03. Indexing, Merge, Concatenate, and Combine.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f4e5e202",
   "metadata": {},
   "source": [
    "## <font color=green> Tratamento de dados: indexação, junção, concatenação e combinação"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "581ed7fe",
   "metadata": {},
   "source": [
    "- Em muitas aplicações, os dados podem estar espalhados em vários arquivos ou bancos de dados, ou podem estar organizados em um formato que não seja fácil de analisar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a289a907",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7713acc",
   "metadata": {},
   "source": [
    "### Indexação Hierárquica"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02fe83c4",
   "metadata": {},
   "source": [
    "- A indexação hierárquica é um recurso que permite ter vários níveis de índices (dois ou mais) em um eixo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ad25007a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a  1    0.005646\n",
       "   2    0.967006\n",
       "   3    0.260935\n",
       "b  1    0.492904\n",
       "   3    0.318336\n",
       "c  1    0.365956\n",
       "   2    0.111986\n",
       "d  2    0.592394\n",
       "   3    0.761145\n",
       "dtype: float64"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.Series(np.random.rand(9),\n",
    "                    index=[['a','a','a','b','b','c','c','d','d'],\n",
    "                           [1,2,3,1,3,1,2,2,3]])\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6af17a7a",
   "metadata": {},
   "source": [
    "- O que você está vendo é um visão elegante de um Series com um MultiIndex como índice. As “lacunas” na exibição do índice significam “utilize o rótulo imediatamente anterior”:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cc93dbc5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "MultiIndex([('a', 1),\n",
       "            ('a', 2),\n",
       "            ('a', 3),\n",
       "            ('b', 1),\n",
       "            ('b', 3),\n",
       "            ('c', 1),\n",
       "            ('c', 2),\n",
       "            ('d', 2),\n",
       "            ('d', 3)],\n",
       "           )"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.index"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c683994d",
   "metadata": {},
   "source": [
    "- Com um objeto hierarquicamente indexado, a chamada indexação parcial é possível, permitindo selecionar subconjuntos dos dados de forma concisa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "11d5ec25",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    0.492904\n",
       "3    0.318336\n",
       "dtype: float64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['b']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4d1919e3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "b  1    0.492904\n",
       "   3    0.318336\n",
       "d  2    0.592394\n",
       "   3    0.761145\n",
       "dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.loc[['b','d']]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "432a402a",
   "metadata": {},
   "source": [
    "- A seleção é até mesmo possível a partir de um nível “mais interno”"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "81ecd548",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a    0.967006\n",
       "c    0.111986\n",
       "d    0.592394\n",
       "dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.loc[:,2]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b285a1ca",
   "metadata": {},
   "source": [
    "- A indexação hierárquica desempenha um papel na reformatação dos dados e nas operações baseadas em grupos, como compor uma tabela pivô.\n",
    "- Podemos reorganizar, por exemplo, os dados em um DataFrame usando o seu método unstack."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "907d5d39",
   "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>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0.005646</td>\n",
       "      <td>0.967006</td>\n",
       "      <td>0.260935</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>0.492904</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.318336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c</th>\n",
       "      <td>0.365956</td>\n",
       "      <td>0.111986</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>d</th>\n",
       "      <td>NaN</td>\n",
       "      <td>0.592394</td>\n",
       "      <td>0.761145</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          1         2         3\n",
       "a  0.005646  0.967006  0.260935\n",
       "b  0.492904       NaN  0.318336\n",
       "c  0.365956  0.111986       NaN\n",
       "d       NaN  0.592394  0.761145"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.unstack()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "120ad3d7",
   "metadata": {},
   "source": [
    "- Em um DataFrame, qualquer eixo pode ter um índice hierárquico"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ca18b7fb",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">Ohio</th>\n",
       "      <th>Colorado</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "      <th>Green</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">a</th>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">b</th>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Ohio     Colorado\n",
       "    Green Red    Green\n",
       "a 1     0   1        2\n",
       "  2     3   4        5\n",
       "b 1     6   7        8\n",
       "  2     9  10       11"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame = pd.DataFrame(np.arange(12).reshape((4,3)), \n",
    "                     index=[['a','a','b','b'],[1,2,1,2]],\n",
    "                     columns=[['Ohio', 'Ohio', 'Colorado'], ['Green', 'Red', 'Green']])\n",
    "frame"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9da45cf",
   "metadata": {},
   "source": [
    "- Os níveis hierárquicos podem ter nomes (como strings ou quaisquer objetos Python)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ab842651",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th colspan=\"2\" halign=\"left\">Ohio</th>\n",
       "      <th>Colorado</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>color</th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "      <th>Green</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>key1</th>\n",
       "      <th>key2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">a</th>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">b</th>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "state      Ohio     Colorado\n",
       "color     Green Red    Green\n",
       "key1 key2                   \n",
       "a    1        0   1        2\n",
       "     2        3   4        5\n",
       "b    1        6   7        8\n",
       "     2        9  10       11"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame.index.names = ['key1', 'key2']\n",
    "frame.columns.names = ['state', 'color']\n",
    "frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "917d431a",
   "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>color</th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>key1</th>\n",
       "      <th>key2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">a</th>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">b</th>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "color      Green  Red\n",
       "key1 key2            \n",
       "a    1         0    1\n",
       "     2         3    4\n",
       "b    1         6    7\n",
       "     2         9   10"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame['Ohio']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0bdd0b5b",
   "metadata": {},
   "source": [
    "### Reorganizando e ordenando níveis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6af2916f",
   "metadata": {},
   "source": [
    "- Ocasionalmente precisaremos reorganizar a ordem dos níveis em um eixo ou ordenar os dados de acordo com os valores em um nível específico.\n",
    "- swaplevel aceita dois números ou nomes de níveis e devolve um novo objetos com os níveis trocados (os dados, porém, permanecem inalterados)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "00892fba",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th colspan=\"2\" halign=\"left\">Ohio</th>\n",
