lewiuberg/tutorial_3.ipynb

## tutorial_3.ipynb
{
  "cells": [
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "# Activity sheet 3\n# Linear Regression\n\n**Objective**\nIn this activity sheet we will learn how to implement linear regression models using scikit-learn. The dataset that is going to be used for training your models has ben provided to you. You should load it using appropriate Python code and Pandas libraries and then split the data into training and testing datasets as indicated below. Create and train different linear regression models, evaluate and compare them following the steps below.\n\n**Dataset description**\nA governmental authority, responsible for environmental issues of your country, wants to decrease the CO2 emissions of cars on the streets. So, it is going to apply a fee policy to car owners based on the predicted emissions of their owning cars. In order to predict the CO2 emission of each car, not being able to measure all the cars individually, is going to use the Fuel Consumption dataset of 2015. It’s a fairly small dataset comprising of around 1000 entries (so we are not going to put your computers into test here)."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Basic data preprocessing\n**A)** Load the data using the Pandas library"
    },
    {
      "metadata": {
        "run_control": {
          "marked": false
        },
        "trusted": true
      },
      "cell_type": "code",
      "source": "import sys\nimport numpy as np\nimport pandas as pd\nimport seaborn as sns\nimport matplotlib.pyplot as plt\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.linear_model import LinearRegression\nfrom sklearn.metrics import mean_absolute_error, mean_squared_error , r2_score\n\n# %matplotlib notebook\n%matplotlib inline\n# plt.rcParams[\"figure.figsize\"] = (10, 6)\n# plt.rcParams.update({\"axes.titlesize\": 16})\n# plt.rcParams.update({\"xtick.labelsize\": 12})\n# plt.rcParams.update({\"ytick.labelsize\": 12})\n# pd.set_option(\"display.max.columns\", None)\n# pd.set_option('display.max_rows', 100)\n# np.set_printoptions(threshold=sys.maxsize)",
      "execution_count": 1,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "def unpack_list(lst):  # Oxford comma\n    if len(lst) == 0:\n        return\n    if len(lst) == 1:\n        return \", \".join(lst)\n    if len(lst) == 2:\n        return \", and \".join(lst)\n    else:\n        first_part = lst[:-1]\n        last_part = lst[-1]\n        return \", \".join(first_part) + \", and \" + last_part\n\n# lst = [element + ', ' for element in lst[:-1] + [lst[-1]]\n    \n#     if len(lst) == 1:\n#         return \" \".join(map(str, lst))\n#     else:\n#         for i in range(0, len(lst)-1):\n#             lst[i] = lst[i] + \", \"\n#         if len(lst) > 2:\n#             lst.insert(-1, \"and \")\n#         return \"\".join(lst)",
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "def word_search(df, *words):\n    words = [word for word in words]\n    sum_words: int = 0\n    found_words: str = []\n    list_of_words: list = []\n        \n    if not words: return\n\n    for word in words:\n        col_count = 0\n        sum_word = 0\n        for column in df:\n            if df[column].dtype == object or df[column].dtype == str:\n                col_count += 1\n                sum_word += df[column].str.contains(f\"^{word}$\").sum()\n                if df[column].str.contains(f\"^{word}$\").any():\n                    if word not in found_words:\n                        found_words.append(word)\n        sum_words += sum_word\n\n    if len(found_words) == 0:\n        found_words = words\n\n    print(\"Columns of dtype str or object:\", col_count)\n    print(f\"Instances of {unpack_list(found_words)} in the dataframe: {sum_words}\")",
      "execution_count": 3,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df = pd.read_csv(\"FuelConsumption.csv\")",
      "execution_count": 4,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**B)** Explore the data: are there any errors, any missing values? are all the attributes numerical or there are also categorical ones?"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.head(10)",
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 5,
          "data": {
            "text/plain": "   MODELYEAR   MAKE       MODEL VEHICLECLASS  ENGINESIZE  CYLINDERS  \\\n0       2014  ACURA         ILX      COMPACT         2.0          4   \n1       2014  ACURA         ILX      COMPACT         2.4          4   \n2       2014  ACURA  ILX HYBRID      COMPACT         1.5          4   \n3       2014  ACURA     MDX 4WD  SUV - SMALL         3.5          6   \n4       2014  ACURA     RDX AWD  SUV - SMALL         3.5          6   \n5       2014  ACURA         RLX     MID-SIZE         3.5          6   \n6       2014  ACURA          TL     MID-SIZE         3.5          6   \n7       2014  ACURA      TL AWD     MID-SIZE         3.7          6   \n8       2014  ACURA      TL AWD     MID-SIZE         3.7          6   \n9       2014  ACURA         TSX      COMPACT         2.4          4   \n\n  TRANSMISSION FUELTYPE  FUELCONSUMPTION_CITY  FUELCONSUMPTION_HWY  \\\n0          AS5        Z                   9.9                  6.7   \n1           M6        Z                  11.2                  7.7   \n2          AV7        Z                   6.0                  5.8   \n3          AS6        Z                  12.7                  9.1   \n4          AS6        Z                  12.1                  8.7   \n5          AS6        Z                  11.9                  7.7   \n6          AS6        Z                  11.8                  8.1   \n7          AS6        Z                  12.8                  9.0   \n8           M6        Z                  13.4                  9.5   \n9          AS5        Z                  10.6                  7.5   \n\n   FUELCONSUMPTION_COMB  FUELCONSUMPTION_COMB_MPG  CO2EMISSIONS  \n0                   8.5                        33           196  \n1                   9.6                        29           221  \n2                   5.9                        48           136  \n3                  11.1                        25           255  \n4                  10.6                        27           244  \n5                  10.0                        28           230  \n6                  10.1                        28           232  \n7                  11.1                        25           255  \n8                  11.6                        24           267  \n9                   9.2                        31           212  ",
            "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>MODELYEAR</th>\n      <th>MAKE</th>\n      <th>MODEL</th>\n      <th>VEHICLECLASS</th>\n      <th>ENGINESIZE</th>\n      <th>CYLINDERS</th>\n      <th>TRANSMISSION</th>\n      <th>FUELTYPE</th>\n      <th>FUELCONSUMPTION_CITY</th>\n      <th>FUELCONSUMPTION_HWY</th>\n      <th>FUELCONSUMPTION_COMB</th>\n      <th>FUELCONSUMPTION_COMB_MPG</th>\n      <th>CO2EMISSIONS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>ILX</td>\n      <td>COMPACT</td>\n      <td>2.0</td>\n      <td>4</td>\n      <td>AS5</td>\n      <td>Z</td>\n      <td>9.9</td>\n      <td>6.7</td>\n      <td>8.5</td>\n      <td>33</td>\n      <td>196</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>ILX</td>\n      <td>COMPACT</td>\n      <td>2.4</td>\n      <td>4</td>\n      <td>M6</td>\n      <td>Z</td>\n      <td>11.2</td>\n      <td>7.7</td>\n      <td>9.6</td>\n      <td>29</td>\n      <td>221</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>ILX HYBRID</td>\n      <td>COMPACT</td>\n      <td>1.5</td>\n      <td>4</td>\n      <td>AV7</td>\n      <td>Z</td>\n      <td>6.0</td>\n      <td>5.8</td>\n      <td>5.9</td>\n      <td>48</td>\n      <td>136</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>MDX 4WD</td>\n      <td>SUV - SMALL</td>\n      <td>3.5</td>\n      <td>6</td>\n      <td>AS6</td>\n      <td>Z</td>\n      <td>12.7</td>\n      <td>9.1</td>\n      <td>11.1</td>\n      <td>25</td>\n      <td>255</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>RDX AWD</td>\n      <td>SUV - SMALL</td>\n      <td>3.5</td>\n      <td>6</td>\n      <td>AS6</td>\n      <td>Z</td>\n      <td>12.1</td>\n      <td>8.7</td>\n      <td>10.6</td>\n      <td>27</td>\n      <td>244</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>RLX</td>\n      <td>MID-SIZE</td>\n      <td>3.5</td>\n      <td>6</td>\n      <td>AS6</td>\n      <td>Z</td>\n      <td>11.9</td>\n      <td>7.7</td>\n      <td>10.0</td>\n      <td>28</td>\n      <td>230</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>TL</td>\n      <td>MID-SIZE</td>\n      <td>3.5</td>\n      <td>6</td>\n      <td>AS6</td>\n      <td>Z</td>\n      <td>11.8</td>\n      <td>8.1</td>\n      <td>10.1</td>\n      <td>28</td>\n      <td>232</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>TL AWD</td>\n      <td>MID-SIZE</td>\n      <td>3.7</td>\n      <td>6</td>\n      <td>AS6</td>\n      <td>Z</td>\n      <td>12.8</td>\n      <td>9.0</td>\n      <td>11.1</td>\n      <td>25</td>\n      <td>255</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>TL AWD</td>\n      <td>MID-SIZE</td>\n      <td>3.7</td>\n      <td>6</td>\n      <td>M6</td>\n      <td>Z</td>\n      <td>13.4</td>\n      <td>9.5</td>\n      <td>11.6</td>\n      <td>24</td>\n      <td>267</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>2014</td>\n      <td>ACURA</td>\n      <td>TSX</td>\n      <td>COMPACT</td>\n      <td>2.4</td>\n      <td>4</td>\n      <td>AS5</td>\n      <td>Z</td>\n      <td>10.6</td>\n      <td>7.5</td>\n      <td>9.2</td>\n      <td>31</td>\n      <td>212</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.describe()",
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 6,
          "data": {
            "text/plain": "       MODELYEAR   ENGINESIZE    CYLINDERS  FUELCONSUMPTION_CITY  \\\ncount     1067.0  1067.000000  1067.000000           1067.000000   \nmean      2014.0     3.346298     5.794752             13.296532   \nstd          0.0     1.415895     1.797447              4.101253   \nmin       2014.0     1.000000     3.000000              4.600000   \n25%       2014.0     2.000000     4.000000             10.250000   \n50%       2014.0     3.400000     6.000000             12.600000   \n75%       2014.0     4.300000     8.000000             15.550000   \nmax       2014.0     8.400000    12.000000             30.200000   \n\n       FUELCONSUMPTION_HWY  FUELCONSUMPTION_COMB  FUELCONSUMPTION_COMB_MPG  \\\ncount          1067.000000           1067.000000               1067.000000   \nmean              9.474602             11.580881                 26.441425   \nstd               2.794510              3.485595                  7.468702   \nmin               4.900000              4.700000                 11.000000   \n25%               7.500000              9.000000                 21.000000   \n50%               8.800000             10.900000                 26.000000   \n75%              10.850000             13.350000                 31.000000   \nmax              20.500000             25.800000                 60.000000   \n\n       CO2EMISSIONS  \ncount   1067.000000  \nmean     256.228679  \nstd       63.372304  \nmin      108.000000  \n25%      207.000000  \n50%      251.000000  \n75%      294.000000  \nmax      488.000000  ",
