{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2c124ef2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5249e11d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>act</th>\n",
       "      <th>year</th>\n",
       "      <th>chapter</th>\n",
       "      <th>section_no</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>legal_topic</th>\n",
       "      <th>procedural_stage</th>\n",
       "      <th>actors</th>\n",
       "      <th>keywords</th>\n",
       "      <th>bail_related</th>\n",
       "      <th>investigation_related</th>\n",
       "      <th>court_related</th>\n",
       "      <th>rights_related</th>\n",
       "      <th>chapter_no</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>1</td>\n",
       "      <td>Short title, extent and commencement</td>\n",
       "      <td>(1) This Act may be called the Bharatiya Nagar...</td>\n",
       "      <td>Legislation</td>\n",
       "      <td>Enactment</td>\n",
       "      <td>[Central Government]</td>\n",
       "      <td>[Act, Bharatiya Nagarik Suraksha Sanhita]</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>2</td>\n",
       "      <td>Definitions</td>\n",
       "      <td>(1) In this Sanhita, unless the context otherw...</td>\n",
       "      <td>Definitions</td>\n",
       "      <td>Preliminary</td>\n",
       "      <td>[State Government, Magistrate, Police Officer]</td>\n",
       "      <td>[definition, offence, bail, cognizable]</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>3</td>\n",
       "      <td>Construction of references</td>\n",
       "      <td>(1) Unless the context otherwise requires, any...</td>\n",
       "      <td>Jurisdiction</td>\n",
       "      <td>Preliminary</td>\n",
       "      <td>[Magistrate, Judicial Magistrate, Executive Ma...</td>\n",
       "      <td>[reference, magistrate, judicial magistrate, e...</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>4</td>\n",
       "      <td>Trial of offences under Bharatiya Nyaya Sanhit...</td>\n",
       "      <td>(1) All offences under \\nthe Bharatiya Nyaya S...</td>\n",
       "      <td>Trial of Offences</td>\n",
       "      <td>Trial</td>\n",
       "      <td>[Court]</td>\n",
       "      <td>[trial, offences, Bharatiya Nyaya Sanhita]</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>5</td>\n",
       "      <td>Saving</td>\n",
       "      <td>Nothing contained in this Sanhita shall, in th...</td>\n",
       "      <td>Special Laws</td>\n",
       "      <td>Preliminary</td>\n",
       "      <td>[Legislature]</td>\n",
       "      <td>[special law, local law, jurisdiction]</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    act  year      chapter  section_no  \\\n",
       "0  BNSS  2023  PRELIMINARY           1   \n",
       "1  BNSS  2023  PRELIMINARY           2   \n",
       "2  BNSS  2023  PRELIMINARY           3   \n",
       "3  BNSS  2023  PRELIMINARY           4   \n",
       "4  BNSS  2023  PRELIMINARY           5   \n",
       "\n",
       "                                               title  \\\n",
       "0               Short title, extent and commencement   \n",
       "1                                        Definitions   \n",
       "2                         Construction of references   \n",
       "3  Trial of offences under Bharatiya Nyaya Sanhit...   \n",
       "4                                             Saving   \n",
       "\n",
       "                                                body        legal_topic  \\\n",
       "0  (1) This Act may be called the Bharatiya Nagar...        Legislation   \n",
       "1  (1) In this Sanhita, unless the context otherw...        Definitions   \n",
       "2  (1) Unless the context otherwise requires, any...       Jurisdiction   \n",
       "3  (1) All offences under \\nthe Bharatiya Nyaya S...  Trial of Offences   \n",
       "4  Nothing contained in this Sanhita shall, in th...       Special Laws   \n",
       "\n",
       "  procedural_stage                                             actors  \\\n",
       "0        Enactment                               [Central Government]   \n",
       "1      Preliminary     [State Government, Magistrate, Police Officer]   \n",
       "2      Preliminary  [Magistrate, Judicial Magistrate, Executive Ma...   \n",
       "3            Trial                                            [Court]   \n",
       "4      Preliminary                                      [Legislature]   \n",
       "\n",
       "                                            keywords  bail_related  \\\n",
       "0          [Act, Bharatiya Nagarik Suraksha Sanhita]         False   \n",
       "1            [definition, offence, bail, cognizable]          True   \n",
       "2  [reference, magistrate, judicial magistrate, e...         False   \n",
       "3         [trial, offences, Bharatiya Nyaya Sanhita]         False   \n",
       "4             [special law, local law, jurisdiction]         False   \n",
       "\n",
       "   investigation_related  court_related  rights_related chapter_no  \n",
       "0                  False          False           False  CHAPTER I  \n",
       "1                   True          False            True  CHAPTER I  \n",
       "2                   True          False           False  CHAPTER I  \n",
       "3                  False           True           False  CHAPTER I  \n",
