OCR_EDA.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h24KpW2YL9Cm"
      },
      "source": [
        "## EDA 할 내용\n",
        "- 이미지 당 단어 개수 분포 \n",
        "- 전체 단어 길이 분포\n",
        "- image width, height\n",
        "- 전체 BBOX 크기 분포\n",
        "    - 넓이 기준\n",
        "- 전체 태그 별 분포\n",
        "    - 언어\n",
        "    - orientation\n",
        "    - word tag\n",
        "    - image tag\n",
        "- aspect ratio (가로/세로)\n",
        "   - polygon은 rectify하기"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2pR_L1BHL9Co"
      },
      "outputs": [],
      "source": [
        "'''\n",
        "original code source : https://www.kaggle.com/aidapearson/eda-starter-notebook\n",
        "'''"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "eO7OcR7bL9Cp"
      },
      "outputs": [],
      "source": [
        "%matplotlib inline\n",
        "import glob\n",
        "import json\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "from pathlib import Path\n",
        "from PIL import Image\n",
        "from collections import Counter\n",
        "import numpy as np"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Y75ZoSi5L9Cq"
      },
      "outputs": [],
      "source": [
        "from IPython.core.interactiveshell import InteractiveShell\n",
        "InteractiveShell.ast_node_interactivity = \"all\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2yOmyrp5L9Cq"
      },
      "outputs": [],
      "source": [
        "def read_json(filename):\n",
        "    with Path(filename).open(encoding='utf8') as handle:\n",
        "        ann = json.load(handle)\n",
        "    return ann"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "F5_5YmOdL9Cq"
      },
      "outputs": [],
      "source": [
        "# gt file과 raw image file path를 이용해 데이터를 불러온다. 아래 경로는 예시.\n",
        "data = read_json(\"../input/data/ICDAR17_Korean/ufo/train.json\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "dCiJnhnBL9Cr"
      },
      "outputs": [],
      "source": [
        "def get_box_size(quads):\n",
        "    \"\"\" 단어 영역의 사각형 좌표가 주어졌을 때 가로, 세로길이를 계산해주는 함수.\n",
        "    TODO: 각 변의 길이를 단순히 max로 처리하기때문에 직사각형에 가까운 형태가 아니면 약간 왜곡이 있다.\n",
        "    Args:\n",
        "        quads: np.ndarray(n, 4, 2) n개 단어 bounding-box의 4개 점 좌표 (단위 pixel)\n",
        "    Return:\n",
        "        sizes: np.ndarray(n, 2) n개 box의 (height, width)쌍\n",
        "    \"\"\"\n",
        "    dists = []\n",
        "    for i, j in [(1, 2), (3, 0), (0, 1), (2, 3)]: # [right(height), left(height), upper(width), lower(width)] sides\n",
        "        dists.append(np.linalg.norm(quads[:, i] - quads[:, j], ord=2, axis=1))\n",
        "\n",
        "    dists = np.stack(dists, axis=-1).reshape(-1, 2, 2) # shape (n, 2, 2) widths, heights into separate dim\n",
        "    return np.rint(dists.mean(axis=-1)).astype(int)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "24cTfnCrL9Cr"
      },
      "outputs": [],
      "source": [
        "def rectify_poly(poly, direction, img_w, img_h):\n",
        "    \"\"\"일반 polygon형태인 라벨을 크롭하고 rectify해주는 함수.\n",
        "    Args:\n",
        "        poly: np.ndarray(2n+4, 2) (where n>0), 4, 6, 8\n",
        "        image: np.ndarray opencv 포멧의 이미지\n",
        "        direction: 글자의 읽는 방향과 진행 방향의 수평(Horizontal) 혹은 수직(Vertical) 여부\n",
        "    Return:\n",
        "        rectified: np.ndarray(2, ?) rectify된 단어 bbox의 사이즈.\n",
        "    \"\"\"\n",
        "    \n",
        "    n_pts = poly.shape[0]\n",
        "    assert n_pts % 2 == 0\n",
        "    if n_pts == 4:\n",
        "        size = get_box_size(poly[None])\n",
        "        h = size[:, 0] / img_h\n",
        "        w = size[:, 1] / img_w\n",
        "        return np.stack((h,w))\n",
        "\n",
        "    def unroll(indices):\n",
        "        return list(zip(indices[:-1], indices[1:]))\n",
        "\n",
        "    # polygon하나를 인접한 사각형 여러개로 쪼갠다.\n",
        "    indices = list(range(n_pts))\n",
        "    if direction == 'Horizontal':\n",
        "        upper_pts = unroll(indices[:n_pts // 2]) # (0, 1), (1, 2), ... (4, 5)\n",
        "        lower_pts = unroll(indices[n_pts // 2:])[::-1] # (8, 9), (7, 8), ... (6, 7)\n",
        "\n",
        "        quads = np.stack([poly[[i, j, k, l]] for (i, j), (k, l) in zip(upper_pts, lower_pts)])\n",
        "    else:\n",
        "        right_pts = unroll(indices[1:n_pts // 2 + 1]) # (1, 2), (2, 3), ... (4, 5)\n",
        "        left_pts = unroll([0] + indices[:n_pts // 2:-1]) # (0, 9), (9, 8), ... (7, 6)\n",
        "\n",
        "        quads = np.stack([poly[[i, j, k, l]] for (j, k), (i, l) in zip(right_pts, left_pts)])\n",
        "\n",
        "    sizes = get_box_size(quads)\n",
        "    if direction == 'Horizontal':\n",
        "        h = sizes[:, 0].max() / img_h\n",
        "        widths = sizes[:, 1]\n",
        "        w = np.sum(widths) / img_w\n",
        "        return np.stack((h,w)).reshape(2,-1)\n",
        "        #return np.stack((h,w))\n",
        "    elif direction == 'Vertical':\n",
        "        heights = sizes[:, 0]\n",
        "        w = sizes[:, 1].max() / img_w\n",
        "        h = np.sum(heights) / img_h\n",
        "        return np.stack((h,w)).reshape(2,-1)\n",
        "    else:\n",
        "        h = sizes[:, 0] / img_h\n",
        "        w = sizes[:, 1] / img_w\n",
        "        return np.stack((h,w),-1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "E9_VkNJcL9Cs"
      },
      "outputs": [],
      "source": [
        "df = {}\n",
        "df['image'] = []\n",
        "df['word_counts'] = []\n",
        "df['image_width'] = []\n",
        "df['image_height'] = []\n",
        "df['image_tags'] = []\n",
        "img_tags = []\n",
        "\n",
        "quads = []\n",
        "polys = []\n",
        "seq_length = []\n",
        "hor_sizes = []\n",
        "ver_sizes = []\n",
        "irr_sizes = []\n",
        "languages = []\n",
        "orientation = []\n",
        "word_tags = []\n",
        "aspect_ratio = []\n",
        "ver_string = []\n",
        "\n",
        "for image_key, image_value in data[\"images\"].items():\n",
        "    df['image'].append(image_key)\n",
        "    img_w = image_value['img_w']\n",
        "    img_h = image_value['img_h']\n",
        "    df['image_width'].append(img_w)\n",
        "    df['image_height'].append(img_h)\n",
        "    df['image_tags'].append(image_value['tags'])\n",
        "    df['image_tags']= [['None'] if v is None else v for v in df['image_tags']] # our data does not inlcude multi-tag images \n",
        "    word_ann = image_value['words']\n",
