jupyter_ml_.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv('Resume.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>Category</th>\n",
       "      <th>Resume</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Skills * Programming Languages: Python (pandas...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Education Details \\r\\nMay 2013 to May 2017 B.E...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Areas of Interest Deep Learning, Control Syste...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Skills â¢ R â¢ Python â¢ SAP HANA â¢ Table...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Education Details \\r\\n MCA   YMCAUST,  Faridab...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>957</th>\n",
       "      <td>Testing</td>\n",
       "      <td>Computer Skills: â¢ Proficient in MS office (...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>958</th>\n",
       "      <td>Testing</td>\n",
       "      <td>â Willingness to accept the challenges. â ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>959</th>\n",
       "      <td>Testing</td>\n",
       "      <td>PERSONAL SKILLS â¢ Quick learner, â¢ Eagerne...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>960</th>\n",
       "      <td>Testing</td>\n",
       "      <td>COMPUTER SKILLS &amp; SOFTWARE KNOWLEDGE MS-Power ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>961</th>\n",
       "      <td>Testing</td>\n",
       "      <td>Skill Set OS Windows XP/7/8/8.1/10 Database MY...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>962 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Category                                             Resume\n",
       "0    Data Science  Skills * Programming Languages: Python (pandas...\n",
       "1    Data Science  Education Details \\r\\nMay 2013 to May 2017 B.E...\n",
       "2    Data Science  Areas of Interest Deep Learning, Control Syste...\n",
       "3    Data Science  Skills â¢ R â¢ Python â¢ SAP HANA â¢ Table...\n",
       "4    Data Science  Education Details \\r\\n MCA   YMCAUST,  Faridab...\n",
       "..            ...                                                ...\n",
       "957       Testing  Computer Skills: â¢ Proficient in MS office (...\n",
       "958       Testing  â Willingness to accept the challenges. â ...\n",
       "959       Testing  PERSONAL SKILLS â¢ Quick learner, â¢ Eagerne...\n",
       "960       Testing  COMPUTER SKILLS & SOFTWARE KNOWLEDGE MS-Power ...\n",
       "961       Testing  Skill Set OS Windows XP/7/8/8.1/10 Database MY...\n",
       "\n",
       "[962 rows x 2 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      Skills * Programming Languages: Python (pandas...\n",
       "1      Education Details \\r\\nMay 2013 to May 2017 B.E...\n",
       "2      Areas of Interest Deep Learning, Control Syste...\n",
       "3      Skills â¢ R â¢ Python â¢ SAP HANA â¢ Table...\n",
       "4      Education Details \\r\\n MCA   YMCAUST,  Faridab...\n",
       "                             ...                        \n",
       "957    Computer Skills: â¢ Proficient in MS office (...\n",
       "958    â Willingness to accept the challenges. â ...\n",
       "959    PERSONAL SKILLS â¢ Quick learner, â¢ Eagerne...\n",
       "960    COMPUTER SKILLS & SOFTWARE KNOWLEDGE MS-Power ...\n",
       "961    Skill Set OS Windows XP/7/8/8.1/10 Database MY...\n",
       "Name: Resume, Length: 962, dtype: object"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Resume']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Category    0\n",
       "Resume      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "k=df['Category'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Java Developer', 'Testing', 'DevOps Engineer', 'Python Developer',\n",
       "       'Web Designing', 'HR', 'Hadoop', 'Data Science', 'Sales',\n",
       "       'Mechanical Engineer', 'ETL Developer', 'Operations Manager',\n",
       "       'Blockchain', 'Arts', 'Database', 'Electrical Engineering',\n",
       "       'Health and fitness', 'PMO', 'DotNet Developer', 'Business Analyst',\n",
       "       'Automation Testing', 'Network Security Engineer', 'SAP Developer',\n",
       "       'Civil Engineer', 'Advocate'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "k.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "values=k.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "counts=k.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([84, 70, 55, 48, 45, 44, 42, 40, 40, 40, 40, 40, 40, 36, 33, 30, 30,\n",
       "       30, 28, 28, 26, 25, 24, 24, 20], dtype=int64)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize =(16, 9))\n",
    "ax.barh(values, counts)\n",
    "for s in ['top', 'bottom', 'left', 'right']:\n",
    "    ax.spines[s].set_visible(False)\n",
    "ax.xaxis.set_ticks_position('none')\n",
    "ax.yaxis.set_ticks_position('none')\n",
    "ax.xaxis.set_tick_params(pad = 5)\n",
    "ax.yaxis.set_tick_params(pad = 10)\n",
    "ax.invert_yaxis()\n",
    "for i in ax.patches:\n",
    "    plt.text(i.get_width()+0.2, i.get_y()+0.5,\n",
    "             str(round((i.get_width()), 2)),\n",
    "             fontsize = 10, fontweight ='bold',\n",
    "             color ='grey')\n",
    "fig.text(0.9, 0.15, 'Niranjan', fontsize = 12,\n",
    "         color ='grey', ha ='right', va ='bottom',\n",
    "         alpha = 0.7)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "import nltk\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.corpus import stopwords"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from wordcloud import WordCloud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['cleaned_text']=''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>Category</th>\n",
       "      <th>Resume</th>\n",
       "      <th>cleaned_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Skills * Programming Languages: Python (pandas...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Education Details \\r\\nMay 2013 to May 2017 B.E...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Areas of Interest Deep Learning, Control Syste...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Skills â¢ R â¢ Python â¢ SAP HANA â¢ Table...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Education Details \\r\\n MCA   YMCAUST,  Faridab...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>957</th>\n",
       "      <td>Testing</td>\n",
       "      <td>Computer Skills: â¢ Proficient in MS office (...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>958</th>\n",
       "      <td>Testing</td>\n",
       "      <td>â Willingness to accept the challenges. â ...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>959</th>\n",
       "      <td>Testing</td>\n",
       "      <td>PERSONAL SKILLS â¢ Quick learner, â¢ Eagerne...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>960</th>\n",
       "      <td>Testing</td>\n",
       "      <td>COMPUTER SKILLS &amp; SOFTWARE KNOWLEDGE MS-Power ...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>961</th>\n",
       "      <td>Testing</td>\n",
       "      <td>Skill Set OS Windows XP/7/8/8.1/10 Database MY...</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>962 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Category                                             Resume  \\\n",
       "0    Data Science  Skills * Programming Languages: Python (pandas...   \n",
       "1    Data Science  Education Details \\r\\nMay 2013 to May 2017 B.E...   \n",
       "2    Data Science  Areas of Interest Deep Learning, Control Syste...   \n",
       "3    Data Science  Skills â¢ R â¢ Python â¢ SAP HANA â¢ Table...   \n",
       "4    Data Science  Education Details \\r\\n MCA   YMCAUST,  Faridab...   \n",
       "..            ...                                                ...   \n",
       "957       Testing  Computer Skills: â¢ Proficient in MS office (...   \n",
       "958       Testing  â Willingness to accept the challenges. â ...   \n",
       "959       Testing  PERSONAL SKILLS â¢ Quick learner, â¢ Eagerne...   \n",
       "960       Testing  COMPUTER SKILLS & SOFTWARE KNOWLEDGE MS-Power ...   \n",
       "961       Testing  Skill Set OS Windows XP/7/8/8.1/10 Database MY...   \n",
       "\n",
       "    cleaned_text  \n",
       "0                 \n",
       "1                 \n",
       "2                 \n",
       "3                 \n",
       "4                 \n",
       "..           ...  \n",
       "957               \n",
       "958               \n",
       "959               \n",
       "960               \n",
       "961               \n",
       "\n",
       "[962 rows x 3 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "##clean th resume\n",
    "##remove tags special non essential charecters\n",
    "\n",
    "def clean(resume):\n",
    "    resume=re.sub('httpS+s*',' ',resume)\n",
    "    resume=re.sub('RT|cc',' ',resume)\n",
    "    resume=re.sub('#S+',' ',resume)\n",
    "    resume=re.sub('@S+','  ',resume)\n",
    "    resume=re.sub('[%s]' % re.escape(\"\"\"!\"#$%&'()*+,-./:;<=>?@[]^_`{|}~\"\"\"), ' ', resume)\n",
    "    resume=re.sub(r'[^x00-x7f]',r' ', resume)\n",
    "    resume=re.sub('s+',' ',resume)\n",
    "    resume=resume.lower()\n",
    "    return resume\n",
    "\n",
    "    \n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['cleaned_text']=df['Resume'].apply(lambda x: clean(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>Category</th>\n",
