ml_03_1-HW.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## First analisys"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from scipy import stats\n",
    "from sklearn import preprocessing as pr\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1000025</th>\n",
       "      <th>5</th>\n",
       "      <th>1</th>\n",
       "      <th>1.1</th>\n",
       "      <th>1.2</th>\n",
       "      <th>2</th>\n",
       "      <th>1.3</th>\n",
       "      <th>3</th>\n",
       "      <th>1.4</th>\n",
       "      <th>1.5</th>\n",
       "      <th>2.1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1002945</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1015425</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1016277</td>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1017023</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1017122</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "      <td>10</td>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   1000025  5   1  1.1  1.2  2 1.3  3  1.4  1.5  2.1\n",
       "0  1002945  5   4    4    5  7  10  3    2    1    2\n",
       "1  1015425  3   1    1    1  2   2  3    1    1    2\n",
       "2  1016277  6   8    8    1  3   4  3    7    1    2\n",
       "3  1017023  4   1    1    3  2   1  3    1    1    2\n",
       "4  1017122  8  10   10    8  7  10  9    7    1    4"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('datasets/breast-cancer-wisconsin.data')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000025     int64\n",
      "5           int64\n",
      "1           int64\n",
      "1.1         int64\n",
      "1.2         int64\n",
      "2           int64\n",
      "1.3        object\n",
      "3           int64\n",
      "1.4         int64\n",
      "1.5         int64\n",
      "2.1         int64\n",
      "dtype: object\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 698 entries, 0 to 697\n",
      "Data columns (total 11 columns):\n",
      " #   Column   Non-Null Count  Dtype \n",
      "---  ------   --------------  ----- \n",
      " 0   1000025  698 non-null    int64 \n",
      " 1   5        698 non-null    int64 \n",
      " 2   1        698 non-null    int64 \n",
      " 3   1.1      698 non-null    int64 \n",
      " 4   1.2      698 non-null    int64 \n",
      " 5   2        698 non-null    int64 \n",
      " 6   1.3      698 non-null    object\n",
      " 7   3        698 non-null    int64 \n",
      " 8   1.4      698 non-null    int64 \n",
      " 9   1.5      698 non-null    int64 \n",
      " 10  2.1      698 non-null    int64 \n",
      "dtypes: int64(10), object(1)\n",
      "memory usage: 60.1+ KB\n"
     ]
    }
   ],
   "source": [
    "print(df.dtypes)\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1000025</th>\n",
       "      <th>5</th>\n",
       "      <th>1</th>\n",
       "      <th>1.1</th>\n",
       "      <th>1.2</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>1.4</th>\n",
       "      <th>1.5</th>\n",
       "      <th>2.1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>6.980000e+02</td>\n",
       "      <td>698.000000</td>\n",
       "      <td>698.000000</td>\n",
       "      <td>698.000000</td>\n",
       "      <td>698.000000</td>\n",
       "      <td>698.000000</td>\n",
       "      <td>698.000000</td>\n",
       "      <td>698.000000</td>\n",
       "      <td>698.000000</td>\n",
       "      <td>698.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.071807e+06</td>\n",
       "      <td>4.416905</td>\n",
       "      <td>3.137536</td>\n",
       "      <td>3.210602</td>\n",
       "      <td>2.809456</td>\n",
       "      <td>3.217765</td>\n",
       "      <td>3.438395</td>\n",
       "      <td>2.869628</td>\n",
       "      <td>1.590258</td>\n",
       "      <td>2.690544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6.175323e+05</td>\n",
       "      <td>2.817673</td>\n",
       "      <td>3.052575</td>\n",
       "      <td>2.972867</td>\n",
       "      <td>2.856606</td>\n",
       "      <td>2.215408</td>\n",
       "      <td>2.440056</td>\n",
       "      <td>3.055004</td>\n",
       "      <td>1.716162</td>\n",
       "      <td>0.951596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>6.163400e+04</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>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>8.702582e+05</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.171710e+06</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.238354e+06</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.345435e+07</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            1000025           5           1         1.1         1.2  \\\n",
       "count  6.980000e+02  698.000000  698.000000  698.000000  698.000000   \n",
       "mean   1.071807e+06    4.416905    3.137536    3.210602    2.809456   \n",
       "std    6.175323e+05    2.817673    3.052575    2.972867    2.856606   \n",
       "min    6.163400e+04    1.000000    1.000000    1.000000    1.000000   \n",
