rdu-weather-history-Ridge.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Introdução\n",
    "<p>Este <i>dataset</i> contém dados meteorológicos do Aeroporto Internacional de Raleigh Durham retirados do serviço da Web do NOAA.</p>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Lendo os Dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 300,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# As colunas estão separadas pelo símbolo ';' #. \n",
    "weather_history = pd.read_csv('rdu-weather-history.csv', parse_dates=['date'], sep=';')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Visão geral\n",
    "<p>O <i>dataset</i> contém informação sobre dados meteorológicos do Aeroporto Internacional de Raleigh Durham, desde 2007. Possui dados como temperaturas mínimas e máximas, volume de chuva, velocidade do vento e etc.</p>   \n",
    "</p>Visualizando as primeiras cinco entradas do DataFrame:</p>    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 301,
   "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>date</th>\n",
       "      <th>temperaturemin</th>\n",
       "      <th>temperaturemax</th>\n",
       "      <th>precipitation</th>\n",
       "      <th>snowfall</th>\n",
       "      <th>snowdepth</th>\n",
       "      <th>avgwindspeed</th>\n",
       "      <th>fastest2minwinddir</th>\n",
       "      <th>fastest2minwindspeed</th>\n",
       "      <th>fastest5secwinddir</th>\n",
       "      <th>...</th>\n",
       "      <th>drizzle</th>\n",
       "      <th>snow</th>\n",
       "      <th>freezingrain</th>\n",
       "      <th>smokehaze</th>\n",
       "      <th>thunder</th>\n",
       "      <th>highwind</th>\n",
       "      <th>hail</th>\n",
       "      <th>blowingsnow</th>\n",
       "      <th>dust</th>\n",
       "      <th>freezingfog</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2009-10-03</td>\n",
       "      <td>55.0</td>\n",
       "      <td>82.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.91</td>\n",
       "      <td>240.0</td>\n",
       "      <td>16.11</td>\n",
       "      <td>230.0</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2009-10-10</td>\n",
       "      <td>59.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>7.83</td>\n",
       "      <td>220.0</td>\n",
       "      <td>17.00</td>\n",
       "      <td>220.0</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2009-10-14</td>\n",
       "      <td>46.9</td>\n",
       "      <td>61.0</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.72</td>\n",
       "      <td>40.0</td>\n",
       "      <td>14.99</td>\n",
       "      <td>50.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2009-10-17</td>\n",
       "      <td>45.0</td>\n",
       "      <td>57.9</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.26</td>\n",
       "      <td>30.0</td>\n",
       "      <td>14.09</td>\n",
       "      <td>40.0</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2009-10-29</td>\n",
       "      <td>48.0</td>\n",
       "      <td>68.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.82</td>\n",
       "      <td>80.0</td>\n",
       "      <td>14.99</td>\n",
       "      <td>70.0</td>\n",
       "      <td>...</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        date  temperaturemin  temperaturemax  precipitation  snowfall  \\\n",
       "0 2009-10-03            55.0            82.0           0.00       0.0   \n",
       "1 2009-10-10            59.0            79.0           0.02       0.0   \n",
       "2 2009-10-14            46.9            61.0           0.14       0.0   \n",
       "3 2009-10-17            45.0            57.9           0.00       0.0   \n",
       "4 2009-10-29            48.0            68.0           0.00       0.0   \n",
       "\n",
       "   snowdepth  avgwindspeed  fastest2minwinddir  fastest2minwindspeed  \\\n",
       "0        0.0          2.91               240.0                 16.11   \n",
       "1        0.0          7.83               220.0                 17.00   \n",
       "2        0.0          8.72                40.0                 14.99   \n",
       "3        0.0          6.26                30.0                 14.09   \n",
       "4        0.0          5.82                80.0                 14.99   \n",
       "\n",
       "   fastest5secwinddir     ...      drizzle snow freezingrain smokehaze  \\\n",
       "0               230.0     ...           No   No           No        No   \n",
       "1               220.0     ...           No   No           No        No   \n",
       "2                50.0     ...          Yes   No           No        No   \n",
       "3                40.0     ...           No   No           No        No   \n",
       "4                70.0     ...           No   No           No        No   \n",
       "\n",
       "  thunder highwind hail blowingsnow dust freezingfog  \n",
       "0      No       No   No          No   No          No  \n",
       "1      No       No  Yes          No   No          No  \n",
       "2      No       No  Yes          No   No          No  \n",
       "3      No       No   No          No   No          No  \n",
       "4      No       No   No          No   No          No  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 301,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "weather_history.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número de linhas do dataset: 4137\n",
      "\n",
      "\n",
      "Resumo estatístico do DataFrame:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>temperaturemin</th>\n",
       "      <th>temperaturemax</th>\n",
       "      <th>precipitation</th>\n",
       "      <th>snowfall</th>\n",
       "      <th>snowdepth</th>\n",
       "      <th>avgwindspeed</th>\n",
       "      <th>fastest2minwinddir</th>\n",
       "      <th>fastest2minwindspeed</th>\n",
       "      <th>fastest5secwinddir</th>\n",
       "      <th>fastest5secwindspeed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4136.000000</td>\n",
       "      <td>4136.000000</td>\n",
       "      <td>4136.000000</td>\n",
       "      <td>4135.000000</td>\n",
       "      <td>4136.000000</td>\n",
       "      <td>4134.000000</td>\n",
       "      <td>4135.000000</td>\n",
       "      <td>4135.000000</td>\n",
       "      <td>4118.000000</td>\n",
       "      <td>4118.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>50.540063</td>\n",
       "      <td>72.017021</td>\n",
       "      <td>0.126630</td>\n",
       "      <td>0.012965</td>\n",
       "      <td>0.017384</td>\n",
       "      <td>5.860614</td>\n",
       "      <td>172.541717</td>\n",
       "      <td>15.957151</td>\n",
       "      <td>177.056824</td>\n",
       "      <td>21.801610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>16.229527</td>\n",
       "      <td>16.530515</td>\n",
       "      <td>0.371318</td>\n",
       "      <td>0.195214</td>\n",
       "      <td>0.213953</td>\n",
       "      <td>2.958446</td>\n",
       "      <td>94.603272</td>\n",
       "      <td>5.270319</td>\n",
       "      <td>96.850988</td>\n",
       "      <td>7.096004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.100000</td>\n",
       "      <td>23.200000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>4.920000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>6.930000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>37.000000</td>\n",
       "      <td>60.100000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.580000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>12.970000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>17.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>52.000000</td>\n",
       "      <td>73.900000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.370000</td>\n",
       "      <td>210.000000</td>\n",
       "      <td>14.990000</td>\n",
       "      <td>210.000000</td>\n",
       "      <td>21.030000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>64.900000</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>0.040000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.610000</td>\n",
       "      <td>240.000000</td>\n",