       "      <th>Resume</th>\n",
       "      <th>cleaned_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Skills * Programming Languages: Python (pandas...</td>\n",
       "      <td>skill    programming language   p thon  panda ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Education Details \\r\\nMay 2013 to May 2017 B.E...</td>\n",
       "      <td>education detail    ma  2013 to ma  2017 b e  ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Areas of Interest Deep Learning, Control Syste...</td>\n",
       "      <td>area  of intere t deep learning  control s  te...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Skills â¢ R â¢ Python â¢ SAP HANA â¢ Table...</td>\n",
       "      <td>skill      r     p thon     sap hana     table...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Data Science</td>\n",
       "      <td>Education Details \\r\\n MCA   YMCAUST,  Faridab...</td>\n",
       "      <td>education detail     mca   ymcaust   faridabad...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>957</th>\n",
       "      <td>Testing</td>\n",
       "      <td>Computer Skills: â¢ Proficient in MS office (...</td>\n",
       "      <td>computer skill       proficient in ms office  ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>958</th>\n",
       "      <td>Testing</td>\n",
       "      <td>â Willingness to accept the challenges. â ...</td>\n",
       "      <td>willingne  to a ept the challenge       po...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>959</th>\n",
       "      <td>Testing</td>\n",
       "      <td>PERSONAL SKILLS â¢ Quick learner, â¢ Eagerne...</td>\n",
       "      <td>personal skills     quick learner      eagerne...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>960</th>\n",
       "      <td>Testing</td>\n",
       "      <td>COMPUTER SKILLS &amp; SOFTWARE KNOWLEDGE MS-Power ...</td>\n",
       "      <td>computer skills   software knowledge ms power ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>961</th>\n",
       "      <td>Testing</td>\n",
       "      <td>Skill Set OS Windows XP/7/8/8.1/10 Database MY...</td>\n",
       "      <td>skill set os window  xp 7 8 8 1 10 databa e my...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>962 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Category                                             Resume  \\\n",
       "0    Data Science  Skills * Programming Languages: Python (pandas...   \n",
       "1    Data Science  Education Details \\r\\nMay 2013 to May 2017 B.E...   \n",
       "2    Data Science  Areas of Interest Deep Learning, Control Syste...   \n",
       "3    Data Science  Skills â¢ R â¢ Python â¢ SAP HANA â¢ Table...   \n",
       "4    Data Science  Education Details \\r\\n MCA   YMCAUST,  Faridab...   \n",
       "..            ...                                                ...   \n",
       "957       Testing  Computer Skills: â¢ Proficient in MS office (...   \n",
       "958       Testing  â Willingness to accept the challenges. â ...   \n",
       "959       Testing  PERSONAL SKILLS â¢ Quick learner, â¢ Eagerne...   \n",
       "960       Testing  COMPUTER SKILLS & SOFTWARE KNOWLEDGE MS-Power ...   \n",
       "961       Testing  Skill Set OS Windows XP/7/8/8.1/10 Database MY...   \n",
       "\n",
       "                                          cleaned_text  \n",
       "0    skill    programming language   p thon  panda ...  \n",
       "1    education detail    ma  2013 to ma  2017 b e  ...  \n",
       "2    area  of intere t deep learning  control s  te...  \n",
       "3    skill      r     p thon     sap hana     table...  \n",
       "4    education detail     mca   ymcaust   faridabad...  \n",
       "..                                                 ...  \n",
       "957  computer skill       proficient in ms office  ...  \n",
       "958      willingne  to a ept the challenge       po...  \n",
       "959  personal skills     quick learner      eagerne...  \n",
       "960  computer skills   software knowledge ms power ...  \n",
       "961  skill set os window  xp 7 8 8 1 10 databa e my...  \n",
       "\n",
       "[962 rows x 3 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "962"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "corpus=\" \"\n",
    "for i in range(0,962):\n",
    "    corpus= corpus+ df[\"cleaned_text\"][i]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'urance  technology assisted review  tar  technolog  a i ted review  a i t  in a elerating the review proce  and run anal tic  and generate report      core member of a team helped in developing automated review platform tool from  cratch for a i ting e di cover  domain  thi  tool implement  predictive coding and topic modelling b  automating review   re ulting in reduced labor co t  and time  pent during the law er  review     under tand the end to end flow of the  olution  doing re earch and development for cla ification model   predictive anal  i  and mining of the information pre ent in text data  worked on anal  ing the output  and preci ion monitoring for the entire tool     tar a i t  in predictive coding  topic modelling from the evidence b  following ey  tandard   developed the cla ifier model  in order to identif   red flag   and fraud related i ue      tool    technologie   p thon   cikit learn  tfidf  word2vec  doc2vec  co ine  imilarit   na  ve ba e   lda  nmf for topic modelling  vader and text blob for  entiment anal  i   matplot lib  tableau da hboard for reporting     multiple data science and analytic projects  usa clients   text analytics   motor vehicle customer review data   received cu tomer feedback  urve  data for pa t one  ear  performed  entiment  po itive  negative   neutral  and time  erie  anal  i  on cu tomer comment  acro  all 4 categorie      created heat map of term  b   urve  categor  ba ed on frequenc  of word    extracted po itive and negati'"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corpus[1000:2500]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "le=LabelEncoder()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "y=df['Category']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "y=le.fit_transform(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,\n",
       "        6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,  6,\n",
       "        6,  6,  6,  6,  6,  6, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12,\n",
       "       12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12,\n",
       "       12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12,  0,\n",
       "        0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,\n",
       "        0,  0,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,  1,\n",
       "        1,  1,  1,  1, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,\n",
       "       24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24,\n",
       "       24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 16, 16,\n",
       "       16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,\n",
       "       16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,\n",
       "       16, 16, 16, 16, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,\n",
       "       22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22,\n",
       "       22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 14, 14, 14, 14, 14, 14, 14,\n",
       "       14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14,\n",
       "       14, 14, 14, 14, 14, 14,  5,  5,  5,  5,  5,  5,  5,  5,  5,  5,  5,\n",
       "        5,  5,  5,  5,  5,  5,  5,  5,  5,  5,  5,  5,  5, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,  4,  4,  4,  4,  4,\n",
       "        4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,  4,\n",
       "        4,  4,  4,  4,  4,  4, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,\n",
       "       21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,  2,  2,  2,  2,\n",
       "        2,  2,  2,  2,  2,  2,  2,  2,  2,  2,  2,  2,  2,  2,  2,  2,  2,\n",
       "        2,  2,  2,  2,  2, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11,\n",
       "       11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11,\n",
       "       11, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,\n",
       "       18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18, 18,\n",
       "       18, 18, 18, 18, 18, 18, 18, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,\n",
       "       20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,\n",
       "       20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20,\n",
       "       20, 20, 20, 20,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,\n",
       "        8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,\n",
       "        8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,  8,\n",
       "        8,  8,  8,  8,  8,  8,  8,  8, 17, 17, 17, 17, 17, 17, 17, 17, 17,\n",
       "       17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 17, 19,\n",
       "       19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,\n",
       "       19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19,  7,  7,  7,  7,  7,\n",
       "        7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7,\n",