       "      <th>Colorado</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>color</th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "      <th>Green</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>key2</th>\n",
       "      <th>key1</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <th>a</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <th>a</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <th>b</th>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <th>b</th>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "state      Ohio     Colorado\n",
       "color     Green Red    Green\n",
       "key2 key1                   \n",
       "1    a        0   1        2\n",
       "2    a        3   4        5\n",
       "1    b        6   7        8\n",
       "2    b        9  10       11"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame.swaplevel('key1','key2')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ea1ebca",
   "metadata": {},
   "source": [
    "- sort_index, por outro lado, ordena os dados usando os valores de um só nível. Ao trocar níveis, não é incomum usar sort_index também para que o resultado seja ordenado em ordem lexicográfica de acordo com o nível indicado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a8a77bd8",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th colspan=\"2\" halign=\"left\">Ohio</th>\n",
       "      <th>Colorado</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>color</th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "      <th>Green</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>key1</th>\n",
       "      <th>key2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <th>2</th>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "state      Ohio     Colorado\n",
       "color     Green Red    Green\n",
       "key1 key2                   \n",
       "a    1        0   1        2\n",
       "b    1        6   7        8\n",
       "a    2        3   4        5\n",
       "b    2        9  10       11"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame.sort_index(level=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1fec168c",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th colspan=\"2\" halign=\"left\">Ohio</th>\n",
       "      <th>Colorado</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>color</th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "      <th>Green</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>key2</th>\n",
       "      <th>key1</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">1</th>\n",
       "      <th>a</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">2</th>\n",
       "      <th>a</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "state      Ohio     Colorado\n",
       "color     Green Red    Green\n",
       "key2 key1                   \n",
       "1    a        0   1        2\n",
       "     b        6   7        8\n",
       "2    a        3   4        5\n",
       "     b        9  10       11"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame.swaplevel(0,1).sort_index(level=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba6c1aa7",
   "metadata": {},
   "source": [
    "### Estatísticas de resumo por níveis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5baed9d2",
   "metadata": {},
   "source": [
    "- Muitas estatísticas descritivas ou de resumo em DataFrame e em Series têm uma opção level; com ela, podemos especificar o nível de acordo com o qual queremos fazer uma agregação, em um eixo em particular."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c03a5682",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th>state</th>\n",
       "      <th colspan=\"2\" halign=\"left\">Ohio</th>\n",
       "      <th>Colorado</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>color</th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "      <th>Green</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>key2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "state  Ohio     Colorado\n",
       "color Green Red    Green\n",
       "key2                    \n",
       "1         6   8       10\n",
       "2        12  14       16"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame.sum(level='key2')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "82a752a1",
   "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>color</th>\n",
       "      <th>Green</th>\n",
       "      <th>Red</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>key1</th>\n",
       "      <th>key2</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">a</th>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">b</th>\n",
       "      <th>1</th>\n",
       "      <td>14</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>20</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "color      Green  Red\n",
       "key1 key2            \n",
       "a    1         2    1\n",
       "     2         8    4\n",
       "b    1        14    7\n",
       "     2        20   10"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame.sum(level='color',axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d6317ad0",
   "metadata": {},
   "source": [
    "### Indexando com as colunas de um DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1b3a4364",
   "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>a</th>\n",
       "      <th>b</th>\n",
       "      <th>c</th>\n",
       "      <th>d</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>one</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>one</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>one</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>two</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>two</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>two</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>two</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   a  b    c  d\n",
       "0  0  7  one  0\n",
       "1  1  6  one  1\n",
       "2  2  5  one  2\n",
       "3  3  4  two  0\n",
       "4  4  3  two  1\n",
       "5  5  2  two  2\n",
       "6  6  1  two  3"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame = pd.DataFrame({'a': range(7), \n",
    "                      'b': range(7, 0, -1), \n",
    "                      'c': ['one', 'one', 'one', 'two', 'two', 'two', 'two'],\n",
    "                      'd': [0,1,2,0,1,2,3]})\n",
    "frame"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06d6fad9",
   "metadata": {},
   "source": [
    "- A função set_index de DataFrame cria um novo DataFrame usando uma ou mais colunas índice de linha"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "603f0371",
   "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></th>\n",
       "      <th>a</th>\n",
       "      <th>b</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c</th>\n",
       "      <th>d</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">one</th>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">two</th>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       a  b\n",
       "c   d      \n",
       "one 0  0  7\n",
       "    1  1  6\n",
       "    2  2  5\n",
       "two 0  3  4\n",
       "    1  4  3\n",
       "    2  5  2\n",
       "    3  6  1"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame2 = frame.set_index(['c', 'd'])\n",
    "frame2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0971470d",
   "metadata": {},
   "source": [
    "- Por padrão, as colunas são removidas do DataFrame, embora possam ser mantidas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "04545833",
   "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></th>\n",
       "      <th>a</th>\n",
       "      <th>b</th>\n",
       "      <th>c</th>\n",
       "      <th>d</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c</th>\n",