       "      <td>18.120000</td>\n",
       "      <td>240.000000</td>\n",
       "      <td>25.050000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>80.100000</td>\n",
       "      <td>105.100000</td>\n",
       "      <td>6.450000</td>\n",
       "      <td>6.690000</td>\n",
       "      <td>5.910000</td>\n",
       "      <td>19.010000</td>\n",
       "      <td>360.000000</td>\n",
       "      <td>59.950000</td>\n",
       "      <td>360.000000</td>\n",
       "      <td>86.120000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       temperaturemin  temperaturemax  precipitation     snowfall  \\\n",
       "count     4136.000000     4136.000000    4136.000000  4135.000000   \n",
       "mean        50.540063       72.017021       0.126630     0.012965   \n",
       "std         16.229527       16.530515       0.371318     0.195214   \n",
       "min          4.100000       23.200000       0.000000     0.000000   \n",
       "25%         37.000000       60.100000       0.000000     0.000000   \n",
       "50%         52.000000       73.900000       0.000000     0.000000   \n",
       "75%         64.900000       86.000000       0.040000     0.000000   \n",
       "max         80.100000      105.100000       6.450000     6.690000   \n",
       "\n",
       "         snowdepth  avgwindspeed  fastest2minwinddir  fastest2minwindspeed  \\\n",
       "count  4136.000000   4134.000000         4135.000000           4135.000000   \n",
       "mean      0.017384      5.860614          172.541717             15.957151   \n",
       "std       0.213953      2.958446           94.603272              5.270319   \n",
       "min       0.000000      0.000000           10.000000              4.920000   \n",
       "25%       0.000000      3.580000           80.000000             12.970000   \n",
       "50%       0.000000      5.370000          210.000000             14.990000   \n",
       "75%       0.000000      7.610000          240.000000             18.120000   \n",
       "max       5.910000     19.010000          360.000000             59.950000   \n",
       "\n",
       "       fastest5secwinddir  fastest5secwindspeed  \n",
       "count         4118.000000           4118.000000  \n",
       "mean           177.056824             21.801610  \n",
       "std             96.850988              7.096004  \n",
       "min             10.000000              6.930000  \n",
       "25%             90.000000             17.000000  \n",
       "50%            210.000000             21.030000  \n",
       "75%            240.000000             25.050000  \n",
       "max            360.000000             86.120000  "
      ]
     },
     "execution_count": 302,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Número de entradas\n",
    "print(\"Número de linhas do dataset:\", weather_history.shape[0])\n",
    "\n",
    "print(\"\\n\\nResumo estatístico do DataFrame:\")\n",
    "weather_history.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 4137 entries, 0 to 4136\n",
      "Data columns (total 28 columns):\n",
      "date                    4137 non-null datetime64[ns]\n",
      "temperaturemin          4136 non-null float64\n",
      "temperaturemax          4136 non-null float64\n",
      "precipitation           4136 non-null float64\n",
      "snowfall                4135 non-null float64\n",
      "snowdepth               4136 non-null float64\n",
      "avgwindspeed            4134 non-null float64\n",
      "fastest2minwinddir      4135 non-null float64\n",
      "fastest2minwindspeed    4135 non-null float64\n",
      "fastest5secwinddir      4118 non-null float64\n",
      "fastest5secwindspeed    4118 non-null float64\n",
      "fog                     4137 non-null object\n",
      "fogheavy                4137 non-null object\n",
      "mist                    4137 non-null object\n",
      "rain                    4137 non-null object\n",
      "fogground               4137 non-null object\n",
      "ice                     4137 non-null object\n",
      "glaze                   4137 non-null object\n",
      "drizzle                 4137 non-null object\n",
      "snow                    4137 non-null object\n",
      "freezingrain            4137 non-null object\n",
      "smokehaze               4137 non-null object\n",
      "thunder                 4137 non-null object\n",
      "highwind                4137 non-null object\n",
      "hail                    4137 non-null object\n",
      "blowingsnow             4137 non-null object\n",
      "dust                    4137 non-null object\n",
      "freezingfog             4137 non-null object\n",
      "dtypes: datetime64[ns](1), float64(10), object(17)\n",
      "memory usage: 905.0+ KB\n"
     ]
    }
   ],
   "source": [
    "weather_history.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.1 Utilizando a função Seaborn de mapas de calor\n",
    "<p>Seaborn é uma biblioteca em python para criação de visualizações estatísticas.</p>\n",
    "<p>O método ```.corr() ```, calcula o <b>coeficiente de correlação de Pearson</b> entre todos os pares de colunas numéricas do DataFrame.</p>\n",
    "<p>Interpretando o coeficiente:\n",
    "- 0.9 para mais ou para menos indica uma correlação muito forte.\n",
    "- 0.7 a 0.9 positivo ou negativo indica uma correlação forte.\n",
    "- 0.5 a 0.7 positivo ou negativo indica uma correlação moderada.\n",
    "- 0.3 a 0.5 positivo ou negativo indica uma correlação fraca.\n",
    "- 0 a 0.3 positivo ou negativo indica uma correlação desprezível."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns; sns.set()\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.figure(figsize=(18,10))\n",
    "weather_map = sns.heatmap(weather_history.corr(), annot=True, square=True, cmap=\"YlGnBu\", linewidths=.3)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Pré-processamento dos dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Substituindo as string por números binários -- No = 0.0 e Yes = 1.0\n",
    "\n",
    "weather_history = weather_history.replace('No', 0.0)\n",
    "weather_history = weather_history.replace('Yes', 1.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date                     0\n",
       "temperaturemin           1\n",
       "temperaturemax           1\n",
       "precipitation            1\n",
       "snowfall                 2\n",
       "snowdepth                1\n",
       "avgwindspeed             3\n",
       "fastest2minwinddir       2\n",
       "fastest2minwindspeed     2\n",
       "fastest5secwinddir      19\n",
       "fastest5secwindspeed    19\n",
       "fog                      0\n",
       "fogheavy                 0\n",
       "mist                     0\n",
       "rain                     0\n",
       "fogground                0\n",
       "ice                      0\n",
       "glaze                    0\n",
       "drizzle                  0\n",
       "snow                     0\n",
       "freezingrain             0\n",
       "smokehaze                0\n",
       "thunder                  0\n",
       "highwind                 0\n",
       "hail                     0\n",
       "blowingsnow              0\n",
       "dust                     0\n",
       "freezingfog              0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 305,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Verificando a soma dos valores nulos em cada feature\n",
    "\n",
    "weather_history.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "date                    0\n",
       "temperaturemin          0\n",
       "temperaturemax          0\n",
       "precipitation           0\n",
       "snowfall                0\n",
       "snowdepth               0\n",
       "avgwindspeed            0\n",
       "fastest2minwinddir      0\n",
       "fastest2minwindspeed    0\n",
       "fastest5secwinddir      0\n",
       "fastest5secwindspeed    0\n",
       "fog                     0\n",
       "fogheavy                0\n",
       "mist                    0\n",
       "rain                    0\n",
       "fogground               0\n",
       "ice                     0\n",
       "glaze                   0\n",