            "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>MODELYEAR</th>\n      <th>ENGINESIZE</th>\n      <th>CYLINDERS</th>\n      <th>FUELCONSUMPTION_CITY</th>\n      <th>FUELCONSUMPTION_HWY</th>\n      <th>FUELCONSUMPTION_COMB</th>\n      <th>FUELCONSUMPTION_COMB_MPG</th>\n      <th>CO2EMISSIONS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>count</th>\n      <td>1067.0</td>\n      <td>1067.000000</td>\n      <td>1067.000000</td>\n      <td>1067.000000</td>\n      <td>1067.000000</td>\n      <td>1067.000000</td>\n      <td>1067.000000</td>\n      <td>1067.000000</td>\n    </tr>\n    <tr>\n      <th>mean</th>\n      <td>2014.0</td>\n      <td>3.346298</td>\n      <td>5.794752</td>\n      <td>13.296532</td>\n      <td>9.474602</td>\n      <td>11.580881</td>\n      <td>26.441425</td>\n      <td>256.228679</td>\n    </tr>\n    <tr>\n      <th>std</th>\n      <td>0.0</td>\n      <td>1.415895</td>\n      <td>1.797447</td>\n      <td>4.101253</td>\n      <td>2.794510</td>\n      <td>3.485595</td>\n      <td>7.468702</td>\n      <td>63.372304</td>\n    </tr>\n    <tr>\n      <th>min</th>\n      <td>2014.0</td>\n      <td>1.000000</td>\n      <td>3.000000</td>\n      <td>4.600000</td>\n      <td>4.900000</td>\n      <td>4.700000</td>\n      <td>11.000000</td>\n      <td>108.000000</td>\n    </tr>\n    <tr>\n      <th>25%</th>\n      <td>2014.0</td>\n      <td>2.000000</td>\n      <td>4.000000</td>\n      <td>10.250000</td>\n      <td>7.500000</td>\n      <td>9.000000</td>\n      <td>21.000000</td>\n      <td>207.000000</td>\n    </tr>\n    <tr>\n      <th>50%</th>\n      <td>2014.0</td>\n      <td>3.400000</td>\n      <td>6.000000</td>\n      <td>12.600000</td>\n      <td>8.800000</td>\n      <td>10.900000</td>\n      <td>26.000000</td>\n      <td>251.000000</td>\n    </tr>\n    <tr>\n      <th>75%</th>\n      <td>2014.0</td>\n      <td>4.300000</td>\n      <td>8.000000</td>\n      <td>15.550000</td>\n      <td>10.850000</td>\n      <td>13.350000</td>\n      <td>31.000000</td>\n      <td>294.000000</td>\n    </tr>\n    <tr>\n      <th>max</th>\n      <td>2014.0</td>\n      <td>8.400000</td>\n      <td>12.000000</td>\n      <td>30.200000</td>\n      <td>20.500000</td>\n      <td>25.800000</td>\n      <td>60.000000</td>\n      <td>488.000000</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.info()",
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": "<class 'pandas.core.frame.DataFrame'>\nRangeIndex: 1067 entries, 0 to 1066\nData columns (total 13 columns):\n #   Column                    Non-Null Count  Dtype  \n---  ------                    --------------  -----  \n 0   MODELYEAR                 1067 non-null   int64  \n 1   MAKE                      1067 non-null   object \n 2   MODEL                     1067 non-null   object \n 3   VEHICLECLASS              1067 non-null   object \n 4   ENGINESIZE                1067 non-null   float64\n 5   CYLINDERS                 1067 non-null   int64  \n 6   TRANSMISSION              1067 non-null   object \n 7   FUELTYPE                  1067 non-null   object \n 8   FUELCONSUMPTION_CITY      1067 non-null   float64\n 9   FUELCONSUMPTION_HWY       1067 non-null   float64\n 10  FUELCONSUMPTION_COMB      1067 non-null   float64\n 11  FUELCONSUMPTION_COMB_MPG  1067 non-null   int64  \n 12  CO2EMISSIONS              1067 non-null   int64  \ndtypes: float64(4), int64(4), object(5)\nmemory usage: 108.5+ KB\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "The MODELYEAR can be classified as ordinal (qualitative), so the data type will be changed."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df[\"MODELYEAR\"] = df[\"MODELYEAR\"].astype(\"category\")",
      "execution_count": 8,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "df[\"MODELYEAR\"].dtype",
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 9,
          "data": {
            "text/plain": "CategoricalDtype(categories=[2014], ordered=False)"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "num_cols = df.select_dtypes(include=[\"float64\", \"int64\"]).columns.tolist()\n# print(f\"There are {df.select_dtypes(include=['float64', 'int64']).any().sum()} numeric columns named:\\n {str(num_cols)[1:-2]}\")\n\n#print(f\"There are {df.select_dtypes(include=['float64', 'int64']).any().sum()} numeric columns named:\\n {*num_cols,}\")\n\nprint(f\"There are {df.select_dtypes(include=['float64', 'int64']).any().sum()} numeric columns named: {', '.join(map(str, num_cols))}\")",
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": "There are 7 numeric columns named: ENGINESIZE, CYLINDERS, FUELCONSUMPTION_CITY, FUELCONSUMPTION_HWY, FUELCONSUMPTION_COMB, FUELCONSUMPTION_COMB_MPG, CO2EMISSIONS\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "run_control": {
          "marked": false
        },
        "trusted": true
      },
      "cell_type": "code",
      "source": "df.info()",
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": "<class 'pandas.core.frame.DataFrame'>\nRangeIndex: 1067 entries, 0 to 1066\nData columns (total 13 columns):\n #   Column                    Non-Null Count  Dtype   \n---  ------                    --------------  -----   \n 0   MODELYEAR                 1067 non-null   category\n 1   MAKE                      1067 non-null   object  \n 2   MODEL                     1067 non-null   object  \n 3   VEHICLECLASS              1067 non-null   object  \n 4   ENGINESIZE                1067 non-null   float64 \n 5   CYLINDERS                 1067 non-null   int64   \n 6   TRANSMISSION              1067 non-null   object  \n 7   FUELTYPE                  1067 non-null   object  \n 8   FUELCONSUMPTION_CITY      1067 non-null   float64 \n 9   FUELCONSUMPTION_HWY       1067 non-null   float64 \n 10  FUELCONSUMPTION_COMB      1067 non-null   float64 \n 11  FUELCONSUMPTION_COMB_MPG  1067 non-null   int64   \n 12  CO2EMISSIONS              1067 non-null   int64   \ndtypes: category(1), float64(4), int64(3), object(5)\nmemory usage: 101.3+ KB\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(f\"Columns with missing data: {df.isnull().any().sum()}\")\nprint(f\"Instances of missing data: {df.isnull().sum().sum()}\")",
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Columns with missing data: 0\nInstances of missing data: 0\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "word_search(df, \"None\", \"none\", \"0\")",
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Columns of dtype str or object: 5\nInstances of None, none, and 0 in the dataframe: 0\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "It does not appear to bee any missing values, at least not np.nan. or any of my manual search words."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**C)** Split the data into training and testing datasets with a ratio of 80% - 20%"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "train, test = train_test_split(df, test_size=0.2, random_state=42, shuffle=True)",
      "execution_count": 14,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "train.head()",
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 15,
          "data": {
            "text/plain": "    MODELYEAR       MAKE                      MODEL    VEHICLECLASS  \\\n333      2014       FIAT       500 ABARTH HATCHBACK     MINICOMPACT   \n106      2014        BMW      650i xDRIVE CABRIOLET      SUBCOMPACT   \n585      2014       JEEP  GRAND CHEROKEE 4X4 DIESEL  SUV - STANDARD   \n55       2014       AUDI                         S4         COMPACT   \n213      2014  CHEVROLET     EXPRESS 1500 CARGO AWD     VAN - CARGO   \n\n     ENGINESIZE  CYLINDERS TRANSMISSION FUELTYPE  FUELCONSUMPTION_CITY  \\\n333         1.4          4           M5        X                   8.5   \n106         4.4          8           A8        Z                  15.0   \n585         3.0          6           A8        D                  11.2   \n55          3.0          6           A7        Z                  13.2   \n213         5.3          8           A4        X                  18.3   \n\n     FUELCONSUMPTION_HWY  FUELCONSUMPTION_COMB  FUELCONSUMPTION_COMB_MPG  \\\n333                  6.9                   7.8                        36   \n106                  9.8                  12.7                        22   \n585                  8.4                   9.9                        29   \n55                   9.2                  11.4                        25   \n213                 14.2                  16.5                        17   \n\n     CO2EMISSIONS  \n333           179  \n106           292  \n585           267  \n55            262  \n213           380  ",
            "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>MODELYEAR</th>\n      <th>MAKE</th>\n      <th>MODEL</th>\n      <th>VEHICLECLASS</th>\n      <th>ENGINESIZE</th>\n      <th>CYLINDERS</th>\n      <th>TRANSMISSION</th>\n      <th>FUELTYPE</th>\n      <th>FUELCONSUMPTION_CITY</th>\n      <th>FUELCONSUMPTION_HWY</th>\n      <th>FUELCONSUMPTION_COMB</th>\n      <th>FUELCONSUMPTION_COMB_MPG</th>\n      <th>CO2EMISSIONS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>333</th>\n      <td>2014</td>\n      <td>FIAT</td>\n      <td>500 ABARTH HATCHBACK</td>\n      <td>MINICOMPACT</td>\n      <td>1.4</td>\n      <td>4</td>\n      <td>M5</td>\n      <td>X</td>\n      <td>8.5</td>\n      <td>6.9</td>\n      <td>7.8</td>\n      <td>36</td>\n      <td>179</td>\n    </tr>\n    <tr>\n      <th>106</th>\n      <td>2014</td>\n      <td>BMW</td>\n      <td>650i xDRIVE CABRIOLET</td>\n      <td>SUBCOMPACT</td>\n      <td>4.4</td>\n      <td>8</td>\n      <td>A8</td>\n      <td>Z</td>\n      <td>15.0</td>\n      <td>9.8</td>\n      <td>12.7</td>\n      <td>22</td>\n      <td>292</td>\n    </tr>\n    <tr>\n      <th>585</th>\n      <td>2014</td>\n      <td>JEEP</td>\n      <td>GRAND CHEROKEE 4X4 DIESEL</td>\n      <td>SUV - STANDARD</td>\n      <td>3.0</td>\n      <td>6</td>\n      <td>A8</td>\n      <td>D</td>\n      <td>11.2</td>\n      <td>8.4</td>\n      <td>9.9</td>\n      <td>29</td>\n      <td>267</td>\n    </tr>\n    <tr>\n      <th>55</th>\n      <td>2014</td>\n      <td>AUDI</td>\n      <td>S4</td>\n      <td>COMPACT</td>\n      <td>3.0</td>\n      <td>6</td>\n      <td>A7</td>\n      <td>Z</td>\n      <td>13.2</td>\n      <td>9.2</td>\n      <td>11.4</td>\n      <td>25</td>\n      <td>262</td>\n    </tr>\n    <tr>\n      <th>213</th>\n      <td>2014</td>\n      <td>CHEVROLET</td>\n      <td>EXPRESS 1500 CARGO AWD</td>\n      <td>VAN - CARGO</td>\n      <td>5.3</td>\n      <td>8</td>\n      <td>A4</td>\n      <td>X</td>\n      <td>18.3</td>\n      <td>14.2</td>\n      <td>16.5</td>\n      <td>17</td>\n      <td>380</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "train.shape",
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 16,
          "data": {
            "text/plain": "(853, 13)"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Christos method:\nNote! He has picked out some particular features. I was under the impression that we should not make such decisions before splitting the data, and I don't see the reasoning behind the decision."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# df2 = df.copy()\n# cdf=df2[['ENGINESIZE','CYLINDERS','FUELCONSUMPTION_COMB','CO2EMISSIONS']]\n# #Split the data keeping 80% as training and 20% as testing data\n# msk=np.random.rand(len(df))<0.8  # array with boolean values.\n# train2=cdf[msk]  # Selecting the ros that are True\n# test2=cdf[~msk]  # Inverting the selection",
      "execution_count": 17,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Investigate correlations between attributes\nCompute the linear correlation between each attribute and the target value, using the corr() method. Plot also histograms to visualise the correlation between them."
