aula_ml.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = \"all\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn import dummy, metrics, linear_model, model_selection, neighbors\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introdução ao Aprendizado de Máquina"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Trabalhando com dados"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Carregando os datasets\n",
    "\n",
    "Nós iremos ver duas vertentes do mesmo dataset, com propósitos diferentes. `Weather` é um dataset educacional (composto por dados fictícios) amplamente usado para ensinar conceitos de aprendizado de máquina. Ele contém 14 instâncias, 4 features e é considerado um dataset de classificação binária."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "      <th>play</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>sunny</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>sunny</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>True</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>overcast</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>rainy</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>rainy</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>rainy</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>True</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>overcast</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>True</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>sunny</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>sunny</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>rainy</td>\n",
       "      <td>mild</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>sunny</td>\n",
       "      <td>mild</td>\n",
       "      <td>normal</td>\n",
       "      <td>True</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>overcast</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>True</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>overcast</td>\n",
       "      <td>hot</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>rainy</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>True</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     outlook temperature humidity  windy play\n",
       "0      sunny         hot     high  False   no\n",
       "1      sunny         hot     high   True   no\n",
       "2   overcast         hot     high  False  yes\n",
       "3      rainy        mild     high  False  yes\n",
       "4      rainy        cool   normal  False  yes\n",
       "5      rainy        cool   normal   True   no\n",
       "6   overcast        cool   normal   True  yes\n",
       "7      sunny        mild     high  False   no\n",
       "8      sunny        cool   normal  False  yes\n",
       "9      rainy        mild   normal  False  yes\n",
       "10     sunny        mild   normal   True  yes\n",
       "11  overcast        mild     high   True  yes\n",
       "12  overcast         hot   normal  False  yes\n",
       "13     rainy        mild     high   True   no"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "      <th>play</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>sunny</td>\n",
       "      <td>85</td>\n",
       "      <td>85</td>\n",
       "      <td>False</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>sunny</td>\n",
       "      <td>80</td>\n",
       "      <td>90</td>\n",
       "      <td>True</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>overcast</td>\n",
       "      <td>83</td>\n",
       "      <td>86</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>rainy</td>\n",
       "      <td>70</td>\n",
       "      <td>96</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>rainy</td>\n",
       "      <td>68</td>\n",
       "      <td>80</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>rainy</td>\n",
       "      <td>65</td>\n",
       "      <td>70</td>\n",
       "      <td>True</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>overcast</td>\n",
       "      <td>64</td>\n",
       "      <td>65</td>\n",
       "      <td>True</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>sunny</td>\n",
       "      <td>72</td>\n",
       "      <td>95</td>\n",
       "      <td>False</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>sunny</td>\n",
       "      <td>69</td>\n",
       "      <td>70</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>rainy</td>\n",
       "      <td>75</td>\n",
       "      <td>80</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>sunny</td>\n",
       "      <td>75</td>\n",
       "      <td>70</td>\n",
       "      <td>True</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>overcast</td>\n",
       "      <td>72</td>\n",
       "      <td>90</td>\n",
       "      <td>True</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>overcast</td>\n",
       "      <td>81</td>\n",
       "      <td>75</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>rainy</td>\n",
       "      <td>71</td>\n",
       "      <td>91</td>\n",
       "      <td>True</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     outlook  temperature  humidity  windy play\n",
       "0      sunny           85        85  False   no\n",
       "1      sunny           80        90   True   no\n",
       "2   overcast           83        86  False  yes\n",
       "3      rainy           70        96  False  yes\n",
       "4      rainy           68        80  False  yes\n",
       "5      rainy           65        70   True   no\n",
       "6   overcast           64        65   True  yes\n",
       "7      sunny           72        95  False   no\n",
       "8      sunny           69        70  False  yes\n",