        "    count_ill = 0 \n",
        "    for word in word_ann.values():\n",
        "        if word['illegibility']== False:\n",
        "            orientation.append(word['orientation'])\n",
        "            orientation = [v for v in orientation]\n",
        "            seq_length.append(len(word['transcription']))\n",
        "            languages.append(word['language'])\n",
        "            languages = [['None'] if v is None else v for v in languages] # our data does not inlcude multi-language words\n",
        "            if word['word_tags'] != None:\n",
        "                word_tags.extend(word['word_tags'][:])\n",
        "            elif word['word_tags']== None:\n",
        "                word_tags.append('None')\n",
        "            poly = np.int32(word['points'])\n",
        "            size = rectify_poly(poly, word['orientation'], img_w, img_h)\n",
        "            if word['orientation'] == 'Horizontal':\n",
        "                hor_sizes.append(size)\n",
        "            elif word['orientation'] == 'Vertical':\n",
        "                ver_sizes.append(size)\n",
        "            else:\n",
        "                irr_sizes.append(size)\n",
        "            \n",
        "        else:\n",
        "            count_ill += 1\n",
        "    df['word_counts'].append(len(word_ann)-count_ill)\n",
        "\n",
        "        \n",
        "all_sizes = hor_sizes + ver_sizes + irr_sizes\n",
        "quad_area = [all_sizes[i][0]*all_sizes[i][1] for i in range(len(all_sizes))]\n",
        "total_area = []\n",
        "for s in quad_area:\n",
        "    if s.shape[0] == 1:\n",
        "        total_area.append(np.sum(s[0])) \n",
        "    else:\n",
        "        total_area.append(np.sum(s))\n",
        "\n",
        "hor_aspect_ratio = [hor_sizes[i][1]/hor_sizes[i][0] for i in range(len(hor_sizes))]\n",
        "ver_aspect_ratio = [ver_sizes[i][1]/ver_sizes[i][0] for i in range(len(ver_sizes))]\n",
        "\n",
        "image_df = pd.DataFrame.from_dict(df)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "tnSGtrrpL9Ct",
        "outputId": "90512162-bf4e-4a24-e777-0306d3922c14"
      },
      "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</th>\n",
              "      <th>word_counts</th>\n",
              "      <th>image_width</th>\n",
              "      <th>image_height</th>\n",
              "      <th>image_tags</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>AB_naver_0329.jpg</td>\n",
              "      <td>6</td>\n",
              "      <td>510</td>\n",
              "      <td>832</td>\n",
              "      <td>outfocus</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>A_naver_0009.jpg</td>\n",
              "      <td>62</td>\n",
              "      <td>620</td>\n",
              "      <td>827</td>\n",
              "      <td>outfocus</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>A_naver_0359.jpg</td>\n",
              "      <td>5</td>\n",
              "      <td>512</td>\n",
              "      <td>512</td>\n",
              "      <td>outfocus</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>BBnaver_0176.jpg</td>\n",
              "      <td>45</td>\n",
              "      <td>340</td>\n",
              "      <td>255</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>BBnaver_0337.jpg</td>\n",
              "      <td>28</td>\n",
              "      <td>683</td>\n",
              "      <td>512</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>145</th>\n",
              "      <td>Y_naver_0010.jpg</td>\n",
              "      <td>10</td>\n",
              "      <td>588</td>\n",
              "      <td>512</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>146</th>\n",
              "      <td>Y_naver_0073.jpg</td>\n",
              "      <td>47</td>\n",
              "      <td>620</td>\n",
              "      <td>827</td>\n",
              "      <td>handwriting</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>147</th>\n",
              "      <td>Y_naver_0079.jpg</td>\n",
              "      <td>11</td>\n",
              "      <td>468</td>\n",
              "      <td>832</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>148</th>\n",
              "      <td>Y_naver_0143.jpg</td>\n",
              "      <td>1</td>\n",
              "      <td>340</td>\n",
              "      <td>449</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>149</th>\n",
              "      <td>Y_naver_0298.jpg</td>\n",
              "      <td>12</td>\n",
              "      <td>554</td>\n",
              "      <td>832</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>150 rows × 5 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                 image  word_counts  image_width  image_height   image_tags\n",
              "0    AB_naver_0329.jpg            6          510           832     outfocus\n",
              "1     A_naver_0009.jpg           62          620           827     outfocus\n",
              "2     A_naver_0359.jpg            5          512           512     outfocus\n",
              "3     BBnaver_0176.jpg           45          340           255         None\n",
              "4     BBnaver_0337.jpg           28          683           512         None\n",
              "..                 ...          ...          ...           ...          ...\n",
              "145   Y_naver_0010.jpg           10          588           512         None\n",
              "146   Y_naver_0073.jpg           47          620           827  handwriting\n",
              "147   Y_naver_0079.jpg           11          468           832         None\n",
              "148   Y_naver_0143.jpg            1          340           449         None\n",
              "149   Y_naver_0298.jpg           12          554           832         None\n",
              "\n",
              "[150 rows x 5 columns]"
            ]
          },
          "execution_count": 90,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "image_df['image_tags'] = image_df['image_tags'].apply(lambda x: ','.join(map(str, x)))\n",
        "image_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "xOcHV3lXL9Ct",
        "outputId": "5209e90e-6ea6-4aa7-dee0-a746bd08857e"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "count      150\n",
              "unique       4\n",
              "top       None\n",
              "freq       107\n",
              "Name: image_tags, dtype: object"
            ]
          },
