model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "from prepare import prep_create_labels\n",
    "from preprocess import prep_job_data, split_job_data, add_columns\n",
    "\n",
    "from pprint import pprint\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import confusion_matrix, classification_report"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Wrangle and Split Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>job_title</th>\n",
       "      <th>company</th>\n",
       "      <th>location</th>\n",
       "      <th>is_remote</th>\n",
       "      <th>salary</th>\n",
       "      <th>post_date</th>\n",
       "      <th>date_accessed</th>\n",
       "      <th>job_description</th>\n",
       "      <th>label</th>\n",
       "      <th>clean</th>\n",
       "      <th>stemmed</th>\n",
       "      <th>lemmatized</th>\n",
       "      <th>words</th>\n",
       "      <th>doc_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>651</th>\n",
       "      <td>Sr. Machine Learning Engineer - Discovery</td>\n",
       "      <td>Twitter</td>\n",
       "      <td>Seattle, WA</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>6 days ago</td>\n",
       "      <td>2021-03-05</td>\n",
       "      <td>Company Description Twitter is what’s happenin...</td>\n",
       "      <td>MLE</td>\n",
       "      <td>company twitter whats happening people talking...</td>\n",
       "      <td>compani descript twitter is what happen and wh...</td>\n",
       "      <td>company description twitter is whats happening...</td>\n",
       "      <td>[company, twitter, whats, happening, people, t...</td>\n",
       "      <td>366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>Data Scientist</td>\n",
       "      <td>EagleView</td>\n",
       "      <td>Rochester, NY 14607</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>2 days ago</td>\n",
       "      <td>2021-03-05</td>\n",
       "      <td>We are looking for a talented Data Scientist t...</td>\n",
       "      <td>DS</td>\n",
       "      <td>looking talented data scientist join machine l...</td>\n",
       "      <td>we are look for a talent data scientist to joi...</td>\n",
       "      <td>we are looking for a talented data scientist t...</td>\n",
       "      <td>[looking, talented, data, scientist, join, mac...</td>\n",
       "      <td>144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>Data Scientist/Engineer</td>\n",
       "      <td>Oracle</td>\n",
       "      <td>United States</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>10 days ago</td>\n",
       "      <td>2021-03-05</td>\n",
       "      <td>Data Scientist/Engineer -210009CG  Applicants ...</td>\n",
       "      <td>DS</td>\n",
       "      <td>data scientistengineer 210009cg applicant requ...</td>\n",
       "      <td>data scientistengin 210009cg applic are requir...</td>\n",
       "      <td>data scientistengineer 210009cg applicant are ...</td>\n",
       "      <td>[data, scientistengineer, 210009cg, applicant,...</td>\n",
       "      <td>557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84</th>\n",
       "      <td>Data &amp; Applied Scientist</td>\n",
       "      <td>Microsoft</td>\n",
       "      <td>Bellevue, WA</td>\n",
       "      <td>0</td>\n",
       "      <td></td>\n",
       "      <td>5 days ago</td>\n",
       "      <td>2021-03-05</td>\n",
       "      <td>What if your job description were: make tomorr...</td>\n",
       "      <td>DS</td>\n",
       "      <td>make tomorrow better essence role within micro...</td>\n",
       "      <td>what if your job descript were make tomorrow b...</td>\n",
       "      <td>what if your job description were make tomorro...</td>\n",
       "      <td>[make, tomorrow, better, essence, role, within...</td>\n",
       "      <td>353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>296</th>\n",
       "      <td>Data Analyst (Sales Operations)</td>\n",
       "      <td>Businessolver</td>\n",
       "      <td>Remote</td>\n",
       "      <td>0</td>\n",
       "      <td>$80,000 - $90,000 a year</td>\n",
       "      <td>7 days ago</td>\n",
       "      <td>2021-03-05</td>\n",
       "      <td>Businessolver is a technology company deliveri...</td>\n",
       "      <td>DA</td>\n",
       "      <td>businessolver technology company delivering ma...</td>\n",
       "      <td>businessolv is a technolog compani deliv marke...</td>\n",
       "      <td>businessolver is a technology company deliveri...</td>\n",
       "      <td>[businessolver, technology, company, deliverin...</td>\n",
