asd-prediction-with-logistic-regression (1).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "81534723ea4f827e0a57e577af85b2138a89ef62"
   },
   "source": [
    "### Classifying autistic patients upon the screening results with Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "_uuid": "6b0494e07bcb84a5373d1b9065c7fa61ebce5975"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score, roc_auc_score, classification_report, roc_curve\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.model_selection import train_test_split, cross_val_score, StratifiedKFold, cross_val_predict\n",
    "from sklearn.preprocessing import StandardScaler, LabelEncoder, LabelBinarizer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "d7647b18545d041ec1004d7a3767e5aac61b0a67"
   },
   "source": [
    "### EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "_uuid": "8c707cd0536ef8d6ff4f9ee610f36e48333072c0"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(\n",
    "    'Autism_Data.arff', \n",
    "    na_values='?',\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "_uuid": "0d60aec3d0205ef110daeca67a0694cf27301d95"
   },
   "outputs": [],
   "source": [
    "data = df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "_uuid": "28a7884576ac90d5dedb841d3256f68367a0504a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(704, 21)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "_uuid": "aee81ac7a7045719929bea0c3ea03e9bc1208731"
   },
   "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>A1_Score</th>\n",
       "      <th>A2_Score</th>\n",
       "      <th>A3_Score</th>\n",
       "      <th>A4_Score</th>\n",
       "      <th>A5_Score</th>\n",
       "      <th>A6_Score</th>\n",
       "      <th>A7_Score</th>\n",
       "      <th>A8_Score</th>\n",
       "      <th>A9_Score</th>\n",
       "      <th>A10_Score</th>\n",
       "      <th>...</th>\n",
       "      <th>gender</th>\n",
       "      <th>ethnicity</th>\n",
       "      <th>jundice</th>\n",
       "      <th>austim</th>\n",
       "      <th>contry_of_res</th>\n",
       "      <th>used_app_before</th>\n",
       "      <th>result</th>\n",
       "      <th>age_desc</th>\n",
       "      <th>relation</th>\n",
       "      <th>Class/ASD</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>White-European</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>'United States'</td>\n",
       "      <td>no</td>\n",
       "      <td>6</td>\n",
       "      <td>'18 and more'</td>\n",
       "      <td>Self</td>\n",
       "      <td>NO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>m</td>\n",
       "      <td>Latino</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>Brazil</td>\n",
       "      <td>no</td>\n",
       "      <td>5</td>\n",
       "      <td>'18 and more'</td>\n",
       "      <td>Self</td>\n",
       "      <td>NO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>m</td>\n",
       "      <td>Latino</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>Spain</td>\n",
       "      <td>no</td>\n",
       "      <td>8</td>\n",
       "      <td>'18 and more'</td>\n",
       "      <td>Parent</td>\n",
       "      <td>YES</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>White-European</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>'United States'</td>\n",
       "      <td>no</td>\n",
       "      <td>6</td>\n",
       "      <td>'18 and more'</td>\n",
       "      <td>Self</td>\n",
       "      <td>NO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>NaN</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>Egypt</td>\n",
       "      <td>no</td>\n",
       "      <td>2</td>\n",
       "      <td>'18 and more'</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NO</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   A1_Score  A2_Score  A3_Score  A4_Score  A5_Score  A6_Score  A7_Score  \\\n",
       "0         1         1         1         1         0         0         1   \n",
       "1         1         1         0         1         0         0         0   \n",
       "2         1         1         0         1         1         0         1   \n",
       "3         1         1         0         1         0         0         1   \n",
       "4         1         0         0         0         0         0         0   \n",
       "\n",
       "   A8_Score  A9_Score  A10_Score  ...  gender       ethnicity jundice austim  \\\n",
       "0         1         0          0  ...       f  White-European      no     no   \n",
       "1         1         0          1  ...       m          Latino      no    yes   \n",
       "2         1         1          1  ...       m          Latino     yes    yes   \n",
       "3         1         0          1  ...       f  White-European      no    yes   \n",
       "4         1         0          0  ...       f             NaN      no     no   \n",
       "\n",
       "     contry_of_res used_app_before result       age_desc relation Class/ASD  \n",
       "0  'United States'              no      6  '18 and more'     Self        NO  \n",
       "1           Brazil              no      5  '18 and more'     Self        NO  \n",
       "2            Spain              no      8  '18 and more'   Parent       YES  \n",
       "3  'United States'              no      6  '18 and more'     Self        NO  \n",
       "4            Egypt              no      2  '18 and more'      NaN        NO  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "466df875fdbb305462048cd5cbbe6854210dfcff"
   },
   "source": [
    "Dataset contains ten behavioural features (AQ-10-Adult) plus ten individuals characteristics that have proved to be effective in detecting the ASD cases from controls in behaviour science."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "_uuid": "e331ebd04d280d2b3b2ca27cb5db9ba33afb7a6e"
   },
   "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>A1_Score</th>\n",
       "      <th>A2_Score</th>\n",
       "      <th>A3_Score</th>\n",
       "      <th>A4_Score</th>\n",
       "      <th>A5_Score</th>\n",
       "      <th>A6_Score</th>\n",
       "      <th>A7_Score</th>\n",
       "      <th>A8_Score</th>\n",
       "      <th>A9_Score</th>\n",
       "      <th>A10_Score</th>\n",
       "      <th>age</th>\n",
       "      <th>result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>704.000000</td>\n",
       "      <td>702.000000</td>\n",
       "      <td>704.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.721591</td>\n",
       "      <td>0.453125</td>\n",
       "      <td>0.457386</td>\n",
       "      <td>0.495739</td>\n",
       "      <td>0.498580</td>\n",
       "      <td>0.284091</td>\n",
       "      <td>0.417614</td>\n",
       "      <td>0.649148</td>\n",
       "      <td>0.323864</td>\n",
       "      <td>0.573864</td>\n",
       "      <td>29.698006</td>\n",
       "      <td>4.875000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.448535</td>\n",
       "      <td>0.498152</td>\n",
       "      <td>0.498535</td>\n",
       "      <td>0.500337</td>\n",
       "      <td>0.500353</td>\n",
       "      <td>0.451301</td>\n",
       "      <td>0.493516</td>\n",
       "      <td>0.477576</td>\n",
       "      <td>0.468281</td>\n",
       "      <td>0.494866</td>\n",
       "      <td>16.507465</td>\n",
       "      <td>2.501493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>383.000000</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         A1_Score    A2_Score    A3_Score    A4_Score    A5_Score    A6_Score  \\\n",
       "count  704.000000  704.000000  704.000000  704.000000  704.000000  704.000000   \n",