          "execution_count": 140,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "image_df['image_tags'].describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "KtkwFihkL9Cu",
        "outputId": "9a682578-f3f9-4970-d29e-6fe044558ec9"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'None', 'document', 'handwriting', 'outfocus'}"
            ]
          },
          "execution_count": 141,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# check if there are image with multiple image tags (which would apper as e.g. \"document, outfocus\")\n",
        "set(image_df['image_tags'].tolist())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ROlVnxDVL9Cu",
        "outputId": "384ee963-0b77-464c-da94-c7881f4465bd"
      },
      "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>word_counts</th>\n",
              "      <th>image_width</th>\n",
              "      <th>image_height</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>150.00000</td>\n",
              "      <td>150.000000</td>\n",
              "      <td>150.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>17.52000</td>\n",
              "      <td>625.386667</td>\n",
              "      <td>618.033333</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>40.30633</td>\n",
              "      <td>359.637117</td>\n",
              "      <td>474.094169</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>1.00000</td>\n",
              "      <td>320.000000</td>\n",
              "      <td>191.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>3.00000</td>\n",
              "      <td>468.000000</td>\n",
              "      <td>453.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>6.50000</td>\n",
              "      <td>620.000000</td>\n",
              "      <td>512.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>14.00000</td>\n",
              "      <td>683.000000</td>\n",
              "      <td>751.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>422.00000</td>\n",
              "      <td>3264.000000</td>\n",
              "      <td>3281.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       word_counts  image_width  image_height\n",
              "count    150.00000   150.000000    150.000000\n",
              "mean      17.52000   625.386667    618.033333\n",
              "std       40.30633   359.637117    474.094169\n",
              "min        1.00000   320.000000    191.000000\n",
              "25%        3.00000   468.000000    453.000000\n",
              "50%        6.50000   620.000000    512.000000\n",
              "75%       14.00000   683.000000    751.000000\n",
              "max      422.00000  3264.000000   3281.000000"
            ]
          },
          "execution_count": 142,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "image_df.describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "cKJu9bOQL9Cu",
        "outputId": "04064846-a42b-4326-89ce-fca74423fb33"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "'AB_naver_0329.jpg'"
            ]
          },
          "execution_count": 143,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# just for checking\n",
        "list(data['images'].keys())[0]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Ew54X0ZwL9Cu",
        "outputId": "07f736ae-3c2b-4c1b-dd01-74be4ee16020"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "dict_keys(['img_h', 'img_w', 'words', 'tags', 'license_tag'])"
            ]
          },
          "execution_count": 144,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "first_img = list(data['images'].keys())[0]\n",
        "data[\"images\"][first_img].keys()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "DuHJ4j7rL9Cv",
        "outputId": "0e9c6daa-b6bd-448e-eaf2-f4cfebad5cc2"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/home/juhyun/anaconda3/envs/school/lib/python3.8/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.JointGrid at 0x7f2b01244bb0>"
            ]
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x432 with 3 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# image width, height \n",
        "g = sns.jointplot(\"image_width\", \"image_height\", data=image_df, kind=\"kde\", space=0, color=\"r\")\n",
        "g.set_axis_labels(\"Image Width\", \"Image Height\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lUAasu_WL9Cv"
      },
      "outputs": [],
      "source": [
        "def create_count_df(df, field, index):\n",
        "    count=df.groupby(field)[index].count().sort_values(ascending=False)\n",
        "    count_df = count.to_frame().reset_index()\n",
        "    count_df.columns = [field, field + '_count']\n",
        "    return count_df\n",
        "\n",
        "def plot_count_df(df, field, random_sample, color, rotation, xlabel, ylabel, title):\n",
        "    fig, ax = plt.subplots(figsize=(10,6))\n",
        "    if random_sample:\n",
        "        df = df.sample(n=50, random_state=1)\n",
        "    bars = ax.bar(df[field], df[field + '_count'], color=color, align='center',alpha=0.5)\n",
        "    for i, b in enumerate(bars):\n",
        "        ax.text(b.get_x()+b.get_width()*(1/2),b.get_height()+0.1, \\\n",
        "                df.iloc[i][field + '_count'],ha='center',fontsize=13)\n",
        "    ax.set_xticklabels(df[field],rotation=rotation, fontsize=13)\n",
        "    ax.set_xlabel(xlabel, fontsize=13)\n",
        "    ax.set_ylabel(ylabel, fontsize=13)\n",
        "    ax.set_title(title, fontsize=20)\n",
        "    plt.show()\n",
        "    "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "87mcouiRL9Cv",
        "outputId": "7c38a51d-a649-4317-8714-b38cfa3f54cb"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_118210/2687934052.py:15: UserWarning: FixedFormatter should only be used together with FixedLocator\n",
            "  ax.set_xticklabels(df[field],rotation=rotation, fontsize=13)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "img_tag_df = create_count_df(df=image_df, field='image_tags', index='image')\n",
        "plot_count_df(df=img_tag_df, field='image_tags', random_sample=False, color='g', rotation=0, xlabel='image tag', \\\n",
        "              ylabel='Number of image tag', title='Image Tag Distribution')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "4rPu6i9FL9Cv",
        "outputId": "2d0cc815-372c-4fa2-8aa4-775644f1c15b"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_118210/2687934052.py:15: UserWarning: FixedFormatter should only be used together with FixedLocator\n",
            "  ax.set_xticklabels(df[field],rotation=rotation, fontsize=13)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Total word tags \n",