       "4                  False           True           False  CHAPTER I  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_json(\"bnss_all_sections.json\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6d0ee93e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 531 entries, 0 to 530\n",
      "Data columns (total 15 columns):\n",
      " #   Column                 Non-Null Count  Dtype \n",
      "---  ------                 --------------  ----- \n",
      " 0   act                    531 non-null    object\n",
      " 1   year                   531 non-null    int64 \n",
      " 2   chapter                531 non-null    object\n",
      " 3   section_no             531 non-null    int64 \n",
      " 4   title                  531 non-null    object\n",
      " 5   body                   531 non-null    object\n",
      " 6   legal_topic            531 non-null    object\n",
      " 7   procedural_stage       531 non-null    object\n",
      " 8   actors                 531 non-null    object\n",
      " 9   keywords               531 non-null    object\n",
      " 10  bail_related           531 non-null    bool  \n",
      " 11  investigation_related  531 non-null    bool  \n",
      " 12  court_related          531 non-null    bool  \n",
      " 13  rights_related         531 non-null    bool  \n",
      " 14  chapter_no             531 non-null    object\n",
      "dtypes: bool(4), int64(2), object(9)\n",
      "memory usage: 47.8+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e274b3f6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(531, 15)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d6d7a458",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['act', 'year', 'chapter', 'section_no', 'title', 'body', 'legal_topic',\n",
       "       'procedural_stage', 'actors', 'keywords', 'bail_related',\n",
       "       'investigation_related', 'court_related', 'rights_related',\n",
       "       'chapter_no'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b43bf4cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>section_no</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>531.0</td>\n",
       "      <td>531.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2023.0</td>\n",
       "      <td>266.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.0</td>\n",
       "      <td>153.430766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2023.0</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2023.0</td>\n",
       "      <td>133.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2023.0</td>\n",
       "      <td>266.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2023.0</td>\n",
       "      <td>398.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2023.0</td>\n",
       "      <td>531.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         year  section_no\n",
       "count   531.0  531.000000\n",
       "mean   2023.0  266.000000\n",
       "std       0.0  153.430766\n",
       "min    2023.0    1.000000\n",
       "25%    2023.0  133.500000\n",
       "50%    2023.0  266.000000\n",
       "75%    2023.0  398.500000\n",
       "max    2023.0  531.000000"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "925c9704",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "num_cols = [\"year\",\"section_no\"]\n",
    "\n",
    "for i in num_cols:\n",
    "    sns.histplot(df,x=i,bins=25,kde=True)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4f8b35c6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "act                      0\n",
       "year                     0\n",
       "chapter                  0\n",
       "section_no               0\n",
       "title                    0\n",
       "body                     0\n",
       "legal_topic              0\n",
       "procedural_stage         0\n",
       "actors                   0\n",
       "keywords                 0\n",
       "bail_related             0\n",
       "investigation_related    0\n",
       "court_related            0\n",
       "rights_related           0\n",
       "chapter_no               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "458c04d3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  1,   2,   3,   4,   5,   6,   7,   8,   9,  10,  11,  12,  13,\n",
       "        14,  15,  16,  17,  18,  19,  20,  21,  22,  23,  24,  25,  26,\n",
       "        27,  28,  29,  30,  31,  32,  33,  34,  35,  36,  37,  38,  39,\n",
       "        40,  41,  42,  43,  44,  45,  46,  47,  48,  49,  50,  51,  52,\n",
       "        53,  54,  55,  56,  57,  58,  59,  60,  61,  62,  63,  64,  65,\n",
       "        66,  67,  68,  69,  70,  71,  72,  73,  74,  75,  76,  77,  78,\n",
       "        79,  80,  81,  82,  83,  84,  85,  86,  87,  88,  89,  90,  91,\n",
       "        92,  93,  94,  95,  96,  97,  98,  99, 100, 101, 102, 103, 104,\n",
       "       105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117,\n",
       "       118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130,\n",
       "       131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143,\n",
       "       144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156,\n",
       "       157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169,\n",
       "       170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182,\n",
       "       183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195,\n",
       "       196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208,\n",
       "       209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221,\n",