       "mean     0.721591    0.453125    0.457386    0.495739    0.498580    0.284091   \n",
       "std      0.448535    0.498152    0.498535    0.500337    0.500353    0.451301   \n",
       "min      0.000000    0.000000    0.000000    0.000000    0.000000    0.000000   \n",
       "25%      0.000000    0.000000    0.000000    0.000000    0.000000    0.000000   \n",
       "50%      1.000000    0.000000    0.000000    0.000000    0.000000    0.000000   \n",
       "75%      1.000000    1.000000    1.000000    1.000000    1.000000    1.000000   \n",
       "max      1.000000    1.000000    1.000000    1.000000    1.000000    1.000000   \n",
       "\n",
       "         A7_Score    A8_Score    A9_Score   A10_Score         age      result  \n",
       "count  704.000000  704.000000  704.000000  704.000000  702.000000  704.000000  \n",
       "mean     0.417614    0.649148    0.323864    0.573864   29.698006    4.875000  \n",
       "std      0.493516    0.477576    0.468281    0.494866   16.507465    2.501493  \n",
       "min      0.000000    0.000000    0.000000    0.000000   17.000000    0.000000  \n",
       "25%      0.000000    0.000000    0.000000    0.000000   21.000000    3.000000  \n",
       "50%      0.000000    1.000000    0.000000    1.000000   27.000000    4.000000  \n",
       "75%      1.000000    1.000000    1.000000    1.000000   35.000000    7.000000  \n",
       "max      1.000000    1.000000    1.000000    1.000000  383.000000   10.000000  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "_uuid": "98b101dab80838e621f0fef32119a06d6b761c44"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 704 entries, 0 to 703\n",
      "Data columns (total 21 columns):\n",
      "A1_Score           704 non-null int64\n",
      "A2_Score           704 non-null int64\n",
      "A3_Score           704 non-null int64\n",
      "A4_Score           704 non-null int64\n",
      "A5_Score           704 non-null int64\n",
      "A6_Score           704 non-null int64\n",
      "A7_Score           704 non-null int64\n",
      "A8_Score           704 non-null int64\n",
      "A9_Score           704 non-null int64\n",
      "A10_Score          704 non-null int64\n",
      "age                702 non-null float64\n",
      "gender             704 non-null object\n",
      "ethnicity          609 non-null object\n",
      "jundice            704 non-null object\n",
      "austim             704 non-null object\n",
      "contry_of_res      704 non-null object\n",
      "used_app_before    704 non-null object\n",
      "result             704 non-null int64\n",
      "age_desc           704 non-null object\n",
      "relation           609 non-null object\n",
      "Class/ASD          704 non-null object\n",
      "dtypes: float64(1), int64(11), object(9)\n",
      "memory usage: 115.6+ KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "_uuid": "8bd90b2e4d8f3a783fab6898b93cb053aec929f1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "int64      11\n",
       "object      9\n",
       "float64     1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.dtypes.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "d540231ef070adf8a337d8ecdca5ec5bd5e2112d"
   },
   "source": [
    "Dataset has **704** examples and **21** features. 11 of them are ints, 9 are objects and 1 is float."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "2d21b848c25cb476ba2dc9235ce63ee51e8df169"
   },
   "source": [
    "### MISSING VALUES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "_uuid": "5706923c0677dc620e19b0e1646d5f7a07244608"
   },
   "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>Total</th>\n",
       "      <th>Percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ethnicity</th>\n",
       "      <td>95</td>\n",
       "      <td>13.494318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>relation</th>\n",
       "      <td>95</td>\n",
       "      <td>13.494318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>age</th>\n",
       "      <td>2</td>\n",
       "      <td>0.284091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A9_Score</th>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2_Score</th>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Total    Percent\n",
       "ethnicity     95  13.494318\n",
       "relation      95  13.494318\n",
       "age            2   0.284091\n",
       "A9_Score       0   0.000000\n",
       "A2_Score       0   0.000000"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_missing_data = data.isnull().sum().sort_values(ascending=False)\n",
    "\n",
    "percent_of_missing_data = (data.isnull().sum()/data.isnull().count()*100).sort_values(ascending=False)\n",
    "\n",
    "missing_data = pd.concat(\n",
    "    [\n",
    "        total_missing_data, \n",
    "        percent_of_missing_data\n",
    "    ], \n",
    "    axis=1, \n",
    "    keys=['Total', 'Percent']\n",
    ")\n",
    "\n",
    "missing_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "_uuid": "304aaedc23f9427d388bfe97311b38efbdd7631b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    6\n",
       "1    5\n",
       "2    8\n",
       "3    6\n",
       "4    2\n",
       "5    9\n",
       "6    2\n",
       "7    5\n",
       "8    6\n",
       "9    8\n",
       "dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.iloc[:, 0:10].sum(axis=1).head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "df14bdcd00e72728768b8862805a454420ea38bb"
   },
   "source": [
    "**result** column is a sum of all a*_score columns."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "6d337231c9cf611df9b2ef9ba33006fe05e16ddd"
   },
   "source": [
    "### YES-NO into 0-1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "_uuid": "ba0971c142e60e0b6fa2b72fd0636abe53c7c186"
   },
   "outputs": [],
   "source": [
    "data.columns = map(lambda x: x.strip().lower(), data.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "_uuid": "f5d10d041a0dc36d1673ea3432eb6e482132e244"
   },
   "outputs": [],
   "source": [
    "data.rename(columns={'class/asd': 'decision_class'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "_uuid": "2d824b94d12b640c42d2dbfee71a9d770651caa3"
   },
   "outputs": [],
   "source": [
    "data.jundice = data.jundice.apply(lambda x: 0 if x == 'no' else 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "_uuid": "fa0f6f2471eb6d3595aabafe6e79763a6d39d8d4"
   },
   "outputs": [],
   "source": [
    "data.austim = data.austim.apply(lambda x: 0 if x == 'no' else 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "_uuid": "f7c682bdcaca966bfe714e94b013e82701d7b2f2"
   },
   "outputs": [],
   "source": [
    "data.used_app_before = data.used_app_before.apply(lambda x: 0 if x == 'no' else 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "_uuid": "4f8422e8bac04609de6126b238627c9758c4eb16"
   },
   "outputs": [],
   "source": [
    "data.decision_class = data.decision_class.apply(lambda x: 0 if x == 'NO' else 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "213b6d7285ecbd73b13d8024c62b84c83a3c8e60"
   },
   "source": [
    "### GENDER M-F to 0-1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "_uuid": "b34ede3e102aba86c2e697a4bdb92ce4f26254f5"
   },
   "outputs": [],
   "source": [
    "le = LabelEncoder()\n",
    "\n",
    "data.gender = le.fit_transform(data.gender) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "76a469868f8ca167a64e1dd55303aeccff3803b4"
   },
   "source": [
    "#### STRING VALUES CLEANING"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "_uuid": "a793c863d5b9d054c8d0aad187a0a671dc547347"