       "drizzle                 0\n",
       "snow                    0\n",
       "freezingrain            0\n",
       "smokehaze               0\n",
       "thunder                 0\n",
       "highwind                0\n",
       "hail                    0\n",
       "blowingsnow             0\n",
       "dust                    0\n",
       "freezingfog             0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 306,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import Imputer\n",
    "import numpy as np\n",
    "\n",
    "#Substituindo os valores nulos pela média dos valores referentes a feature\n",
    "imputer = Imputer(missing_values=np.nan, strategy='median', axis=0)\n",
    "weather_history[['temperaturemin']] = imputer.fit_transform(weather_history[['temperaturemin']])\n",
    "weather_history[['temperaturemax']] = imputer.fit_transform(weather_history[['temperaturemax']])\n",
    "weather_history[['precipitation']] = imputer.fit_transform(weather_history[['precipitation']])\n",
    "weather_history[['snowfall']] = imputer.fit_transform(weather_history[['snowfall']])\n",
    "weather_history[['snowdepth']] = imputer.fit_transform(weather_history[['snowdepth']])\n",
    "weather_history[['avgwindspeed']] = imputer.fit_transform(weather_history[['avgwindspeed']])\n",
    "weather_history[['fastest2minwinddir']] = imputer.fit_transform(weather_history[['fastest2minwinddir']])\n",
    "weather_history[['fastest2minwindspeed']] = imputer.fit_transform(weather_history[['fastest2minwindspeed']])\n",
    "weather_history[['fastest5secwinddir']] = imputer.fit_transform(weather_history[['fastest5secwinddir']])\n",
    "weather_history[['fastest5secwindspeed']] = imputer.fit_transform(weather_history[['fastest5secwindspeed']])\n",
    "\n",
    "weather_history.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Agrupamento dos dados para teste\n",
    "\n",
    "Como o clima de uma região possui uma grande relação com as estações do ano, os dados para teste foram agrupados de acordo com a estação que occorre em uma data específica.\n",
    "\n",
    "- Primavera: 20 de março a 20 de junho \n",
    "- Verão: 21 de junho a 21 de setembro\n",
    "- Outono: 22 de setembro a 20 de dezembro\n",
    "- Inverno: 21 de dezembro a 19 de março"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Criando nova coluna \"Season\"\n",
    "\n",
    "A coluna \"season\" foi criada com base nos periodos de inicio e de fim das estações do ano no local em que os dados foram colhidos."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Antes de utilizar o algoritimo para definir as estações a partir das datas é feita uma pequena alteração no formato da data.  \n",
    "Antes: AAAA-MM-DD  \n",
    "Depois: AAAAMMDD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "metadata": {},
   "outputs": [],
   "source": [
    "season_array = []\n",
    "\n",
    "for date in weather_history['date']:\n",
    "    if int(date.strftime('%m%d')) >= 1221 or int(date.strftime('%m%d')) <= 319:\n",
    "        season = 0 #'winter'\n",
    "    elif int(date.strftime('%m%d')) >= 320 and int(date.strftime('%m%d')) <= 620:\n",
    "        season = 1 #'spring'\n",
    "    elif int(date.strftime('%m%d')) >= 621 and int(date.strftime('%m%d')) <= 921:\n",
    "        season = 2 #'summer'\n",
    "    elif int(date.strftime('%m%d')) >= 922 and int(date.strftime('%m%d')) <= 1220:\n",
    "        season = 3 #'autumn'    \n",
    "    \n",
    "    season_array.append(season)\n",
    "\n",
    "weather_history['season'] = season_array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "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>date</th>\n",
       "      <th>temperaturemin</th>\n",
       "      <th>temperaturemax</th>\n",
       "      <th>precipitation</th>\n",
       "      <th>snowfall</th>\n",
       "      <th>snowdepth</th>\n",
       "      <th>avgwindspeed</th>\n",
       "      <th>fastest2minwinddir</th>\n",
       "      <th>fastest2minwindspeed</th>\n",
       "      <th>fastest5secwinddir</th>\n",
       "      <th>...</th>\n",
       "      <th>snow</th>\n",
       "      <th>freezingrain</th>\n",
       "      <th>smokehaze</th>\n",
       "      <th>thunder</th>\n",
       "      <th>highwind</th>\n",
       "      <th>hail</th>\n",
       "      <th>blowingsnow</th>\n",
       "      <th>dust</th>\n",
       "      <th>freezingfog</th>\n",
       "      <th>season</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2070</th>\n",
       "      <td>2007-01-01</td>\n",
       "      <td>48.9</td>\n",
       "      <td>68.0</td>\n",
       "      <td>0.45</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12.75</td>\n",
       "      <td>190.0</td>\n",
       "      <td>25.05</td>\n",
       "      <td>180.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>813</th>\n",
       "      <td>2007-01-02</td>\n",
       "      <td>32.0</td>\n",
       "      <td>55.9</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.13</td>\n",
       "      <td>320.0</td>\n",
       "      <td>12.97</td>\n",
       "      <td>330.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>814</th>\n",
       "      <td>2007-01-03</td>\n",
       "      <td>28.9</td>\n",
       "      <td>62.1</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.24</td>\n",
       "      <td>220.0</td>\n",
       "      <td>14.09</td>\n",
       "      <td>220.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>815</th>\n",
       "      <td>2007-01-04</td>\n",
       "      <td>46.0</td>\n",
       "      <td>69.1</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.47</td>\n",
       "      <td>220.0</td>\n",
       "      <td>14.09</td>\n",
       "      <td>230.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2449</th>\n",
       "      <td>2007-01-05</td>\n",
       "      <td>57.0</td>\n",
       "      <td>72.0</td>\n",
       "      <td>0.86</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.05</td>\n",
       "      <td>190.0</td>\n",
       "      <td>21.03</td>\n",
       "      <td>190.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           date  temperaturemin  temperaturemax  precipitation  snowfall  \\\n",
       "2070 2007-01-01            48.9            68.0           0.45       0.0   \n",
       "813  2007-01-02            32.0            55.9           0.00       0.0   \n",
       "814  2007-01-03            28.9            62.1           0.00       0.0   \n",
       "815  2007-01-04            46.0            69.1           0.00       0.0   \n",
       "2449 2007-01-05            57.0            72.0           0.86       0.0   \n",
       "\n",
       "      snowdepth  avgwindspeed  fastest2minwinddir  fastest2minwindspeed  \\\n",
       "2070        0.0         12.75               190.0                 25.05   \n",
       "813         0.0          3.13               320.0                 12.97   \n",
       "814         0.0          2.24               220.0                 14.09   \n",
       "815         0.0          4.47               220.0                 14.09   \n",
       "2449        0.0          8.05               190.0                 21.03   \n",
       "\n",
       "      fastest5secwinddir   ...    snow  freezingrain  smokehaze  thunder  \\\n",
       "2070               180.0   ...     0.0           0.0        0.0      0.0   \n",
       "813                330.0   ...     0.0           0.0        0.0      0.0   \n",
       "814                220.0   ...     0.0           0.0        0.0      0.0   \n",
       "815                230.0   ...     0.0           0.0        0.0      0.0   \n",
       "2449               190.0   ...     0.0           0.0        0.0      0.0   \n",
       "\n",
       "      highwind  hail  blowingsnow  dust  freezingfog  season  \n",
       "2070       0.0   0.0          0.0   0.0          0.0       0  \n",
       "813        0.0   0.0          0.0   0.0          0.0       0  \n",
       "814        0.0   0.0          0.0   0.0          0.0       0  \n",
       "815        0.0   0.0          0.0   0.0          0.0       0  \n",
       "2449       0.0   0.0          0.0   0.0          0.0       0  \n",