       "25%    8.702582e+05    2.000000    1.000000    1.000000    1.000000   \n",
       "50%    1.171710e+06    4.000000    1.000000    1.000000    1.000000   \n",
       "75%    1.238354e+06    6.000000    5.000000    5.000000    4.000000   \n",
       "max    1.345435e+07   10.000000   10.000000   10.000000   10.000000   \n",
       "\n",
       "                2           3         1.4         1.5         2.1  \n",
       "count  698.000000  698.000000  698.000000  698.000000  698.000000  \n",
       "mean     3.217765    3.438395    2.869628    1.590258    2.690544  \n",
       "std      2.215408    2.440056    3.055004    1.716162    0.951596  \n",
       "min      1.000000    1.000000    1.000000    1.000000    2.000000  \n",
       "25%      2.000000    2.000000    1.000000    1.000000    2.000000  \n",
       "50%      2.000000    3.000000    1.000000    1.000000    2.000000  \n",
       "75%      4.000000    5.000000    4.000000    1.000000    4.000000  \n",
       "max     10.000000   10.000000   10.000000   10.000000    4.000000  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1000025</th>\n",
       "      <th>5</th>\n",
       "      <th>1</th>\n",
       "      <th>1.1</th>\n",
       "      <th>1.2</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>1.4</th>\n",
       "      <th>1.5</th>\n",
       "      <th>2.1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1000025</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.055276</td>\n",
       "      <td>-0.041735</td>\n",
       "      <td>-0.041717</td>\n",
       "      <td>-0.065003</td>\n",
       "      <td>-0.045630</td>\n",
       "      <td>-0.060082</td>\n",
       "      <td>-0.052188</td>\n",
       "      <td>-0.034961</td>\n",
       "      <td>-0.080378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>-0.055276</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.645366</td>\n",
       "      <td>0.655089</td>\n",
       "      <td>0.486699</td>\n",
       "      <td>0.522108</td>\n",
       "      <td>0.558511</td>\n",
       "      <td>0.536176</td>\n",
       "      <td>0.350176</td>\n",
       "      <td>0.716509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.041735</td>\n",
       "      <td>0.645366</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.906814</td>\n",
       "      <td>0.705397</td>\n",
       "      <td>0.751674</td>\n",
       "      <td>0.755824</td>\n",
       "      <td>0.722699</td>\n",
       "      <td>0.458548</td>\n",
       "      <td>0.817772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.1</th>\n",
       "      <td>-0.041717</td>\n",
       "      <td>0.655089</td>\n",
       "      <td>0.906814</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.682871</td>\n",
       "      <td>0.719524</td>\n",
       "      <td>0.736066</td>\n",
       "      <td>0.719272</td>\n",
       "      <td>0.438756</td>\n",
       "      <td>0.818794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.2</th>\n",
       "      <td>-0.065003</td>\n",
       "      <td>0.486699</td>\n",
       "      <td>0.705397</td>\n",
       "      <td>0.682871</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.599402</td>\n",
       "      <td>0.666759</td>\n",
       "      <td>0.603132</td>\n",
       "      <td>0.417476</td>\n",
       "      <td>0.696605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.045630</td>\n",
       "      <td>0.522108</td>\n",
       "      <td>0.751674</td>\n",
       "      <td>0.719524</td>\n",
       "      <td>0.599402</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.616108</td>\n",
       "      <td>0.628704</td>\n",
       "      <td>0.478975</td>\n",
       "      <td>0.682618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.060082</td>\n",
       "      <td>0.558511</td>\n",
       "      <td>0.755824</td>\n",
       "      <td>0.736066</td>\n",
       "      <td>0.666759</td>\n",
       "      <td>0.616108</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.665914</td>\n",
       "      <td>0.344118</td>\n",
       "      <td>0.756732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.4</th>\n",
       "      <td>-0.052188</td>\n",
       "      <td>0.536176</td>\n",
       "      <td>0.722699</td>\n",
       "      <td>0.719272</td>\n",
       "      <td>0.603132</td>\n",
       "      <td>0.628704</td>\n",
       "      <td>0.665914</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.428185</td>\n",
       "      <td>0.712067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.5</th>\n",
       "      <td>-0.034961</td>\n",
       "      <td>0.350176</td>\n",
       "      <td>0.458548</td>\n",
       "      <td>0.438756</td>\n",
       "      <td>0.417476</td>\n",
       "      <td>0.478975</td>\n",
       "      <td>0.344118</td>\n",
       "      <td>0.428185</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.423008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.1</th>\n",
       "      <td>-0.080378</td>\n",
       "      <td>0.716509</td>\n",
       "      <td>0.817772</td>\n",