       "      <td>476</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     job_title        company  \\\n",
       "651  Sr. Machine Learning Engineer - Discovery        Twitter   \n",
       "79                              Data Scientist      EagleView   \n",
       "92                     Data Scientist/Engineer         Oracle   \n",
       "84                    Data & Applied Scientist      Microsoft   \n",
       "296            Data Analyst (Sales Operations)  Businessolver   \n",
       "\n",
       "                location  is_remote                    salary    post_date  \\\n",
       "651          Seattle, WA          0                             6 days ago   \n",
       "79   Rochester, NY 14607          0                             2 days ago   \n",
       "92         United States          0                            10 days ago   \n",
       "84          Bellevue, WA          0                             5 days ago   \n",
       "296               Remote          0  $80,000 - $90,000 a year   7 days ago   \n",
       "\n",
       "    date_accessed                                    job_description label  \\\n",
       "651    2021-03-05  Company Description Twitter is what’s happenin...   MLE   \n",
       "79     2021-03-05  We are looking for a talented Data Scientist t...    DS   \n",
       "92     2021-03-05  Data Scientist/Engineer -210009CG  Applicants ...    DS   \n",
       "84     2021-03-05  What if your job description were: make tomorr...    DS   \n",
       "296    2021-03-05  Businessolver is a technology company deliveri...    DA   \n",
       "\n",
       "                                                 clean  \\\n",
       "651  company twitter whats happening people talking...   \n",
       "79   looking talented data scientist join machine l...   \n",
       "92   data scientistengineer 210009cg applicant requ...   \n",
       "84   make tomorrow better essence role within micro...   \n",
       "296  businessolver technology company delivering ma...   \n",
       "\n",
       "                                               stemmed  \\\n",
       "651  compani descript twitter is what happen and wh...   \n",
       "79   we are look for a talent data scientist to joi...   \n",
       "92   data scientistengin 210009cg applic are requir...   \n",
       "84   what if your job descript were make tomorrow b...   \n",
       "296  businessolv is a technolog compani deliv marke...   \n",
       "\n",
       "                                            lemmatized  \\\n",
       "651  company description twitter is whats happening...   \n",
       "79   we are looking for a talented data scientist t...   \n",
       "92   data scientistengineer 210009cg applicant are ...   \n",
       "84   what if your job description were make tomorro...   \n",
       "296  businessolver is a technology company deliveri...   \n",
       "\n",
       "                                                 words  doc_length  \n",
       "651  [company, twitter, whats, happening, people, t...         366  \n",
       "79   [looking, talented, data, scientist, join, mac...         144  \n",
       "92   [data, scientistengineer, 210009cg, applicant,...         557  \n",
       "84   [make, tomorrow, better, essence, role, within...         353  \n",
       "296  [businessolver, technology, company, deliverin...         476  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_json('indeed-data-jobs-FINAL.json')\n",
    "df = prep_create_labels(df).reset_index(drop=True)\n",
    "df = prep_job_data(df, 'job_description', extra_words=['job', 'description']).reset_index(drop=True)\n",
    "df = add_columns(df)\n",
    "df.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train: 0.6\n",
      "validate: 0.2\n",
      "test: 0.2\n"
     ]
    }
   ],
   "source": [
    "train, validate, test = split_job_data(df)\n",
    "print(f'train: {round(train.shape[0]/len(df),2)}')\n",
    "print(f'validate: {round(validate.shape[0]/len(df),2)}')\n",
    "print(f'test: {round(test.shape[0]/len(df),2)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>label</th>\n",
       "      <th>job_title</th>\n",
       "      <th>company</th>\n",
       "      <th>location</th>\n",
       "      <th>is_remote</th>\n",
       "      <th>clean</th>\n",
       "      <th>words</th>\n",
       "      <th>doc_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>472</th>\n",
       "      <td>DE</td>\n",
       "      <td>Data Integration Engineer / Architect</td>\n",
       "      <td>Rayn Solutions LLC</td>\n",
       "      <td>Remote</td>\n",