       "        7,  7,  7,  7,  7,  7,  7,  7,  7,  7,  7, 13, 13, 13, 13, 13, 13,\n",
       "       13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13,\n",
       "       13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13,\n",
       "       13, 13, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10,\n",
       "       10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10, 10,\n",
       "       10, 10, 10, 10, 10, 10, 10, 10,  9,  9,  9,  9,  9,  9,  9,  9,  9,\n",
       "        9,  9,  9,  9,  9,  9,  9,  9,  9,  9,  9,  9,  9,  9,  9,  9,  9,\n",
       "        9,  9,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,\n",
       "        3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,\n",
       "        3,  3,  3,  3,  3,  3,  3,  3, 23, 23, 23, 23, 23, 23, 23, 23, 23,\n",
       "       23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,\n",
       "       23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,\n",
       "       23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23, 23,\n",
       "       23, 23, 23, 23, 23, 23, 23, 23, 23, 23])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=df['cleaned_text']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer,CountVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "vect=CountVectorizer(encoding='utf-8',lowercase=False)\n",
    "tf=TfidfVectorizer(max_features=2000,sublinear_tf=True,norm='l1',stop_words='english',lowercase=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "#vect.fit(X)\n",
    "#X=vect.transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TfidfVectorizer(lowercase=False, max_features=2000, norm='l1',\n",
       "                stop_words='english', sublinear_tf=True)"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_trans=tf.transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import StratifiedShuffleSplit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.multiclass import OneVsRestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "lr=OneVsRestClassifier(LogisticRegression())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "sgd=SGDClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import f1_score,accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "m=OneVsRestClassifier(MultinomialNB())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "s=StratifiedShuffleSplit(n_splits=5,test_size=0.3,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "f1=[]\n",
    "accuracy=[]\n",
    "clf=OneVsRestClassifier(KNeighborsClassifier())\n",
    "for t_index,te_index in s.split(X_trans,y):\n",
    "    X_train,y_train=X_trans[t_index],y[t_index]\n",
    "    X_test,y_test=X_trans[te_index],y[te_index]\n",
    "    sgd.fit(X_train,y_train)\n",
    "    f1.append(f1_score(y_test,sgd.predict(X_test),average='micro'))\n",
    "    accuracy.append(accuracy_score(y_test,sgd.predict(X_test)))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.9965397923875432,\n",
       " 0.9965397923875432,\n",
       " 0.986159169550173,\n",
       " 0.9930795847750865,\n",
       " 0.9896193771626297]"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.9965397923875432,\n",
       " 0.9965397923875432,\n",
       " 0.986159169550173,\n",
       " 0.9930795847750865,\n",
       " 0.9896193771626297]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_t,X_te,y_t,y_te=train_test_split(X_trans,y,test_size=0.2,random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OneVsRestClassifier(estimator=MultinomialNB())"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m.fit(X_t,y_t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=m.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 12, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 12, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 12, 15,\n",
       "       15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15,\n",
       "       15, 15, 12, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 12])"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.multiclass import OneVsRestClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = OneVsRestClassifier(KNeighborsClassifier())\n",
    "clf.fit(X_t, y_t)\n",
    "prediction = clf.predict(X_te)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9222797927461139\n"
     ]
    }
   ],
   "source": [
    "print(accuracy_score(y_te,prediction))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "di={\n",
    "    'C':[1.0,0.9,0.8,1.2,1.5,2,0.5],\n",
    "    'kernel':['rbf','linear']\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "search=RandomizedSearchCV(SVC(),di,n_iter=10,cv=5,random_state=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "k=search.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'mean_fit_time': array([0.26016054, 0.1701046 , 0.25721765, 0.15109754, 0.25802503,\n",
       "        0.26113739, 0.25922174, 0.10923004, 0.25945783, 0.21423697]),\n",
       " 'std_fit_time': array([0.0031369 , 0.00213354, 0.00486291, 0.00206037, 0.00377484,\n",
       "        0.00561412, 0.00430829, 0.0015906 , 0.00551678, 0.00281887]),\n",
       " 'mean_score_time': array([0.04245906, 0.03488688, 0.04231687, 0.03192763, 0.04191852,\n",
       "        0.04241352, 0.04271631, 0.0228384 , 0.04232597, 0.04140606]),\n",
       " 'std_score_time': array([0.00165752, 0.00188949, 0.00092385, 0.00129184, 0.00137144,\n",
       "        0.00135282, 0.00140652, 0.00075461, 0.0007439 , 0.00139166]),\n",
       " 'param_kernel': masked_array(data=['linear', 'rbf', 'linear', 'rbf', 'linear', 'linear',\n",
       "                    'linear', 'rbf', 'linear', 'rbf'],\n",
       "              mask=[False, False, False, False, False, False, False, False,\n",
       "                    False, False],\n",
       "        fill_value='?',\n",
       "             dtype=object),\n",
       " 'param_C': masked_array(data=[2, 0.8, 0.8, 1.0, 1.5, 0.9, 1.0, 2, 1.2, 0.5],\n",
       "              mask=[False, False, False, False, False, False, False, False,\n",
       "                    False, False],\n",
       "        fill_value='?',\n",
       "             dtype=object),\n",
       " 'params': [{'kernel': 'linear', 'C': 2},\n",
       "  {'kernel': 'rbf', 'C': 0.8},\n",
       "  {'kernel': 'linear', 'C': 0.8},\n",
       "  {'kernel': 'rbf', 'C': 1.0},\n",
       "  {'kernel': 'linear', 'C': 1.5},\n",
       "  {'kernel': 'linear', 'C': 0.9},\n",
       "  {'kernel': 'linear', 'C': 1.0},\n",
       "  {'kernel': 'rbf', 'C': 2},\n",
       "  {'kernel': 'linear', 'C': 1.2},\n",
       "  {'kernel': 'rbf', 'C': 0.5}],\n",
       " 'split0_test_score': array([0.1037037 , 0.97037037, 0.08148148, 0.97037037, 0.1037037 ,\n",
       "        0.08148148, 0.08148148, 0.97037037, 0.08148148, 0.94814815]),\n",
       " 'split1_test_score': array([0.11851852, 0.95555556, 0.08888889, 0.98518519, 0.1037037 ,\n",
       "        0.08888889, 0.08888889, 0.98518519, 0.0962963 , 0.94814815]),\n",
       " 'split2_test_score': array([0.11111111, 1.        , 0.08888889, 1.        , 0.1037037 ,\n",
       "        0.08888889, 0.08888889, 1.        , 0.1037037 , 0.97037037]),\n",
       " 'split3_test_score': array([0.10447761, 0.92537313, 0.08955224, 0.94029851, 0.10447761,\n",
       "        0.08955224, 0.08955224, 0.97761194, 0.08955224, 0.91791045]),\n",
       " 'split4_test_score': array([0.09701493, 0.98507463, 0.08955224, 0.98507463, 0.09701493,\n",
       "        0.08955224, 0.08955224, 0.99253731, 0.09701493, 0.98507463]),\n",
       " 'mean_test_score': array([0.10696517, 0.96727474, 0.08767275, 0.97618574, 0.10252073,\n",
       "        0.08767275, 0.08767275, 0.98514096, 0.09360973, 0.95393035]),\n",
       " 'std_test_score': array([0.0073007 , 0.02565325, 0.00310982, 0.02024269, 0.00276917,\n",
       "        0.00310982, 0.00310982, 0.01049148, 0.00754001, 0.02282341]),\n",
       " 'rank_test_score': array([5, 3, 8, 2, 6, 8, 8, 1, 7, 4])}"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "k.cv_results_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'kernel': 'rbf', 'C': 2}"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "k.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "sv=SVC(C=2,kernel='rbf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=2)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sv.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.986159169550173"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1_score(y_test,sv.predict(X_test),average='micro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 8,  1, 15, 15, 22, 20, 16, 14,  2, 13,  1, 11, 17,  3,  0, 15,  1,\n",
       "        5, 23, 15, 10, 18,  4,  6, 19,  7, 15,  8, 11, 24, 23, 21, 24, 14,\n",
       "       20, 18, 15, 10,  9, 20,  9,  4,  4, 20,  1, 16, 24, 12, 21,  2,  9,\n",
       "       21, 22,  1,  9, 17, 16, 16, 16, 15, 15,  8, 11, 23, 19, 24, 23, 13,\n",
       "       23, 11, 21, 24, 12,  2, 24, 15, 23,  4, 24, 14,  7, 22, 24, 13,  1,\n",
       "        8, 15, 15, 22,  3,  3, 15, 23,  6, 22, 18, 11, 24, 15, 10,  8,  8,\n",
       "       13, 12,  0, 16,  7, 10, 17,  6,  2, 10, 17,  7,  3, 23,  7,  6,  8,\n",
       "        8,  6, 23, 17, 11,  8, 10,  3, 15, 23,  1,  1, 23,  5, 23, 19, 20,\n",
       "        4, 12, 10,  6, 19,  7, 15,  5, 10, 18, 21, 13,  8, 23, 18, 15, 18,\n",
       "       23, 12, 21, 17, 11, 16, 15,  3, 15, 19, 23, 13, 22, 12, 15, 15, 19,\n",
