11 Lavorare su un DataFrame.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Statistiche descrittive\n",
    "\n",
    "Consideriamo nuovamente il DataFrame degli [incidenti](http://dati.comune.milano.it/dataset/9f7bcc9c-20a4-4e48-a7cd-99b15ed11102/resource/38d2171d-1067-4252-9f96-02867a2cc617/download/ds177_trafficotrasporti_incidenti_stradali_persone_infortunate_mese_zona_2001-2016.json)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "incidenti = pd.read_json(\"data/incidenti.json\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La funzione `describe` permette di iniziare l'esplorazione del dataset, calcolando alcune statistiche essenziali."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1920.000000</td>\n",
       "      <td>1920.000000</td>\n",
       "      <td>1920.000000</td>\n",
       "      <td>1920.000000</td>\n",
       "      <td>1920.000000</td>\n",
       "      <td>1728.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2008.500000</td>\n",
       "      <td>145.287500</td>\n",
       "      <td>108.352083</td>\n",
       "      <td>6.500000</td>\n",
       "      <td>0.515104</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.610973</td>\n",
       "      <td>65.982854</td>\n",
       "      <td>48.726844</td>\n",
       "      <td>3.452952</td>\n",
       "      <td>0.752125</td>\n",
       "      <td>2.582736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2001.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2004.750000</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>73.000000</td>\n",
       "      <td>3.750000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2008.500000</td>\n",
       "      <td>137.000000</td>\n",
       "      <td>103.000000</td>\n",
       "      <td>6.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2012.250000</td>\n",
       "      <td>188.000000</td>\n",
       "      <td>140.000000</td>\n",
       "      <td>9.250000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2016.000000</td>\n",
       "      <td>557.000000</td>\n",
       "      <td>439.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Anno       Feriti    Incidenti         Mese        Morti  \\\n",
       "count  1920.000000  1920.000000  1920.000000  1920.000000  1920.000000   \n",
       "mean   2008.500000   145.287500   108.352083     6.500000     0.515104   \n",
       "std       4.610973    65.982854    48.726844     3.452952     0.752125   \n",
       "min    2001.000000    22.000000    17.000000     1.000000     0.000000   \n",
       "25%    2004.750000    96.000000    73.000000     3.750000     0.000000   \n",
       "50%    2008.500000   137.000000   103.000000     6.500000     0.000000   \n",
       "75%    2012.250000   188.000000   140.000000     9.250000     1.000000   \n",
       "max    2016.000000   557.000000   439.000000    12.000000     4.000000   \n",
       "\n",
       "              Zona  \n",
       "count  1728.000000  \n",
       "mean      5.000000  \n",
       "std       2.582736  \n",
       "min       1.000000  \n",
       "25%       3.000000  \n",
       "50%       5.000000  \n",
       "75%       7.000000  \n",
       "max       9.000000  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notare che `count` conta il numero di righe per cui la variabile non è nulla."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Rimuovere valori mancanti\n",
    "\n",
    "La presenza di valori mancanti può creare problemi. Per questo motivo possiamo usare `dropna` per rimuovere tutte le righe che contengono valori mancanti."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001</td>\n",
       "      <td>187</td>\n",
       "      <td>141</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona\n",
       "1  2001     209        171     1      0   1.0\n",
       "2  2001     115         91     1      1   2.0\n",
       "3  2001     187        141     1      1   3.0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_puliti = incidenti.dropna()\n",
    "incidenti_puliti.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Stiamo però attenti che il nuovo DataFrame potrebbe avere caratteristiche diverse dal DataFrame originale. Andiamo a calcolare il numero medio di `Feriti` e di `Incidenti` per i DataFrame `incidenti` e `incidenti_puliti`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Feriti       145.287500\n",
       "Incidenti    108.352083\n",
       "Name: mean, dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.describe()[['Feriti', 'Incidenti']].loc['mean']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Feriti       152.524306\n",
       "Incidenti    113.753472\n",
       "Name: mean, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_puliti.describe()[['Feriti', 'Incidenti']].loc['mean']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creare una nuova colonna\n",
    "\n",
    "Per creare una nuova colonna a partire da altre già esistenti, bisogna scrivere l'espressione corrispondente. Ad esempio, per creare una colonna con il numero di feriti per incidente, bisogna scrivere"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "incidenti['rapporto'] = incidenti['Feriti'] / incidenti['Incidenti']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Come usuale, è opportuno verificare il risultato dell'operazione, guardando le prime righe del dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>69</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.380000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.222222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.263736</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto\n",
       "0  2001      69         50     1      1   NaN  1.380000\n",
       "1  2001     209        171     1      0   1.0  1.222222\n",
       "2  2001     115         91     1      1   2.0  1.263736"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Assegnare valori ad una colonna\n",
    "\n",