       "9      rainy           75        80  False  yes\n",
       "10     sunny           75        70   True  yes\n",
       "11  overcast           72        90   True  yes\n",
       "12  overcast           81        75  False  yes\n",
       "13     rainy           71        91   True   no"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "weather_nominal = pd.read_csv('./weather.nominal.csv')\n",
    "weather_nominal\n",
    "\n",
    "weather_numeric = pd.read_csv('./weather.csv')\n",
    "weather_numeric"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Aqui vamos conferir os possíveis valores únicos em colunas categóricas:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Outlook: ['sunny' 'overcast' 'rainy']\n",
      "Temperature: ['hot' 'mild' 'cool']\n",
      "Humidity: ['high' 'normal']\n",
      "Windy: [False  True]\n",
      "Play: ['no' 'yes']\n"
     ]
    }
   ],
   "source": [
    "print('Outlook:', weather_nominal.outlook.unique())\n",
    "print('Temperature:', weather_nominal.temperature.unique())\n",
    "print('Humidity:', weather_nominal.humidity.unique())\n",
    "print('Windy:', weather_nominal.windy.unique())\n",
    "print('Play:', weather_nominal.play.unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Mostrando dataset com `head` (5 primeiros valores) e `tail` (5 últimos valores), bom para datasets extensos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "      <th>play</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>sunny</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>sunny</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>True</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>overcast</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>rainy</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>rainy</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    outlook temperature humidity  windy play\n",
       "0     sunny         hot     high  False   no\n",
       "1     sunny         hot     high   True   no\n",
       "2  overcast         hot     high  False  yes\n",
       "3     rainy        mild     high  False  yes\n",
       "4     rainy        cool   normal  False  yes"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "      <th>play</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>rainy</td>\n",
       "      <td>75</td>\n",
       "      <td>80</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>sunny</td>\n",
       "      <td>75</td>\n",
       "      <td>70</td>\n",
       "      <td>True</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>overcast</td>\n",
       "      <td>72</td>\n",
       "      <td>90</td>\n",
       "      <td>True</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>overcast</td>\n",
       "      <td>81</td>\n",
       "      <td>75</td>\n",
       "      <td>False</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>rainy</td>\n",
       "      <td>71</td>\n",
       "      <td>91</td>\n",
       "      <td>True</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     outlook  temperature  humidity  windy play\n",
       "9      rainy           75        80  False  yes\n",
       "10     sunny           75        70   True  yes\n",
       "11  overcast           72        90   True  yes\n",
       "12  overcast           81        75  False  yes\n",
       "13     rainy           71        91   True   no"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "weather_nominal.head()\n",
    "weather_numeric.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Transformando os dados categóricos em numéricos\n",
    "\n",
    "Para usar os datasets, nós devemos transformar os dados categóricos em valores numéricos, já que tanto nosso método de classificação quanto o de regressão trabalham apenas com números. Iremos usar uma técnica de encoding básica, mapeando os valores categóricos para números.\n",
    "\n",
    "Uma forma mais \"profissional\" de fazer a transformação de dados é o `One-hot encoding`, mas não será tratado aqui."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "      <th>play</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>85</td>\n",
       "      <td>85</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>80</td>\n",
       "      <td>90</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>83</td>\n",
       "      <td>86</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>70</td>\n",
       "      <td>96</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>68</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   outlook  temperature  humidity  windy  play\n",
       "0        2           85        85      0     0\n",
       "1        2           80        90      1     0\n",
       "2        0           83        86      0     1\n",
       "3        1           70        96      0     1\n",
       "4        1           68        80      0     1"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Mudando o tipo das colunas categóricas\n",
    "weather_numeric['outlook'] = weather_numeric['outlook'].astype('category')\n",
    "weather_numeric['windy'] = weather_numeric['windy'].astype('category')\n",
    "weather_numeric['play'] = weather_numeric['play'].astype('category')\n",
    "\n",
    "# Aplicando enconding às colunas selecionadas\n",
    "cat_columns = weather_numeric.select_dtypes(['category']).columns\n",
    "weather_numeric[cat_columns] = weather_numeric[cat_columns].apply(lambda x: x.cat.codes)\n",
    "weather_numeric.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dividindo datasets em treino e teste"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>rainy</td>\n",