   },
   "outputs": [],
   "source": [
    "data.contry_of_res = data.contry_of_res.astype('str')\n",
    "data.contry_of_res = data.contry_of_res.str.lower()\n",
    "data.contry_of_res = data.contry_of_res.str.replace(\"'\", \"\")\n",
    "data.contry_of_res = data.contry_of_res.str.strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "_uuid": "3739e9eefa956a07ac2befe864c5ad2039013216"
   },
   "outputs": [],
   "source": [
    "data.relation = data.relation.replace(np.nan, 'unknown', regex=True)\n",
    "data.relation = data.relation.astype('str')\n",
    "data.relation = data.relation.str.lower()\n",
    "data.relation = data.relation.str.replace(\"'\", \"\")\n",
    "data.relation = data.relation.str.strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "_uuid": "91865e69104a68ee54bd498665b874c9e75d93fe"
   },
   "outputs": [],
   "source": [
    "data.ethnicity = data.ethnicity.replace(np.nan, 'unknown', regex=True)\n",
    "data.ethnicity = data.ethnicity.astype('str')\n",
    "data.ethnicity = data.ethnicity.str.lower()\n",
    "data.ethnicity = data.ethnicity.str.replace(\"'\", \"\")\n",
    "data.ethnicity = data.ethnicity.str.strip()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "44b3b6c4c6292c668fa3e9d590d843d5191f6d89"
   },
   "source": [
    "#### GENDER VS DECISSION CLASS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "_uuid": "fb32ecfd88acb38604bf6574c587686423656676"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    281\n",
       "0    234\n",
       "Name: gender, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.gender[data.decision_class == 0].value_counts() # 0-female, 1-male"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "_uuid": "4d0a470c8487e8fc07728410cb70d7d2d729d774"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    103\n",
       "1     86\n",
       "Name: gender, dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.gender[data.decision_class == 1].value_counts() # 0-female, 1-male"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "d8227c1a895a44486b5b294418e2e7961ab0a1a7"
   },
   "source": [
    "ASD is diagnosed more commonly in **females** than males."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "287d0d72d226f67cf4568803d4ea080e333fe75d"
   },
   "source": [
    "#### ETHNICITY VS DECISSION CLASS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "_uuid": "46187a0f1bb5d5d270ab42c2a65c5816ff1f8b08"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "white-european    124\n",
       "asian             107\n",
       "unknown            86\n",
       "middle eastern     84\n",
       "south asian        33\n",
       "black              25\n",
       "others             22\n",
       "pasifika           11\n",
       "latino             10\n",
       "hispanic            8\n",
       "turkish             5\n",
       "Name: ethnicity, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.ethnicity[data.decision_class == 0].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "_uuid": "731be39a009a27eac2aa4e6f48251d93e8445d31"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1b7ff1fae48>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.ethnicity[data.decision_class == 0].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "_uuid": "36f76b6da30e821f3747e7e1c22580e3017ccc85"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "white-european    109\n",
       "black              18\n",
       "asian              16\n",
       "latino             10\n",
       "unknown             9\n",
       "others              9\n",
       "middle eastern      8\n",
       "hispanic            5\n",
       "south asian         3\n",
       "turkish             1\n",
       "pasifika            1\n",
       "Name: ethnicity, dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.ethnicity[data.decision_class == 1].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "_uuid": "04f5d2fd207b08364da695ff1cc07b7a026010e6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1b7ff51d898>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.ethnicity[data.decision_class == 1].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "f2b086e11c40d700da74f2e549c7bd1e6f6d9fad"
   },
   "source": [
    "ASD prevalence is reported to be highest among **white-european** people."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "fc72f45eee6b045b3ad10bd9fe7f2d561a4ac483"
   },
   "source": [
    "#### RELATION (WHO DID TEST) VS DECISSION CLASS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "_uuid": "1d37d52b616faa3bb53fb29f84d41849ba35882a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "self                        365\n",
       "unknown                      86\n",
       "parent                       38\n",
       "relative                     19\n",
       "others                        4\n",
       "health care professional      3\n",
       "Name: relation, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.relation[data.decision_class == 0].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "_uuid": "6b26c0e2b824378b3f1493eddc968bcbc710e683"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1b7ff5a6828>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.relation[data.decision_class == 0].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "_uuid": "c21fd301e713f153f2cc984812ed6263cd30d785"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "self                        157\n",
       "parent                       12\n",
       "relative                      9\n",
       "unknown                       9\n",
       "health care professional      1\n",
       "others                        1\n",
       "Name: relation, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.relation[data.decision_class == 1].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "_uuid": "1489b034e989b7257497122f492f766d57fa2ffc"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1b7ff5fc128>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.relation[data.decision_class == 1].value_counts().plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "_uuid": "7b6a0062c488761718a0c416e87d3af84231fcd5"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "united arab emirates    79\n",
       "india                   75\n",
       "new zealand             66\n",
       "united states           60\n",
       "united kingdom          48\n",
       "jordan                  45\n",
       "australia               15\n",
       "sri lanka               14\n",
       "afghanistan             11\n",
       "iran                     7\n",