       "\n",
       "[5 rows x 29 columns]"
      ]
     },
     "execution_count": 308,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Ordenando o dataset pela coluna data\n",
    "\n",
    "weather_history = weather_history.sort_values('date', ascending=True)\n",
    "weather_history.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Criando DataFrames separados para cada estação."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "metadata": {},
   "outputs": [],
   "source": [
    "winter_weather_history = weather_history[weather_history['season'] == 0]\n",
    "spring_weather_history = weather_history[weather_history['season'] == 1]\n",
    "summer_weather_history = weather_history[weather_history['season'] == 2]\n",
    "autumn_weather_history = weather_history[weather_history['season'] == 3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 310,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "len_winter = len(winter_weather_history['date'])\n",
    "len_summer = len(summer_weather_history['date'])\n",
    "len_spring = len(spring_weather_history['date'])\n",
    "len_autumn = len(autumn_weather_history['date'])\n",
    "\n",
    "X = ['winter', 'summer', 'spring', 'autumn']\n",
    "Y = [len_winter, len_summer, len_spring, len_autumn]\n",
    "\n",
    "plt.bar(X, Y)\n",
    "plt.xlabel('Seasons')\n",
    "plt.ylabel('Amount of data for each season')\n",
    "plt.title('Season per amount')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 311,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fad357e5fd0>"
      ]
     },
     "execution_count": 311,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Exemplo para mostrar que a correlação com a divisão do dataframe em estações aumentou. Nesse exemplo\n",
    "# tem os dados do verão de 2007\n",
    "\n",
    "plt.figure(figsize=(10,8))\n",
    "\n",
    "winter_weather_history = winter_weather_history[winter_weather_history['date'] <= '2007-12-31']     \n",
    "winter_weather_history = winter_weather_history.drop(columns=['snowdepth', 'highwind', 'hail', 'blowingsnow', 'dust', 'freezingfog', 'season'])\n",
    "sns.heatmap(winter_weather_history.corr(), square=True, cmap=\"YlGnBu\", linewidths=.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 6. Ridge Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.1  Exemplo com Linear Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nesse exemplo, utiliza-se o Linear Regression para fazer a previsão dos dados, em que utiliza-se apenas uma feature. Um gráfico bidimensional é construído para representar essa técnica de regressão. A linha reta pode ser vista no gráfico, mostrando como a regressão linear tenta desenhar uma linha reta que melhor minimiza a soma residual de quadrados entre as respostas observadas no conjunto de dados e as respostas preditas pela aproximação linear.\n",
    "\n",
    "* Métricas:\n",
    "\n",
    " 1. Mean squared error: Essa métrica quantifica da variação a variável target que é prevista pelas features.\n",
    " 2. Coeficientes: Representam a mudança média na variável de resposta para uma unidade de mudança na variável preditor enquanto mantém outros preditores na constante do modelo.\n",
    " 3. Score: Precisão do modelo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 312,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Métricas:\n",
      "\n",
      "Coefficients: \n",
      " [[0.56899981]]\n",
      "Mean squared error: 40.87\n",
      "Variance score: 0.63\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from sklearn import linear_model\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "\n",
    "\n",
    "# Definindo os valores de x e y\n",
    "y = winter_weather_history['temperaturemin'].values\n",
    "x = winter_weather_history['temperaturemax'].values #Direção do vento\n",
    "\n",
    "y = y[:len(y)-1]\n",
    "x = x[:len(x)-1]\n",
    "\n",
    "# Modificando a dimensão de 1D para 2D\n",
    "y = y.reshape(-1, 1)\n",
    "x = x.reshape(-1, 1)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(x, y, test_size = 0.50, random_state=42)\n",
    "\n",
    "\n",
    "reg = linear_model.LinearRegression();\n",
    "reg.fit(X_train, y_train)\n",
    "\n",
    "plt.scatter(X_test, y_test,  color='black')\n",
    "plt.plot(X_test, reg_pred, color='blue', linewidth=3)\n",
    "\n",
    "plt.xticks(())\n",
    "plt.yticks(())\n",
    "plt.xlabel('Temperatura máxima')\n",
    "plt.ylabel('Temperatura mínima')\n",
    "plt.show()\n",
    "\n",
    "reg_pred = reg.predict(X_test)\n",
    "\n",
    "print('Métricas:\\n')\n",
    "#Coeficientes\n",
    "print('Coefficients: \\n', reg.coef_)\n",
    "# Mean squared error\n",
    "print(\"Mean squared error: %.2f\"\n",
    "      % mean_squared_error(y_test, reg_pred))\n",
    "print('Variance score: %.2f' % r2_score(y_test, reg_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.2 Example Ridge Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 317,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Métricas\n",
      "Mean squared error: 38.26\n",
      "R2:  [[0.64155439]]\n",
      "Score:  0.6506760369607618\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import Ridge\n",
    "\n",
    "ridge = Ridge(fit_intercept=True, alpha=0.0, random_state=0, normalize=True)\n",
    "ridge.fit(x, y)\n",
    "ridge_pred = ridge.predict(X_test)\n",
    "# Mean squared error\n",
    "print('Métricas')\n",
    "print(\"Mean squared error: %.2f\"\n",
    "      % mean_squared_error(y_test, ridge_pred))\n",
    "print('R2: ', ridge.coef_)\n",
    "print('Score: ', ridge.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K-fold Cross Validation - K-fold CV\n",
    "\n",
    "Muito importante para avaliar o modelo, pois aumenta a quantidade de dados para o treino do modelo. Usando 5 cv, o método retornará 5 scores e depois a média deles será calculada."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 318,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.62683938 -0.76137831  0.79578137 -0.62575381  0.74280225]\n",
      "0.1556581735669766\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "reg = LinearRegression()\n",
    "cv_results = cross_val_score(ridge, X_test, y_test, cv=5)\n",
    "\n",
    "print(cv_results)\n",
    "print(np.mean(cv_results))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 319,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fad36a57710>]"
      ]
     },
     "execution_count": 319,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD7CAYAAABkO19ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xl4E3XiP/D3JGl6Nz0o0wKhHG3laLlFdlt0SWnr9pCjLbprWXZXdFXUdfXnuVqfxVV31T3YdUX57n7rin51VayoxWO5LFXuwwDlpqGltGnpfdCmTeb3R2kgFEihSZNJ36/n4aGZzEzeDeE9M59JJoIkSRKIiMhjKVwdgIiInItFT0Tk4Vj0REQejkVPROThWPRERB6ORU9E5OH6VPRFRUVITU1FcnIyVq1a1ev+M2fOYPHixZg/fz4yMzPxzTffWO978803kZycjNTUVGzZssVxyYmIqG8kO7q6uqSkpCSprKxM6ujokDIzM6Vjx47ZzPPMM89I7777riRJknTs2DFpzpw51p8zMzOljo4OqaysTEpKSpK6urrsPSQRETmQyt6GQK/XIyoqClqtFgCQnp6ODRs2IDo62jqPIAhoaWkBADQ3N2Po0KEAgA0bNiA9PR1qtRparRZRUVHQ6/WYOnXqFR+vpqa5XxuuywkJ8UN9fZvD1+sMcskql5yAfLLKJScgn6xyyQn0P2t4eOAV77Nb9EajEREREdbboihCr9fbzPPAAw/grrvuwjvvvINz584hPz/fuuzkyZNtljUajdf8C/SXSqUc8Me8XnLJKpecgHyyyiUnIJ+scskJODer3aLvi8LCQixYsAC//OUvsXfvXjz++OP4/PPPr2tdISF+TvmFr7a1czdyySqXnIB8ssolJyCfrHLJCTgvq92iF0URVVVV1ttGoxGiKNrM89FHH+Gf//wnAGDq1Kno6OhAfX19n5a9lDMOs8LDA50yJOQMcskql5yAfLLKJScgn6xyyQn0P+vVNhJ233UTHx8Pg8GA8vJymEwmFBYWQqfT2cwTGRmJrVu3AgBOnDiBjo4OhIaGQqfTobCwECaTCeXl5TAYDJg0adJ1/yJERHTt7O7Rq1Qq5OXlYenSpTCbzcjKykJMTAxWrFiBuLg4JCUl4cknn8QzzzyDt956C4Ig4A9/+AMEQUBMTAx+/OMfIy0tDUqlEnl5eVAq5TNmRkTkCQRJcq/LFDvjMGswHb4NFLnkBOSTVS45AflklUtOwMVDN0REJG8seiIiD8eiJxogFknC2uJSHDTUuToKDTIOeR89EdnXYTLjy+1l6Cg249abRmLhzWOgUnJfi5yPrzKiAeLrrcKTd06DGOKLL7eX4Q/v7sHZhnOujkWDAIueaABFRQQi7+c3YtZEESfPNOG5/J3Ydbja1bHIw7HoiQaYr7cKd2dMwC/SxsFsseD1Tw5g9ddHYOo0uzoaeSiO0RO5gCAImD1pGMYO02Dl2gPYtKcChqpm3J0xARGhfq6ORx6Ge/RELjRsiD+e/dkM3DJlGErPNOF3b+3EtoNV9hckugYseiIXU3spseTWcXgsdzoAYNVnJchfdwgdHMohB+HQDZGbuHnqCIT6eWHl2gPYoq/Eycom3D8/DpFh/q6ORjLHPXoiNyKG+uG3i6djzrThqKhpxe/e2onvDlS6OhbJHIueyM14qZRYnHID7p8fB6VCwD8/P4T8dYf4rhy6bhy6IXJTM8YNxUgxACs/OYgt+kqUVjbhPg7l0HXgHj2RGxsa4oenF0/DnGnDcbqmFcv/vQvbSviuHLo2LHoiN9czlHPvvIkAgFWfluDtr46gs4tDOdQ3HLohkomZ40WMFAPxesEBbN5bgdIzTbhvQRyGBvu6Ohq5Oe7RE8lIRKgfnvnZdMyeFIlTxmb8Ln8n9hytcXUscnMseiKZUXsp8Yu08bgrfTzMZgte+3g/Pth4HF1mi6ujkZti0RPJVEJ8JJ752QyIoX74ckcZXnlvL+qbO1wdi9wQi55IxkYMDUDekhmYMW4ojp1uxO/yd+AQv8GKLsGiJ5I5X28V7ps3ET+ZG4PW9i68+p99KNxqgEWSXB2N3ASLnsgDCIKA5BlaPHHnNAQHeGPNNyfx2pr9aG3vdHU0cgMseiIPEj1cg+d+cSPGR4Vg3/GzWP7WTpQZm10di1ysT++jLyoqwgsvvACLxYKcnBzcc889Nve/+OKL2L59OwCgvb0dtbW12LVrFwBg/PjxiI2NBQBERkbijTfecGR+IrpEkJ8aj94+BZ8Un8Tn353CC6t3IzclFrMnDXN1NHIRu0VvNpuxfPly5OfnQxRFZGdnQ6fTITo62jrP008/bf159erVKCkpsd728fHB2rVrHRybiK5GoRCw8OaxGDNMg39+VoL8dYdx8kwTfjo3Fl4qHsgPNnb/xfV6PaKioqDVaqFWq5Geno4NGzZccf7CwkJkZGQ4NCQRXZ8p0UOQ9/MZ0A4NwDf7zuAP7+5GbWO7q2PRALO7R280GhEREWG9LYoi9Hr9ZeetqKjA6dOnMWvWLOu0jo4OLFy4ECqVCvfccw/mzp171ccLCfGDSqXsa/4+Cw8PdPg6nUUuWeWSE5BPVmfkDA8PxJ9/E4bXP/oem3afxvNv78LjuTMwOTa83+uVA7nkBJyX1aHXuiksLERqaiqUygtFvWnTJoiiiPLycixZsgSxsbEYOXLkFddRX9/myEgAup+8mhp5nJCSS1a55ATkk9XZOXPnxmBEmB/+b/0xPLvqOyy8eQzSZkVBEIRrXhefU8frb9arbSTsDt2IooiqqguXRTUajRBF8bLzrlu3Dunp6b2WBwCtVouZM2fajN8T0cARBAFzpo3Akxe9BfP1Tw7gXEeXq6ORk9kt+vj4eBgMBpSXl8NkMqGwsBA6na7XfCdOnEBTUxOmTp1qndbY2AiTyQQAqKurw549e2xO4hLRwBs7XIPnfn4jbtAGY/eRGrywejeq6hx/JE3uw+7QjUqlQl5eHpYuXQqz2YysrCzExMRgxYoViIuLQ1JSEoDuvfm0tDSbw8ATJ07gueeegyAIkCQJd999N4ueyA0E+avx6B1T8OGmE/jvrnI8/++duDtjIqbEDHF1NHICQZLc63PSzhhPG0zjdANFLjkB+WR1Vc6tB6vw7y8Ow9RlwW0Jo3Bb4mgo7Izb8zl1PJeO0RORZ/vBxAg8vXg6hmh88Om3BvztIz3aeOkEj8KiJyKMFAOR9/MbMXF0KPQnavH827tx5myrq2ORg7DoiQgAEODrhd/kTMaPbxoJY10bfv/2Luw7dtbVscgBWPREZKVQCMiZE417bpsAi0XC39fo8dm3pbzkscyx6Imol1kTIvBU7nSEBnmjYEspVhYcQLuJ77eXKxY9EV1WVEQgnu15v/3R7vfbVzvhk+vkfCx6IrqiIL/u99snTRuBippWPP/vXTjIryqUHRY9EV2VSqnAnSmx+PmPx6Gj04y//Od7fF58Em72ERy6ChY9EfXJzZOH4fGfTEOArwpvFuzH6q+OoMtscXUs6gMWPRH1WfQIDZ5dciNGDwvC5n1n8Kf396G5zeTqWGQHi56IrkmYxgcvPzAb02PDcaS8Ac//excqalpcHYuugkVPRNfMx1uF+xbE4baEUTjb2I4XVu/GvuP8cJW7YtET0XVRCALmzx6De+dNhNki4e8f6fHF9lM8SeuGWPRE1C8zx4t4KncaggO98eGmE/hX4SF0dvEkrTth0RNRv42KCMKzS2ZgdGQQvjtQhT+9vxct53gFTHfBoicihwgO8MYTP52KGeOG4ujpRvz+7V385io3waInIodReylx77yJSP9BFKrrz+GFt3fhSFm9q2MNeix6InIohSAg65ax+GXaeLSbzHj1/X0o1le6OtagxqInIqdInBSJR2+fAh+1Ev+77hDWfHOClzt2ERY9ETnNuKgQPL14OoYG+6Jw6ym8ufYgTJ1mV8cadFj0RORUkWH++O3PpiN2hAY7D1fj5ff2orGVl00YSCx6InK6QD81Hr1jKn4wUcTJM034/b938TtpBxCLnogGhJdKgaUZEzA/cTRqm9rx0ju7cbS8wdWxBoU+FX1RURFSU1ORnJyMVatW9br/xRdfxLx58zBv3jykpqZixowZ1vsKCgqQkpKClJQUFBQUOC45EcmOIAi4LXE07kq/8I6cXYerXR3L46nszWA2m7F8+XLk5+dDFEVkZ2dDp9MhOjraOs/TTz9t/Xn16tUoKSkBADQ0NOC1117DmjVrIAgCFi5cCJ1OB41G44RfhYjkIiE+EpoANf5RcAArPzmAO+bGIHmG1tWxPJbdPXq9Xo+oqChotVqo1Wqkp6djw4YNV5y/sLAQGRkZAIDi4mIkJCQgODgYGo0GCQkJ2LJli+PSE5FsxY0Ow5M/nYYgfzXeW38MH2w8zrdfOondojcajYiIiLDeFkURRqPxsvNWVFTg9OnTmDVr1jUvS0SDT1REIH67eDoiw/zw5Y4yrPr0IC+I5gR2h26uRWFhIVJTU6FUKq97HSEhflCprn/5KwkPD3T4Op1FLlnlkhOQT1a55AQclzU8PBB/evgWPP+v7dhxqBrnTBY8/YuZCPD1ctj65cJZWe0WvSiKqKqqst42Go0QRfGy865btw55eXk2y+7YscNm2ZkzZ1718errHX8RpPDwQNTUNDt8vc4gl6xyyQnIJ6tccgLOyfrrrHj8z2cl2H20Bv9vxTf4Tc5khAb59Gudg+k5vdpGwu7QTXx8PAwGA8rLy2EymVBYWAidTtdrvhMnTqCpqQlTp061TktMTERxcTEaGxvR2NiI4uJiJCYmXuevQUSeTO2lxH3z4zB3+ghU1LTihdW7cbqaX1HoCHb36FUqFfLy8rB06VKYzWZkZWUhJiYGK1asQFxcHJKSkgB0782npaVBEATrssHBwbj//vuRnZ0NAFi2bBmCg4Od9KsQkdwpFAJ+MjcGoUE++GDTcbz07m48lDUJN4wMcXU0WRMkN/veL2ccZg2mw7eBIpecgHyyyiUnMDBZt5VU4V+fH4JCIWDZgjhMGjvkmtcxmJ7Tfg3dEBG5wqwJEXgoexIEAH9fsx/bSqrsLkOXx6InIrcVPyYMj9w+BWovJf7n0xJs2lvh6kiyxKInIrcWqw3GEz+digA/L6z+6ggKtxpcHUl2WPRE5PZGioF4Knc6woK8seabk/hw03G42elFt8aiJyJZiAj1w1O50xER6ocvtpfh7a+OwGJh2fcFi56IZCM0yAdP5k5DlBiIb/adwZufHkSXmZdMsIdFT0SyEuSnxmM/mWr9xqq/rdGjg19PeFUseiKSHT8fFX5z+xRMGhuGAyfr8Kf/7ENbe6erY7ktFj0RyZK3lxIPLIzHzPFDcfx0I17+v71oauN30V4Oi56IZEulVOCezIn40ZRhKKtuwavvsewvh0VPRLKmUAhYnHoDkqaNwOmaVpb9ZbDoiUj2BEHAT5NjbMq+mWVvxaInIo/QU/a6acNxuqYVr7y3D40tHa6O5RZY9ETkMQRBwJ3JsZgzbThO17TgmTe+4549WPRE5GEEQUDu+bI3VDbh1ff3DfqyZ9ETkcfpKfsf/3AUyqtb8Or7+9BybvC+z55FT0QeSRAE3LtgEuZMHY7y6ha88t7eQVv2LHoi8lgKhYA7U2Lxo/Nl/+ogLXsWPRF5NIUgIDcl1uZDVYOt7Fn0ROTxFIKA3NQbcEtP2b8/uMqeRU9Eg4JC6P4E7S1ThqHM2II//2cf2k1dro41IFj0RDRo9JR9QnwEDFXNeL3gwKC4nj2LnogGFYUg4Oc/Htd9iePSOuSvOwSLh38tIYueiAYdpUKB++bFYcywIGw9aMRHm0+4OpJT9anoi4qKkJqaiuTkZKxateqy86xbtw5paWlIT0/Ho48+ap0+fvx4zJs3D/PmzcO9997rmNRERP3krVbi19mTEBHqhy+3l+HrHWWujuQ0KnszmM1mLF++HPn5+RBFEdnZ2dDpdIiOjrbOYzAYsGrVKrz33nvQaDSora213ufj44O1a9c6Jz0RUT8E+qnxyO2T8cLq3Xh/43EE+asxa2KEq2M5nN09er1ej6ioKGi1WqjVaqSnp2PDhg0283zwwQe48847odFoAABhYWHOSUtE5GBDNL54ZNEU+Hqr8K/CQzhYWufqSA5nt+iNRiMiIi5s4URRhNFotJnHYDCgtLQUd9xxBxYtWoSioiLrfR0dHVi4cCEWLVqE9evXOzA6EZFjaIcG4KGseAiCgNcK9sNQ1eTqSA5ld+imL8xmM06dOoXVq1ejqqoKubm5+OyzzxAUFIRNmzZBFEWUl5djyZIliI2NxciRI6+4rpAQP6hUSkfEshEeHujwdTqLXLLKJScgn6xyyQnIJ2tfc4aHB0LwUuGPb+/E3z7aj5cfnI3IIf5OTtc7gzPYLXpRFFFVVWW9bTQaIYpir3kmT54MLy8vaLVajBo1CgaDAZMmTbLOq9VqMXPmTJSUlFy16Ovr2673d7mi8PBA1NQ0O3y9ziCXrHLJCcgnq1xyAvLJeq05YyMDkZsci9VfH8VvVxbj6cUzoPFXOzHhBf19Tq+2kbA7dBMfHw+DwYDy8nKYTCYUFhZCp9PZzDN37lzs2LEDAFBXVweDwQCtVovGxkaYTCbr9D179ticxCUicjdzpo1Axg9HoaahHX/94Huc65D/p2ft7tGrVCrk5eVh6dKlMJvNyMrKQkxMDFasWIG4uDgkJSVh9uzZ+Pbbb5GWlgalUonHH38cISEh2LNnD5577jkIggBJknD33Xez6InI7S2YPRpNrR0o+r4S/yjYj4dzJkOllO/HjgRJcq+PhDnjcFAuh5mAfLLKJScgn6xyyQnIJ2t/cpotFvzj4wPYd/wsbpog4u7MCVAIgoMTXuDSoRsiosFIqVDgV/MmInq4BttLjPhyu3w/UMWiJyK6Am8vJR5YGI+QQG+s+eYEDhnk+R57Fj0R0VUE+atx3/w4KAQBb3x6EPXNHa6OdM1Y9EREdkQP1+B2XTSa2zrx+if7ZXdpYxY9EVEfJE0fgZsmiDhR0YQPNh53dZxrwqInIuoDQRCw5NYbMGyIP9bvPo3tJUb7C7kJFj0RUR/5qFVYtiAO3mol3vriMCrOtro6Up+w6ImIrkFkmD/uShuPjk4zXi/YL4tPzrLoiYiu0YxxQ5FyoxaVtW3I/+Iw3Oxzp72w6ImIrkP2j8YiZoQGuw5X47+7Trs6zlWx6ImIroNKqcB98+MQ5K/Gh5uO42h5g6sjXRGLnojoOgUHeOO+eRMhScDKtQfQ2OKeH6Zi0RMR9cMNI0OQ/aOxaGwx4Y21B2G2uN+HqVj0RET9lDpTi+mx4ThS3oCColJXx+mFRU9E1E+CIOCX6eMRHuyDL7eXoby6xdWRbLDoiYgcwNdbhdyUG2CRJKz++ggsbvSWSxY9EZGDxI8Jw/TYcBw/3Yjv9lfZX2CAsOiJiBzoJ3NjoPZS4MPNx9Ha3unqOABY9EREDhUa5IPbEkajua0TH39z0tVxALDoiYgcLuVGLSLD/LB5bwVKK5tcHYdFT0TkaCqlArkpN0AC8M7XR2CxuPbELIueiMgJxkeF4KYJIkorm1GkP+PSLCx6IiInuV0XDR+1Ems2n0Bzm8llOfpU9EVFRUhNTUVycjJWrVp12XnWrVuHtLQ0pKen49FHH7VOLygoQEpKClJSUlBQUOCY1EREMhAc4I35s8egtb0LH20+4bIcKnszmM1mLF++HPn5+RBFEdnZ2dDpdIiOjrbOYzAYsGrVKrz33nvQaDSora0FADQ0NOC1117DmjVrIAgCFi5cCJ1OB41G47zfiIjIjSRNH45ifSW26Csxe/IwRA8f+P6zu0ev1+sRFRUFrVYLtVqN9PR0bNiwwWaeDz74AHfeeae1wMPCwgAAxcXFSEhIQHBwMDQaDRISErBlyxYn/BpERO5JqVAgNyUWAPDOV0dcctEzu0VvNBoRERFhvS2KIoxG2y/FNRgMKC0txR133IFFixahqKioz8sSEXm6WG0wEuIiUFbdgk17Kgb88e0O3fSF2WzGqVOnsHr1alRVVSE3NxefffbZda0rJMQPKpXSEbFshIcHOnydziKXrHLJCcgnq1xyAvLJ6i45782egn0nNuCT4lLcmjAGIUE+veZxVla7RS+KIqqqLlyzwWg0QhTFXvNMnjwZXl5e0Gq1GDVqFAwGA0RRxI4dO2yWnTlz5lUfr76+7Vp/B7vCwwNRU9Ps8PU6g1yyyiUnIJ+scskJyCeru+VcMHs03vn6KN767AByU26wua+/Wa+2kbA7dBMfHw+DwYDy8nKYTCYUFhZCp9PZzDN37lxrodfV1cFgMECr1SIxMRHFxcVobGxEY2MjiouLkZiYeN2/CBGRnN0yZRg0AWpsO2hEZ5d5wB7X7h69SqVCXl4eli5dCrPZjKysLMTExGDFihWIi4tDUlISZs+ejW+//RZpaWlQKpV4/PHHERISAgC4//77kZ2dDQBYtmwZgoODnfsbERG5KaVCgR9MjMCX28uw73gtbhw3dEAeV5AkN7poMuCUwyx3O3y7GrlklUtOQD5Z5ZITkE9Wd8xZUdOCZ/+1A5PGhuHhnMnW6S4duiEiIscZHh6A0ZGB2H+yFg0D9GXiLHoiogH2w7hISBKw7eDAvN2cRU9ENMBumiBCpRTw7f5KDMToOYueiGiABfh6YXL0EFScbcUpo/PPIbDoiYhcICE+EgDwrd753y3LoicicoG40aEI8ldjW0kVOruce/0bFj0RkQuolArMmiCitb0L+hNnnfpYLHoiIhexDt/sd+7wDYueiMhFtEMDMFIMgP5ELeqb2532OCx6IiIXSoiPhEWS8I0TL1/MoicicqFZE0QoFQI27ipz2mOw6ImIXCjQT41JY8NQeqYJVXWOv0w7wKInInK5EeEBAICmVpNT1s+iJyJyMR9197fqtZucc416Fj0RkYt5ny/6jk4WPRGRR/L26tmj73LK+ln0REQu1lP0HRy6ISLyTD4cuiEi8mzePBlLROTZOHRDROThrG+v5NANEZFn8larAHCPnojIY/l48WQsEZFH8/JSQBCcdzJW1ZeZioqK8MILL8BisSAnJwf33HOPzf0ff/wxXn75ZYiiCADIzc1FTk4OAGD8+PGIjY0FAERGRuKNN95wZH4iItlTCAJ81EqnDd3YLXqz2Yzly5cjPz8foigiOzsbOp0O0dHRNvOlpaUhLy+v1/I+Pj5Yu3at4xITEXkgH7XKdSdj9Xo9oqKioNVqoVarkZ6ejg0bNjglDBHRYCRJEgQBMLmq6I1GIyIiIqy3RVGE0WjsNd/XX3+NzMxMPPTQQ6isrLRO7+jowMKFC7Fo0SKsX7/eQbGJiDzH4bIG1DV1IHq4xinr79MYvT1z5sxBRkYG1Go13n//fTzxxBN4++23AQCbNm2CKIooLy/HkiVLEBsbi5EjR15xXSEhflCplI6IZSM8PNDh63QWuWSVS05APlnlkhOQT1Y55HzzsxIAwKLkG5yS127Ri6KIqqoL31BuNBqtJ117hISEWH/OycnBK6+8YrM8AGi1WsycORMlJSVXLfr6esd/w0p4eCBqapodvl5nkEtWueQE5JNVLjkB+WSVQ87q+jbsOFiFGG0wQv1U1533ahsIu0M38fHxMBgMKC8vh8lkQmFhIXQ6nW3Q6mrrzxs3bsTYsWMBAI2NjTCZur8xpa6uDnv27Ol1EpeIaDDbsLsCEoDbbh4LQRCc8hh29+hVKhXy8vKwdOlSmM1mZGVlISYmBitWrEBcXBySkpKwevVqbNy4EUqlEhqNBi+99BIA4MSJE3juuecgCAIkScLdd9/NoiciOu9cRxe26M9AE6BGwqRhaKhvdcrjCJIkSU5Z83VyxmGWHA7fesglq1xyAvLJKpecgHyyunvO/+4qx3vrj2HB7NH45fxJ/crar6EbIiJyPIskYcPu01ApFbhl6nCnPhaLnojIBfQnalFdfw6zJooI8lM79bFY9ERELvDfneUAgOQZWqc/FoueiGiAna5pwaFT9Rg3MhjaoQFOfzwWPRHRAFu/6zSAgdmbB1j0REQDqrnNhK0HqxAe7IPJ0UMG5DFZ9EREA6jo+zPo7LIgaboWCoVzPiB1KRY9EdEAaWzpwIbdp+GtViIxPnLAHtchFzUjIqKrKzM2429r9GhoMSHjh6Pg5zNw9cuiJyJyst1HqvE/n5fA1GlB1i1jkDYrakAfn0VPROQkkiTh8+8MKNhSCm8vJR5YGI9pseEDnoNFT0TkBKZOM/K/OIztJUaEBXnjwaxJGCm65tr4LHoiIgerb+7Aax/rUVrZjOjhGixbGA+Nv3Mvc3A1LHoiIgcqrWzC38+fdE2Ii8DPbh0HL5Vr3+DIoicicpAdh4z438JD6OyyYNGcaKTO1Drty0SuhUcVvUWSIABu8cQS0eBhkSR8WlyKT781wEetxIPZkzBlgD712hceU/RNbSY89eY2KARg2BB/DB/ib/1b5eMFSZK4ASAih+swmfGvwhLsOlKDIRof/Dp7EoaHO/9CZdfCY4reV63EtNghOH66EccrGnHsdKPN/QG+XhgW5odh4QE2G4EgF54gISJ5q2tqx9/W6FFmbEGsNhjLFsQh0MnXlr8eHlP0Xiol7kqfAADo7DKjsrYNZ862ouJsK842daD0THf5H73cBuCSI4Bh4f5O/yIAIpK3E2ca8dqa/WhsNeHmyZHITbkBKqV7XlXGY4r+Yl4qJUaKgdb3rPZ8b6Sp04yqujZUnG3t3gjUtOJMbSuOlTfgaHmDzToCfL2spT/8og2BO26tiWhgbT1Yhfx1h2G2WPCTpBjMnTHCrYeGPbLor0TtZbsB6GHqtD0COHP+z9HyBhy5ZAMQ6Odlu/fPDQDRoNDZZcG+42dRrK/E/pO18PVW4aF58YgbE+bqaHYNqqK/ErWXElERgYiKsN0AdHSaUVXbhoqzLag424rKs90/HylrwOEy2w1AkF/PEFAAhoX7Y1iYH4aHByDA12sgfxUiciBJklBmbEGxvhLbSqrQ2t4FABg7LAi/TB+PyDB/FyfsGxb9VXhfaQNgMqOy7vzQT88w0NlWHL7cBsBffdkjAG4AiNz8Ax1GAAAPj0lEQVRXU6sJ2w5WoXh/JU7XtALo/r9868yRSIiPcLt31djTp6IvKirCCy+8AIvFgpycHNxzzz0293/88cd4+eWXIYoiACA3Nxc5OTkAgIKCAqxcuRIAcN9992HBggWOzO8S3molRkUEYVREkM30DpMZZ2pty//M2VYcOlWPQ6fqbebV+Kt7lf/wcH8M/OWOiAgAuswW7D9Ri+L9ldCfqIXZIkGpEDAtNhyJ8ZGIGxPqtidb7bFb9GazGcuXL0d+fj5EUUR2djZ0Oh2io6Nt5ktLS0NeXp7NtIaGBrz22mtYs2YNBEHAwoULodPpoNFoHPtbuAlvtRKjI4MwOtJ2A9Bu6up1DqCi5vIbgNAgb0SE+l0o//N//Hx4BEDkDKerW1C8vxLbDlahqa0TAKAdGoDE+EjcNFH0iHfg2S16vV6PqKgoaLXdX2Kbnp6ODRs29Cr6yykuLkZCQgKCg4MBAAkJCdiyZQsyMjL6GVtefNSqq24AeoaAKs62oqq+DSWGepQYbDcAwQHdQ0CR1vIPwLAh/gP65QVEnqLlXCe2lxhRvL8Sp6qaAXS/027u9BFInBTpsqtMOovdljAajYiIiLDeFkURer2+13xff/01du7cidGjR+Opp55CZGTkZZc1Go0Oii5/l9sAhIcHoux0ffcG4GyLdQNQebYVBw31OHjJBiAk0Lt77z+se+in52duAIhsWSwSDpTWoXh/JfYdq0GXWYJCEDB5bBgSJ0VicvQQ2Q7N2OOQNpgzZw4yMjKgVqvx/vvv44knnsDbb799XesKCfGDSqV0RCwb4eHy2UKPHBGCkSNCek1va+9EubEZZVXNKLvo74OldThYWmcz7xCND7RiIEZGBGFkRGD3HzHQoUNAcnpO5ZJVLjkB+WTtkID1O8qwafdp1DW1AwC0YgDm3jgSP5quRWiQj4sTXuCs59Ru0YuiiKqqKutto9FoPenaIyTkQinl5OTglVdesS67Y8cOm2Vnzpx51cerr2/rW/Jr0POBKTmwlzXUzwuhY0IxZUyoddq5jq5enwGoONuKvUdrsPdoje3yQd4YFmb7KeBhYf7w9b62bb4nPafuQi45AffP2tbehZ2Hjdh+qBqHz58H8/NWYc7U4UiIj8ToyEAIggBzRydqajpdnLZbf5/Tq20k7P7vjo+Ph8FgQHl5OURRRGFhIf70pz/ZzFNdXY2hQ4cCADZu3IixY8cCABITE/HnP/8ZjY3dlx0oLi7GI488ct2/CF2er7cKY4drMHa47Unutvauy74L6EBpHQ5ccgQQFuRtHf+3fh5giB981BwCInmwSBIOn6pH8f5K7DlSA1OXBYIATBwdisT4SEyLHQIvJ4wWyIHd/8UqlQp5eXlYunQpzGYzsrKyEBMTgxUrViAuLg5JSUlYvXo1Nm7cCKVSCY1Gg5deegkAEBwcjPvvvx/Z2dkAgGXLlllPzJLz+fmoED1cg+heG4BOnDnbhjO1PZ8F6P5A2IGTdThwsvcGYNiQ7gvBRYb5ISTQG4F+aijUKnSZLR47pknura29C1V1baiqa0VlbRuq6tpQWtmEuqYOAMDQEF8kxkci4+ZooKvLxWldT5AkSXJ1iIs543DQ3Q8zL+bKrK3tndZP/1ZcdBTQ2GK64jIBvl4I8lcjyK/nb3X33+f/aPzVCPTzgsZf7bK9Kbn8+8slJzAwWc0WC842tKOyrg1V58u8509Ta+/XpK+3CjNuCEdCfCRiRmggCMKgek77NXRDg4e/jxeiR2gQPcL2CKC1vRNnznbvOTW2mtDcakKHWUJNXSsaW01obOnAmbOtdtfv66203RBc8rPGX41Afy8E+anho1a69UWiyHFaznWiqrYNlXWtNoVeXX8OZovtfqgAIEzjg7gxoYgI9UNkmD8iQv0QEeqH4AA1XzNXwKInu/x9vBAzIhgxIy4Mu12699FltqC5rRNNrSY0tZm6/241ofH87eZWExpbO9HUZkJ1QyPsHUeqVYpLNgi2Rwya8/cF+qnh76Pif3A312W2oLr+3IW98osKveVc75Ohft4qREUEWks8ItQPEWF+EEN8B+04e3+w6MkhVEoFQgK9ERLobXdei0VCy7nujULj+Y1Az8/dG4gLG4xTVc299uoupVQIlxwheNkcLYwcdg5SpxlB/moE+HpBoeBGwRkkSUJTWyeqalutJd4zfn62oR2WS7buCkFAeLAPxg4LQkTYxYXujyA/L268HYhFTwNO0VPM/mqMsDOvJElo6+iyPUJoNaGp5+jhoiOIytpWnDJefYxTEIDAnvMKlw4h9Tpa8OLJ5svo7DLDUNmEQ8dreo2fn+vofeIzwNcLY4YFnR9qubB3Hh7sy+d3gLDoya0JggB/Hy/4+3jZvSSsJEloN5nR3NZ9VNAzbGSWgMqaFpshpdqmdutVCa/G30d1mfMJthsKzfn71F4XhhQkSYJFkmCx4Pzf0vlp3Uc0PdMs56dJFgntFqC2tsV2HkmCdH4dZkvP7d7rvnTei9dtsVnukvvPT5PO/3zhMWBdZ8+8ja0mVNW2obaxHZceYykVAoaG+GLcyGDr3nlkqD8iwvx4pVY3wKInjyEIAny9VfD1VmHoRR8svtK7GUyd5u7zB22dF44ULjlK6DlyqKy1/0E+lVKAdL5Y3eqtbA4U5K9GjDYYo4droPH1QkSYHyJD/TAk2AdKBffO3RWLngYttZcSQzS+GKLxtTvvZU82t9luHNo6uqAQBAgKAQpBgELoHqZSCIL1b6HXNFiX8fdTw2RdB2yWU/Tctq7/ovVctK6ex1b2PN5Fy126rr7mVCq65w3wUVkvoSGnty0Si56oT67lZPP1YnmSs/BYi4jIw7HoiYg8HIueiMjDseiJiDwci56IyMOx6ImIPByLnojIw7HoiYg8nNt98QgRETkW9+iJiDwci56IyMOx6ImIPByLnojIw7HoiYg8HIueiMjDybLoi4qKkJqaiuTkZKxatarX/SaTCQ8//DCSk5ORk5OD06dPAwA+/fRTzJs3z/pn3LhxOHToEABg8eLFSE1Ntd5XW1vr9Jw7d+7EggULMGHCBHz55Zc29xUUFCAlJQUpKSkoKCiwTj9w4AAyMzORnJyM3//+93DUu2OvN+uhQ4dw++23Iz09HZmZmVi3bp31vieffBI6nc76nPY8167ICQDjx4+3Zrn33nut08vLy5GTk4Pk5GQ8/PDDMJlMLsu5bds2m9dofHw81q9fD8A5z2dfsubn5yMtLQ2ZmZlYsmQJKioqrPcN5Ov0enMO9Gu0P1kBJ71OJZnp6uqSkpKSpLKyMqmjo0PKzMyUjh07ZjPPO++8Iz377LOSJEnS559/Lv3617/utZ7Dhw9LSUlJ1tu5ubmSXq8f0Jzl5eXSoUOHpMcee0z64osvrNPr6+slnU4n1dfXSw0NDZJOp5MaGhokSZKkrKwsae/evZLFYpHuuusuafPmzS7NevLkSam0tFSSJEmqqqqSEhISpMbGRkmSJOmJJ56wmdeVOSVJkqZMmXLZ9T700EPS559/LkmSJD377LPSu+++69KcPerr66Ubb7xRamtrkyTJ8c9nX7Nu3brVmuHdd9+1/n8ayNdpf3IO5Gu0v1klyTmvU9nt0ev1ekRFRUGr1UKtViM9PR0bNmywmWfjxo1YsGABACA1NRVbt27ttUdRWFiI9PR0l+YcMWIExo0bB8Ul37VZXFyMhIQEBAcHQ6PRICEhAVu2bEF1dTVaWlowZcoUCIKA+fPn91rnQGcdPXo0Ro0aBQAQRRGhoaGoq6vrdyZH57wSSZKwbds2pKamAgAWLFjQ7+fUUTm/+uorzJ49G76+9r/q0JlZZ82aZc0wZcoUVFVVARjY12l/cg7ka7S/Wa+kv69T2RW90WhERESE9bYoijAajb3miYyMBACoVCoEBgaivr7eZp5169b1Kvqnn34a8+bNwz/+8Y9+H2r2Jee1Lnvp9IiIiD6v01lZL6bX69HZ2YmRI0dap/3lL39BZmYmXnzxxX4PifQ3Z0dHBxYuXIhFixZZh0Pq6+sRFBQElar7WzUd8Zw66vksLCxERkaGzTRHPp/Xk/Wjjz7CzTfffNVlnfE67U/Oizn7NeqIrM54nQ7K74z9/vvv4evri9jYWOu0V199FaIooqWlBQ899BDWrl2L+fPnuzClvFRXV+Oxxx7DH//4R+te6iOPPILw8HB0dnbi2WefxapVq/DAAw+4LOOmTZsgiiLKy8uxZMkSxMbGIiAgwGV5rqa6uhpHjx5FYmKidZqrn8+1a9fiwIEDeOeddwbsMa/HlXK642v0clmd8TqV3R69KIo2hzlGoxGiKPaap7KyEgDQ1dWF5uZmhISEWO+/3LBNzzoCAgKQkZEBvV7v9JzXuuyl06uqqvq8TmdlBYCWlhb86le/wm9+8xtMmTLFOn3o0KEQBAFqtRoLFy7E/v37XZqzZ16tVouZM2eipKQEISEhaGpqQldXFwDHPKf9zQkAX3zxBZKTk+Hl5WWd5ujn81qyfvfdd3jjjTewcuVKqNXqqy7rjNdpf3ICA/cadURWZ7xOZVf08fHxMBgMKC8vh8lkQmFhIXQ6nc08Op3O+g6Ar776CrNmzYIgCAAAi8WCL774wqbou7q6rGN2nZ2d2Lx5M2JiYpye80oSExNRXFyMxsZGNDY2ori4GImJiRg6dCgCAgKwb98+SJKETz75BElJSf3K2d+sJpMJy5Ytw7x583Drrbfa3FddXQ2ge3xx/fr1Ln1OGxsbrYfldXV12LNnD6KjoyEIAm666SZ89dVXALrfRdLXdTojZ4/L7Yw4+vnsa9aSkhLk5eVh5cqVCAsLs04fyNdpf3IO5Gu0v1md9jrt82lbN7J582YpJSVFSkpKkl5//XVJkiTpr3/9q7R+/XpJkiSpvb1devDBB6W5c+dKWVlZUllZmXXZbdu2STk5OTbra21tlRYsWCBlZGRIaWlp0vPPPy91dXU5Pef3338vzZ49W5o8ebI0c+ZMKS0tzbrshx9+KM2dO1eaO3eu9NFHH1mn6/V6KT09XUpKSpJ+97vfSRaLpd85+5P1k08+kSZMmCDddttt1j8lJSWSJEnS4sWLpYyMDCk9PV169NFHpZaWFpfl3L17t5SRkSFlZmZKGRkZ0gcffGBdZ1lZmZSVlSXNnTtXevDBB6WOjg6X5ZSk7nfkJCYmSmaz2Wadzng++5J1yZIl0g9+8APrv++vfvUr67ID+Tq93pwD/RrtT1ZnvU55mWIiIg8nu6EbIiK6Nix6IiIPx6InIvJwLHoiIg/Hoici8nAseiIiD8eiJyLycCx6IiIP9/8BonxFArCvFoUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "alpha_space = np.logspace(-4, 0, 50)\n",
    "ridge_scores = []\n",
    "ridge_scores_std = []\n",
    "\n",
    "for alpha in alpha_space:\n",
    "    ridge.alpha = alpha\n",
    "    ridge_cv_score = cross_val_score(ridge, X_test, y_test, cv=10)\n",
    "    ridge_scores.append(np.mean(ridge_cv_score))\n",
    "    ridge_scores_std.append(np.std(ridge_cv_score))\n",
    "    \n",
    "plt.plot(ridge_scores, ridge_scores_std)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A regressão de Ridge basicamente minimiza uma versão penalizada da função de mínimos quadrados. A penalização encolhe o valor dos coeficientes de regressão. Apesar dos poucos pontos de dados em cada dimensão, a inclinação da previsão é muito mais estável e a variação na própria linha é bastante reduzida, em comparação com a da regressão linear