data_eda.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "fwoxMwWLnEyI"
      },
      "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": 2,
      "metadata": {
        "id": "_eCzdzcYnEyI"
      },
      "outputs": [],
      "source": [
        "from IPython.core.interactiveshell import InteractiveShell\n",
        "InteractiveShell.ast_node_interactivity = \"all\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "JHcbJ8gpnEyI"
      },
      "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": 4,
      "metadata": {
        "id": "5zltUmUxnEyI"
      },
      "outputs": [],
      "source": [
        "# ! gt file과 raw image file path를 이용해 데이터를 불러온다. \n",
        "# ! train.json 파일 경로를 사용하도록 수정 필요.\n",
        "data = read_json(\"/opt/ml/input/data/medical/ufo/train.json\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "3YY8SzYnnEyJ"
      },
      "outputs": [],
      "source": [
        "def get_box_size(quads):\n",
        "    \"\"\" 단어 영역의 사각형 좌표가 주어졌을 때 가로, 세로 길이를 계산해주는 함수.\n",
        "    TODO: 각 변의 길이를 단순히 max로 처리하기 때문에 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) heights, widths into separate dim\n",
        "    return np.rint(dists.mean(axis=-1)).astype(int)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "XbncN2oCnEyJ"
      },
      "outputs": [],
      "source": [
        "def rectify_poly(poly, direction):\n",
        "    \"\"\"일반 polygon 형태인 라벨을 크롭하고 rectify한 형태의 가로, 세로 길이를 계산해주는 함수.\n",
        "    Args:\n",
        "        poly: np.ndarray(2n+4, 2) (where n>0), 4, 6, 8\n",
        "        direction: 글자의 읽는 방향과 진행 방향의 수평(Horizontal) 혹은 수직(Vertical) 여부\n",
        "    Return:\n",
        "        rectified: np.ndarray(2, ?) rectify된 단어 bbox의 가로, 세로 길이.\n",
        "    \"\"\"\n",
        "    \n",
        "    n_pts = poly.shape[0]\n",
        "    assert n_pts % 2 == 0\n",
        "    if direction not in ['Horizontal', 'Vertical', 'Irregular']:\n",
        "        return None\n",
        "    if n_pts == 4:\n",
        "        size = get_box_size(poly[None])\n",
        "        h = size[:, 0]\n",
        "        w = size[:, 1]\n",
        "        return np.stack((h,w))\n",
        "\n",
        "    def unroll(indices):\n",
        "        return list(zip(indices[:-1], indices[1:]))\n",
        "\n",
        "    # ! polygon 하나를 인접한 사각형 여러개로 쪼갬\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 in ['Horizontal', 'Irregular']:\n",
        "        h = sizes[:, 0].max()\n",
        "        h = np.rint(h).astype(np.int32)\n",
        "        resize_ratios = h / sizes[:, 0]\n",
        "        widths = np.rint(sizes[:, 1] * resize_ratios).astype(np.int32)      # widths = sizes[:, 1]\n",
        "        w = np.sum(widths)\n",
        "            \n",
        "        return np.vstack((h,w))\n",
        "    \n",
        "    elif direction == 'Vertical':\n",
        "        w = sizes[:, 1].max()\n",
        "        w = np.rint(w).astype(np.int32)\n",
        "        resize_ratios = w / sizes[:, 1]\n",
        "        heights = np.rint(sizes[:, 0] * resize_ratios).astype(np.int32)\n",
        "        h = np.sum(heights)\n",
        "        return np.vstack((h,w))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "e5qzWpR5nEyK"
      },
      "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_height = []\n",
        "word_width = []\n",
        "bbox_size = []\n",
        "word_tags = []\n",
        "aspect_ratio = []\n",
        "ver_string = []\n",
        "\n",
        "for image_key, image_value in list(data[\"images\"].items())[:]:\n",
        "\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']] \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] \n",
        "            if word['tags'] != None:\n",
        "                word_tags.extend(word['tags'][:])\n",
        "            elif word['tags']== None:\n",
        "                word_tags.append('None')\n",
        "            poly = np.int32(word['points'])\n",
        "            size = rectify_poly(poly, word['orientation'])\n",
        "            if word['orientation'] == 'Horizontal':\n",
        "                hor_sizes.append(size)\n",
        "                # word_height.append(size[0])\n",
        "                # word_width.append(size[1])\n",
        "                bbox_size.append(np.sum(size[0]* size[1]))\n",
        "            elif word['orientation'] == 'Vertical':\n",
        "                ver_sizes.append(size)\n",
        "                # word_height.append(size[0])\n",
        "                # word_width.append(size[1])\n",
        "                bbox_size.append(np.sum(size[0]* size[1]))\n",
        "            else:\n",
        "                irr_sizes.append(size)\n",
        "                # word_height.append(size[0])\n",
        "                # word_width.append(size[1])\n",
        "                bbox_size.append(np.sum(size[0]* size[1]))\n",
        "        else:\n",
        "            count_ill += 1\n",
        "    df['word_counts'].append(len(word_ann)-count_ill)\n",
        "    \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": 8,
      "metadata": {
        "id": "y1LY32FIFcOs",
        "outputId": "5881d3ad-e85d-4afa-8521-0a93682118df"
      },
      "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>drp.en_ko.in_house.deepnatural_002411.jpg</td>\n",
              "      <td>375</td>\n",
              "      <td>1884</td>\n",