       "Name: contry_of_res, dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.contry_of_res[data.decision_class == 0].value_counts().head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "_uuid": "cf47f95d21754191e395071eb55fe974e8ab585b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1b7ff6790f0>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAFbCAYAAADFmXWwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3XucZFV57vHfA4iIiIA0SEAYVIQQFdRRURDlFiSioAJKNEFDJDkhAokxojFRiYmoJEpM1CCEjHcQNaAYFREBLwFnuMrtoIiIIIwcUMSIoM/5Y+1iapru6erae0/t3vN8P5/5dNfu6rcXTddba6/Lu2SbiIhY+NaadAMiIqIZSegRET2RhB4R0RNJ6BERPZGEHhHRE0noERE9kYQeEdETSegRET2RhB4R0RPrrM4ftummm3rRokWr80dGRCx4y5Yt+4ntqbmet1oT+qJFi1i6dOnq/JEREQuepB+M8rwMuURE9EQSekRET4yU0CX9haSrJH1H0ickrSdpW0kXSbpe0mmS1m27sRERMbs5E7qkLYGjgMW2nwisDbwceCfwHtvbAXcCh7fZ0IiIWLVRh1zWAR4maR1gfeBWYE/gjOrrS4ADm29eRESMas6EbvtHwAnATZRE/lNgGXCX7furp90MbNlWIyMiYm6jDLlsDBwAbAv8FvBwYL8Znjrj0UeSjpC0VNLS5cuX12lrRESswihDLnsD37e93PZ9wGeAZwMbVUMwAFsBt8z0zbZPsr3Y9uKpqTnXxUdExJhGSeg3AbtIWl+SgL2Aq4HzgIOq5xwGnNlOEyMiYhRz7hS1fZGkM4BLgPuBS4GTgLOBT0p6e3XtlHEasOjYs+f1/BuPf8E4PyYiovdG2vpv+y3AW6ZdvgF4RuMtioiIsWSnaERETyShR0T0RBJ6RERPJKFHRPREEnpERE8koUdE9EQSekRETyShR0T0xGo9U3R1m88u1OxAjYiFLj30iIieSEKPiOiJJPSIiJ5IQo+I6Ikk9IiInkhCj4joiST0iIieSEKPiOiJJPSIiJ6YM6FL2l7SZUP/fibpGEmbSDpH0vXVx41XR4MjImJmcyZ029fZ3tn2zsDTgF8AnwWOBc61vR1wbvU4IiImZL5DLnsB37P9A+AAYEl1fQlwYJMNi4iI+ZlvQn858Inq881t3wpQfdxspm+QdISkpZKWLl++fPyWRkTEKo2c0CWtC7wI+NR8foDtk2wvtr14ampqvu2LiIgRzaeHvh9wie3bqse3SdoCoPp4e9ONi4iI0c0noR/KiuEWgLOAw6rPDwPObKpRERExfyMldEnrA/sAnxm6fDywj6Trq68d33zzIiJiVCOdWGT7F8Cjpl27g7LqJSIiOiA7RSMieiIJPSKiJ5LQIyJ6Igk9IqInktAjInpipFUusbJFx5498nNvPP4FLbYkImKF9NAjInoiCT0ioieS0CMieiIJPSKiJ5LQIyJ6Igk9IqInktAjInoiCT0ioieS0CMieiIJPSKiJ5LQIyJ6YtQj6DaSdIakayVdI+lZkjaRdI6k66uPG7fd2IiImN2oPfQTgS/a3gHYCbgGOBY41/Z2wLnV44iImJA5E7qkDYHdgVMAbP/K9l3AAcCS6mlLgAPbamRERMxtlB76Y4HlwKmSLpV0sqSHA5vbvhWg+rhZi+2MiIg5jFIPfR3gqcBrbV8k6UTmMbwi6QjgCICtt956rEauKeZTZx1Saz0iVjZKD/1m4GbbF1WPz6Ak+NskbQFQfbx9pm+2fZLtxbYXT01NNdHmiIiYwZwJ3faPgR9K2r66tBdwNXAWcFh17TDgzFZaGBERIxn1CLrXAh+TtC5wA/BqypvB6ZIOB24CDm6niRERMYqRErrty4DFM3xpr2abExER48pO0YiInkhCj4joiST0iIieSEKPiOiJJPSIiJ5IQo+I6Ikk9IiInkhCj4joiST0iIieSEKPiOiJJPSIiJ5IQo+I6Ikk9IiInkhCj4joiST0iIieSEKPiOiJJPSIiJ4Y6cQiSTcCdwO/Bu63vVjSJsBpwCLgRuAQ23e208yIiJjLfHroe9je2fbgKLpjgXNtbwecWz2OiIgJqTPkcgCwpPp8CXBg/eZERMS4Rk3oBr4saZmkI6prm9u+FaD6uFkbDYyIiNGMNIYO7Gr7FkmbAedIunbUH1C9ARwBsPXWW4/RxIiIGMVIPXTbt1Qfbwc+CzwDuE3SFgDVx9tn+d6TbC+2vXhqaqqZVkdExIPMmdAlPVzSIwafA78LfAc4CzisetphwJltNTIiIuY2ypDL5sBnJQ2e/3HbX5T0beB0SYcDNwEHt9fMiIiYy5wJ3fYNwE4zXL8D2KuNRkVExPyNOikaC9yiY88e+bk3Hv+CFlsSEW3J1v+IiJ5IQo+I6Ikk9IiInkhCj4joiST0iIieSEKPiOiJJPSIiJ7IOvSoJevbI7ojPfSIiJ5IQo+I6Ikk9IiInkhCj4joiST0iIieSEKPiOiJJPSIiJ5IQo+I6ImRE7qktSVdKunz1eNtJV0k6XpJp0lat71mRkTEXObTQz8auGbo8TuB99jeDrgTOLzJhkVExPyMlNAlbQW8ADi5eixgT+CM6ilLgAPbaGBERIxm1B76e4G/Bn5TPX4UcJft+6vHNwNbNty2iIiYhzkTuqT9gdttLxu+PMNTPcv3HyFpqaSly5cvH7OZERExl1F66LsCL5J0I/BJylDLe4GNJA2qNW4F3DLTN9s+yfZi24unpqYaaHJERMxkzoRu+422t7K9CHg58FXbrwDOAw6qnnYYcGZrrYyIiDnVWYf+BuAvJX2XMqZ+SjNNioiIcczrgAvbXwO+Vn1+A/CM5psUERHjyE7RiIieSEKPiOiJJPSIiJ5IQo+I6Ikk9IiInkhCj4joiST0iIieSEKPiOiJJPSIiJ5IQo+I6Ikk9IiInkhCj4joiST0iIieSEKPiOiJJPSIiJ5IQo+I6Ikk9IiInpgzoUtaT9LFki6XdJWkt1XXt5V0kaTrJZ0mad32mxsREbMZpYd+L7Cn7Z2AnYHnS9oFeCfwHtvbAXcCh7fXzIiImMucCd3Fz6uHD6n+GdgTOKO6vgQ4sJUWRkTESEYaQ5e0tqTLgNuBc4DvAXfZvr96ys3Alu00MSIiRjFSQrf9a9s7A1sBzwB+e6anzfS9ko6QtFTS0uXLl4/f0oiIWKV5rXKxfRfwNWAXYCNJ61Rf2gq4ZZbvOcn2YtuLp6am6rQ1IiJWYZRVLlOSNqo+fxiwN3ANcB5wUPW0w4Az22pkRETMbZ25n8IWwBJJa1PeAE63/XlJVwOflPR24FLglBbbGRERc5gzodu+AnjKDNdvoIynR0REB2SnaERETyShR0T0RBJ6RERPJKFHRPREEnpERE8koUdE9EQSekRETyShR0T0RBJ6RERPJKFHRPREEnpERE8koUdE9EQSekRETyShR0T0RBJ6RERPJKFHRPREEnpERE8koUdE9MQoh0Q/RtJ5kq6RdJWko6vrm0g6R9L11ceN229uRETMZpQe+v3A62z/NrALcKSkHYFjgXNtbwecWz2OiIgJmTOh277V9iXV53cD1wBbAgcAS6qnLQEObKuRERExt3mNoUtaBDwFuAjY3PatUJI+sNks33OEpKWSli5fvrxeayMiYlYjJ3RJGwCfBo6x/bNRv8/2SbYX2148NTU1ThsjImIEIyV0SQ+hJPOP2f5Mdfk2SVtUX98CuL2dJkZExChGWeUi4BTgGtv/PPSls4DDqs8PA85svnkRETGqdUZ4zq7AHwBXSrqsuvYm4HjgdEmHAzcBB7fTxIiIGMWcCd321wHN8uW9mm1ORESMKztFIyJ6Igk9IqInktAjInoiCT0ioieS0CMieiIJPSKiJ5LQIyJ6YpSNRRGr3aJjzx75uTce/4IWWxKxcKSHHhHRE0noERE9kSGXWONkOCf6Kj30iIieSEKPiOiJJPSIiJ5IQo+I6Ikk9IiInkhCj4joiVHOFP0PSbdL+s7QtU0knSPp+urjxu02MyIi5jLKOvT/BP4V+PDQtWOBc20fL+nY6vEbmm9exMKR9e0xaXP20G1fAPy/aZcPAJZUny8BDmy4XRERMU/jjqFvbvtWgOrjZs01KSIixtH6pKikIyQtlbR0+fLlbf+4iIg11rgJ/TZJWwBUH2+f7Ym2T7K92PbiqampMX9cRETMZdyEfhZwWPX5YcCZzTQnIiLGNcqyxU8A3wK2l3SzpMOB44F9JF0P7FM9joiICZpz2aLtQ2f50l4NtyUiImrITtGIiJ5IQo+I6ImcWBTRcdmBGqNKDz0ioieS0CMieiJDLhFrsLaGczJMNBnpoUdE9EQSekRET2TIJSIWjAzlrFp66BERPZEeekQE/ZggTg89IqInktAjInoiCT0ioieS0CMieiIJPSKiJ5LQIyJ6olZCl/R8SddJ+q6kY5tqVEREzN/YCV3S2sC/AfsBOwKHStqxqYZFRMT81OmhPwP4ru0bbP8K+CRwQDPNioiI+aqT0LcEfjj0+ObqWkRETIBsj/eN0sHAvrb/uHr8B8AzbL922vOOAI6oHm4PXDfij9gU+MlYjZtM3DZjL7S4bcZeaHHbjL3Q4rYZu+9xt7E9NdeT6tRyuRl4zNDjrYBbpj/J9knASfMNLmmp7cXjN2/1xm0z9kKL22bshRa3zdgLLW6bsRO3qDPk8m1gO0nbSloXeDlwVjPNioiI+Rq7h277fkl/DnwJWBv4D9tXNdayiIiYl1rlc21/AfhCQ22Zbt7DNBOO22bshRa3zdgLLW6bsRda3DZjJy41JkUjIqJbsvU/IqInktAjInqiMwld0uMkPbT6/HmSjpK00aTbNWmS1pK04aTbERHd15kxdEmXAYuBRZSVM2cB29v+vUm2ayaS7gZm/cXZrpWAJX0c+FPg18Ay4JHAP9t+d8242wKvpfyOH5gQt/2imnFfArwT2AxQ9c91fw9tkjQFvIFSh2i9wXXbe06sURMkaRtgO9tfkfQwYB3bdzcQ93HAzbbvlfQ84MnAh23fVTd2PFiXDon+TbUU8sXAe22/T9KlTQRu+sVr+xFV3OOAHwMfoSSxVwCPqNteYEfbP5P0CsoqojdQEnuthA78F3AK8DngNzVjDXsX8ELb1zQY8wGStgPewYP//z22RtiPAacBL6C8eR4GLK8R7wHV39trePAb5x91Mbak11B2c28CPI6ySfCDwF512lr5NLBY0uMpf3tnAR8HanXU2nhDlvSXq/q67X8eN3YV/2jgVOBu4GTgKcCxtr9cJ+6wLiX0+yQdSnlhvbC69pCGYrf14t3X9jOHHn9A0kWUBFfHQyQ9BDgQ+Ffb90lq4lbql7b/pYE4093WVjKvnAq8BXgPsAfwasobaB2Psn2KpKNtnw+cL+n8mjEHzgQuBL5CuctqUhuxj6QU27sIwPb1kjZrKHZbHbU2XtNNdMZW5Y9snyhpX2CK8nd8KtDLhP5qyv+Yf7D9/Wp44KMNxW7rxfvrqhf9ScoQzKE08yL7d+BG4HLggup2+GcNxD1R0lsof0D3Di7avqRm3KWSTqPcAQzH/UzNuAMPs32uJNn+AfBWSRdSkvy47qs+3irpBZSyFVvVbWhlfdtvaCjW6oh9r+1fSeU9UtI6rGJIcZ7a6qg1/pq2/bYG2rUqg07I7wGn2r5cg196QzqT0G1fLekNwNbV4+8DxzcUvq0X7+8DJ1b/DHyjulZL1Yse7kn/QNIedeMCTwL+ANiTFUMurh7XsSHwC+B3h64ZaCqh/1LSWsD11e7kH1HG6+t4u6RHAq8D3kf5b/iLmjEHPi/p96qNd01rI/b5kt4EPEzSPsCfUYblmtBWR63x17Skv7b9LknvY4Y3NNtH1YkPLJP0ZWBb4I2SHkGzQ5+dmhR9IXACsK7tbSXtDBxXd8Kuir0/5Tb1Max48b7Ndidrz0jaHPhH4Lds71cdHPIs26fUjHst8OSqfv2CIenpwDXARsDfUyaJ32X7f2rE3MT2/5t2bduqI1FLNWn+cMrdyn00OEncRuzqzfJwyhuyKIsSTnZXksMM2nhNS3qh7c9JOmymr9teMm7sKv5awM7ADbbvkvQoYEvbV9SJu9LP6Mr/M0nLKD3Fr9l+SnXtSttPmmzLZtfW5Jek/6aMrf2N7Z2qW+BL6/4uqmGR19q+vU6cGeKuR0kIv8PKE1S1JwHbIukbwH62f1Y9/m3gU7afONmWrX6SHk6ZX/l19Xht4KG2f1Ej5pWseiXYk8eN3TZJzwG+Ofh9VNeeWndoshpeeQXwWNvHSdoaeLTti+u1eIXODLkA99v+6bQhpVrvNqvhFqqtya9NbZ8u6Y3wQCG0JuJvDlwr6dusPNZd9y7oI8C1wL7AcZQ/2tqTpJLea/sYSZ9j5v9/ddr9j8Dnqtv17YEPU9rdCEkbA9ux8hvcBR2NfS6wN/Dz6vHDKPMsz64Rc//q45HVx49UH19BGZ4by2yv5YEGXtNQ7lC+LekQ27dV104Gnloz7vspQyx7Ul4nd1NWAT29ZtwHdCmhf0fS7wNrV8vUjgK+WTPmIKksrRlnNm1Nft1T3Y4ZQNIuwE8biFtnEnFVHm/7YEkH2F5SraP/UgNxB0nghAZircT22dVKoi9TVjccaPv6JmJL+mPgaMqY7mXALsC3qD9X0Vbs9WwPkjm2fy5p/TrtrCavkbSr7V2HvnRsdXd03JihB6/lXSlLFk+rHh9MWdrbhOsoS4S/Julw29+k/qoqgGfafupglY/tO1VKjzemSwn9tcDfUHqOg4Tw93UC2v5c9bHW2NcqtDX59ZeU9bqPq/74pyh/sLXYPr8anx/0CC5uaPhlMEF1l6QnUtbmL6ob1Pay6mNTywln6uFtCNwAvFZSUz28oym/4/+xvYekHYCmVlC0Efue4SEFSU8D/rdmzIGHS9rN9ter2M+mzAGMZfBalvQqYA/b91WPP0hzy/9s+/OSrgNOk/QfNLPq575qOGvQUZui4UnRLiX0F9j+G0pSBx445u5T4wac7VZ9oIGhhqOBN0lqevLrKuC5lKEAUXoMtcs0SDqEqudRxX2fpNfbPqNm6JOqYYC/pbwRbQD8Xc2YbY3DTr9ba6pXN+yXtn8pCUkPtX2tpO07HPsY4FOSBieObQG8rGbMgcOB/6hWFAHcBTQxt/JblDurwcT2BtW1JggeWI//HMp8VhNj/v8CfBbYTNI/AAcBb24g7gO6NCl6ie2nznVtnjGfW336EuDRrFgudShwo+03jRu7TW38LqoYlwP7DHrlVQ/hK7Z3qhO3LSrr72c1uK3vGkmfpSzXO4YyFHIn8BA3UMairdjV8NOgA3HtoOfbFJV6RLLdxNAhkl4NvBU4r7r0XOCtbd2NS9ra9k0NxNmBsgNXwLlueEPexBO6pP0oC+0PYcV4GJRb4R1tP6OBn3GB7d3nujZm7MYmqCQ9GtiS8sbz+6wYt9sQ+KDtHWq2daVVQ9UyqsvHXT2jlrdKt0nSrpSEsA3lTnVwd1WnnMBMP+e5lGWW/91CkhzE/mLdpajVUMgiVl6t9eFaDSxxHwq8dIbY446hD8d+NDDYqX2R7R/XjVnFbW3VVjXksjkr/y5qv1EMdGHI5RbKbfCLWPn2926a2+gxJemxtm+AB4pUzXmC9lxamKDaF3hVFW84Gd4NNHE38UVJXwI+UT1+GfVOnBpsld6eMq47WAP8QqCRFR3wwKTw+4DfBtalHHl4T82hrVMof1/LaHh7vqSP2P4DWDH+L+kjlE1ddeKuBVwxWFrZ1NxC1bbHUf6GB78LU1b+1HUmZUJ/GUMrqxpyL3ArJek+QdITGlpJ1NaqrddSFibcRvk9i/J7bm4Jp+1O/KPcNrYV+/nATZSx469RttXv20DcKyl/TJdVj3cATmsg7ktb/F28lPJm8R7gxQ3F/DLwiKHHj6D0Gptq81Lg8cCllGT+asrOwzoxL2rxd3zJtMdrA1c3FPtjwNYNt/caqrv1Fn4X32kp7h9Xr787KcMu/wt8taHYl1Yfr6g+PqSJ2MB3KSULWvm7s92JHvrAIklNV9QbxPhitRRyMGRxre0megutTH7Z/nS1Pnr6LV/t21Tbn6asfW3S1sDwLf+vaGCVyzDb35W0tstmj1Ml1V3Sep6kd1PKEzRS16baNzDYQj+ovSPK76OpMyS3AK6SdDFwz+Ci603wf4cyx3RrzbbN5JuSnmT7yobjtrmSqJVVW8APaWb58ay6lNDbqKg37GmsGMfbqVqiVveW8maVQzj+CzhH0p2UIaRaqiVY61N+DydTZsPH3k2mluu3U25RL64m7Ay8GGhycuoX1XrdyyS9i5J4xl76VhmMvS4eularro3tdwDvkPQO22+s07hVaKOA1KbA1dWbRJMbzgB2A14l6ftV7MFcRd1hhjZXEg