       "       222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234,\n",
       "       235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247,\n",
       "       248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260,\n",
       "       261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273,\n",
       "       274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286,\n",
       "       287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299,\n",
       "       300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312,\n",
       "       313, 314, 315, 316, 317, 318, 319, 320, 321, 322, 323, 324, 325,\n",
       "       326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338,\n",
       "       339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351,\n",
       "       352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364,\n",
       "       365, 366, 367, 368, 369, 370, 371, 372, 373, 374, 375, 376, 377,\n",
       "       378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390,\n",
       "       391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403,\n",
       "       404, 405, 406, 407, 408, 409, 410, 411, 412, 413, 414, 415, 416,\n",
       "       417, 418, 419, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429,\n",
       "       430, 431, 432, 433, 434, 435, 436, 437, 438, 439, 440, 441, 442,\n",
       "       443, 444, 445, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455,\n",
       "       456, 457, 458, 459, 460, 461, 462, 463, 464, 465, 466, 467, 468,\n",
       "       469, 470, 471, 472, 473, 474, 475, 476, 477, 478, 479, 480, 481,\n",
       "       482, 483, 484, 485, 486, 487, 488, 489, 490, 491, 492, 493, 494,\n",
       "       495, 496, 497, 498, 499, 500, 501, 502, 503, 504, 505, 506, 507,\n",
       "       508, 509, 510, 511, 512, 513, 514, 515, 516, 517, 518, 519, 520,\n",
       "       521, 522, 523, 524, 525, 526, 527, 528, 529, 530, 531])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"section_no\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2231dec0",
   "metadata": {},
   "outputs": [],
   "source": [
    "def int_to_str(num):\n",
    "    return str(num)\n",
    "\n",
    "def bool_to_str(bol):\n",
    "    if bol:\n",
    "        return \"Yes\"\n",
    "    else:\n",
    "        return \"No\"\n",
    "\n",
    "def list_to_str(lis):\n",
    "    return \" \".join(lis)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "db1e3b32",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 531 entries, 0 to 530\n",
      "Data columns (total 15 columns):\n",
      " #   Column                 Non-Null Count  Dtype \n",
      "---  ------                 --------------  ----- \n",
      " 0   act                    531 non-null    object\n",
      " 1   year                   531 non-null    object\n",
      " 2   chapter                531 non-null    object\n",
      " 3   section_no             531 non-null    object\n",
      " 4   title                  531 non-null    object\n",
      " 5   body                   531 non-null    object\n",
      " 6   legal_topic            531 non-null    object\n",
      " 7   procedural_stage       531 non-null    object\n",
      " 8   actors                 531 non-null    object\n",
      " 9   keywords               531 non-null    object\n",
      " 10  bail_related           531 non-null    int64 \n",
      " 11  investigation_related  531 non-null    object\n",
      " 12  court_related          531 non-null    object\n",
      " 13  rights_related         531 non-null    object\n",
      " 14  chapter_no             531 non-null    object\n",
      "dtypes: int64(1), object(14)\n",
      "memory usage: 62.4+ KB\n"
     ]
    }
   ],
   "source": [
    "df[\"year\"] = df[\"year\"].apply(int_to_str)\n",
    "df[\"section_no\"] = df[\"section_no\"].apply(int_to_str)\n",
    "df[\"bail_related\"] = df[\"bail_related\"].astype(int)\n",
    "df[\"investigation_related\"] = df[\"investigation_related\"].apply(bool_to_str)\n",
    "df[\"court_related\"] = df[\"court_related\"].apply(bool_to_str)\n",
    "df[\"rights_related\"] = df[\"rights_related\"].apply(bool_to_str)\n",
    "df[\"actors\"] = df[\"actors\"].apply(list_to_str)\n",
    "df[\"keywords\"] = df[\"keywords\"].apply(list_to_str)\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d459471a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['CHAPTER I', 'CHAPTER II', 'CHAPTER III', 'CHAPTER IV',\n",
       "       'CHAPTER V', 'CHAPTER VI', 'CHAPTER VII', 'CHAPTER VIII',\n",
       "       'CHAPTER IX', 'CHAPTER X', 'CHAPTER XI', 'CHAPTER XII',\n",
       "       'CHAPTER XIII', 'CHAPTER XIV', 'CHAPTER XV', 'CHAPTER XVI',\n",
       "       'CHAPTER XVII', 'CHAPTER XVIII', 'CHAPTER XIX', 'CHAPTER XX',\n",
       "       'CHAPTER XXI', 'CHAPTER XXII', 'CHAPTER XXIII', 'CHAPTER XXIV',\n",
       "       'CHAPTER XXV', 'CHAPTER XXVI', 'CHAPTER XXVII', 'CHAPTER XXVIII',\n",
       "       'CHAPTER XXIX', 'CHAPTER XXX', 'CHAPTER XXXI', 'CHAPTER XXXII',\n",
       "       'CHAPTER XXXIII', 'CHAPTER XXXIV', 'CHAPTER XXXV', 'CHAPTER XXXVI',\n",