    "Un caso particolare si ha quando si vuole assegnare lo stesso valore a tutte le righe di un DataFrame. In questo caso assegnamo il valore `0` ad una nuova colonna che chiamiamo `prova`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "incidenti['prova'] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "      <th>prova</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>69</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.380000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.222222</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.263736</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001</td>\n",
       "      <td>187</td>\n",
       "      <td>141</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.326241</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2001</td>\n",
       "      <td>154</td>\n",
       "      <td>105</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.466667</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto  prova\n",
       "0  2001      69         50     1      1   NaN  1.380000      0\n",
       "1  2001     209        171     1      0   1.0  1.222222      0\n",
       "2  2001     115         91     1      1   2.0  1.263736      0\n",
       "3  2001     187        141     1      1   3.0  1.326241      0\n",
       "4  2001     154        105     1      0   4.0  1.466667      0"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Convertire una colonna\n",
    "\n",
    "Possiamo notare che la colonna `Zona` contiene numeri con virgola (*float*). Siccome in realtà `Zona` è un numero intero, convertiamo il numero usando il metodo `astype`. Dobbiamo dare un argomento a `astype` che è un dizionario che associa ad ogni colonna da convertire, il tipo da usare nella conversione."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001</td>\n",
       "      <td>187</td>\n",
       "      <td>141</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona\n",
       "1  2001     209        171     1      0     1\n",
       "2  2001     115         91     1      1     2\n",
       "3  2001     187        141     1      1     3"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_puliti = incidenti_puliti.astype({'Zona': int})\n",
    "incidenti_puliti.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Però, `astype` richiede che i valori da convertire siano tutti non nulli: per questo motivo abbiamo lavorato su `incidenti_puliti`, invece che su `incidenti`. \n",
    "\n",
    "Cosa succede se si eseguono le stesse operazioni su `incidenti`?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Operare solo su alcune righe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Cannot convert non-finite values (NA or inf) to integer",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-22-95ccdedc5ac9>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mincidenti\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mincidenti\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m'Zona'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mint\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/util/_decorators.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    176\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    177\u001b[0m                     \u001b[0mkwargs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnew_arg_name\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnew_arg_value\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 178\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    179\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    180\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0m_deprecate_kwarg\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/core/generic.py\u001b[0m in \u001b[0;36mastype\u001b[0;34m(self, dtype, copy, errors, **kwargs)\u001b[0m\n\u001b[1;32m   4988\u001b[0m             \u001b[0;32mfor\u001b[0m \u001b[0mcol_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcol\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0miteritems\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   4989\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0mcol_name\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 4990\u001b[0;31m                     \u001b[0mresults\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcol\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcol_name\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   4991\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   4992\u001b[0m                     \u001b[0mresults\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mresults\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcol\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mcopy\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mcol\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/util/_decorators.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    176\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    177\u001b[0m                     \u001b[0mkwargs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mnew_arg_name\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnew_arg_value\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 178\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    179\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    180\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0m_deprecate_kwarg\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/core/generic.py\u001b[0m in \u001b[0;36mastype\u001b[0;34m(self, dtype, copy, errors, **kwargs)\u001b[0m\n\u001b[1;32m   4999\u001b[0m             \u001b[0;31m# else, only a single dtype is given\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   5000\u001b[0m             new_data = self._data.astype(dtype=dtype, copy=copy, errors=errors,\n\u001b[0;32m-> 5001\u001b[0;31m                                          **kwargs)\n\u001b[0m\u001b[1;32m   5002\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_constructor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnew_data\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__finalize__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   