       "       23,  6,  8, 15,  3, 19, 15,  8, 18, 19,  8,  3, 16, 23, 24, 22,  9,\n",
       "       16, 23,  1,  2,  9, 12,  3, 12, 20, 18, 24, 13, 12,  7, 20, 14,  6,\n",
       "       14, 22, 12, 11,  3,  7, 23, 12,  4,  9, 20, 20, 14, 22,  8, 10,  0,\n",
       "       24, 10, 20,  8, 23, 13,  7,  5,  6,  5, 15, 13,  9, 11, 10, 23, 18,\n",
       "        6,  6,  7, 17, 15, 24,  0,  1, 20, 12, 16,  1, 20,  6,  3, 12, 18,\n",
       "       14, 16, 13, 13,  5, 13,  4, 15, 18,  3, 17, 10, 20, 22, 22, 15, 24,\n",
       "        5,  8, 12, 14,  8,  2, 13, 21, 22, 16, 14, 15,  4, 19, 18, 20, 12])"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sv.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import VotingClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "vi=VotingClassifier(estimators=[\n",
    "    ('LogisticRegression',lr),\n",
    "    ('SGD classifier',sgd),\n",
    "    ('OnevsRestKnn',clf),\n",
    "    ('SVM_rbf_kernel',sv)\n",
    "    #(random forest),\n",
    "    #(XGboost),\n",
    "    #()\n",
    "],voting='hard')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from sklearn.metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "f1=[]\n",
    "accuracy=[]\n",
    "clf=OneVsRestClassifier(KNeighborsClassifier())\n",
    "for t_index,te_index in s.split(X_trans,y):\n",
    "    X_train,y_train=X_trans[t_index],y[t_index]\n",
    "    X_test,y_test=X_trans[te_index],y[te_index]\n",
    "    vi.fit(X_train,y_train)\n",
    "    f1.append(f1_score(y_test,vi.predict(X_test),average='micro'))\n",
    "    accuracy.append(accuracy_score(y_test,vi.predict(X_test)))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.986159169550173,\n",
       " 1.0,\n",
       " 0.9826989619377162,\n",
       " 0.9930795847750865,\n",
       " 0.9930795847750865]"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.986159169550173,\n",
       " 1.0,\n",
       " 0.9826989619377162,\n",
       " 0.9930795847750865,\n",
       " 0.9930795847750865]"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'slate'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-89-e532bb050ccb>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[1;32mimport\u001b[0m \u001b[0mslate\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'slate'"
     ]
    }
   ],
   "source": [
    "import slate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'/CropBox': [0, 0, 612, 792], '/Parent': IndirectObject(1, 0), '/Contents': [IndirectObject(9, 0), IndirectObject(11, 0), IndirectObject(13, 0), IndirectObject(14, 0), IndirectObject(15, 0), IndirectObject(17, 0), IndirectObject(18, 0), IndirectObject(19, 0)], '/Rotate': 0, '/MediaBox': [0, 0, 612, 792], '/Resources': {'/Font': {'/F15': IndirectObject(7, 0), '/F16': IndirectObject(8, 0), '/F17': IndirectObject(10, 0), '/F18': IndirectObject(12, 0), '/F19': IndirectObject(16, 0)}, '/ProcSet': ['/PDF', '/Text', '/ImageB', '/ImageC']}, '/Type': '/Page'}\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "PdfReadWarning: Xref table not zero-indexed. ID numbers for objects will be corrected. [pdf.py:1736]\n"
     ]
    }
   ],
   "source": [
    "import PyPDF2\n",
    "fl = open(\"c10.pdf\",\"rb\")\n",
    "reader = PyPDF2.PdfFileReader(fl)\n",
    "page1 = reader.getPage(0)\n",
    "print(page1)\n",
    "pdfData = page1.extractText()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'filename' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-104-d8a55aabb363>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mf\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mopen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"CVs\"\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0mfilename\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m\"rb\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mdoc\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mslate\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mPDF\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mf\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0meach_resume\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;34m\"\"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mj\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdoc\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m         \u001b[0meach_resume\u001b[0m \u001b[1;33m+=\u001b[0m \u001b[0mdoc\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mj\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'filename' is not defined"
     ]
    }
   ],
   "source": [
    "f = open(\"CVs\" + filename, \"rb\")\n",
    "doc = slate.PDF(f)\n",
    "each_resume = \"\"\t\n",
    "for j in range(len(doc)):\n",
    "        each_resume += doc[j]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'RAHUL SRIVASTAVARoom No: C-342, LLR Hallof Residence , IIT Kharagpur\\nPh: 9732961707, Email : \\nrahusr@gmail.com\\n                                         \\nAcademic Qualifications\\nExamination Year Board/University\\n%/CGPA Dual Degree(Hons.),Computer Science & Engineering\\n2011IIT Kharagpur8.18/10Class XII2003CBSE88%Class X2001CBSE84%Academic Distinctions\\nAll India Rank 482\\nin IIT-JEEexamination among 300,000 candidates, 2006\\nState Rank 55\\nin UPSEAT(Uttar Pradesh State Engineering Admission Test) among 80,000 candidates, 2006\\nAll India Rank 1078\\n(with 98.99 percentile\\n) in GATEexamination among 107,086 candidates, 2010.\\n9.75/10 GPA\\nin 8thsemester (10/10in Project, 10/10 in 4 Courses) at IIT Kharagpur, 2010\\nRecipient of GATE MHRD scholarship\\ngiven by Gov. of India, May 2010 to April 2011\\nOffered scholarshipfor summer internship at Universidad de Guanajuatoin \\nMexico\\n, 2010 Work Experience/ Internship\\nLaboratory of Vision, Robotics and Artificial Intelligence, UGTO, Mexico\\n                 (20\\nthMay-20\\nthJuly,2010)Project Name:\\nReducing Complexity of 3D Mesh\\nArea:Computer Graphics\\n3D objects are represented as a set of polygons, PLY files represents this set as a text file.\\nAim was to reduce the number of polygonsrepresenting the 3D object, keeping visual similarity intact.\\nSoftware made(inC++using OpenGL) to parse the PLY fileand display its content.\\nAn innovative approachfor polygon reduction using curvature estimate proposed.\\nNew approach successfully implemented &\\n15-20 % reduction\\nin number of polygons achieved.\\nTata Research Development & Design Center(TRDDC), Pune                                  \\n(10thMay Œ10\\nthJuly, 2009)Project Name: \\nQuest 2009 (Web Application)\\nArea:Databases(MDX & OLAP), GUI\\nAim was to create a survey analysis toolin the form of a web interfacefor cubedata structure.\\nMDX(Multi Dimensional Expression) queries were mappedtoanintuitive 2D GUI.\\nWebsite successfully madedisplaying MDX queriesin the form of tables and charts.\\n3000 lines of codewritten in C#&ASP.NET. Visual Studio 2005and SQL Server 2005were used.\\nHighest Grade 10/10in summer internship evaluation at TRDDC and IIT Kharagpur, 2009\\nProjects\\nDesign Lab, IIT Kharagpur\\n                                                                                  (20\\nthJuly Œ20\\nth  November, 2010)\\nProject Name: \\nCreating 3D photographs and videos\\nArea:3D Technology\\nEasy way of creating anaglyph 3D photos and videossuggested using normal 2D camera.\\nProgram writtenwhich took as input two images of the same object from slightly different perspectives. \\nOutputwas a \\n3D anaglyph image\\n. Similar technique used to produce \\n3D anaglyph video\\n.Program written in C++using anopen sourcelibrary \\nImageMagick\\n.B.Tech. Project, IIT Kharagpur\\n                                                                                      (August 2009 ŒApril 2010)\\nProject Name: \\nEstimating curvature of digital curves\\nArea:Computational Geometry, Graphics\\nAn innovative approachusing digital osculatingcircle proposed for curvature estimationof digital curves..\\nIdea was to approximate a part of the digital curve withbest fitting osculating circle\\nThenew algorithmwas implemented in Java using NetBeans6.5\\nIt produced improved resultthan all other known methods and used only integer calculations\\nPositionsof  Responsibility\\nTeaching Assistant for Computer Organization and Architecture Lab, CSE , 20\\nthJuly Œ30\\nthNovember 2010\\nMember of sponsorship team of BitWise 2010, annual programming contest organized by CSE, IIT Kharagpur\\nExtra Curricular Activities\\nNational Cadet Corpse (NCC)of Indian Army, Annual Training Camp: completed successfully, 2006.\\nParticipant of Inter Hall English Dramaticscompetition, 2008.\\nInvolved in teaching70 studentsof KendriyaVidyalaya, IIT KGPas a part of National Services Scheme(NSS) \\nVolunteer, MANZARannual literary fest of IIT Kharagpur. \\n'"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pdfData"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "text='''Skills * Programming Languages: Python (pandas, numpy, scipy, scikit-learn, matplotlib), Sql, Java, JavaScript/JQuery. * Machine learning: Regression, SVM, NaÃ¯ve Bayes, KNN, Random Forest, Decision Trees, Boosting techniques, Cluster Analysis, Word Embedding, Sentiment Analysis, Natural Language processing, Dimensionality reduction, Topic Modelling (LDA, NMF), PCA & Neural Nets. * Database Visualizations: Mysql, SqlServer, Cassandra, Hbase, ElasticSearch D3.js, DC.js, Plotly, kibana, matplotlib, ggplot, Tableau. * Others: Regular Expression, HTML, CSS, Angular 6, Logstash, Kafka, Python Flask, Git, Docker, computer vision - Open CV and understanding of Deep learning.Education Details \n",