              "      <td>2448</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_002418.jpg</td>\n",
              "      <td>207</td>\n",
              "      <td>2481</td>\n",
              "      <td>3507</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_002432.jpg</td>\n",
              "      <td>364</td>\n",
              "      <td>3024</td>\n",
              "      <td>4032</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_002441.jpg</td>\n",
              "      <td>368</td>\n",
              "      <td>3120</td>\n",
              "      <td>4160</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_002451.jpg</td>\n",
              "      <td>364</td>\n",
              "      <td>2480</td>\n",
              "      <td>3507</td>\n",
              "      <td>[autoannotated]</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>95</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003932.jpg</td>\n",
              "      <td>238</td>\n",
              "      <td>2481</td>\n",
              "      <td>3507</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>96</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003951.jpg</td>\n",
              "      <td>464</td>\n",
              "      <td>2488</td>\n",
              "      <td>2304</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>97</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003958.jpg</td>\n",
              "      <td>259</td>\n",
              "      <td>1440</td>\n",
              "      <td>1440</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003976.jpg</td>\n",
              "      <td>250</td>\n",
              "      <td>1080</td>\n",
              "      <td>1440</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003979.jpg</td>\n",
              "      <td>372</td>\n",
              "      <td>1700</td>\n",
              "      <td>2338</td>\n",
              "      <td>[autoannotated]</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>100 rows × 5 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                        image  word_counts  image_width   \n",
              "0   drp.en_ko.in_house.deepnatural_002411.jpg          375         1884  \\\n",
              "1   drp.en_ko.in_house.deepnatural_002418.jpg          207         2481   \n",
              "2   drp.en_ko.in_house.deepnatural_002432.jpg          364         3024   \n",
              "3   drp.en_ko.in_house.deepnatural_002441.jpg          368         3120   \n",
              "4   drp.en_ko.in_house.deepnatural_002451.jpg          364         2480   \n",
              "..                                        ...          ...          ...   \n",
              "95  drp.en_ko.in_house.deepnatural_003932.jpg          238         2481   \n",
              "96  drp.en_ko.in_house.deepnatural_003951.jpg          464         2488   \n",
              "97  drp.en_ko.in_house.deepnatural_003958.jpg          259         1440   \n",
              "98  drp.en_ko.in_house.deepnatural_003976.jpg          250         1080   \n",
              "99  drp.en_ko.in_house.deepnatural_003979.jpg          372         1700   \n",
              "\n",
              "    image_height       image_tags  \n",
              "0           2448  [autoannotated]  \n",
              "1           3507  [autoannotated]  \n",
              "2           4032  [autoannotated]  \n",
              "3           4160  [autoannotated]  \n",
              "4           3507  [autoannotated]  \n",
              "..           ...              ...  \n",
              "95          3507  [autoannotated]  \n",
              "96          2304  [autoannotated]  \n",
              "97          1440  [autoannotated]  \n",
              "98          1440  [autoannotated]  \n",
              "99          2338  [autoannotated]  \n",
              "\n",
              "[100 rows x 5 columns]"
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "image_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "EYlvgOE1nEyK",
        "outputId": "9fcff85c-a568-4794-b2b9-f41caa8ed35d"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "0     None\n",
              "1     None\n",
              "2     None\n",
              "3     None\n",
              "4     None\n",
              "      ... \n",
              "95    None\n",
              "96    None\n",
              "97    None\n",
              "98    None\n",
              "99    None\n",
              "Name: image_tags, Length: 100, dtype: object"
            ]
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "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>drp.en_ko.in_house.deepnatural_002411.jpg</td>\n",
              "      <td>375</td>\n",
              "      <td>1884</td>\n",
              "      <td>2448</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_002418.jpg</td>\n",
              "      <td>207</td>\n",
              "      <td>2481</td>\n",
              "      <td>3507</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_002432.jpg</td>\n",
              "      <td>364</td>\n",
              "      <td>3024</td>\n",
              "      <td>4032</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_002441.jpg</td>\n",
              "      <td>368</td>\n",
              "      <td>3120</td>\n",
              "      <td>4160</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_002451.jpg</td>\n",
              "      <td>364</td>\n",
              "      <td>2480</td>\n",
