-6631-42fb-ded5-6aa530f47d32"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/content\n"
     ]
    }
   ],
   "source": [
    "#https://drive.google.com/open?id=1NVfn_L9YoGJloB_rzKh06YCqQNe6DfJj\n",
    "%cd /content/\n",
    "!pip install -U -q PyDrive\n",
    "\n",
    "from pydrive.auth import GoogleAuth\n",
    "from pydrive.drive import GoogleDrive\n",
    "from google.colab import auth\n",
    "from oauth2client.client import GoogleCredentials\n",
    "\n",
    "# 1. Authenticate and create the PyDrive client.\n",
    "auth.authenticate_user()\n",
    "gauth = GoogleAuth()\n",
    "gauth.credentials = GoogleCredentials.get_application_default()\n",
    "drive = GoogleDrive(gauth)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 68
    },
    "colab_type": "code",
    "id": "SXopnWmKigES",
    "outputId": "1b45822c-0d95-486b-f20c-6c1a768c9063"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/content\n",
      "Uploaded file with ID 1qmDX_URx9rKqjBZczX0hINk5uA4kI9SA\n",
      "Uploaded file with ID 1vBVqBh-caYYat7ZXBleJWHGV0al1I4nJ\n"
     ]
    }
   ],
   "source": [
    "# Upload it on the drive.\n",
    "%cd /content\n",
    "uploaded = drive.CreateFile({'title': 'Training Data QA.pickle'})\n",
    "uploaded.SetContentFile('Training Data QA.pickle')\n",
    "uploaded.Upload()\n",
    "print('Uploaded file with ID {}'.format(uploaded.get('id')))\n",
    "\n",
    "uploaded = drive.CreateFile({'title': 'Validation Data QA.pickle'})\n",
    "uploaded.SetContentFile('Validation Data QA.pickle')\n",
    "uploaded.Upload()\n",
    "print('Uploaded file with ID {}'.format(uploaded.get('id')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Co2CvJSMB5on"
   },
   "source": [
    "### COCO image dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "LN1M7UC9fX0O"
   },
   "source": [
    "VGG16 trained images, which have been used in Visual Qa datasets. Since we are going to use VGG16 architecture, so we wont be training all the images, instead we will be using the image feature matrix from stanford, which have been already trained on VGG16 architecture. All these images are present in the training and validation data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 272
    },
    "colab_type": "code",
    "id": "LSpaWJx31RbX",
    "outputId": "25e14b89-f86b-4c7f-cb39-1c75addae014"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--2019-06-15 12:21:01--  http://cs.stanford.edu/people/karpathy/deepimagesent/coco.zip\n",
      "Resolving cs.stanford.edu (cs.stanford.edu)... 171.64.64.64\n",
      "Connecting to cs.stanford.edu (cs.stanford.edu)|171.64.64.64|:80... connected.\n",
      "HTTP request sent, awaiting response... 301 Moved Permanently\n",
      "Location: https://cs.stanford.edu/people/karpathy/deepimagesent/coco.zip [following]\n",
      "--2019-06-15 12:21:02--  https://cs.stanford.edu/people/karpathy/deepimagesent/coco.zip\n",
      "Connecting to cs.stanford.edu (cs.stanford.edu)|171.64.64.64|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 764830639 (729M) [application/zip]\n",
      "Saving to: ‘coco.zip’\n",
      "\n",
      "coco.zip            100%[===================>] 729.40M  16.5MB/s    in 50s     \n",
      "\n",
      "2019-06-15 12:21:52 (14.6 MB/s) - ‘coco.zip’ saved [764830639/764830639]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!wget http://cs.stanford.edu/people/karpathy/deepimagesent/coco.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 85
    },
    "colab_type": "code",
    "id": "WmP2bSbMiEc0",
    "outputId": "e25614d0-2b87-4414-972b-0c3f538b5790"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Archive:  coco.zip\n",
      "  inflating: coco/dataset.json       \n",
      "  inflating: coco/readme.txt         \n",
      "  inflating: coco/vgg_feats.mat      \n"
     ]
    }
   ],
   "source": [
    "!unzip coco.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "HidDfGjlstqn",
    "outputId": "4f5da553-a74b-4d73-c1f0-c362ecc503b8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/content/coco\n"
