P4_Tree.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true,
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Decision Tree Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 408,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "import jenkspy as jenkspy\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import tree, metrics\n",
    "from sklearn.metrics import confusion_matrix, classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 409,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "           age  hypertension  heart_disease  avg_glucose_level       bmi  \\\n0     0.931846             0              1           2.500974  0.859059   \n1     1.673346             0              1          -0.056516  0.279850   \n2    -0.094846             0              0           1.303993  0.548264   \n3     1.616307             1              0           1.364196 -0.920950   \n4     1.730384             0              0           1.616050 -0.214597   \n...        ...           ...            ...                ...       ...   \n4189 -0.551154             0              0          -0.801661 -0.227499   \n4190 -0.608193             0              0           1.718958  0.345098   \n4191 -0.323000             1              0          -0.283580  0.402209   \n4192 -0.608193             0              0          -0.514394  0.025377   \n4193  1.673346             1              0          -0.518352  0.179082   \n\n      work_type_Govt_job  work_type_Never_worked  work_type_Private  \\\n0                      0                       0                  1   \n1                      0                       0                  1   \n2                      0                       0                  1   \n3                      0                       0                  0   \n4                      0                       0                  1   \n...                  ...                     ...                ...   \n4189                   0                       0                  1   \n4190                   0                       0                  1   \n4191                   1                       0                  0   \n4192                   0                       0                  1   \n4193                   0                       0                  1   \n\n      work_type_Self-employed  smoking_status_Unknown  ...  \\\n0                           0                       0  ...   \n1                           0                       0  ...   \n2                           0                       0  ...   \n3                           1                       0  ...   \n4                           0                       0  ...   \n...                       ...                     ...  ...   \n4189                        0                       0  ...   \n4190                        0                       0  ...   \n4191                        0                       0  ...   \n4192                        0                       0  ...   \n4193                        0                       0  ...   \n\n      smoking_status_never smoked  smoking_status_smokes  gender_Female  \\\n0                               0                      0              0   \n1                               1                      0              0   \n2                               0                      1              1   \n3                               1                      0              1   \n4                               0                      0              0   \n...                           ...                    ...            ...   \n4189                            0                      0              0   \n4190                            0                      1              0   \n4191                            0                      1              1   \n4192                            0                      1              0   \n4193                            1                      0              1   \n\n      gender_Male  gender_Other  ever_married_No  ever_married_Yes  \\\n0               1             0                0                 1   \n1               1             0                0                 1   \n2               0             0                0                 1   \n3               0             0                0                 1   \n4               1             0                0                 1   \n...           ...           ...              ...               ...   \n4189            1             0                1                 0   \n4190            1             0                0                 1   \n4191            0             0                0                 1   \n4192            1             0                0                 1   \n4193            0             0                0                 1   \n\n      Residence_type_Rural  Residence_type_Urban  stroke  \n0                        0                     1       1  \n1                        1                     0       1  \n2                        0                     1       1  \n3                        1                     0       1  \n4                        0                     1       1  \n...                    ...                   ...     ...  \n4189                     1                     0       0  \n4190                     0                     1       0  \n4191                     1                     0       0  \n4192                     1                     0       0  \n4193                     0                     1       0  \n\n[4194 rows x 21 columns]",
      "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>age</th>\n      <th>hypertension</th>\n      <th>heart_disease</th>\n      <th>avg_glucose_level</th>\n      <th>bmi</th>\n      <th>work_type_Govt_job</th>\n      <th>work_type_Never_worked</th>\n      <th>work_type_Private</th>\n      <th>work_type_Self-employed</th>\n      <th>smoking_status_Unknown</th>\n      <th>...</th>\n      <th>smoking_status_never smoked</th>\n      <th>smoking_status_smokes</th>\n      <th>gender_Female</th>\n      <th>gender_Male</th>\n      <th>gender_Other</th>\n      <th>ever_married_No</th>\n      <th>ever_married_Yes</th>\n      <th>Residence_type_Rural</th>\n      <th>Residence_type_Urban</th>\n      <th>stroke</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>0.931846</td>\n      <td>0</td>\n      <td>1</td>\n      <td>2.500974</td>\n      <td>0.859059</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>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1.673346</td>\n      <td>0</td>\n      <td>1</td>\n      <td>-0.056516</td>\n      <td>0.279850</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>1</td>\n      <td>0</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    </tr>\n    <tr>\n      <th>2</th>\n      <td>-0.094846</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1.303993</td>\n      <td>0.548264</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>0</td>\n      <td>1</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>1</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>1.616307</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1.364196</td>\n      <td>-0.920950</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>...</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>1</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>1.730384</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1.616050</td>\n      <td>-0.214597</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>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>4189</th>\n      <td>-0.551154</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.801661</td>\n      <td>-0.227499</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>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>4190</th>\n      <td>-0.608193</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1.718958</td>\n      <td>0.345098</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>0</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>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>4191</th>\n      <td>-0.323000</td>\n      <td>1</td>\n      <td>0</td>\n      <td>-0.283580</td>\n      <td>0.402209</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</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    </tr>\n    <tr>\n      <th>4192</th>\n      <td>-0.608193</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.514394</td>\n      <td>0.025377</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>0</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>0</td>\n    </tr>\n    <tr>\n      <th>4193</th>\n      <td>1.673346</td>\n      <td>1</td>\n      <td>0</td>\n      <td>-0.518352</td>\n      <td>0.179082</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>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>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>4194 rows × 21 columns</p>\n</div>"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 409