5003\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/core/internals.py\u001b[0m in \u001b[0;36mastype\u001b[0;34m(self, dtype, **kwargs)\u001b[0m\n\u001b[1;32m   3712\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3713\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3714\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'astype'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3715\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3716\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mconvert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/core/internals.py\u001b[0m in \u001b[0;36mapply\u001b[0;34m(self, f, axes, filter, do_integrity_check, consolidate, **kwargs)\u001b[0m\n\u001b[1;32m   3579\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3580\u001b[0m             \u001b[0mkwargs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'mgr'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3581\u001b[0;31m             \u001b[0mapplied\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mb\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3582\u001b[0m             \u001b[0mresult_blocks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_extend_blocks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mapplied\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mresult_blocks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3583\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/core/internals.py\u001b[0m in \u001b[0;36mastype\u001b[0;34m(self, dtype, copy, errors, values, **kwargs)\u001b[0m\n\u001b[1;32m    573\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'raise'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalues\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    574\u001b[0m         return self._astype(dtype, copy=copy, errors=errors, values=values,\n\u001b[0;32m--> 575\u001b[0;31m                             **kwargs)\n\u001b[0m\u001b[1;32m    576\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    577\u001b[0m     def _astype(self, dtype, copy=False, errors='raise', values=None,\n",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/core/internals.py\u001b[0m in \u001b[0;36m_astype\u001b[0;34m(self, dtype, copy, errors, values, klass, mgr, **kwargs)\u001b[0m\n\u001b[1;32m    662\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    663\u001b[0m                 \u001b[0;31m# _astype_nansafe works fine with 1-d only\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 664\u001b[0;31m                 \u001b[0mvalues\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mastype_nansafe\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mravel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    665\u001b[0m                 \u001b[0mvalues\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mvalues\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    666\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/.miniconda/lib/python3.6/site-packages/pandas/core/dtypes/cast.py\u001b[0m in \u001b[0;36mastype_nansafe\u001b[0;34m(arr, dtype, copy)\u001b[0m\n\u001b[1;32m    700\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    701\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misfinite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 702\u001b[0;31m             raise ValueError('Cannot convert non-finite values (NA or inf) to '\n\u001b[0m\u001b[1;32m    703\u001b[0m                              'integer')\n\u001b[1;32m    704\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: Cannot convert non-finite values (NA or inf) to integer"
     ]
    }
   ],
   "source": [
    "incidenti = incidenti.astype({'Zona': int})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La presenza di valori mancanti rende impossibile l'esecuzione di `astype`. Possiamo sfruttare il fatto che `incidenti` e `incidenti_puliti` hanno lo stesso indice."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "      <th>prova</th>\n",
       "      <th>Zonaint</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>69</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.380000</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.222222</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.263736</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto  prova  Zonaint\n",
       "0  2001      69         50     1      1   NaN  1.380000      0      NaN\n",
       "1  2001     209        171     1      0   1.0  1.222222      0      1.0\n",
       "2  2001     115         91     1      1   2.0  1.263736      0      2.0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.loc[:, 'Zonaint'] = incidenti_puliti.loc[:, 'Zona']\n",
    "incidenti.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La nuova colonna aggiunta (il risultato di `incidenti_puliti.loc[:, 'Zona'].astype(int)`) contiene solo alcune delle righe di `incidenti`. Ma pandas sfrutta gli indici di entrambi i DataFrame per aggiungere la colonna, mettendo valori mancanti dove non trova un indice corrispondente. Ad esempio, `indici_puliti` non ha una riga con indice `0`.\n",
    "Notare che `Zonaint` non ha numeri interi. Ciò è dovuto alla presenza di valori nulli (`NaN`). Per avere interi dobbiamo eliminarli."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Operare solo su alcune righe.\n",
    "\n",
    "Se non si vuole passare per un DataFrame intermedio come `incidenti_puliti`, diventa necessario estendere la chiamata a `loc` esplicitando anche una selezione delle righe, per invocare `astype(int)` solo sulle righe in cui la zona non sia mancante.\n",
    "Questo passo permette anche di gestire adeguatamente la zona mancante, che verrà posta a *0*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "      <th>prova</th>\n",
       "      <th>Zonaint</th>\n",