        "word_tag_df = pd.DataFrame(word_tags, columns = ['word_tags'])\n",
        "word_tag_df['index'] = [i for i in range(len(word_tags))]\n",
        "word_tag_df = create_count_df(word_tag_df, field='word_tags',index='index')\n",
        "plot_count_df(df=word_tag_df, field='word_tags', random_sample=False, color='g', rotation=0, xlabel='word tags', \\\n",
        "              ylabel='Count of each word tag', title='Word tag Distribution')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-XCARHM-L9Cv",
        "outputId": "c6c332be-9c92-4121-b5b4-cd75ba5e02f0"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "2500"
            ]
          },
          "execution_count": 148,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "word_tag_df.iloc[0]['word_tags_count']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "CPGWzkbGL9Cw"
      },
      "outputs": [],
      "source": [
        "def plot_dist(df, field, bins, color, xlabel, ylabel, title):\n",
        "    sns.set(color_codes=True)\n",
        "    fig, ax = plt.subplots(figsize=(18,6))\n",
        "    sns.distplot(df[field], bins=bins, color=color, ax=ax)\n",
        "    \n",
        "    ax.set_xlabel(xlabel, fontsize=13)\n",
        "    ax.set_ylabel(ylabel, fontsize=13)\n",
        "    ax.set_title(title, fontsize=20)\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "GfyWtMeKL9Cw",
        "outputId": "0c0202c6-d40c-432f-8496-f1d3a70f1c77"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/home/juhyun/anaconda3/envs/school/lib/python3.8/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plot_dist(df=image_df, field='word_counts', bins=50, color='b', xlabel='number of words per Image', \\\n",
        "          ylabel='Frequency', title='Words per Image Distribution')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kSC9LAWXL9Cw"
      },
      "outputs": [],
      "source": [
        "word_df ={}\n",
        "word_df['index'] = [i for i in range(len(total_area))]\n",
        "word_df['orientation'] = orientation\n",
        "word_df['language'] = languages\n",
        "word_df['bbox_size'] = total_area"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gYs2m8x4L9Cw"
      },
      "outputs": [],
      "source": [
        "word_df = pd.DataFrame.from_dict(word_df)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "TQIQ3X3NL9Cw",
        "outputId": "91bd0c4c-b9fa-4fde-c08a-a9af1362d11a"
      },
      "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>index</th>\n",
              "      <th>orientation</th>\n",
              "      <th>language</th>\n",
              "      <th>bbox_size</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>0.011595</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>0.010657</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>0.012026</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>en</td>\n",
              "      <td>0.002800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>en</td>\n",
              "      <td>0.004921</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2623</th>\n",
              "      <td>2623</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>0.001041</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2624</th>\n",
              "      <td>2624</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>0.009787</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2625</th>\n",
              "      <td>2625</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>0.000909</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2626</th>\n",
              "      <td>2626</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>en</td>\n",
              "      <td>0.002246</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2627</th>\n",
              "      <td>2627</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>en</td>\n",
              "      <td>0.005971</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2628 rows × 4 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      index orientation language  bbox_size\n",
              "0         0  Horizontal       ko   0.011595\n",
              "1         1  Horizontal       ko   0.010657\n",
              "2         2  Horizontal       ko   0.012026\n",
              "3         3  Horizontal       en   0.002800\n",
              "4         4  Horizontal       en   0.004921\n",
              "...     ...         ...      ...        ...\n",
              "2623   2623  Horizontal       ko   0.001041\n",
              "2624   2624  Horizontal       ko   0.009787\n",
              "2625   2625  Horizontal       ko   0.000909\n",
              "2626   2626  Horizontal       en   0.002246\n",
              "2627   2627  Horizontal       en   0.005971\n",
              "\n",
              "[2628 rows x 4 columns]"
            ]
          },
          "execution_count": 153,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "word_df['language'] = word_df['language'].apply(lambda x: ','.join(map(str, x)))\n",
        "word_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "YBRAOke8L9Cw",
        "outputId": "1a316ea3-f4ed-4cb3-a701-c866e96076ee"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "count    2628.000000\n",
              "mean        0.004943\n",
              "std         0.018550\n",
              "min         0.000039\n",
              "25%         0.000531\n",
              "50%         0.001195\n",
              "75%         0.003259\n",
              "max         0.413370\n",
              "Name: bbox_size, dtype: float64"
            ]
          },
          "execution_count": 154,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "word_df.describe()['bbox_size']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "1fx0H_6YL9Cw",
        "outputId": "e30ec843-a795-4e84-91f7-97e278b60460"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Index(['index', 'orientation', 'language', 'bbox_size'], dtype='object')"
            ]
          },
          "execution_count": 155,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "word_df.columns"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "rl7ox33cL9Cw",
        "outputId": "a8d0a2bc-c42b-4600-aa5c-c9da49beb721"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_118210/2687934052.py:15: UserWarning: FixedFormatter should only be used together with FixedLocator\n",
            "  ax.set_xticklabels(df[field],rotation=rotation, fontsize=13)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "orientation = create_count_df(df=word_df, field='orientation', index='index')\n",
        "plot_count_df(df=orientation, field='orientation', random_sample=False, color='g', rotation=0, xlabel='orientation', \\\n",