       "      <td>1</td>\n",
       "      <td>looking talented experienced consultant long t...</td>\n",
       "      <td>[looking, talented, experienced, consultant, l...</td>\n",
       "      <td>295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>DS</td>\n",
       "      <td>Data Scientist (all levels)</td>\n",
       "      <td>Cascade Data Labs</td>\n",
       "      <td>Portland, OR 97202</td>\n",
       "      <td>1</td>\n",
       "      <td>cascade data lab boutique consulting agency en...</td>\n",
       "      <td>[cascade, data, lab, boutique, consulting, age...</td>\n",
       "      <td>236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>464</th>\n",
       "      <td>DE</td>\n",
       "      <td>Data Analytics Engineer</td>\n",
       "      <td>Ford Motor Company</td>\n",
       "      <td>Dearborn, MI</td>\n",
       "      <td>0</td>\n",
       "      <td>dearborn ford motor company product developmen...</td>\n",
       "      <td>[dearborn, ford, motor, company, product, deve...</td>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>DS</td>\n",
       "      <td>Senior Data Scientist</td>\n",
       "      <td>Verusen</td>\n",
       "      <td>Atlanta, GA 30308</td>\n",
       "      <td>0</td>\n",
       "      <td>company verusen leading technology company us ...</td>\n",
       "      <td>[company, verusen, leading, technology, compan...</td>\n",
       "      <td>367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>693</th>\n",
       "      <td>MLE</td>\n",
       "      <td>Machine Learning Intern</td>\n",
       "      <td>Knock</td>\n",
       "      <td>United States</td>\n",
       "      <td>1</td>\n",
       "      <td>knock mission empower people move freely knock...</td>\n",
       "      <td>[knock, mission, empower, people, move, freely...</td>\n",
       "      <td>311</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    label                              job_title             company  \\\n",
       "472    DE  Data Integration Engineer / Architect  Rayn Solutions LLC   \n",
       "75     DS            Data Scientist (all levels)   Cascade Data Labs   \n",
       "464    DE                Data Analytics Engineer  Ford Motor Company   \n",
       "131    DS                  Senior Data Scientist             Verusen   \n",
       "693   MLE                Machine Learning Intern               Knock   \n",
       "\n",
       "               location  is_remote  \\\n",
       "472              Remote          1   \n",
       "75   Portland, OR 97202          1   \n",
       "464        Dearborn, MI          0   \n",
       "131   Atlanta, GA 30308          0   \n",
       "693       United States          1   \n",
       "\n",
       "                                                 clean  \\\n",
       "472  looking talented experienced consultant long t...   \n",
       "75   cascade data lab boutique consulting agency en...   \n",
       "464  dearborn ford motor company product developmen...   \n",
       "131  company verusen leading technology company us ...   \n",
       "693  knock mission empower people move freely knock...   \n",
       "\n",
       "                                                 words  doc_length  \n",
       "472  [looking, talented, experienced, consultant, l...         295  \n",
       "75   [cascade, data, lab, boutique, consulting, age...         236  \n",
       "464  [dearborn, ford, motor, company, product, deve...         334  \n",
       "131  [company, verusen, leading, technology, compan...         367  \n",
       "693  [knock, mission, empower, people, move, freely...         311  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.sample(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Will run Bag of words and TFIDF models and try additional models if time permits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# bag of words model\n",
    "vectorizer = CountVectorizer(stop_words='english', \n",
    "                             min_df=30, \n",
    "                             ngram_range=(1,2), \n",
    "                             binary=True)\n",
    "\n",
    "# Learn vocabulary in sentences. \n",
    "vectorizer.fit(train.clean)\n",
    "\n",
    "# Get dictionary. \n",
    "vectorizer.get_feature_names()\n",
    "\n",
    "# Transform each sentences in vector space.\n",
    "bow = vectorizer.transform(train.clean)\n",
    "\n",
    "# this is just to see the array of 0's and 1's\n",
    "bow_array = bow.toarray()\n",
    "\n",
    "X_bow = bow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TF-IDF\n",