              "      <td>3507</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>95</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003932.jpg</td>\n",
              "      <td>238</td>\n",
              "      <td>2481</td>\n",
              "      <td>3507</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>96</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003951.jpg</td>\n",
              "      <td>464</td>\n",
              "      <td>2488</td>\n",
              "      <td>2304</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>97</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003958.jpg</td>\n",
              "      <td>259</td>\n",
              "      <td>1440</td>\n",
              "      <td>1440</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003976.jpg</td>\n",
              "      <td>250</td>\n",
              "      <td>1080</td>\n",
              "      <td>1440</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99</th>\n",
              "      <td>drp.en_ko.in_house.deepnatural_003979.jpg</td>\n",
              "      <td>372</td>\n",
              "      <td>1700</td>\n",
              "      <td>2338</td>\n",
              "      <td>None</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>100 rows × 5 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                        image  word_counts  image_width   \n",
              "0   drp.en_ko.in_house.deepnatural_002411.jpg          375         1884  \\\n",
              "1   drp.en_ko.in_house.deepnatural_002418.jpg          207         2481   \n",
              "2   drp.en_ko.in_house.deepnatural_002432.jpg          364         3024   \n",
              "3   drp.en_ko.in_house.deepnatural_002441.jpg          368         3120   \n",
              "4   drp.en_ko.in_house.deepnatural_002451.jpg          364         2480   \n",
              "..                                        ...          ...          ...   \n",
              "95  drp.en_ko.in_house.deepnatural_003932.jpg          238         2481   \n",
              "96  drp.en_ko.in_house.deepnatural_003951.jpg          464         2488   \n",
              "97  drp.en_ko.in_house.deepnatural_003958.jpg          259         1440   \n",
              "98  drp.en_ko.in_house.deepnatural_003976.jpg          250         1080   \n",
              "99  drp.en_ko.in_house.deepnatural_003979.jpg          372         1700   \n",
              "\n",
              "    image_height image_tags  \n",
              "0           2448       None  \n",
              "1           3507       None  \n",
              "2           4032       None  \n",
              "3           4160       None  \n",
              "4           3507       None  \n",
              "..           ...        ...  \n",
              "95          3507       None  \n",
              "96          2304       None  \n",
              "97          1440       None  \n",
              "98          1440       None  \n",
              "99          2338       None  \n",
              "\n",
              "[100 rows x 5 columns]"
            ]
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "image_df['image_tags'].apply(lambda x: x.remove(\"autoannotated\"))\n",
        "image_df['image_tags'] = image_df['image_tags'].apply(lambda x: ','.join(map(str, x)) if len(x) != 0 else \"None\")\n",
        "image_df"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OFxNQVQvnEyL",
        "outputId": "b61b37f0-db2a-4b6f-e4b6-d3344c4c0e59"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'None'}"
            ]
          },
          "execution_count": 11,
          "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": "YBOJBM5WnEyL",
        "outputId": "da326b5a-f2ad-4f6b-b844-6883f71f3e27"
      },
      "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>100.000000</td>\n",
              "      <td>100.000000</td>\n",
              "      <td>100.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>311.130000</td>\n",
              "      <td>2042.700000</td>\n",
              "      <td>2612.88000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>91.962515</td>\n",
              "      <td>856.654374</td>\n",
              "      <td>1098.84429</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>157.000000</td>\n",
              "      <td>645.000000</td>\n",
              "      <td>803.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>216.750000</td>\n",
              "      <td>1285.750000</td>\n",
              "      <td>1440.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>354.500000</td>\n",
              "      <td>2115.000000</td>\n",
              "      <td>2717.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>366.000000</td>\n",
              "      <td>2652.250000</td>\n",
              "      <td>3507.00000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>595.000000</td>\n",
              "      <td>4032.000000</td>\n",
              "      <td>4160.00000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       word_counts  image_width  image_height\n",
              "count   100.000000   100.000000     100.00000\n",
              "mean    311.130000  2042.700000    2612.88000\n",
              "std      91.962515   856.654374    1098.84429\n",
              "min     157.000000   645.000000     803.00000\n",
              "25%     216.750000  1285.750000    1440.00000\n",
              "50%     354.500000  2115.000000    2717.00000\n",