     ]
    }
   ],
   "source": [
    "%cd /content/coco\n",
    "import json\n",
    "\n",
    "with open('dataset.json') as f:\n",
    "    data = json.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "JM9j0it0kl0y",
    "outputId": "a5fe7c8c-51b3-4c8d-e4b6-ac1ea2e585ef"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "123287"
      ]
     },
     "execution_count": 27,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data[\"images\"])   # This number of images is equal to the number of images in training and validation data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data\n",
    "'''\n",
    "The output looks something like this:\n",
    "{'dataset': 'coco',\n",
    " 'images': [{'cocoid': 391895,\n",
    "   'filename': 'COCO_val2014_000000391895.jpg',\n",
    "   'filepath': 'val2014',\n",
    "   'imgid': 0,\n",
    "   'sentences': [{'imgid': 0,\n",
    "     'raw': 'A man with a red helmet on a small moped on a dirt road. ',\n",
    "     'sentid': 770337,\n",
    "     'tokens': ['a',\n",
    "      'man',\n",
    "      'with',\n",
    "      'a',\n",
    "      'red',\n",
    "      'helmet',\n",
    "      'on',\n",
    "      'a',\n",
    "      'small',\n",
    "      'moped',\n",
    "      'on',\n",
    "      'a',\n",
    "      'dirt',\n",
    "      'road']},\n",
    "    {'imgid': 0,\n",
    "     'raw': 'Man riding a motor bike on a dirt road on the countryside.',\n",
    "     'sentid': 771687,\n",
    "     'tokens': ['man',\n",
    "      'riding',\n",
    "      'a',\n",
    "      'motor',\n",
    "      'bike',\n",
    "      'on',\n",
    "      'a',\n",
    "      'dirt',\n",
    "      'road',\n",
    "      'on',\n",
    "      'the',\n",
    "      'countryside']},\n",
    "    {'imgid': 0,\n",
    "     'raw': 'A man riding on the back of a motorcycle.',\n",
    "     'sentid': 772707,\n",
    "     'tokens': ['a',\n",
    "      'man',\n",
    "      'riding',\n",
    "      'on',\n",
    "      'the',\n",
    "      'back',\n",
    "      'of',\n",
    "      'a',\n",
    "      'motorcycle']},\n",
    "    {'imgid': 0,\n",
    "     'raw': 'A dirt path with a young person on a motor bike rests to the foreground of a verdant area with a bridge and a background of cloud-wreathed mountains. ',\n",
    "     'sentid': 776154,\n",
    "     'tokens': ['a',\n",
    "      'dirt',\n",
    "      'path',\n",
    "      'with',\n",
    "      'a',\n",
    "      'young',\n",
    "      'person',\n",
    "      'on',\n",
    "      'a',\n",
    "      'motor',\n",
    "      'bike',\n",
    "      'rests',\n",
    "      'to',\n",
    "      'the',\n",
    "      'foreground',\n",
    "      'of',\n",
    "      'a',\n",
    "      'verdant',\n",
    "      'area',\n",
    "      'with',\n",
    "      'a',\n",
    "      'bridge',\n",
    "      'and',\n",
    "      'a',\n",
    "      'background',\n",
    "      'of',\n",
    "      'cloud',\n",
    "      'wreathed',\n",
    "      'mountains']},\n",
    "      ...\n",
    "      {'imgid': 999,\n",
    "     'raw': 'A vase filled with assorted color flowers sitting on a table.',\n",
    "     'sentid': 75170,\n",
    "     'tokens': ['a',\n",
    "      'vase',\n",
    "      'filled',\n",
    "      'with',\n",
    "      'assorted',\n",
    "      'color',\n",
    "      'flowers',\n",
    "      'sitting',\n",
    "      'on',\n",
    "      'a',\n",
    "      'table']}],\n",
    "   'sentids': [69920, 71639, 71693, 73331, 75170],\n",
    "   'split': 'restval'},\n",
    "  ...]}\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "9maPLhU_hrG7",
    "outputId": "59b0942c-4ccb-4d7a-c6b9-77e32b7b052b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "391895"
      ]
     },
     "execution_count": 29,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Lets create a relationaship between this coco data and trainnig data!\n",
    "data[\"images\"][0][\"cocoid\"]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 142
    },
    "colab_type": "code",