       "       'CHAPTER XXXVII', 'CHAPTER XXXVIII', 'CHAPTER XXXIX'], dtype=object)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"chapter_no\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "e41f21e1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAGxCAYAAACEFXd4AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAALQRJREFUeJzt3Qt0VOW5//EnFxKuCSWQhJRAURQIBNDgwRwtpYBEQI6U1AsioObAgQItYAMnLkAu1SBaQTlc1KWALRTFih5RgYhcVIJAFMWAKXBoiSUXqyYBbAIk81/Pu9aefwYSQJpkT958P2vtNZm99+x592TB/PK+z7t3gMfj8QgAAIClAt1uAAAAQG0i7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArBbsdgP8QUVFhZw8eVJatGghAQEBbjcHAABcAb0u8qlTpyQmJkYCA6vvvyHsiJigExsbeyWfKwAA8DO5ubnSrl27arcTdkRMj47zYYWFhdXdbwcAAFy1kpIS01nhfI9Xh7Aj4h260qBD2AEAoH65XAkKBcoAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAVvObsLNw4UJzi/apU6d615WWlsqkSZMkIiJCmjdvLsnJyVJQUODzuhMnTsjQoUOladOmEhkZKampqXL+/HkXzgAAAPijYPED+/btk+eee0569Ojhs37atGny9ttvy4YNGyQ8PFwmT54sI0aMkI8++shsLy8vN0EnOjpadu/eLXl5eTJmzBhp1KiRPP744+JPElJfdrsJgF/KenKM200AYDnXe3ZOnz4to0aNkhdeeEF+9KMfedcXFxfLiy++KE8//bT0799fEhISZNWqVSbU7Nmzx+yzdetWOXTokPzxj3+UXr16yeDBg2XBggWybNkyOXv2rItnBQAA/IXrYUeHqbR3ZuDAgT7rs7Ky5Ny5cz7ru3TpIu3bt5fMzEzzXB/j4+MlKirKu09SUpKUlJRIdnZ2te9ZVlZm9qm8AAAAO7k6jLV+/Xr55JNPzDDWhfLz8yUkJERatmzps16DjW5z9qkcdJztzrbqpKeny7x582roLAAAgD9zrWcnNzdXfvOb38jatWulcePGdfreaWlpZpjMWbQtAADATq6FHR2mKiwslBtvvFGCg4PNsnPnTnn22WfNz9pDo3U3RUVFPq/T2VhakKz08cLZWc5zZ5+qhIaGSlhYmM8CAADs5FrYGTBggBw8eFAOHDjgXXr37m2KlZ2fdVbVtm3bvK/JyckxU80TExPNc33UY2hocmRkZJjwEhcX58p5AQAA/+JazU6LFi2ke/fuPuuaNWtmrqnjrE9JSZHp06dLq1atTICZMmWKCTg333yz2T5o0CATakaPHi2LFi0ydTqzZs0yRc/aewMAAOAX19mpzuLFiyUwMNBcTFBnUOlMq+XLl3u3BwUFyaZNm2TixIkmBGlYGjt2rMyfP9/VdgMAAP8R4PF4PNLA6dRzvWihFivXVv0OFxUEqsZFBQHU9ve369fZAQAAqE2EHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1VwNOytWrJAePXpIWFiYWRITE+Xdd9/1bu/Xr58EBAT4LBMmTPA5xokTJ2To0KHStGlTiYyMlNTUVDl//rwLZwMAAPxRsJtv3q5dO1m4cKFcd9114vF4ZM2aNXLnnXfKp59+Kt26dTP7jBs3TubPn+99jYYaR3l5uQk60dHRsnv3bsnLy5MxY8ZIo0aN5PHHH3flnAAAgH9xNewMGzbM5/ljjz1menv27NnjDTsabjTMVGXr1q1y6NAhee+99yQqKkp69eolCxYskJkzZ8rcuXMlJCSkTs4DAAD4L7+p2dFemvXr18uZM2fMcJZj7dq10rp1a+nevbukpaXJ999/792WmZkp8fHxJug4kpKSpKSkRLKzs+v8HAAAgP9xtWdHHTx40ISb0tJSad68uWzcuFHi4uLMtvvuu086dOggMTEx8vnnn5sem5ycHHn99dfN9vz8fJ+go5znuq06ZWVlZnFoOAIAAHZyPex07txZDhw4IMXFxfLaa6/J2LFjZefOnSbwjB8/3ruf9uC0bdtWBgwYIMeOHZNrr732qt8zPT1d5s2bV0NnAAAA/Jnrw1haV9OpUydJSEgwIaRnz57yzDPPVLlvnz59zOPRo0fNo9byFBQU+OzjPK+uzkfpcJiGK2fJzc2twTMCAAD+xPWwc6GKigqfIabKtAdIaQ+P0uEvHQYrLCz07pORkWGmsTtDYVUJDQ31Tnd3FgAAYCdXh7G0h2Xw4MHSvn17OXXqlKxbt0527NghW7ZsMUNV+nzIkCESERFhanamTZsmffv2NdfmUYMGDTKhZvTo0bJo0SJTpzNr1iyZNGmSCTQAAACuhh3tkdHr4uj1ccLDw02I0aBz2223maElnVK+ZMkSM0MrNjZWkpOTTZhxBAUFyaZNm2TixImml6dZs2am5qfydXkAAEDDFuDRq/k1cDobS8OW1u/U1pBWQurLtXJcoL7LenKM200AYPn3t9/V7AAAANQkwg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDVXw86KFSukR48eEhYWZpbExER59913vdtLS0tl0qRJEhERIc2bN5fk5GQpKCjwOcaJEydk6NCh0rRpU4mMjJTU1FQ5f/68C2cDAAD8kathp127drJw4ULJysqS/fv3S//+/eXOO++U7Oxss33atGny1ltvyYYNG2Tnzp1y8uRJGTFihPf15eXlJuicPXtWdu/eLWvWrJHVq1fLnDlzXDwrAADgTwI8Ho9H/EirVq3kySeflF/+8pfSpk0bWbdunflZffnll9K1a1fJzMyUm2++2fQC3XHHHSYERUVFmX1WrlwpM2fOlK+//lpCQkKu6D1LSkokPDxciouLTQ9TbUhIfblWjgvUd1lPjnG7CQDqqSv9/vabmh3tpVm/fr2cOXPGDGdpb8+5c+dk4MCB3n26dOki7du3N2FH6WN8fLw36KikpCRz8k7vUFXKysrMPpUXAABgJ9fDzsGDB009TmhoqEyYMEE2btwocXFxkp+fb3pmWrZs6bO/BhvdpvSxctBxtjvbqpOenm6SoLPExsbWyrkBAAD3uR52OnfuLAcOHJCPP/5YJk6cKGPHjpVDhw7V6numpaWZLi9nyc3NrdX3AwAA7gkWl2nvTadOnczPCQkJsm/fPnnmmWfknnvuMYXHRUVFPr07OhsrOjra/KyPe/fu9TmeM1vL2acq2oukCwAAsJ/rPTsXqqioMDU1GnwaNWok27Zt827LyckxU821pkfpow6DFRYWevfJyMgwRUo6FAYAAOBqz44OJw0ePNgUHZ86dcrMvNqxY4ds2bLF1NKkpKTI9OnTzQwtDTBTpkwxAUdnYqlBgwaZUDN69GhZtGiRqdOZNWuWuTYPPTcAAMD1sKM9MmPGjJG8vDwTbvQCgxp0brvtNrN98eLFEhgYaC4mqL09OtNq+fLl3tcHBQXJpk2bTK2PhqBmzZqZmp/58+e7eFYAAMCf+N11dtzAdXYA93CdHQAN5jo7AAAAtYGwAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwmqthJz09XW666SZp0aKFREZGyvDhwyUnJ8dnn379+klAQIDPMmHCBJ99Tpw4IUOHDpWmTZua46Smpsr58+fr+GwAAIA/CnbzzXfu3CmTJk0ygUfDySOPPCKDBg2SQ4cOSbNmzbz7jRs3TubPn+99rqHGUV5eboJOdHS07N69W/Ly8mTMmDHSqFEjefzxx+v8nAAAgH9xNexs3rzZ5/nq1atNz0xWVpb07dvXJ9xomKnK1q1bTTh67733JCoqSnr16iULFiyQmTNnyty5cyUkJKTWzwMAAPgvv6rZKS4uNo+tWrXyWb927Vpp3bq1dO/eXdLS0uT777/3bsvMzJT4+HgTdBxJSUlSUlIi2dnZddh6AADgj1zt2amsoqJCpk6dKrfccosJNY777rtPOnToIDExMfL555+bHhut63n99dfN9vz8fJ+go5znuq0qZWVlZnFoMAIAAHbym7CjtTtffPGFfPjhhz7rx48f7/1Ze3Datm0rAwYMkGPHjsm111571YXR8+bN+5fbDAAA/J9fDGNNnjxZNm3aJNu3b5d27dpdct8+ffqYx6NHj5pHreUpKCjw2cd5Xl2djw6F6ZCZs+Tm5tbQmQAAAH/jatjxeDwm6GzcuFHef/996dix42Vfc+DAAfOoPTwqMTFRDh48KIWFhd59MjIyJCwsTOLi4qo8RmhoqNleeQEAAHYKdnvoat26dfLmm2+aa+04NTbh4eHSpEkTM1Sl24cMGSIRERGmZmfatGlmplaPHj3MvjpVXUPN6NGjZdGiReYYs2bNMsfWUAMAABo2V3t2VqxYYYaR9MKB2lPjLK+88orZrtPGdUq5BpouXbrIww8/LMnJyfLWW295jxEUFGSGwPRRe3nuv/9+c52dytflAQAADVew28NYlxIbG2suPHg5OlvrnXfeqcGWAQAAW/hFgTIAAEBtIewAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFa7qrDTv39/KSoqumh9SUmJ2QYAAFCvw86OHTvk7NmzF60vLS2VDz74oCbaBQAAUPf3xtK7jjsOHTrkvUu5Ki8vl82bN8uPf/zjmmkZAABAXYedXr16SUBAgFmqGq5q0qSJLF26tCbaBQAAUPdh5/jx4+ZO5ddcc43s3btX2rRp490WEhIikZGREhQUVDMtAwAAqOuw06FDB/NYUVFRE+8NAADgX2GnsiNHjsj27dulsLDwovAzZ86cmmgbAACAO2HnhRdekIkTJ0rr1q0lOjra1PA49GfCDgAAqNdh53e/+5089thjMnPmzJpvEQAAgNvX2fnuu+/krrvuqsl2AAAA+E/Y0aCzdevWmm8NAACAPwxjderUSWbPni179uyR+Ph4adSokc/2X//61zXVPgAAgLoPO88//7w0b95cdu7caZbKtECZsAMAAOp12NGLCwIAAFhbswMAAGB1z85DDz10ye0vvfTS1bYHAADA/bCjU88rO3funHzxxRdSVFRU5Q1CAQAA6lXY2bhx40Xr9JYRelXla6+9tibaBQAA4F81O4GBgTJ9+nRZvHhxTR0SAADAvwqUjx07JufPn6/JQwIAANT9MJb24FTm8XgkLy9P3n77bRk7duy/1iIAAAC3w86nn3560RBWmzZt5Pe///1lZ2oBAAD4fdjZvn17zbcEAADAX8KO4+uvv5acnBzzc+fOnU3vDgAAQL0vUD5z5owZrmrbtq307dvXLDExMZKSkiLff//9FR8nPT1dbrrpJmnRooVERkbK8OHDveHJUVpaKpMmTZKIiAhzP67k5GQpKCjw2efEiRMydOhQadq0qTlOamoqhdIAAODqw44WKOsNQN966y1zIUFd3nzzTbPu4YcfvuLj6P4aZPTu6RkZGebihIMGDTJhyjFt2jTzPhs2bDD7nzx5UkaMGOHdXl5eboLO2bNnZffu3bJmzRpZvXq1zJkz52pODQAAWCbAo1OpfqDWrVvLa6+9Jv369buolufuu+82w1tXQ1+nPTMaarS3qLi42AyNrVu3Tn75y1+afb788kvp2rWrZGZmys033yzvvvuu3HHHHSYERUVFmX1WrlwpM2fONMcLCQm57PuWlJRIeHi4eb+wsDCpDQmpL9fKcYH6LuvJMW43AUA9daXf31fVs6NDVU6wqEyDyg8ZxrqQNla1atXKPGZlZZnenoEDB3r36dKli7Rv396EHaWP8fHxPu1JSkoyH0B2dnaV71NWVma2V14AAICdrirsJCYmyqOPPmrqaRz//Oc/Zd68eWbb1dDbTUydOlVuueUW6d69u1mXn59vemZatmzps68GG93m7HNh8HKeO/tUVSukSdBZYmNjr6rNAADA0tlYS5Yskdtvv13atWsnPXv2NOs+++wzCQ0Nla1bt15VQ7R2R28m+uGHH0ptS0tL87kwovbsEHgAALDTVYUdHTY6cuSIrF271tTQqJEjR8qoUaOkSZMmP/h4kydPlk2bNsmuXbtMgHJER0ebwmMtgK7cu6OzsXSbs8/evXt9jufM1nL2uZCGMl0AAID9rirs6DCQDhWNGzfOZ/1LL71kioK1OPhKaG30lClTzF3Ud+zYIR07dvTZnpCQII0aNZJt27aZKedKp6brVHNnuEwfH3vsMSksLDQ1Q0pndmmhUlxc3NWcHgAAaOg1O88995wpFL5Qt27dzEyoHzJ09cc//tHMttJr7WiNjS5a/6O0nkav3aNDTjrTSwuWH3zwQRNwdCaW0qnqGmpGjx5thtK2bNkis2bNMsem9wYAAFxVz44GEr2g4IV0mrjeEPRKrVixwjxeOIV91apV8sADD5ifFy9ebO69pT07OotKZ1otX77cu29QUJAZAps4caIJQc2aNTM3I50/fz6/XQAAcHVhR4t5P/roo4uGnXSdXkn5Sl3JJX4aN24sy5YtM0t1OnToIO+8884Vvy8AAGg4rirsaK2OThPXa+D079/frNO6mhkzZvygKygDAAD4ZdjRe09988038qtf/crMlnJ6YLQwWad1AwAA1OuwExAQIE888YTMnj1bDh8+bKabX3fddRQEAwAAv3NVYcehdyHXu5YDAABYNfUcAACgviDsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaq6GnV27dsmwYcMkJiZGAgIC5I033vDZ/sADD5j1lZfbb7/dZ59vv/1WRo0aJWFhYdKyZUtJSUmR06dP1/GZAAAAf+Vq2Dlz5oz07NlTli1bVu0+Gm7y8vK8y5/+9Cef7Rp0srOzJSMjQzZt2mQC1Pjx4+ug9QAAoD4IdvPNBw8ebJZLCQ0Nlejo6Cq3HT58WDZv3iz79u2T3r17m3VLly6VIUOGyFNPPWV6jAAAQMPm9zU7O3bskMjISOncubNMnDhRvvnmG++2zMxMM3TlBB01cOBACQwMlI8//tilFgMAAH/ias/O5egQ1ogRI6Rjx45y7NgxeeSRR0xPkIacoKAgyc/PN0GosuDgYGnVqpXZVp2ysjKzOEpKSmr1PAAAgHv8Ouzce++93p/j4+OlR48ecu2115rengEDBlz1cdPT02XevHk11EoAAODP/H4Yq7JrrrlGWrduLUePHjXPtZansLDQZ5/z58+bGVrV1fmotLQ0KS4u9i65ubm13nYAAOCOehV2vvrqK1Oz07ZtW/M8MTFRioqKJCsry7vP+++/LxUVFdKnT59LFj3rVPXKCwAAsJOrw1h6PRynl0YdP35cDhw4YGpudNGhpuTkZNNLozU7M2bMkE6dOklSUpLZv2vXrqauZ9y4cbJy5Uo5d+6cTJ482Qx/MRMLAAC43rOzf/9+ueGGG8yipk+fbn6eM2eOKUD+/PPP5T/+4z/k+uuvNxcLTEhIkA8++MD0zDjWrl0rXbp0MTU8OuX81ltvleeff97FswIAAP7E1Z6dfv36icfjqXb7li1bLnsM7QFat25dDbcMAADYol7V7AAAAPxQhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAVgt2uwEAUN8lpL7sdhMAv5T15BjxB/TsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNVcDTu7du2SYcOGSUxMjAQEBMgbb7zhs93j8cicOXOkbdu20qRJExk4cKAcOXLEZ59vv/1WRo0aJWFhYdKyZUtJSUmR06dP1/GZAAAAf+Vq2Dlz5oz07NlTli1bVuX2RYsWybPPPisrV66Ujz/+WJo1ayZJSUlSWlrq3UeDTnZ2tmRkZMimTZtMgBo/fnwdngUAAPBnwW6++eDBg81SFe3VWbJkicyaNUvuvPNOs+7ll1+WqKgo0wN07733yuHDh2Xz5s2yb98+6d27t9ln6dKlMmTIEHnqqadMjxEAAGjY/LZm5/jx45Kfn2+Grhzh4eHSp08fyczMNM/1UYeunKCjdP/AwEDTE1SdsrIyKSkp8VkAAICd/DbsaNBR2pNTmT53tuljZGSkz/bg4GBp1aqVd5+qpKenm+DkLLGxsbVyDgAAwH1+G3ZqU1pamhQXF3uX3Nxct5sEAAAaWtiJjo42jwUFBT7r9bmzTR8LCwt9tp8/f97M0HL2qUpoaKiZvVV5AQAAdvLbsNOxY0cTWLZt2+Zdp7U1WouTmJhonutjUVGRZGVlefd5//33paKiwtT2AAAAuDobS6+Hc/ToUZ+i5AMHDpiam/bt28vUqVPld7/7nVx33XUm/MyePdvMsBo+fLjZv2vXrnL77bfLuHHjzPT0c+fOyeTJk81MLWZiAQAA18PO/v375ec//7n3+fTp083j2LFjZfXq1TJjxgxzLR69bo724Nx6661mqnnjxo29r1m7dq0JOAMGDDCzsJKTk821eQAAAFwPO/369TPX06mOXlV5/vz5ZqmO9gKtW7eulloIAADqO7+t2QEAAKgJhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGp+HXbmzp0rAQEBPkuXLl2820tLS2XSpEkSEREhzZs3l+TkZCkoKHC1zQAAwL/4ddhR3bp1k7y8PO/y4YcferdNmzZN3nrrLdmwYYPs3LlTTp48KSNGjHC1vQAAwL8Ei58LDg6W6Ojoi9YXFxfLiy++KOvWrZP+/fubdatWrZKuXbvKnj175Oabb3ahtQAAwN/4fc/OkSNHJCYmRq655hoZNWqUnDhxwqzPysqSc+fOycCBA7376hBX+/btJTMz85LHLCsrk5KSEp8FAADYya/DTp8+fWT16tWyefNmWbFihRw/flx++tOfyqlTpyQ/P19CQkKkZcuWPq+Jiooy2y4lPT1dwsPDvUtsbGwtnwkAAHCLXw9jDR482Ptzjx49TPjp0KGDvPrqq9KkSZOrPm5aWppMnz7d+1x7dgg8AADYya97di6kvTjXX3+9HD161NTxnD17VoqKinz20dlYVdX4VBYaGiphYWE+CwAAsFO9CjunT5+WY8eOSdu2bSUhIUEaNWok27Zt827PyckxNT2JiYmuthMAAPgPvx7G+u1vfyvDhg0zQ1c6rfzRRx+VoKAgGTlypKm1SUlJMcNRrVq1Mr0zU6ZMMUGHmVgAAKBehJ2vvvrKBJtvvvlG2rRpI7feequZVq4/q8WLF0tgYKC5mKDOsEpKSpLly5e73WwAAOBH/DrsrF+//pLbGzduLMuWLTMLAABAva/ZAQAA+KEIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAq1kTdpYtWyY/+clPpHHjxtKnTx/Zu3ev200CAAB+wIqw88orr8j06dPl0UcflU8++UR69uwpSUlJUlhY6HbTAACAy6wIO08//bSMGzdOHnzwQYmLi5OVK1dK06ZN5aWXXnK7aQAAwGX1PuycPXtWsrKyZODAgd51gYGB5nlmZqarbQMAAO4LlnruH//4h5SXl0tUVJTPen3+5ZdfVvmasrIysziKi4vNY0lJSa21s7zsn7V2bKA+q81/d3WFf9+AO/++neN7PB67w87VSE9Pl3nz5l20PjY21pX2AA1Z+NIJbjcBQD3/933q1CkJDw+3N+y0bt1agoKCpKCgwGe9Po+Ojq7yNWlpaaag2VFRUSHffvutRERESEBAQK23Ge7SvwQ02Obm5kpYWBi/DsAi/PtuWDwejwk6MTExl9yv3oedkJAQSUhIkG3btsnw4cO94UWfT548ucrXhIaGmqWyli1b1kl74T806BB2ADvx77vhCL9Ej441YUdpL83YsWOld+/e8m//9m+yZMkSOXPmjJmdBQAAGjYrws4999wjX3/9tcyZM0fy8/OlV69esnnz5ouKlgEAQMNjRdhROmRV3bAVUJkOYeoFKC8cygRQ//HvG1UJ8FxuvhYAAEA9Vu8vKggAAHAphB0AAGA1wg4AALAaYQdWeuCBB8wFIhcuXOiz/o033uDCkUA9pOWles/DpKSki7YtX77cXCvtq6++cqVt8H+EHVircePG8sQTT8h3333ndlMA/Iv0j5dVq1bJxx9/LM8995x3/fHjx2XGjBmydOlSadeuHZ8zqkTYgbX0r0C9ZYjeC606f/7zn6Vbt25muupPfvIT+f3vf1+nbQRw5fQ2L88884z89re/NSFHe3tSUlJk0KBBcsMNN8jgwYOlefPm5hpro0ePNjeKdrz22msSHx8vTZo0MbcG0v8f9OKzaBgIO7CW3jPt8ccfN3/xVdW9nZWVJXfffbfce++9cvDgQZk7d67Mnj1bVq9e7Up7AVyeXi1/wIAB8tBDD8n//M//yBdffGF6evr3728Cz/79+81FZfX+iPrvW+Xl5cnIkSPNaw4fPiw7duyQESNGXPZO2bAH19mBtTU7RUVFpkYnMTFR4uLi5MUXXzTPf/GLX5j/5EaNGmWuvL1161bv67Q7/O2335bs7GxX2w+geoWFhaZHVm/grL2zGng++OAD2bJli3cf/QNHe4JycnLk9OnT5h6Kf/3rX6VDhw58tA0QPTuwntbtrFmzxvxFV5k+v+WWW3zW6fMjR45IeXl5HbcSwJWKjIyU//qv/5KuXbuaG0B/9tlnsn37djOE5SxdunQx+x47dkx69uxpeoN0GOuuu+6SF154gVq+BoawA+v17dvXzOBIS0tzuykAakhwcLBZlPbcDBs2TA4cOOCz6B8u+u9fh7QzMjLk3XffNb28OrTduXNnU/eDhsGae2MBl6JT0PUGsfofnEP/Kvzoo4989tPn119/vfnPEUD9cOONN5rhLJ1k4ASgqmZzac+tLnrTaB3O2rhxo0yfPr3O24u6R88OGgTtvtYanWeffda77uGHH5Zt27bJggUL5C9/+YsZ6tKCR53pAaD+mDRpkqnf0SLkffv2maErrd958MEHzZC0TlfXyQpavHzixAl5/fXXTb2e/sGDhoGwgwZj/vz5UlFR4fPX4Kuvvirr16+X7t27m7/2dB8tbgZQf8TExJheWQ02Og1d/7iZOnWqudBgYGCghIWFya5du2TIkCGm53bWrFnmMhM6VR0NA7