       "      <td>0.818794</td>\n",
       "      <td>0.696605</td>\n",
       "      <td>0.682618</td>\n",
       "      <td>0.756732</td>\n",
       "      <td>0.712067</td>\n",
       "      <td>0.423008</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          1000025         5         1       1.1       1.2         2         3  \\\n",
       "1000025  1.000000 -0.055276 -0.041735 -0.041717 -0.065003 -0.045630 -0.060082   \n",
       "5       -0.055276  1.000000  0.645366  0.655089  0.486699  0.522108  0.558511   \n",
       "1       -0.041735  0.645366  1.000000  0.906814  0.705397  0.751674  0.755824   \n",
       "1.1     -0.041717  0.655089  0.906814  1.000000  0.682871  0.719524  0.736066   \n",
       "1.2     -0.065003  0.486699  0.705397  0.682871  1.000000  0.599402  0.666759   \n",
       "2       -0.045630  0.522108  0.751674  0.719524  0.599402  1.000000  0.616108   \n",
       "3       -0.060082  0.558511  0.755824  0.736066  0.666759  0.616108  1.000000   \n",
       "1.4     -0.052188  0.536176  0.722699  0.719272  0.603132  0.628704  0.665914   \n",
       "1.5     -0.034961  0.350176  0.458548  0.438756  0.417476  0.478975  0.344118   \n",
       "2.1     -0.080378  0.716509  0.817772  0.818794  0.696605  0.682618  0.756732   \n",
       "\n",
       "              1.4       1.5       2.1  \n",
       "1000025 -0.052188 -0.034961 -0.080378  \n",
       "5        0.536176  0.350176  0.716509  \n",
       "1        0.722699  0.458548  0.817772  \n",
       "1.1      0.719272  0.438756  0.818794  \n",
       "1.2      0.603132  0.417476  0.696605  \n",
       "2        0.628704  0.478975  0.682618  \n",
       "3        0.665914  0.344118  0.756732  \n",
       "1.4      1.000000  0.428185  0.712067  \n",
       "1.5      0.428185  1.000000  0.423008  \n",
       "2.1      0.712067  0.423008  1.000000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "corr_ = df.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(corr_, annot=True, fmt='.2f')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5 2.1 0.7165085190765071\n",
      "1 1.1 0.9068136687904309\n",
      "1 1.2 0.7053969744849616\n",
      "1 2 0.751674461765537\n",
      "1 3 0.7558236148428459\n",
      "1 1.4 0.7226987041005679\n"
     ]
    }
   ],
   "source": [
    "CorField = []\n",
    "for i in corr_:\n",
    "    for j in corr_.index[corr_[i] > 0.7]:\n",
    "        if i != j and j not in CorField and i not in CorField:\n",
    "            CorField.append(j)\n",
    "            print (i,j, corr_[i][corr_.index==j].values[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "\"None of [Index(['bore', 'stroke', 'compression-ratio', 'horsepower'], dtype='object')] are in the [columns]\"",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-9-4112d6e7ee8a>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mcorr_\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mdf\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m'bore'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'stroke'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'compression-ratio'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'horsepower'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcorr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mcorr_\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\frame.py\u001b[0m in \u001b[0;36m__getitem__\u001b[1;34m(self, key)\u001b[0m\n\u001b[0;32m   3028\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mis_iterator\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3029\u001b[0m                 \u001b[0mkey\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 3030\u001b[1;33m             \u001b[0mindexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mloc\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_listlike_indexer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mraise_missing\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   3031\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   3032\u001b[0m         \u001b[1;31m# take() does not accept boolean indexers\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py\u001b[0m in \u001b[0;36m_get_listlike_indexer\u001b[1;34m(self, key, axis, raise_missing)\u001b[0m\n\u001b[0;32m   1264\u001b[0m             \u001b[0mkeyarr\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mindexer\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mnew_indexer\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0max\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_reindex_non_unique\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkeyarr\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1265\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1266\u001b[1;33m         