1Wbb2ZFau2/raBuDdQ1rafzcq/58bmmiY+KTogaZntpw1P3Em60PZzGog94xihm1lzPPgZTU5QXWH7yUMfNwA+Y/t35/zmVcedsX677brlfpH0VGDw/+oC243Usq9ib0MZd1yXMu79SOD9tr/b1M9oUjXhepnteyS9krLD8ER3d1XOc2e67gbG6GdbqVT3d9Hiap+1gINsn14nziyxZ9zY5warPHYpoX+DkhDOAL5Kqah3vO3a77qSrqGsmGnkP1bSJqv6uqcVfRoj/kW2nynpfyhLLu+gjEVu10Tcua7NM+ZKE3VNq1YFLLH9yhZitzKsJekKYCfKZNdHKBOwL7E9Y+IcMebXbe82w91W50+HAlCprz78e25sZUeTnakqXiMr4CahS0Mux1CGGY6i7BDdg1JDuQlNjxEuo7yoZhoSMlB33P/z1VDOu4FLqpgn14wJLdRvt/0bSZc3tU53hvi/ljQlad0mXqwDTQ9rTXO/bUs6gNIzP0WzVPAble3dqo+NH8LQ0iqiQewXAf9E2fRzO2WZ6DWUN9K6sQdLAAcVMh9NWfxQ1zmS/oqyjHp4nmKsjprarUm08s/qQg+9+h9zvO3XtxT/PErZyjbGCBtXjQneO/ic0rP5Zd2JXEmLKLXbd2VF/fZjbN9YM+5XKRNUTU7UDcf/d8qwxVnT4o899tjWsFYV+3zgi5Qhgd0pJ+lc5o5WDpW0FHg5ZVf2YuAPKeeL1l4qq7KZbU/KBranqNT1P9T2ETXjDi8BfKC2fwNj81Tj/TMl3rE6apKeZntZm0NbA53ooVe9sKdJUlPDItO8tYWYQCuV76CsZX9qFete4F5Jl1Cz2luVuA+o2baZtH3Syy3Vv7VYsfa97t/JoFbJLyT9FmVYa9uaMQdeRtkYdrjtH6uUSa17HmyrWlhFNHCf7TskrSVpLdvnSXpnA3GPphwif0cDsabbkXLIx26Uv7MLKWesjsVDNYmqyf0dqrjXNXnXCR1J6JVLgTMlfYqVe2G1T71p8h1wWNMbi7Rip+jDJD2FlXeK1qp+V8Wf6TzRnwJLbZ85bly3V/Rr4GrbK9X0UanzU0dbw1q47Fj856HHN9HMJp22tLGKaOCu6u7nAuBjkm4H7m8gbptLAJdQjnwcvF4Ora4dUidoNWfzQeB7lNf2tpL+xPZ/14m70s/owpALgKRTZ7hs1zvNvNWJJJXiUYO1sDsP1sLaHquwUTXO+irKbe+3WZHQ7wb+s+6bm6STKL2DQXJ8KaUQ2GMop6gcM2bc6UW/ngM0UfRrEL+V2jZDsR5KKSFbK0Es1InLNlcRqTo8gxWrqh4JfKxuz1rSKZQdyo0vAZR0uafVN5rp2hhxrwX2H/xeJT0OONs1S3oM60wP3farW4jZ2kRSpdG1sC6FhZZIeqnLBqCmPR7Y0/b9AJI+QNnluQ9l1924/gZ4uqcV/aKsWBqbVtT52XLa3cWGjNnLk/SSVXyt1h3havh7a1w1f/UP1SqiX9Lw8Jnte4YeNrk34abq37rVvyZdKmkXV0ccSnomZb6prtunvUneQJkobszEE7raP1WoTa1sLAK2UqlOdzfwIcrY+bG269Z73pJyKz3oiT6ccm7pr1VKAI9rrWlDLHfQQLlf2qnz88JVfK2xg63VchGmprS4imhwlzKoV/LAl2jgbqXJtdsDWlGu+SHAH0q6qXq8DXB1Az/iKklfAE6v4h5MORnpJdDM8PLEEzrtnyrUGtsvrj59a7WS5pGU1Q11/ZHtEyXtSznd/tWUnbR1E/q7KOOkX6O8sHYH/rG6Lf5KjbjTi369HKg9Lmj7cuBySR/3ioMMNgYeY/vOMWM2fic43WwrMGiyCFOzbgS+IamxVURt36VUd4F/zYP3EdQ5uWn/uZ9Sy3qUv4nBapflwCaUTkYjnYnOjKEvVJJ2oyzxOrX6I9vANU+OH1pKdyLl0OzPSrrU1eHZNWNvATyDktAvtt3EHcVgKGPXKu4Ftv+ribhV7K9ReunrUCaflwPn215l+d5JkfRdynFjbazAaIyqqpCS7qKU3FhJ3V6wWtx0JunLlHXifwX8KWXPynK3cyTkgtGFHjoAkhZTxmIH9amBsU+lWS1UtvIupkzOnEq5VfsoJbHVsaz6g90WeKOkR9DcUVVrURLiOsDjJT1+3GWWM0wCDiZxXyPpN5TTZN5t+/012/xI2z+rVhWdavstKrsxu6r1IkwNeVo1IXoTZWNRo9zuprNHuWzYOrpaxXZ+tf6/syQ9lrIPZBfK6+VblH0gtTqAwzqT0CllQV9PmZxr9Jy9Fr0YeApl2Ru2b6mSb12HUzZC3WD7FyoHRtceKqjW/76MsrJleChgrIQ+1yRg1e5vUk47r2Od6s7iEIaOKOyw1oswNeSDlCHCbVl5yHMw7t3EYR9tVIeEFRURb62WA95CWT7cZR8H/o2SN6AMTX6SFYXiautSQl9u+6y5n9Ypv7JtSYNDXxtZu2v7N1RvEtXjOygTjXUdSNmM0fRBAzOqNpQ8r4FQx1EODf+67W9XPZ3rxwkkaU/bX51ttUsTE1O0uwKjMbb/BfgXSR+w/X9a+jFtbTp7u8o5pa+j3F1sSCkf0mWy/ZGhxx+V9OdN/4Am441N0l6UBfznsnKvppFVB21QqfewHWXZ3zsoh99+onqhdI6k/wYOtv3zSbdlUiS9rRqyaXzfQ8ysGtbZzvZXJK0PrG377pqx5t7hAAANI0lEQVQxlwBH276rerwJcEKX//9JOp5ySPagltLLgIdSeu21i/pBtxL6RymbXlYaDujy/yAASfsAv0u5Tf2S7XMm3KRZSfo0pQrg9DfNLi8NHWw6m2lJ61h/G2qxRGoVv40VGAuSpNcARwCb2H6cykEzH7S9V824D1ok0NTCgbao1IiZjd3AYT5dGnLZyR0tXjQbSX9L2cF5ztC1I2yPdTqNWi7LSylutdCGtQA+P/T5epQxyLFX51STdX9OWQ/cho9RVmDsz9AKjJZ+VtcdSVlVdRGA7etVSunWtZakjQfLV6vXTpfy2YPYbqpW0Ky61EP/EPAe200s4F8tVOpS/AQ40vZ51bWxt6RrRZU3UY51u7P6fCPgptXxB7EQVD3sr9Tp8VZvxv9LQyVSp8UeHNZyxWCVlqTzXaMe+kKlFbX9L3WptrgO5czVWqvXJP0h8EbKbmRTJsz/YdoYdSespnkboFvvaLsBh6n5o6ra9CNK9cJPSTrD9rth/GPzBglbpVb3Wba/UD3eD9h73LiSTrd9yNBOuOk/t8u/45lsR3nDq2MwXHPk0LWmVnYsxBUYbTlf0uCc1X0oVQw/Vzeo7Q+rlP3dk/Kae0mHO4PPpRzaM9Mu5cZ2J0O3euitHFXVpqFex3rAByhnDz7JNYvtDHp4064ttb14tu+ZI94Wtm9diL9jWGkbOdXH2yilEDo5YS5pf0rJ1cewYgXG2xbgKq7aqrupwxmaZwJOdlcST890JqFDO7su2yTpQ7ZfM/T4SOB1dSc3qm30F1I2KRl4JbC77X1rxt3P00p1SvpT22PXel5dqjHS4brzHndDVBXvYMqRZXdLejOlXs7fu8GzUCMAJK1yR3OT+xM6k9CHd13afoLKoQOfsl131+WCUyWvt1BqrQw2/hxXd3xX5dCCN9v+avX4DcDzbO9Xs8mt0ix152uOoQ/KK+xGWXJ6AvAm1zhfdSj2FPAaYBEr73ru9IqtNqgcmP1WVuwAHwylNjG0tSBolsOhB9zTQ6Ivo9p1OVh6NDyptCaStEGTa8YlbUpZMfJ64PmUZaIvd1X4qqvUcN35KuZguOwdwJW2P97UsrfqjfNCSoXIB85sdTslkTtNpQb4X/Dg30Wn69wsVF2aFG1l1+VCJOnZlNNzNgC2lrQT8Ce2/6xOXNs/UTm09yuUF9hBC2Qss9G685UfqZxVujfwTpVDLpoo+QuwvtfwIlFDfjp9mG9No9VYIrxLCf306gW2UbUZ4Y8otcA7S9Jjbd/QQuj3APtSrRm3fbmk3ccNpgefoLMuZTXHQSrHuHbyJJ0hbdSdP4Ryl3KC7btUasU0dUj55yX93mCV0ppI0mDp7nmS3k1ZyTG8me2SGb+xn1ZbifDODLnAwtp1CSDpAsqhEd+mjHNfaLvOyT+DuCut3a2u1T4Cqw9UTk5/JGVCs9EDduvSylUnH05JYPdBt4+ga4PK+QCz8Zq4a3Z16FIPnSqBdzqJD7O9u8rhuk8HngecXY17r3LH5wh+WA27uIp/FCve5ddobunA7yZ4AR091zbbe8DMd7EqxdXWOJKeQKnfvoiVJ8sbe3PrVA99oalWSDyn+rcRZQXGhbY/scpvnDvuppS6yXtTendfBo5qYhdjtG9ouGHYT4EfuDrPdU0x087pmfZZrAkkXU4pWTx9gnjZrN80T53qoS9A51PGxd4BfKHBIYDtbb9i+EK1/KuJg2qjfe+nrGsfDL89CbgceFS17r/uUYKdV61E+h3gkdO2vG/IUMGyNcz9tj/Q5g/oVEKvhhd2oIxDXte1MdIZPIpyOtHuwFEqp/R8y/bf1oz7PkpCmOvaSFZD0a9Y2Y3A4bavApC0I2XC9e8pk4O9T+iUU7z2p9y5Dm95v5uyRn+NMfT6+1y1+XD6BHFjr7/OJPSq5sUHge9Rhhm2lfQnXV7yVK2OuIGyxXsr4NmUY+jGIulZVYypabvLNgTWrtHUZayi6BflxJpozg6DZA5g+2pJT7F9gzR2qZ8FxfaZwJmSnmX7W5Nuz4QNv/6gHMoxrLE5hc4kdOCfgD1sfxdA0uOAs2ng9Pi2SPoecB3wdcqb0atr3lWsS1l7vg4wPMH2M+CgcYO2VfQrZnWdpA9QDjKAcpDB/63Wund6E1cL7pB0LrC57SdKejLwIttvn3TDVpeh19/DKMXJdqMk+AspeaMxnZkUlXSB7d2HHotysvvY66/bJmktl+Pimo67TRsFs5ou+hUzm/bCFeUN//3ALymbjtaYE6NUDm5+PfDvQ0twv2P7iZNt2eon6XRK5+xj1aVDgY1sH9LUz5h4D31owuQqSV+gHDpg4GDK+u4ue3zVE2uk9yHpvbaPAf51sGN2mOsfrPuTqhDVcNGvbMFuiKRzXU7ieWu1U/SfZnjaGpPMK+vbvnjaUNMatdJnyPbT9pKcV618aczEEzorT5jcRqkdDOWEl41Xf3Pm5UNUvQ8A21dI+jgw7u3koDj/CQ20bSaHUop+fZYVRb8ObelnrYm2qDY+vUjSJ5lWG38N2x058JNq+HRQ0uMg4NbJNmliLpW0i+3/AZD0TBpeudaZIZeFSNK3bT992o7Oy2zvPOm2rUrTRb+iqJLV4ZShlunbvNfI3ZHVJqKTKJP9dwLfB17RxpBi10m6hrL656bq0taUDYO/oaHDfLrQQwdA5ZCIw3nwwbpdLjnaSu+jrZKjbRX9isL2GcAZKsfb/SvwBMrf8prca/oRcCpwHrAJZQz5MOC4STZqQp7f9g/oTEKnDDdcSylKdRzwCrq/3f1ISu9jB0k/ovQ+XtlA3FOYoeRoAxot+hWz+jFlOGul+u2U49LWNGcCdwGXUL+g2oK2Ou5KupTQH2/7YEkH2F5SjUV/adKNWpWqRsXeVanftWzf3VDo1kqO2v7htAmqJt8wojiKFfXb96h2TTZ2iMECs5Xt1numUXQpoQ/W594l6YmUXs6iyTVnbtW64pdSFdsZJErbdW8n2yo5mqJfq0cb9dsXqm9KelITVUhjbl1K6CdJ2hh4M2VIYAOg7hb6tp1JKbq0jKHE24DBMWjD68NN/Vv2P6UU/doSuJmyBT3j581ro377QrUb8CpJ36e8RgbzQWvsSWRt6sQqF5WTwQ+yffqk2zIfC22DhKRdbX9jrmvRnC7Xb18dJG0z0/U1cZXL6tCJhA4P3im6EEg6CXhf07eTkv5uput1h3JmKWX6oGsRsTB1acjlHEl/BZwG3DO42PFKgG3dTt4z9Pl6lKp1Y491t1j0KyI6pEsJfbDe/Miha6bBSmQt2K+NoLZX2jIu6QSqpYZjaqXoV0R0S2eGXGJ21WTxxba3qxmnlaJfEdENXeqhUy1X3JGVd4p+eHItmgxJV7Jid+HawBQ1dtathqJfEdEBnemhS3oL5aDlHYEvUIYzvm57jRsSmLYy4H7gtjpnUUp6mu1l1YqLB+nywcsRMbouJfQrgZ2AS23vJGlz4GTbL5zjWyMigm4Nufyv7d9Iul/ShsDtdHtCdMFpq+hXRHRDlxL60mp33YcoOy9/Dlw82Sb1TltFvyKiAzoz5DJM0iJgQ9tXTLgpvSLpItvPnPuZEbEQdTKhRzskHU9ZNdN00a+I6IAk9DWIpPNmuLxGnqQT0UdJ6BERPTHxSVFJm6zq6x2v5bKgtFX0KyK6YeIJnbLiwpQldFtTDpIVsBHlMNVtJ9e03mm06FdEdEtnhlwkfRA4y/YXqsf7AXvbft1kW9Zf1YlLZ9ned9JtiYj61pp0A4Y8fZDMAaozNWfcqh6NWZ9s3orojS4MuQz8RNKbgY9ShmBeCdwx2Sb1S9NFvyKiW7o05LIJ8BZgd0rSuQA4LpOizWm66FdEdEtnEvqApA1s/3zS7YiIWGg6M4Yu6dmSrgaurh7vJOn9E25WRMSC0ZmEDrwH2Jdq3Nz25ZThl4iIGEGXEjq2fzjtUioCRkSMqEurXH4o6dmAJa0LHEU2vUREjKwzk6KSNgVOBPam7BT9MnBUVrlERIymSwl9V9vfmOtaRETMrEsJ/RLbT53rWkREzGziY+iSngU8G5iS9JdDX9qQspsxIiJGMPGEDqwLbEBpyyOGrv8MOGgiLYqIWIC6NOSyje0fTLodEREL1cQTuqT32j5G0udYUTjqAbZfNIFmRUQsOF0YcvlI9fGEibYiImKBm3gPPSIimtGFHjpQ1pwDbwW2obRLlBPpcwBDRMQIOtNDl3Qt8BeUM0YfqOFiO4dcRESMoDM9dOCn1bFzERExhi710I+nbCT6DHDv4LrtSybWqIiIBaRLCf28GS7b9p6rvTEREQtQZxJ6RETU05kxdEl/N9N12zmVPiJiBJ1J6MA9Q5+vB+xPDriIiBhZZ4dcJD0UOMv2vpNuS0TEQtCpM0WnWR/IpqKIiBF1ZshF0pWsKM61NjAFZPw8ImJEnRlykbTN0MP7gdts3z+p9kRELDSdSegREVFPl8fQIyJiHpLQIyJ6Igk9IqInktAjInoiCT0ioif+PyFkrm+SJcpbAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.contry_of_res[data.decision_class == 0].value_counts().head(15).plot(kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "_uuid": "7d158ee68ce4d18dcde8c2ccb1812f43e44719ab"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "united states     53\n",
       "united kingdom    29\n",
       "new zealand       15\n",
       "australia         12\n",