       "      <th>d</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">one</th>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>one</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>one</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>one</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">two</th>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>two</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>two</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>two</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>two</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       a  b    c  d\n",
       "c   d              \n",
       "one 0  0  7  one  0\n",
       "    1  1  6  one  1\n",
       "    2  2  5  one  2\n",
       "two 0  3  4  two  0\n",
       "    1  4  3  two  1\n",
       "    2  5  2  two  2\n",
       "    3  6  1  two  3"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame.set_index(['c','d'], drop=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34bee5c1",
   "metadata": {},
   "source": [
    "- reset_index, por outro lado, faz o inverso de set_index; os níveis dos índices hierárquicos são passados para as colunas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "aae849af",
   "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>c</th>\n",
       "      <th>d</th>\n",
       "      <th>a</th>\n",
       "      <th>b</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>one</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>one</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>one</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>two</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>two</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>two</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>two</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     c  d  a  b\n",
       "0  one  0  0  7\n",
       "1  one  1  1  6\n",
       "2  one  2  2  5\n",
       "3  two  0  3  4\n",
       "4  two  1  4  3\n",
       "5  two  2  5  2\n",
       "6  two  3  6  1"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "frame2.reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76f366c1",
   "metadata": {},
   "source": [
    "### Junções no DataFrame no estilo de bancos de dados"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e1392e9",
   "metadata": {},
   "source": [
    "- Operações de merge ou de join combinam conjuntos de dados associando linhas por meio de uma ou mais chaves em bancos de dados relacionais\n",
    "- A função merge do pandas é o ponto de entrada principal para usar essesw algoritmos em seus dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "dcbec1f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = pd.DataFrame({'key': ['b', 'b', 'a', 'c', 'a', 'a', 'b'],\n",
    "                    'data1': range(7)})\n",
    "\n",
    "df2 = pd.DataFrame({'key': ['a', 'b', 'd'], \n",
    "                    'data2': range(3)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ec63d830",
   "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>key</th>\n",
       "      <th>data1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>a</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>b</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  data1\n",
       "0   b      0\n",
       "1   b      1\n",
       "2   a      2\n",
       "3   c      3\n",
       "4   a      4\n",
       "5   a      5\n",
       "6   b      6"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "a3a905f2",
   "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>key</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>d</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  data2\n",
       "0   a      0\n",
       "1   b      1\n",
       "2   d      2"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2a7c0798",
   "metadata": {},
   "source": [
    "- Fazendo junção de muitos-para-um\n",
    "- Os dados em df1 têm várias linhas de rótulos a e b, enquanto df2 tem apenas uma linha para cada valor na coluna key"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "3d8e08fa",
   "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>key</th>\n",
       "      <th>data1</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>b</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>a</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  data1  data2\n",
       "0   b      0      1\n",
       "1   b      1      1\n",
       "2   b      6      1\n",
       "3   a      2      0\n",
       "4   a      4      0\n",
       "5   a      5      0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(df1, df2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3e96293",
   "metadata": {},
   "source": [
    "- Observe que a coluna não foi especificada para fazer a junção. A função merge utilizou como chaves os nomes das colunas que se sobrepõem"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "111f278d",
   "metadata": {},
   "source": [
    "- No entanto, é possível especificá-la explicitamente com o parâmetro on"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "b4c80392",
   "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>key</th>\n",
       "      <th>data1</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>b</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>a</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  data1  data2\n",
       "0   b      0      1\n",
       "1   b      1      1\n",
       "2   b      6      1\n",
       "3   a      2      0\n",
       "4   a      4      0\n",
       "5   a      5      0"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(df1, df2, on='key')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "8761e5cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "df3 = pd.DataFrame({'lkey': ['b', 'b', 'a', 'c', 'a', 'a', 'b'], \n",
    "                    'data1': range(7)})\n",
    "df4 = pd.DataFrame({'rkey': ['a', 'b', 'd'], \n",
    "                    'data2': range(3)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "c175cb3e",
   "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>lkey</th>\n",
       "      <th>data1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>a</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>b</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  lkey  data1\n",
       "0    b      0\n",
       "1    b      1\n",
       "2    a      2\n",
       "3    c      3\n",
       "4    a      4\n",
       "5    a      5\n",
       "6    b      6"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "f5a9f160",
   "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>rkey</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>d</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  rkey  data2\n",
       "0    a      0\n",
       "1    b      1\n",
       "2    d      2"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "b1ec3d8a",
   "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>lkey</th>\n",
       "      <th>data1</th>\n",
       "      <th>rkey</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>b</td>\n",
       "      <td>6</td>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>a</td>\n",
       "      <td>5</td>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  lkey  data1 rkey  data2\n",
       "0    b      0    b      1\n",
       "1    b      1    b      1\n",
       "2    b      6    b      1\n",
       "3    a      2    a      0\n",
       "4    a      4    a      0\n",
       "5    a      5    a      0"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(df3, df4, left_on='lkey', right_on='rkey')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "049876f4",
   "metadata": {},
   "source": [
    "- No nosso exemplo os valores 'c' e 'd' e os dados associados estão ausentes no resultado.\n",
    "- É importante destacar que por padrão, merge executa uma junção do tipo 'inner'."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "058dcb8c",
   "metadata": {},
   "source": [
    "- Outras opções possíveis são: 'left', 'right' e 'outer'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "224ad0b7",
   "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>key</th>\n",
       "      <th>data1</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>b</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>a</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>c</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>d</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  data1  data2\n",