OxAACA1ejZAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBLNCvXz9zxVh/Vpdt1HskLVmyRPydm7+3HTt2mPtFFRUVufL+QF3iRqCABfReP40aNRJ/oF+iP//5z+W7774zl+uvzTauXr3ahIULv7D1/kjNmjUT21T32QK4NMIOYIFWrVqJv6vLNrZp00bcdPbsWQkJCXG1DQD+P4axAAtUHg7RIRy9w/NDDz0kLVq0kPbt28vzzz/v3fff//3fZebMmT6v1ztAa6+L3ixRlZWVmbu///jHPzY9JH369DG9Co6//e1vMmzYMPnRj35ktnfr1k3eeecd+etf/2p6HpRu02ES58aqFw7Z5OXlydChQ6VJkybSsWNHWbdu3UXDT08//bS5qaO+R2xsrPzqV7+S06dPm23aHr2rdXFxsXkfXebOnev9DCofR+90feedd0rz5s3NTSHvvvtuKSgo8G7X1/Xq1Uv+8Ic/mNeGh4fLvffeK6dOnbriz3/y5Mnm/Fq3bi1JSUlm/RdffGFuNqnvGxUVJaNHj5Z//OMf1R5H3793797m9xYdHS333XefFBYWmm2X+mz1Brfp6enmc9TPs2fPnvLaa6/5HFt/P3oTTN2ux9HjAQ0FYQewkN7RWb80P/30UxMQJk6cKDk5OWbbqFGjzJ3ePR6Pd/9XXnnF3Dn6pz/9qXmuX9yZmZlmv88//1zuuusuuf322+XIkSNm+6RJk0wg0nB08OBBeeKJJ8wXugaSP//5z2YffT8NNM8880yVbRwzZoycPHnShBZ9jQYy54vdoXesfvbZZyU7O1vWrFkj77//vsyYMcMb2jTQaHjR99FFA9qFNAho0Pn2229l586dkpGRIf/3f/8n99xzj89+x44dkzfeeEM2bdpkFt134cKFV/yZa/u0N0fvvr1y5UoztNa/f3+54YYbZP/+/bJ582YTsDRoVefcuXOyYMEC+eyzz0xbNJA4geZSn60GnZdfftm8r35W06ZNk/vvv9+cg8rNzZURI0aYgHrgwAH5z//8T/nv//7vKz43oN7zAKj3fvazn3l+85vfmJ87dOjguf/++73bKioqPJGRkZ4VK1aY54WFhZ7g4GDPrl27vPskJiZ6Zs6caX7+29/+5gkKCvL8/e9/93mPAQMGeNLS0szP8fHxnrlz51bZlu3bt2uK8nz33XfVtvHw4cNmn3379nm3HzlyxKxbvHhxtee5YcMGT0REhPf5qlWrPOHh4Rftp5+Bc5ytW7ea8zlx4oR3e3Z2tnmvvXv3muePPvqop2nTpp6SkhLvPqmpqZ4+ffpU25YLz+2GG27wWbdgwQLPoEGDfNbl5uaa983JybnoM6mKfj66/6lTp6r9bEtLS03bd+/e7fPalJQUz8iRI83P+nuLi4vz2a6/76p+T4CNqNkBLNSjRw/vzzrcoUMiTq+J1rMMGjRI1q5da3pyjh8/bnpxnnvuObNde2rKy8vNkEdl2pMTERFhfv71r39teou2bt0qAwcOlOTkZJ/3vBztmQgODpYbb7zRu65Tp05meKay9957z/RafPnll1JSUiLnz5+X0tJS+f7776Vp06ZX9F6HDx82vSK6OOLi4kyBr2676aabzDodvtLhI0fbtm0v6mm6lISEBJ/n2juzfft20+N1Ie1FuvDzVVlZWWZITV+rRcjaK+UMw2mbq3L06FHzedx2220X1Q1pr5LzGehQZGWJiYlXfG5AfUfYASx04awnDTzOF6czlKWBZenSpaZWRutidFFaExMUFGS+ePWxMueLW4dBtC7l7bffNoFHA4kOnU2ZMqXGzkGHcO644w4Tqh577DFT4Pzhhx9KSkqK+SK/0rBTU5/Z5Vw4+0s/Rx020iG+C2mQutCZM2fMZ6qLBlENpRpy9Lmeb3WcGib9XWiNVWWhoaFX3H7AZoQdoAHSGpbx48ebOhINO1o/49DeAO3Z0V4Np4anKtpTMmHCBLOkpaXJCy+8YMKOMwtJj1Gdzp07m14arSlyekS0h0J7MxwatjRsaIjS2h316quv+hxH3+tS76O6du1qalZ0cXp3Dh06ZGpqqustqQnaa6U1NtpjpL1Yl6O9V998842pE3LaqbU+lVX12eo5aKjRYPSzn/2s2s/gf//3f33W7dmz56rOC6iPKFAGGiDthRg+fLjMnj3bDHGMHDnSu02HV7TnRwOQXhtHh7n27t1rem+090DprKMtW7aYbZ988okZrtEvVNWhQwfTK6JFvjrLy+l5qKxLly5m+EsDlx5bQ4/+rDOF9LXOsJYW7GrvkxYU60wlLcCtTIOEHn/btm1mlpMO51xI30d7rfSctK36fnpuGgy0iLu2aBG3FkXrZ6vX/dGhK/3MdAZZVQFNZ81pmHHOV8OJFitXVtVnq0NvWpitRclaJK3vo+epx9HnSgOpFpenpqaaIUQNuHqNIqChIOwADZR++WttiPbe6BdtZatWrTKB4OGHHza9MBqM9Avb2U+/rPXLXAOOztLSgLR8+XKzTYdS5s2bZ2b76HRrndlVFZ09pNv79u0rv/jFL2TcuHHmi7tx48Zmu06f1qnnOgzUvXt3M7SjgasynZGlX+Q6s0qHfRYtWnTR+2g4ePPNN009kL6Xhp9rrrnGzECrTTq7TWdm6WelNVIauDQkaq2Q01NVmbZfA8iGDRtMb4328Dz11FM++1T32Woo0uCqn4/zO9FgqlPRlf7etJdJZ3jp56qhUS9PADQUAVql7HYjAOCrr74ywzdalDxgwAA+EAA1hrADwBV6zRwdhtEeD71mjF4/5+9//7v85S9/8ZtbXwCwAwXKAFyh9TiPPPKIqU/R4SsdktKhKn8LOpea9u0UO184DAjAv9CzAwCXoLPGLnVrhSudbQXAPYQdAABgNWZjAQAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABis/8HIXmW2c6TX3cAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cat_cols = [\"bail_related\",\"investigation_related\",\"rights_related\",\"chapter_no\"]\n",
    "for i in cat_cols:\n",
    "    sns.countplot(df,x=i)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "35c1eb51",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>act</th>\n",
       "      <th>year</th>\n",
       "      <th>chapter</th>\n",
       "      <th>section_no</th>\n",
       "      <th>title</th>\n",
       "      <th>body</th>\n",
       "      <th>legal_topic</th>\n",
       "      <th>procedural_stage</th>\n",
       "      <th>actors</th>\n",
       "      <th>keywords</th>\n",
       "      <th>bail_related</th>\n",
       "      <th>investigation_related</th>\n",
       "      <th>court_related</th>\n",
       "      <th>rights_related</th>\n",
       "      <th>chapter_no</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>1</td>\n",
       "      <td>Short title, extent and commencement</td>\n",
       "      <td>(1) This Act may be called the Bharatiya Nagar...</td>\n",
       "      <td>Legislation</td>\n",
       "      <td>Enactment</td>\n",
       "      <td>Central Government</td>\n",
       "      <td>Act Bharatiya Nagarik Suraksha Sanhita</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>2</td>\n",
       "      <td>Definitions</td>\n",
       "      <td>(1) In this Sanhita, unless the context otherw...</td>\n",
       "      <td>Definitions</td>\n",
       "      <td>Preliminary</td>\n",
       "      <td>State Government Magistrate Police Officer</td>\n",
       "      <td>definition offence bail cognizable</td>\n",
       "      <td>1</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>3</td>\n",
       "      <td>Construction of references</td>\n",
       "      <td>(1) Unless the context otherwise requires, any...</td>\n",
       "      <td>Jurisdiction</td>\n",
       "      <td>Preliminary</td>\n",
       "      <td>Magistrate Judicial Magistrate Executive Magis...</td>\n",
       "      <td>reference magistrate judicial magistrate execu...</td>\n",
       "      <td>0</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>4</td>\n",
       "      <td>Trial of offences under Bharatiya Nyaya Sanhit...</td>\n",
       "      <td>(1) All offences under \\nthe Bharatiya Nyaya S...</td>\n",
       "      <td>Trial of Offences</td>\n",
       "      <td>Trial</td>\n",
       "      <td>Court</td>\n",
       "      <td>trial offences Bharatiya Nyaya Sanhita</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BNSS</td>\n",