padrão"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQ0AAADSCAYAAABQDZVGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xl0VPed5/33TzvCQhICsUgCIVZhIRYJAQYKEDsYcMDtbZy0254mceJsnWeeZ57Tz5nuk5nnnMzT05mO+3gyodOedLonjh2Pu5vxEgdJVSptpdKOkFgkEKANbQgtaK2q3/PHlcpCBlNlaynB93UOh6p7f/fW9xbio1u/+t37U1prhBDCU37TXYAQYmaR0BBCeEVCQwjhFQkNIYRXJDSEEF6R0BBCeEVCQ7gppaqUUrsfsG63UqphiksSPihgugsQvkNr/eR01yB8n5xpCACUUvILRHhEQuMxppS6rpT6v5RS54G7SqkGpdS+kXWzlFK/Ukp1KqWqgc3jtt2klCpTSvUopX6nlHpXKfWfxqx/WilVrpS6o5TKV0olT+3RickioSFeBI4CEYBjzPK/AJaP/DkI/PHoCqVUEPDPwK+AucA7wNfGrN8IvA18E4gCfgGcVUoFT+JxiCkioSHe1FrXa637xy1/Dvh/tda3tdb1wJtj1m3F6A97U2s9rLX+ALCPWX8a+IXWulBr7dRa/wMwOLKdmOEkNET9A5YvHrfuxrh1jfreqx3Htl0K/Gjko8kdpdQdIG5kOzHDqem4ynXevHk6Pj5+yl9XCPFgJSUl7Vrr+Q9rNy095vHx8RQXF0/HSwshHkApdePhreTjiRDCSxIaQgivSGgIIbzy0NBQSr2tlGpVSl14wHqllHpTKVWrlDqvlNo08WUKIXyFJ2cavwIOfcH6w8DKkT+ngZ9/9bKEEL7qoaGhtbYCt7+gyQng19pgAyKUUosmqkAhhG+ZiD6NGO4d2NMwskwI8Qia0o5QpdRppZQM0BBiBpuI0GjEGCI8KnZk2edorc9orVMn4DWFENNkIkLjLPCNkW9RtgJdWuvmCdivEMIHPXQYuVLqHWA3MG/kdm9/AQQCaK3/O/AxcASoBfqAP5msYoUQ0++hoaG1fvEh6zXwnQmrSAjh02REqBACp9PpcVu5L6QQjzGXy0VlZSWfflrg8TYSGkI8hrTWXLx4kU8/tfLJJwnk5/8x8LpH20poCPEY0VpTU1PDuXPZfPLJYnJyvk5392yv9iGhIcRjoq6ujowMMx9/PBer9Tlu3w4HIDnZyZ49mfzsZ57tR0JDiEdcfX09mZlmPv44BKv1BC0tUQCsXu3iwIFcIiPN+HnxlYiEhhCPqObmZrKyzHzyicZiOUhj4wIAli7VHD1qZ968T/HzM+4RvHix5/d8ltAQ4hHT1taGxWLh44/vYjbv48aNWAAWLdIcP36e+fPPEhDgAiA6Opqvfe1rLFy4kNOnT3u0fwkNIR4RnZ2dWCwWfv/7DszmdGpqEgCIitI888xlFi78gMDA4ZFlUZw6dYpFi4y7WHR0dHj8OhIaQsxw3d3dZGdn8+mnjWRl7eHixdUAhIVpTp26zuLF7xEUNABAREQEp06dIjbWOPtoamriww8/pLnZ88vFJDSEmKF6e3vJzc3l00+vkpVlorLyabRWzJoFp041sWTJbwkK6gFgzpw5nDp1iiVLlgBQW1vL73//e/cZRnh4uMevK6EhxAzT399Pfn4+f/hDNZmZT1FWdgCXy4/AQDh5sp34+P/JrFl3AJg9ezYnT54kISEBrTVlZWWYzWZ6eowwmT9/PkeOHCE+Pp4f/vCHHr2+hIYQM8Tg4CA2m41z5yrIzEyjpOTbDA/74+cHX/taFytX/obQ0FYAZs2axTPPPMOqVasYGhoiOzubgoICBgcHAYiJieHYsWMsWLDA6zokNITwccPDwxQVFXHuXBGZmRux219ncDAQgKNH75KY+FueeKIBgODgYI4fP87atWvp6uriww8/pLy83H1B2rJlyzh69ChRUVFfuh4JDSF8lNPppKSkhIyMArKykigo+BZ9fcEA7N07QHLy7wgPvwZAYGAgR44cYcOGDTQ0NPCb3/yGmpoa975Wr17NoUOHiIiI+Mp1SWgI4WNcLhcVFRVkZOSSlbWS/PzTdHfPAmD79iFSU/+FyMiLAAQEBLBv3z42b95MdXU1Z86cueebkKSkJPbu3TshYTFKQkMIH6G1pqqqiszMbDIz48jN/RM6O58AICXFwbZtHxEVVY5S4O/vz+7du0lNTaWkpIS/+Zu/cXduKqVITk5m9+7dExoWoyQ0hJhmWmsuX75MZqYZszkaq/Xf0NZm/Gdfu9bFrl1/IDq6EKXAz8+Pp556ivXr12O32/npT3/K8LAxYEspxYYNGzCZTJMSFqMkNISYJlprrl27RmZmFmZzGFbrH9HUNA+A5ctd7N1rZeHCbPz8jEBISUlhzZo12O123nrrLZRSgLFu48aN7Ny5c1LDYpSEhhDT4MaNG2RmZmGx+JOdfYybNxcCEBen2b/fRkxMBv7+LpRSJCUlER8fj91up7i4mICAAHdgTGVYjJLQEGIKNTY2YjabMZsHyc7ez9WrxgjN6GjN4cOlxMZ+QmCg8fXoihUrWLRoEaWlpVRWVjJr1iyUUjidzmkJi1ESGkJMgdbW1pGw6CQ7ez+XLi0HICJCc/RoFUuWnCU42OibiI2NZe7cuVRVVVFbW0t4eDh+fn4MDAywYcMGdu7cSWRk5LQdi4SGEJOoo6MDi8WC2dyM1bqXyspEAGbP1hw9WktCwgeEhBgXk82bN4/Q0FBu3rzJrVu3iIqKor29nZ6eHtavXz/tYTFKQkOISXDnzh2ys7OxWK5jte6mvPwkLpciOFhz9OgNVqz4X4SG9gLGxWT+/v60t7cze/ZsYmNjaWpqoq2tzSfOLMbzKDSUUoeAnwH+wC+11j8Zt34J8A9AxEibf6+1/niCaxXC5/X09JCTk4PFchmrdQclJcdwOPzw99c8/XQTq1e/R1hYF2BcH+J0Ounu7mbBggUkJCRw/fp1+vr6fDIsRnkyLaM/8BawH2gAipRSZ7XW1WOa/T/Ae1rrnyul1mJM1Rg/CfUK4ZP6+vrIy8vDYqnEat2K3f49hob8UUpz4EAra9f+joiIdsAY8u1wOBgYGGDFihUEBARw+fJlWltbfTosRnlyppEG1GqtrwEopX4LnADGhoYG5ow8DgeaJrJIIXzVwMAABQUFWCylWK0p2O3fpa/PuJhs9+7brFv3PlFRxrBuPz8/XC6Xe1yF0+mkqqoKp9PJ+vXrMZlMPh0WozwJjRigfszzBmDLuDZ/CfxBKfVdYDaw7347UkqdBk6npKR4X6kQPmRoaAi73Y7FUojVug6b7Q16eoyLybZu7WbTpg+Ijr5xzzZz5sxhw4YN9PT0UFFRMePCYtREdYS+CPxKa/3XSqltwD8qpZK01q6xjbTWZ4AzqampeoJeV4gp5XA4KC4uxmLJJzd3NXl5r3PnTigAGzbcZfPmfyEmpvaebZYuXUpycjItLS3k5ua6w2Lnzp3MnTt3Og7jK/EkNBqBuDHPY0eWjfUacAhAa12glAoB5gGtE1GkENPN6XRSXl6OxZJDbm48ubl/Snt7GACJif1s3fohS5ZUMzJQE4B169aRnJxMbW0tn3zyyYwPi1GehEYRsFIptQwjLF4AXhrX5iawF/iVUioRCAHaJrJQIabD6ATJZnM2+fkLyc39Y5qbjY8SCQmDbN/+KcuWlbknGwoICGDr1q2sXbuWiooK3n333UcmLEY9NDS01g6l1BvApxhfp76tta5SSv0YKNZanwV+BPydUuqHGJ2ir2it5SOImLFGJ0g2my0UFIRjtb5Iff18AGJjh9m+PYPVq4vckw2FhISwd+9eli9fjt1u5+23337kwmKUmo7/26mpqbq4uHjKX1eIh9FaU1tbS1ZWFjZbMFbrAa5dM2Yfi452sHOnhcTEAvdkQ2FhYRw9epTFixeTn59PcXHxjA0LpVSJ1jr1Ye1kRKgQI+rq6jCbzdhsTnJyDnHp0lIAIiOdbN+eS3JyLkFBjpFlkZw6dYrw8HDy8vJ4//33cTqdJCcnYzKZZlRYeEtCQzz2GhoaRs4sesnJOcCFCysAmD3bxfbtBWzaZCUkZAgwZiZ74YUXCAkJIS8vz31mkZyczM6dO7/SDXtnCgkN8di6desWZrOZgoI2cnP3UV6eiNaKkBAXW7cWk5ZmITS0HzDC4qWXXiIoKOixDYtREhrisTM6QXJBQT15eemUlLyA06kICHCxeXMZTz1lJizMuJhs7ty5vPjii4SEhJCfn09RUdFjGxajJDTEY6Ozs5Ps7Gzy82vJzzdht58cmWxIk5Jynh07soiMNC4mi4yM5Nlnn2XOnDkSFuNIaIhHXnd3N1arlfz8avLzn6Kw8GkGBowf/aSkanbtymL+/M/mND1x4gTR0dHk5eW5w2LdunWYTKbHOixGSWiIR9bdu3fJyckhP7+CgoLNFBT8kLt3jYvJVq26wp49WSxa1AJAaGgox44dIy4ujry8PN555x0cDoeExX1IaIhHzugEybm5xdhs6yko+D5dXSEAxMfXkZ6exZIlxjSGISEhHDx4kBUrVlBQUMAHH3zgDoudO3cyb9686TwUnyShIR4Zg4ODFBYWkpNTQGFhIvn536Wjw7iYLCamgfT0LBIS6lDKuKdFeno6SUlJFBQU8PHHH0tYeEhCQ8x4oxMk5+TkUVSUQG7u67S0GLd3iY5uIT09i9Wrr7hnJtuxYwcpKSnYbDbefPNNCQsvSWiIGcvpdFJaWkp2tpWSksXk5r5GQ4MxEnPu3A727LHw5JMX3JMNpaWlsW3bNux2O3/7t3+Lw+EgKSkJk8kkYeEFCQ0x44xOkGyxZFNWNher9WVu3FgAQHh4FyZTNhs2VODvb1wfMjpVYXFxMW+99RbDw8PuDk4JC+9JaIgZY3SCZIvFQlnZLKzWU9TWGrd6mT27l507c0hJKXFPNpSYmMjevXspKyvj5z//uYTFBJHQED5vdIJks9lMRYUiO/uwe7KhkJABtm/PIy2t0D3ZUEJCAvv37+fChQv84he/cIfFzp07mT9//nQeyiNBQkP4rNEJko2wGCA7O53KyrUABAUNsXVrIdu25TFr1iBgzEx28OBBLl26xNtvvy1hMUkkNMSXt3u38bfFMuG7vnnzJpmZmVRUdGG17qG8fB1a+xEQ4GDz5mK2b8/hiSf6AFiwYAEHDx7k6tWr/PrXv2Z4eNjdwSlhMfEkNIRPaWpqIiMjg/Pn28jJ2UlxcQoulz9+fi42bSrBZMomPLwHMK4POXjwIPX19bzzzjsSFlNEQkN8TklJCRkZGQDs27ePsVNOlJSUEP3880SEhxNWWgpA/YoVRISHc+XMGaxWKyaTCcD92JMpK1pbW/nDH/5AZWUTeXnbKSx8EYcjEKU0ycmV7NplJiqqEzDulrVv3z7a2tr44IMPGBoakrCYQhIa4nOsVisDAwPux2P/01utVr7mcHCnq4uwkWXOkedWq9V9cRh8dqHYF4VGe3s7n3zyCdXVDdhs2ygoeJbBQWPId2LiJXbvzmLBAuMe1bNmzWLPnj10dXXx0UcfSVhMEwkN8Tkmk8l9pjF61jB23T+P/B33ox/R09PDP7/8MiaTCRPc90zjfjo6Ovjwww+pqWnAbt9MXt4p+vqMId8rVlxlz54sYmKMifqCg4PZsWMHg4ODZGRkSFhMM7mxsPjyvkRHaFtbG2fPnuX69SbKyjZhtZro6THOWZYsqWfv3iyWLr0OfDYdgNaaoqIihoaGePLJJzGZTERHR0/ssQi5sbCYAl6ERUNDAx999BFNTS2cP5+M1fost2+HA7Bo0S3S0zNZufIqoPH392fTpk0EBgZit9slLHyMhIaYNKODss6dO0d7+20uXkzEav0jWlqM60Pmz29j924za9deQimNUoqkpHWEhYVRXFwsYeGjPAoNpdQh4GcYkyX9Umv9k/u0eQ5jImgNVGitx8/CJh4TDoeD8vJysrOz6enppbZ2BdnZL9DQYPQ/REbeYdcuMxs2VAHGkO/Vq1cTGRlJaWmphIWPe2hoKKX8gbeA/Rgzxhcppc5qravHtFkJ/N/Adq11p1JK/qUfQ729vdjtdgoLCxkaGuL69aXk5LzE1auLAAgL68FkspKaWoFSxpDvZcuWER0dTVlZGZcvX2bt2rXs2rVLwsKHeXKmkQbUaq2vASilfgucAKrHtPlT4C2tdSeA1lomfn6MtLS0kJ+fz4ULF3C5XDQ2LiY39xAXLxoXk4WG9rF9ey7bt5cDxpQAMTExLF68mIqKCurq6iQsZhBPQiMGqB/zvAHYMq7NKgClVB7GR5i/1Fr/fkIqFD5Ja01NTQ0FBQVcv34dgJaW+eTnH6aiYhkAwcGDbNuWj8lUip+fMSVAdHQ0sbGxXLhwgcbGRtauXYvJZGLBggXTdSjCSxPVERoArAR2A7GAVSm1Tmt9Z2wjpdRp4LQnIwSFbxoaGqK8vJzCwkJu376Nn58fHR2R2GyHKC5eidaKwMBh0tLs7NlTREDAZ1MCLF26lIsXL9La2iphMYN5EhqNQNyY57Ejy8ZqAAq11sNAnVLqCkaIFI1tpLU+A5xJTU2VGeVnmK6uLux2O6WlpQwMDBAYGEhXVxh2+wFstidxOhX+/k42bSrhwIEiAgPbAWPI99KlS6mpqaG8vFzC4hHgSWgUASuVUsswwuIFYPw3I/8CvAj8D6XUPIyPK9cmslAxPRoaGrDZbFRXG11YISEh9PaGYrfvxWbbwNCQH0q52LChnMOHiwgObgaMId/x8fFcu3aNCxcukJiYyK5duyQsHgEPDQ2ttUMp9QbwKUZ/xdta6yql1I+BYq312ZF1B5RS1Rjfof07rXXHZBYuJo/L5eLixYvYbDYaGhoIDAwkLCyMlpZBcnJM5OVtpr/fH4Ann7zA0aNFhIbeBCAoKIglS5ZQX1/PxYsXJSweQTKMXLgNDAxQWlqK3W6nq6uLOXPmEBwcTGPjHUpLd5Cb+xS9vcbvmZUrr3D8uJ2wsKuAMeQ7Li6OpqYmBgcHJSxmIBlGLjx2+/ZtbDYb5eXlDA8Ps3jxYubMmcO1a01UVKwjL89EZ2cQYEw2dPy4nXnzruByufDz8yMmJobW1lbq6upITEzEZDKxcOHCaT4qMVkkNB5TWmtu3LiBzWbj8uXL+Pn5sWrVKpxOJ5cvX6WyMoXc3JdoazMuU4+JaeDEiSIWLqzC5XKitWLRokXcvn2b+vp6CYvHiITGY8bhcHDhwgUKCwu5desWoaGhbNmyhf7+fs6fv0BVVRJ5eX9GU9NswJhs6JlnioiLq8DpdOByGWMturq6aG5uZs2aNezatUvC4jEiofGYuHv3LsXFxRQXF9Pb28v8+fM5cOAAd+7coaiomEuXVpKX931u3DBmJps7t4Pjx0tYsaIEh2MIpxOioqLo7e2ltbVVwuIxJqHxiGttbcVms3H+/HmcTicrVqxg06ZNNDY2kpVl5sqVOAoKvk1NjXHl6Zw5XRw7VsqTTxYzNNSHwwERERH09/fT0dEhYSEkNB5FWmtqa2ux2Wxcu3aNgIAANmzYwMaNG7l69Sr/+q//Sk3NfAoL/y1VVca1HrNn93L4cBmbNhUzNNTN0JAxMGtwcJA7d+6wZs0aTCYTixYtmuajE9PtK4eGUuq7wD+NXqwmps/w8DAVFRUUFhbS3t5OWFgY6enpJCcnU11dzW9+8xuuXg3Dbv86ZWUxAISE9HPgQAXbt5fQ39/O0BCEhobicDjo6emRsBCfMxFnGgswLpcvBd4GPtXTMfjjMdbd3e0e4t3f38/ixYs5efIkq1evpqKigl/+8pdcvx6M3f5H2O3xAAQGDrF3byV79pTQ399Mf78x2tPlctHX18fq1avZtWuXhIX4nAkZ3KWUUsAB4E+AVOA94O+11lfv114Gd02MpqYmbDYbVVVVaK1Zs2YNW7duJSYmhsrKSrKzs6mr0xQXH6GgYCUul8Lf38GuXVUcPFhGf/8NAAIDAwHjTEXC4vE1pYO7tNZaKXULuAU4gEjgfaXUOa31/zkRryEMLpeLS5cuYbPZqK+vJygoiLS0NNLS0oiIiKCqqmrkxr2DFBcfJC9vLQ6HH35+TnbsuMSxYxX099fQ32+M4lRKSVgIr0xEn8b3gW8A7cAvMa47GVZK+QE1gITGBBgYGKCsrAy73c6dO3eIiIjg4MGDbNy4kaCgIK5cucK7775LXV03ZWX7sFrXMzjoD2jS0i5z6tQF+vsv0N8P/v7+KKVwOBwSFsJrE3GmMRc4qbW+MXah1tqllHp6Avb/WOvs7KSwsJCysjKGhoZYsmQJBw4cYPXq1Sil3BMkX73aRkXFHrKzU+jrMz5urF9/jRdeqGZoqJT+fuPGvf7+/u6wMJlMLF68eJqPUMw0Xzk0tNZ/8QXrLn7V/T+OtNbcvHkTm83GpUuX8PPzIykpiS1btrj/k9+8eZOsrCxqaxs5f34nFss36Okxrg9JTLzJyy9fxum0MTjocoeF0+kkISGBXbt2SViIL03GafgQp9NJVVUVNpuN5uZmZs2axY4dO0hLSyMszJhQqKmpCbPZzKVL16iq2orF8gKdncb1IQkJTbzySi1+fjkMDzsA3GGxfPlyCQsxISQ0fEBfXx/FxcUUFRXR29vLvHnzePrpp0lOTnZ/s9Ha2orZbKa6+jIXL6Zgtf4ZLS3G9SGxsW28+upVgoPNDA8P4XKBn58fLpdLwkJMOAmNadTW1uYe4u1wOFi+fDknTpxg+fLlGN9iG5etWywWKioqqalZh9X6AxobjetDFi68zSuv1BERkcXAQB/Dw5+FxYoVKyQsxKSQ0JhiWmuuXr2KzWbj6tWrBAQEkJyczJYtW+65fX9XVxfZ2dmUlZVz9epKcnO/S12dcX3I3LldvPrqTRYsyODu3W4GBkAphdZawkJMOgmNKTI8PMz58+cpLCykra2NJ554gj179pCSksLs2bPd7Xp7e8nJyaGkpIS6ujhyc1/nyhVjZrI5c3p55ZV64uOz6O5u5+7dz4dFTEzMdB2ieExIaEyynp4eioqKKC4upr+/n4ULF/LMM8+QlJSEv7+/u11fXx95eXnY7XZu3IjGZnuVykrjbGH27H6+/vVGEhPNdHY20d0tYSGmj4TGJGlubsZms7lnHRsd4r1kyRJ3fwUYg7ZsNhsFBQU0NIRTWPgyJSVLAQgJGeTFF5vZtMlKR0cdnWMuCVy+fDm7d++WsBBTTkJjArlcLq5cuYLNZuPGjRsEBQWRmprKli1bmDt37j1th4aGsNvt5Ofn09AQQlHRcxQWJrgnG3r22Vvs3FlAa+tFOsbc133FihUSFmJaSWhMgMHBQfcQ787OTsLDwzlw4AAbN24kJCTknrYOh4OSkhJycnJoavKjpOQYubmrcbn88Pd3cuJEC/v3F9HSUk7rmBlxJSyEr5DQ+Aru3LnjHuI9ODhIXFwc+/btY82aNfj5+d3T1ul0UlFRQXZ2Nk1NDkpLD2K1rsXhCEApF0eOtHDiRAXNzQW0tHy23ejHkNjY2Ck+OiHuT0LDS1pr6uvr3UO8lVKsXbvWfUn6eC6XiwsXLpCdnU1j413Ky/dhsaxncNAYtJWe3s7zz1dz65aF5ubPblOQkJDAnj17JCyEz/EoNJRSh4CfYcyw9kut9U8e0O4U8D6wWWv9SN0ww+l0Ul1djc1mo6mpiZCQEJ566inS0tKYM2fO59prrbl06RJms5nGxjucP7+bzMwU+vuDAdi+/TZf/3oNbW3naG52urdbtmwZ6enpEhbCZz00NJRS/sBbwH6MiZ6LlFJntdbV49qFAd8HCiej0OnS399PSUkJdrudnp4eoqKiOHLkCOvXrycoKOhz7Ufvz2k2m7l5s5Wqqh1kZqbR0xMKwMaNXXzzmzfo6PiQW7eG3dvFx8ezd+9eCQvh8zw500gDarXW1wCUUr8FTgDV49r9R+A/A/9uQiucJu3t7dhsNioqKnA4HCQkJHDs2DFWrFhxz1emY12/fp2srCyuX2/g0qWtZGS8SGencaFZYmIP3/lOEz09/8KtWwPubZYuXcq+ffskLMSM4UloxAD1Y543AFvGNlBKbQLitNYfKaVmbGhoramrq8Nms1FTU4O/v797iPcXzUna0NCA2WymtvYaNTWbyMg4SVtbBAAJCXf5/vfbGBx8n/b2u+5t4uLi2L9/P3FxcZN+XEJMpIm4c5cf8FPgFQ/angZOp6SkfNWXnVAOh8M9xLu1tZXZs2eze/duUlNT7xniPd6tW7cwm81cvnyFurokMjLeoKkpCoDY2D6+//1O/Px+R2dnl3ub2NhYDhw4IGEhZixPQqMRGPsTHjuybFQYkARYRk7bFwJnlVLHx3eGaq3PAGdSU1N94m7lvb297iHefX19LFiwgBMnTpCUlERAwIPfmvb2diwWCxcuVFFfv4rMzG9x44ZxJhIdPcAPftDF7Nn/i87ONvc2ixYt4vDhwxIWYsbzJDSKgJVKqWUYYfEC8NLoSq11FzBv9LlSygL8H7787cmtW7fcQ7ydTierVq1i69atxMfHP7C/Aoxb72VnZ3P+/HkaG5dgNr9GTY3RFxEZOcT3vtfLvHkf0NHRyNCQsc3ChQs5fPgwS5YsmYpDE2LSPTQ0tNYOpdQbwKcYX7m+rbWuUkr9GCjWWp+d7CIngtbaPcT7+vXrBAYGsmnTJrZs2UJUVNQXbtvd3Y3VaqWsrIxbtxaSnf0NqqriAXjiiWHeeKOPpUvP0tJyzT3kOzo6mqNHj0pYiEfOhMx74q2pnPdkaGiI8vJyCgsLuX37NnPmzCEtLY1NmzYxa9asL9z27t275ObmUlxcTEtLBLm5BygrWwlASIiDb31rgMTEj2lu/uxWqFFRURw7doylS5dO6nEJMdGmdN4TX9TV1UVhYSGlpaUMDg4SGxtLeno6iYmJnxviPV5/fz8FBQXYbDZaW0MpKPgaRUVr0NqPwEAnr702RGpqBg0NpTQ3G9tERkZy7Ngxli1bNgVHJ8T0eeRCo6GhAZvNRnW1MYxkdIi3J+MghoaCZzwUAAANJElEQVSG3Jept7UFUFh4hIKCdTid/vj7u3j55QFMJisNDTYaGowztPDwcI4dO8by5csn9biE8BWPRGi4XC73EO/GxkaCg4PZtm0baWlphIeHP3T74eFhiouLyc3Npb1dU1SUTm7uBoaHA1FK89xzgxw6VEh9vYX6eiMswsLCOHbsGCtXrpzswxPCp8zo0Ojv76e0tBS73U53dzdz587l8OHDbNiw4b5DvMdzOp2UlpaSk5NDW9sgpaUmsrNTGBw0Lmd/+ukhTp6soKHh99y86QJg9uzZPP3006xZs2ZSj00IXzUjQ6Ojo4PCwkLKy8sZHh5m2bJlHDlyhFWrVn3hV6ajXC4X58+fJzs7m7a2XsrLt2M2p9HXZ1wfkp4+zEsvXaS5+Sw3bxoXk82aNYsjR46QlJT08AJ37zb+tli+5BEK4btmTGhorbl+/To2m40rV67g7+/PunXr2LJlCwsXLvR4H9XV1ZjNZlpaOqmq2kpGxlZ6eozrQ7Ztc/Dqq9dobX2fhgbjYrLg4GAOHTrEhg0bJu3YhJhJfD40HA4HlZWVFBYW0tLSQmhoKCaTic2bN/PEE088dPuSkhKsViurVq2ivr6e5uZWLl5MISPj39DZGQnA+vVOvvnNeu7c+S2NjYMABAUFsX//flJTH/oN1GdGzzCys+95XvLXf01GRgYA+/btIyUlhZKSks8tu1/dJpPpc+uEmE4+Gxp37951D/G+e/cu0dHRHD9+nHXr1n3hEO+xtNZkZWXR19dHUVExNTXJnDt3irY2Y0qA1audfOc7LfT3/09aW/sACAgIID09nW3btk3YsVitVgYGBtyPU1JS7rts/Dajg8okNIQv8bnQaGlpwWazUVlZidPpZOXKlWzdupVly5Z51F8xanSC5Lt3+3A6A8jL24bZnA7A0qVOvvvd22j9T9y+3Q0Yc56aTCZ27tzp1evcY7QPY1yfhmnMWYXJZHL/PX7ZWCaTyX2mIYQv8YkRoVprampqsNls1NXVERgYyPr169myZQvz5s37gj193ugEybW1tdTXL+Pcud3cvGkM5V640MX3vtdNUNA/0tt7GzCmMXzqqadIT0//8mExnnSEihloRowIHRoaoqKigsLCQjo6OggLC2Pv3r2kpKQ8dIj3eK2trVgsFi5evMitW3GcO/cyV68aA67mznXxve/dJSzsn+jpaWVoyJhsaPPmzRw8ePChI0S9JmEhHmHTEhpOp5OMjAxKSkoYGBggJiaGU6dOkZiYeM+sY54YnSC5srKS9vYFZGY+z8WLxhiKsDAX3/72AAsWvEN3dwM9PUZYrF+/nqNHj3rcNyKE+My0/K9pbW0lPz+fxMRE9xBvbz8afDZBchm3b8/FYjlJZWUSWitmzdKcPj1EfPx7dHVdo9votuDJJ5/k2LFjBAcHT8JRCfF4mJY+jVWrVmm73U5ERITX245OkFxcXMydO09gtZooLd2Ay+VPYKDm1VeHWbPmn+nqujT29XjmmWe8/sgjxOPEp/s05syZ43VgjE6QXFhYSFdXMHl5+7HbU3E6A/Dz03z96w42bvzfdHefp2vk7nrx8fGcPHmSsLCwSTgKIR5PPv+hfnBwkIKCAgoKCujqUhQU7KKgYAvDw8a1JadOOdm27ff09ha7P4aM9pFERkZOY+VCPJp8NjSGhoYoKioiNzeX7m4ndvtWcnO3MTBgfMQ4csTF7t2Z9PXl09trbLNgwQJOnjxJdHT0NFYuxKPN50JjdIJkY0TkIKWlqWRn7+DuXWPI+K5dLg4dymFoKJu+PqM/Zu7cuTzzzDNy014hpoDPhMboBMkWi4U7d3o5f34jZrOJ7m7jfhibN7s4ftyO0/kHBgeNsJgzZw4nTpwgISFhOksX4rEy7aExOkGyxWKho6OT6up1ZGXt4vZt42a/SUmakydL8ff/GKfTuKdFaGio+54WEzaKUwjhkWkLjdEJkrOysmhra+fKldVkZj5Pa6sxf8jKlZpnn60kOPhfARdaG5epHz58mOTkZAkLIabJtITG4OAgZ86cobn5FnV1CWRmnqCx0biHZ1yc5rnnLhEW9gHgACAwMJB9+/aRmpo68UO+hRBemZbQ6OjooKgokMzMb3D9unH37uhozfPPXyUq6n3AuKeFv78/u3btYtu2bTLkWwgfMS0jQkNCkvTg4AUAIiI0zz13nYULf4efXz9gXHm6bds2du7cKUO+hZgiEzoiVCl1CPgZxgxrv9Ra/2Tc+j8D/i3G54k24FWt9Y0H7W9wMITZszXPPttAXNx7BAT0utelpKSwd+9eGfIthI96aGgopfyBt4D9QANQpJQ6q7WuHtOsDEjVWvcppV4H/j/g+QftMzJymNdff5OgoDvuZUlJSRw4cECGfAvh4zw500gDarXW1wCUUr8FTgDu0NBam8e0twEvf9EOQ0La3IGxatUqDh06JEO+hZghPAmNGKB+zPMGYMsXtH8N+OR+K5RSp4HTixYtYunSpRw5ckSGfAsxw0zo95dKqZeBVOCv7rdea31Ga506f/58XnnlFQkMIWYgT840GoGxF3XEjiy7h1JqH/DnwC6t9eAX7TAwMNCbGoUQPsSTM40iYKVSaplSKgh4ATg7toFSaiPwC+C41rp14ssUQviKh4aG1toBvAF8ClwE3tNaVymlfqyUOj7S7K+AJ4DfKaXKlVJnH7A7IcQM59E4Da31x8DH45b9hzGP901wXUIIHyUXcgghvCKhIYTwioSGEMIrEhpCCK9IaAghvCKhIYTwioSGEMIrEhpCCK9IaAghvCKhIYTwioSGEMIrEhpCCK9IaAghvCKhIYTwioSGEMIrEhpCCK9IaAghvCKhIYTwioSGEMIrEhpCCK9IaAghvCKhIYTwioSGEMIrHoWGUuqQUuqyUqpWKfXv77M+WCn17sj6QqVU/EQXKoTwDQ8NDaWUP/AWcBhYC7yolFo7rtlrQKfWegXwX4H/PNGFCiF8gydnGmlArdb6mtZ6CPgtcGJcmxPAP4w8fh/Yq5RSE1emEMJXeBIaMUD9mOcNI8vu22Zk7tcuIGoiChRC+JYp7QhVSp1WShVP5WsKISaWJxNANwJxY57Hjiy7X5sGpVQAEA50jN+R1voMcEYp1aaUCsWYhX4mSWRm1TzT6gWpeSo8qN6lnmzsSWgUASuVUsswwuEF4KVxbc4CfwwUAM8CWVpr/aAdaq3nK6WKtdapnhTpK2ZazTOtXpCap8JXrfehoaG1diil3gA+BfyBt7XWVUqpHwPFWuuzwN8D/6iUqgVuYwSLEOIR5MmZBlrrj4GPxy37D2MeDwB/5OVrn/GyvS+YaTXPtHpBap4KX6le9QWfIoS4L6VUHGAFUrTWt5VSkUApsEdrfX1aixOTToaRC69preuBnwM/GVn0E+CMBMbjYdJDY6YNQfeg3j9TSlUrpc4rpTKVUh71OE+mh9U8pt0ppZRWSk1Ep91/BbYqpX4A7AD+izcbe1KzUuq5kfe6Sin1mwmo+Uvz4OdiiVLKrJQqG/nZODIddY6p522lVKtS6sID1iul1Jsjx3NeKbXJ451rrSftD0bH6VUgAQgCKoC149p8G/jvI49fAN6dzJomoN49QOjI49ens15Pax5pF4bxkcIGpE7Qax8ENLB/Et7nlUAZEDnyPNqX32OMfoLXRx6vBa5P88+FCdgEXHjA+iPAJ4ACtgKFnu57ss80ZtoQ9IfWq7U2a637Rp7aMMatTCdP3mOA/4hxTdDABL72YaAZSPJyO09q/lPgLa11J4DWuvWrFvsVeFKvBuaMPA4Hmqawvs/RWlsxvsl8kBPAr7XBBkQopRZ5su/JDo2ZNgTdk3rHeg0jrafTQ2seOfWM01p/NFEvqpTaAOzH+C31Q09/4EZ48j6vAlYppfKUUjal1KGvVPBX40m9fwm8rJRqwPim8btTU9qX5u3Pupt0hH5JSqmXgVTgr6a7li+ilPIDfgr8aAL3qTA6Qn+gtb6J8R541afhgQCMjyi7gReBv1NKRUzwa0ykF4Ffaa1jMU79/3HkvX/kTPZBeTMEnS8agj5FPKkXpdQ+4M+B41rrwSmq7UEeVnMYxscHi1LqOsaZwdmv2Bn6p8BNrfW5kef/DUhUSu2aoJrB+M13Vms9rLWuA65ghMh08KTe14D3ALTWBUAIMG9KqvtyPPpZv69J7owJAK4By/isA+nJcW2+w70doe9NY+eRJ/VuxOgUWzlddXpb87j2FiaoI3SS3+dDwD+MPJ6HcSod5cP1fgK8MvI4EaNPQ03z+xzPgztCj3JvR6jd4/1OQeFHMH5LXAX+fGTZjzF+S4ORyL8DagE7kDDNb/TD6s0AWoDykT9np7NeT2oe13baQ8PD91lhfKyqBiqBF3y83rVA3kiglAMHprnedzA6qYcxztpeA74FfGvM+/vWyPFUevMzISNChRBeeSQ7aoQQk0dCQwjhFQkNIYRXJDSEEF6R0BBCeEVCQwjhFQkNIYRXJDSEEF75/wEHcSDZArzwCAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 288x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn import linear_model\n",
    "\n",
    "np.random.seed(0)\n",
    "\n",
    "classifiers = dict(ols=linear_model.LinearRegression(),\n",
    "                   ridge=linear_model.Ridge(alpha=.1))\n",
    "\n",
    "fignum = 1\n",
    "for name, clf in classifiers.items():\n",
    "    fig = plt.figure(fignum, figsize=(4, 3))\n",
    "    plt.clf()\n",
    "    plt.title(name)\n",
    "    ax = plt.axes([.12, .12, .8, .8])\n",
    "\n",
    "    for _ in range(6):\n",
    "        this_X = .1 * np.random.normal(size=(2, 1)) + X_train\n",
    "        clf.fit(this_X, y_train)\n",
    "\n",
    "        ax.plot(X_test, clf.predict(X_test), color='.5')\n",
    "        ax.scatter(this_X, y_train, s=3, c='.5', marker='o', zorder=10)\n",
    "\n",
    "    clf.fit(X_train, y_train)\n",
    "    ax.plot(X_test, clf.predict(X_test), linewidth=2, color='blue')\n",
    "    ax.scatter(X_train, y_train, s=30, c='r', marker='+', zorder=10)\n",
    "\n",
    "    ax.set_xticks(())\n",
    "    ax.set_yticks(())\n",
    "    ax.set_ylim((0, 1.6))\n",
    "    ax.set_xlabel('X')\n",
    "    ax.set_ylabel('y')\n",
    "    ax.set_xlim(0, 2)\n",
    "    fignum += 1\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 7. Referências   \n",
    "\n",
    "> [Coeficiente de correlação de Pearson](https://pt.wikipedia.org/wiki/Coeficiente_de_correla%C3%A7%C3%A3o_de_Pearson#Refer%C3%AAncias)\n",
    "\n",
    "> [Why isn't Logistic Regression called Logistic Classification?](https://stats.stackexchange.com/questions/127042/why-isnt-logistic-regression-called-logistic-classification)\n",
    "\n",
    "\n",
    "\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}