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "df=pd.read_csv('prepared_data.csv')\n",
    "df\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 410,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "           age  hypertension  heart_disease  avg_glucose_level       bmi  \\\n674  -0.608193             1              0           2.153504 -0.299359   \n1332  0.475538             0              0          -0.919151  0.138579   \n530  -1.349693             0              0          -0.281497 -0.497138   \n1319 -0.551154             0              0          -0.119636  0.943821   \n1730 -0.608193             0              0          -0.591262 -0.030945   \n...        ...           ...            ...                ...       ...   \n2555  0.190346             0              0          -0.475439 -1.245872   \n2540  0.931846             0              0          -0.898319  0.378739   \n2944 -0.323000             0              0          -0.181714  0.053817   \n228   1.160000             0              0           1.866028 -0.454757   \n3670 -0.722270             1              0          -1.010393  1.282871   \n\n      work_type_Govt_job  work_type_Never_worked  work_type_Private  \\\n674                    1                       0                  0   \n1332                   0                       0                  1   \n530                    0                       0                  1   \n1319                   0                       0                  0   \n1730                   0                       0                  1   \n...                  ...                     ...                ...   \n2555                   0                       0                  1   \n2540                   0                       0                  1   \n2944                   0                       0                  1   \n228                    0                       0                  1   \n3670                   0                       0                  1   \n\n      work_type_Self-employed  smoking_status_Unknown  ...  \\\n674                         0                       0  ...   \n1332                        0                       0  ...   \n530                         0                       0  ...   \n1319                        1                       0  ...   \n1730                        0                       0  ...   \n...                       ...                     ...  ...   \n2555                        0                       0  ...   \n2540                        0                       0  ...   \n2944                        0                       1  ...   \n228                         0                       0  ...   \n3670                        0                       0  ...   \n\n      smoking_status_never smoked  smoking_status_smokes  gender_Female  \\\n674                             1                      0              0   \n1332                            1                      0              0   \n530                             1                      0              1   \n1319                            0                      0              1   \n1730                            0                      0              0   \n...                           ...                    ...            ...   \n2555                            0                      0              0   \n2540                            0                      0              0   \n2944                            0                      0              0   \n228                             0                      0              0   \n3670                            1                      0              1   \n\n      gender_Male  gender_Other  ever_married_No  ever_married_Yes  \\\n674             1             0                0                 1   \n1332            1             0                0                 1   \n530             0             0                0                 1   \n1319            0             0                0                 1   \n1730            1             0                0                 1   \n...           ...           ...              ...               ...   \n2555            1             0                0                 1   \n2540            1             0                0                 1   \n2944            1             0                0                 1   \n228             1             0                0                 1   \n3670            0             0                0                 1   \n\n      Residence_type_Rural  Residence_type_Urban  stroke  \n674                      1                     0       0  \n1332                     1                     0       0  \n530                      0                     1       0  \n1319                     1                     0       0  \n1730                     1                     0       0  \n...                    ...                   ...     ...  \n2555                     0                     1       0  \n2540                     1                     0       0  \n2944                     1                     0       0  \n228                      0                     1       0  \n3670                     1                     0       0  \n\n[3355 rows x 21 columns]",
      "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>age</th>\n      <th>hypertension</th>\n      <th>heart_disease</th>\n      <th>avg_glucose_level</th>\n      <th>bmi</th>\n      <th>work_type_Govt_job</th>\n      <th>work_type_Never_worked</th>\n      <th>work_type_Private</th>\n      <th>work_type_Self-employed</th>\n      <th>smoking_status_Unknown</th>\n      <th>...</th>\n      <th>smoking_status_never smoked</th>\n      <th>smoking_status_smokes</th>\n      <th>gender_Female</th>\n      <th>gender_Male</th>\n      <th>gender_Other</th>\n      <th>ever_married_No</th>\n      <th>ever_married_Yes</th>\n      <th>Residence_type_Rural</th>\n      <th>Residence_type_Urban</th>\n      <th>stroke</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>674</th>\n      <td>-0.608193</td>\n      <td>1</td>\n      <td>0</td>\n      <td>2.153504</td>\n      <td>-0.299359</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>...</td>\n      <td>1</td>\n      <td>0</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>0</td>\n    </tr>\n    <tr>\n      <th>1332</th>\n      <td>0.475538</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.919151</td>\n      <td>0.138579</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>1</td>\n      <td>0</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>0</td>\n    </tr>\n    <tr>\n      <th>530</th>\n      <td>-1.349693</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.281497</td>\n      <td>-0.497138</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>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>0</td>\n    </tr>\n    <tr>\n      <th>1319</th>\n      <td>-0.551154</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.119636</td>\n      <td>0.943821</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>...