       "canada            10\n",
       "india              6\n",
       "france             5\n",
       "brazil             5\n",
       "mexico             4\n",
       "italy              4\n",
       "Name: contry_of_res, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.contry_of_res[data.decision_class == 1].value_counts().head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "_uuid": "7ce69bb026b3af30c66f41b177f74f62c7c27d4d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1b7ff6f1dd8>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.contry_of_res[data.decision_class == 1].value_counts().head(15).plot(kind='bar')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "dfa9f1aafb94ee3c25717a9df34ddc0ad34f069b"
   },
   "source": [
    "ASD prevalence is reported to be the highest among **USA** and **UK** citizens."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "2f4fad97943d8e5df33e79c014c18108ba3188eb"
   },
   "source": [
    "### LABEL BINARIZING"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "881e0fec42e8e1a5fa684432c546c0d0bb1a967f"
   },
   "source": [
    "#### contry_of_res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "_uuid": "6e2069b80a3639a92264f20fcffddd2866450d1c"
   },
   "outputs": [],
   "source": [
    "lb = LabelBinarizer()\n",
    "\n",
    "lb.fit(data.contry_of_res.values)\n",
    "\n",
    "binarized_data = lb.transform(data.contry_of_res.values)\n",
    "\n",
    "binarized_contry_of_res_matrix_data = np.vstack(binarized_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "_uuid": "12778d973ea1af878acea979e250bb45731b07e0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(704, 67)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "binarized_contry_of_res_matrix_data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "35a49563c453a056505e9b6f16ea99289832ef31"
   },
   "source": [
    "#### relation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "_uuid": "f8f2a20ebadc42daeedacb313570c62e27ad8503"
   },
   "outputs": [],
   "source": [
    "lb = LabelBinarizer()\n",
    "\n",
    "lb.fit(data.relation.values)\n",
    "\n",
    "binarized_data = lb.transform(data.relation.values)\n",
    "\n",
    "binarized_result_matrix_data = np.vstack(binarized_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "_uuid": "4967dacf4f50dbce083c68c1bf42d7d30d37da18"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(704, 6)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "binarized_result_matrix_data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "296e3bb518fe35ae25057638fb7d68568ee18a43"
   },
   "source": [
    "#### ethnicity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "_uuid": "4591813fa9b333bd953e1611f628d799ec0702c5"
   },
   "outputs": [],
   "source": [
    "lb = LabelBinarizer()\n",
    "\n",
    "lb.fit(data.ethnicity.values)\n",
    "\n",
    "binarized_data = lb.transform(data.ethnicity.values)\n",
    "\n",
    "binarized_ethnicity_matrix_data = np.vstack(binarized_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "_uuid": "fac0d33ac494656719817a2bdfd15640f2197b78"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(704, 11)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "binarized_ethnicity_matrix_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "_uuid": "286848ecf874eaef57ae223f06af6d003393b931"
   },
   "outputs": [],
   "source": [
    "data.drop(['age_desc', 'result'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "4f36957ed138e99230129b5d00177a6669382836"
   },
   "source": [
    "I'm dropping **age_desc** and **result** columns, because there's no such relation between other attributes.\n",
    "\n",
    "We can assume **age_desc** column represents the information whether the patient is an adult or not. Every row has the same value - \"18 and more\", so in our situation its redundant.\n",
    "\n",
    "Also, **result** column is a sum of all a*_score columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "_uuid": "aed86ff31bb236191c7e628f37c56200cdb529c3"
   },
   "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>a1_score</th>\n",
       "      <th>a2_score</th>\n",
       "      <th>a3_score</th>\n",
       "      <th>a4_score</th>\n",
       "      <th>a5_score</th>\n",
       "      <th>a6_score</th>\n",
       "      <th>a7_score</th>\n",
       "      <th>a8_score</th>\n",
       "      <th>a9_score</th>\n",
       "      <th>a10_score</th>\n",
       "      <th>age</th>\n",
       "      <th>gender</th>\n",
       "      <th>ethnicity</th>\n",
       "      <th>jundice</th>\n",
       "      <th>austim</th>\n",
       "      <th>contry_of_res</th>\n",
       "      <th>used_app_before</th>\n",
       "      <th>relation</th>\n",
       "      <th>decision_class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>white-european</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>united states</td>\n",
       "      <td>0</td>\n",
       "      <td>self</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>24.0</td>\n",
       "      <td>1</td>\n",
       "      <td>latino</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>brazil</td>\n",
       "      <td>0</td>\n",
       "      <td>self</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>27.0</td>\n",
       "      <td>1</td>\n",
       "      <td>latino</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>spain</td>\n",
       "      <td>0</td>\n",
       "      <td>parent</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>white-european</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>united states</td>\n",
       "      <td>0</td>\n",
       "      <td>self</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>egypt</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   a1_score  a2_score  a3_score  a4_score  a5_score  a6_score  a7_score  \\\n",
       "0         1         1         1         1         0         0         1   \n",
       "1         1         1         0         1         0         0         0   \n",
       "2         1         1         0         1         1         0         1   \n",
       "3         1         1         0         1         0         0         1   \n",
       "4         1         0         0         0         0         0         0   \n",
       "\n",
       "   a8_score  a9_score  a10_score   age  gender       ethnicity  jundice  \\\n",
       "0         1         0          0  26.0       0  white-european        0   \n",
       "1         1         0          1  24.0       1          latino        0   \n",
       "2         1         1          1  27.0       1          latino        1   \n",
       "3         1         0          1  35.0       0  white-european        0   \n",
       "4         1         0          0  40.0       0         unknown        0   \n",
       "\n",
       "   austim  contry_of_res  used_app_before relation  decision_class  \n",
       "0       0  united states                0     self               0  \n",
       "1       1         brazil                0     self               0  \n",
       "2       1          spain                0   parent               1  \n",
       "3       1  united states                0     self               0  \n",
       "4       0          egypt                0  unknown               0  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "_uuid": "c1d5fb73964612151dbef3477784708c23c4f671"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(383.0, 17.0)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.age.max(), data.age.min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "_uuid": "baf13216ec5cc4e3b537292c3e2e69e2ef35a138"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The oldest patient: 383.0 years.\n",