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_validate_read_indexer\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkeyarr\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mindexer\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mraise_missing\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mraise_missing\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1267\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mkeyarr\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mindexer\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1268\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py\u001b[0m in \u001b[0;36m_validate_read_indexer\u001b[1;34m(self, key, indexer, axis, raise_missing)\u001b[0m\n\u001b[0;32m   1306\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mmissing\u001b[0m \u001b[1;33m==\u001b[0m \u001b[0mlen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mindexer\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1307\u001b[0m                 \u001b[0maxis_name\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mobj\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_axis_name\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0maxis\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1308\u001b[1;33m                 \u001b[1;32mraise\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34mf\"None of [{key}] are in the [{axis_name}]\"\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1309\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1310\u001b[0m             \u001b[0max\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mobj\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_axis\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0maxis\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyError\u001b[0m: \"None of [Index(['bore', 'stroke', 'compression-ratio', 'horsepower'], dtype='object')] are in the [columns]\""
     ]
    }
   ],
   "source": [
    "corr_ = df[['bore', 'stroke', 'compression-ratio', 'horsepower']].corr()\n",
    "corr_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.regplot(x='engine-size', y='price', data=df)\n",
    "plt.ylim(0,)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.regplot(x='highway-mpg', y='price', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[['highway-mpg', 'price']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.regplot(x='peak-rpm', y='price', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[['peak-rpm', 'price']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.regplot(x='stroke', y='price', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[['stroke', 'price']].corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.boxplot(x='body-style', y='price', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.boxplot(x='engine-location', y='price', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.boxplot(x='drive-wheels', y='price', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.describe(include='object')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['drive-wheels'].value_counts().to_frame()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "drive_wheel_counts = df['drive-wheels'].value_counts().to_frame()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "drive_wheel_counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "engine_loc_counts = df['engine-location'].value_counts().to_frame()\n",
    "engine_loc_counts.rename(columns={'engine-localtion': 'value_counts'}, inplace=True)\n",
    "engine_loc_counts.index.name = 'engine-location'\n",
    "engine_loc_counts.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['drive-wheels'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_group_one = df[['drive-wheels', 'body-style', 'price']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_group_one = df_group_one.groupby(['drive-wheels'], as_index=False).mean()\n",
    "df_group_one"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_gptest = df[['drive-wheels', 'body-style', 'price']]\n",
    "grouped_test1 = df_gptest.groupby(['drive-wheels', 'body-style'], as_index=False).mean()\n",
    "grouped_test1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "grouped_pivot = grouped_test1.pivot(index='drive-wheels', columns='body-style')\n",
    "grouped_pivot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.pcolor(grouped_pivot, cmap='RdBu')\n",
    "plt.colorbar()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots()\n",
    "im = ax.pcolor(grouped_pivot, cmap='RdBu')\n",
    "\n",
    "row_labels = grouped_pivot.columns.levels[1]\n",
    "col_labels = grouped_pivot.index\n",
    "\n",
    "ax.set_xticks(np.arange(grouped_pivot.shape[1]) + 0.5, minor=False)\n",
    "ax.set_yticks(np.arange(grouped_pivot.shape[0]) + 0.5, minor=False)\n",
    "\n",
    "ax.set_xticklabels(row_labels, minor=False)\n",
    "ax.set_yticklabels(col_labels, minor=False)\n",
    "\n",
    "plt.xticks(rotation=90)\n",
    "\n",
    "fig.colorbar(im)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.heatmap(grouped_pivot, annot=True, fmt='.2f')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols = ['wheel-base', 'horsepower', 'length', 'width', 'curb-weight', 'engine-size', 'bore', 'city-mpg', 'highway-mpg']\n",
    "for col in cols:\n",
    "    pearson_coef, p_value = stats.pearsonr(df[col], df['price'])\n",