</td>\n      <td>0</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>1</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1730</th>\n      <td>-0.608193</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.591262</td>\n      <td>-0.030945</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>0</td>\n      <td>0</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>0</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>2555</th>\n      <td>0.190346</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.475439</td>\n      <td>-1.245872</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>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2540</th>\n      <td>0.931846</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.898319</td>\n      <td>0.378739</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>0</td>\n      <td>0</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>0</td>\n    </tr>\n    <tr>\n      <th>2944</th>\n      <td>-0.323000</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.181714</td>\n      <td>0.053817</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>0</td>\n      <td>0</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>0</td>\n    </tr>\n    <tr>\n      <th>228</th>\n      <td>1.160000</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1.866028</td>\n      <td>-0.454757</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>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3670</th>\n      <td>-0.722270</td>\n      <td>1</td>\n      <td>0</td>\n      <td>-1.010393</td>\n      <td>1.282871</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>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>1</td>\n      <td>0</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>3355 rows × 21 columns</p>\n</div>"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 410
    }
   ],
   "source": [
    "train, test = train_test_split(df,test_size=0.2)\n",
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 411,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<AxesSubplot:>"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 411
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "train['stroke'].hist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### stroke = 1 <<<< stroke =0 => need to add"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 412,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<AxesSubplot:title={'center':'count (Stroke)'}>"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 412
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "class_count_1, class_count_2 = train['stroke'].value_counts()\n",
    "#seprate 2 classes of Outcome\n",
    "class_1 = train[train['stroke'] == 0]\n",
    "class_2 = train[train['stroke'] == 1]# print the shape of the class\n",
    "#print('class 1:', class_1.shape)\n",
    "#print('class 2:', class_2.shape)\n",
    "class_2_over = class_2.sample(class_count_1, replace=True)\n",
    "test_over = pd.concat([class_2_over, class_1], axis=0)\n",
    "#print(\"total Outcome of 1 and 2:\",test_over['Outcome'].value_counts())# plot the count after under-sampeling\n",
    "test_over['stroke'].value_counts().plot(kind='bar', title='count (Stroke)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 413,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%% \n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "           age  hypertension  heart_disease  avg_glucose_level       bmi  \\\n201   0.874807             0              0          -0.670214 -1.316507   \n192   1.616307             0              0          -0.227126 -1.274126   \n4024  0.703692             0              0           1.852071  0.111889   \n86    1.730384             0              1          -0.623551 -1.570794   \n38    1.445192             1              0           2.811156  0.435247   \n...        ...           ...            ...                ...       ...   \n2555  0.190346             0              0          -0.475439 -1.245872   \n2540  0.931846             0              0          -0.898319  0.378739   \n2944 -0.323000             0              0          -0.181714  0.053817   \n228   1.160000             0              0           1.866028 -0.454757   \n3670 -0.722270             1              0          -1.010393  1.282871   \n\n      work_type_Govt_job  work_type_Never_worked  work_type_Private  \\\n201                    0                       0                  1   \n192                    0                       0                  1   \n4024                   1                       0                  0   \n86                     0                       0                  0   \n38                     0                       0                  1   \n...                  ...                     ...                ...   \n2555                   0                       0                  1   \n2540                   0                       0                  1   \n2944                   0                       0                  1   \n228                    0                       0                  1   \n3670                   0                       0                  1   \n\n      work_type_Self-employed  smoking_status_Unknown  \\\n201                         0                       0   \n192                         0                       0   \n4024                        0                       0   \n86                          1                       1   \n38                          0                       0   \n...                       ...                     ...   \n2555                        0                       0   \n2540                        0                       0   \n2944                        0                       1   \n228                         0                       0   \n3670                        0                       0   \n\n      smoking_status_formerly smoked  smoking_status_never smoked  \\\n201                                1                            0   \n192                                0                            0   \n4024                               0                            1   \n86                                 0                            0   \n38                                 0                            1   \n...                              ...                          ...   \n2555                               1                            0   \n2540                               1                            0   \n2944                               0                            0   \n228                                1                            0   \n3670                               0                            1   \n\n      smoking_status_smokes  gender_Female  gender_Male  gender_Other  \\\n201                       0              0            1             0   \n192                       1              1            0             0   \n4024                      0              1            0             0   \n86                        0              1            0             0   \n38                        0              0            1             0   \n...                     ...            ...          ...           ...   \n2555                      0              0            1             0   \n2540                      0              0            1             0   \n2944                      0              0            1             0   \n228                       0              0            1             0   \n3670                      0              1            0             0   \n\n      ever_married_No  ever_married_Yes  Residence_type_Rural  \\\n201                 0                 1                     1   \n192                 0                 1                     0   \n4024                0                 1                     0   \n86                  0                 1                     1   \n38                  0                 1                     1   \n...               ...               ...                   ...   \n2555                0                 1                     0   \n2540                0                 1                     1   \n2944                0                 1                     1   \n228                 0                 1                     0   \n3670                0                 1                     1   \n\n      Residence_type_Urban  \n201                      0  \n192                      1  \n4024                     1  \n86                       0  \n38                       0  \n...                    ...  \n2555                     1  \n2540                     0  \n2944                     0  \n228                      1  \n3670                     0  \n\n[6316 rows x 20 columns]",