       "0   b    0.0    1.0\n",
       "1   b    1.0    1.0\n",
       "2   b    6.0    1.0\n",
       "3   a    2.0    0.0\n",
       "4   a    4.0    0.0\n",
       "5   a    5.0    0.0\n",
       "6   c    3.0    NaN\n",
       "7   d    NaN    2.0"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(df1, df2, how='outer')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6b5a0ca",
   "metadata": {},
   "source": [
    "- Fazendo junção de muitos-para-muitos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "5637765b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = pd.DataFrame({'key': ['b', 'b', 'a', 'c', 'a', 'b'], \n",
    "                    'data1': range(6)})\n",
    "\n",
    "df2 = pd.DataFrame({'key': ['a', 'b', 'a', 'b', 'd'],\n",
    "                    'data2': range(5)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "797b04d3",
   "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>key</th>\n",
       "      <th>data1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>b</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  data1\n",
       "0   b      0\n",
       "1   b      1\n",
       "2   a      2\n",
       "3   c      3\n",
       "4   a      4\n",
       "5   b      5"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "7a947c84",
   "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>key</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>b</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>d</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  data2\n",
       "0   a      0\n",
       "1   b      1\n",
       "2   a      2\n",
       "3   b      3\n",
       "4   d      4"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "8f454a05",
   "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>key</th>\n",
       "      <th>data1</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>c</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>b</td>\n",
       "      <td>5</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>b</td>\n",
       "      <td>5</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   key  data1  data2\n",
       "0    b      0    1.0\n",
       "1    b      0    3.0\n",
       "2    b      1    1.0\n",
       "3    b      1    3.0\n",
       "4    a      2    0.0\n",
       "5    a      2    2.0\n",
       "6    c      3    NaN\n",
       "7    a      4    0.0\n",
       "8    a      4    2.0\n",
       "9    b      5    1.0\n",
       "10   b      5    3.0"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(df1, df2, on='key', how='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "801bcda1",
   "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>key</th>\n",
       "      <th>data1</th>\n",
       "      <th>data2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>b</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>b</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>a</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  data1  data2\n",
       "0   b      0      1\n",
       "1   b      0      3\n",
       "2   b      1      1\n",
       "3   b      1      3\n",
       "4   b      5      1\n",
       "5   b      5      3\n",
       "6   a      2      0\n",
       "7   a      2      2\n",
       "8   a      4      0\n",
       "9   a      4      2"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(df1, df2, on='key', how='inner')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53205578",
   "metadata": {},
   "source": [
    "- Para um merge com vários chaves, passe uma lista de nomes de coluna"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "8a7eafa3",
   "metadata": {},
   "outputs": [],
   "source": [
    "left = pd.DataFrame({'key1': ['foo', 'foo', 'bar'], \n",
    "                     'key2': ['one', 'two', 'one'],\n",
    "                     'lval': [1,2,3]})\n",
    "\n",
    "right = pd.DataFrame({'key1': ['foo', 'foo', 'bar', 'bar'], \n",
    "                      'key2': ['one', 'one', 'one', 'two'], \n",
    "                      'rval': [4, 5, 6, 7]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "aea9a82c",
   "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>key1</th>\n",
       "      <th>key2</th>\n",
       "      <th>lval</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>foo</td>\n",
       "      <td>two</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>bar</td>\n",
       "      <td>one</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key1 key2  lval\n",
       "0  foo  one     1\n",
       "1  foo  two     2\n",
       "2  bar  one     3"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "left"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "ed72b64f",
   "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>key1</th>\n",
       "      <th>key2</th>\n",
       "      <th>rval</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>bar</td>\n",
       "      <td>one</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>bar</td>\n",
       "      <td>two</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key1 key2  rval\n",
       "0  foo  one     4\n",
       "1  foo  one     5\n",
       "2  bar  one     6\n",
       "3  bar  two     7"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "right"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "2017a075",
   "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>key1</th>\n",
       "      <th>key2</th>\n",
       "      <th>lval</th>\n",
       "      <th>rval</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>1.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>foo</td>\n",
       "      <td>two</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>bar</td>\n",
       "      <td>one</td>\n",
       "      <td>3.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>bar</td>\n",
       "      <td>two</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key1 key2  lval  rval\n",
       "0  foo  one   1.0   4.0\n",
       "1  foo  one   1.0   5.0\n",
       "2  foo  two   2.0   NaN\n",
       "3  bar  one   3.0   6.0\n",
       "4  bar  two   NaN   7.0"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(left, right, on=['key1', 'key2'], how='outer')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11c21391",
   "metadata": {},
   "source": [
    "- merge tem uma opção suffixes para especificar strings a serem concatenadas nos nomes que se sobrepõem, nos objetos DataFrame à esquerda e à direita"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "4c4ae623",
   "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>key1</th>\n",
       "      <th>key2_x</th>\n",
       "      <th>lval</th>\n",
       "      <th>key2_y</th>\n",
       "      <th>rval</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>1</td>\n",
       "      <td>one</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>1</td>\n",
       "      <td>one</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>foo</td>\n",
       "      <td>two</td>\n",
       "      <td>2</td>\n",
       "      <td>one</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>foo</td>\n",
       "      <td>two</td>\n",
       "      <td>2</td>\n",
       "      <td>one</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>bar</td>\n",
       "      <td>one</td>\n",
       "      <td>3</td>\n",
       "      <td>one</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>bar</td>\n",
       "      <td>one</td>\n",
       "      <td>3</td>\n",
       "      <td>two</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key1 key2_x  lval key2_y  rval\n",
       "0  foo    one     1    one     4\n",
       "1  foo    one     1    one     5\n",
       "2  foo    two     2    one     4\n",
       "3  foo    two     2    one     5\n",
       "4  bar    one     3    one     6\n",
       "5  bar    one     3    two     7"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(left, right, on='key1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "fa086662",
   "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>key1</th>\n",
       "      <th>key2_left</th>\n",
       "      <th>lval</th>\n",
       "      <th>key2_right</th>\n",
       "      <th>rval</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>1</td>\n",