    },
    {
      "metadata": {
        "scrolled": false,
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Note! I dont see the point of histograms here, they dont show the correlation, but distrubution. I guess it was an error.\n\ntrain.hist(bins=50, figsize=(14,10));",
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 1008x720 with 9 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "train.corr()",
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 19,
          "data": {
            "text/plain": "                          ENGINESIZE  CYLINDERS  FUELCONSUMPTION_CITY  \\\nENGINESIZE                  1.000000   0.934908              0.835716   \nCYLINDERS                   0.934908   1.000000              0.797885   \nFUELCONSUMPTION_CITY        0.835716   0.797885              1.000000   \nFUELCONSUMPTION_HWY         0.779448   0.722597              0.965585   \nFUELCONSUMPTION_COMB        0.821930   0.776898              0.995514   \nFUELCONSUMPTION_COMB_MPG   -0.816321  -0.776885             -0.939937   \nCO2EMISSIONS                0.874302   0.847746              0.896145   \n\n                          FUELCONSUMPTION_HWY  FUELCONSUMPTION_COMB  \\\nENGINESIZE                           0.779448              0.821930   \nCYLINDERS                            0.722597              0.776898   \nFUELCONSUMPTION_CITY                 0.965585              0.995514   \nFUELCONSUMPTION_HWY                  1.000000              0.985767   \nFUELCONSUMPTION_COMB                 0.985767              1.000000   \nFUELCONSUMPTION_COMB_MPG            -0.901552             -0.933554   \nCO2EMISSIONS                         0.860135              0.890314   \n\n                          FUELCONSUMPTION_COMB_MPG  CO2EMISSIONS  \nENGINESIZE                               -0.816321      0.874302  \nCYLINDERS                                -0.776885      0.847746  \nFUELCONSUMPTION_CITY                     -0.939937      0.896145  \nFUELCONSUMPTION_HWY                      -0.901552      0.860135  \nFUELCONSUMPTION_COMB                     -0.933554      0.890314  \nFUELCONSUMPTION_COMB_MPG                  1.000000     -0.907383  \nCO2EMISSIONS                             -0.907383      1.000000  ",
            "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>ENGINESIZE</th>\n      <th>CYLINDERS</th>\n      <th>FUELCONSUMPTION_CITY</th>\n      <th>FUELCONSUMPTION_HWY</th>\n      <th>FUELCONSUMPTION_COMB</th>\n      <th>FUELCONSUMPTION_COMB_MPG</th>\n      <th>CO2EMISSIONS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>ENGINESIZE</th>\n      <td>1.000000</td>\n      <td>0.934908</td>\n      <td>0.835716</td>\n      <td>0.779448</td>\n      <td>0.821930</td>\n      <td>-0.816321</td>\n      <td>0.874302</td>\n    </tr>\n    <tr>\n      <th>CYLINDERS</th>\n      <td>0.934908</td>\n      <td>1.000000</td>\n      <td>0.797885</td>\n      <td>0.722597</td>\n      <td>0.776898</td>\n      <td>-0.776885</td>\n      <td>0.847746</td>\n    </tr>\n    <tr>\n      <th>FUELCONSUMPTION_CITY</th>\n      <td>0.835716</td>\n      <td>0.797885</td>\n      <td>1.000000</td>\n      <td>0.965585</td>\n      <td>0.995514</td>\n      <td>-0.939937</td>\n      <td>0.896145</td>\n    </tr>\n    <tr>\n      <th>FUELCONSUMPTION_HWY</th>\n      <td>0.779448</td>\n      <td>0.722597</td>\n      <td>0.965585</td>\n      <td>1.000000</td>\n      <td>0.985767</td>\n      <td>-0.901552</td>\n      <td>0.860135</td>\n    </tr>\n    <tr>\n      <th>FUELCONSUMPTION_COMB</th>\n      <td>0.821930</td>\n      <td>0.776898</td>\n      <td>0.995514</td>\n      <td>0.985767</td>\n      <td>1.000000</td>\n      <td>-0.933554</td>\n      <td>0.890314</td>\n    </tr>\n    <tr>\n      <th>FUELCONSUMPTION_COMB_MPG</th>\n      <td>-0.816321</td>\n      <td>-0.776885</td>\n      <td>-0.939937</td>\n      <td>-0.901552</td>\n      <td>-0.933554</td>\n      <td>1.000000</td>\n      <td>-0.907383</td>\n    </tr>\n    <tr>\n      <th>CO2EMISSIONS</th>\n      <td>0.874302</td>\n      <td>0.847746</td>\n      <td>0.896145</td>\n      <td>0.860135</td>\n      <td>0.890314</td>\n      <td>-0.907383</td>\n      <td>1.000000</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "correlation_by_CO2EMISSIONS = train[train.columns[1:]].corr()[\"CO2EMISSIONS\"][:-1]\n\nprint(f\"Number of features correlated with CO2EMISSIONS: {len(correlation_by_CO2EMISSIONS.index)}\")\nprint(correlation_by_CO2EMISSIONS.sort_values(ascending=False))",
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Number of features correlated with CO2EMISSIONS: 6\nFUELCONSUMPTION_CITY        0.896145\nFUELCONSUMPTION_COMB        0.890314\nENGINESIZE                  0.874302\nFUELCONSUMPTION_HWY         0.860135\nCYLINDERS                   0.847746\nFUELCONSUMPTION_COMB_MPG   -0.907383\nName: CO2EMISSIONS, dtype: float64\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "plt.figure(figsize=(8, 8))\n\nax = sns.heatmap(\n    train.corr(),\n    vmin=-1, vmax=1, center=0,\n    cmap=sns.diverging_palette(20, 220, n=200),\n    square=True,\n    annot=True,\n    linewidths=0.5,\n    cbar_kws={\"shrink\": .82},\n    fmt=\".2f\",\n    annot_kws={\"size\": 10}\n)\n\nax.set_xticklabels(\n    ax.get_xticklabels(),\n    rotation=45,\n    horizontalalignment='right'\n);",
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 576x576 with 2 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "train.corr().columns",
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 22,
          "data": {
            "text/plain": "Index(['ENGINESIZE', 'CYLINDERS', 'FUELCONSUMPTION_CITY',\n       'FUELCONSUMPTION_HWY', 'FUELCONSUMPTION_COMB',\n       'FUELCONSUMPTION_COMB_MPG', 'CO2EMISSIONS'],\n      dtype='object')"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "scrolled": false,
        "trusted": true
      },
      "cell_type": "code",
      "source": "for feature in correlation_by_CO2EMISSIONS.index:\n    train.plot(kind=\"scatter\", x = feature, y = \"CO2EMISSIONS\")",
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 432x288 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 432x288 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 432x288 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 432x288 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 432x288 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 432x288 with 1 Axes>",
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEICAYAAACwDehOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAA2GklEQVR4nO3de3xU5bno8d+TEMJVwYCBEiho8IIKKTtFEGWLVLGiYiv10lK0R4v7bNsq1YpUd2s9UMFaUXd3PbXaKrWtUrCCdxTweANtxBAEVCIESeQaAbmGhDznj7UmTCZrzSRk1sxk5vl+Pvlk5llrZt41hHnnvT2vqCrGGGMMQFayC2CMMSZ1WKVgjDGmgVUKxhhjGlilYIwxpoFVCsYYYxpYpWCMMaZBuyCfXEQqgD3AYaBOVYtF5DjgaaA/UAFcoao7RUSAB4GLgP3Ataq6Itrz9+jRQ/v37x9Y+Y0xJh29//77O1S1p9exQCsF12hV3RF2/3ZgsarOFJHb3ftTgW8CA92fM4GH3d+++vfvT0lJSTClNsaYNCUiG/2OJaP7aDzwhHv7CeCysPgcdSwHuolI7ySUzxhjMlbQlYICi0TkfRGZ7MbyVXWze3sLkO/e7gNsCntspRszxhiTIEF3H52tqlUicjzwqoh8FH5QVVVEWpRnw61cJgP069cvfiU1xhgTbEtBVavc39uAfwLDgK2hbiH39zb39Cqgb9jDC9xY5HM+oqrFqlrcs6fnOIkxxpijFFilICKdRaRr6DZwAfAhsBC4xj3tGmCBe3shMEkcw4HdYd1MxhhjEiDI7qN84J/OTFPaAX9T1ZdF5F/AXBG5DtgIXOGe/yLOdNRynCmpPwiwbC1WvbeGyp0HKOjekbwuuckujjHGBCKwSkFV1wNDPOLVwBiPuAI3BlWe1lhQWsXU+WXkZGVRW1/PvZcP5tIiGwM3xqQfW9EcQ/XeGqbOL+NgbT17auo4WFvPbfPLqN5bk+yiGWNM3FmlEEPlzgPkZDV+m3KysqjceSBJJTLGmOBYpRBDQfeO1NbXN4rV1tdT0L1jkkpkjDHBsUohhrwuudx7+WA65GTRNbcdHXKyuPfywTbYbIxJS4nIfdTmXVrUh5GFPWz2kTEm7Vml0Ex5XXKtMjDGpD3rPjLGGNPAKgVjjDENrFIwxhjTwCoFY4wxDaxSMMYY08AqBWOMMQ2sUjDGGNPAKgVjjDENrFIwxhjTwCoFY4wxDaxSMMYY0yDwSkFEskXkAxF53r3/uIhsEJFS96fIjYuIPCQi5SJSJiJDgy6bMcaYxhKREO8mYC1wTFjsZ6o6L+K8bwID3Z8zgYfd38YYYxIk0JaCiBQA44BHm3H6eGCOOpYD3USkd5DlM8YY01jQ3UcPALcB9RHxGW4X0WwRCeWj7gNsCjun0o0ZY4xJkMAqBRG5GNimqu9HHJoGnAJ8HTgOmNrC550sIiUiUrJ9+/b4FNYYYwwQbEthJHCpiFQATwHniciTqrrZ7SKqAf4MDHPPrwL6hj2+wI01oqqPqGqxqhb37NkzwOIbY0zmCaxSUNVpqlqgqv2Bq4AlqjoxNE4gIgJcBnzoPmQhMMmdhTQc2K2qm4MqnzHGmKaSsR3nX0WkJyBAKfAfbvxF4CKgHNgP/CAJZTPGmIyWkEpBVV8HXndvn+dzjgI3JqI8iVK9t4bKnQco6N7R9nc2xrQJyWgpZIQFpVXcNq+M7CzhcL3ymwmDubTIJlMZY1KbpbkIQPXeGm79x0pq6urZf+gwNXX13PKPlVTvrUl20YwxJiqrFAKw+vPd1B7WRrHaw8rqz3cnqUTGGNM8VikEQloYN8aY1GCVQgA65Xi/rX5xY4xJFfYpFYCK6v0tihtjTKqwSiEARX27tShujDGpwiqFABTmd2XSiH6NYpNG9KMwv2uSSmSMMc1jlUJQ1P9QyYZq7l/0MSUbqhNXHmOMaQZbvBaA8q17mLP8s0axOcs+Y9Lw/tz13GreKncqg4eWlHNOYR5/uX54MoppjDFNWEshAKWbdnnG//lBZUOFEPJmebW1GIwxKcMqhQD4DSjvPlDnGX9j3Y4AS2OMMc2XkZVC+dY9zCvZRPnWPYE8v99A82VFX/E8f9TAHoGUwxhjWirjxhR+8eyqRv39k0b04+7xZ8T9de4efwaThvendNMuivp2ozC/q28l1K1T+7i/vjHGHI2Main4DQAH2WKYUNy3YSqq31iDX9wYYxItoyqFt8q9++794vFmi9qMMakuoyqFHl28u2n84sYYk2kCrxREJFtEPhCR5937A0TkXREpF5GnRaS9G89175e7x/vHuywjTuxBVkSi0ixx4onwyuotLYobY0yiJaKlcBOwNuz+LGC2qhYCO4Hr3Ph1wE43Pts9L67yuuTywJVFtM8Wcttl0T5beODKogRulem3zDnK8mdjjEmgQGcfiUgBMA6YAfxURAQ4D/iue8oTwF3Aw8B49zbAPOB3IiLu3s1xc2lRH0YW9kjK3slnDsgD1vnEjTEm+YKekvoAcBsQygSXB+xS1dAqrkogtHFxH2ATgKrWichu9/y4jwLndclNaGUQktMuG6Fxu0DcuDHGpILAuo9E5GJgm6q+H+fnnSwiJSJSsn379ng+deA6t89u0lGkbtwYY1JBkGMKI4FLRaQCeAqn2+hBoJuIhFooBUCVe7sK6AvgHj8WaJIUSFUfUdViVS3u2bNngMWPv893H4gat+ypxphkC6z7SFWnAdMARORc4FZV/Z6I/AOYgFNRXAMscB+y0L2/zD2+JN7jCcnnv3fzxEeXW/ZUY0zSJWOdwlScQedynDGDx9z4Y0CeG/8pcHsSyhao075yDO0i3vF2WVBbd9iypxpjUkJCKgVVfV1VL3Zvr1fVYapaqKrfUdUaN37QvV/oHl+fiLIlUl6XXO6/oojcdkKnnGxy2wn3X1HEysrdnudb9lRjTKJlXEK8ZPOaErt4jffitSEFxya4dMaYTGeVQhJETont0bUDWQL1YSMoWeLEjTEmkTIq91Gq6tw+u1GFAE4FYVNVjTGJZpVCCog1VdUYYxLFKoUUsGLjzhbFjTEmKFYppIAnl21sUdwYY4JilUIK2HmgtkVxY4wJilUKKWDwV45pUdwYY4JilUIKuOPiQTHj1XtrWLlpF9V7axJVLGNMBrJ1ChGq99YkfK+Fiur9vvHiAXksKK3i1rmlKE72pN9eUcSlRX08H2OMMa1hlUKYBaVVTJ1fRk5WFrX19dx7+eCEfPgW9e3mG6/eW8NNT5U2iv/kqVJGFvZIyp4Qxpj0Zt1Hruq9NUydX8bB2nr21NRxsLae2+aXJaS7pjC/K5NG9GsUmzSiH4X5XXlmRaXnY/zixhjTGi1uKYhId5zd09IqrXXlzgPkZGVxkPqGWE5WFpU7DyTkG/nd489g0vD+lG7aRVHfbhTmO5vVvfHxNs/z3/h4Gz8cdWLg5TLGZJaoLQUR+YWInOLezhWRpcCnwFYR+UYiCpgoBd07su9QXaPYvkN1FHTvmLAyFOZ3ZUJx34YKAaBbp/ae5/rFjTGmNWJ1H10JfOzevsb93RP4d+DXQRUqGTZs3+uZf2jD9r3JKZBrzKnHtyhujDGtEatSOBTWTTQWeEpVD6vqWtJskNpv74Jk72mwbpt3peQXN8aY1ohVKdSIyOki0hMYDSwKO9YpuGIl3qiBPaLGy7fuYV7JJsq37klksVhd9WWL4sYY0xqxvu3fDMzD6TKaraobAETkIuCDYIuWWMUD8jinMI83w7bFPKcwj+IBefzi2VXMWf5ZQ3zSiH7cPf6MhJSret+hFsWNMaY1olYKqrocOMUj/iLwYrTHikgH4A0g132dear6SxF5HGdMIrQH5bWqWioiAjwIXATsd+MrWnY5rfOX64dTsqGaN9btYNTAHhQPyKN8655GFQLAnGWfMWl4/0YDwkHZV1PXorgxxrRG1EpBRCZFOayq+pcox2uA81R1r4jkAG+JyEvusZ+p6ryI878JDHR/zgQedn8nVPEAp3UQUrppl+d5pZt2JaRSqPOZ+RuKJ2MFtjEmfcXqPvq6T/xSoA/gWym4A9Sh0dAc9yfa2obxwBz3cctFpJuI9FbVzTHKGKhoq40TYWi/7nz2RdPNdob2686C0ipuefpI+ov7r7T0F8aY1ok60KyqPw79AD8B3gXOBZYDQ2M9uYhki0gpsA14VVXfdQ/NEJEyEZktIqGvt32ATWEPr3RjSdW9c3uypHEsS5x4IvxodKFnfOKZ/bjpqVLqFA4r1KmT/sIS5hljWiNmmgsRaSci1wNrgW8AE1T1SlUti/VYd/pqEVAADBOR04FpOOMUXweOA6a2pMAiMllESkSkZPv27S156FFxVjo3rhVysoTKnYnZKtMvBcYHPt1alv7CGNMasVY03wisAf4NuFBVr1XVj6M9xouq7gKWus+xWR01wJ+BYe5pVUDfsIcVuLHI53pEVYtVtbhnz54tLUqLdW6fTc3hxr1eNYeVzu2zA3/tkLvHn8FrU0Zx34TBvDZlFHePP4Nl673XT/jFjTGmOWK1FP4bOAY4G1jodvmUicgqEYnaUhCRniLSzb3dETgf+EhEersxAS4DPnQfshCYJI7hwO5kjycAvFD2eYviQYlMgZHftYPneX5xY4xpjlgDzQNa8dy9gSdEJBun8pmrqs+LyBJ3MZwApcB/uOe/iDMdtRxnSuoPWvHacfOmz4rmN9ft4KbzT05waY4YWdiDv/+raVfRyELvRXjGGNMcsdYpbAQQkQHAaW54jaquj/XE7pjD1zzi5/mcr8CNsZ430Qbmd6Hks12e8WTqmOPdfeUXN8aY5og1pnCMiMwFFgP/y/15TUT+ISIZsYHwdWefEDWerPQXDy0pb1HcGGOaI1b30UM4A81XqWo9NIwF/BfwOyDa4ra0EJr9M2dZ4zQXhfldk5r+onybd+4jv7gxxjRHrEphpKpeGx5wu3nuFpF1gZUqxXhtgJPs9Bf5XXNZX33QMw5OCyZywx5jjImlNemvJfYp6aMwv2ujD9e3yr0HoN8q35GQD+Hvntmf6S9+5BlPZgvGGNO2xZqS+o67+1qjCkBE/gtYFlyxUl8PnzxDfvF4+9bQAs94Ud9uni2YRI95GGPapliVwo+BM4ByEZnv/nwKDHGPZawRJ+aRHbHSOTtLGHFins8j4iuvSy4PXVVETpbQLstZZf3QVUVUVO/3PN8vsV9rVe+tYeWmXZZew5g0EWtK6pfAd0TkRGCQG16jqp8GXrIUl9cll9lXDOGWuStRFEH47RVDEpqp9NKiPows7NEoS+qzKzZ5ntsuZkKTlltQWsXU+WXkZGVRW1/PvZcPtoR8xrRxsaakDhWRocCxOCknqoBjw+IZraTiC2rrlbp6qK1XSjZ+kfAy5HXJZUjfbg2VUbSxjniq3lvD1PllHKytZ09NHQdr67ltfpm1GIxp42INNP82yjEFPBeiZYJkzz7ys93nQ9kvfrScRIFZHKS+IZaTlUXlzgO2r4MxbVis7qPRiSpIW/NWuXeG1rfKtye1Uth/8HCL4keroHtHauvrG8Vq6+sp6N4xrq9jjEmsWN1HXxeRXmH3J4nIAhF5SESOC754qatHF+/Ec37xRPHbxSg8/uyKTVz/xL98xx+aI69LLvdePpgOOVl0zW1Hh5ws7r18sLUSjGnjYnUf/QFnDwVEZBQwE2fWURHwCDAhyMKlMr9ZRomafeTn4KHaqPHhv36VLV8eAuC1tduY9fJHLPv5+Uf1Wl4D3caYti3WnJRsVQ2Nnl4JPKKq81X1vwDvLcEyxM59h1oUT5TyHd5TUst37OfZFZsaKoSQzV8eanWLIXyg2xjTtsWsFEQk1JoYAywJO9aa1dBtnt+8/6DWAzTXqELvlsqowjyeLvHelc0vbozJPLEqhb8D/09EFgAHgDcBRKQQ2B1w2VJaUd9uLYonyiPXnukbzz/G+9u8X9wYk3miVgqqOgO4BXgcONtNhhd6XEavaPbbOzkVks9VzBzHBaf0oEM74YJTelAxcxwAB2u8ZyD5xY0xmSdqF5CIdALeV9Va9/7JOLujbVTVZxJQvpTmlT01VXi1GNZs8U6r7Rc3xmSeWN1HLwP9oaHLaBlwAnCjiNwT7YEi0kFE3hORlSKyWkR+5cYHiMi7IlIuIk+LSHs3nuveL3eP92/ltSVE5N7JqSy/a/uo8Sff2cB3/u87PPnOhkQWyxiTQmJVCt1VNbRvwjXA31X1x8A3gYtjPLYGOE9Vh+BMYb1QRIYDs4DZqloI7ASuc8+/Dtjpxme755k42vjFAd/4kLte5s6Fa/hXxU7uXLiGIXe9nODSGWNSQaxKIXzN03nAqwCqegio93xE6IGOve7dHPcnlBpjnht/ArjMvT3evY97fExkym7TOnmdvVsKIOyOWPG8++BhazEYk4FiVQplInKfiEzBWZewCEBEujXnyUUkW0RKgW04FcqnwC5VrXNPqQRCaTX7AJsA3OO7geSuBEszt449xTOek+1d9/71vc8848aY9BWrUvghsANnXOECVQ2tjBoE3BfryVX1sKoWAQXAMMD7U6kFRGSyiJSISMn27d75h1LF7FfWcu5vljL7lbXJLgoAYwb14uT8zo1iJ+d35nC9d6Nv575gMp7aHgzGpC45Mss04BcS+QXOWoepQC9VrROREcBdqjpWRF5xby9zF8xtAXpqlAIWFxdrSUlJQsrfUgOnvUBtWMlzBNbdMy55BQqzeM0WFq3ZygWD8hkzqBeDf/kyX3pMSz0mN5uyX10Y19e2PRiMST4ReV9Vi72OxUqIt0pEyvx+Yjy2Z6ibSUQ6AucDa4GlHMmZdA2wwL290L2Pe3xJtAohlc1+ZW2jCgGgVmloMST7m/KYQb2YNWEIYwY5uQ6/cqz34jW/+NGyPRiMSX2xUlXEmmEUTW/gCRHJxql85qrq8yKyBnhKRKYDHwCPuec/BvxFRMqBL4CrWvHaSbWgbItv/IT8Y7htXhmCM+r+mwnJ/6b86TbvfEnh8WdXbOL5VVu4+IxeXDa071G9ju3BYEzqi7WfwsajfWJVLQO+5hFfjzO+EBk/CHznaF8vlYwf3IsHl65vEj//lJ78dO5KDtcfaUZMmbuSkYU9kvqh6J1X9Ug8XplVY+3BUL23xjKuGpNksbqP9ojIl2E/e8J/J6qQbc2UsaeSEzGhJ0dgSL/jGlUIAIfrlWWfViewdE11bOc9+6hjO4lrZtVoezAsKK1i5KwlTHz0XUbOWsLC0qqjuhZjTOvEmn20GFgDTAdOV9WuqnpM6HfwxWu71t0zjptGn0D/vE7cNPoE1t0zjrWbd3me6xdPlJvHDPSNP/z6p57H/OKxXFrUh7ennseT15/J21PP49KiPjbWYEwKiZUQ7zJgLLAd+KOI/D8R+c9M33WtuaaMPZXXfzaaKWNPBaBT+xzP8/ziiXLD6IFNWgsd2wk3jB7IoXrvsX6/eHNE7sEQGmsIFxprMMYkVqyWAqq6W1X/jJPa4g/A3cC1AZcrLY09rVeL4om0dvpFTBt7Eqf27sq0sSexdvpFAGT7bPDpFz8aBd07su9QXaPYvkN1tt+zMUkQc6McETkLuBo4B3gL+Jaqvhl0wdJRd580E37xRLth9EBuGN24K+nTHd7f1kPx2a+sZUHZFsYP7tXQImqpnfsOEdnwqFcnbgPOxiRWrIHmCuD3QBUwGfgTsE9EhorI0OCLl15Wf+49Nu8XTwV+fyBZOAv0Hly6norq/Ty4dD0Dp71wVK+RqrvYGZOJYrUUKnCm0491f8KFktuZZvrEZ9+CT7Z8yaiTelK+dU/K7c1wXOcctu9rOmk1Jxsicug1LNBraYshVXexMyYTxVqncG6CypERtu7xnk2zdU8Nv3h2FXOWH0lAN2lEP+4ef0aiiubr+rMHcM8rnzSJa8Pyu8aeWLaxxZVCaBe7OcsaX3+qVIzGZJJY3Ue3hd3+TsSxXwdVqHQ1dlC+Z/y03l0bVQgAc5Z9RvnWPYkoVlR+M5O65GZ7np+ddXTZzu8efwavTRnFfRMG89qUUSlRIRqTiWLNPgpPNTEt4lh8M6VlgOIBeZxT2Dgb+DmFeVTtOuh5/iurvdNlJJrXzKTex3gPAPvFm+NodrFLdh4pY9JNrDEF8bntdd80w1+uH07JhmreWLeDUQN7UDwgj2nzV3qeW7nTOydRMkTOTCrfsc/zvFD8zmdW8tLqrXzztHymf3tIIGWyjKvGxF+sSkF9bnvdN81UPCCP4gFHWgy1h733M/CLpwK/XqIsgf63H5mF9OR7lTz5XiUVM+ObNjx8FXQowd5t88uSnkfKmLYuVvfRkFCuI2BweA4kwDp94+S4zt4fYn7xVNCpvff3ibqm2zIATsshnmwVtDHBiJXmIjss11E793bofnJzM6QR9Wl0+cVTwfeG9fOM+7UgXlq9Na6vHyvjqjHm6MRMc2GCN3aQT/oLn3gq8MsEe3xX79ZN55z4/qlFy7hqjDl6UccURGQw8AjQB3gJmKqqO91j76lqk30RTMt16+Sd5iI8nop7Day7Z1yTNBfh4wnhPtsV/9lBlxb1YWRhj5R7X4xpy2INNP8euAtYDlwPvCUil6rqp4B1H8VJtDQPhfldWVBaxc/+sRIRQVW57ztDUmaWzZSxp7Z4sdrkx9/ljfJqRhXm8ci1Z7bq9fO65FplYEwcxWrTd1XVl1V1l6reB/wIeFlEhhNj9pGI9BWRpSKyRkRWi8hNbvwuEakSkVL356Kwx0wTkXIR+VhEItNqpK1oaR6q99Yw5elSDh1WaurqOXRYufnp0jY7L7//7S+w6KMdHKxTFn20w7dlYYxJjpgdvSJybOi2qi4FLgf+Anw1xkPrgFtUdRAwHLhRRAa5x2arapH786L7OoNwFsudhrMw7vfu/s5pL5TmIVwozcOyT3d4ZhBd9umOBJaw+fz+wbJxWghe/OLGmMSL1X00CzgVp/sIcPZeFpExwH9Fe6CqbgY2u7f3iMhanLEJP+OBp1S1BtggIuU4ezkvi3kVaeDu8WcwaXj/Jgnx1m72TqK3dvOXXDwkNbqQwl09rIAn36v0jP/VIw7w6kepWcEZk4liTUn9m6ouBxCRLiLSxY1/pqo/bO6LiEh/4GtA6Cvhj0SkTET+JCLd3VgfIHzj30qiVyJpxyvNQ/W+Q57n+sWTzW/18vRvD8FvAlKcJyZFZWkxjImuOd1H/1tEPgM2Ap+JyEYR+c/mvoBbkcwHblbVL4GHgROBIpyWxG9bUmARmSwiJSJSsn379pY8tE1q5zPx3y+eCipmjmPisALyOucwcVhBw2rmIQXe23qH4rc+vYIhd73MrU+vCKRcC0qrGDlrCRMffZeRs5awsLQqkNcxpi0TVf/xYhG5EzgL+JGqrndjJwAPAu+q6vSoTy6SAzwPvKKq93sc7w88r6qni8g0AFW9xz32CnCXqvp2HxUXF2tJSUn0K2zjSjZUM+EPy5vE590wvFGqjLZg8C9f4suapqk7jsnN8ozHMzVG9d4aRs5awsHaI6/TISeLt6eeZ7OXTMYRkfdVtdjrWKyWwveBb4cqBAD39hXApBgvKsBjwNrwCkFEeoed9i3gQ/f2QuAqEckVkQHAQOC9GOVLe36ZVdtahQBQU+udy8mrQgDi2mKwtBjGNE/MhHiq2iSvs6oeEJFY2dpG4lQqq0Sk1I39HLhaRIpwprRWADe4z7laROYCa3BmLt2oqj6ZdDKLV2bVtuiSIb2Z98HmZp//zAebue/K+Ly2pcUwpnliVQpVIjJGVReHB93ZR1H/d6vqW3in134xymNmADNilCkjRWZWDUnFlc5+7rtyKPM+aLouIQsIOh9sKC3GbRGptlP9PTMm0WJVCj8BFojIW8D7bqwYpxUwPsiCmdja4n4CFTPHcevTK3h17TbOP/V47rtyKKNmvuaZBqOgm/OBfe2jy3hnwxecNeA4Hr9+xFG/tqXFMCa2WAPNhUAv4CScRWXgdO98DGx2010kTSYMNPtJp4HTk+94gRqPjsLcbDzj8d6bwZhM05qB5geA3ar6J1W9xf15DNjtHjNJEmvgtHzrHuaVbEqJfZ5jEZ/ZtV4VAjgth0RqS++lMa0Vq/soX1VXRQZVdZU7ndQkSbSB0188u4o5yz9riE8a0Y+7x6funkjfP/Or/PHtjc0+//XyLxpuB73tZ1t7L41prVgthW5Rjtm0jSTy209g575DjT7EAOYs+yylv+XeccnpTf4Qm7PIuf/tL/Dke5VU76vlyfcq455cr3zrnqjvpa2ONukoVkuhRER+qKp/DA+KyPUcGXg2SeI1cDqvZJPnuaE03Klq/cxxzHjuQ57/cAsXn96LOy45PeqHvN/2nnc+szJuLYZoKc1Xb/4yZdOZG9MasSqFm4F/isj3aDz7qD3OwjOTZJH7CURLw53q7rjkdO645PSG++cWHteoqyg87re950urtzL92/Epj9971j+vE1c8stzNXutM1Lj56VJGFvZoc4P8xkSKlRBvq6qeBfwKZ6FZBfArVR2hqluCL55pqe6d2zfZJzlLnHhb4zf99PHrR6D13qPQfvGj4ZfSfMuXB9tUOnNjWiJWSwFo2EdhacBlMXGw+vPdnh9Yqz/fzaiTjk9OoVqhYuY4z3UKXxzwXu7mFz9aXinNH397g+e5O/amZuZaY1qiWZWCaUv8sqemblbVWI5mwVrk3tGtUZjftdF4zNmFPTzP84sb05YkMJO9SYTTvnIM7SL+VdtlOfF04veHmwUMnPYCDy5dT0X1fh5cup6B0+I7KynaTnnGtHVWKaSZvC653H9FEbnthE452eS2E+6/oijtBkBHFR7nGS/olkttRPdZrToth3i6e/wZvDZlFPdNGMxrU0bZ2gWTNqKmuUh1mZzmIpa2lCjvaHlNWe2YDQc8xpo7ZsPaGU2nvRqTiaKlubBKwbRpkYPQ0dY2RGZjzcJZH2FMpolWKdhAs2nTIgehu+Rmsddj055sILIBUQ/MeO5DazEYE8bGFExa+fG5hZ5xv9ULfw7LufTkOxv4zv99hyff8Z5yakwmsJaCSSs3jB7IA4vXcaDuSLdox3bS6H64Ovf3kLteZvdBp+r4V8VOfrPoY1bedWHQxTUm5QTWUhCRviKyVETWiMhqEbnJjR8nIq+KyDr3d3c3LiLykIiUi0iZiAwNqmwmva2dfhHTxp7Eqb27Mm3sSaydflHUKaxPvrOhoUII2X3wsLUYTEYKsvuoDrhFVQcBw4EbRWQQcDuwWFUHAovd+wDfBAa6P5OBhwMsm0lzN4weyEs3jeKG0QMB+O6wAs/zvjusgAVl3jvL+sWNSWeBVQqqullVV7i39wBrgT4423g+4Z72BHCZe3s8MEcdy4FuItI7qPKZzOKXOXX6t4eQI95dS6H4tY8u46Q7Xkj45j7GJENCBprdDXm+BryLs3FP6CvYFiDfvd0HCM/7XOnGjImLipnjmDisgLzOOUwcVtCwrWfJxl2e55ds3EX/21/g9fIvOHTY2dwn3ns2tIbt52CCEPhAs4h0AeYDN6vqlxK296KqqojP1zT/55uM071Ev379YpxtTGPTvz2kSWrtwz459A75xK99dNlR5WOKpwWlVdw2rwzBSd79mwmDbT8HExeBthREJAenQvirqj7jhreGuoXc39vceBXQN+zhBW6sEVV9RFWLVbW4Z8+ewRXeZIyLTm9Z9tjQHg8znvuQEfe8xoznPgyiWL6q99bw07krqamr52BdPTV19UyZu7KhxWB7SpvWCKylIE6T4DFgrareH3ZoIXANMNP9vSAs/iMReQo4E9gd1s1kTGD+e+LXea6F3UIn3P5Cw+roP769kcfe3piw1dHLPq3mcER+9MP1yrJPq3lvQ7XtKW1aJciWwkjg+8B5IlLq/lyEUxmcLyLrgG+49wFeBNYD5cAfgf8MsGzGNFIxcxyXnH48ndtnccnpx1Mxcxzn+iTd63NMeyJ7lkKroxNhh88YwtrNu9vc/twm9QTWUlDVt/BP4j/G43wFbgyqPMbE8t8Tv97ovl8upe0+m+k8sWxjQ8qMZ1ds4vlVW7j4jF5cNrSv5/lHy2/fhk7tvf87h+/PXbKhmjfW7WDUwB4UD8iLa7lMerA0F8ZEEWoxtM929oaumDnOdwA6FB/+61e5eW4Zr63dxs1zyxjx61fjWia//RzGntbL8/zQXtMTH13OhD8s56El5Uz4w3K+/+jyuJbLpAdLc2FMDJEzjdpnec9Map/ltBC2fNm4JbH5y0M8u2ITlw3tG7cWhNc2oeBUDnOWNR5TKMzvSsmGat4qr270HG+WV1OyodpaDKYRS51tTAvNeO5D/hiWSC/khyO/yjMfVFK9v2n6vbxO2eS0y25UYfQ+pj3Lfn5+3MtXvnVPk8ri/kUf89CS8ibn/uS8Qn56wcmejzHpK1rqbOs+MqaF7rjk9Cb/cbLceGQOpZCd+w/7tiDirTC/KxOK+zb6cB810HscYtTAHvzi2VV8Y/Yb3DqvjG/MfoNfLFgV9zKZtsMqBWOOwvqZ4/jhyK/S+9hcfjjyqw3TUc87yfvDt3sn757a51dtAYJf81A8II9zCht3E51TmEe3Tu1txpJpxLqPjIkzrxlLpxzfiY+27feMf7Jtf8J2hIucfTSvZBO3zitrct59EwYzoTi+s6ZM6rDuI2MSqGLmOC44pQcd2gkXnNKDipnjWF/dtEIA+CiiQoCmax7imeOoeEAeP73g5IbB5dDMpEh+cZP+bPaRMQF45NozG90/a8BxDekxwkXuGx3ydEkld1xyOgtKq7h1bim4WY5+e0VRXHMchaa3es1YMpnJuo9MyqjeW0PlzgMUdO9IXpfcZBcn7ry6lY7vnMO2fbWe8Zem/Dv/Nv21Jsfev/MbcX9/bPZRZrHuI5PyFpRWMXLWEiY++i4jZy1hYWmTXIhtntdCuJ+MGeh57k/GDGTR6i2ex0LxPyxdxzcffIM/LF3X6rJ5zVgymclaCibpqvfWMHLWEg7WHulI6ZCTxdtTz0vLFkOk8P2hAY7tkM3Kuy7kx399n+dWNa0YLjmjF6+t3dpkH+q10y9KSHlN22ctBZPSKnce8Mz6WbnzQJJKlFgr77qQ6ZcO4uv9uzP90kGsvOtCAHJzsj3P37B9X6MKAeBAnTa0GBav2cLUeStZvMa7pWFMNDbQbJKutu4wtYcbf8jVHlZq65xvz+k+1gAw8awBTDxrQKPYVV/vy7wVTbvRtu7xriyffHcj80ur+GTrPsAZrD45vzOvTDk33sU1acwqBZN0FT7TNSuq91O1+yBT55eRk5VFbX09916eOTuMhRacvRmWs+icwjx27T/E9n1NF5dliTRUCCEfb93H4jVbGDPISZa3eM0WFq3ZygWD8htixoSzSsEknd+c+P55nZj4p/c4WFvPQXfi5m3zyxhZ2CNtWwyR/nL98CYLzhav2cJ1c95vcm6n9t7dTY+/XcGYQb24YPbr1oowMdmYgkk6v1TQOe2yyclq/Ceak5WVMWMNIZELzsYM6sXJ+Z0bnXNyfmffilJRFq/Z4tuKgPgukDNtm7UUTErwSgVdvbeG2vrGS7tq6+sp6N4xSaVMHa9MObdJV9DiNVt469OmC+R+MHIAi9Zs9XyeRWu2svfQ4YztojNNBdZSEJE/icg2EfkwLHaXiFRFbM8ZOjZNRMpF5GMRGRtUuUzqipwrn9cll3svH0yHnCy65rajQ04W914+OGO6jmIZM6gXsyYMaRgb8GtBjBnUiwsG5Xs+x4gTjmPq/DIO1tazp6aOg7X13Da/zFoMGSzIlsLjwO+AORHx2ap6X3hARAYBVwGnAV8BXhORk1TVOw+xyRiXFvVhZGGPtJ99FC9eLQg4UmF8HNaFdHJ+Zwb07EpOVlbDmA0c6aKz9zozBblH8xsi0r+Zp48HnlLVGmCDiJQDw4BlQZXPtB15XXLtA6oFxgzq5TmzyKvCsC46EykZA80/EpEyt3upuxvrA4TvNlLpxowxcRTZ5dSaLrqSDdXcv+hjSjZUxzzXtB2JHmh+GPg/gLq/fwv8r5Y8gYhMBiYD9OvXL8bZxphYjqaLbuKjyxv2fH5oSTnnFObxl+uHA5Zcr61LaKWgqg1TIETkj8Dz7t0qIHxHjwI35vUcjwCPgJP7KJiSGpNZWtJFV7KhuqFCCHmzvJqSDdUsXPl5o53cJo3ox93jzwCssmgrElopiEhvVd3s3v0WEJqZtBD4m4jcjzPQPBB4L5FlM8Y0zxvrdnjGny39nCffbbq156Th/ZmzrMK3sjCpJcgpqX/HGSg+WUQqReQ64F4RWSUiZcBoYAqAqq4G5gJrgJeBG23mkTGpadRA732oj+3o/R3zldVbYu4DbeMTqSPI2UdXe4Qfi3L+DGBGUOUxxsSHX06mb32tgP95fX2zn6d00y4K87tGHZ8wiWcrmo0xLeaVkwnw3Npz7Gm9+M2iT5o8R1HfblHHJ0LPaRLLKgVjzFEpHpDX5IPbK10JeFcWhflduX/Rx57P/ca6HVYpJIlVCsaYuCrM79pkdpFfZTFqYA8eWlLe5Dn8xi2aIxP23wiSVQrGmITwqiz8xieOtpWwoLTKkvu1klUKxpik8hufaKnqvTUNyf0ydf+NeLBKwRiTdF7jEy1VufNA1OR+tniueaxSMMakhYLuHX2T+/3i2VW+i+dsDKIx23nNGJMW/JL77dx3yHfx3ILSKkbOWsLER99l5KwlLCz1zK6TUaylYIxJG17J/eaVbPI8963yHcx8+SMbg4hglYIxJq1EJvcr6tvN87weXXJtgyEP1n1kjElrhfldmTSicZr9SSP6MeLEPNtgyIO1FIwxac9v8dy9lw/mtoh1DZncSgCrFIwxGcJr8ZztAd6UVQrGmIzWFvcAD3IarVUKxhjThgSdysMGmo0xpo0IT+Wxp6aOg7X13Da/jOq9NXF7DasUjDGmjQil8ggXmkYbL0Fux/knEdkmIh+GxY4TkVdFZJ37u7sbFxF5SETKRaRMRIYGVS5jjGmroqXyiJcgWwqPAxdGxG4HFqvqQGCxex/gm8BA92cy8HCA5TLGmDbJL5VHPAebg9yj+Q0R6R8RHg+c695+AngdmOrG56iqAstFpJuI9FbVzUGVzxhj2qKgp9EmevZRftgH/RYg373dBwhPUFLpxqxSMMaYCEFOo03aQLPbKtCWPk5EJotIiYiUbN++PYCSGWNM5kp0pbBVRHoDuL+3ufEqoG/YeQVurAlVfURVi1W1uGfPnoEW1hhjMk2iK4WFwDXu7WuABWHxSe4spOHAbhtPMMaYxAtsTEFE/o4zqNxDRCqBXwIzgbkich2wEbjCPf1F4CKgHNgP/CCochljjPEX5Oyjq30OjfE4V4EbgyqLMcaY5hHn87htEpHtOC2OaHoAOxJQnFSUydcOmX39mXztkNnX35xr/6qqeg7KtulKoTlEpERVi5NdjmTI5GuHzL7+TL52yOzrb+21W+4jY4wxDaxSMMYY0yATKoVHkl2AJMrka4fMvv5MvnbI7Otv1bWn/ZiCMcaY5suEloIxxphmSqtKoSV7OKQbEekrIktFZI2IrBaRm9x42l+/iHQQkfdEZKV77b9y4wNE5F13n46nRaR9sssaFBHJFpEPROR5934mXXuFiKwSkVIRKXFjaf93H+JmlZ4nIh+JyFoRGdGa60+rSoGW7eGQbuqAW1R1EDAcuFFEBpEZ118DnKeqQ4Ai4EI3XcosYLaqFgI7geuSV8TA3QSsDbufSdcOMFpVi8KmYmbC333Ig8DLqnoKMATn7+Dor19V0+oH6A98GHb/Y6C3e7s38HGyy5ig92EBcH6mXT/QCVgBnImzgKedGx8BvJLs8gV0zQXuf/zzgOcByZRrd6+vAugREcuIv3vgWGAD7vhwPK4/3VoKXvz2cEhb7uZGXwPeJUOu3+0+KcXJvPsq8CmwS1Xr3FNCe3SkoweA24DQPo15ZM61g5OCf5GIvC8ik91YRvzdAwOA7cCf3e7DR0WkM624/kyoFBqoU22m9XQrEekCzAduVtUvw4+l8/Wr6mFVLcL51jwMOCW5JUoMEbkY2Kaq7ye7LEl0tqoOxdnW90YRGRV+MJ3/7nHy1w0FHlbVrwH7iOgqaun1Z0Kl4LeHQ9oRkRycCuGvqvqMG86Y6wdQ1V3AUpwuk24iEkr66LtHRxs3ErhURCqAp3C6kB4kM64dAFWtcn9vA/6J86UgU/7uK4FKVX3XvT8Pp5I46uvPhErBbw+HtCIiAjwGrFXV+8MOpf31i0hPEenm3u6IM5ayFqdymOCelpbXrqrTVLVAVfsDVwFLVPV7ZMC1A4hIZxHpGroNXAB8SAb83QOo6hZgk4ic7IbGAGtoxfWn1eK18D0cgK04ezg8C8wF+uHu4aCqXySpiIERkbOBN4FVHOlb/jnOuEJaX7+IDAaeALJxvujMVdW7ReQEnG/PxwEfABNVtSZ5JQ2WiJwL3KqqF2fKtbvX+U/3bjvgb6o6Q0TySPO/+xARKQIeBdoD63H2o8niKK8/rSoFY4wxrZMJ3UfGGGOaySoFY4wxDaxSMMYY08AqBWOMMQ2sUjDGGNPAKgVjjDENrFIwvkTksJuOOPTTX0SuFZHfRZz3uogUu7fD0xiXishDbvxxEZng8RoniciLborfFSIyV0Ty3WNnuymxP3J/Joc97i4R2S8ix4fF9obdvsNNo13mluPMsPL1CDvv3LB009eKiIrIN8KOX+bGJoRd68fipOl+W0ROFpF/uq9RLiK7w679rIj35lgRmeOe96l7+1j3WH/3dX4c9tq/E5FrY/wb3eq+N6Ui8i8RmeTG24vIA+5rrRORBSJSEPY4FZEnw+63E5HtEe/Fdvd5V4uTmrlTlHLc5T5nYVjsZjcW+bdRJiKLRKSXG+8iIg+778kKcXIY/TDadZvgWKVgojmgTjri0E9FMx83OuwxP/E7SUQ6AC/g5G0Z6Oav+T3Q0/3A+BvwH+qkBD4buEFExoU9xQ7gFo/nHQFcDAxV1cHAN4BNzSz7KpyVwSFXAysjzvmeOmm6nwB+o6rfcvMuXQ+8GXbt70Q87jFgvaoWquqJONktHw07vg24SZq594GI/AfO6u1h7uuPwcmQCvBroCtwsjrpk58FnhGR0PF9wOnirADHfZ7IVBhPu9dxGnAIuDJGkSLfu+8AqyPOGe3+m5TgLK4E5z3YCYT+Bi7EWXRnksAqBZNM3wWWqepzoYCqvq6qHwI3Ao+r6go3vgMnE2h4sq8/AVeKSOQHSG9gR2gFr6ruUNXPm1mmN4FhIpIjTnLBQqDU59w33OMxud+g/w34P2Hhu4FiETnRvb8dJwX2NTTPz4H/HUp8qKpfquoT7jf6HwBTVPWwe+zPuPtOhD3+RSBUyV4N/N2n7O2Azjgf3NE8C4x3H3MisBun4vbyBlDonjcMuFNV692yblfVWTFeywTEKgUTTcewrpB/xj69wdKwx02Jct7pgF92z9M8jpW48ZC9OBXDTRHnLQL6isgnIvJ7Efn3FpRdgdeAsTgfcAujnHsJzrfj5hgElIY+pMHJ7IpT4YRf0yzgVhHJjvZkInIM0FVV13scLgQ+i8ySS9P37yngKrfFNhgnJUq4K8VJR16F8839OaL7EicPz+k4LYano5x7Mc57dxqwMlQhmOSzSsFEE9599C035pcXJTwe3n00O+AyPgRcI25SNABV3YvzrXwyzrfvp8P65r3KHxl7CudD7Sq8vz3/1f2wHAnc2prCNymI8yH/Lk4rKlCqWoazKdXVOK2GSE+73VK9cD7Af9aMpw29d5dxJCdRuKXue3cMcE/kQXcsqFREmtuyM3FmlYJpqWogcr/X4/DvJohmNc6Ht5c1Hsf+jYg+ajdV9t9wupvC44fdrqhfAj8CLncPRZa/SdlV9T3gDJzdvD7xKNv33ArvMlVt7ljFGqBIRBr+z7m3i9xj4X4NTOXI+EATbitgrzgJ4SJ9CvQLryhdTd4/nJbQffh0HbmvpTithFF+54R5Hvg+3i0VOPKFYZL7b7cGGBJ6X1R1hlsRHdOM1zIBsErBtNS/gJFhM0eKgVyaP5Ab7m/AWeGDxyIyyu1++B/gWnEyQCJO1stZwL0ez3M/cANOlkzcGUEDw44X4WSKBHgd50MLt4tmIk6a6Ui3c2QgtNVUtRwnW+mdYeE7gRXusfBzP8L5sLwkxtPeA/yP25UUmsUzSVX34QyC3x/qhnJnJXUClkQ8x5+AX6lqrG6ws3Eqm6hUdT9OhTYj1rnu+eU43VrTw8ragSgVoglWu9inGHOEqm4VkZuAF91vd3uBqyP6hJeKSKjvvExVJ7m3/yAiD7i3N6nqCHF2DnvAjdcCZcBN7utMBP7ofuMV4IHwQemwMu1wxzxC4xddgP8WZ4+FOqAcpysJnIHeh0VkpfucLwNPEkFVX2rZO9Ms17nlCn24LnNjXmbgVCLRPIxzrf8SkVqc9++37rFpOC2AT0SkHvgI+JZGpEVW1UqcLjgvV4qTkj0LZzOXa2OUJ/ScTzXnvDDXA78BykWkGjiAM6nAJIGlzjbGGNPAuo+MMcY0sO4jY1KciPwPzkyncA+6aw8SXZY7cBalhfuHqjZrDMGkPus+MsYY08C6j4wxxjSwSsEYY0wDqxSMMcY0sErBGGNMA6sUjDHGNPj/zbUIf5cPrHcAAAAASUVORK5CYII=\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Simple linear regression\n**A)** If you should use only one of the independent variables of the dataset to predict CO2 emissions, which one would that be and why?"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "FUELCONSUMPTION_CITY has the highest correlation with CO2EMISSIONS, but I would use FUELCONSUMPTION_COMB which has about the same score, but is a bit more varied. "
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**B)** Use sklearn packages to train a simple linear regression model with the chosen variable"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "train.plot(kind=\"scatter\", x=\"FUELCONSUMPTION_COMB\", y=\"CO2EMISSIONS\");",
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 432x288 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**Example of train_test_split for a linear regression:**\n\n```python\nX_train, X_test, y_train, y_test = train_test_split(\n    df.FUELCONSUMPTION_COMB,\n    df.CO2EMISSIONS,\n    test_size=.2,\n    random_state=42\n)\n```"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**Example of reshape:**\n\n```python\n# X = X.reshape(2, 5) # <-- 2 is the number of columns.\n# y = y.reshape(2, 5) # <-- 5 is the number of elements in each of those columns.\n# X = X.reshape(-1, 1) # <-- -1 is getting the index length of the column.\n# y = y.reshape(-1, 1) # <-- 1 is the amount of data points in each column.\n\n>>> X = np.arange(0,100)\n>>> y = np.arange(0,100)\n>>>\n>>> X.ndim\n    1\n>>> X.shape\n    (100,)\n>>> X = X.reshape(2,50)\n>>>\n>>> X.ndim\n    2\n>>> X.shape\n    (2,50)\n```"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Input for independent variable needs to be 2 dimmensional.\nX_train = train[\"FUELCONSUMPTION_COMB\"].values.reshape(-1, 1)\ny_train = train[\"CO2EMISSIONS\"]\n\nX_test = test[\"FUELCONSUMPTION_COMB\"].values.reshape(-1, 1)\ny_test = test[\"CO2EMISSIONS\"]",
      "execution_count": 25,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "display(X_train.shape)\ny_train.ndim",
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "(853, 1)"
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "execution_count": 26,
          "data": {
            "text/plain": "1"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Creating linear regression object\nmodel = LinearRegression()",
      "execution_count": 27,
      "outputs": []
    },
    {
      "metadata": {
        "scrolled": false,
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Training the model\nmodel.fit(X_train, y_train)",
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 28,
          "data": {
            "text/plain": "LinearRegression()"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**C)** Print the model’s coefTicients"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(\"Coefficient:\", model.coef_[0]) #[0][0])\nprint(\"Intercept:\", model.intercept_) #[0])",
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Coefficient: 16.180900781199195\nIntercept: 69.10302617988444\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**D)** Plot the linear regression line on the training data"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Using model to predict on the test data\ny_pred = model.predict(X_test)",
      "execution_count": 30,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "plt.plot(X_test, y_pred, label=\"Linear Regression\", color=\"r\")\nplt.scatter(X_test, y_test, label=\"Test Data\", color=\"b\");",
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<Figure size 432x288 with 1 Axes>",
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Some tests for specific prediction."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "round(model.predict(np.array([[15]]))[0], 4)",
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 32,
          "data": {
            "text/plain": "311.8165"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**E)** Evaluate your model using evaluation scores of a) Mean absolute error, b)\nResidual sum of squares and c) R2-score</font>"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "<br/>\nCalculating SSR (RSS), which is the Sum of Squares (Residuals) due to the Regression, and measures the difference between the predicted value and the mean of the dependent variable, or the variability left unexplained after performing the regression in other words.\n\nSSR Should be **low**."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "def SSR_score(X_test_y_pred, y_test, model=None):\n    if model:\n        ssr = ((model.predict(X_test_y_pred) - y_test.mean()) ** 2).sum()\n        return ssr\n    else:\n        ssr = ((X_test_y_pred - y_test.mean()) ** 2).sum()\n        return ssr",
      "execution_count": 33,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "ssr = SSR_score(X_test, y_test, model)\nprint(\"SSR:\", round(ssr, 4))",
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": "SSR: 700304.0871\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "<br/>Calculating SSE, which is the Sum of Squares Error, and measures the difference between the observed value and the predicted value, or the variability of the errors/residuals."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "sse = sum((y_test - y_pred)**2)\nprint(\"SSE:\", round(sse, 4))",
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": "SSE: 170651.011\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "markdown",