    "id": "q_cWImqgjKxr",
    "outputId": "83448203-4604-41ca-9f38-385840d2bac4"
   },
   "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>image_id</th>\n",
       "      <th>questions</th>\n",
       "      <th>multiple_choice_answer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>107517</th>\n",
       "      <td>391895</td>\n",
       "      <td>is the road paved</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107518</th>\n",
       "      <td>391895</td>\n",
       "      <td>what type of road structure can be seen in the...</td>\n",
       "      <td>bridge</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107519</th>\n",
       "      <td>391895</td>\n",
       "      <td>is the person wearing a helmet</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        image_id  ... multiple_choice_answer\n",
       "107517    391895  ...                     no\n",
       "107518    391895  ...                 bridge\n",
       "107519    391895  ...                    yes\n",
       "\n",
       "[3 rows x 3 columns]"
      ]
     },
     "execution_count": 137,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_val_qa.loc[data_val_qa[\"image_id\"] == 391895]\n",
    "# Here, each cocoid is related to image_id of training and validation data. So we are going to collect cocoid and create a mapping file for it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "YGRpRGsjmKAu"
   },
   "outputs": [],
   "source": [
    "# creating coco_vgg_id_map file(Used later on)\n",
    "\n",
    "map_list = [ [data[\"images\"][i][\"cocoid\"], i]  for i in range(len(data[\"images\"]))]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 187
    },
    "colab_type": "code",
    "id": "EagxISnXoElY",
    "outputId": "14dfc697-174b-46e2-81de-e58d54a17f7c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[391895, 0],\n",
       " [522418, 1],\n",
       " [184613, 2],\n",
       " [318219, 3],\n",
       " [554625, 4],\n",
       " [397133, 5],\n",
       " [574769, 6],\n",
       " [60623, 7],\n",
       " [309022, 8],\n",
       " [5802, 9]]"
      ]
     },
     "execution_count": 38,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "map_list[0:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "XAJgY5kG1WKH"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "datafrm = pd.DataFrame(map_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4e39hEzP4DCx"
   },
   "outputs": [],
   "source": [
    "datafrm.to_csv('coco_vgg_id_map.txt', header = None, index = None, sep=' ', mode = 'a')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "6uPnvFXUm_2M",
    "outputId": "83bb9ec9-537c-421b-ec41-035f33fc3981"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Uploaded file with ID 1iLrNPNQ3GMZquyE7QhB7l9Yt4mS46r_p\n"
     ]
    }
   ],
   "source": [
    "uploaded = drive.CreateFile({'title': 'coco_vgg_id_map.txt'})\n",
    "uploaded.SetContentFile('coco_vgg_id_map.txt')\n",
    "uploaded.Upload()\n",
    "print('Uploaded file with ID {}'.format(uploaded.get('id')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "0XfVBrkRCihv"
   },
   "source": [
    "### Downoad Image Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "eS8R7fi7Cm8o"
   },
   "source": [
    "Finding the image_id of each image."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 391
    },
    "colab_type": "code",
    "id": "IR0nO6VKdtMt",
    "outputId": "144d2bad-d154-46d4-d496-3effc87dd99e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--2019-06-15 13:16:35--  http://images.cocodataset.org/zips/train2014.zip\n",
      "Resolving images.cocodataset.org (images.cocodataset.org)... 52.216.17.144\n",
      "Connecting to images.cocodataset.org (images.cocodataset.org)|52.216.17.144|:80... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 13510573713 (13G) [application/zip]\n",
      "Saving to: ‘train2014.zip’\n",
      "\n",
      "train2014.zip       100%[===================>]  12.58G  19.8MB/s    in 10m 51s \n",
      "\n",
      "2019-06-15 13:27:27 (19.8 MB/s) - ‘train2014.zip’ saved [13510573713/13510573713]\n",