       "      <td>one</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>foo</td>\n",
       "      <td>one</td>\n",
       "      <td>1</td>\n",
       "      <td>one</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>foo</td>\n",
       "      <td>two</td>\n",
       "      <td>2</td>\n",
       "      <td>one</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>foo</td>\n",
       "      <td>two</td>\n",
       "      <td>2</td>\n",
       "      <td>one</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>bar</td>\n",
       "      <td>one</td>\n",
       "      <td>3</td>\n",
       "      <td>one</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>bar</td>\n",
       "      <td>one</td>\n",
       "      <td>3</td>\n",
       "      <td>two</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key1 key2_left  lval key2_right  rval\n",
       "0  foo       one     1        one     4\n",
       "1  foo       one     1        one     5\n",
       "2  foo       two     2        one     4\n",
       "3  foo       two     2        one     5\n",
       "4  bar       one     3        one     6\n",
       "5  bar       one     3        two     7"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(left, right, on='key1', suffixes=('_left', '_right'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9196c9c5",
   "metadata": {},
   "source": [
    "### Fazendo merge com base no índice"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "faba3bd9",
   "metadata": {},
   "source": [
    "- Você pode passar left_index=True ou right_index=True (ou ambos) para informar que o índice deverá ser usado como a chave do merge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "276b1b70",
   "metadata": {},
   "outputs": [],
   "source": [
    "left1 = pd.DataFrame({'key': ['a', 'b', 'a', 'a', 'b', 'c'], \n",
    "                      'value': range(6)})\n",
    "\n",
    "right1 = pd.DataFrame({'group_val': [3.5, 7]}, index=['a', 'b'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "b29fcc24",
   "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>key</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>b</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>c</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  value\n",
       "0   a      0\n",
       "1   b      1\n",
       "2   a      2\n",
       "3   a      3\n",
       "4   b      4\n",
       "5   c      5"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "left1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "153a91e8",
   "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>group_val</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   group_val\n",
       "a        3.5\n",
       "b        7.0"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "right1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "429191a7",
   "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>key</th>\n",
       "      <th>value</th>\n",
       "      <th>group_val</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>3</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>b</td>\n",
       "      <td>4</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  value  group_val\n",
       "0   a      0        3.5\n",
       "2   a      2        3.5\n",
       "3   a      3        3.5\n",
       "1   b      1        7.0\n",
       "4   b      4        7.0"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(left1, right1, left_on='key', right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "77ea2ee5",
   "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>key</th>\n",
       "      <th>value</th>\n",
       "      <th>group_val</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>3</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>b</td>\n",
       "      <td>4</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>c</td>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  value  group_val\n",
       "0   a      0        3.5\n",
       "2   a      2        3.5\n",
       "3   a      3        3.5\n",
       "1   b      1        7.0\n",
       "4   b      4        7.0\n",
       "5   c      5        NaN"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(left1, right1, left_on='key', right_index=True, how='outer')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2d0455a",
   "metadata": {},
   "source": [
    "- O DataFrane tem uma instância conveniente de join para o merge pelo índice, que pode ser usada para combinar vários objetos DataFrame com índices iguais ou semelhantes, porém com colunas que não se sobrepõem"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "71cc5ff4",
   "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>key</th>\n",
       "      <th>value</th>\n",
       "      <th>group_val</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>0</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>a</td>\n",
       "      <td>2</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>a</td>\n",
       "      <td>3</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>1</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>b</td>\n",
       "      <td>4</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>c</td>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  key  value  group_val\n",
       "0   a      0        3.5\n",
       "2   a      2        3.5\n",
       "3   a      3        3.5\n",
       "1   b      1        7.0\n",
       "4   b      4        7.0\n",
       "5   c      5        NaN"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "left1.join(right1, on='key', how='outer')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03333cca",
   "metadata": {},
   "source": [
    "### Concatenando ao longo de um eixo"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2c5d3817",
   "metadata": {},
   "source": [
    "- A função concat do pandas oferece uma forma consistente de tratar a concatenação de dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "81926ea5",
   "metadata": {},
   "outputs": [],
   "source": [
    "s1 = pd.Series([0, 1], index=['a', 'b'])\n",
    "s2 = pd.Series([2, 3, 4], index=['c', 'd', 'e'])\n",
    "s3 = pd.Series([5, 6], index=['f', 'g'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "a9f2fab0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a    0\n",
       "b    1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "17210811",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "c    2\n",
       "d    3\n",
       "e    4\n",
       "dtype: int64"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "30d387b8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "f    5\n",
       "g    6\n",
       "dtype: int64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "bd75c2cc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a    0\n",
       "b    1\n",
       "c    2\n",
       "d    3\n",
       "e    4\n",
       "f    5\n",
       "g    6\n",
       "dtype: int64"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([s1,s2,s3])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a20be85",
   "metadata": {},
   "source": [
    "- Por padrão, concat atua em axis=0, gerando outra Series.\n",
    "- Se você passar axis=1, o resultado será um DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "5114f745",
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c</th>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>d</th>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>e</th>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>f</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>g</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     0    1    2\n",
       "a  0.0  NaN  NaN\n",
       "b  1.0  NaN  NaN\n",
       "c  NaN  2.0  NaN\n",
       "d  NaN  3.0  NaN\n",
       "e  NaN  4.0  NaN\n",
       "f  NaN  NaN  5.0\n",
       "g  NaN  NaN  6.0"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([s1, s2, s3], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fec4b902",
   "metadata": {},
   "source": [
    "- Não há sobreposição no outro eixo, o que gera a união ordenada (a junção 'outer') dos índices"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7efc1bd4",
   "metadata": {},
   "source": [