      "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>age</th>\n      <th>hypertension</th>\n      <th>heart_disease</th>\n      <th>avg_glucose_level</th>\n      <th>bmi</th>\n      <th>work_type_Govt_job</th>\n      <th>work_type_Never_worked</th>\n      <th>work_type_Private</th>\n      <th>work_type_Self-employed</th>\n      <th>smoking_status_Unknown</th>\n      <th>smoking_status_formerly smoked</th>\n      <th>smoking_status_never smoked</th>\n      <th>smoking_status_smokes</th>\n      <th>gender_Female</th>\n      <th>gender_Male</th>\n      <th>gender_Other</th>\n      <th>ever_married_No</th>\n      <th>ever_married_Yes</th>\n      <th>Residence_type_Rural</th>\n      <th>Residence_type_Urban</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>201</th>\n      <td>0.874807</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.670214</td>\n      <td>-1.316507</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</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>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>192</th>\n      <td>1.616307</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.227126</td>\n      <td>-1.274126</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</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>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>4024</th>\n      <td>0.703692</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1.852071</td>\n      <td>0.111889</td>\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>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    </tr>\n    <tr>\n      <th>86</th>\n      <td>1.730384</td>\n      <td>0</td>\n      <td>1</td>\n      <td>-0.623551</td>\n      <td>-1.570794</td>\n      <td>0</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>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>38</th>\n      <td>1.445192</td>\n      <td>1</td>\n      <td>0</td>\n      <td>2.811156</td>\n      <td>0.435247</td>\n      <td>0</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>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    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>2555</th>\n      <td>0.190346</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.475439</td>\n      <td>-1.245872</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</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>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>2540</th>\n      <td>0.931846</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.898319</td>\n      <td>0.378739</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</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>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>2944</th>\n      <td>-0.323000</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.181714</td>\n      <td>0.053817</td>\n      <td>0</td>\n      <td>0</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>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    </tr>\n    <tr>\n      <th>228</th>\n      <td>1.160000</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1.866028</td>\n      <td>-0.454757</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</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>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>3670</th>\n      <td>-0.722270</td>\n      <td>1</td>\n      <td>0</td>\n      <td>-1.010393</td>\n      <td>1.282871</td>\n      <td>0</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>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n  </tbody>\n</table>\n<p>6316 rows × 20 columns</p>\n</div>"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 413
    }
   ],
   "source": [
    "y_train=test_over['stroke']\n",
    "x_train=test_over.drop(columns='stroke')\n",
    "x_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 414,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<AxesSubplot:>"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 414
    },
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD4CAYAAAAAczaOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAAsTAAALEwEAmpwYAAATeUlEQVR4nO3df6zd9X3f8ecrOKQMMiClvWLGm6nqqHOCSugVUHXaLmUFw6Q4VdsIRINJ2Vx1MLUrmuZ0f5CFISXaSKQwSusIC9LROPRHaivQIUa5QpnmBNNQjGGMW3CKPYrXmDi9oWVz9t4f5+vs1LV9j88993tz+3k+pKP7/X6+n+/383nfa17ne77new6pKiRJbXjbck9AktQfQ1+SGmLoS1JDDH1JaoihL0kNWbXcEziZ8847r9auXTv2/t/61rc488wzJzehFaC1mlurF6y5FYup+emnn/6zqvq+4237rg79tWvXsnv37rH3n52dZWZmZnITWgFaq7m1esGaW7GYmpN87UTbvLwjSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kN+a7+RO5i7TlwmJu2PNz7uPs+/k96H1PS0li7DBkCcP+GpfnaCc/0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMWDP0k35PkK0n+KMneJP+2a78wyZeTzCX5fJLTu/Z3dOtz3fa1Q8f6SNf+YpKrl6wqSdJxjXKm/xbw41X1w8DFwIYklwOfAD5VVT8IvAHc3PW/GXija/9U148k64HrgPcAG4BfTXLaBGuRJC1gwdCvgflu9e3do4AfB367a38A+EC3vLFbp9t+ZZJ07dur6q2qegWYAy6dRBGSpNGM9Inc7oz8aeAHgXuAPwa+UVVHui77gdXd8mrgVYCqOpLkMPC9XfuuocMO7zM81mZgM8DU1BSzs7OnVtGQqTPgtouOLNxxwhYz58Wan59f1vH71lq9YM19W44MgaWreaTQr6pvAxcnOQf4AvBDE5/J/x9rK7AVYHp6uhbzP0O++8Ed3LWn/2+a2HfDTO9jHtXa/0C6tXrBmvu2HF/lAoOvYViKmk/p7p2q+gbwBPCjwDlJjibqBcCBbvkAsAag23428PXh9uPsI0nqwSh373xfd4ZPkjOAnwBeYBD+P9112wTs6JZ3dut02/+gqqprv667u+dCYB3wlQnVIUkawSjXPs4HHuiu678NeKiqvpjkeWB7kn8HfBW4r+t/H/AbSeaAQwzu2KGq9iZ5CHgeOALc0l02kiT1ZMHQr6pngfcdp/1ljnP3TVX9JfAzJzjWncCdpz5NSdIk+IlcSWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQxYM/SRrkjyR5Pkke5P8Ytf+0SQHkjzTPa4d2ucjSeaSvJjk6qH2DV3bXJItS1OSJOlEVo3Q5whwW1X9YZJ3Ak8neazb9qmq+g/DnZOsB64D3gP8HeC/JHl3t/ke4CeA/cBTSXZW1fOTKESStLAFQ7+qXgNe65b/PMkLwOqT7LIR2F5VbwGvJJkDLu22zVXVywBJtnd9DX1J6kmqavTOyVrgSeC9wC8DNwHfBHYzeDXwRpL/COyqqv/U7XMf8PvdITZU1T/t2j8EXFZVtx4zxmZgM8DU1NSPbN++feziDh46zOt/MfbuY7to9dn9D9qZn5/nrLPOWrbx+9ZavWDNfdtz4PCyjHvh2aeNXfMVV1zxdFVNH2/bKJd3AEhyFvA7wC9V1TeT3AvcAVT38y7g58aa4ZCq2gpsBZienq6ZmZmxj3X3gzu4a8/IJU7Mvhtmeh/zqNnZWRbzO1tpWqsXrLlvN215eFnGvX/DmUtS80iJmOTtDAL/war6XYCqen1o+2eAL3arB4A1Q7tf0LVxknZJUg9GuXsnwH3AC1X1yaH284e6/STwXLe8E7guyTuSXAisA74CPAWsS3JhktMZvNm7czJlSJJGMcqZ/o8BHwL2JHmma/sV4PokFzO4vLMP+HmAqtqb5CEGb9AeAW6pqm8DJLkVeBQ4DdhWVXsnVokkaUGj3L3zJSDH2fTISfa5E7jzOO2PnGw/SdLS8hO5ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDVkwdBPsibJE0meT7I3yS927e9K8liSl7qf53btSfLpJHNJnk1yydCxNnX9X0qyaenKkiQdzyhn+keA26pqPXA5cEuS9cAW4PGqWgc83q0DXAOs6x6bgXth8CQB3A5cBlwK3H70iUKS1I8FQ7+qXquqP+yW/xx4AVgNbAQe6Lo9AHygW94IfLYGdgHnJDkfuBp4rKoOVdUbwGPAhkkWI0k6uVO6pp9kLfA+4MvAVFW91m36U2CqW14NvDq02/6u7UTtkqSerBq1Y5KzgN8BfqmqvpnkO9uqqpLUJCaUZDODy0JMTU0xOzs79rGmzoDbLjoyiWmdksXMebHm5+eXdfy+tVYvWHPfliNDYOlqHin0k7ydQeA/WFW/2zW/nuT8qnqtu3xzsGs/AKwZ2v2Cru0AMHNM++yxY1XVVmArwPT0dM3MzBzbZWR3P7iDu/aM/Lw2MftumOl9zKNmZ2dZzO9spWmtXrDmvt205eFlGff+DWcuSc2j3L0T4D7ghar65NCmncDRO3A2ATuG2m/s7uK5HDjcXQZ6FLgqybndG7hXdW2SpJ6Mchr8Y8CHgD1JnunafgX4OPBQkpuBrwEf7LY9AlwLzAFvAh8GqKpDSe4Anur6fayqDk2iCEnSaBYM/ar6EpATbL7yOP0LuOUEx9oGbDuVCUqSJsdP5EpSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhqyYOgn2ZbkYJLnhto+muRAkme6x7VD2z6SZC7Ji0muHmrf0LXNJdky+VIkSQsZ5Uz/fmDDcdo/VVUXd49HAJKsB64D3tPt86tJTktyGnAPcA2wHri+6ytJ6tGqhTpU1ZNJ1o54vI3A9qp6C3glyRxwabdtrqpeBkiyvev7/KlPWZI0rgVD/yRuTXIjsBu4rareAFYDu4b67O/aAF49pv2y4x00yWZgM8DU1BSzs7NjT3DqDLjtoiNj7z+uxcx5sebn55d1/L61Vi9Yc9+WI0Ng6WoeN/TvBe4Aqvt5F/Bzk5hQVW0FtgJMT0/XzMzM2Me6+8Ed3LVnMc9r49l3w0zvYx41OzvLYn5nK01r9YI19+2mLQ8vy7j3bzhzSWoeKxGr6vWjy0k+A3yxWz0ArBnqekHXxknaJUk9GeuWzSTnD63+JHD0zp6dwHVJ3pHkQmAd8BXgKWBdkguTnM7gzd6d409bkjSOBc/0k3wOmAHOS7IfuB2YSXIxg8s7+4CfB6iqvUkeYvAG7RHglqr6dnecW4FHgdOAbVW1d9LFSJJObpS7d64/TvN9J+l/J3DncdofAR45pdlJkibKT+RKUkMMfUlqiKEvSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IasmDoJ9mW5GCS54ba3pXksSQvdT/P7dqT5NNJ5pI8m+SSoX02df1fSrJpacqRJJ3MKGf69wMbjmnbAjxeVeuAx7t1gGuAdd1jM3AvDJ4kgNuBy4BLgduPPlFIkvqzYOhX1ZPAoWOaNwIPdMsPAB8Yav9sDewCzklyPnA18FhVHaqqN4DH+OtPJJKkJbZqzP2mquq1bvlPgalueTXw6lC//V3bidr/miSbGbxKYGpqitnZ2TGnCFNnwG0XHRl7/3EtZs6LNT8/v6zj9621esGa+7YcGQJLV/O4of8dVVVJahKT6Y63FdgKMD09XTMzM2Mf6+4Hd3DXnkWXeMr