       "      <td>mild</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>rainy</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>sunny</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>sunny</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>overcast</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>rainy</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>sunny</td>\n",
       "      <td>hot</td>\n",
       "      <td>high</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>overcast</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>overcast</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     outlook temperature humidity  windy\n",
       "9      rainy        mild   normal  False\n",
       "3      rainy        mild     high  False\n",
       "0      sunny         hot     high  False\n",
       "7      sunny        mild     high  False\n",
       "2   overcast         hot     high  False\n",
       "4      rainy        cool   normal  False\n",
       "1      sunny         hot     high   True\n",
       "11  overcast        mild     high   True\n",
       "6   overcast        cool   normal   True"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>rainy</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>overcast</td>\n",
       "      <td>hot</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>sunny</td>\n",
       "      <td>cool</td>\n",
       "      <td>normal</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>rainy</td>\n",
       "      <td>mild</td>\n",
       "      <td>high</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>sunny</td>\n",
       "      <td>mild</td>\n",
       "      <td>normal</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     outlook temperature humidity  windy\n",
       "5      rainy        cool   normal   True\n",
       "12  overcast         hot   normal  False\n",
       "8      sunny        cool   normal  False\n",
       "13     rainy        mild     high   True\n",
       "10     sunny        mild   normal   True"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "9     yes\n",
       "3     yes\n",
       "0      no\n",
       "7      no\n",
       "2     yes\n",
       "4     yes\n",
       "1      no\n",
       "11    yes\n",
       "6     yes\n",
       "Name: play, dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "5      no\n",
       "12    yes\n",
       "8     yes\n",
       "13     no\n",
       "10    yes\n",
       "Name: play, dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_nominal_train, X_nominal_test, y_nominal_train, y_nominal_test = train_test_split(\n",
    "    weather_nominal.iloc[:, :-1], \n",
    "    weather_nominal.iloc[:, -1], \n",
    "    test_size=0.3\n",
    ")\n",
    "\n",
    "X_nominal_train\n",
    "X_nominal_test\n",
    "y_nominal_train\n",
    "y_nominal_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>65</td>\n",
       "      <td>70</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>64</td>\n",
       "      <td>65</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1</td>\n",
       "      <td>71</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>80</td>\n",
       "      <td>90</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1</td>\n",
       "      <td>75</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>70</td>\n",
       "      <td>96</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>83</td>\n",
       "      <td>86</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>85</td>\n",
       "      <td>85</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0</td>\n",
       "      <td>81</td>\n",
       "      <td>75</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    outlook  temperature  humidity  windy\n",
       "5         1           65        70      1\n",
       "6         0           64        65      1\n",
       "13        1           71        91      1\n",
       "1         2           80        90      1\n",
       "9         1           75        80      0\n",
       "3         1           70        96      0\n",
       "2         0           83        86      0\n",
       "0         2           85        85      0\n",
       "12        0           81        75      0"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>outlook</th>\n",
       "      <th>temperature</th>\n",
       "      <th>humidity</th>\n",
       "      <th>windy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>68</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2</td>\n",
       "      <td>75</td>\n",
       "      <td>70</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0</td>\n",
       "      <td>72</td>\n",
       "      <td>90</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2</td>\n",
       "      <td>69</td>\n",
       "      <td>70</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2</td>\n",
       "      <td>72</td>\n",
       "      <td>95</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    outlook  temperature  humidity  windy\n",
       "4         1           68        80      0\n",
       "10        2           75        70      1\n",
       "11        0           72        90      1\n",
       "8         2           69        70      0\n",
       "7         2           72        95      0"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "5     0\n",
       "6     1\n",
       "13    0\n",
       "1     0\n",
       "9     1\n",
       "3     1\n",
       "2     1\n",
       "0     0\n",
       "12    1\n",
       "Name: play, dtype: int8"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "4     1\n",
       "10    1\n",
       "11    1\n",
       "8     1\n",
       "7     0\n",