    "- Podemos fazer a sua intersecção passando join='inner'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "1a781670",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a    0\n",
       "b    1\n",
       "f    5\n",
       "g    6\n",
       "dtype: int64"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s4 = pd.concat([s1, s3])\n",
    "s4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "d0d66308",
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>f</th>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>g</th>\n",
       "      <td>NaN</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     0  1\n",
       "a  0.0  0\n",
       "b  1.0  1\n",
       "f  NaN  5\n",
       "g  NaN  6"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([s1, s4], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "a08386b0",
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   0  1\n",
       "a  0  0\n",
       "b  1  1"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([s1, s4], axis=1, join='inner')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "432e8344",
   "metadata": {},
   "source": [
    "- Os rótulos 'f' e 'g' desapareceram por causa da opção join='inner'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9457175",
   "metadata": {},
   "source": [
    "- Suponha que quiséssemos criar um índice hieráquico no eixo da concatenação. Para isso, utilize o argumeto keys"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "867d0f49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a    0\n",
       "b    1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "1ca88ca2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "f    5\n",
       "g    6\n",
       "dtype: int64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "9d24ce7b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "one    a    0\n",
       "       b    1\n",
       "two    a    0\n",
       "       b    1\n",
       "three  f    5\n",
       "       g    6\n",
       "dtype: int64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = pd.concat([s1, s1, s3], keys=['one', 'two', 'three'])\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "5f4f0863",
   "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>a</th>\n",
       "      <th>b</th>\n",
       "      <th>f</th>\n",
       "      <th>g</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>one</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>two</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>three</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         a    b    f    g\n",
       "one    0.0  1.0  NaN  NaN\n",
       "two    0.0  1.0  NaN  NaN\n",
       "three  NaN  NaN  5.0  6.0"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.unstack()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a766186a",
   "metadata": {},
   "source": [
    "- No caso da combinação de Series ao longo de axis=1, as keys passam a ser os cabeçalhos das colunas do DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "e4eebb3a",
   "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>one</th>\n",
       "      <th>two</th>\n",
       "      <th>three</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c</th>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>d</th>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>e</th>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>f</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>g</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   one  two  three\n",
       "a  0.0  NaN    NaN\n",
       "b  1.0  NaN    NaN\n",
       "c  NaN  2.0    NaN\n",
       "d  NaN  3.0    NaN\n",
       "e  NaN  4.0    NaN\n",
       "f  NaN  NaN    5.0\n",
       "g  NaN  NaN    6.0"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([s1, s2, s3], axis=1, keys=['one', 'two', 'three'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59b6fddc",
   "metadata": {},
   "source": [
    "- A mesma lógica se estende aos objetos DataFrame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "f56b11a4",
   "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>one</th>\n",
       "      <th>two</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c</th>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   one  two\n",
       "a    0    1\n",
       "b    2    3\n",
       "c    4    5"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = pd.DataFrame(np.arange(6).reshape(3, 2), index=['a', 'b', 'c'], columns=['one', 'two'])\n",
    "df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "278dd262",
   "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>three</th>\n",
       "      <th>four</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c</th>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   three  four\n",
       "a      5     6\n",
       "c      7     8"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = pd.DataFrame(5 + np.arange(4).reshape(2,2), index=['a', 'c'], columns=['three', 'four'])\n",
    "df2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "0c66edc0",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">level1</th>\n",
       "      <th colspan=\"2\" halign=\"left\">level2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>one</th>\n",
       "      <th>two</th>\n",
       "      <th>three</th>\n",
       "      <th>four</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>a</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>b</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>c</th>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>7.0</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  level1     level2     \n",
       "     one two  three four\n",
       "a      0   1    5.0  6.0\n",
       "b      2   3    NaN  NaN\n",
       "c      4   5    7.0  8.0"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([df1, df2], axis=1, keys=['level1', 'level2'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d70ecf9",
   "metadata": {},
   "source": [
    "- Quando o índice das linhas dos DataFrames não contém nenhum dado relevante, podemos passar ignore_index=True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "d4823bcc",
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = pd.DataFrame(np.random.randn(3,4), columns=['a', 'b', 'c', 'd'])\n",
    "df2 = pd.DataFrame(np.random.randn(2,3), columns=['b', 'd', 'a'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "9aeb9bc7",
   "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>a</th>\n",
       "      <th>b</th>\n",
       "      <th>c</th>\n",
       "      <th>d</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.102494</td>\n",
       "      <td>-1.381333</td>\n",
       "      <td>0.283748</td>\n",
       "      <td>-0.250660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-2.248611</td>\n",
       "      <td>-0.992999</td>\n",
       "      <td>-0.188599</td>\n",
       "      <td>0.102097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.523714</td>\n",
       "      <td>-1.038862</td>\n",
       "      <td>0.722294</td>\n",
       "      <td>-1.517398</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          a         b         c         d\n",
       "0 -0.102494 -1.381333  0.283748 -0.250660\n",
       "1 -2.248611 -0.992999 -0.188599  0.102097\n",
       "2  0.523714 -1.038862  0.722294 -1.517398"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "f5de3b1b",
   "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>b</th>\n",
       "      <th>d</th>\n",
       "      <th>a</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.160190</td>\n",
       "      <td>-1.452414</td>\n",
       "      <td>-2.336241</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.353083</td>\n",
       "      <td>-0.136983</td>\n",
       "      <td>-1.020427</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          b         d         a\n",
       "0 -0.160190 -1.452414 -2.336241\n",
       "1  0.353083 -0.136983 -1.020427"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "c8d9a9b4",
   "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>a</th>\n",
       "      <th>b</th>\n",
       "      <th>c</th>\n",
       "      <th>d</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.102494</td>\n",
       "      <td>-1.381333</td>\n",
       "      <td>0.283748</td>\n",