    "\n",
    "Data Science Assurance Associate \n",
    "\n",
    "Data Science Assurance Associate - Ernst & Young LLP\n",
    "Skill Details \n",
    "JAVASCRIPT- Exprience - 24 months\n",
    "jQuery- Exprience - 24 months\n",
    "Python- Exprience - 24 monthsCompany Details \n",
    "company - Ernst & Young LLP\n",
    "description - Fraud Investigations and Dispute Services   Assurance\n",
    "TECHNOLOGY ASSISTED REVIEW\n",
    "TAR (Technology Assisted Review) assists in accelerating the review process and run analytics and generate reports.\n",
    "* Core member of a team helped in developing automated review platform tool from scratch for assisting E discovery domain, this tool implements predictive coding and topic modelling by automating reviews, resulting in reduced labor costs and time spent during the lawyers review.\n",
    "* Understand the end to end flow of the solution, doing research and development for classification models, predictive analysis and mining of the information present in text data. Worked on analyzing the outputs and precision monitoring for the entire tool.\n",
    "* TAR assists in predictive coding, topic modelling from the evidence by following EY standards. Developed the classifier models in order to identify \"red flags\" and fraud-related issues.\n",
    "\n",
    "Tools & Technologies: Python, scikit-learn, tfidf, word2vec, doc2vec, cosine similarity, NaÃ¯ve Bayes, LDA, NMF for topic modelling, Vader and text blob for sentiment analysis. Matplot lib, Tableau dashboard for reporting.\n",
    "\n",
    "MULTIPLE DATA SCIENCE AND ANALYTIC PROJECTS (USA CLIENTS)\n",
    "TEXT ANALYTICS - MOTOR VEHICLE CUSTOMER REVIEW DATA * Received customer feedback survey data for past one year. Performed sentiment (Positive, Negative & Neutral) and time series analysis on customer comments across all 4 categories.\n",
    "* Created heat map of terms by survey category based on frequency of words * Extracted Positive and Negative words across all the Survey categories and plotted Word cloud.\n",
    "* Created customized tableau dashboards for effective reporting and visualizations.\n",
    "CHATBOT * Developed a user friendly chatbot for one of our Products which handle simple questions about hours of operation, reservation options and so on.\n",
    "* This chat bot serves entire product related questions. Giving overview of tool via QA platform and also give recommendation responses so that user question to build chain of relevant answer.\n",
    "* This too has intelligence to build the pipeline of questions as per user requirement and asks the relevant /recommended questions.\n",
    "\n",
    "Tools & Technologies: Python, Natural language processing, NLTK, spacy, topic modelling, Sentiment analysis, Word Embedding, scikit-learn, JavaScript/JQuery, SqlServer\n",
    "\n",
    "INFORMATION GOVERNANCE\n",
    "Organizations to make informed decisions about all of the information they store. The integrated Information Governance portfolio synthesizes intelligence across unstructured data sources and facilitates action to ensure organizations are best positioned to counter information risk.\n",
    "* Scan data from multiple sources of formats and parse different file formats, extract Meta data information, push results for indexing elastic search and created customized, interactive dashboards using kibana.\n",
    "* Preforming ROT Analysis on the data which give information of data which helps identify content that is either Redundant, Outdated, or Trivial.\n",
    "* Preforming full-text search analysis on elastic search with predefined methods which can tag as (PII) personally identifiable information (social security numbers, addresses, names, etc.) which frequently targeted during cyber-attacks.\n",
    "Tools & Technologies: Python, Flask, Elastic Search, Kibana\n",
    "\n",
    "FRAUD ANALYTIC PLATFORM\n",
    "Fraud Analytics and investigative platform to review all red flag cases.\n",
    "â¢ FAP is a Fraud Analytics and investigative platform with inbuilt case manager and suite of Analytics for various ERP systems.\n",
    "* It can be used by clients to interrogate their Accounting systems for identifying the anomalies which can be indicators of fraud by running advanced analytics\n",
    "Tools & Technologies: HTML, JavaScript, SqlServer, JQuery, CSS, Bootstrap, Node.js, D3.js, DC.js'''\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "cleaned_text=clean(pdfData)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "cleaned_text=[cleaned_text]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "idf=tf.transform(cleaned_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<1x2000 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 162 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "idf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics.pairwise import cosine_similarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_index_from_title(title):\n",
    "    return df[df.Category == title][\"index\"].values[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "similarity=[]\n",
    "for i in range(962):\n",
    "    c=cosine_similarity(X_trans[i],idf)\n",
    "    similarity.append((c,y[i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(array([[0.12875242]]), 6),\n",
       " (array([[0.07305243]]), 6),\n",
       " (array([[0.0978387]]), 6),\n",
       " (array([[0.1717803]]), 6),\n",
       " (array([[0.08681563]]), 6),\n",
       " (array([[0.05784156]]), 6),\n",
       " (array([[0.10286779]]), 6),\n",
       " (array([[0.13010398]]), 6),\n",
       " (array([[0.11637348]]), 6),\n",
       " (array([[0.16625304]]), 6),\n",
       " (array([[0.12875242]]), 6),\n",
       " (array([[0.07305243]]), 6),\n",
       " (array([[0.0978387]]), 6),\n",
       " (array([[0.1717803]]), 6),\n",
       " (array([[0.08681563]]), 6),\n",
       " (array([[0.05784156]]), 6),\n",
       " (array([[0.10286779]]), 6),\n",
       " (array([[0.13010398]]), 6),\n",
       " (array([[0.11637348]]), 6),\n",
       " (array([[0.16625304]]), 6),\n",
       " (array([[0.12875242]]), 6),\n",
       " (array([[0.07305243]]), 6),\n",
       " (array([[0.0978387]]), 6),\n",
       " (array([[0.1717803]]), 6),\n",
       " (array([[0.08681563]]), 6),\n",
       " (array([[0.05784156]]), 6),\n",
       " (array([[0.10286779]]), 6),\n",
       " (array([[0.13010398]]), 6),\n",
       " (array([[0.11637348]]), 6),\n",
       " (array([[0.16625304]]), 6),\n",
       " (array([[0.12875242]]), 6),\n",
       " (array([[0.07305243]]), 6),\n",
       " (array([[0.0978387]]), 6),\n",
       " (array([[0.1717803]]), 6),\n",
       " (array([[0.08681563]]), 6),\n",
       " (array([[0.05784156]]), 6),\n",
       " (array([[0.10286779]]), 6),\n",
       " (array([[0.13010398]]), 6),\n",
       " (array([[0.11637348]]), 6),\n",
       " (array([[0.16625304]]), 6),\n",
       " (array([[0.00972124]]), 12),\n",
       " (array([[0.11138335]]), 12),\n",
       " (array([[0.00734833]]), 12),\n",
       " (array([[0.00746898]]), 12),\n",
       " (array([[0.00746898]]), 12),\n",
       " (array([[0.03401845]]), 12),\n",
       " (array([[0.03768677]]), 12),\n",
       " (array([[0.10330955]]), 12),\n",
       " (array([[0.09546385]]), 12),\n",
       " (array([[0.04190392]]), 12),\n",
       " (array([[0.08503756]]), 12),\n",
       " (array([[0.00972124]]), 12),\n",
       " (array([[0.11138335]]), 12),\n",
       " (array([[0.00734833]]), 12),\n",
       " (array([[0.00746898]]), 12),\n",
       " (array([[0.00746898]]), 12),\n",
       " (array([[0.03401845]]), 12),\n",
       " (array([[0.03768677]]), 12),\n",
       " (array([[0.10330955]]), 12),\n",
       " (array([[0.09546385]]), 12),\n",
       " (array([[0.04190392]]), 12),\n",
       " (array([[0.08503756]]), 12),\n",
       " (array([[0.00972124]]), 12),\n",
       " (array([[0.11138335]]), 12),\n",
       " (array([[0.00734833]]), 12),\n",
       " (array([[0.00746898]]), 12),\n",
       " (array([[0.00746898]]), 12),\n",
       " (array([[0.03401845]]), 12),\n",
       " (array([[0.03768677]]), 12),\n",
       " (array([[0.10330955]]), 12),\n",
       " (array([[0.09546385]]), 12),\n",
       " (array([[0.04190392]]), 12),\n",
       " (array([[0.08503756]]), 12),\n",
       " (array([[0.00972124]]), 12),\n",
       " (array([[0.11138335]]), 12),\n",
       " (array([[0.00734833]]), 12),\n",
       " (array([[0.00746898]]), 12),\n",
       " (array([[0.00746898]]), 12),\n",
       " (array([[0.03401845]]), 12),\n",
       " (array([[0.03768677]]), 12),\n",
       " (array([[0.10330955]]), 12),\n",
       " (array([[0.09546385]]), 12),\n",
       " (array([[0.04190392]]), 12),\n",
       " (array([[0.08503756]]), 12),\n",
       " (array([[0.04320077]]), 0),\n",
       " (array([[0.01413643]]), 0),\n",
       " (array([[0.00537592]]), 0),\n",
       " (array([[0.03631038]]), 0),\n",
       " (array([[0.10364461]]), 0),\n",
       " (array([[0.03007]]), 0),\n",
       " (array([[0.0378435]]), 0),\n",
       " (array([[0.04322667]]), 0),\n",
       " (array([[0.03373551]]), 0),\n",
       " (array([[0.06756097]]), 0),\n",
       " (array([[0.04320077]]), 0),\n",
       " (array([[0.01413643]]), 0),\n",
       " (array([[0.00537592]]), 0),\n",
       " (array([[0.03631038]]), 0),\n",