       "Name: play, dtype: int8"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_numeric_train, X_numeric_test, y_numeric_train, y_numeric_test = train_test_split(\n",
    "    weather_numeric.iloc[:, :-1], \n",
    "    weather_numeric.iloc[:, -1], \n",
    "    test_size=0.3\n",
    ")\n",
    "\n",
    "X_numeric_train\n",
    "X_numeric_test\n",
    "y_numeric_train\n",
    "y_numeric_test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modelos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Classificação\n",
    "\n",
    "Vamos usar o KNN para classificar as instâncias do `Weather`. Como funciona o KNN? Imagine que suas instâncias estão distribuídas pelo espaço (então você tem a ideia de pontos e o conceito de distâncias) e que instâncias próximas são semelhantes. Logo, você pode \"adivinhar\" a classe de uma instância olhando para seus vizinhos.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "           metric_params=None, n_jobs=None, n_neighbors=3, p=2,\n",
       "           weights='uniform')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classification = neighbors.KNeighborsClassifier(n_neighbors=3)\n",
    "classification.fit(X_numeric_train, y_numeric_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1, 0, 1, 0], dtype=int8)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "class_predict = classification.predict(X_numeric_test)\n",
    "class_predict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regressão\n",
    "\n",
    "Vamos agora usar a Regressão Linear para fazer a regressão das instâncias do `Weather`. A ideia da regressão linear é explicar seus dados como uma função `y = a * x + b + e`, onde `a` e `b` são coeficientes e `e` é um termo de erro, ou seja, uma medida que mostra que existem outros fatores não explicados pelo modelo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None,\n",
       "         normalize=False)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regression = linear_model.LinearRegression()\n",
    "regression.fit(X_numeric_train, y_numeric_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.91362021, -0.28710622,  0.65729027,  0.45110359,  0.46459723])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regr_predict = regression.predict(X_numeric_test)\n",
    "regr_predict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Avaliação de resultados\n",
    "\n",
    "Existem várias formas de medir quão \"bom\" é um modelo (apesar de isso ser subjetivo), as métricas são boas formas. Elas testam resultados gerados pelos modelos com valores reais, e dão uma ideia de quão preciso é o seu modelo na classificação (ou quão bem sua curva explica os dados na regressão)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.33      1.00      0.50         1\n",
      "           1       1.00      0.50      0.67         4\n",
      "\n",
      "   micro avg       0.60      0.60      0.60         5\n",
      "   macro avg       0.67      0.75      0.58         5\n",
      "weighted avg       0.87      0.60      0.63         5\n",
      "\n",
      "Regressão:\n",
      "0.4597383395421244\n"
     ]
    }
   ],
   "source": [
    "print('KNN:')\n",
    "print(metrics.classification_report(y_numeric_test, class_predict))\n",
    "\n",
    "print('Regressão:')\n",
    "print(metrics.mean_squared_error(y_numeric_test, regr_predict))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cross validation\n",
    "\n",
    "A ideia é evitar o overfitting fazendo passos intermediários de validação com uma parte do dataset de treino. Por exemplo, uma validação cruzada de 10 folds treina o modelo com 9 pedaços e valida os resultados obtidos com o pedaço restante, e repete esse processo mais 9 vezes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN:\n",
      "validação cruzada: {'fit_time': array([0.015661  , 0.00339317]), 'score_time': array([0.02187538, 0.0083282 ]), 'test_score': array([0.6 , 0.25]), 'train_score': array([0.75, 0.4 ])}\n",
      "Regressão:\n",
      "validação cruzada: {'fit_time': array([0.0084095 , 0.02097845]), 'score_time': array([0.00701356, 0.01187325]), 'test_score': array([-0.85396599, -8.81199539]), 'train_score': array([1., 1.])}\n"
     ]
    }
   ],
   "source": [
    "print('KNN:')\n",
    "print('validação cruzada:', model_selection.cross_validate(\n",
    "    classification, \n",
    "    X_numeric_train, \n",
    "    y_numeric_train,\n",
    "    cv=2\n",
    "))\n",
    "\n",
    "print('Regressão:')\n",
    "print('validação cruzada:', model_selection.cross_validate(\n",
    "    regression, \n",
    "    X_numeric_train, \n",
    "    y_numeric_train,\n",
    "    cv=2\n",
    "))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extras"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Baseline classifier\n",
    "\n",
    "É muito difícil avaliar o resultado de um modelo (descobrir se o valor resultante das métricas é bom ou não), mas existe uma forma de descobrir qual seria o pior valor possível. A ideia é usar um classificador \"dummy\", isto é, uma estratégia ingênua de classificação. Neste caso apresentado abaixo, o dummy checa qual a classe dominante e classifica as instâncias de teste com aquela mesma classe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DummyClassifier(constant=None, random_state=None, strategy='most_frequent')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classification = dummy.DummyClassifier(strategy='most_frequent')\n",
    "classification.fit(X_nominal_train, y_nominal_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = classification.predict(X_nominal_test)\n",
    "\n",
    "metrics.accuracy_score(y_nominal_test, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}