23TDT+5hHzc7Ospjf2UrTWr1gzX27acvDyzLu/RvOXJKax7175/Xusg3dz4Nd+wFgzVC/C7q2E7VLkno0bujvBI7egbMJ2DHUfmN3F8/lwOHuMtCjwFVJzu3ewL2qa5Mk9WjBax9JPgfMAOcl2c/gLpyPAw8luRn4GvDBrvsjwLXAHPAm8GGAqjqU5A7gqa7fx6rq2DeHJUlLbMHQr6rrT7DpyuP0LeCWExxnG7DtlGYnSZooP5ErSQ0x9CWpIYa+JDXE0Jekhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqyKJCP8m+JHuSPJNkd9f2riSPJXmp+3lu154kn04yl+TZJJdMogBJ0ugmcaZ/RVVdXFXT3foW4PGqWgc83q0DXAOs6x6bgXsnMLYk6RQsxeWdjcAD3fIDwAeG2j9bA7uAc5KcvwTjS5JOIFU1/s7JK8AbQAG/XlVbk3yjqs7ptgd4o6rOSfJF4ONV9aVu2+PAv66q3cccczODVwJMTU39yPbt28ee38FDh3n9L8befWwXrT67/0E78/PznHXWWcs2ft9aqxesuW97DhxelnEvPPu0sWu+4oornh66+vJXrFrUrOAfVNWBJN8PPJbkvw9vrKpKckrPKlW1FdgKMD09XTMzM2NP7u4Hd3DXnsWWeOr23TDT+5hHzc7Ospjf2UrTWr1gzX27acvDyzLu/RvOXJKaF3V5p6oOdD8PAl8ALgVeP3rZpvt5sOt+AFgztPsFXZskqSdjh36SM5O88+gycBXwHLAT2NR12wTs6JZ3Ajd2d/FcDhyuqtfGnrkk6ZQt5trHFPCFwWV7VgG/WVX/OclTwENJbga+Bnyw6/8IcC0wB7wJfHgRY0uSxjB26FfVy8APH6f968CVx2kv4JZxx5MkLZ6fyJWkhhj6ktQQQ1+SGmLoS1JDDH1JaoihL0kNMfQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQwx9SWqIoS9JDTH0Jakhhr4kNcTQl6SGGPqS1BBDX5IaYuhLUkMMfUlqiKEvSQ0x9CWpIYa+JDWk99BPsiHJi0nmkmzpe3xJalmvoZ/kNOAe4BpgPXB9kvV9zkGSWtb3mf6lwFxVvVxV/xvYDmzseQ6S1KxVPY+3Gnh1aH0/cNlwhySbgc3d6nySFxcx3nnAny1i/7HkE32P+FcsS83LqLV6wZqbcMUnFlXz3zvRhr5Df0FVtRXYOoljJdldVdOTONZK0VrNrdUL1tyKpaq578s7B4A1Q+sXdG2SpB70HfpPAeuSXJjkdOA6YGfPc5CkZvV6eaeqjiS5FXgUOA3YVlV7l3DIiVwmWmFaq7m1esGaW7EkNaeqluK4kqTvQn4iV5IaYuhLUkNWfOgv9LUOSd6R5PPd9i8nWbsM05yoEWr+5STPJ3k2yeNJTnjP7kox6td3JPmpJJVkxd/eN0rNST7Y/a33JvnNvuc4aSP82/67SZ5I8tXu3/e1yzHPSUmyLcnBJM+dYHuSfLr7fTyb5JJFD1pVK/bB4M3gPwZ+ADgd+CNg/TF9/jnwa93ydcDnl3vePdR8BfC3uuVfaKHmrt87gSeBXcD0cs+7h7/zOuCrwLnd+vcv97x7qHkr8Avd8npg33LPe5E1/0PgEuC5E2y/Fvh9IMDlwJcXO+ZKP9Mf5WsdNgIPdMu/DVyZJD3OcdIWrLmqnqiqN7vVXQw+D7GSjfr1HXcAnwD+ss/JLZFRav5nwD1V9QZAVR3seY6TNkrNBfztbvls4H/2OL+Jq6ongUMn6bIR+GwN7ALOSXL+YsZc6aF/vK91WH2iPlV1BDgMfG8vs1sao9Q87GYGZwor2YI1dy9711TVw31ObAmN8nd+N/DuJP81ya4kG3qb3dIYpeaPAj+bZD/wCPAv+pnasjnV/94X9F33NQyanCQ/C0wD/2i557KUkrwN+CRw0zJPpW+rGFzimWHwau7JJBdV1TeWc1JL7Hrg/qq6K8mPAr+R5L1V9X+Xe2IrxUo/0x/lax2+0yfJKgYvCb/ey+yWxkhfZZHkHwP/Bnh/Vb3V09yWykI1vxN4LzCbZB+Da587V/ibuaP8nfcDO6vq/1TVK8D/YPAksFKNUvPNwEMAVfXfgO9h8GVsf1NN/KtrVnroj/K1DjuBTd3yTwN/UN07JCvUgjUneR/w6wwCf6Vf54UFaq6qw1V1XlWtraq1DN7HeH9V7V6e6U7EKP+2f4/BWT5JzmNwueflHuc4aaPU/CfAlQBJ/j6D0P9fvc6yXzuBG7u7eC4HDlfVa4s54Iq+vFMn+FqHJB8DdlfVTuA+Bi8B5xi8YXLd8s148Uas+d8DZwG/1b1n/SdV9f5lm/QijVjz3ygj1vwocFWS54FvA/+qqlbsq9gRa74N+EySf8ngTd2bVvJJXJLPMXjiPq97n+J24O0AVfVrDN63uBaYA94EPrzoMVfw70uSdIpW+uUdSdIpMPQlqSGGviQ1xNCXpIYY+pLUEENfkhpi6EtSQ/4fJ5MTcYVZdVAAAAAASUVORK5CYII=\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_train.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 415,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "           age  hypertension  heart_disease  avg_glucose_level       bmi  \\\n190   1.673346             1              0          -0.887487  0.689534   \n3402 -1.178577             0              0           0.092638  1.593666   \n4130  1.274076             1              0           2.274952  0.649279   \n854  -1.178577             0              0           0.039934  0.053817   \n1972  1.559269             0              0           0.369906 -1.302380   \n...        ...           ...            ...                ...       ...   \n2938  0.760730             0              0          -0.386696 -0.271105   \n1610  1.045923             1              0           2.511807 -0.059200   \n1154 -0.437077             0              0          -0.556057 -0.398249   \n510  -1.007462             0              0           2.319323  1.282871   \n3987  1.730384             0              0           0.555932  0.746043   \n\n      work_type_Govt_job  work_type_Never_worked  work_type_Private  \\\n190                    0                       0                  1   \n3402                   0                       0                  1   \n4130                   0                       0                  1   \n854                    1                       0                  0   \n1972                   0                       0                  0   \n...                  ...                     ...                ...   \n2938                   0                       0                  1   \n1610                   0                       0                  1   \n1154                   0                       0                  1   \n510                    0                       0                  1   \n3987                   0                       0                  1   \n\n      work_type_Self-employed  smoking_status_Unknown  \\\n190                         0                       0   \n3402                        0                       1   \n4130                        0                       0   \n854                         0                       0   \n1972                        1                       0   \n...                       ...                     ...   \n2938                        0                       0   \n1610                        0                       0   \n1154                        0                       0   \n510                         0                       0   \n3987                        0                       1   \n\n      smoking_status_formerly smoked  smoking_status_never smoked  \\\n190                                0                            1   \n3402                               0                            0   \n4130                               0                            1   \n854                                0                            0   \n1972                               0                            0   \n...                              ...                          ...   \n2938                               0                            1   \n1610                               0                            0   \n1154                               0                            0   \n510                                0                            1   \n3987                               0                            0   \n\n      smoking_status_smokes  gender_Female  gender_Male  gender_Other  \\\n190                       0              1            0             0   \n3402                      0              0            1             0   \n4130                      0              1            0             0   \n854                       1              1            0             0   \n1972                      1              0            1             0   \n...                     ...            ...          ...           ...   \n2938                      0              0            1             0   \n1610                      1              0            1             0   \n1154                      1              0            1             0   \n510                       0              0            1             0   \n3987                      0              0            1             0   \n\n      ever_married_No  ever_married_Yes  Residence_type_Rural  \\\n190                 1                 0                     0   \n3402                1                 0                     0   \n4130                0                 1                     1   \n854                 0                 1                     1   \n1972                0                 1                     1   \n...               ...               ...                   ...   \n2938                0                 1                     1   \n1610                0                 1                     0   \n1154                0                 1                     0   \n510                 0                 1                     1   \n3987                0                 1                     1   \n\n      Residence_type_Urban  \n190                      1  \n3402                     1  \n4130                     0  \n854                      0  \n1972                     0  \n...                    ...  \n2938                     0  \n1610                     1  \n1154                     1  \n510                      0  \n3987                     0  \n\n[839 rows x 20 columns]",