       "      <th>Zona_int</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>69</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.380000</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.222222</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.263736</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto  prova  Zonaint  \\\n",
       "0  2001      69         50     1      1   NaN  1.380000      0      NaN   \n",
       "1  2001     209        171     1      0   1.0  1.222222      0      1.0   \n",
       "2  2001     115         91     1      1   2.0  1.263736      0      2.0   \n",
       "\n",
       "   Zona_int  \n",
       "0         0  \n",
       "1         1  \n",
       "2         2  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti['Zona_int'] = 0\n",
    "incidenti.loc[incidenti['Zona'].notnull(), 'Zona_int'] = \\\n",
    "incidenti.loc[incidenti['Zona'].notnull(), 'Zona'].astype(int)\n",
    "incidenti.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creare una categoria\n",
    "\n",
    "Una colonna viene chiamata *categoria* se i suoi valori permettono di classificare le righe.\n",
    "Adesso andremo a creare una nuova variabile `tipo` che ha tre valori:\n",
    "\n",
    "*  `a` se il rapporto fra il numero di feriti e il numero di incidenti è minore di 1.3;\n",
    "*  `b` se il rapporto è fra 1.3 (incluso) e 1.4 (escluso);\n",
    "*  `c` se il rapporto è maggiore o uguale a 1.4;\n",
    "\n",
    "Il modo più semplice, ma più lungo, è simile a quanto fatto in precedenza: per ogni valore possibile di `tipo`, si selezionano le righe e si cambia il valore solo di tali righe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "      <th>prova</th>\n",
       "      <th>Zonaint</th>\n",
       "      <th>Zona_int</th>\n",
       "      <th>tipo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>69</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.380000</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.222222</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.263736</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001</td>\n",
       "      <td>187</td>\n",
       "      <td>141</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.326241</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2001</td>\n",
       "      <td>154</td>\n",
       "      <td>105</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.466667</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto  prova  Zonaint  \\\n",
       "0  2001      69         50     1      1   NaN  1.380000      0      NaN   \n",
       "1  2001     209        171     1      0   1.0  1.222222      0      1.0   \n",
       "2  2001     115         91     1      1   2.0  1.263736      0      2.0   \n",
       "3  2001     187        141     1      1   3.0  1.326241      0      3.0   \n",
       "4  2001     154        105     1      0   4.0  1.466667      0      4.0   \n",
       "\n",
       "   Zona_int tipo  \n",
       "0         0    b  \n",
       "1         1    a  \n",
       "2         2    a  \n",
       "3         3    b  \n",
       "4         4    c  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.loc[incidenti['rapporto'] < 1.3, 'tipo'] = 'a'\n",
    "incidenti.loc[(incidenti['rapporto'] >= 1.3) & (incidenti['rapporto'] < 1.4), 'tipo'] = 'b'\n",
    "incidenti.loc[incidenti['rapporto'] >= 1.4, 'tipo'] = 'c'\n",
    "incidenti.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cancellare una colonna"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "del incidenti['Zonaint']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "      <th>prova</th>\n",
       "      <th>Zona_int</th>\n",
       "      <th>tipo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>69</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.380000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.222222</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.263736</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto  prova  Zona_int tipo\n",
       "0  2001      69         50     1      1   NaN  1.380000      0         0    b\n",
       "1  2001     209        171     1      0   1.0  1.222222      0         1    a\n",
       "2  2001     115         91     1      1   2.0  1.263736      0         2    a"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Funzioni"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Funzioni aggregate (Serie)\n",
    "\n",
    "Alcune funzioni possono essere applicate su una Serie intera. Ad esempio la funzione `len` permette di ottenere il numero di righe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1920"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(incidenti)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Sicuramente più interessanti sono le funzioni che calcolano minimo, massimo o la somma. Queste funzioni sono applicate a Serie."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "278952"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti['Feriti'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "22"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti['Feriti'].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "557"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti['Feriti'].max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Di particolare rilievo è che eventuali valori mancanti vengono gestiti correttamente (vengono ignorati)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Funzioni aggregate (DataFrame)\n",
    "\n",
    "Le stesse funzioni aggregate, se vengono invocate su un DataFrame, vengono applicate su tutte le colonne, incluse quelle che non contengono valori numerici.\n",
    "\n",
    "Il risultato è sempre una Serie."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Anno         2001\n",
       "Feriti         22\n",
       "Incidenti      17\n",
       "Mese            1\n",
       "Morti           0\n",
       "Zona            1\n",
       "rapporto        1\n",
       "prova           0\n",
       "Zona_int        0\n",
       "tipo            a\n",