       "      <td>-0.250660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-2.248611</td>\n",
       "      <td>-0.992999</td>\n",
       "      <td>-0.188599</td>\n",
       "      <td>0.102097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.523714</td>\n",
       "      <td>-1.038862</td>\n",
       "      <td>0.722294</td>\n",
       "      <td>-1.517398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-2.336241</td>\n",
       "      <td>-0.160190</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-1.452414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.020427</td>\n",
       "      <td>0.353083</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.136983</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          a         b         c         d\n",
       "0 -0.102494 -1.381333  0.283748 -0.250660\n",
       "1 -2.248611 -0.992999 -0.188599  0.102097\n",
       "2  0.523714 -1.038862  0.722294 -1.517398\n",
       "3 -2.336241 -0.160190       NaN -1.452414\n",
       "4 -1.020427  0.353083       NaN -0.136983"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.concat([df1, df2], ignore_index=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10cd9e1f",
   "metadata": {},
   "source": [
    "### Combinando dados com sobreposição"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1e9e3b9",
   "metadata": {},
   "source": [
    "- Podemos ter dois conjuntos de dados cujos índices se sobreponham de forma total ou parcial. Nesse caso, considere a função where do Numpy, que executa o equivalente a uma expressão if-else, porém orientado a arrays."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "42855322",
   "metadata": {},
   "outputs": [],
   "source": [
    "a = pd.Series([np.nan, 2.5, 0, 3.5, 4.5, np.nan], index=['f', 'e', 'd', 'c', 'b', 'a'])\n",
    "b = pd.Series([0, np.nan, 2, np.nan, np.nan, 5], index=['a', 'b', 'c', 'd', 'e', 'f'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "2da65033",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "f    NaN\n",
       "e    2.5\n",
       "d    0.0\n",
       "c    3.5\n",
       "b    4.5\n",
       "a    NaN\n",
       "dtype: float64"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "179a4468",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a    0.0\n",
       "b    NaN\n",
       "c    2.0\n",
       "d    NaN\n",
       "e    NaN\n",
       "f    5.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "44cb51a6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "a    0.0\n",
       "b    4.5\n",
       "c    2.0\n",
       "d    0.0\n",
       "e    2.5\n",
       "f    5.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b.combine_first(a)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2270ed2e",
   "metadata": {},
   "source": [
    "- Com  DataFrames, combine_first faz o mesmo coluna a coluna"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "c0715cc1",
   "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>a</th>\n",
       "      <th>b</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NaN</td>\n",
       "      <td>6.0</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     a    b   c\n",
       "0  1.0  NaN   2\n",
       "1  NaN  2.0   6\n",
       "2  5.0  NaN  10\n",
       "3  NaN  6.0  14"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = pd.DataFrame({'a': [1, np.nan, 5, np.nan], \n",
    "                    'b': [np.nan, 2, np.nan, 6], \n",
    "                    'c': range(2,18,4)})\n",
    "df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "68ea75bf",
   "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>a</th>\n",
       "      <th>b</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7.0</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     a    b\n",
       "0  5.0  NaN\n",
       "1  4.0  3.0\n",
       "2  NaN  4.0\n",
       "3  3.0  6.0\n",
       "4  7.0  8.0"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = pd.DataFrame({'a': [5, 4, np.nan, 3, 7], \n",
    "                    'b': [np.nan, 3, 4, 6, 8]})\n",
    "df2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "4d5021aa",
   "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>a</th>\n",
       "      <th>b</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     a    b     c\n",
       "0  1.0  NaN   2.0\n",
       "1  4.0  2.0   6.0\n",
       "2  5.0  4.0  10.0\n",
       "3  3.0  6.0  14.0\n",
       "4  7.0  8.0   NaN"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.combine_first(df2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37d16a99",
   "metadata": {},
   "source": [
    "### Reformatando um DataFrame"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56cf0d9e",
   "metadata": {},
   "source": [
    "- Podemos formatar um DataFrame fazendo com que valores se tornem coluna e vice-versa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "02165c03",
   "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>estudante</th>\n",
       "      <th>escola</th>\n",
       "      <th>matéria</th>\n",
       "      <th>nota</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>André</td>\n",
       "      <td>Z</td>\n",
       "      <td>inglês</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Bernardo</td>\n",
       "      <td>Y</td>\n",
       "      <td>inglês</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Cristina</td>\n",
       "      <td>Z</td>\n",
       "      <td>inglês</td>\n",
       "      <td>1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Debóra</td>\n",
       "      <td>Y</td>\n",
       "      <td>inglês</td>\n",
       "      <td>10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>André</td>\n",
       "      <td>Z</td>\n",
       "      <td>matemática</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Bernardo</td>\n",
       "      <td>Y</td>\n",
       "      <td>matemática</td>\n",
       "      <td>200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Cristina</td>\n",
       "      <td>Z</td>\n",
       "      <td>matemática</td>\n",
       "      <td>2000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Debóra</td>\n",
       "      <td>Y</td>\n",
       "      <td>matemática</td>\n",
       "      <td>20000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>André</td>\n",
       "      <td>Z</td>\n",
       "      <td>física</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Bernardo</td>\n",
       "      <td>Y</td>\n",
       "      <td>física</td>\n",
       "      <td>500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Cristina</td>\n",
       "      <td>Z</td>\n",
       "      <td>física</td>\n",
       "      <td>5000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Debóra</td>\n",
       "      <td>Y</td>\n",
       "      <td>física</td>\n",
       "      <td>50000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   estudante escola     matéria   nota\n",
       "0      André      Z      inglês     10\n",
       "1   Bernardo      Y      inglês    100\n",
       "2   Cristina      Z      inglês   1000\n",
       "3     Debóra      Y      inglês  10000\n",
       "4      André      Z  matemática     20\n",
       "5   Bernardo      Y  matemática    200\n",
       "6   Cristina      Z  matemática   2000\n",
       "7     Debóra      Y  matemática  20000\n",
       "8      André      Z      física     50\n",
       "9   Bernardo      Y      física    500\n",
       "10  Cristina      Z      física   5000\n",
       "11    Debóra      Y      física  50000"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_longo = pd.DataFrame({\n",
    "        \"estudante\":\n",
    "            [\"André\", \"Bernardo\", \"Cristina\", \"Debóra\",\n",
    "             \"André\", \"Bernardo\", \"Cristina\", \"Debóra\",\n",
    "             \"André\", \"Bernardo\", \"Cristina\", \"Debóra\"],\n",
    "        \"escola\":\n",
    "            [\"Z\", \"Y\", \"Z\", \"Y\",\n",
    "             \"Z\", \"Y\", \"Z\", \"Y\",\n",
    "             \"Z\", \"Y\", \"Z\", \"Y\"],\n",
    "        \"matéria\":\n",
    "            [\"inglês\", \"inglês\", \"inglês\", \"inglês\",\n",
    "             \"matemática\", \"matemática\", \"matemática\", \"matemática\",\n",
    "             \"física\", \"física\", \"física\", \"física\"],\n",
    "        \"nota\":\n",