      "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>age</th>\n      <th>hypertension</th>\n      <th>heart_disease</th>\n      <th>avg_glucose_level</th>\n      <th>bmi</th>\n      <th>work_type_Govt_job</th>\n      <th>work_type_Never_worked</th>\n      <th>work_type_Private</th>\n      <th>work_type_Self-employed</th>\n      <th>smoking_status_Unknown</th>\n      <th>smoking_status_formerly smoked</th>\n      <th>smoking_status_never smoked</th>\n      <th>smoking_status_smokes</th>\n      <th>gender_Female</th>\n      <th>gender_Male</th>\n      <th>gender_Other</th>\n      <th>ever_married_No</th>\n      <th>ever_married_Yes</th>\n      <th>Residence_type_Rural</th>\n      <th>Residence_type_Urban</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>190</th>\n      <td>1.673346</td>\n      <td>1</td>\n      <td>0</td>\n      <td>-0.887487</td>\n      <td>0.689534</td>\n      <td>0</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>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>3402</th>\n      <td>-1.178577</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.092638</td>\n      <td>1.593666</td>\n      <td>0</td>\n      <td>0</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>0</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    </tr>\n    <tr>\n      <th>4130</th>\n      <td>1.274076</td>\n      <td>1</td>\n      <td>0</td>\n      <td>2.274952</td>\n      <td>0.649279</td>\n      <td>0</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>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>854</th>\n      <td>-1.178577</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.039934</td>\n      <td>0.053817</td>\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>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>1</td>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>1972</th>\n      <td>1.559269</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.369906</td>\n      <td>-1.302380</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>0</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    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>2938</th>\n      <td>0.760730</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.386696</td>\n      <td>-0.271105</td>\n      <td>0</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>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    </tr>\n    <tr>\n      <th>1610</th>\n      <td>1.045923</td>\n      <td>1</td>\n      <td>0</td>\n      <td>2.511807</td>\n      <td>-0.059200</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</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>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>1154</th>\n      <td>-0.437077</td>\n      <td>0</td>\n      <td>0</td>\n      <td>-0.556057</td>\n      <td>-0.398249</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</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>1</td>\n      <td>0</td>\n      <td>0</td>\n      <td>1</td>\n      <td>0</td>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>510</th>\n      <td>-1.007462</td>\n      <td>0</td>\n      <td>0</td>\n      <td>2.319323</td>\n      <td>1.282871</td>\n      <td>0</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>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    </tr>\n    <tr>\n      <th>3987</th>\n      <td>1.730384</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.555932</td>\n      <td>0.746043</td>\n      <td>0</td>\n      <td>0</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>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    </tr>\n  </tbody>\n</table>\n<p>839 rows × 20 columns</p>\n</div>"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 415
    }
   ],
   "source": [
    "y_test=test['stroke']\n",
    "x_test=test.drop(columns='stroke')\n",
    "x_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 416,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEWCAYAAABhffzLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAAsTAAALEwEAmpwYAAAl4UlEQVR4nO3de7xVc/7H8ddHdXRPVFRHpaTQTKlDuYuILiRKbiWqMTMujxkG85th24aZMYwZzDAulYZQKiKkMsilpNyF0lEqqpNO6X7qnM/vj71PTjmxT+esvfbl/Xw8enT22nvt9dmH3nut73d91jJ3R0REssdeYRcgIiLJpeAXEckyCn4RkSyj4BcRyTIKfhGRLKPgFxHJMgp+yWpmvGjGkARet8GM1smoSSRopvP4JdWZsRjYH9gOFAPzgf8CD7pTEmJplWLGhjIPawNbiX0+gF+4Mzb5VUk2qB52ASIJ6uvODDMaACcCdwNdgaHhlrXn3Klb+nP8y22YOzN2fZ0Z1d3ZnszaJLNpqEfSijvr3HkWOA8YYkYHADP2NuNOM74yY6UZ/zGjVul6ZpxlxvtmfGfGIjNOjy9/1Yxh8Z8PNuM1M9aZsdqMcWXWdzMOjv/cwIz/mlFgxhIz/mgW+7dkxiVmvBGvpdCML804oyKf0YyTzFhmxvVmrABGm7GXGTfEa//WjPFm7FtmnW5mvGXGWjM+MOOkPf0dS+ZT8EtacmcOsAw4Pr7or8AhQCfgYKA5cBOAGUcRGxr6HbAPcAKwuJy3/RMwDWgI5AL37mbz9wINgNbEjj4Gs/ORR1fgc6AR8DdgpBlWwY94ALAv0BIYAVwJ9ItvrxlQCPw7/vmaA88Dt8bXuRaYaEbjCm5TsoSCX9LZ18C+8VAdAfzGnTXurAf+DAyKv+4yYJQ7090pcWe5O5+V837biAVtM3e2uPPGri8wo1r8fX/vznp3FgN/By4u87Il7jzkTjEwBmhKbI6iIkqAiDtb3dkMXA78wZ1l7mwFbgbONaM6cBHwgjsvxD/fdGAu0KuC25QsoeCXdNYcWAM0JjY5Oi8+1LEWmBpfDnAgsCiB97sOMGCOGZ+YcWk5r2kE1ACWlFm2JF5LqRWlP7izKf5jXSqmwJ0tZR63BJ4u8/k+JTYRvH/8uQGlz8WfP47YF47ID2hyV9KSGUcSC9s3gNXAZuBwd5aX8/KlQJufek93VgDD4+9/HDDDjJnufFHmZav5/shgfnxZCyh3u5Wx6+l2S4FL3Xlz1xeasRR41D1Wu8hP0R6/pBUz6pvRB3gSeMydj+KndD4E/MOMJvHXNTejZ3y1kcBQM06JT5I2N6N9Oe89wIzc+MNCYuG70+mi8eGb8cBtZtQzoyXwW+CxAD5uWf+Jb7NlvNbGZpwVf+4xoK8ZPc2oZkbN+ARx7m7fTbKagl/SxXNmrCe25/sH4C52nlC9HvgCmG3Gd8AMoB3smAgeCvwDWAe8RmyPfVdHAm/Hz69/FrjanfxyXnclsBHIJ3bE8TgwqrIf8CfcHa9pWvz3MJvYJDLuLAXOAv4PKCD2O/od+vctu6EGLhGRLKM9AhGRLKPgFxHJMgp+EZEso+AXEckyaXEef6NGjbxVq1ZhlyEiklbmzZu32t1/cOmOtAj+Vq1aMXfu3LDLEBFJK2a2pLzlGuoREckyCn4RkSyj4BcRyTIKfhGRLKPgFxHJMgp+EZEso+AXEUlR24q3BfK+Cn4RkRTj7kyYP4H2/27PjPwZVf7+Cn4RkRQya+ksjh11LAOeGkB+YT4PvftQlW8jLTp3RUQy3aI1i/j9y7/nqflPAdCkThNuOekWLut8WZVvS8EvIhKiNZvXcOvMW/nXnH+xrWQbNavX5Jqjr+G6Y6+j/t71A9mmgl9EJARbt2/lvnfu408z/0ThlkIMY0jHIdx68q3k1g/2dskKfhGRJHJ3Jn46ketnXE9+YeyWzicfdDJ3nnonRzQ9Iik1BBr8FrXfAMMABz4idsPrpsCTwH7APOBij3hRkHWIiKSCWUtncc20a5i1bBYAhzY6lDtOvYNebXthZkmrI7CzeixqzYGrgDyPeAegGjAIuB34h0f8YKAQqPqZCxGRFJJfmM95E87jmFHHMGvZLJrUacJ/ev+HD3/5Ib0P6Z3U0Ifgh3qqA7UsatuA2sA3wMnABfHnxwA3A/cHXIeISNKt2byG22bexr1z7k3axG0iAgt+j/hyi9qdwFfAZmAasaGdtR7x7fGXLQOal7e+mY0ARgC0aNEiqDJFRKpcUXER/57z750mbgd3HMxtJ98W+MRtIgILfotaQ+As4CBgLfAUcHqi67v7g8CDAHl5eR5AiSIiVap04vaGGTewqHARAN1bdefO0+6kc9POIVf3vSCHenoAX3rECwAsapOAY4F9LGrV43v9ucDyAGsQEUmKVJm4TUSQwf8V0M2iVpvYUM8pwFzgFeBcYmf2DAEmB1iDiEig8gvz+f3Lv2f8J+OBnTtuq++VmmfMBznG/7ZFbQLwLrAdeI/Y0M3zwJMWtVvjy0YGVYOISFAKNxdy68xbU27iNhHmnvrD53l5eT537tywyxARoai4iPveuY9bXrtlx8TtxR0v5tbut3JggwPDLm8nZjbP3fN2XZ6axyEiIikmXSZuE6HgFxH5CbOXzeaaadfw1tK3gNSeuE2Egl9EZDd2nbhtXLsxt3S/hWGdh6XsxG0i0rdyEZGAFG4u5LbXYx23RcVFaTVxmwgFv4hIXHkTt4M7Dk7JidvKUPCLSNZzdyZ9OonrZ1yf9hO3iVDwi0hWyy/MZ/DTg3lz6ZsAtG/UnjtOvYPebZN/1cxkUfCLSNbaULSBvk/0ZX7B/IyZuE1EZn86EZHdcHeGPTuM+QXzObTRobx12VvsU3OfsMtKisBuxCIiksr+OfufjPtkHPVy6jHpvElZE/qg4BeRLDRzyUx+N/13ADzS7xHaN2ofckXJpeAXkazy9fqvGfjUQIq9mOuOuY7+h/YPu6SkU/CLSNYoKi5iwFMDWLlxJScfdDK3nXJb2CWFQsEvIlnj2mnX8tbSt8itn8sT5zyR8Wfv7I6CX0SywtgPx3LvnHvJqZbDhAETaFKnSdglhUbBLyIZ78OVHzL8ueEA3HP6PXTN7RpyReFS8ItIRlu7ZS39x/Vn8/bNXNLpEkZ0GRF2SaFT8ItIxirxEgY/PZhFhYs44oAjuK/XfRl7GYaKUPCLSMb6y+t/4bkFz9GwZkMmDpxIrRq1wi4pJSj4RSQjvfTFS9z4yo0Yxtj+Yzmo4UFhl5QysvNcJhHJaIvXLuaCSRfgONGTopzR9oywS0op2uMXkYyyZfsWzhl/Dms2r6F329788YQ/hl1SylHwi0hGueKFK3j3m3dp3bA1j579KHuZYm5X+o2ISMZ4aN5DjHxvJDWr12TSwEk0rNUw7JJSkoJfUt6M/BlcN/06Vm9aHXYpksLeWf4OV7x4BQAP9nmQjgd0DLmi1KXgl5Tl7vz9rb9z2qOnccdbd9Dt4W4s+HZB2GVJClq9aTXnjD+HouIifn3kr7m448Vhl5TSFPySkoqKi7js2cu4dvq1OE7LBi1ZVLiIo0cezetLXg+7PEkhxSXFnD/xfJZ+t5Ruud24q+ddYZeU8hT8knIKNhbQ4789GP3+aGpVr8X4c8fz8a8+ps8hfVizeQ09Hu3B2A/Hhl2mpIgbX7mRGfkzaFKnCRMGTCCnWk7YJaU8Bb+klI9XfcxRDx/F61+9TrN6zXh96OsMOHwAdXPq8sx5z3DlUVdSVFzERU9fxC2v3YK7h12yhOiZz57hL2/8hWpWjXHnjqN5/eZhl5QWFPySMp5f8DzHjDyGxWsXk9csj3eGv0OXZl12PF9tr2rcc8Y93H363RhG5NUIQycPpai4KMSqJSwLvl3AkGeGAHB7j9s5qdVJ4RaURhT8Ejp3565Zd9H3ib6sL1rPwMMH8tolr9GsXrNyX39V16t4ZtAz1K5RmzEfjKHnYz0p3FyY5KolTBuKNtB/XH++2/odAw4bwG+P/m3YJaUVBb+Eqqi4iGHPDuOaadfgODefeDNPnvMktWvU/tH1zmx3JjMvmckBdQ/g1cWvcsyoY8gvzE9S1RImd2fYs8P4pOATDm10KCPPHKkrblaQgl9CUzqJO+r9UTsmcSMnRRL+R9ylWRfeHvY2HZp04LPVn9Ht4W7MXjY74KolbPe8fQ/jPhlH3Zy6TDpvEvX2rhd2SWlHwS+h2HUSd+bQmQw4fECF36dFgxa8eembnNbmNAo2FdB9THee+uSpACqWVPD6kte5dvq1ADxy1iO0b9Q+5IrSk4Jfkm7XSdw5w+aQ1yxvj9+v/t71mXL+FIZ3Hs6W7VsYOGEgf3vzbzrjJ8N8s/4bBk