      "The youngest patient: 17.0 years.\n",
      "Average age: 29.698005698005698 years.\n",
      "Median age: 27.0 years.\n"
     ]
    }
   ],
   "source": [
    "print('The oldest patient: {} years.'.format(data.age.max()))\n",
    "print('The youngest patient: {} years.'.format(data.age.min()))\n",
    "print('Average age: {} years.'.format(data.age.mean()))\n",
    "print('Median age: {} years.'.format(data.age.median(skipna=True)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "_uuid": "04238fc72703be9fbf7b82f241642e4a1c423f6c"
   },
   "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>a1_score</th>\n",
       "      <th>a2_score</th>\n",
       "      <th>a3_score</th>\n",
       "      <th>a4_score</th>\n",
       "      <th>a5_score</th>\n",
       "      <th>a6_score</th>\n",
       "      <th>a7_score</th>\n",
       "      <th>a8_score</th>\n",
       "      <th>a9_score</th>\n",
       "      <th>a10_score</th>\n",
       "      <th>age</th>\n",
       "      <th>gender</th>\n",
       "      <th>ethnicity</th>\n",
       "      <th>jundice</th>\n",
       "      <th>austim</th>\n",
       "      <th>contry_of_res</th>\n",
       "      <th>used_app_before</th>\n",
       "      <th>relation</th>\n",
       "      <th>decision_class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>383.0</td>\n",
       "      <td>0</td>\n",
       "      <td>pasifika</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>new zealand</td>\n",
       "      <td>0</td>\n",
       "      <td>self</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    a1_score  a2_score  a3_score  a4_score  a5_score  a6_score  a7_score  \\\n",
       "52         1         0         0         0         0         0         0   \n",
       "\n",
       "    a8_score  a9_score  a10_score    age  gender ethnicity  jundice  austim  \\\n",
       "52         0         0          0  383.0       0  pasifika        0       0   \n",
       "\n",
       "   contry_of_res  used_app_before relation  decision_class  \n",
       "52   new zealand                0     self               0  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.loc[(data.age == 383)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "18410de435fe7ff19da65317283070e06974f165"
   },
   "source": [
    "We can either assume it's been put by a mistake and calculate mean/median value or delete that observation.\n",
    "\n",
    "Here I will calculate a median value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "_uuid": "1de627dd52ad21922e096f996aa85d3469f7d345"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "27.0"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.age.median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "_uuid": "598c0fb191144bae97d0d7c365560a46505222c3"
   },
   "outputs": [],
   "source": [
    "data.age.replace(data.age.max(), data.age.median(), inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "_uuid": "96926270292bb9813bc1139a69552fd4f4bc2777"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Iqra Altaf\\python\\lib\\site-packages\\statsmodels\\nonparametric\\kde.py:448: RuntimeWarning: invalid value encountered in greater\n",
      "  X = X[np.logical_and(X > clip[0], X < clip[1])] # won't work for two columns.\n",
      "C:\\Users\\Iqra Altaf\\python\\lib\\site-packages\\statsmodels\\nonparametric\\kde.py:448: RuntimeWarning: invalid value encountered in less\n",
      "  X = X[np.logical_and(X > clip[0], X < clip[1])] # won't work for two columns.\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,8))\n",
    "\n",
    "sns.kdeplot(\n",
    "    data.age[data.decision_class == 1], \n",
    "    color=\"darkturquoise\", \n",
    "    shade=True\n",
    ")\n",
    "\n",
    "sns.kdeplot(\n",
    "    data.age[data.decision_class == 0], \n",
    "    color=\"lightcoral\", \n",
    "    shade=True\n",
    ")\n",
    "\n",
    "plt.legend(['ASD', 'NOT ASD'])\n",
    "plt.title('age vs decision_class')\n",
    "plt.xlim(data.age.min() - 10, data.age.max() + 10)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "_uuid": "a512b209e1fddea568bd8be2e59f88ca6ab3a3d8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    515\n",
       "1    189\n",
       "Name: decision_class, dtype: int64"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.decision_class.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "_uuid": "33831923a65ae568dd3b52cb608368a2fa9db288"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    103\n",
       "1     86\n",
       "Name: gender, dtype: int64"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.gender[data.decision_class == 1].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "_uuid": "40c99ebb790c561db8907e2a45ad0c9ee70da92e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    281\n",
       "0    234\n",
       "Name: gender, dtype: int64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.gender[data.decision_class == 0].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "_uuid": "ea55c3a174ed30846a3148f85e78d630690362e0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a1_score           0\n",
       "a2_score           0\n",
       "a3_score           0\n",
       "a4_score           0\n",
       "a5_score           0\n",
       "a6_score           0\n",
       "a7_score           0\n",
       "a8_score           0\n",
       "a9_score           0\n",
       "a10_score          0\n",
       "age                2\n",
       "gender             0\n",
       "ethnicity          0\n",
       "jundice            0\n",
       "austim             0\n",
       "contry_of_res      0\n",
       "used_app_before    0\n",
       "relation           0\n",
       "decision_class     0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "d90a1aa1b8fbea83e38370e15e1984e702438f91"
   },
   "source": [
    "### MISSING VALUES IMPUTATION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "_uuid": "f8a19087d84a9a032fb24056ea61a2bc14b39eef"
   },
   "outputs": [],
   "source": [
    "to_update_nans_dict = {}\n",
    "\n",
    "columns = [\n",
    "    'age'\n",
    "]\n",
    "\n",
    "for _decision_class in [0, 1]:\n",
    "    for column in columns:\n",
    "        vals = data[data.decision_class == _decision_class][column].value_counts()\n",
    "        \n",
    "        to_update_nans_dict['{decision_class}_{column}'.format(\n",
    "            decision_class=_decision_class,\n",
    "            column=column\n",
    "        )] = vals.idxmax()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "_uuid": "a2c4b191b3f1bdf2d88d21861273e0ed1f8d4822"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'0_age': 21.0, '1_age': 30.0}"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "to_update_nans_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "_uuid": "0225cbb26d69a0db44017621d5f558c4915940c0"
   },
   "outputs": [],
   "source": [
    "data.iloc[62, data.columns.get_loc('age')] = to_update_nans_dict.get('0_age')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "_uuid": "ba6add993932672c8fe8f1d538247351e9c1f207"
   },
   "outputs": [],
   "source": [
    "data.iloc[91, data.columns.get_loc('age')] = to_update_nans_dict.get('1_age')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "f7e7eb5e2021f5843a0d8d5c1404e2125382a8c0"
   },
   "source": [
    "### HEATMAP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "_uuid": "b08d22daac38ef4cb867368b67ddc6698c3dd9be"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2880x1440 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "corr = data.corr()\n",