       " (array([[0.10364461]]), 0),\n",
       " (array([[0.03007]]), 0),\n",
       " (array([[0.0378435]]), 0),\n",
       " (array([[0.04322667]]), 0),\n",
       " (array([[0.03373551]]), 0),\n",
       " (array([[0.06756097]]), 0),\n",
       " (array([[0.11000713]]), 1),\n",
       " (array([[0.09841115]]), 1),\n",
       " (array([[0.13727697]]), 1),\n",
       " (array([[0.0331845]]), 1),\n",
       " (array([[0.09579542]]), 1),\n",
       " (array([[0.16013758]]), 1),\n",
       " (array([[0.11000713]]), 1),\n",
       " (array([[0.09841115]]), 1),\n",
       " (array([[0.13727697]]), 1),\n",
       " (array([[0.0331845]]), 1),\n",
       " (array([[0.09579542]]), 1),\n",
       " (array([[0.16013758]]), 1),\n",
       " (array([[0.11000713]]), 1),\n",
       " (array([[0.09841115]]), 1),\n",
       " (array([[0.13727697]]), 1),\n",
       " (array([[0.0331845]]), 1),\n",
       " (array([[0.09579542]]), 1),\n",
       " (array([[0.16013758]]), 1),\n",
       " (array([[0.11000713]]), 1),\n",
       " (array([[0.09841115]]), 1),\n",
       " (array([[0.13727697]]), 1),\n",
       " (array([[0.0331845]]), 1),\n",
       " (array([[0.09579542]]), 1),\n",
       " (array([[0.16013758]]), 1),\n",
       " (array([[0.11000713]]), 1),\n",
       " (array([[0.09841115]]), 1),\n",
       " (array([[0.13727697]]), 1),\n",
       " (array([[0.0331845]]), 1),\n",
       " (array([[0.09579542]]), 1),\n",
       " (array([[0.16013758]]), 1),\n",
       " (array([[0.11000713]]), 1),\n",
       " (array([[0.09841115]]), 1),\n",
       " (array([[0.13727697]]), 1),\n",
       " (array([[0.0331845]]), 1),\n",
       " (array([[0.09579542]]), 1),\n",
       " (array([[0.16013758]]), 1),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.11971527]]), 24),\n",
       " (array([[0.17336991]]), 24),\n",
       " (array([[0.10847563]]), 24),\n",
       " (array([[0.20179246]]), 24),\n",
       " (array([[0.11328753]]), 16),\n",
       " (array([[0.13094654]]), 16),\n",
       " (array([[0.06416533]]), 16),\n",
       " (array([[0.1104385]]), 16),\n",
       " (array([[0.08011548]]), 16),\n",
       " (array([[0.11328753]]), 16),\n",
       " (array([[0.13094654]]), 16),\n",
       " (array([[0.06416533]]), 16),\n",
       " (array([[0.1104385]]), 16),\n",
       " (array([[0.08011548]]), 16),\n",
       " (array([[0.11328753]]), 16),\n",
       " (array([[0.13094654]]), 16),\n",
       " (array([[0.06416533]]), 16),\n",
       " (array([[0.1104385]]), 16),\n",
       " (array([[0.08011548]]), 16),\n",
       " (array([[0.11328753]]), 16),\n",
       " (array([[0.13094654]]), 16),\n",
       " (array([[0.06416533]]), 16),\n",
       " (array([[0.1104385]]), 16),\n",
       " (array([[0.08011548]]), 16),\n",
       " (array([[0.11328753]]), 16),\n",
       " (array([[0.13094654]]), 16),\n",
       " (array([[0.06416533]]), 16),\n",
       " (array([[0.1104385]]), 16),\n",
       " (array([[0.08011548]]), 16),\n",
       " (array([[0.11328753]]), 16),\n",
       " (array([[0.13094654]]), 16),\n",
       " (array([[0.06416533]]), 16),\n",
       " (array([[0.1104385]]), 16),\n",
       " (array([[0.08011548]]), 16),\n",
       " (array([[0.11328753]]), 16),\n",
       " (array([[0.13094654]]), 16),\n",
       " (array([[0.06416533]]), 16),\n",
       " (array([[0.1104385]]), 16),\n",
       " (array([[0.08011548]]), 16),\n",
       " (array([[0.11328753]]), 16),\n",
       " (array([[0.13094654]]), 16),\n",
       " (array([[0.06416533]]), 16),\n",
       " (array([[0.1104385]]), 16),\n",
       " (array([[0.08011548]]), 16),\n",
       " (array([[0.05892059]]), 22),\n",
       " (array([[0.03421838]]), 22),\n",
       " (array([[0.10043654]]), 22),\n",
       " (array([[0.05962047]]), 22),\n",
       " (array([[0.05987812]]), 22),\n",
       " (array([[0.05892059]]), 22),\n",
       " (array([[0.03421838]]), 22),\n",
       " (array([[0.10043654]]), 22),\n",
       " (array([[0.05962047]]), 22),\n",
       " (array([[0.05987812]]), 22),\n",
       " (array([[0.05892059]]), 22),\n",
       " (array([[0.03421838]]), 22),\n",
       " (array([[0.10043654]]), 22),\n",
       " (array([[0.05962047]]), 22),\n",
       " (array([[0.05987812]]), 22),\n",
       " (array([[0.05892059]]), 22),\n",
       " (array([[0.03421838]]), 22),\n",
       " (array([[0.10043654]]), 22),\n",
       " (array([[0.05962047]]), 22),\n",
       " (array([[0.05987812]]), 22),\n",
       " (array([[0.05892059]]), 22),\n",
       " (array([[0.03421838]]), 22),\n",
       " (array([[0.10043654]]), 22),\n",
       " (array([[0.05962047]]), 22),\n",
       " (array([[0.05987812]]), 22),\n",
       " (array([[0.05892059]]), 22),\n",
       " (array([[0.03421838]]), 22),\n",
       " (array([[0.10043654]]), 22),\n",
       " (array([[0.05962047]]), 22),\n",
       " (array([[0.05987812]]), 22),\n",
       " (array([[0.05892059]]), 22),\n",
       " (array([[0.03421838]]), 22),\n",
       " (array([[0.10043654]]), 22),\n",
       " (array([[0.05962047]]), 22),\n",
       " (array([[0.05987812]]), 22),\n",
       " (array([[0.05892059]]), 22),\n",
       " (array([[0.03421838]]), 22),\n",
       " (array([[0.10043654]]), 22),\n",
       " (array([[0.05962047]]), 22),\n",
       " (array([[0.05987812]]), 22),\n",
       " (array([[0.04094977]]), 14),\n",
       " (array([[0.08174847]]), 14),\n",
       " (array([[0.05028285]]), 14),\n",
       " (array([[0.07853587]]), 14),\n",
       " (array([[0.09709415]]), 14),\n",
       " (array([[0.09803334]]), 14),\n",
       " (array([[0.04094977]]), 14),\n",
       " (array([[0.08174847]]), 14),\n",
       " (array([[0.05028285]]), 14),\n",
       " (array([[0.07853587]]), 14),\n",
       " (array([[0.09709415]]), 14),\n",
       " (array([[0.09803334]]), 14),\n",
       " (array([[0.04094977]]), 14),\n",
       " (array([[0.08174847]]), 14),\n",
       " (array([[0.05028285]]), 14),\n",
       " (array([[0.07853587]]), 14),\n",
       " (array([[0.09709415]]), 14),\n",
       " (array([[0.09803334]]), 14),\n",
       " (array([[0.04094977]]), 14),\n",
       " (array([[0.08174847]]), 14),\n",
       " (array([[0.05028285]]), 14),\n",
       " (array([[0.07853587]]), 14),\n",
       " (array([[0.09709415]]), 14),\n",
       " (array([[0.09803334]]), 14),\n",
       " (array([[0.04094977]]), 14),\n",
       " (array([[0.08174847]]), 14),\n",
       " (array([[0.05028285]]), 14),\n",
       " (array([[0.07853587]]), 14),\n",
       " (array([[0.09709415]]), 14),\n",
       " (array([[0.09803334]]), 14),\n",
       " (array([[0.09341116]]), 5),\n",
       " (array([[0.0304334]]), 5),\n",
       " (array([[0.1006042]]), 5),\n",
       " (array([[0.08204503]]), 5),\n",
       " (array([[0.09591418]]), 5),\n",
       " (array([[0.13312563]]), 5),\n",
       " (array([[0.09341116]]), 5),\n",
       " (array([[0.0304334]]), 5),\n",
       " (array([[0.1006042]]), 5),\n",
       " (array([[0.08204503]]), 5),\n",
       " (array([[0.09591418]]), 5),\n",
       " (array([[0.13312563]]), 5),\n",
       " (array([[0.09341116]]), 5),\n",
       " (array([[0.0304334]]), 5),\n",
       " (array([[0.1006042]]), 5),\n",
       " (array([[0.08204503]]), 5),\n",
       " (array([[0.09591418]]), 5),\n",
       " (array([[0.13312563]]), 5),\n",
       " (array([[0.09341116]]), 5),\n",
       " (array([[0.0304334]]), 5),\n",
       " (array([[0.1006042]]), 5),\n",
       " (array([[0.08204503]]), 5),\n",
       " (array([[0.09591418]]), 5),\n",
       " (array([[0.13312563]]), 5),\n",
       " (array([[0.08706315]]), 15),\n",
       " (array([[0.08395754]]), 15),\n",
       " (array([[0.08855681]]), 15),\n",
       " (array([[0.13497456]]), 15),\n",
       " (array([[0.10628888]]), 15),\n",
       " (array([[0.03820588]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.12592502]]), 15),\n",
       " (array([[0.05404185]]), 15),\n",
       " (array([[0.06923819]]), 15),\n",
       " (array([[0.12415626]]), 15),\n",
       " (array([[0.05745342]]), 15),\n",
       " (array([[0.07683715]]), 15),\n",
       " (array([[0.08706315]]), 15),\n",
       " (array([[0.08395754]]), 15),\n",
       " (array([[0.08855681]]), 15),\n",
       " (array([[0.13497456]]), 15),\n",
       " (array([[0.10628888]]), 15),\n",
       " (array([[0.03820588]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.12592502]]), 15),\n",
       " (array([[0.05404185]]), 15),\n",
       " (array([[0.06923819]]), 15),\n",
       " (array([[0.12415626]]), 15),\n",
       " (array([[0.05745342]]), 15),\n",
       " (array([[0.07683715]]), 15),\n",
       " (array([[0.08706315]]), 15),\n",
       " (array([[0.08395754]]), 15),\n",
       " (array([[0.08855681]]), 15),\n",
       " (array([[0.13497456]]), 15),\n",
       " (array([[0.10628888]]), 15),\n",
       " (array([[0.03820588]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.12592502]]), 15),\n",
       " (array([[0.05404185]]), 15),\n",
       " (array([[0.06923819]]), 15),\n",
       " (array([[0.12415626]]), 15),\n",
       " (array([[0.05745342]]), 15),\n",
       " (array([[0.07683715]]), 15),\n",
       " (array([[0.08706315]]), 15),\n",
       " (array([[0.08395754]]), 15),\n",
       " (array([[0.08855681]]), 15),\n",
       " (array([[0.13497456]]), 15),\n",
       " (array([[0.10628888]]), 15),\n",
       " (array([[0.03820588]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.12592502]]), 15),\n",
       " (array([[0.05404185]]), 15),\n",