4YyPaS7fzumN9xzmHnhF1S2lLwS9LsbhK3Kk7Bq1GtBg/0eYDbe9wOwPUzrufyKZezrXhbpd9bwreteBsDnhrAig0r6N6qO38+5c9hl5TWArsev0WtHTCuzKLWwE3APsBwoCC+/P884i8EVYekhqLiIn455ZeMen8UADefeDM3nXhTlU7KmRnXHXsdrRu25uKnL+bBdx9k8brFjD93PA1qNqiy7UjyXTvtWt5c+ibN6zXnyXOfpPpeupVIZVgyDoctatWA5UBXYCiwwSN+Z6Lr5+Xl+dy5c4MqTwJWeh2VmUtmUqt6Lcb0G7NH4/kVMXvZbM584kwKNhXQoUkHnr/geVo0aBHoNiUYj3/0OBdOupAae9Vg5tCZdMvtFnZJacPM5rn7D8ZRkzXUcwqwyCO+JEnbkxTxyapPOOqho5i5ZGalJnErqltuN2YPm037Ru35eNXHdH24K/O+nhf4dqVqfbTyI4Y/NxyAe864R6FfRZIV/IOAJ8o8vsKi9qFFbZRFrdwLZpvZCDOba2ZzCwoKynuJpLgXFr7A0SOP5su1X1bJJG5FtW7YmrcufYvurbqzYsMKTnjkBJ79/NmkbV8qZ92WdfQf359N2zYxpOMQftHlF2GXlDECD36LWg5wJlB6jt39QBugE/AN8Pfy1nP3B909z93zGjduHHSZUoXcnX/M+kcgk7gV1bBWQ6ZeNJUhHYewadsm+j3Zj3vevifpdUjFlHgJg58ZzBdrvqDTAZ24v/f9atKqQsnY4z8DeNcjvhLAI77SI17sES8BHgKOSkINkiRFxUUMf244v532W0q8JOFO3CDlVMth9FmjueWkW3Ccq6dezVUvXkVxSXFoNcmP++sbf+XZz5+lYc2GTBw4kVo1aoVdUkZJRvCfT5lhHota0zLPnQ18nIQaJAlWb1rNqY+eysj3Ru5RJ26QzIwbT7yRx85+jJxqOdw75176jevHhqINYZcmu5i2aBp//N8fMYyx/cfSumHrsEvKOIEGv0WtDnAqMKnM4r9Z1D6yqH0IdAd+E2QNkhxhTeJW1IU/v5DpF09n31r7MmXBFE4YfQJfr/867LIkbsnaJVww8QIcJ3JihDPanhF2SRkpKadzVpZO50xtLyx8gUETBrG+aD15zfJ45rxnUv666Au+XUDvx3vzxZovyK2fy/MXPM/P9/952GVltS3bt3DcqOOY9808erXtxXPnP8deph7Tygj7dE7JQKk0iVtRh+x3CLMum8WxBx7Lsu+Wcdyo43jpi5fCLiurXfnClcz7Zh6tG7bmsbMfU+gHSL9Z2SOpOIlbUY1qN2LG4BkM6hA7Wun9eG8emPtA2GVVyrbibby+5HUemPsAby19iy3bt4RdUkIefvdhHn7vYWpWr8nEgRNpWKvcs7yliqjvWSqsbCduzeo1GdNvDAMPHxh2WXukZvWajO0/ljYN23Db67dx+fOX88WaL7j91NvTZo/zy8IveWnRS7y06CX+9+X/+G7rdzueq7FXDTo37Uy33G4cnXs0Rx94NAfWPzAlJtxLzf16Lle8cAUAD/R5gE4HdAq3oCygMX6pkPkF8+n7RF/yC/NpWrcpkwdN5sjmR4ZdVpUY/d5oRkwZwfaS7Zxz6Dk8evajKXka4caijby6+FWmfjGVlxa9xMI1C3d6vn2j9hxxwBF8tOojPln1Cc7O/8ab1WsW+xKIfxF0btqZmtVrJvMj7LB602q6PNiFr9Z9xa/yfsW/e/87lDoy1e7G+BX8krAXF77IoImD+G7rd3Rp2oXJgyanxXh+Rbyc/zLnjD+HdVvX0bV5VyYPmsz+dfcPtSZ358OVH+7Yq3/jqzd2us9wg70bcErrU+jZpic92/Sk5T4tdzy3bss65iyfw6xls5i1bBazl81m7Za1O71/WEcFxSXFnDH2DKbnT6dbbjdeu+Q1cqrlBLrNbKPglz3m7vxz9j+5dvq1lHgJAw8fyOizRqfVeH5FzC+YT6+xvViybgmt9mnFCxe8wKGND01qDas3rWb6ouk7wn7FhhU7njOMI5sfuSPou+Z2TfhqlSVewuerP499ESyNfRnML5gfylHBH17+A39+4880qdOEeSPmkVs/t0rfXxT8soeKiov41fO/YuR7I4FgLqecilZuWMmZT57JnOVzaLB3AyadN4mTDzo5sO1tK97G7GWzdwT9vK/n7RTGTes2pefBPTm9zen0aN2D/WrvV2XbrshRQekXQbfcbpU6Kpj82WT6jetHNavGjMEzOKnVSZX/IPIDCn6psEyaxN0Tm7Zt4uKnL2bSp5Oovld1Hur7EJd0uqTK3n/x2sW89EUs6F/+8uWdJmVzquVwQssTduzVd2jSIWlftkEfFSz8diF5D+Xx3dbvuOPUO7j2mGuD+ihZT8EvFZLJk7gVUeIlXD/9eu6cFbt9xB+P/yO3dL9lj0K4dFK2dK9+wbcLdnq+3X7tYkF/cE9ObHkidXLqVMlnqArrtqzj7eVvM3vZ7N0eFeRUy+GIA47Y8UVwdO7RHNjgwJ1es7FoI91GduPjVR9z7mHnMv7c8Rl/9BgmBb8kLBsmcSvq/nfu54oXr6DES7jgZxcw6sxR7F197x9dx935aNVHvPTFS0xdNPUHk7L1965Pj9Y9yp2UTXWJHhU0r9d8p0njf835F098/ATtG7VnzrA51Nu7XkifIDso+OUn7TqJO+CwATzS75GMncStqBcXvsjACQPZULSB41ocxzPnPfODsfayk7LTFk3jmw3f7HjOMPKa5XH6wadXeFI2HZQeFcxaOovZy2eXe1QAUDenLnOGzUn6hHk2UvDLT7p15q3c+MqNAEROjHDTiTelTRNTsnyw4gN6P96b5euX03bftkweNJlvN3+7Y69+d5OyPdv0pEfrHjSq3SjE6pOrvKOCJeuW8N9+/+XsQ88Ou7ysoOCXH1XiJTT9e1NWbVzFo2c/ykU/vyjsklLW8u+W0/eJvry34r0fPJdTLYfjWxy/Y68+mZOyIrvaXfBnznGmVMqc5XNYtXEVrfZpxYU/uzDsclJa8/rNmTl0JhdMvIDnFjyX0pOyIuVR8AsAz33+HAB9D+mrPdQE1M2py+RBkyncUsi+tfYNuxyRCtEArgAwZeEUAPoc0ifkStKHmSn0JS0p+IUla5fw4coPqZtTlxNbnhh2OSISMAW/MGVBbG+/Z5ueP3luuoikPwW/8NyC78f3RSTzKfiz3Pqt63ll8SsYphtbi2SJhILfjElm9DbTF0WmmZE/g6LiIrrldqNJnSZhlyMiSZBokN8HXAAsNOOvZrQLsCZJIg3ziGSfhILfnRnuXAh0BhYDM8x4y4yhZtQIskAJTomX8PzC5wHo207BL5ItEh66MWM/4BJgGPAecDexL4LpgVQmgSvbrXt448PDLkdEkiShzl0zngbaAY8Cfd0pveTgODN0EZ00VXoaZ5+2fdStK5JFEr1kwz3uvFLeE+784AJAkh52jO9rmEckqyQ61HOYGfuUPjCjoRm/CqYkSQZ164pkr0SDf7g7a0sfuFMIDA+kIkkKdeuKZK9Eg7+aGTsGgc2oBuQEU5IkQ+kwjy7KJpJ9Eh3jn0psIveB+ONfxJdJGtpQtGFHt26vtr3CLkdEkizR4L+eWNj/Mv54OvBwIBVJ4KYvmk5RcRFH5x6tbl2RLJRQ8LtTAtwf/yNpTt26Itkt0fP42wJ/AQ4DapYud6d1QHVJQNStKyKJTu6OJra3vx3oDvwXeCyooiQ47yx/h1UbV9GyQUt164pkqUSDv5Y7LwPmzhJ3bgZ6B1eWBKXsMI+6dUWyU6KTu1vjl2ReaMYVwHKg7o+tYFFrB4wrs6g1cBOxo4VxQCtiF3wb6BEvrFjZsqfUrSsiie7xXw3UBq4CugAXAUN+bAWP+Oce8U4e8U7xdTYBTwM3AC97xNsCL8cfSxKoW1dEIIHgjzdrnefOBneWuTPUnXPcmV2B7ZwCLPKILwHOAsbEl48B+lW0aNkzpd26p7U5Td26IlnsJ4PfnWLguEpuZxDwRPzn/T3ipVf3XAHsX94KZjbCzOaa2dyCgoJKbl4ApiyMBb9O4xTJbokO9bxnxrNmXGxG/9I/iaxoUcsBzgSe2vU5j7gDXt567v6gu+e5e17jxo0TLFN2Z0PRBv735f/UrSsiCQd/TeBb4GSgb/xPohd5OQN41yO+Mv54pUWtKUD871WJlyt7qrRbV/fWFZFEO3eHVmIb5/P9MA/As8Qmhv8a/3tyJd5bEqRuXREplWjn7mjKGZJx59IfXS9qdYBTiV3np9RfgfEWtcuAJcDAhKuVPVK2W1dX4xSRRM/jn1Lm55rA2cDXP7WSR3wjsN8uy74ldpaPJEnZbt0OTTqEXY6IhCzRoZ6JZR+b8QTwRiAVSZVTt66IlJXo5O6u2gKaIUwT6tYVkbISHeNfz85j/CuIXaNfUpy6dUVkV4kO9dQLuhAJRumkrrp1RaRUQkM9ZpxtRoMyj/cx06UW0oFO4xSRXSU6xh9xZ13pA3fWApFAKpIqo25dESlPosFf3usSPRVUQqJuXREpT6LBP9eMu8xoE/9zFzAvyMKk8kqHedS0JSJlJRr8VwJFxG6g8iSwBfh1UEVJ5e10b12N74tIGYme1bMR3TAlrahbV0R2J9GzeqabsU+Zxw3NeCmwqqTS1K0rIruT6FBPo/iZPAC4U4g6d1OaxvdFZHcSDf4SM1qUPjCjFbu5gYqE76t1X+3o1j2p1UlhlyMiKSbRUzL/ALxhxmuAAccDIwKrSipF99YVkR+T6OTuVDPyiIX9e8AzwOYA65JKULeuiPyYRC/SNgy4GsgF3ge6AbOI3YpRUoi6dUXkpyQ6xn81cCSwxJ3uwBHw/WSvpI7Sbt2uuV3VrSsi5Uo0+Le4swXAjL3d+QxoF1xZsqdKx/c1zCMiu5Po5O6y+Hn8zwDTzSgkdr9cSSHq1hWRRCQ6uXt2/MebzXgFaABMDawq2SPvLH+HlRtXqltXRH5Uha+w6c5rQRQiladuXRFJxJ7ec1dSkLp1RSQRCv4MoW5dEUmUgj9DqFtXRBKl4M8Q6tYVkUQp+DOAunVFpCIU/BlgRv4MdeuKSMIU/Bnguc81zCMiiVPwpzl164pIRSn405y6dUWkohT8aa5s05a6dUUkEQr+NKercYpIRSn409hX677ig5UfqFtXRCpEwZ/G1K0rIntCwZ/Gdozvt9VF2UQkcRW+LHNFWNT2AR4GOgAOXAr0BIYDBfGX/Z9H/IUg68hEZbt1ex/SO+xyRCSNBBr8wN3AVI/4uRa1HKA2seD/h0f8zoC3ndFKu3W75XZTt66IVEhgwW9RawCcAFwC4BEvAoosqlMOq4K6dUVkTwW5x38QseGc0Ra1jsA84Or4c1dY1AYDc4FrPOKFAdaRcdStKyKVEeTkbnWgM3C/R/wIYCNwA3A/0AboBHwD/L28lc1shJnNNbO5BQUF5b0ka5V267Zo0ELduiJSYUEG/zJgmUf87fjjCUBnj/hKj3ixR7wEeAg4qryV3f1Bd89z97zGjRsHWGb6Kdu0pW5dEamowILfI74CWGpRaxdfdAow36LWtMzLzgY+DqqGTKWbrohIZQR9Vs+VwNj4GT35wFDgHotaJ2Kndy4GfhFwDRmltFu3To066tYVkT0SaPB7xN8H8nZZfHGQ28x06tYVkcpS526a0UXZRKSyFPxpZGPRRnXrikilKfjTyPT86Wwt3qp764pIpSj404i6dUWkKij400TZbt0+h+hqnCKy5xT8aWLu13N3dOv+rMnPwi5HRNKYgj9NlB3mUbeuiFSGgj9NqFtXRKqKgj8NlO3WPbHViWGXIyJpTsGfBsp269asXjPkakQk3Sn404C6dUWkKin4U5y6dUWkqin4U5y6dUWkqin4U1zpaZx92qppS0SqhoI/he10b912Gt8Xkaqh4E9h6tYVkSAo+FOYunVFJAgK/hRW2q2ri7KJSFVS8Kco3VtXRIKi4E9Rzy+ITeqqW1dEqpqCP0XpomwiEhQFfwoq263bq22vsMsRkQyj4E9Bpd26RzU/iv3r7h92OSKSYRT8KUgXZRORICn4U0yJl3wf/OrWFZEAKPhTjLp1RSRoCv4UU/aibOrWFZEgKPhTzI7TODXMIyIBUfCnkKXrlqpbV0QCp+BPIbq3rogkg4I/hahbV0SSQcGfItStKyLJouBPEerWFZFkUfCnCHXrikiyKPhTgLp1RSSZFPwpoLRb98D6B6pbV0QCVz3IN7eo7QM8DHQAHLgU+BwYB7QCFgMDPeKFQdaR6nRvXRFJpqD3+O8GpnrE2wMdgU+BG4CXPeJtgZfjj7PalIUa5hGR5Aks+C1qDYATgJEAHvEij/ha4CxgTPxlY4B+QdWQDpauW8r7K95Xt66IJE2QQz0HAQXAaItaR2AecDWwv0f8m/hrVgDlnrtoZiOAEQAtWrQIsMxwqVtXRJItyKGe6kBn4H6P+BHARnYZ1vGIO7Gx/x9w9wfdPc/d8xo3bhxgmeEq7dbtc0ifkCsRkWwRZPAvA5Z5xN+OP55A7ItgpUWtKUD871UB1pDSynbr9m7bO+xyRCRLBBb8HvEVwFKLWrv4olOA+cCzwJD4siHA5KBqSHUz8meoW1dEki7Q0zmBK4GxFrUcIB8YSuzLZrxF7TJgCTAw4BpSli7KJiJhCDT4PeLvA3nlPHVKkNtNB2W7dTW+LyLJpM7dkJTt1v35/j8PuxwRySIK/pCUvSibunVFJJkU/CHRvXVFJCwK/hCoW1dEwhT0WT2hWrN5DcUlxWGX8QNPzX8KgFPbnKpuXRFJuowO/uNHH8/8gvlhl7FbOo1TRMKQ0cHfsGZDGtVuFHYZ5WrZoCX9D+0fdhkikoUyOvjfuPSNsEsQEUk5mtwVEckyCn4RkSyj4BcRyTIKfhGRLKPgFxHJMgp+EZEso+AXEckyCn4RkSxj7uXe6zylmFkBsbt17YlGwOoqLKeqqK6KUV0Vo7oEoKW7N951YVoEf2WY2Vx3L+8uYKFSXRWjuipGdcmP0VCPiEiWUfCLiGSZbAj+B8MuYDdUV8WoropRXbJbGT/GLyIiO8uGPX4RESlDwS8ikmUy9kYsFrWawExgb2Kfc4JHPBJuVd+zqFUD5gLLPeJ9wq4HwKK2GFgPFAPbPZIap91Z1PYBHgY6AA5c6hGfFXJN7YBxZRa1Bm7yiP8znIq+Z1H7DTCM2O/qI2CoR3xLuFWBRe1qYDhgwEOp8LvKVpm8x78VONkj3hHoBJxuUesWbkk7uRr4NOwiytHdI94pVUI/7m5gqke8PdCRFPi9ecQ/j/+eOgFdgE3A0+FWBRa15sBVQJ5HvANQDRgUblVgUetALPSPIvbfsI9F7eBwq8peGbvH7xF3YEP8YY34n5SYybao5QK9gduA34ZcTkqzqDUATgAuAfCIFwFFYdZUjlOARR7xPe0ur2rVgVoWtW1AbeDrkOsBOBR42yO+CcCi9hrQH/hbqFVlqUze48eiVs2i9j6wCpjuEX875JJK/RO4DigJuY5dOTDNojbPojYi7GLiDgIKgNEWtfcsag9b1OqEXdQuBgFPhF0EgEd8OXAn8BXwDbDOIz4t3KoA+Bg43qK2n0WtNtALODDkmrJWRge/R7w4fiieCxwVP9wMlUWtD7DKIz4v7FrKcZxHvDNwBvBri9oJYRdEbO+1M3C/R/wIYCNwQ7glfc+ilgOcCTwVdi0AFrWGwFnEvjCbAXUsaheFWxV4xD8FbgemAVOB94nNJUkIMjr4S3nE1wKvAKeHXArAscCZ8YnUJ4GTLWqPhVtSTHxvEY/4KmLj1UeFWxEAy4BlZY7WJhD7IkgVZwDvesRXhl1IXA/gS494gUd8GzAJOCbkmgDwiI/0iHfxiJ8AFAILwq4pW2Vs8FvUGsfPBsGiVgs4Ffgs1KIAj/jvPeK5HvFWxIYI/ucRD32PzKJWx6JWr/Rn4DRih+eh8oivAJbGz6KB2Hj6/BBL2tX5pMgwT9xXQDeLWm2LmhH7fYU+GQ5gUWsS/7sFsfH9x8OtKHtl7OQu0BQYEz9tci9gvEd8Ssg1pbL9gactahD7/+Jxj/jUcEva4UpgbHxYJR8YGnI9wI4vyFOBX4RdSymP+NsWtQnAu8B24D1S5zIJEy1q+wHbgF/Hj8QlBLpkg4hIlsnYoR4RESmfgl9EJMso+EVEsoyCX0Qkyyj4RUSyjIJfpBxm3GzGtXuwXiczelX2fUSCpOAXqVqd4PvgF0lFCn6RODP+YMYCM94A2sWXtTFjqhnzzHjdjPbx5Y+Y8R8z5sbX6WNGDnALcJ4Z75txXvytDzPjVTPyzbgqnE8n8r1M7twVSZgZXYhdQqMTsX8X7wLziHW9Xu7OQjO6AvcBJ8dXa0XsekZtiF0L6mDgJiDPnSvi73sz0B7oDtQDPjfjfne2JeWDiZRDwS8SczzwtDux68UbzwI1iV3g7CmzHa/bu8w6490pARaakU8s4MvzvDtbga1mrCJ2eYxlVf8RRBKj4BfZvb2Ate502s3zu17vZHfXP9la5udi9O9OQqYxfpGYmUA/M2qZUQ/oS+x2il+aMQDADDOjY5l1BpixlxltiN1z93Ni9yyul+TaRSpEwS8CuPMusZunfwC8CLwTf+pC4DIzPgA+IXaTk1JfAXPir7/cnS3ExvoP22VyVySl6OqcInvAjEeAKe5MCLsWkYrSHr+ISJbRHr+ISJbRHr+ISJZR8IuIZBkFv4hIllHwi4hkGQW/iEiW+X/0K80IjGiP/QAAAABJRU5ErkJggg==\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "accuracy_list = []\n",