    "    print(f'PearsonR между {col} и price {pearson_coef} c Р-значением Р={p_value}')\n",
    "    \n",
    "    if p_value < 0.001:\n",
    "        print(f'Корреляция между {col} и price статистически значимая')\n",
    "    elif p_value < 0.05:\n",
    "        print(f'Корреляция между {col} и price средняя')\n",
    "    elif p_value < 0.1:\n",
    "        print(f'Корреляция между {col} и price слабая')\n",
    "    \n",
    "    if pearson_coef > 0:\n",
    "        if pearson_coef > 0.85:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает положительную очень сильную связь\\n')\n",
    "        elif pearson_coef > 0.75:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает положительную достаточно сильную связь\\n')\n",
    "        elif pearson_coef > 0.60:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает положительную относительно сильную связь\\n')\n",
    "        elif pearson_coef > 0.50:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает положительную среднюю сильную связь\\n')\n",
    "        else:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает положительную слабую связь')\n",
    "    else:\n",
    "        if abs(pearson_coef) > 0.85:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает негативную очень сильную связь\\n')\n",
    "        elif abs(pearson_coef) > 0.75:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает негативную достаточно сильную связь\\n')\n",
    "        elif abs(pearson_coef) > 0.60:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает негативную относительно сильную связь\\n')\n",
    "        elif abs(pearson_coef) > 0.50:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает положительную среднюю сильную связь\\n')\n",
    "        else:\n",
    "            print(f'Коэффициент ~{pearson_coef} показывает положительную слабую связь')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "grouped_test2 = df_gptest[['drive-wheels', 'price']].groupby(['drive-wheels'])\n",
    "grouped_test2.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_gptest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "grouped_test2.get_group('4wd')['price']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "f_val, p_val = stats.f_oneway(grouped_test2.get_group('fwd')['price'],\n",
    "                              grouped_test2.get_group('rwd')['price'],\n",
    "                              grouped_test2.get_group('4wd')['price'])\n",
    "print(f'ANOVA results: F= {f_val}, P= {p_val}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "f_val, p_val = stats.f_oneway(grouped_test2.get_group('fwd')['price'], grouped_test2.get_group('rwd')['price'])\n",
    "print(f'ANOVA results: F= {f_val}, P= {p_val}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "f_val, p_val = stats.f_oneway(grouped_test2.get_group('4wd')['price'], grouped_test2.get_group('rwd')['price'])\n",
    "print(f'ANOVA results: F= {f_val}, P= {p_val}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "f_val, p_val = stats.f_oneway(grouped_test2.get_group('4wd')['price'], grouped_test2.get_group('fwd')['price'])\n",
    "print(f'ANOVA results: F= {f_val}, P= {p_val}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing by sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "class MultiColumnLabelEncoder:\n",
    "    def __init__(self, columns=None):\n",
    "        self.columns = columns\n",
    "            \n",
    "    def fit(self, X, y=None):\n",
    "        return self\n",
    "    \n",
    "    def transform(self, X):\n",
    "        output = X.copy()\n",
    "        if self.columns is not None:\n",
    "            for col in self.columns:\n",
    "                output[col] = pr.LabelEncoder().fit_transform(output[col].astype(str))\n",
    "        else:\n",
    "            for colname, col in output.iteritems():\n",
    "                output[colname] = pr.LabelEncoder().fit_transform(col.astype(str))\n",
    "        return output\n",
    "    \n",
    "    def fit_transform(self, X, y=None):\n",
    "        return self.fit(X, y).transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = MultiColumnLabelEncoder(columns=['make', 'aspiration', 'num-of-doors', 'body-style', 'drive-wheels', 'engine-location', 'engine-type', 'num-of-cylinders', 'fuel-system', 'horsepower-binned']).fit_transform(df)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df.price"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop('price', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sc = StandardScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "regr = RandomForestRegressor(max_depth=2, random_state=0, n_estimators=100)\n",
    "regr.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(regr.feature_importances_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop('price', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "importances = list(zip(regr.feature_importances_, df.columns))\n",
    "importances.sort(reverse=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "importances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}