    "tfidf = TfidfVectorizer(stop_words='english', \n",
    "                        min_df=30, \n",
    "                        ngram_range=(1,2), \n",
    "                        binary=True)\n",
    "\n",
    "tfidf_sparse_matrix = tfidf.fit_transform(train.clean)\n",
    "\n",
    "# Get vocabularies.\n",
    "tfidf.vocabulary_\n",
    "\n",
    "# Transform to document-term matrix\n",
    "vector_spaces = tfidf.transform(train.clean)\n",
    "vector_spaces.toarray()\n",
    "\n",
    "X_tfidf = tfidf_sparse_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create our y (label) dataset\n",
    "y = train['label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# BOW Log Reg Model\n",
    "lm = LogisticRegression().fit(X_bow, y)\n",
    "train['predict_bow'] = lm.predict(X_bow)\n",
    "\n",
    "# TF-IDF Log Reg Model\n",
    "lm_tfidf = LogisticRegression().fit(X_tfidf, y)\n",
    "train['pred_tfidf'] = lm_tfidf.predict(X_tfidf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bag of Words Model \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "          DA       0.99      1.00      1.00       112\n",
      "          DE       0.99      1.00      1.00       103\n",
      "          DS       1.00      0.99      1.00       103\n",
      "         MLE       1.00      0.99      1.00       108\n",
      "\n",
      "    accuracy                           1.00       426\n",
      "   macro avg       1.00      1.00      1.00       426\n",
      "weighted avg       1.00      1.00      1.00       426\n",
      "\n"
     ]
    },
    {
     "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>predict_bow</th>\n",
       "      <th>DA</th>\n",
       "      <th>DE</th>\n",
       "      <th>DS</th>\n",
       "      <th>MLE</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>label</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DA</th>\n",
       "      <td>112</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DE</th>\n",
       "      <td>0</td>\n",
       "      <td>103</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DS</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>102</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MLE</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>107</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "predict_bow   DA   DE   DS  MLE\n",
       "label                          \n",
       "DA           112    0    0    0\n",
       "DE             0  103    0    0\n",
       "DS             1    0  102    0\n",
       "MLE            0    1    0  107"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('Bag of Words Model \\n', classification_report(train.label, train.predict_bow))\n",
    "pd.crosstab(train.label, train.predict_bow)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TF-IDF Model \n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "          DA       0.98      0.97      0.98       112\n",
      "          DE       0.98      0.98      0.98       103\n",
      "          DS       0.96      0.97      0.97       103\n",
      "         MLE       0.96      0.96      0.96       108\n",
      "\n",
      "    accuracy                           0.97       426\n",
      "   macro avg       0.97      0.97      0.97       426\n",
      "weighted avg       0.97      0.97      0.97       426\n",
      "\n"
     ]
    },
    {
     "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>pred_tfidf</th>\n",
       "      <th>DA</th>\n",
       "      <th>DE</th>\n",
       "      <th>DS</th>\n",
       "      <th>MLE</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>label</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DA</th>\n",
       "      <td>109</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DE</th>\n",
       "      <td>1</td>\n",
       "      <td>101</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DS</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>100</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MLE</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>104</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "pred_tfidf   DA   DE   DS  MLE\n",
       "label                         \n",
       "DA          109    0    3    0\n",
       "DE            1  101    0    1\n",
       "DS            0    0  100    3\n",
       "MLE           1    2    1  104"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('TF-IDF Model \\n', classification_report(train.label, train.pred_tfidf))\n",
    "pd.crosstab(train.label, train.pred_tfidf)"
   ]
  },
  {
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}