    "for i in range(1,10):\n",
    "    DecisionTree = DecisionTreeClassifier(max_depth = i, criterion='gini')\n",
    "    DecisionTree.fit(x_train, y_train)\n",
    "    y_pred = DecisionTree.predict(x_test)\n",
    "    accuracy_list.append(accuracy_score(y_test,y_pred)*100)\n",
    "plt.figure()\n",
    "plt.xticks([0,1,2,3,4,5,6],[3,4,5,6,7,8,9])\n",
    "plt.plot(accuracy_list, color=\"green\", linewidth=2)\n",
    "plt.xlabel(\"depth\",c=\"b\")\n",
    "plt.ylabel(\"accuracy\",c=\"b\")\n",
    "plt.yticks(c= 'green')\n",
    "plt.xticks(c= 'green')\n",
    "plt.title(\"Decision Tree\",c=\"b\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## 10 is the best for max depth"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 417,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "[Text(279.0, 475.65000000000003, 'X[0] <= 0.162\\ngini = 0.5\\nsamples = 6316\\nvalue = [3158, 3158]'),\n Text(139.5, 339.75, 'X[0] <= -0.751\\ngini = 0.265\\nsamples = 2193\\nvalue = [1849, 344]'),\n Text(69.75, 203.85000000000002, 'gini = 0.0\\nsamples = 891\\nvalue = [891, 0]'),\n Text(209.25, 203.85000000000002, 'gini = 0.389\\nsamples = 1302\\nvalue = [958, 344]'),\n Text(418.5, 339.75, 'X[0] <= 0.96\\ngini = 0.433\\nsamples = 4123\\nvalue = [1309, 2814]'),\n Text(348.75, 203.85000000000002, 'X[3] <= 0.038\\ngini = 0.493\\nsamples = 1695\\nvalue = [746, 949]'),\n Text(279.0, 67.94999999999999, 'gini = 0.487\\nsamples = 858\\nvalue = [499, 359]'),\n Text(418.5, 67.94999999999999, 'gini = 0.416\\nsamples = 837\\nvalue = [247, 590]'),\n Text(488.25, 203.85000000000002, 'gini = 0.356\\nsamples = 2428\\nvalue = [563, 1865]')]"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 417
    },
    {
     "data": {
      "text/plain": "<Figure size 720x720 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "DT = DecisionTreeClassifier(criterion = \"gini\", max_leaf_nodes=5,max_depth=10).fit(x_train,y_train)\n",
    "fig, axs = plt.subplots(1, 1, figsize=(10,10))\n",
    "tree.plot_tree(DT, ax=axs)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 418,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "0.7437425506555423"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 418
    }
   ],
   "source": [
    "y_pred = DT.predict(x_test)\n",
    "metrics.accuracy_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 419,
   "outputs": [
    {
     "name": "stdout",
     "text": [
      "acc for training data: 76.045\n",
      "acc for test data: 74.374\n"
     ],
     "output_type": "stream"
    }
   ],
   "source": [
    "print('acc for training data: {:.3f}'.format(DT.score(x_train, y_train)*100))\n",
    "print('acc for test data: {:.3f}'.format(DT.score(x_test, y_test)*100))"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n",
     "is_executing": false
    }
   }
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## classification report for gini"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 420,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.75      0.85       789\n",
      "           1       0.15      0.68      0.24        50\n",
      "\n",
      "    accuracy                           0.74       839\n",
      "   macro avg       0.56      0.71      0.54       839\n",
      "weighted avg       0.92      0.74      0.81       839\n",
      "\n"
     ],
     "output_type": "stream"
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "### ٍ Let's test Entropy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 421,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 432x288 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "accuracy_list = []\n",
    "for i in range(1,10):\n",
    "    DecisionTree = DecisionTreeClassifier(max_depth = i, criterion=\"entropy\")\n",
    "    DecisionTree.fit(x_train, y_train)\n",
    "    y_pred = DecisionTree.predict(x_test)\n",
    "    accuracy_list.append(accuracy_score(y_test,y_pred)*100)\n",
    "plt.figure()\n",
    "plt.xticks([0,1,2,3,4,5,6],[3,4,5,6,7,8,9])\n",
    "plt.plot(accuracy_list, color=\"r\", linewidth=2)\n",
    "plt.xlabel(\"depth\",c=\"b\")\n",
    "plt.ylabel(\"accuracy\",c=\"b\")\n",
    "plt.yticks(c= 'r')\n",
    "plt.xticks(c= 'r')\n",
    "plt.title(\"Decision Tree\",c=\"b\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10 is the best for max depth"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 422,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "[Text(279.0, 475.65000000000003, 'X[0] <= -0.009\\nentropy = 1.0\\nsamples = 6316\\nvalue = [3158, 3158]'),\n Text(139.5, 339.75, 'X[0] <= -0.751\\nentropy = 0.52\\nsamples = 1857\\nvalue = [1640, 217]'),\n Text(69.75, 203.85000000000002, 'entropy = 0.0\\nsamples = 891\\nvalue = [891, 0]'),\n Text(209.25, 203.85000000000002, 'entropy = 0.769\\nsamples = 966\\nvalue = [749, 217]'),\n Text(418.5, 339.75, 'X[0] <= 0.96\\nentropy = 0.925\\nsamples = 4459\\nvalue = [1518, 2941]'),\n Text(348.75, 203.85000000000002, 'X[3] <= 0.038\\nentropy = 0.997\\nsamples = 2031\\nvalue = [955, 1076]'),\n Text(279.0, 67.94999999999999, 'entropy = 0.978\\nsamples = 1097\\nvalue = [645, 452]'),\n Text(418.5, 67.94999999999999, 'entropy = 0.917\\nsamples = 934\\nvalue = [310, 624]'),\n Text(488.25, 203.85000000000002, 'entropy = 0.781\\nsamples = 2428\\nvalue = [563, 1865]')]"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 422
    },
    {
     "data": {
      "text/plain": "<Figure size 720x720 with 1 Axes>",
      "image/png": "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\n"
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "DT = DecisionTreeClassifier(criterion = \"entropy\", max_leaf_nodes=5,max_depth=10).fit(x_train,y_train)\n",
    "fig, axs = plt.subplots(1, 1, figsize=(10,10))\n",
    "tree.plot_tree(DT, ax=axs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 423,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "0.7234803337306317"
     },
     "metadata": {},
     "output_type": "execute_result",
     "execution_count": 423
    }
   ],
   "source": [
    "y_pred = DT.predict(x_test)\n",
    "metrics.accuracy_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 424,
   "outputs": [
    {
     "name": "stdout",
     "text": [
      "acc for training data: 75.586\n",
      "acc for test data: 72.348\n"
     ],
     "output_type": "stream"
    }
   ],
   "source": [
    "print('acc for training data: {:.3f}'.format(DT.score(x_train, y_train)*100))\n",
    "print('acc for test data: {:.3f}'.format(DT.score(x_test, y_test)*100))"
   ],
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n",
     "is_executing": false
    }
   }
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## classification report for entropy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 425,
   "metadata": {
    "pycharm": {
     "is_executing": false,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.72      0.83       789\n",
      "           1       0.14      0.70      0.23        50\n",
      "\n",
      "    accuracy                           0.72       839\n",
      "   macro avg       0.56      0.71      0.53       839\n",
      "weighted avg       0.92      0.72      0.80       839\n",
      "\n"
     ],
     "output_type": "stream"
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred))\n",
    "\n"
   ]
  }
 ],
 "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": "3.7.2"
  },
  "pycharm": {
   "stem_cell": {
    "cell_type": "raw",
    "source": [],
    "metadata": {
     "collapsed": false
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}