       "dtype: object"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.min()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se si vuole applicare la funzione solo alle colonne numeriche, bisogna usare l'opzione `numeric_only`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Anno         2001.0\n",
       "Feriti         22.0\n",
       "Incidenti      17.0\n",
       "Mese            1.0\n",
       "Morti           0.0\n",
       "Zona            1.0\n",
       "rapporto        1.0\n",
       "prova           0.0\n",
       "Zona_int        0.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.min(numeric_only = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Raggruppare righe\n",
    "\n",
    "Talvolta il fatto che più righe condividano uno stesso valore può essere sfruttare per costruire gruppi di righe, che poi possono essere analizzati separatamente.\n",
    "\n",
    "Ad esempio, nel dataset `incidenti`, vogliamo contare quanti incidenti abbiamo avuto per ogni zona.\n",
    "Ciò richiede di raggruppare con una `groupby`.\n",
    "\n",
    "In realtà andiamo a raggruppare per `Zona_int` solo per motivi estetici (non è gradevole vedere una zona con il punto decimale)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<pandas.core.groupby.groupby.DataFrameGroupBy object at 0x7f7d467103c8>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.groupby('Zona_int')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Il risultato di una `groupby` è un oggetto opaco, che può essere sfruttato tramite operatori su gruppi. \n",
    "Pertanto da solo non serve a molto. Ma diventa molto utile se abbinato a `count`, `min`, `max`, `mean`, `sum`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Raggruppare righe con operatori\n",
    "\n",
    "Per contare il numero di incidenti, dobbiamo sommare il valore della colonna `Incidenti`, distinguendo per ogni zona.\n",
    "\n",
    "\n",
    "Notare in questo caso che le righe che hanno *Zona* mancante non vengono considerate nei raggruppamenti."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Zona_int\n",
       "0    11470\n",
       "1    24846\n",
       "2    13958\n",
       "3    25470\n",
       "4    20248\n",
       "5    20000\n",
       "6    17524\n",
       "7    21432\n",
       "8    25696\n",
       "9    27392\n",
       "Name: Incidenti, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.groupby('Zona_int')['Incidenti'].sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Estrarre un gruppo\n",
    "\n",
    "Per estrarre un singolo gruppo si può usare il metodo `get_group`. Ad esempio, per estrarre tutti i dati relativi al gruppo della zona 3:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "      <th>prova</th>\n",
       "      <th>Zona_int</th>\n",
       "      <th>tipo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001</td>\n",
       "      <td>187</td>\n",
       "      <td>141</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.326241</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2001</td>\n",
       "      <td>200</td>\n",
       "      <td>147</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.360544</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2001</td>\n",
       "      <td>260</td>\n",
       "      <td>178</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.460674</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>2001</td>\n",
       "      <td>250</td>\n",
       "      <td>177</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.412429</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>2001</td>\n",
       "      <td>287</td>\n",
       "      <td>207</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.386473</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>2001</td>\n",
       "      <td>315</td>\n",
       "      <td>211</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.492891</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>63</th>\n",
       "      <td>2001</td>\n",
       "      <td>301</td>\n",
       "      <td>223</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.349776</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>2001</td>\n",
       "      <td>167</td>\n",
       "      <td>112</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.491071</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>2001</td>\n",
       "      <td>261</td>\n",
       "      <td>182</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.434066</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>2001</td>\n",
       "      <td>311</td>\n",
       "      <td>225</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.382222</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>2001</td>\n",
       "      <td>285</td>\n",
       "      <td>211</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.350711</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113</th>\n",
       "      <td>2001</td>\n",
       "      <td>239</td>\n",
       "      <td>162</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.475309</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>2002</td>\n",
       "      <td>208</td>\n",
       "      <td>149</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.395973</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>2002</td>\n",
       "      <td>193</td>\n",
       "      <td>153</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.261438</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143</th>\n",
       "      <td>2002</td>\n",
       "      <td>264</td>\n",
       "      <td>190</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.389474</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>2002</td>\n",
       "      <td>266</td>\n",
       "      <td>180</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.477778</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>163</th>\n",
       "      <td>2002</td>\n",
       "      <td>282</td>\n",
       "      <td>208</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.355769</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>173</th>\n",