    "            [10, 100, 1000, 10000,\n",
    "             20, 200, 2000, 20000,\n",
    "             50, 500, 5000, 50000]\n",
    "})\n",
    "\n",
    "df_longo"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "10dfe71b",
   "metadata": {},
   "source": [
    "- Tornando o DataFrame menos longo e mais amplo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "8371e6da",
   "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>matéria</th>\n",
       "      <th>física</th>\n",
       "      <th>inglês</th>\n",
       "      <th>matemática</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>estudante</th>\n",
       "      <th>escola</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>André</th>\n",
       "      <th>Z</th>\n",
       "      <td>50</td>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bernardo</th>\n",
       "      <th>Y</th>\n",
       "      <td>500</td>\n",
       "      <td>100</td>\n",
       "      <td>200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cristina</th>\n",
       "      <th>Z</th>\n",
       "      <td>5000</td>\n",
       "      <td>1000</td>\n",
       "      <td>2000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Debóra</th>\n",
       "      <th>Y</th>\n",
       "      <td>50000</td>\n",
       "      <td>10000</td>\n",
       "      <td>20000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "matéria           física  inglês  matemática\n",
       "estudante escola                            \n",
       "André     Z           50      10          20\n",
       "Bernardo  Y          500     100         200\n",
       "Cristina  Z         5000    1000        2000\n",
       "Debóra    Y        50000   10000       20000"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_longo.pivot_table(index=[\"estudante\", \"escola\"], \n",
    "                    columns='matéria', \n",
    "                    values='nota')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62090f4a",
   "metadata": {},
   "source": [
    "- Adicionando margens para com a soma dos valores por linhas e por colunas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "6a789c25",
   "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>matéria</th>\n",
       "      <th>física</th>\n",
       "      <th>inglês</th>\n",
       "      <th>matemática</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>estudante</th>\n",
       "      <th>escola</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>André</th>\n",
       "      <th>Z</th>\n",
       "      <td>50</td>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bernardo</th>\n",
       "      <th>Y</th>\n",
       "      <td>500</td>\n",
       "      <td>100</td>\n",
       "      <td>200</td>\n",
       "      <td>800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cristina</th>\n",
       "      <th>Z</th>\n",
       "      <td>5000</td>\n",
       "      <td>1000</td>\n",
       "      <td>2000</td>\n",
       "      <td>8000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Debóra</th>\n",
       "      <th>Y</th>\n",
       "      <td>50000</td>\n",
       "      <td>10000</td>\n",
       "      <td>20000</td>\n",
       "      <td>80000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <th></th>\n",
       "      <td>55550</td>\n",
       "      <td>11110</td>\n",
       "      <td>22220</td>\n",
       "      <td>88880</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "matéria           física  inglês  matemática    All\n",
       "estudante escola                                   \n",
       "André     Z           50      10          20     80\n",
       "Bernardo  Y          500     100         200    800\n",
       "Cristina  Z         5000    1000        2000   8000\n",
       "Debóra    Y        50000   10000       20000  80000\n",
       "All                55550   11110       22220  88880"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_longo.pivot_table(index=[\"estudante\", \"escola\"], \n",
    "                    columns='matéria', \n",
    "                    values='nota',\n",
    "                    margins=True,  # add margins\n",
    "                    aggfunc='sum')  # sum margins (rows/columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2bdff7dd",
   "metadata": {},
   "source": [
    "- Se o parâmetro columns não for especificado, serão apresentadas as médias de todas as colunas numéricas "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "3ff9888a",
   "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></th>\n",
       "      <th>nota</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>estudante</th>\n",
       "      <th>escola</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>André</th>\n",
       "      <th>Z</th>\n",
       "      <td>26.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bernardo</th>\n",
       "      <th>Y</th>\n",
       "      <td>266.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cristina</th>\n",
       "      <th>Z</th>\n",
       "      <td>2666.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Debóra</th>\n",
       "      <th>Y</th>\n",
       "      <td>26666.666667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          nota\n",
       "estudante escola              \n",
       "André     Z          26.666667\n",
       "Bernardo  Y         266.666667\n",
       "Cristina  Z        2666.666667\n",
       "Debóra    Y       26666.666667"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_longo.pivot_table(index=[\"estudante\", \"escola\"])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "180834ed",
   "metadata": {},
   "source": [
    "- É possível também especificar o preenchemento dos valores faltantes "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "a21b6fcf",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th>escola</th>\n",
       "      <th colspan=\"3\" halign=\"left\">Y</th>\n",
       "      <th colspan=\"3\" halign=\"left\">Z</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>matéria</th>\n",
       "      <th>física</th>\n",
       "      <th>inglês</th>\n",
       "      <th>matemática</th>\n",
       "      <th>física</th>\n",
       "      <th>inglês</th>\n",
       "      <th>matemática</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>estudante</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>André</th>\n",
       "      <td>-5</td>\n",
       "      <td>-5</td>\n",
       "      <td>-5</td>\n",
       "      <td>50</td>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bernardo</th>\n",
       "      <td>500</td>\n",
       "      <td>100</td>\n",
       "      <td>200</td>\n",
       "      <td>-5</td>\n",
       "      <td>-5</td>\n",
       "      <td>-5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cristina</th>\n",
       "      <td>-5</td>\n",
       "      <td>-5</td>\n",
       "      <td>-5</td>\n",
       "      <td>5000</td>\n",
       "      <td>1000</td>\n",
       "      <td>2000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Debóra</th>\n",
       "      <td>50000</td>\n",
       "      <td>10000</td>\n",
       "      <td>20000</td>\n",
       "      <td>-5</td>\n",
       "      <td>-5</td>\n",
       "      <td>-5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "escola         Y                        Z                  \n",
       "matéria   física inglês matemática física inglês matemática\n",
       "estudante                                                  \n",
       "André         -5     -5         -5     50     10         20\n",
       "Bernardo     500    100        200     -5     -5         -5\n",
       "Cristina      -5     -5         -5   5000   1000       2000\n",
       "Debóra     50000  10000      20000     -5     -5         -5"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_longo.pivot_table(index=\"estudante\", \n",
    "                    columns=['escola', 'matéria'], \n",
    "                    values='nota', \n",
    "                    fill_value=-5)  # replace NaN with -5"
   ]
  }
 ],
 "metadata": {
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  },
  "vscode": {
   "interpreter": {
    "hash": "4ef1bb06cfdb40804b9995cc70859f43ff54c1a5223e7f7279b6762f9929c1e2"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}