        "              ylabel='Count of each orientation', title='Orientation Distribution')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6bmqpwAwL9Cw",
        "outputId": "0228b552-f197-4c92-e83d-e4688b514930"
      },
      "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>orientation</th>\n",
              "      <th>orientation_count</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Horizontal</td>\n",
              "      <td>2423</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Vertical</td>\n",
              "      <td>186</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Irregular</td>\n",
              "      <td>19</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  orientation  orientation_count\n",
              "0  Horizontal               2423\n",
              "1    Vertical                186\n",
              "2   Irregular                 19"
            ]
          },
          "execution_count": 233,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "orientation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "sTQ615PrL9Cw",
        "outputId": "42befbb7-92cb-4b1e-b3fc-2094f4000fb6"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_118210/2687934052.py:15: UserWarning: FixedFormatter should only be used together with FixedLocator\n",
            "  ax.set_xticklabels(df[field],rotation=rotation, fontsize=13)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# None means it's either a different language or a special symbol\n",
        "lang = create_count_df(df=word_df, field='language', index='index')\n",
        "plot_count_df(df=lang, field='language', random_sample=False, color='g', rotation=0, xlabel='language', \\\n",
        "              ylabel='Count of each language', title='Language Distribution')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6ZQP1V_nL9Cw",
        "outputId": "1abb0219-8472-4780-d6b1-5d938ebaaa1b"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/home/juhyun/anaconda3/envs/school/lib/python3.8/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABCoAAAGOCAYAAABR4OKVAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAA9WklEQVR4nO3deXxU1f3/8fcsSSaBhJCQQDaBxBJTkEUDiBYtIAUVxKUIXxSqfsFSC6WilLgAyipL67egiEvLT6xLpVaQYIm1SFm0IAKyVRFkCSQETAhJSDJJZu7vD5qpIxAGssxN5vV8PPIw95y7fDJwHsDbc861GIZhCAAAAAAAwASs/i4AAAAAAACgGkEFAAAAAAAwDYIKAAAAAABgGgQVAAAAAADANAgqAAAAAACAaRBUAAAAAAAA0yCoAAAAkJSamqqRI0f6uwwAAAKe3d8FAACAhpGamnpOW1BQkGJjY9W9e3c99NBDSklJ8erfvHmzRo0adc51DodDiYmJ6tOnj0aPHq3IyMjzPvPgwYNatmyZPv30U+Xl5ckwDLVu3Vq9evXSqFGjlJyc7DnX7XZr1KhR+uyzz/Tb3/5WgwYNOud+2dnZuv322xUUFKT3339fbdq0ucRPAQAAmJ3FMAzD30UAAID6Vx1UjBs3ztNWXFysnTt3avv27QoLC9Obb76ptLQ0T391UJGQkKA777xTkmQYhk6dOqUNGzYoOztb7dq101//+lc1a9bM63nLli3Ts88+K7fbre7du6tTp06SpD179mjLli2yWq3KyMjwCkKOHj2qIUOGyGazadWqVWrdurWnz+Vy6d5779X27dv13HPP6dZbb63Tz+fAgQMKDQ1VfHx8nd4XAABcGmZUAAAQYMaPH39O24wZM/SnP/1Jr732mp599tlz+hMSEs65rqKiQsOHD9eePXuUlZWlu+66y9O3YsUKzZo1S5GRkXr++efVvXt3r2u3bt2qX/7yl5o1a5YiIiJ0xx13SJISExP15JNP6vHHH1dGRob++Mc/ymKxSJJeeuklbd++XYMGDarzkELSObNJAACAf7BHBQAA0A033CBJKigo8Pma4OBg9ejR45zrSkpKNHv2bEnSggULzgkpJCk9PV3z58+XJM2ZM0clJSWevrvuukv9+/fXJ598otdee02StGvXLi1evFhxcXGaNm2azzVWVFRo2bJluvPOO9W9e3d16dJFffv21S9+8Qt98sknXud+f4+KzZs3KzU1tcavzZs3e93jwIEDysjI0E033aROnTrp+uuv16OPPqpvvvnG55oBAAh0zKgAAACef7RXL8/wRWVlpbZs2XLOdVlZWTp9+rQ6d+6s3r17X/D6G2+8UVdffbV27dqlrKws3X333Z6+6dOna/v27frd736na6+9VpMmTVJVVZWeffZZRURE+Fzj448/rszMTHXo0EFDhgyRw+HQiRMn9Pnnn2vDhg26/vrrL3htQkKC1zKZalVVVVq6dKmcTqdCQ0M97evXr9f48eNVVVWlPn366IorrlBeXp4+/PBDrVu3TsuWLVPHjh19rh0AgEBFUAEAQIBZtGiR5/uSkhLt2rVL27ZtU58+ffTggw+e95pjx455rjMMQ4WFhdq4caNycnL00EMP6brrrvOc+/nnn0tSjSFAtRtuuMHz/O8GFVFRUZo5c6bGjh2rESNGqKKiQg888IDXcy6muLhYq1evVseOHbV8+XLZbDav/lOnTtV4fWJi4nmXyWRkZMjpdOpnP/uZOnfuLEk6ffq0Hn30UTkcDr3xxhu68sorPefv27dPw4YN01NPPaX33nvP5/oBAAhUBBUAAASY559//py2K6+8UrfddpuaN29+3muOHTt23ut69+6tfv36ebWdPHlSknx6I0f1OSdOnDinr0+fPurRo4e2bNmi+Ph4TZw48aL3+y6LxSLDMBQcHCyr9dzVri1btryk+0lnP7v33ntP/fr1U0ZGhqd9xYoVKioq0tSpU71CCknq0KGDhg4dqtdee0379+8/px8AAHgjqAAAIMB89dVXnu9LS0u1f/9+LViwQI899pj279+vRx555JxrevTooddff91zfOrUKW3fvl0zZ87UfffdpxdeeEE33XRTndb56aef6rPPPpMkHT9+XF988cV597u4kObNm6tPnz76+OOPNWTIEP3kJz9Renq6unTp4rVkw1fvv/++Fi1apE6dOum3v/2tV/ixY8cOSdKXX37pNWOl2qFDhySd3cOCoAIAgJoRVAAAEMDCwsLUuXNnPf/887rpppv06quvavjw4YqLi6vxupYtW6pv375yOBx64IEHNHv2bE9Q0apVK0lnw4WLqT4nNjbWq72oqEiPP/647Ha7nnzySc2YMUOTJ0/W+++/f8FZH+fzf//3f3rllVeUmZnpCRBCQkI0YMAATZ482VPrxWzZskVPPPGEEhIS9NJLL50TdBQWFkqS3nnnnRrvU1pa6nPtAAAEKt76AQAAFBERofbt26uqqkp79uzx+brqPRoOHTqk4uJiSdK1114rSee8VeN8qs+55pprvNqfeeYZ5ebmavz48fqf//kfjRkzRseOHdOcOXN8rk2SHA6Hxo8fr6ysLK1bt07z58/Xtddeq/fff1+/+tWvfLrHgQMHNG7cODkcDr388svnDTfCw8MlSStXrtRXX311wa8777zzkuoHACAQEVQAAABJZzeElM5ulumroqIiz/dut1uSNHDgQEVERGjnzp3atGnTBa/dtGmTdu7cqcjISA0YMMDT/sEHHygzM1PXXHONxowZI0kaN26c0tLS9Je//EUff/zxJf1c1eLi4nT77bfrD3/4g9q2bavPP//8ohtqFhQU6Oc//7lKS0u1aNGiCy7b6NKli6T/biQKAAAuH0EFAADQRx99pKNHjyooKEjdunXz+bqlS5dKklJTU9WiRQtJZ/eGmDx5siTp0UcfPe8/3rdt26ZHH31U0tlXiFYv58jLy9PTTz+tsLAwzZs3z7MPRFBQkObNm6fg4GBNmTJFBQUFF62toKDAaz+OaqWlpSotLZXdbldQUNAFr3c6nfrFL36h7OxsPfPMM+rVq9cFz73rrrsUERGh559/Xjt37jyn3+12a/PmzRetGQAAsEcFAAAB57ubPZaWlurAgQNav369JOmRRx4579KG776eVDo7+2Lbtm3as2ePHA6Hpk6d6nX+T3/6UxUXF2v+/Pm699571aNHD3Xs2FEWi0V79uzR5s2bZbVa9cQTT+iOO+6QdHYmR0ZGhk6fPq2ZM2cqKSnJ654dOnTQhAkTNH/+fD399NNauHBhjT9nXl6e7rjjDnXo0EGpqamKi4tTSUmJ1q1bp5MnT2rkyJE17nexbNky7dixQ0lJScrJyTnvJpl33nmnEhMT1bJlSy1cuFC//OUvdc8996hXr1668sorZbFYdPz4cW3fvl2FhYXatWtXjTUDAADJYlzK/E4AANBopaamntNms9kUFRWlq6++Wvfdd59uuOEGr/7Nmzdr1KhR51wXFBSk1q1b67rrrtP//u//Kjk5+bzPPHDggJYtW6Z//etfno0z27Rpo+uuu06jRo1SSkqK59zXXntNs2fPVr9+/bR48eLz3s/tdmvkyJHaunWr5s2bpyFDhlzw5y0qKtKyZcu0ZcsWHTx4UKdOnVJkZKTat2+vYcOG6bbbbpPFYvH6fL77dpNFixad95Ws37Vs2TL17NnTc3z06FH98Y9/1MaNG5Wbm6ugoCDFxsbq6quv1oABA3TzzTfXeD8AAEBQAQAAAAAATIQ9KgAAAAAAgGkQVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANu78LqG+nTp2R280bWKtFRzdXfn6Jv8sAGjXGEVA7jCGgdhhDQO0whvzParWoZctmF+xv8kGF220QVHwPnwdQe4wjoHYYQ0DtMIaA2mEMmRtLPwAAAAAAgGkQVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAaBBUAAAAAAMA0CCoAAAAAAIBpEFQAAAAAAADTIKgAAAAAAACmQVABAAAAAABMg6ACAAAAAACYht3fBaBhVRaXyHbmtFebxeFQlS3ETxUBAAAAAPBfBBUBxlVWqvyt273aotO7Sc0IKgAAAAAA/sfSDwAAAAAAYBoEFQAAAAAAwDQIKgAAAAAAgGkQVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAaBBUAAAAAAMA0CCoAAAAAAIBp2BvqQQ8//LCOHj0qq9WqsLAwTZkyRWlpaTp48KAyMjJUWFioyMhIzZ07V+3atZOkGvsAAAAAAEDT02AzKubOnav3339fK1as0IMPPqgnnnhCkjRt2jSNGDFCWVlZGjFihKZOneq5pqY+AAAAAADQ9DRYUBEeHu75vqSkRBaLRfn5+dq7d68GDRokSRo0aJD27t2rgoKCGvsAAAAAAEDT1GBLPyTpySef1KZNm2QYhl599VXl5uaqdevWstlskiSbzabY2Fjl5ubKMIwL9kVFRTVk2QAAAAAAoIE0aFAxa9YsSdKKFSs0b948TZgwod6fGR3dvN6f0ZiUnyhTeHiIV1tYWLAcMeEXuALA+cQwZoBaYQwBtcMYAmqHMWRuDRpUVLvjjjs0depUtWnTRnl5eXK5XLLZbHK5XDpx4oTi4uJkGMYF+y5Ffn6J3G6jnn6SxidcUnGx06stuLRCxSeL/VMQ0AjFxITrJGMGuGyMIaB2GENA7TCG/M9qtdQ4qaBB9qg4c+aMcnNzPcdr165VixYtFB0drbS0NGVmZkqSMjMzlZaWpqioqBr7AAAAAABA09QgMyrKyso0YcIElZWVyWq1qkWLFlqyZIksFouefvppZWRkaPHixYqIiNDcuXM919XUBwAAAAAAmh6LYRhNel0ESz+8hRtlyl6/2astOr2bXM1a+KkioPFhuiBQO4whoHYYQ0DtMIb8zxRLPwAAAAAAAHxBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAaBBUAAAAAAMA0CCoAAAAAAIBpEFQAAAAAAADTIKgAAAAAAACmQVABAAAAAABMg6ACAAAAAACYBkEFAAAAAAAwDYIKAAAAAABgGgQVAAAAAADANAgqAAAAAACAaRBUAAAAAAAA0yCoAAAAAAAApkFQAQAAAAAATIOgAgAAAAAAmAZBBQAAAAAAMA2CCgAAAAAAYBoEFQAAAAAAwDQIKgAAAAAAgGkQVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAaBBUAAAAAAMA0CCoAAAAAAIBpEFQAAAAAAADTIKgAAAAAAACmQVABAAAAAABMg6ACAAAAAACYBkEFAAAAAAAwDYIKAAAAAABgGgQVAAAAAADANAgqAAAAAACAaRBUAAAAAAAA07A3xENOnTql3/zmNzpy5IiCg4PVtm1bTZ8+XVFRUUpNTVWHDh1ktZ7NTObNm6fU1FRJ0tq1azVv3jy5XC517NhRc+bMUWhoaEOUDAAAAAAA/KBBZlRYLBaNHj1aWVlZWrVqlZKSkrRgwQJP/9tvv62VK1dq5cqVnpDizJkzmjJlipYsWaK///3vatasmf7whz80RLkAAAAAAMBPGiSoiIyMVM+ePT3HXbt2VU5OTo3XrF+/Xp06dVK7du0kScOHD9ff/va3+iwTAAAAAAD4WYMs/fgut9utt956S3379vW0jRw5Ui6XSzfeeKPGjx+v4OBg5ebmKj4+3nNOfHy8cnNzG7pcAAAAAADQgBo8qJgxY4bCwsJ03333SZLWrVunuLg4lZSUaNKkSXrhhRf0yCOP1NnzoqOb19m9moLyE2UKDw/xagsLC5YjJtxPFQGNUwxjBqgVxhBQO4whoHYYQ+bWoEHF3LlzdfjwYS1ZssSzeWZcXJwkqXnz5ho6dKiWLl3qad+8ebPn2pycHM+5lyI/v0Rut1EH1TcN4ZKKi51ebcGlFSo+WeyfgoBGKCYmXCcZM8BlYwwBtcMYAmqHMeR/VqulxkkFDfZ60t/97nfavXu3XnjhBQUHB0uSTp8+rfLycklSVVWVsrKylJaWJknq3bu3du3apUOHDkk6u+HmLbfc0lDlAgAAAAAAP2iQGRVff/21XnrpJbVr107Dhw+XJCUmJmr06NGaOnWqLBaLqqqq1K1bN02YMEHS2RkW06dP189//nO53W6lpaXpySefbIhyAQAAAACAnzRIUPGDH/xAX3311Xn7Vq1adcHrbr75Zt188831VRYAAAAAADCZBlv6AQAAAAAAcDEEFQAAAAAAwDQIKgAAAAAAgGkQVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAaBBUAAAAAAMA0CCoAAAAAAIBpEFQAAAAAAADTIKgAAAAAAACmQVABAAAAAABMg6ACAAAAAACYBkEFAAAAAAAwDYIKAAAAAABgGgQVAAAAAADANAgqAAAAAACAaRBUAAAAAAAA0yCoAAAAAAAApkFQAQAAAAAATIOgAgAAAAAAmAZBBQAAAAAAMA2CCgAAAAAAYBoEFQAAAAAAwDQIKgAAAAAAgGkQVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAaBBUAAAAAAMA0CCoAAAAAAIBpEFQAAAAAAADTIKgAAAAAAACmQVABAAAAAABMg6ACAAAAAACYBkEFAAAAAAAwjQYJKk6dOqUxY8ZowIABGjx4sMaNG6eCggJJ0o4dO3T77bdrwIABevDBB5Wfn++5rqY+AAAAAADQ9DRIUGGxWDR69GhlZWVp1apVSkpK0oIFC+R2uzVp0iRNnTpVWVlZSk9P14IFCySpxj4AAAAAANA0NUhQERkZqZ49e3qOu3btqpycHO3evVshISFKT0+XJA0fPlxr1qyRpBr7AAAAAABA09Tge1S43W699dZb6tu3r3JzcxUfH+/pi4qKktvtVmFhYY19AAAAAACgabI39ANnzJihsLAw3Xffffr73/9e78+Ljm5e789oTMpPlCk8PMSrLSwsWI6YcD9VBDROMYwZoFYYQ0DtMIaA2mEMmVuDBhVz587V4cOHtWTJElmtVsXFxSknJ8fTX1BQIKvVqsjIyBr7LkV+foncbqOufoRGL1xScbHTqy24tELFJ4v9UxDQCMXEhOskYwa4bIwhoHYYQ0DtMIb8z2q11DipwOelHw888ID+9re/qbKy8rIK+d3vfqfdu3frhRdeUHBwsCSpU6dOKi8v19atWyVJb7/9tgYOHHjRPgAAAAAA0DRZDMPwabrB73//e7333ntyOp264447NHToUCUnJ/v0kK+//lqDBg1Su3bt5HA4JEmJiYl64YUXtG3bNk2bNk1Op1MJCQmaP3++WrVqJUk19vmKGRXewo0yZa/f7NUWnd5NrmYt/FQR0PiQwgO1wxgCaocxBNQOY8j/LjajwuegQpIMw9D69ev17rvvau3aterSpYvuueceDRw4UCEhIRe/gR8QVHgjqABqjz/cgNphDAG1wxgCaocx5H91tvRDkiwWi2666SYtXLhQa9asUVlZmSZPnqzevXtrwYIFKikpqXXBAAAAAAAgcF3y60l37typKVOmaMiQIQoKCtLs2bP1/PPPa//+/Ro7dmx91AgAAAAAAAKEz2/9eO211/SXv/xFx48f1+DBg/Xmm28qNTXV09+lSxf17NmzXooEAAAAAACBweegYvXq1br//vt12223eTbE/K6QkBDNnz+/TosDAAAAAACBxeeg4p133rnoOf37969VMQAAAAAAILD5vEfF448/rs8++8yrbcuWLXrqqafqvCgAAAAAABCYfA4q1q1bpy5duni1denSRWvXrq3zogAAAAAAQGDyOahwu92yWCxebRaLRZWVlXVeFAAAAAAACEw+BxVXXnmlVq9e7dW2Zs0apaSk1HlRAAAAAAAgMPm8meaECRM0evRorVu3Tu3atdPhw4e1du1avfzyy/VZHwAAAAAACCA+z6jo0aOH3nnnHbVs2VJ79+5VZGSk/vznP6tnz571WR8AAAAAAAggPs+okKSrrrpK06ZNq69aAAAAAABAgLukoCI3N1f//ve/debMGa/2wYMH12lRAAAAAAAgMPkcVPz5z3/WjBkzFB4ertDQUE+7xWIhqAAAAAAAAHXC56Bi8eLFeu6559S/f//6rAcAAAAAAAQwnzfTLC0tJaQAAAAAAAD1yuegYuDAgVq3bl09lgIAAAAAAAKdz0s/nE6nfv3rX+u6665TTEyMV9+MGTPqvDAAAAAAABB4fA4qbDabbrnlFklSVVVVvRUEAAAAAAACl89BxZw5c+qzDgAAAAAAAN/3qJCk4uJirVq1Sq+88ook6eTJk8rLy6uXwgAAAAAAQODxOajYvXu3+vfvr5dfflmLFy+WJH311VeaOXNmvRUHAAAAAAACi89BxezZszVp0iStWrVKdvvZFSPdunXTjh076qs2AAAAAAAQYHwOKvbv36+77rpLkmSxWCRJzZo1U1lZWf1UBgAAAAAAAo7PQUXLli2Vk5Pj1Xb48GG1bt26zosCAAAAAACByeeg4s4779TEiRO1detWGYah3bt366mnntLQoUPrsz4AAAAAABBAfH496ZgxY1RWVqaHHnpIpaWlGjVqlOcLAAAAAACgLvgcVNhsNj3yyCN65JFHVFBQoKioqPqsCwAAAAAABCCfl358FyEFAAAAAACoDz7PqOjYsaPnbR/ft3v37jorCAAAAAAABC6fg4qlS5d6Hefl5en//b//53llKQAAAAAAQG35HFT06NHjnLZu3bpp4sSJuvfee+u0KAAAAAAAEJgua4+Kaq1bt9b+/fvrqhYAAAAAABDgfJ5RsW3bNq/jsrIyvffee0pJSanzogAAAAAAQGDyOagYMWKE13FYWJg6duyo2bNn13lRAAAAAAAgMPkcVHz55Zf1WQcAAAAAAEDt9qgAAAAAAACoSz7PqBg5cqQsFstFz1u2bFmtCgIAAAAAAIHL56AiLS1Ny5cvV79+/ZSYmKhjx47po48+0j333KPIyMh6LBEAAAAAAAQKn4OKI0eO6Pnnn9cNN9zgafvkk0+0bNkyPf744/VSHAAAAAAACCw+71GxZcsW9erVy6utZ8+e2rJli0/Xz507V3379lVqaqr27dvnae/bt68GDhyoIUOGaMiQIdqwYYOnb8eOHbr99ts1YMAAPfjgg8rPz/e1XAAAAAAA0Aj5HFTExcXpgw8+8GrLyspSmzZtfLq+X79+euONN5SQkHBO38KFC7Vy5UqtXLlSvXv3liS53W5NmjRJU6dOVVZWltLT07VgwQJfywUAAAAAAI2Qz0s/Jk2apPHjx+vtt99WQkKCjh07pi+++EILFy706fr09PRLKmz37t0KCQnxXDd8+HD169dPc+bMuaT7AAAAAACAxsPnGRU//vGPtXr1av3oRz9SaGiobrjhBq1evVp9+vSpdRGPPfaYBg8erKefflpFRUWSpNzcXMXHx3vOiYqKktvtVmFhYa2fBwAAAAAAzMnnGRWSdMUVV2js2LF1WsAbb7yhuLg4VVRUaNasWZo+fXqdLvGIjm5eZ/dqCspPlCk8PMSrLSwsWI6YcD9VBDROMYwZoFYYQ0DtMIaA2mEMmdslBRXV+0icPHlSq1at0meffaZTp07pJz/5yWUXEBcXJ0kKDg7WiBEj9Itf/MLTnpOT4zmvoKBAVqv1kl+Fmp9fIrfbuOz6mppwScXFTq+24NIKFZ8s9k9BQCMUExOuk4wZ4LIxhoDaYQwBtcMY8j+r1VLjpAKfl34sXbpUixYtUu/evZWbmyvp7HKMV1999bKLKy0tVXHx2d8ghmHogw8+UFpamiSpU6dOKi8v19atWyVJb7/9tgYOHHjZzwIAAAAAAObn84yKt956S6+88orat2+vxYsXS5LatWunw4cP+3T9zJkz9eGHH+rbb7/VAw88oMjISC1ZskTjx4+Xy+WS2+1WSkqKpk2bJkmyWq2aN2+epk2bJqfTqYSEBM2fP/8yfkQAAAAAANBY+BxUnD59Wu3bt5ckWSwWSWdnQVR/fzFPPfWUnnrqqXPaV6xYccFrrrnmGq1atcrXEgEAAAAAQCPn89KPlJQUffzxx15tGzZs0FVXXVXnRQEAAAAAgMDk84yKiRMn6qGHHtItt9yiiooKzZgxQ6tXr9ZLL71Un/UBAAAAAIAA4vOMivT0dL3zzjtyOBzq2bOn3G63Xn/9dXXp0qU+6wMAAAAAAAHEpxkVVVVVevjhh7Vo0SJNmTKlvmsCAAAAAAAByqcZFXa7XXv27JHNZqvvegAAAAAAQADzeenH7bffrj/96U/1WQsAAAAAAAhwPm+muXfvXr3++ut68803FR8fL6v1vxnHH//4x3opDgAAAAAABJaLBhVTpkzRjBkz1L17d3Xv3l07duxQ165dG6A0AAAAAAAQaC4aVKxevVozZszQuHHjJEk9evTQq6++Wu+FAQAAAACAwHPRPSoMw6jxGAAAAAAAoK5cNKiwWCw1HgMAAAAAANSViy79qKys1JIlSzzHTqfT61iSxo4dW/eVAQAAAACAgHPRoKJr167atGmT57hz585exxaLhaACAAAAAADUiYsGFa+//npD1AEAAAAAAHDxPSoAAAAAAAAaCkEFAAAAAAAwDYIKAAAAAABgGgQVAAAAAADANAgqAAAAAACAaRBUAAAAAAAA0yCoAAAAAAAApkFQAQAAAAAATIOgAgAAAAAAmAZBBQAAAAAAMA2CCgAAAAAAYBoEFQAAAAAAwDQIKgAAAAAAgGkQVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAaBBUAAAAAAMA0CCoAAAAAAIBpEFQAAAAAAADTIKgAAAAAAACmQVABAAAAAABMg6ACAAAAAACYBkEFAAAAAAAwjQYJKubOnau+ffsqNTVV+/bt87QfPHhQw4YN04ABAzRs2DAdOnTIpz4AAAAAANA0NUhQ0a9fP73xxhtKSEjwap82bZpGjBihrKwsjRgxQlOnTvWpDwAAAAAANE0NElSkp6crLi7Oqy0/P1979+7VoEGDJEmDBg3S3r17VVBQUGMfAAAAAABouuz+enBubq5at24tm80mSbLZbIqNjVVubq4Mw7hgX1RUlL9KBgAAAAAA9cxvQUVDiY5u7