       "      <td>2002</td>\n",
       "      <td>294</td>\n",
       "      <td>211</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.393365</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>2002</td>\n",
       "      <td>277</td>\n",
       "      <td>206</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.344660</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>193</th>\n",
       "      <td>2002</td>\n",
       "      <td>163</td>\n",
       "      <td>109</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.495413</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>203</th>\n",
       "      <td>2002</td>\n",
       "      <td>237</td>\n",
       "      <td>185</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.281081</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>213</th>\n",
       "      <td>2002</td>\n",
       "      <td>324</td>\n",
       "      <td>235</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.378723</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223</th>\n",
       "      <td>2002</td>\n",
       "      <td>276</td>\n",
       "      <td>197</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.401015</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>233</th>\n",
       "      <td>2002</td>\n",
       "      <td>225</td>\n",
       "      <td>162</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.388889</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243</th>\n",
       "      <td>2003</td>\n",
       "      <td>215</td>\n",
       "      <td>156</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.378205</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>253</th>\n",
       "      <td>2003</td>\n",
       "      <td>213</td>\n",
       "      <td>147</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.448980</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>263</th>\n",
       "      <td>2003</td>\n",
       "      <td>229</td>\n",
       "      <td>167</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.371257</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>2003</td>\n",
       "      <td>225</td>\n",
       "      <td>170</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.323529</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>283</th>\n",
       "      <td>2003</td>\n",
       "      <td>301</td>\n",
       "      <td>207</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.454106</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>c</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293</th>\n",
       "      <td>2003</td>\n",
       "      <td>238</td>\n",
       "      <td>188</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.265957</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1623</th>\n",
       "      <td>2014</td>\n",
       "      <td>117</td>\n",
       "      <td>92</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.271739</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1633</th>\n",
       "      <td>2014</td>\n",
       "      <td>60</td>\n",
       "      <td>44</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.363636</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1643</th>\n",
       "      <td>2014</td>\n",
       "      <td>129</td>\n",
       "      <td>104</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.240385</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1653</th>\n",
       "      <td>2014</td>\n",
       "      <td>117</td>\n",
       "      <td>92</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.271739</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1663</th>\n",
       "      <td>2014</td>\n",
       "      <td>97</td>\n",
       "      <td>76</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.276316</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1673</th>\n",
       "      <td>2014</td>\n",
       "      <td>102</td>\n",
       "      <td>76</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.342105</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1683</th>\n",
       "      <td>2015</td>\n",
       "      <td>77</td>\n",
       "      <td>67</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.149254</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1693</th>\n",
       "      <td>2015</td>\n",
       "      <td>39</td>\n",
       "      <td>33</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.181818</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1703</th>\n",
       "      <td>2015</td>\n",
       "      <td>134</td>\n",
       "      <td>106</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.264151</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1713</th>\n",
       "      <td>2015</td>\n",
       "      <td>139</td>\n",
       "      <td>108</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.287037</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1723</th>\n",
       "      <td>2015</td>\n",
       "      <td>106</td>\n",
       "      <td>82</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.292683</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1733</th>\n",
       "      <td>2015</td>\n",
       "      <td>132</td>\n",
       "      <td>106</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.245283</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1743</th>\n",
       "      <td>2015</td>\n",
       "      <td>131</td>\n",
       "      <td>97</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.350515</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1753</th>\n",
       "      <td>2015</td>\n",
       "      <td>52</td>\n",
       "      <td>42</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.238095</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1763</th>\n",
       "      <td>2015</td>\n",
       "      <td>102</td>\n",
       "      <td>81</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.259259</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1773</th>\n",
       "      <td>2015</td>\n",