       "      <td>2023</td>\n",
       "      <td>PRELIMINARY</td>\n",
       "      <td>5</td>\n",
       "      <td>Saving</td>\n",
       "      <td>Nothing contained in this Sanhita shall, in th...</td>\n",
       "      <td>Special Laws</td>\n",
       "      <td>Preliminary</td>\n",
       "      <td>Legislature</td>\n",
       "      <td>special law local law jurisdiction</td>\n",
       "      <td>0</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>CHAPTER I</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    act  year      chapter section_no  \\\n",
       "0  BNSS  2023  PRELIMINARY          1   \n",
       "1  BNSS  2023  PRELIMINARY          2   \n",
       "2  BNSS  2023  PRELIMINARY          3   \n",
       "3  BNSS  2023  PRELIMINARY          4   \n",
       "4  BNSS  2023  PRELIMINARY          5   \n",
       "\n",
       "                                               title  \\\n",
       "0               Short title, extent and commencement   \n",
       "1                                        Definitions   \n",
       "2                         Construction of references   \n",
       "3  Trial of offences under Bharatiya Nyaya Sanhit...   \n",
       "4                                             Saving   \n",
       "\n",
       "                                                body        legal_topic  \\\n",
       "0  (1) This Act may be called the Bharatiya Nagar...        Legislation   \n",
       "1  (1) In this Sanhita, unless the context otherw...        Definitions   \n",
       "2  (1) Unless the context otherwise requires, any...       Jurisdiction   \n",
       "3  (1) All offences under \\nthe Bharatiya Nyaya S...  Trial of Offences   \n",
       "4  Nothing contained in this Sanhita shall, in th...       Special Laws   \n",
       "\n",
       "  procedural_stage                                             actors  \\\n",
       "0        Enactment                                 Central Government   \n",
       "1      Preliminary         State Government Magistrate Police Officer   \n",
       "2      Preliminary  Magistrate Judicial Magistrate Executive Magis...   \n",
       "3            Trial                                              Court   \n",
       "4      Preliminary                                        Legislature   \n",
       "\n",
       "                                            keywords  bail_related  \\\n",
       "0             Act Bharatiya Nagarik Suraksha Sanhita             0   \n",
       "1                 definition offence bail cognizable             1   \n",
       "2  reference magistrate judicial magistrate execu...             0   \n",
       "3             trial offences Bharatiya Nyaya Sanhita             0   \n",
       "4                 special law local law jurisdiction             0   \n",
       "\n",
       "  investigation_related court_related rights_related chapter_no  \n",
       "0                    No            No             No  CHAPTER I  \n",
       "1                   Yes            No            Yes  CHAPTER I  \n",
       "2                   Yes            No             No  CHAPTER I  \n",
       "3                    No           Yes             No  CHAPTER I  \n",
       "4                    No           Yes             No  CHAPTER I  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "35469a9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import string\n",
    "\n",
    "def remove_punc(txt):\n",
    "  return txt.translate(str.maketrans('','',string.punctuation))\n",
    "\n",
    "def tolower(txt):\n",
    "    return txt.lower()\n",
    "\n",
    "def remove_num(txt):\n",
    "    new = \"\"\n",
    "    for i in txt:\n",
    "        if not i.isdigit():\n",
    "            new+=i\n",
    "    return new\n",
    "\n",
    "def remove_emoj(txt):\n",
    "    new = \"\"\n",
    "    for i in txt:\n",
    "        if i.isascii():\n",
    "            new+=i\n",
    "    return new"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "5948c8e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop([\"act\",\"bail_related\"],axis=1)\n",
    "y = df[\"bail_related\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "51d105e2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      0\n",
       "1      1\n",
       "2      0\n",
       "3      0\n",
       "4      0\n",
       "      ..\n",
       "526    0\n",
       "527    0\n",
       "528    0\n",
       "529    0\n",
       "530    1\n",
       "Name: bail_related, Length: 531, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "aab70f99",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2023 PRELIMINARY 1 Short title, extent and com...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2023 PRELIMINARY 2 Definitions (1) In this San...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2023 PRELIMINARY 3 Construction of references ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2023 PRELIMINARY 4 Trial of offences under Bha...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2023 PRELIMINARY 5 Saving Nothing contained in...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text\n",
       "0  2023 PRELIMINARY 1 Short title, extent and com...\n",
       "1  2023 PRELIMINARY 2 Definitions (1) In this San...\n",
       "2  2023 PRELIMINARY 3 Construction of references ...\n",
       "3  2023 PRELIMINARY 4 Trial of offences under Bha...\n",
       "4  2023 PRELIMINARY 5 Saving Nothing contained in..."