      "\n",
      "--2019-06-15 13:27:29--  http://images.cocodataset.org/zips/val2014.zip\n",
      "Resolving images.cocodataset.org (images.cocodataset.org)... 52.216.128.171\n",
      "Connecting to images.cocodataset.org (images.cocodataset.org)|52.216.128.171|:80... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 6645013297 (6.2G) [application/zip]\n",
      "Saving to: ‘val2014.zip’\n",
      "\n",
      "val2014.zip         100%[===================>]   6.19G  20.0MB/s    in 5m 19s  \n",
      "\n",
      "2019-06-15 13:32:49 (19.8 MB/s) - ‘val2014.zip’ saved [6645013297/6645013297]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Download image datasets!\n",
    "!wget http://images.cocodataset.org/zips/train2014.zip\n",
    "!wget http://images.cocodataset.org/zips/val2014.zip      "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Since training is very large, and will take long time to unzip, we are going to unzip validation data.\n",
    "# Although we dont need the image dataset, but still we are going to look at it.\n",
    "!unzip val2014.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 286
    },
    "colab_type": "code",
    "id": "zBdax443AqKK",
    "outputId": "5aa2b5bf-c5eb-44be-fc7d-569d0b20017a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXEAAAD8CAYAAACB3pQWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5yR+xhlnnPGAcU7iZ5xxxhkPGP83JvqOfxbqR6EAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%pylab inline\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "\n",
    "image_directory = '/content/coco/val2014/COCO_val2014_000000324670.jpg'\n",
    "\n",
    "img = mpimg.imread(image_directory)\n",
    "imgplot = plt.imshow(img)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "k3_GcMZK3_0S"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import pickle\n",
    "\n",
    "validation_data_pickle = open(\"Validation Data QA.pickle\", \"rb\")\n",
    "validation_data = pickle.load(validation_data_pickle)\n",
    "validation_data_pickle.close()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 142
    },
    "colab_type": "code",
    "id": "X30hwGMiAjOF",
    "outputId": "60ec4f2c-a51f-42e6-967a-091c057a6c01"
   },
   "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>image_id</th>\n",
       "      <th>questions</th>\n",
       "      <th>multiple_choice_answer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10725</th>\n",
       "      <td>324670</td>\n",
       "      <td>could this be a seabird</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10726</th>\n",
       "      <td>324670</td>\n",
       "      <td>what element of this photograph is in sharp focus</td>\n",
       "      <td>bird</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10727</th>\n",
       "      <td>324670</td>\n",
       "      <td>is it a cold day</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       image_id  ... multiple_choice_answer\n",
       "10725    324670  ...                    yes\n",
       "10726    324670  ...                   bird\n",
       "10727    324670  ...                     no\n",
       "\n",
       "[3 rows x 3 columns]"
      ]
     },
     "execution_count": 139,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's check for the image_id.\n",
    "image_id = \"324670\" # Lst six digit of each image name.\n",
    "validation_data.loc[validation_data[\"image_id\"] == 324670 ]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ZI_eGMhMBGvU"
   },
   "source": [
    "So, the last six digits numbers of the image name is the image_id."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "-ASZfv_fDA2K"
   },
   "source": [
    "Continue to the next notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "S2zLtbRHDJFd"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Data.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python [default]",
   "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": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}