       "      <td>165</td>\n",
       "      <td>130</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.269231</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1783</th>\n",
       "      <td>2015</td>\n",
       "      <td>97</td>\n",
       "      <td>82</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.182927</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1793</th>\n",
       "      <td>2015</td>\n",
       "      <td>82</td>\n",
       "      <td>66</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.242424</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1803</th>\n",
       "      <td>2016</td>\n",
       "      <td>83</td>\n",
       "      <td>67</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.238806</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1813</th>\n",
       "      <td>2016</td>\n",
       "      <td>101</td>\n",
       "      <td>77</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.311688</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1823</th>\n",
       "      <td>2016</td>\n",
       "      <td>100</td>\n",
       "      <td>73</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.369863</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1833</th>\n",
       "      <td>2016</td>\n",
       "      <td>139</td>\n",
       "      <td>111</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.252252</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1843</th>\n",
       "      <td>2016</td>\n",
       "      <td>147</td>\n",
       "      <td>114</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.289474</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1853</th>\n",
       "      <td>2016</td>\n",
       "      <td>127</td>\n",
       "      <td>98</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.295918</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1863</th>\n",
       "      <td>2016</td>\n",
       "      <td>132</td>\n",
       "      <td>98</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.346939</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1873</th>\n",
       "      <td>2016</td>\n",
       "      <td>69</td>\n",
       "      <td>52</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.326923</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1883</th>\n",
       "      <td>2016</td>\n",
       "      <td>110</td>\n",
       "      <td>85</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.294118</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1893</th>\n",
       "      <td>2016</td>\n",
       "      <td>116</td>\n",
       "      <td>94</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.234043</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1903</th>\n",
       "      <td>2016</td>\n",
       "      <td>119</td>\n",
       "      <td>89</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.337079</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1913</th>\n",
       "      <td>2016</td>\n",
       "      <td>86</td>\n",
       "      <td>71</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.211268</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>192 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto  prova  Zona_int  \\\n",
       "3     2001     187        141     1      1   3.0  1.326241      0         3   \n",
       "13    2001     200        147     2      1   3.0  1.360544      0         3   \n",
       "23    2001     260        178     3      1   3.0  1.460674      0         3   \n",
       "33    2001     250        177     4      0   3.0  1.412429      0         3   \n",
       "43    2001     287        207     5      1   3.0  1.386473      0         3   \n",
       "53    2001     315        211     6      0   3.0  1.492891      0         3   \n",
       "63    2001     301        223     7      2   3.0  1.349776      0         3   \n",
       "73    2001     167        112     8      0   3.0  1.491071      0         3   \n",
       "83    2001     261        182     9      0   3.0  1.434066      0         3   \n",
       "93    2001     311        225    10      1   3.0  1.382222      0         3   \n",
       "103   2001     285        211    11      1   3.0  1.350711      0         3   \n",
       "113   2001     239        162    12      0   3.0  1.475309      0         3   \n",
       "123   2002     208        149     1      1   3.0  1.395973      0         3   \n",
       "133   2002     193        153     2      1   3.0  1.261438      0         3   \n",
       "143   2002     264        190     3      1   3.0  1.389474      0         3   \n",
       "153   2002     266        180     4      0   3.0  1.477778      0         3   \n",
       "163   2002     282        208     5      0   3.0  1.355769      0         3   \n",
       "173   2002     294        211     6      0   3.0  1.393365      0         3   \n",
       "183   2002     277        206     7      0   3.0  1.344660      0         3   \n",
       "193   2002     163        109     8      0   3.0  1.495413      0         3   \n",
       "203   2002     237        185     9      0   3.0  1.281081      0         3   \n",
       "213   2002     324        235    10      0   3.0  1.378723      0         3   \n",
       "223   2002     276        197    11      0   3.0  1.401015      0         3   \n",
       "233   2002     225        162    12      0   3.0  1.388889      0         3   \n",
       "243   2003     215        156     1      0   3.0  1.378205      0         3   \n",
       "253   2003     213        147     2      0   3.0  1.448980      0         3   \n",
       "263   2003     229        167     3      1   3.0  1.371257      0         3   \n",
       "273   2003     225        170     4      1   3.0  1.323529      0         3   \n",
       "283   2003     301        207     5      1   3.0  1.454106      0         3   \n",
       "293   2003     238        188     6      0   3.0  1.265957      0         3   \n",