    "\n",
    "sns.heatmap(\n",
    "    data=corr,\n",
    "    annot=True,\n",
    "    fmt='.2f',\n",
    "    linewidths=.5,\n",
    "    cmap='RdYlGn',\n",
    "    xticklabels=corr.columns.values,\n",
    "    yticklabels=corr.columns.values\n",
    ")\n",
    "\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(40, 20)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "92a12187bc5b11b477b8fc96cada36355d31d8e0"
   },
   "source": [
    "### PREPARING DATA FOR MODEL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "_uuid": "4f1a73a7d3191acf9a6f401a59de5180e14705af"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Iqra Altaf\\python\\lib\\site-packages\\ipykernel_launcher.py:5: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  \"\"\"\n"
     ]
    }
   ],
   "source": [
    "feature_names = list(\n",
    "    set(data.columns[0:-1].tolist()).difference(['contry_of_res', 'relation', 'ethnicity'])\n",
    ")\n",
    "\n",
    "X = data[feature_names].as_matrix()\n",
    "\n",
    "X_new = np.hstack((\n",
    "    X, \n",
    "    binarized_contry_of_res_matrix_data,\n",
    "    binarized_ethnicity_matrix_data,\n",
    "    binarized_result_matrix_data,\n",
    "))\n",
    "                   \n",
    "y = data.decision_class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "_uuid": "9475a8d39d06358c545e7abace6e3067db25c539"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((704, 99), (704,))"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_new.shape, y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "_uuid": "653f415ca99a91b84e75a3b4efc4e20dd4337087"
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X_new,\n",
    "    y,\n",
    "    random_state=42,\n",
    "    test_size=0.3\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "_uuid": "23abfd1422ac4530200b4d1be0d86f4bf8865842"
   },
   "outputs": [],
   "source": [
    "log_reg = LogisticRegression(\n",
    "    C=1,\n",
    "    penalty='l1',\n",
    "    solver='liblinear',\n",
    "    random_state=42,\n",
    "    multi_class='ovr'\n",
    "\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "_uuid": "67474554d880b3a7c6ab6718f743eb1fa044b30d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr',\n",
       "          n_jobs=None, penalty='l1', random_state=42, solver='liblinear',\n",
       "          tol=0.0001, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log_reg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "_uuid": "55a26bc8ee8c34440a08aa6259261300fb1d7a5d"
   },
   "outputs": [],
   "source": [
    "log_reg_predict = log_reg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "_uuid": "d7091d8b7e67ce91a58e140076e26262297f749e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log_reg.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "_uuid": "77cdf193e68be4868370aa6df73fcf53598ecc94"
   },
   "outputs": [],
   "source": [
    "preds = log_reg.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "_uuid": "cfb1a27971fd246c5f45a89f0de894f38233763b"
   },
   "outputs": [],
   "source": [
    "log_reg_predict_proba = log_reg.predict_proba(X_test)[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "_uuid": "6b90b5cf2f73e29ac7291c0dc8c8b40a65f5957b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Logistic Regression Accuracy: 100.00%\n",
      "Logistic Regression AUC: 100.00%\n",
      "Logistic Regression Classification report:\n",
      "\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00       155\n",
      "           1       1.00      1.00      1.00        57\n",
      "\n",
      "   micro avg       1.00      1.00      1.00       212\n",
      "   macro avg       1.00      1.00      1.00       212\n",
      "weighted avg       1.00      1.00      1.00       212\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print('\\nLogistic Regression Accuracy: {:.2f}%'.format(accuracy_score(y_test, log_reg_predict) * 100))\n",
    "print('Logistic Regression AUC: {:.2f}%'.format(roc_auc_score(y_test, log_reg_predict) * 100))\n",
    "print('Logistic Regression Classification report:\\n\\n', classification_report(y_test, log_reg_predict))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "_uuid": "dae17ed67cc428bb93a5adebdfb1b1c2921296ea"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[155   0]\n",
      " [  0  57]]\n"
     ]
    }
   ],
   "source": [
    "print(confusion_matrix(y_test, preds))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "b0680fb40b3230c3891f05c12a1701b77c8d5193"
   },
   "source": [
    "### ROC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "_uuid": "396b72dca8b4c2ecf38fd2a751f2be52a34fca75"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fpr, tpr, thresholds = roc_curve(\n",
    "    y_test, \n",
    "    log_reg_predict_proba\n",
    ")\n",
    "\n",
    "plt.plot(\n",
    "    [0, 1], \n",
    "    [0, 1], \n",
    "    'k--'\n",
    ")\n",
    "\n",
    "plt.plot(fpr, tpr)\n",
    "\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.0])\n",
    "\n",
    "plt.rcParams['font.size'] = 14\n",
    "\n",
    "plt.title('ROC curve for Logistic Regression')\n",
    "plt.xlabel('False Positive Rate (1 - Specificity)')\n",
    "plt.ylabel('True Positive Rate (Sensitivity)')\n",
    "\n",
    "plt.grid(True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "3e4b5c3ebf4253344c5e87effe768e7e83a2ce01"
   },
   "source": [
    "### F1 SCORE AFTER 10-FOLD CROSS-VALIDATION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "_uuid": "e3744a3b1544e330070ff4f15d473638cf5742b6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross-validated scores: [0.97297297 1.         1.         1.         1.         1.\n",
      " 1.         1.         1.         1.        ]\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      1.00      1.00       515\n",
      "           1       1.00      0.99      1.00       189\n",
      "\n",
      "   micro avg       1.00      1.00      1.00       704\n",
      "   macro avg       1.00      1.00      1.00       704\n",
      "weighted avg       1.00      1.00      1.00       704\n",
      "\n",
      "\n",
      "LogisticRegression: F1 after 10-fold cross-validation: 99.73% (+/- 0.02%)\n"
     ]
    }
   ],
   "source": [
    "kfold = StratifiedKFold(\n",
    "    n_splits=10, \n",
    "    shuffle=True, \n",
    "    random_state=42\n",
    ")\n",
    "\n",
    "predicted = cross_val_predict(\n",
    "    log_reg, \n",
    "    X_new, \n",
    "    y, \n",
    "    cv=kfold\n",
    ")\n",
    "\n",
    "scores = cross_val_score(\n",
    "    log_reg, \n",
    "    X_new, \n",
    "    y, \n",
    "    cv=kfold,\n",
    "    scoring='f1'\n",
    ")\n",
    "\n",
    "print('Cross-validated scores: {}\\n'.format(scores))\n",
    "\n",
    "print(classification_report(y, predicted))\n",
    "\n",
    "print(\"\\nLogisticRegression: F1 after 10-fold cross-validation: {:.2f}% (+/- {:.2f}%)\".format(\n",
    "    scores.mean() * 100,\n",
    "    scores.std() * 2\n",
    "))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_uuid": "60f3907a0504572e338e79a44c6ac07537eb8df6"
   },
   "source": [
    "Cross-validation minimizes classification error estimation bias and is thus primarily used for small sample datasets. As expected, the classification f1 score dropped from ∼100% for the full dataset to ∼99.75% with 10-fold or percentage-split cross-validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "_uuid": "0ec4afa40f4e90018808f2079edbe7db6895a27c",
    "collapsed": true
   },
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}