      "source": "<br/>Calculating SST, which is the Total Sum of Squares variance, and the squared differences between the observed dependent variable and its mean."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "sst = ((y_test - y_test.mean()) ** 2 ).sum()\nprint(\"SST:\", round(sst, 4))",
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": "SST: 884878.3364\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "<br/>\nCalculating the Mean Absolute Error (MAE), which is the mean of the absolute value of the errors.\n\nMAE should be **low**."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "mae = mean_absolute_error(y_test, y_pred)\nprint(\"MAE:\", round(mae, 4))",
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": "MAE: 20.4419\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "<br/>\nCalculating the Mean Squared Error (MSE), which is the mean of the squared errors.\n\nMSE should be **low**."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "mse = mean_squared_error(y_test, y_pred)\nprint(\"MSE:\", round(mse, 4))",
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": "MSE: 797.4346\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "<br/>Calculating the Root Mean Squared Error (RMSE), which is the deviation between values, therefore the lower this number is the better. We use it to compare y and ŷ.\nRSME should be as **low** as possible."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "rmse = np.sqrt(mean_squared_error(y_test, y_pred)) \nprint(\"RMSE:\", round(rmse, 4))",
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": "RMSE: 28.2389\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "<br/>\nCalculating r2, which is the coefficient of determination and it is the proportion of the variance in the dependent variable that is predictable from the independent variable, or variables when discussing polynomial regression.\n\nr2 indicates how scattered the sample values are compared to the regression line, and it should be as **high** as possible. High SSE (Sum of Squares errors(residuals)) values will result in a high r2 value, and vice versa."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "r2 = r2_score(y_test, y_pred)\nprint(\"r2:\", round(r2, 4))",
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": "r2: 0.8071\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "<br/>\nTesting scikit-learn's built in score function to score the model.\n\nThis function looks like it uses r2 for the evaluation, which would have a maximum of 1.0."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "scikit_score = model.score(X_test, y_test)\nprint(\"Scikit Score:\", round(scikit_score, 4))",
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Scikit Score: 0.8071\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "markdown",
      "source": "<br/>**Manual calculations:**"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "SST = ((y_test - y_test.mean()) ** 2 ).sum()\nSST",
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 42,
          "data": {
            "text/plain": "884878.3364485982"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "SSR = ((y_pred - y_test.mean()) ** 2 ).sum()\nSSR",
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 43,
          "data": {
            "text/plain": "700304.0871465029"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "SSE = ((y_test - y_pred) ** 2 ).sum()\nSSE",
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 44,
          "data": {
            "text/plain": "170651.01103608022"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "MSE = np.mean((y_test - y_pred)**2)\nMSE",
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 45,
          "data": {
            "text/plain": "797.4346310097206"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "RSME = np.sqrt(MSE)\nRSME",
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 46,
          "data": {
            "text/plain": "28.238885087937177"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "MAE = np.mean(abs(y_test - y_pred))\nMAE",
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 47,
          "data": {
            "text/plain": "20.441911472549577"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "R2 = 1 - (SSE/SST)\nR2",
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 48,
          "data": {
            "text/plain": "0.8071474868274242"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "np.sqrt(MSE)",
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 49,
          "data": {
            "text/plain": "28.238885087937177"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "SST = ((y_test - np.mean(y_test)) ** 2).sum()\nSST",
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 50,
          "data": {
            "text/plain": "884878.3364485982"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Multivariable linear regression\n**A)** If you should use a second independent variables of the dataset to predict CO2 emissions, which one would that be and why?"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "I would choose ENGINESIZE, due to correlation rank."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**B)** Use sklearn packages to train a multivariable linear regression model with the 2 chosen variables this time"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Input for independent variable needs to be 2 dimmensional, which they are in this case.\nX_train = train[[\"FUELCONSUMPTION_COMB\", \"ENGINESIZE\"]]\ny_train = train[\"CO2EMISSIONS\"]\n\nX_test = test[[\"FUELCONSUMPTION_COMB\", \"ENGINESIZE\"]]\ny_test = test[\"CO2EMISSIONS\"]",
      "execution_count": 51,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Creating linear regression object\nmodel = LinearRegression()",
      "execution_count": 52,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Training the model\nmodel.fit(X_train, y_train)",
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 53,
          "data": {
            "text/plain": "LinearRegression()"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(\n    f\"Coefficient:\\n\\t{X_train.columns[0]}:\\t{model.coef_[0]}\\\n    \\n\\t{X_train.columns[1]}:\\t\\t{model.coef_[1]}\")\nprint(\"Intercept:\", model.intercept_)",
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Coefficient:\n\tFUELCONSUMPTION_COMB:\t9.618439340589196    \n\tENGINESIZE:\t\t19.592827232761028\nIntercept: 79.60512760457624\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Using model to predict on the test data\ny_pred = model.predict(X_test)",
      "execution_count": 55,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**C)** Evaluate your new model using evaluation scores of a) Mean absolute error, b) Residual sum of squares and c) R2-score"
    },
    {
      "metadata": {
        "trusted": true,
        "scrolled": true
      },
      "cell_type": "code",
      "source": "ssr = SSR_score(X_test, y_test, model)\nprint(\"SSR:\", round(ssr, 4))\nsse = sum((y_test - y_pred)**2)\nprint(\"SSE:\", round(sse, 4))\nsst = ((y_test - y_test.mean()) ** 2 ).sum()\nprint(\"SST:\", round(sst, 4))\nmae = mean_absolute_error(y_test, y_pred)\nprint(\"MAE:\", round(mae, 4))\nmse = mean_squared_error(y_test, y_pred)\nprint(\"MSE:\", round(mse, 4))\nrmse = np.sqrt(mean_squared_error(y_test, y_pred)) \nprint(\"RMSE:\", round(rmse, 4))\nr2 = r2_score(y_test, y_pred)\nprint(\"r2:\", round(r2, 4))",