              "75%     366.000000  2652.250000    3507.00000\n",
              "max     595.000000  4032.000000    4160.00000"
            ]
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "image_df.describe()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "UzGBBtEWnEyL"
      },
      "source": [
        "### Image width, height "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "3TsKyy_-nEyL",
        "outputId": "164531e8-d109-4b78-e749-4c9c20936cf0"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.JointGrid at 0x7f24e7be9310>"
            ]
          },
          "execution_count": 13,
          "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": [
        "g = sns.jointplot(x = \"image_width\", y = \"image_height\", data=image_df, kind=\"kde\", space=0, color=\"r\")\n",
        "g.set_axis_labels(\"Image Width\", \"Image Height\")"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "ZCGiIqnZnEyM"
      },
      "source": [
        "### Image tags, Word tags"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "j5V5w4z5nEyM"
      },
      "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": "Ayeg4QhnnEyM",
        "outputId": "a44e541f-bd13-4e3f-f7c2-3d51593fc926"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_19303/2687934052.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": "SCWPfCOLnEyM",
        "outputId": "2ac52107-8e34-4c34-e41b-f0b1bc370dd1"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_19303/2687934052.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_tags = [i for i in word_tags if i not in ['Auto','UpdatedBbox','UpdatedText','UpdatedTags']]\n",
        "word_tag_df = pd.DataFrame(word_tags, columns = ['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='tags',index='index')\n",
        "plot_count_df(df=word_tag_df, field='tags', random_sample=False, color='g', rotation=0, xlabel='word tags', \\\n",
        "              ylabel='Count of each word tag', title='Word tag Distribution')"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "Qyh3tF1DnEyM"
      },
      "source": [
        "### Words per Image"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "pvf9yub_nEyM"
      },
      "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": "rQUHwxIenEyN",
        "outputId": "7c025601-f0d7-4f1e-b049-4b7477b79f73"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_19303/845028883.py:4: UserWarning: \n",
            "\n",
            "`distplot` is a deprecated function and will be removed in seaborn v0.14.0.\n",
            "\n",
            "Please adapt your code to use either `displot` (a figure-level function with\n",
            "similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "\n",
            "For a guide to updating your code to use the new functions, please see\n",
            "https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751\n",
            "\n",
            "  sns.distplot(df[field], bins=bins, color=color, ax=ax)\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": "eKMPOxFVnEyO"
      },
      "outputs": [],
      "source": [
        "word_df ={}\n",
        "word_df['index'] = [i for i in range(len(bbox_size))]\n",
        "word_df['orientation'] = orientation\n",
        "word_df['language'] = languages\n",
        "# word_df['height'] = word_height\n",
        "# word_df['width'] = word_width\n",
        "word_df['bbox_size'] = bbox_size"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2MOckinvnEyO"
      },
      "outputs": [],
      "source": [
        "word_df = pd.DataFrame.from_dict(word_df)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lFN9wqWtnEyO",
        "outputId": "607c75e1-786e-4711-e3c3-54d8c932c5b0"
      },
      "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>None</td>\n",
              "      <td>3772</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>None</td>\n",
              "      <td>3000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>None</td>\n",
              "      <td>4557</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>None</td>\n",
              "      <td>799</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>None</td>\n",
              "      <td>5800</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>31108</th>\n",
              "      <td>31108</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>399</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31109</th>\n",
              "      <td>31109</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>506</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31110</th>\n",
              "      <td>31110</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>550</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31111</th>\n",
              "      <td>31111</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>483</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31112</th>\n",
              "      <td>31112</td>\n",
              "      <td>Horizontal</td>\n",
              "      <td>ko</td>\n",
              "      <td>483</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>31113 rows × 4 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       index orientation language  bbox_size\n",
              "0          0  Horizontal     None       3772\n",
              "1          1  Horizontal     None       3000\n",
              "2          2  Horizontal     None       4557\n",
              "3          3  Horizontal     None        799\n",