       " (array([[0.06923819]]), 15),\n",
       " (array([[0.12415626]]), 15),\n",
       " (array([[0.05745342]]), 15),\n",
       " (array([[0.07683715]]), 15),\n",
       " (array([[0.08706315]]), 15),\n",
       " (array([[0.08395754]]), 15),\n",
       " (array([[0.08855681]]), 15),\n",
       " (array([[0.13497456]]), 15),\n",
       " (array([[0.10628888]]), 15),\n",
       " (array([[0.03820588]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.12592502]]), 15),\n",
       " (array([[0.05404185]]), 15),\n",
       " (array([[0.06923819]]), 15),\n",
       " (array([[0.12415626]]), 15),\n",
       " (array([[0.05745342]]), 15),\n",
       " (array([[0.07683715]]), 15),\n",
       " (array([[0.08706315]]), 15),\n",
       " (array([[0.08395754]]), 15),\n",
       " (array([[0.08855681]]), 15),\n",
       " (array([[0.13497456]]), 15),\n",
       " (array([[0.10628888]]), 15),\n",
       " (array([[0.03820588]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.06190625]]), 15),\n",
       " (array([[0.12592502]]), 15),\n",
       " (array([[0.05404185]]), 15),\n",
       " (array([[0.06923819]]), 15),\n",
       " (array([[0.12415626]]), 15),\n",
       " (array([[0.05745342]]), 15),\n",
       " (array([[0.07683715]]), 15),\n",
       " (array([[0.09282015]]), 4),\n",
       " (array([[0.06435399]]), 4),\n",
       " (array([[0.14591474]]), 4),\n",
       " (array([[0.12971506]]), 4),\n",
       " (array([[0.07290712]]), 4),\n",
       " (array([[0.11032109]]), 4),\n",
       " (array([[0.14591474]]), 4),\n",
       " (array([[0.12971506]]), 4),\n",
       " (array([[0.07290712]]), 4),\n",
       " (array([[0.11032109]]), 4),\n",
       " (array([[0.14591474]]), 4),\n",
       " (array([[0.12971506]]), 4),\n",
       " (array([[0.07290712]]), 4),\n",
       " (array([[0.11032109]]), 4),\n",
       " (array([[0.09282015]]), 4),\n",
       " (array([[0.06435399]]), 4),\n",
       " (array([[0.14591474]]), 4),\n",
       " (array([[0.12971506]]), 4),\n",
       " (array([[0.07290712]]), 4),\n",
       " (array([[0.11032109]]), 4),\n",
       " (array([[0.14591474]]), 4),\n",
       " (array([[0.12971506]]), 4),\n",
       " (array([[0.07290712]]), 4),\n",
       " (array([[0.11032109]]), 4),\n",
       " (array([[0.14591474]]), 4),\n",
       " (array([[0.12971506]]), 4),\n",
       " (array([[0.07290712]]), 4),\n",
       " (array([[0.11032109]]), 4),\n",
       " (array([[0.1117662]]), 21),\n",
       " (array([[0.14400502]]), 21),\n",
       " (array([[0.10275142]]), 21),\n",
       " (array([[0.08680955]]), 21),\n",
       " (array([[0.0814612]]), 21),\n",
       " (array([[0.08169455]]), 21),\n",
       " (array([[0.1117662]]), 21),\n",
       " (array([[0.14400502]]), 21),\n",
       " (array([[0.10275142]]), 21),\n",
       " (array([[0.08680955]]), 21),\n",
       " (array([[0.0814612]]), 21),\n",
       " (array([[0.08169455]]), 21),\n",
       " (array([[0.1117662]]), 21),\n",
       " (array([[0.14400502]]), 21),\n",
       " (array([[0.10275142]]), 21),\n",
       " (array([[0.08680955]]), 21),\n",
       " (array([[0.0814612]]), 21),\n",
       " (array([[0.08169455]]), 21),\n",
       " (array([[0.1117662]]), 21),\n",
       " (array([[0.14400502]]), 21),\n",
       " (array([[0.10275142]]), 21),\n",
       " (array([[0.08680955]]), 21),\n",
       " (array([[0.0814612]]), 21),\n",
       " (array([[0.08169455]]), 21),\n",
       " (array([[0.10649787]]), 2),\n",
       " (array([[0.13623161]]), 2),\n",
       " (array([[0.10687717]]), 2),\n",
       " (array([[0.0953845]]), 2),\n",
       " (array([[0.09823044]]), 2),\n",
       " (array([[0.12147235]]), 2),\n",
       " (array([[0.06370182]]), 2),\n",
       " (array([[0.0953845]]), 2),\n",
       " (array([[0.09823044]]), 2),\n",
       " (array([[0.0953845]]), 2),\n",
       " (array([[0.09823044]]), 2),\n",
       " (array([[0.0953845]]), 2),\n",
       " (array([[0.09823044]]), 2),\n",
       " (array([[0.10649787]]), 2),\n",
       " (array([[0.13623161]]), 2),\n",
       " (array([[0.10687717]]), 2),\n",
       " (array([[0.0953845]]), 2),\n",
       " (array([[0.09823044]]), 2),\n",
       " (array([[0.12147235]]), 2),\n",
       " (array([[0.06370182]]), 2),\n",
       " (array([[0.0953845]]), 2),\n",
       " (array([[0.09823044]]), 2),\n",
       " (array([[0.0953845]]), 2),\n",
       " (array([[0.09823044]]), 2),\n",
       " (array([[0.0953845]]), 2),\n",
       " (array([[0.09823044]]), 2),\n",
       " (array([[0.09901738]]), 11),\n",
       " (array([[0.05173708]]), 11),\n",
       " (array([[0.07014417]]), 11),\n",
       " (array([[0.04923083]]), 11),\n",
       " (array([[0.04555698]]), 11),\n",
       " (array([[0.09901738]]), 11),\n",
       " (array([[0.05173708]]), 11),\n",
       " (array([[0.07014417]]), 11),\n",
       " (array([[0.04923083]]), 11),\n",
       " (array([[0.04555698]]), 11),\n",
       " (array([[0.09901738]]), 11),\n",
       " (array([[0.05173708]]), 11),\n",
       " (array([[0.07014417]]), 11),\n",
       " (array([[0.04923083]]), 11),\n",
       " (array([[0.04555698]]), 11),\n",
       " (array([[0.09901738]]), 11),\n",
       " (array([[0.05173708]]), 11),\n",
       " (array([[0.07014417]]), 11),\n",
       " (array([[0.04923083]]), 11),\n",
       " (array([[0.04555698]]), 11),\n",
       " (array([[0.09901738]]), 11),\n",
       " (array([[0.05173708]]), 11),\n",
       " (array([[0.07014417]]), 11),\n",
       " (array([[0.04923083]]), 11),\n",
       " (array([[0.04555698]]), 11),\n",
       " (array([[0.09901738]]), 11),\n",
       " (array([[0.05173708]]), 11),\n",
       " (array([[0.07014417]]), 11),\n",
       " (array([[0.04923083]]), 11),\n",
       " (array([[0.04555698]]), 11),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.13541684]]), 18),\n",
       " (array([[0.10696337]]), 18),\n",
       " (array([[0.10434099]]), 18),\n",
       " (array([[0.14189154]]), 18),\n",
       " (array([[0.09162866]]), 20),\n",
       " (array([[0.06813987]]), 20),\n",
       " (array([[0.07718182]]), 20),\n",
       " (array([[0.06766804]]), 20),\n",
       " (array([[0.14578861]]), 20),\n",
       " (array([[0.07380749]]), 20),\n",
       " (array([[0.09162866]]), 20),\n",
       " (array([[0.06813987]]), 20),\n",
       " (array([[0.07718182]]), 20),\n",
       " (array([[0.06766804]]), 20),\n",
       " (array([[0.14578861]]), 20),\n",
       " (array([[0.07380749]]), 20),\n",
       " (array([[0.09162866]]), 20),\n",
       " (array([[0.06813987]]), 20),\n",
       " (array([[0.07718182]]), 20),\n",
       " (array([[0.06766804]]), 20),\n",
       " (array([[0.14578861]]), 20),\n",
       " (array([[0.07380749]]), 20),\n",
       " (array([[0.09162866]]), 20),\n",
       " (array([[0.06813987]]), 20),\n",
       " (array([[0.07718182]]), 20),\n",
       " (array([[0.06766804]]), 20),\n",
       " (array([[0.14578861]]), 20),\n",
       " (array([[0.07380749]]), 20),\n",
       " (array([[0.09162866]]), 20),\n",
       " (array([[0.06813987]]), 20),\n",
       " (array([[0.07718182]]), 20),\n",
       " (array([[0.06766804]]), 20),\n",
       " (array([[0.14578861]]), 20),\n",
       " (array([[0.07380749]]), 20),\n",
       " (array([[0.09162866]]), 20),\n",
       " (array([[0.06813987]]), 20),\n",
       " (array([[0.07718182]]), 20),\n",
       " (array([[0.06766804]]), 20),\n",
       " (array([[0.14578861]]), 20),\n",
       " (array([[0.07380749]]), 20),\n",
       " (array([[0.09162866]]), 20),\n",
       " (array([[0.06813987]]), 20),\n",
       " (array([[0.07718182]]), 20),\n",
       " (array([[0.06766804]]), 20),\n",
       " (array([[0.14578861]]), 20),\n",
       " (array([[0.07380749]]), 20),\n",
       " (array([[0.09162866]]), 20),\n",
       " (array([[0.06813987]]), 20),\n",
       " (array([[0.07718182]]), 20),\n",
       " (array([[0.06766804]]), 20),\n",
       " (array([[0.14578861]]), 20),\n",
       " (array([[0.07380749]]), 20),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.13828398]]), 8),\n",
       " (array([[0.14019608]]), 8),\n",
       " (array([[0.09783996]]), 8),\n",
       " (array([[0.07868944]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.09726865]]), 8),\n",
       " (array([[0.08311021]]), 8),\n",
       " (array([[0.08627657]]), 8),\n",
       " (array([[0.08197775]]), 17),\n",
       " (array([[0.07060154]]), 17),\n",
       " (array([[0.11179472]]), 17),\n",
       " (array([[0.06838308]]), 17),\n",
       " (array([[0.06902366]]), 17),\n",
       " (array([[0.08197775]]), 17),\n",
       " (array([[0.07060154]]), 17),\n",
       " (array([[0.11179472]]), 17),\n",
       " (array([[0.06838308]]), 17),\n",
       " (array([[0.06902366]]), 17),\n",
       " (array([[0.08197775]]), 17),\n",
       " (array([[0.07060154]]), 17),\n",
       " (array([[0.11179472]]), 17),\n",
       " (array([[0.06838308]]), 17),\n",
       " (array([[0.06902366]]), 17),\n",
       " (array([[0.08197775]]), 17),\n",
       " (array([[0.07060154]]), 17),\n",
       " (array([[0.11179472]]), 17),\n",
       " (array([[0.06838308]]), 17),\n",
       " (array([[0.06902366]]), 17),\n",
       " (array([[0.08197775]]), 17),\n",
       " (array([[0.07060154]]), 17),\n",
       " (array([[0.11179472]]), 17),\n",
       " (array([[0.06838308]]), 17),\n",
       " (array([[0.06902366]]), 17),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.09851084]]), 19),\n",
       " (array([[0.07725119]]), 19),\n",