u8STKX8RJnCw0O82sLCguWICfdTRUDjFMOYAWqFMQTUDmMIqB3GkLn5LaiIi4tTXl6eXC6XbDabXC6XTpw4obi4OBmGccG+S5WfXyK326iHn6BxCpdUXOz0agsurVDxyWL/FAQ0QjEx4TrJmAEuG2MIqB3GEFA7jCH/s1otNU4q8NvrSaOjo5WWlqbMzExJUmZmptLS0hQVFVVjHwAAAAAAaLoshmHU+3SDmTNn6sMPP9S3336rli1bKjIyUqtXr9aBAweUkZGhoqIiRUREaO7cuUpOTpakGvsuBTMqvIUbZcpev9mrLTq9m1zNWvipIqDxIYUHaocxBNQOYwioHcaQ/11sRkWDBBX+RFDhjaACqD3+cANqhzEE1A5jCKgdxpD/mXbpBwAAAAAAwPcRVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAadn8XAP+pzP9WVofD32UAAAAAAODBjIoA5S4vV+GHa1T0ySZ/lwIAAAAAgAdBRYAq3btbRlWVKo/nqrKgwN/lAAAAAAAgiaAiILnLy1X21ZcKat1GklT42Wd+rggAAAAAgLMIKgJQ9WyK8J7XKahNnE5v3izDMPxdFgAAAAAABBWBprK4WGVffamQdsmyt4iUI+VKVebnq+zrff4uDQAAAAAAgopAk5f1kYyqKjXr3FmS5Ei6QtaQEBV9stHPlQEAAAAAQFARUKqKi3Ty43We2RSSZAkKUnjXrir+7DO5nU7/FggAAAAACHgEFQHkVNYauSsrPbMpqkX27CnDWa6SbVv9VBkAAAAAAGcRVASIquIiFa79SFHd0z2zKaqFJicrKCZGpzex/AMAAAAA4F8EFQGi6JNNMioqFDfo1nP6LBaLIq7/kcq+/Lcq87/1Q3UAAAAAAJxFUBEgnIcPyx4VLUdcm/P2R/S6XpJU9OknDVkWAAAAAABeCCoChPPoEYUkJl6wP6hVjEKvSlPRpo0yDKMBKwMAAAAA4L8IKgKAu7JCFcePKyTpihrPi+h1gypPnpDz0MEGqgwAAAAAAG8EFQGgIidHcrsVkpRU43lhV6VJksoJKgAAAAAAfkJQEQCc2dmSpJDEmoMKe1SUrM2bq/zw4YYoCwAAAACAc9j9XQDqn/NotizBwQqKbS3JeU6/zSLpzGlJUmhCgioOM6MCAAAAAOAfBBUBwHk0WyEJibJYrdJ59sl0VziV/8VeSZJhD1J5To7clZWyBgU1cKUAAAAAgEDH0o8mzjAMObOPKLiGN358lz26leRyqSLnWD1XBgAAAADAuQgqmriqwkK5z5y56Bs/qtmjoiRJTvapAAAAAAD4AUFFE+fMPiLp4htpVrM1D5c1NFTlRwgqAAAAAAANj6Ciias4Wv3GD9+WflgsFjkSEuQ8cqgeqwIAAAAA4PwIKpo459Fs2aOjZQtr5vM1jsREObOzZbhc9VgZAAAAAADnIqho4pzZ2T7vT1HNkZQko7JSFcdz66kqAAAAAADOj6CiCXNXVKjieK7Pyz6qOf5zPhtqAgAAAAAaGkFFE1aRkyMZhkISL21GRXBsrCzBwSpnnwoAAAAAQAMjqGjCnEf/88aPJN/e+FHNYrUqJOkKZlQAAAAAABocQUUT5szOliU4WEExsZd8raNtW5UfOSLD7a6HygAAAAAAOD+CiibMeTRbIYmJslgv/Zc55Ip2MpzlqjyRVw+VAQAAAABwfgQVTZRhGGff+HGJ+1NUc7RtK0kqZ/kHAAAAAKABEVQ0UVWnTsldeuaS3/hRLTguXha7XU421AQAAAAANCCCiibqvxtpXt6MCovdruDEJGZUAAAAAAAaFEFFE+XMzpYkBSdc3owKSXJc0VbOI4dlGEZdlQUAAAAAQI3s/i5Akvr27avg4GCFhIRIkh577DH17t1bO3bs0NSpU+V0OpWQkKD58+crOjraz9U2DhVHs2Vv1Uq2sLDLvkdI27Y6vX6dqr79VkExMXVYHQAAAAAA52eKoEKSFi5cqA4dOniO3W63Jk2apDlz5ig9PV2LFy/WggULNGfOHD9W2Xic3UgzqVb3cLRtJ0kqP3KIoAIAAAAA0CBMu/Rj9+7dCgkJUXp6uiRp+PDhWrNmjZ+rahzcFRWqyDt+2ftTVAtOSJBsNjnZpwIAAAAA0EBMM6Pisccek2EYuvbaazVx4kTl5uYqPj7e0x8VFSW3263CwkJFRkb6fN/o6Ob1UK25FX+9XzIMxfzwB2oVE+7VV36iTOHhIV5tQUE2r7awsGA5/nNd7hVJcuceVcz37gMEOsYEUDuMIaB2GENA7TCGzM0UQcUbb7yhuLg4VVRUaNasWZo+fbr69+9fJ/fOzy+R2x1Ym0Ge3vWlJMkZEaOTJ4u9+sIlFRc7vdpCKl1ebcGlFSr+z3W2hCtUtP1znThRJIvFUr+FA41ETEz4OWMLgO8YQ0DtMIaA2mEM+Z/VaqlxUoEpln7ExcVJkoKDgzVixAht27ZNcXFxysnJ8ZxTUFAgq9V6SbMpApUzO1uWkJA62VciNDlF7jNnVHkirw4qAwAAAACgZn4PKkpLS1VcfDbNMgxDH3zwgdLS0tSpUyeVl5dr69atkqS3335bAwcO9GepjYbzaLZCEhJlsdb+l9eRkiJJKv/mQK3vBQAAAADAxfh96Ud+fr7Gjx8vl8slt9utlJQUTZs2TVarVfPmzdO0adO8Xk+KmhmGIefRbIV371En9wuOi5fV4VDZNwcU0euGOrknAAAAAAAX4vegIikpSStWrDhv3zXXXKNVq1Y1bEGNXFVBgdylpQpJrN0bP6pZrFY52ier/AAzKgAAAAAA9c/vSz9Qt5xHsyVJIYlJdXZPR3KKnEez5XY6L34yAAAAAAC1QFDRxDizj0iSQpIS6+yejuQUye1W+eFDdXZPAAAAAADOh6CiiXEezVZQTIysjtA6u6cjOVkSG2oCAAAAAOofQUUT4zyareA6XPYhSfbwCAXFxBJUAAAAAADqHUFFE+J2OlWZl1en+1NUcySnqOzAARmGUef3BgAAAACgGkFFE+I8dkwyDIUk1c0bP77LkZIi1+lCVZ0qqPN7AwAAAABQjaCiCamohzd+VAtNvlKSeE0pAAAAAKBeEVQ0Ic6jR2QJcSioVas6v3dIYqIsQUEqY58KAAAAAEA9IqhoQpzZ2WcDBWvd/7Ja7HY52rVnQ00AAAAAQL0iqGgiDMOQ82h2vSz7qOZITpbz8CG5Kyvr7RkAAAAAgMBGUNFEVBXky11WppCk+gwqUmRUVcmZnV1vzwAAAAAABDaCiiaiOjyojzd+VHNUb6jJ8g8AAAAAQD0hqGginNVv/EhIqLdnBLVsKXvLKIIKAAAAAEC9IahoIpzZRxQUEyurI7Ren+NITiaoAAAAAADUG4KKJsJ5NLte96eo5khOUeW3J1V1+nS9PwsAAAAAEHgIKpoAt9OpyhMn6uyNHzaLZDtz2vNldzk9faGefSr218mzAAAAAAD4Lru/C0DtOY8dlQyjzmZUuCucyv9ir+c4Or2b1CxEkhTSrp2soaEq2b5NzbtdWyfPAwAAAACgGjMqmgDPGz8S6++NH9WsQUFqfm13FX/+udxO58UvAAAAAADgEhBUNAHO7COyOhyyR0c3yPMiruslw1mukh3bG+R5AAAAAIDAQVDRBJT+e48cV3aQxdowv5yhHVJlj4pS8b8+aZDnAQAAAAACB0FFI1dx/Lgq8/LUvHPnBnumxWpVeM9eOrNnt6qKihrsuQAAAACApo+gopE7s/MLSVKzzl0a9LkR110vud0q/mxzgz4XAAAAANC0EVQ0cmd2faHg+HgFtYpp0OeGJCQoJOkKFf/r0wZ9LgAAAACgaSOoaMTc5WUq3feVml3dsLMpqoVf10vlB79RxfHjfnk+AAAAAKDpIahoxM7s2SO5XGrWpatfnh/R8zrJYlERm2oCAAAAAOoIQUUjdmbXF7KGhSk05Uq/PN8e2VJhV/1Qxf/6VIZh+KUGAAAAAEDTQlDRSBlut87s/ELNOnaSxWbzWx3h1/VS5bcnVX5gv99qAAAAAAA0HQQVjZTzyGG5iooa/G0f3xd+7bWyBAeriE01AQAAAAB1gKCikTqz8wvJYlFYp6vr/Vk2i2Q7c9rry+5ySpKsjlA173qNirf8S1WnT9d7LQAAAACApo2gopEq2fmFHMkpsodH1Puz3BVO5W/d7vVllJd7+qNuGyyjslLH//CyDLe73usBAAAAADRdBBWNUNXp03IeOqhmV3f2dymSpJCEBMX8z70q3btHp9Z84O9yAAAAAACNGEFFI3Rm905J8vv+FN/VovdNCu/RU9+u+KvKvt7n73IAAAAAAI0UQUUjdGbnF7K3bKmQpCv8XYqHxWJR7Mj7FRTdSrkvL5GrpMTfJQEAAAAAGiGCikbGqKpS6Z7danZ1Z1ksFn+X48UWGqq4sQ/LVVyk40tflWEY/i4JAAAAANDIEFQ0Mqc3rpe7vFzNOnf1ax3ffxNI9VtAHG3bqdVPh+nMFzt08q0/ye10+rVOAAAAAEDjYvd3AfBdxfFcnXznbYWldfT7/hTuCqfyv9jrOY5O7yY1C5EkRfa7WZUnjqtw7T90ZudOxd43Us06mWPjTwAAAACAuTGjopEwqqqU+8pLsgQFqc3/jpbFat5fOovFotgRI5X4m8dlsdt17P9+p9yXX1TV6UJ/lwYAAAAAMDlmVDQS+e+vkPPwIcX9YpzskS39Xc45bBZJZ07/9zjYLltCGzWb9JjyP/pI+R9+qDM7v1Dz9B6K6HmdQlOvMnXYAgAAAADwD4KKRqB031cq+NtqRfyot8KvTfd3Oef1/aUgrbr80HNsbdVaLW+7Xe7coyr6bLOKNq6XLSJCLdLTFXZNdzmSU2QNDvZX6QAAAAAAEzF9UHHw4EFlZGSosLBQkZGRmjt3rtq1a+fvshqMq/SMjr/6soJaxSh2+L3+Luey2Vu0UKsbeynoylQ5jx2V89BBnVq/XgVr10o2mxzt2iv0Bx0U+oMOCrmireyRkaZ7qwkAAAAAoP6ZPqiYNm2aRowYoSFDhmjlypWaOnWqli1b5u+y6p1hGCr76kvlr1qpqsJTSsp4UlaHw99l1ZrFbpejbTs52rZTZKc0lRzLVdm+fSr7ep9O/T1Lp9Z8IEmyhoYqJC5OjsRE2eMTZW8ZpaCoKNmjomQLj2DZCAAAAAA0UaYOKvLz87V3714tXbpUkjRo0CDNmDFDBQUFioqK8nN19cNVekZFn3yi0//8WBW5ObKGNVPsvaMUmpzi79LqXHCoQy1S2qtFSnvplgFSeZly//FPVRUWyFVYqKrCUzr92Wdyl633us5it8varLlsYWGyhoWd/W9oqCx2uyz2IFnstv/81+71JbtdFqtVFptNstpksVllsdqk7/3XYrNJNtvZMMRqO3s/23/uE2T/772rv2fmBwAAAADUGVMHFbm5uWrdurVsNpskyWazKTY2Vrm5uT4HFVZr4/lHZEVujnIWL5JRWSlH0hWKuesuNevcVdagoDp7hsWwyh4W6tVmtdu92i52XFfXyO3S6X9/7TmMSvuBwtq3k9TO0xbd+YeqOFOmqtOnVVVUJPeZYlUWFMpVXiZXebkMp1Ou0jK5i07L7XLJcFVJVS65XVUyqlyS4b6Uj+fyWG1ngwurRZJVslrOhhcWSZazx5JFslhkqT62VJ9jOXuOxSKLRWePrdb/9F2gXdXXWc4+03r2vHPaLd95pixn67J+75n/afeu5xLGzCWca1E9jcVLqrduHukMDVZ5WcXl1VCXDMM/z70UdV1iI/iZjTr/oc9V43j6/u/Hi/329MPv34rQYJV9dwyZnCkDaTPWZDZN+DOqrMsxdN7PycfPrrZ/BjfhXyOYm6tZiM6ccfq7jDpjbx6uZl27NapZ5xf7d7qpg4q60LJlM3+X4LvoDopbvKieH9Jc7W+9+ZzWFu3bXtJxQ14DAAAAAAgcpo5c4uLilJeXJ5fLJUlyuVw6ceKE4uLi/FwZAAAAAACoD6YOKqKjo5WWlqbMzExJUmZmptLS0prs/hQAAAAAAAQ6i2GYe8HvgQMHlJGRoaKiIkVERGju3LlKTk72d1kAAAAAAKAemD6oAAAAAAAAgcPUSz8AAAAAAEBgIagAAAAAAACmQVABAAAAAABMg6ACAAAAAACYBkEFAAAAAAAwDYKKJujgwYMaNmyYBgwYoGHDhunQoUPnnONyufTMM8/o5ptvVv/+/bV8+fKGLxQwMV/G0caNG3XXXXepU6dOmjt3bsMXCZiYL2PohRde0G233abBgwfrrrvu0oYNGxq+UMCkfBlD7777rgYPHqwhQ4Zo8ODBWrZsWcMXCpiUL2Oo2jfffKMuXbrw9zkT4fWkTdCoUaN09913a8iQIVq5cqXefffdc/7gWrFihVatWqVXXnlFhYWFuuOOO/Tmm28qMTHRT1UD5uLLODp8+LBKS0u1Zs0aVVRUaPLkyX6qFjAfX8bQhg0blJ6ertDQUH355Ze67777tHHjRjkcDj9VDZiHL2OopKREzZo1k8ViUUlJiQYPHqwXX3xRV111lZ+qBszDlzEknf0fuPfff79iY2MVGxvL3+dMghkVTUx+fr727t2rQYMGSZIGDRqkvXv3qqCgwOu8Dz74QEOHDpXValVUVJRuvvlmrVmzxh8lA6bj6zhq27at0tLSZLfb/VEmYFq+jqHevXsrNDRUkpSamirDMFRYWNjQ5QKm4+sYat68uSwWiySpvLxclZWVnmMgkPk6hiTp5Zdf1o9//GO1a9eugatETQgqmpjc3Fy1bt1aNptNkmSz2RQbG6vc3NxzzouPj/ccx8XF6fjx4w1aK2BWvo4jAOd3OWNoxYoVuuKKK9SmTZuGKhMwrUsZQ//4xz902223qU+fPho9erRSU1MbulzAdHwdQ19++aU2btyo+++/3w9VoiYEFQAAwK+2bNmi3//+9/rtb3/r71KARqdfv35avXq1srKytHLlSn3zzTf+LgloFCorKzVlyhQ988wznkAD5sF85SYmLi5OeXl5crlcstlscrlcOnHihOLi4s45LycnR507d5Z07gwLIJD5Oo4AnN+ljKHt27dr0qRJWrx4sZKTk/1QLWA+l/PnUHx8vK6++mqtW7eOsYSA58sYOnnypI4cOaKHHnpIklRUVCTDMFRSUqIZM2b4q3T8BzMqmpjo6GilpaUpMzNTkpSZmam0tDRFRUV5nTdw4EAtX75cbrdbBQUF+uijjzRgwAB/lAyYjq/jCMD5+TqGdu7cqUceeUQLFy5Ux44d/VEqYEq+jqEDBw54vi8oKNDmzZvVoUOHBq0VMCNfxlB8fLw2b96stWvXau3atfrZz36me+65h5DCJHjrRxN04MABZWRkqKioSBEREZo7d66Sk5M1ZswY/epXv9LVV18tl8ul6dOna9OmTZKkMWPGaNiwYX6uHDAPX8bR1q1bNXHiRJWUlMgwDIWHh2vWrFnq3bu3v8sH/M6XMXT33Xfr2LFjat26tee6efPmscYekG9jaPbs2dq0aZPsdrsMw9DQoUM1cuRIf5cOmIIvY+i7Fi1apNLSUt76YRIEFQAAAAAAwDRY+gEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADANggoAAAAAAGAaBBUAAAAAAMA0CCoAAECTlpOTo27duikvL8/fpQAAAB/welIAAOCTkSNHavv27QoKCpIktWrVSvfee6/uv/9+zzl9+/bVyZMnZbfbJUnx8fEaO3asBg8e7DnH6XTqxRdf1OrVq5WXl6fmzZvrxhtv1IQJExQXF6fTp09r8ODBGjNmjEaOHOm5bvbs2dq6dav+/Oc/e2oAAABNj93fBQAAgMbj4Ycf1sMPPyxJ2rFjh+6//3794Ac/0A033OA5Z+bMmRoyZIgMw9DatWs1fvx4/fCHP1RKSopcLpceeughffvtt1qwYIE6duyo3NxczZ07V0OHDtW7776r1q1b69lnn9XDDz+s66+/XikpKdq4caOWL1+ud999l5ACAIAmjqUfAADgsnTt2lUpKSnat2/fefstFov69euniIgIHThwQJKUmZmpzz//XIsXL1aXLl1kt9uVlJSk5557Ts2aNdPChQslSddff73uuecePfbYYzpx4oQyMjI0adIkJScnn/dZhmHoueee049+9CN169ZNffv21euvvy5JOnr0qFJTU3X8+HFJUs+ePdWtWzfPV1pammbNmiVJqqqq0pIlSzRgwAClp6dr+PDh2rVrV51+bgAAoGbMqAAAAJfMMAxt27ZN33zzjbp27Xrec1wul9auXauysjJ16tRJkrR+/Xp17txZbdu29To3KChIt9xyi959911P22OPPaa7775bQ4YMUefOnTVixIgL1rNp0ya99957Wr58ueLi4pSfn3/BPSk2b97s+f5f//qXxo0bp1tvvVWStGjRIn3yySd69dVXFR8fr7/+9a8aPXq0PvzwQ7Vo0cKnzwYAANQOMyoAAIDPlixZovT0dHXt2lUjRozQ4MGD1blzZ69zpk2b5jnnV7/6lcaMGaP4+HhJUkFBgVq3bn3ee8fGxqqgoMBzHBwcrG7duqmgoEA//elPa6wrKChITqdTX3/9tZxOp6Kjo/XDH/6wxmv27dunCRMmaM6cOerWrZsMw9CyZcv0m9/8RklJSbLZbBo6dKhiY2O1bt06Hz4dAABQFwgqAACAz8aOHautW7fqiy++0D//+U8dOHBATzzxhNc5zzzzjLZu3apdu3bpgw8+UGZmpmdJR1RU1AVnOpw4cUItW7b0HG/cuFGrVq3SyJEjNWfOHJWUlFywrp49e2rixIl68cUX1atXLz344IM1LtnIy8vTmDFjNG7cOPXv31+SdOrUKZWWlmrs2LFKT0/3fGVnZ/PGEAAAGhBBBQAAuCxt2rTRLbfcog8//PCC57Rv3159+vTR3//+d0lS7969tXPnTh05csTrvKqqKv3tb3/TjTfeKOlsaJCRkaHHH39cTz75pJKSkjR9+vQa6xk2bJjeeustbdq0SVdddZXGjx9/3vNKSko0ZswY3XrrrV5vFWnZsqXCwsK0dOlSbd261fO1Y8cOPfTQQz59JgAAoPYIKgAAwGU5efKk1qxZo6uuuuqC52RnZ2vdunVKTU2VJA0aNEhdu3bVL3/5S+3cuVMul0tHjx7VxIkTVVJS4gkXpk6dqs6dO+uee+6RxWLRs88+q48//lhZWVnnfc7OnTu1detWVVRUKDg4WM2aNZPVeu5fc6qqqjR+/HilpKToN7/5jVefxWLRqFGjNG/ePB06dEiSdObMGW3YsIEZFQAANCCLYRiGv4sAAADmN3LkSG3fvt3zetDQ0FB1795dkydP9uxB0bdvX508eVJ2u12GYSg8PFw33nijJk+erIiICElSeXm5XnzxRWVmZurEiRNq3ry5evfurV//+teKj4/XX/7yFz333HNatWqVoqKiPM/PzMzUjBkztGrVKsXGxnrV9umnn3oCBpvNpg4dOigjI0OdO3fW0aNH1a9fP/3zn/9UVVWV+vXrJ4fD4RVkjBgxQpMmTVJVVZVef/11LV++XMePH1dYWJi6dOmiKVOmqE2bNvX9EQMAABFUAAAAAAAAE2HpBwAAAAAAMA2CCgAAAAAAYBoEFQAAAAAAwDQIKgAAAAAAgGkQVAAAAAAAANMgqAAAAAAAAKZBUAEAAAAAAEyDoAIAAAAAAJgGQQUAAAAAADCN/w8U052e05HlcwAAAABJRU5ErkJggg==",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plot_dist(df=word_df, field='bbox_size', bins=200, color='r', xlabel='BBOX size', \\\n",
        "          ylabel='Frequency', title='BBOX size')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "S5BdAj4wL9Cw"
      },
      "outputs": [],
      "source": [
        "def plot_dist_list(target_list, bins, color, xlabel, ylabel, title):\n",
        "    sns.set(color_codes=True)\n",
        "    fig, ax = plt.subplots(figsize=(18,6))\n",
        "    sns.distplot(target_list, bins=bins, color=color, ax=ax)\n",
        "    ax.set_xlabel(xlabel, fontsize=13)\n",
        "    ax.set_ylabel(ylabel, fontsize=13)\n",
        "    ax.set_title(title, fontsize=20)\n",
        "    plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gU8dLwLhL9Cw",
        "outputId": "ab8bcc0a-9d61-4966-ef18-fc4dbf817f94"
      },
      "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>aspect ratio</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>186.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.220503</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.158984</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.080321</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.153336</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.192214</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.241817</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>1.882353</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       aspect ratio\n",
              "count    186.000000\n",
              "mean       0.220503\n",
              "std        0.158984\n",
              "min        0.080321\n",
              "25%        0.153336\n",
              "50%        0.192214\n",
              "75%        0.241817\n",
              "max        1.882353"
            ]
          },
          "execution_count": 160,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "pd.DataFrame(ver_aspect_ratio, columns=['aspect ratio']).describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fI7YcqtZL9Cx",
        "outputId": "bc3325e7-7a0d-4351-e16f-e0e60dd7754d"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/home/juhyun/anaconda3/envs/school/lib/python3.8/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# 한 두글자짜리 단어나 숫자의 경우 aspect ratio가 1미만인 경우가 있다\n",
        "plot_dist_list(hor_aspect_ratio, bins=20, color='r', xlabel='Aspect Ratio (BBOX Width / BBOX Height)', \\\n",
        "          ylabel='Frequency', title='Aspect Ratio Distribution (Horizontal words)')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "uMylTv-tL9Cx",
        "outputId": "21c54831-ff69-42ae-e42f-9bc5d186b471"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/home/juhyun/anaconda3/envs/school/lib/python3.8/site-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# aspect ratio가 1.88정도 되는 경우가 하나 있음. 이미지에서 글자가 왜곡되어 있었고 박스가 약간 널널하게 잡힌 케이스. \n",
        "plot_dist_list(ver_aspect_ratio, bins=20, color='r', xlabel='Aspect Ratio (BBOX Width / BBOX Height)', \\\n",
        "          ylabel='Frequency', title='Aspect Ratio Distribution (Veritcal words)')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "j0L_YEPTL9Cx"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "yjdP8vmYL9Cx"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "J-54Pb_6L9Cx"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "name": "OCR_EDA_revised.ipynb",
      "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": "3.8.11"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}