              "4          4  Horizontal     None       5800\n",
              "...      ...         ...      ...        ...\n",
              "31108  31108  Horizontal       ko        399\n",
              "31109  31109  Horizontal       ko        506\n",
              "31110  31110  Horizontal       ko        550\n",
              "31111  31111  Horizontal       ko        483\n",
              "31112  31112  Horizontal       ko        483\n",
              "\n",
              "[31113 rows x 4 columns]"
            ]
          },
          "execution_count": 21,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "word_df['language'] = word_df['language'].apply(lambda x: ','.join(map(str, x)))\n",
        "word_df"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "QBGslS9lnEyO"
      },
      "source": [
        "### Orientation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lrl-zCh1nEyO",
        "outputId": "22863978-1169-403c-837f-6cc22b03a284"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_19303/2687934052.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')"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "fPvOZpzynEyO"
      },
      "source": [
        "### Lanaguage"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "hjjSSxV9nEyO",
        "outputId": "34f6fa7e-245d-49fa-9d64-717598e57ed6"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_19303/2687934052.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')"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "qVzvDRUSnEyO"
      },
      "source": [
        "### Aspect Ratio"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wGXgMiSXnEyO"
      },
      "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": "bZNDMV-7nEyO",
        "outputId": "d1a5dd17-e807-42da-817b-17c3d3cbdffa"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_19303/1739589287.py:4: UserWarning: \n",
            "\n",
            "`distplot` is a deprecated function and will be removed in seaborn v0.14.0.\n",
            "\n",
            "Please adapt your code to use either `displot` (a figure-level function with\n",
            "similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "\n",
            "For a guide to updating your code to use the new functions, please see\n",
            "https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751\n",
            "\n",
            "  sns.distplot(target_list, bins=bins, color=color, ax=ax)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# ! horizontal 단어들의 aspect ratio\n",
        "# ! 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": "diEGh80-nEyO",
        "outputId": "dbe56697-3761-45cb-9b2c-c36842aad826"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_19303/1739589287.py:4: UserWarning: \n",
            "\n",
            "`distplot` is a deprecated function and will be removed in seaborn v0.14.0.\n",
            "\n",
            "Please adapt your code to use either `displot` (a figure-level function with\n",
            "similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "\n",
            "For a guide to updating your code to use the new functions, please see\n",
            "https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751\n",
            "\n",
            "  sns.distplot(target_list, bins=bins, color=color, ax=ax)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# ! vertical 단어들의 aspect ration\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)')"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "GDa17Iv1nEyP"
      },
      "source": [
        "### BBOX size"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_bK35BzynEyP",
        "outputId": "1e963a1f-fee9-41b6-b0bf-f30a824f386d"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "count     31113.000000\n",
              "mean       2121.231704\n",
              "std        3598.708463\n",
              "min          15.000000\n",
              "25%         440.000000\n",
              "50%         858.000000\n",
              "75%        2555.000000\n",
              "max      136157.000000\n",
              "Name: bbox_size, dtype: float64"
            ]
          },
          "execution_count": 27,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "word_df['bbox_size'].describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "mUL63E2OnEyP",
        "outputId": "98846dd6-b9c4-46c4-dd78-0a6eaaebb9d1"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_19303/845028883.py:4: UserWarning: \n",
            "\n",
            "`distplot` is a deprecated function and will be removed in seaborn v0.14.0.\n",
            "\n",
            "Please adapt your code to use either `displot` (a figure-level function with\n",
            "similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "\n",
            "For a guide to updating your code to use the new functions, please see\n",
            "https://gist.github.com/mwaskom/de44147ed2974457ad6372750bbe5751\n",
            "\n",
            "  sns.distplot(df[field], bins=bins, color=color, ax=ax)\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# ! bbox size 분포 \n",
        "plot_dist(df=word_df, field='bbox_size', bins=50, color='r', xlabel='BBOX size', \\\n",
        "          ylabel='Frequency', title='BBOX size')"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "gpuClass": "standard",
    "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": "3.8.5"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}