       " (array([[0.08366787]]), 19),\n",
       " (array([[0.08370553]]), 7),\n",
       " (array([[0.08306956]]), 7),\n",
       " (array([[0.07892072]]), 7),\n",
       " (array([[0.06613042]]), 7),\n",
       " (array([[0.08537017]]), 7),\n",
       " (array([[0.08751729]]), 7),\n",
       " (array([[0.09817426]]), 7),\n",
       " (array([[0.06750055]]), 7),\n",
       " (array([[0.08205216]]), 7),\n",
       " (array([[0.09758207]]), 7),\n",
       " (array([[0.09312917]]), 7),\n",
       " (array([[0.08370553]]), 7),\n",
       " (array([[0.08306956]]), 7),\n",
       " (array([[0.07892072]]), 7),\n",
       " (array([[0.06613042]]), 7),\n",
       " (array([[0.08537017]]), 7),\n",
       " (array([[0.08751729]]), 7),\n",
       " (array([[0.09817426]]), 7),\n",
       " (array([[0.06750055]]), 7),\n",
       " (array([[0.08205216]]), 7),\n",
       " (array([[0.09758207]]), 7),\n",
       " (array([[0.09312917]]), 7),\n",
       " (array([[0.08370553]]), 7),\n",
       " (array([[0.08306956]]), 7),\n",
       " (array([[0.07892072]]), 7),\n",
       " (array([[0.06613042]]), 7),\n",
       " (array([[0.08537017]]), 7),\n",
       " (array([[0.08751729]]), 7),\n",
       " (array([[0.09817426]]), 7),\n",
       " (array([[0.06750055]]), 7),\n",
       " (array([[0.08205216]]), 7),\n",
       " (array([[0.09758207]]), 7),\n",
       " (array([[0.09312917]]), 7),\n",
       " (array([[0.10858929]]), 13),\n",
       " (array([[0.079649]]), 13),\n",
       " (array([[0.10815626]]), 13),\n",
       " (array([[0.09867179]]), 13),\n",
       " (array([[0.11590315]]), 13),\n",
       " (array([[0.07598683]]), 13),\n",
       " (array([[0.11869846]]), 13),\n",
       " (array([[0.10858929]]), 13),\n",
       " (array([[0.079649]]), 13),\n",
       " (array([[0.10815626]]), 13),\n",
       " (array([[0.09867179]]), 13),\n",
       " (array([[0.11590315]]), 13),\n",
       " (array([[0.07598683]]), 13),\n",
       " (array([[0.11869846]]), 13),\n",
       " (array([[0.10858929]]), 13),\n",
       " (array([[0.079649]]), 13),\n",
       " (array([[0.10815626]]), 13),\n",
       " (array([[0.09867179]]), 13),\n",
       " (array([[0.11590315]]), 13),\n",
       " (array([[0.07598683]]), 13),\n",
       " (array([[0.11869846]]), 13),\n",
       " (array([[0.10858929]]), 13),\n",
       " (array([[0.079649]]), 13),\n",
       " (array([[0.10815626]]), 13),\n",
       " (array([[0.09867179]]), 13),\n",
       " (array([[0.11590315]]), 13),\n",
       " (array([[0.07598683]]), 13),\n",
       " (array([[0.11869846]]), 13),\n",
       " (array([[0.10858929]]), 13),\n",
       " (array([[0.079649]]), 13),\n",
       " (array([[0.10815626]]), 13),\n",
       " (array([[0.09867179]]), 13),\n",
       " (array([[0.11590315]]), 13),\n",
       " (array([[0.07598683]]), 13),\n",
       " (array([[0.11869846]]), 13),\n",
       " (array([[0.10858929]]), 13),\n",
       " (array([[0.079649]]), 13),\n",
       " (array([[0.10815626]]), 13),\n",
       " (array([[0.09867179]]), 13),\n",
       " (array([[0.11590315]]), 13),\n",
       " (array([[0.07598683]]), 13),\n",
       " (array([[0.11869846]]), 13),\n",
       " (array([[0.08829544]]), 10),\n",
       " (array([[0.11200373]]), 10),\n",
       " (array([[0.08153422]]), 10),\n",
       " (array([[0.14578329]]), 10),\n",
       " (array([[0.05544461]]), 10),\n",
       " (array([[0.08829544]]), 10),\n",
       " (array([[0.11200373]]), 10),\n",
       " (array([[0.08153422]]), 10),\n",
       " (array([[0.14578329]]), 10),\n",
       " (array([[0.05544461]]), 10),\n",
       " (array([[0.08829544]]), 10),\n",
       " (array([[0.11200373]]), 10),\n",
       " (array([[0.08153422]]), 10),\n",
       " (array([[0.14578329]]), 10),\n",
       " (array([[0.05544461]]), 10),\n",
       " (array([[0.08829544]]), 10),\n",
       " (array([[0.11200373]]), 10),\n",
       " (array([[0.08153422]]), 10),\n",
       " (array([[0.14578329]]), 10),\n",
       " (array([[0.05544461]]), 10),\n",
       " (array([[0.08829544]]), 10),\n",
       " (array([[0.11200373]]), 10),\n",
       " (array([[0.08153422]]), 10),\n",
       " (array([[0.14578329]]), 10),\n",
       " (array([[0.05544461]]), 10),\n",
       " (array([[0.08829544]]), 10),\n",
       " (array([[0.11200373]]), 10),\n",
       " (array([[0.08153422]]), 10),\n",
       " (array([[0.14578329]]), 10),\n",
       " (array([[0.05544461]]), 10),\n",
       " (array([[0.08829544]]), 10),\n",
       " (array([[0.11200373]]), 10),\n",
       " (array([[0.08153422]]), 10),\n",
       " (array([[0.14578329]]), 10),\n",
       " (array([[0.05544461]]), 10),\n",
       " (array([[0.08829544]]), 10),\n",
       " (array([[0.11200373]]), 10),\n",
       " (array([[0.08153422]]), 10),\n",
       " (array([[0.14578329]]), 10),\n",
       " (array([[0.05544461]]), 10),\n",
       " (array([[0.12768777]]), 9),\n",
       " (array([[0.09196075]]), 9),\n",
       " (array([[0.12268173]]), 9),\n",
       " (array([[0.08341281]]), 9),\n",
       " (array([[0.1839941]]), 9),\n",
       " (array([[0.13101828]]), 9),\n",
       " (array([[0.13960268]]), 9),\n",
       " (array([[0.12768777]]), 9),\n",
       " (array([[0.09196075]]), 9),\n",
       " (array([[0.12268173]]), 9),\n",
       " (array([[0.08341281]]), 9),\n",
       " (array([[0.1839941]]), 9),\n",
       " (array([[0.13101828]]), 9),\n",
       " (array([[0.13960268]]), 9),\n",
       " (array([[0.12768777]]), 9),\n",
       " (array([[0.09196075]]), 9),\n",
       " (array([[0.12268173]]), 9),\n",
       " (array([[0.08341281]]), 9),\n",
       " (array([[0.1839941]]), 9),\n",
       " (array([[0.13101828]]), 9),\n",
       " (array([[0.13960268]]), 9),\n",
       " (array([[0.12768777]]), 9),\n",
       " (array([[0.09196075]]), 9),\n",
       " (array([[0.12268173]]), 9),\n",
       " (array([[0.08341281]]), 9),\n",
       " (array([[0.1839941]]), 9),\n",
       " (array([[0.13101828]]), 9),\n",
       " (array([[0.13960268]]), 9),\n",
       " (array([[0.1010258]]), 3),\n",
       " (array([[0.10058829]]), 3),\n",
       " (array([[0.13974732]]), 3),\n",
       " (array([[0.08315959]]), 3),\n",
       " (array([[0.15551689]]), 3),\n",
       " (array([[0.1010258]]), 3),\n",
       " (array([[0.10058829]]), 3),\n",
       " (array([[0.13974732]]), 3),\n",
       " (array([[0.08315959]]), 3),\n",
       " (array([[0.15551689]]), 3),\n",
       " (array([[0.1010258]]), 3),\n",
       " (array([[0.10058829]]), 3),\n",
       " (array([[0.13974732]]), 3),\n",
       " (array([[0.08315959]]), 3),\n",
       " (array([[0.15551689]]), 3),\n",
       " (array([[0.1010258]]), 3),\n",
       " (array([[0.10058829]]), 3),\n",
       " (array([[0.13974732]]), 3),\n",
       " (array([[0.08315959]]), 3),\n",
       " (array([[0.15551689]]), 3),\n",
       " (array([[0.1010258]]), 3),\n",
       " (array([[0.10058829]]), 3),\n",
       " (array([[0.13974732]]), 3),\n",
       " (array([[0.08315959]]), 3),\n",
       " (array([[0.15551689]]), 3),\n",
       " (array([[0.1010258]]), 3),\n",
       " (array([[0.10058829]]), 3),\n",
       " (array([[0.13974732]]), 3),\n",
       " (array([[0.08315959]]), 3),\n",
       " (array([[0.15551689]]), 3),\n",
       " (array([[0.1010258]]), 3),\n",
       " (array([[0.10058829]]), 3),\n",
       " (array([[0.13974732]]), 3),\n",
       " (array([[0.08315959]]), 3),\n",
       " (array([[0.15551689]]), 3),\n",
       " (array([[0.1010258]]), 3),\n",
       " (array([[0.10058829]]), 3),\n",
       " (array([[0.13974732]]), 3),\n",
       " (array([[0.08315959]]), 3),\n",
       " (array([[0.15551689]]), 3),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23),\n",
       " (array([[0.09368953]]), 23),\n",
       " (array([[0.06344664]]), 23),\n",
       " (array([[0.06264088]]), 23),\n",
       " (array([[0.06831466]]), 23),\n",
       " (array([[0.07746057]]), 23),\n",
       " (array([[0.06754278]]), 23),\n",
       " (array([[0.10813784]]), 23)]"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "similarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "i=vi.predict(idf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Data Science'], dtype=object)"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "le.inverse_transform(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "output = open('label_encoder.pkl', 'wb')\n",
    "pickle.dump(le, output)\n",
    "output.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "vect = open('tfidf.pkl', 'wb')\n",
    "pickle.dump(tf, vect)\n",
    "output.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = open('models.pkl', 'wb')\n",
    "pickle.dump(vi, model)\n",
    "output.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder = pickle.load(open('label_encoder.pkl', 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "vect = pickle.load(open('tfidf.pkl', 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = pickle.load(open('models.pkl', 'rb'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Data Science']\n"
     ]
    }
   ],
   "source": [
    "print(encoder.inverse_transform(i))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [],
   "source": [
    "trasnf=vect.transform(cleaned_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [],
   "source": [
    "b=model.predict(trasnf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Data Science']\n"
     ]
    }
   ],
   "source": [
    "print(encoder.inverse_transform(b))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}