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cols = X.columns.tolist()\n",
    "\n",
    "X[\"text\"] = X[cols[0]]\n",
    "X.drop(cols[0],axis=1,inplace=True)\n",
    "for i in range(1,len(cols)-1):\n",
    "    X[\"text\"] = X[\"text\"] + \" \" + X[cols[i]]\n",
    "    X.drop(cols[i],axis=1,inplace=True)\n",
    "\n",
    "X.drop(cols[-1],axis=1,inplace=True)\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "ad7e096d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2023 preliminary 1 short title extent and comm...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2023 preliminary 2 definitions 1 in this sanhi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2023 preliminary 3 construction of references ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2023 preliminary 4 trial of offences under bha...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2023 preliminary 5 saving nothing contained in...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text\n",
       "0  2023 preliminary 1 short title extent and comm...\n",
       "1  2023 preliminary 2 definitions 1 in this sanhi...\n",
       "2  2023 preliminary 3 construction of references ...\n",
       "3  2023 preliminary 4 trial of offences under bha...\n",
       "4  2023 preliminary 5 saving nothing contained in..."
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X[\"text\"] = X[\"text\"].apply(remove_punc)\n",
    "X[\"text\"] = X[\"text\"].apply(tolower)\n",
    "X[\"text\"] = X[\"text\"].apply(remove_emoj)\n",
    "# X[\"text\"] = X[\"text\"].apply(remove_num) \n",
    "\n",
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "29c41b0d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\Amir sohail\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\keras\\src\\export\\tf2onnx_lib.py:8: FutureWarning: In the future `np.object` will be defined as the corresponding NumPy scalar.\n",
      "  if not hasattr(np, \"object\"):\n"
     ]
    }
   ],
   "source": [
    "# from tensorflow.keras.models import Sequential\n",
    "# from tensorflow.keras.layers import Input,Embedding,Dense,LSTM,Dropout,Bidirectional\n",
    "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
    "from tensorflow.keras.preprocessing.text import Tokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "1d1923f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "Max_len = 100\n",
    "Max_words = 10000\n",
    "embedding_dim = 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "61ce2f5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.20, random_state=42,stratify=y)\n",
    "\n",
    "tokenizer = Tokenizer(num_words=Max_words,oov_token=\"<OOV>\")\n",
    "tokenizer.fit_on_texts(X_train[\"text\"])\n",
    "train_seq = tokenizer.texts_to_sequences(X_train[\"text\"])\n",
    "train_pad = pad_sequences(train_seq,maxlen=Max_len,padding=\"post\",truncating=\"post\")\n",
    "test_seq = tokenizer.texts_to_sequences(X_test[\"text\"])\n",
    "test_pad = pad_sequences(test_seq,maxlen=Max_len,padding=\"post\",truncating=\"post\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "430bf687",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier, ExtraTreesClassifier ,AdaBoostClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from xgboost import XGBClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "a92184d6",
   "metadata": {},
   "outputs": [],
   "source": [
    "results = []\n",
    "models = {\n",
    "    'Decision Tree_C': DecisionTreeClassifier(random_state=42,class_weight=\"balanced\"),\n",
    "    'Random Forest_C': RandomForestClassifier(random_state=42, n_jobs=-1,class_weight=\"balanced\"),\n",
    "    'Extra Trees_C': ExtraTreesClassifier(random_state=42, n_jobs=-1,class_weight=\"balanced\"),\n",
    "    'Gradient Boosting_C': GradientBoostingClassifier(random_state=42),\n",
    "    'K-Neighbors_C': KNeighborsClassifier(),\n",
    "    \"XGBoost_C\": XGBClassifier(random_state=42),\n",
    "    \"AdaBoost_C\": AdaBoostClassifier(random_state=42),\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "fd2b837d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>accuracy_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Decision Tree_C</td>\n",
       "      <td>0.831776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Random Forest_C</td>\n",
       "      <td>0.878505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Extra Trees_C</td>\n",
       "      <td>0.878505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Gradient Boosting_C</td>\n",
       "      <td>0.878505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>K-Neighbors_C</td>\n",
       "      <td>0.850467</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>XGBoost_C</td>\n",
       "      <td>0.878505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>AdaBoost_C</td>\n",
       "      <td>0.887850</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Name  accuracy_score\n",
       "0      Decision Tree_C        0.831776\n",
       "1      Random Forest_C        0.878505\n",
       "2        Extra Trees_C        0.878505\n",
       "3  Gradient Boosting_C        0.878505\n",
       "4        K-Neighbors_C        0.850467\n",
       "5            XGBoost_C        0.878505\n",
       "6           AdaBoost_C        0.887850"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for name,model in models.items():\n",
    "    model.fit(train_pad,y_train)\n",
    "    y_pred = model.predict(test_pad)\n",
    "    acc = accuracy_score(y_test,y_pred)\n",
    "    results.append({\n",
    "        \"Name\":name,\n",
    "        \"accuracy_score\":acc,\n",
    "    })\n",
    "\n",
    "results_df = pd.DataFrame(results)\n",
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "971c4fe9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8878504672897196"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_model = models[\"AdaBoost_C\"]\n",
    "y_pred = final_model.predict(test_pad)\n",
    "accuracy_score(y_test,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "380649e0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['columns.pkl']"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from joblib import dump\n",
    "\n",
    "dump(final_model,\"model.pkl\")\n",
    "dump(tokenizer,\"tokenizer.pkl\")\n",
    "dump(X.columns.tolist(),\"columns.pkl\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}