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "text": "SSR: 741137.3889\nSSE: 116325.6513\nSST: 884878.3364\nMAE: 17.0338\nMSE: 543.5778\nRMSE: 23.3148\nr2: 0.8685\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**D)** Compare the evaluation scores of the simple and multivariable linear models. Which one performs better? Can you further improve the performance of a regressor by considering additional attributes?"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Input for independent variable needs to be 2 dimmensional, which they are in this case.\nX_train = train[[\"FUELCONSUMPTION_COMB\", \"ENGINESIZE\", \"CYLINDERS\"]]\ny_train = train[\"CO2EMISSIONS\"]\n\nX_test = test[[\"FUELCONSUMPTION_COMB\", \"ENGINESIZE\", \"CYLINDERS\"]]\ny_test = test[\"CO2EMISSIONS\"]",
      "execution_count": 57,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Creating linear regression object\nmodel = LinearRegression()",
      "execution_count": 58,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Training the model\nmodel.fit(X_train, y_train)",
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 59,
          "data": {
            "text/plain": "LinearRegression()"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Using model to predict on the test data\ny_pred = model.predict(X_test)",
      "execution_count": 62,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true,
        "scrolled": true
      },
      "cell_type": "code",
      "source": "ssr = SSR_score(X_test, y_test, model)\nprint(\"SSR:\", round(ssr, 4))\nsse = sum((y_test - y_pred)**2)\nprint(\"SSE:\", round(sse, 4))\nsst = ((y_test - y_test.mean()) ** 2 ).sum()\nprint(\"SST:\", round(sst, 4))\nmae = mean_absolute_error(y_test, y_pred)\nprint(\"MAE:\", round(mae, 4))\nmse = mean_squared_error(y_test, y_pred)\nprint(\"MSE:\", round(mse, 4))\nrmse = np.sqrt(mean_squared_error(y_test, y_pred)) \nprint(\"RMSE:\", round(rmse, 4))\nr2 = r2_score(y_test, y_pred)\nprint(\"r2:\", round(r2, 4))",
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "stream",
          "text": "SSR: 732263.4857\nSSE: 109750.9993\nSST: 884878.3364\nMAE: 16.7216\nMSE: 512.8551\nRMSE: 22.6463\nr2: 0.876\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "markdown",
      "source": "<br/>**Polynomial test**"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from sklearn.preprocessing import PolynomialFeatures",
      "execution_count": 64,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Input for independent variable needs to be 2 dimmensional, which they are in this case.\nX_train = train[[\"FUELCONSUMPTION_COMB\", \"ENGINESIZE\", \"CYLINDERS\"]]\ny_train = train[\"CO2EMISSIONS\"]\n\nX_test = test[[\"FUELCONSUMPTION_COMB\", \"ENGINESIZE\", \"CYLINDERS\"]]\ny_test = test[\"CO2EMISSIONS\"]",
      "execution_count": 134,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "polynomial_feature = PolynomialFeatures(degree=4)\nX_train_poly = polynomial_feature.fit_transform(X_train)\nX_test_poly = polynomial_feature.fit_transform(X_test)",
      "execution_count": 135,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Creating linear regression object\nmodel = LinearRegression()",
      "execution_count": 136,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Training the model\nmodel.fit(X_train_poly, y_train)",
      "execution_count": 137,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 137,
          "data": {
            "text/plain": "LinearRegression()"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "model.intercept_",
      "execution_count": 138,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 138,
          "data": {
            "text/plain": "172.30734120782"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "model.coef_",
      "execution_count": 139,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 139,
          "data": {
            "text/plain": "array([-3.61626964e-10, -2.59201313e+01, -1.31653211e+02,  3.91390356e+01,\n        1.62807966e+01,  7.76541042e+01, -7.57945063e+01, -1.57266665e+02,\n        8.97038296e+01,  3.63033693e+01, -6.97394301e-01, -5.04293847e+00,\n        1.24906202e+00,  5.42098533e+01, -6.27817560e+01,  2.59552703e+01,\n       -3.27504825e+01,  2.35816054e+00,  4.08582374e+01, -2.62181371e+01,\n        5.55942706e-02, -2.58176653e-01, -1.92479009e-01, -6.47681175e-01,\n        2.99945869e+00, -3.82226418e-01,  3.36585343e+00, -1.03100471e+01,\n        4.01237974e+00, -1.30835869e+00, -8.81762462e+00,  2.00342228e+01,\n       -9.62998071e+00,  4.40299278e-01,  1.16150680e+00])"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Using model to predict on the test data\ny_pred_poly = model.predict(X_test_poly)",
      "execution_count": 141,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true,
        "scrolled": false
      },
      "cell_type": "code",
      "source": "mae = mean_absolute_error(y_test, y_pred_poly)\nprint(\"MAE:\", round(mae, 4))\nmse = mean_squared_error(y_test, y_pred_poly)\nprint(\"MSE:\", round(mse, 4))\nrmse = np.sqrt(mean_squared_error(y_test, y_pred_poly)) \nprint(\"RMSE:\", round(rmse, 4))\nr2 = r2_score(y_test, y_pred_poly)\nprint(\"r2:\", round(r2, 4))\n\n# adjusted r2 using formula adj_r2 = 1 - (1- r2) * (n-1) /(n - k - 1)\n# n = sample size = data.shape[0]\n# k = number of predictors = data.shape[1]\nn = X_train_poly.shape[0]\nk = X_train_poly.shape[1]\nadj_r2 = 1 - (1-r2)*(n - 1) / (n - k - 1) \nprint(\"r2̅ :\", round(adj_r2, 4))",
      "execution_count": 142,
      "outputs": [
        {
          "output_type": "stream",
          "text": "MAE: 8.5362\nMSE: 247.3734\nRMSE: 15.7281\nr2: 0.9402\nr2̅ : 0.9376\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "gist": {
      "id": "7b2c750ede0c3e9026ea0dca8e6e1c89",
      "data": {
        "description": "nuc_machine_learning/Session 3/tutorial_3.ipynb",
        "public": true
      }
    },
    "hide_input": false,
    "kernelspec": {
      "name": "venv",
      "display_name": "venv",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.8.3",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "latex_envs": {
      "eqNumInitial": 1,
      "eqLabelWithNumbers": true,
      "current_citInitial": 1,
      "cite_by": "apalike",
      "bibliofile": "biblio.bib",
      "LaTeX_envs_menu_present": true,
      "labels_anchors": false,
      "latex_user_defs": false,
      "user_envs_cfg": false,
      "report_style_numbering": false,
      "autoclose": false,
      "autocomplete": true,
      "hotkeys": {
        "equation": "Ctrl-E",
        "itemize": "Ctrl-I"
      }
    },
    "toc": {
      "nav_menu": {},
      "number_sections": true,
      "sideBar": true,
      "skip_h1_title": true,
      "base_numbering": 1,
      "title_cell": "Table of Contents",
      "title_sidebar": "Contents",
      "toc_cell": false,
      "toc_position": {},
      "toc_section_display": true,
      "toc_window_display": false
    },
    "varInspector": {
      "window_display": false,
      "cols": {
        "lenName": 16,
        "lenType": 16,
        "lenVar": 40
      },
      "kernels_config": {
        "python": {
          "library": "var_list.py",
          "delete_cmd_prefix": "del ",
          "delete_cmd_postfix": "",
          "varRefreshCmd": "print(var_dic_list())"
        },
        "r": {
          "library": "var_list.r",
          "delete_cmd_prefix": "rm(",
          "delete_cmd_postfix": ") ",
          "varRefreshCmd": "cat(var_dic_list()) "
        }
      },
      "types_to_exclude": [
        "module",
        "function",
        "builtin_function_or_method",
        "instance",
        "_Feature"
      ]
    },
    "_draft": {
      "nbviewer_url": "https://gist.github.com/7b2c750ede0c3e9026ea0dca8e6e1c89"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 4
}