       "...    ...     ...        ...   ...    ...   ...       ...    ...       ...   \n",
       "1623  2014     117         92     7      1   3.0  1.271739      0         3   \n",
       "1633  2014      60         44     8      0   3.0  1.363636      0         3   \n",
       "1643  2014     129        104     9      0   3.0  1.240385      0         3   \n",
       "1653  2014     117         92    10      0   3.0  1.271739      0         3   \n",
       "1663  2014      97         76    11      0   3.0  1.276316      0         3   \n",
       "1673  2014     102         76    12      0   3.0  1.342105      0         3   \n",
       "1683  2015      77         67     1      0   3.0  1.149254      0         3   \n",
       "1693  2015      39         33     2      0   3.0  1.181818      0         3   \n",
       "1703  2015     134        106     3      1   3.0  1.264151      0         3   \n",
       "1713  2015     139        108     4      1   3.0  1.287037      0         3   \n",
       "1723  2015     106         82     5      0   3.0  1.292683      0         3   \n",
       "1733  2015     132        106     6      2   3.0  1.245283      0         3   \n",
       "1743  2015     131         97     7      0   3.0  1.350515      0         3   \n",
       "1753  2015      52         42     8      0   3.0  1.238095      0         3   \n",
       "1763  2015     102         81     9      0   3.0  1.259259      0         3   \n",
       "1773  2015     165        130    10      1   3.0  1.269231      0         3   \n",
       "1783  2015      97         82    11      0   3.0  1.182927      0         3   \n",
       "1793  2015      82         66    12      1   3.0  1.242424      0         3   \n",
       "1803  2016      83         67     1      0   3.0  1.238806      0         3   \n",
       "1813  2016     101         77     2      0   3.0  1.311688      0         3   \n",
       "1823  2016     100         73     3      0   3.0  1.369863      0         3   \n",
       "1833  2016     139        111     4      2   3.0  1.252252      0         3   \n",
       "1843  2016     147        114     5      0   3.0  1.289474      0         3   \n",
       "1853  2016     127         98     6      0   3.0  1.295918      0         3   \n",
       "1863  2016     132         98     7      1   3.0  1.346939      0         3   \n",
       "1873  2016      69         52     8      0   3.0  1.326923      0         3   \n",
       "1883  2016     110         85     9      0   3.0  1.294118      0         3   \n",
       "1893  2016     116         94    10      0   3.0  1.234043      0         3   \n",
       "1903  2016     119         89    11      0   3.0  1.337079      0         3   \n",
       "1913  2016      86         71    12      0   3.0  1.211268      0         3   \n",
       "\n",
       "     tipo  \n",
       "3       b  \n",
       "13      b  \n",
       "23      c  \n",
       "33      c  \n",
       "43      b  \n",
       "53      c  \n",
       "63      b  \n",
       "73      c  \n",
       "83      c  \n",
       "93      b  \n",
       "103     b  \n",
       "113     c  \n",
       "123     b  \n",
       "133     a  \n",
       "143     b  \n",
       "153     c  \n",
       "163     b  \n",
       "173     b  \n",
       "183     b  \n",
       "193     c  \n",
       "203     a  \n",
       "213     b  \n",
       "223     c  \n",
       "233     b  \n",
       "243     b  \n",
       "253     c  \n",
       "263     b  \n",
       "273     b  \n",
       "283     c  \n",
       "293     a  \n",
       "...   ...  \n",
       "1623    a  \n",
       "1633    b  \n",
       "1643    a  \n",
       "1653    a  \n",
       "1663    a  \n",
       "1673    b  \n",
       "1683    a  \n",
       "1693    a  \n",
       "1703    a  \n",
       "1713    a  \n",
       "1723    a  \n",
       "1733    a  \n",
       "1743    b  \n",
       "1753    a  \n",
       "1763    a  \n",
       "1773    a  \n",
       "1783    a  \n",
       "1793    a  \n",
       "1803    a  \n",
       "1813    b  \n",
       "1823    b  \n",
       "1833    a  \n",
       "1843    a  \n",
       "1853    a  \n",
       "1863    b  \n",
       "1873    b  \n",
       "1883    a  \n",
       "1893    a  \n",
       "1903    b  \n",
       "1913    a  \n",
       "\n",
       "[192 rows x 10 columns]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.groupby('Zona_int').get_group(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Operatori aggregati\n",
    "\n",
    "L'elenco delle funzioni che possono essere applicate ad una `groupby` sono chiamati operatori aggregati. Sono:\n",
    "\n",
    "*  `mean()` calcola il valore medio\n",
    "*  `median()` calcola il valore mediano\n",
    "*  `sum()`  calcola la somma\n",
    "*  `size()` calcola il numero di elementi dei gruppi\n",
    "*  `count()` calcola il numero di elementi non mancanti nei gruppi\n",
    "*  `std()` calcola la deviazione standard\n",
    "*  `var()` calcola la varianza\n",
    "*  `sem()` calcola l'errore standard della media\n",
    "*  `describe()` calcola le statistiche descrittive\n",
    "*  `first()` calcola il primo valore di ogni gruppo\n",
    "*  `last()` calcola l'ultimo valore di ogni gruppo\n",
    "*  `nth()` estrae alcuni valori\n",
    "*  `min()` calcola il minimo valore di ogni gruppo\n",
    "*  `max()` calcola il massimo valore di ogni gruppo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Size e Count\n",
    "\n",
    "Ricordiamo che `count` esclude dal conteggio i valori mancanti, mentre `size` include i valori mancanti.\n",
    "Andiamo a confrontare i risultati di queste operazioni, considerano solo la colonna `Zona`, visto che è l'unica a contenere valori mancanti.\n",
    "\n",
    "Chiaramente ci aspettiamo che `count` restituisca valori minori di quelli calcolati da `size`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Anno\n",
       "2001    120\n",
       "2002    120\n",
       "2003    120\n",
       "2004    120\n",
       "2005    120\n",
       "2006    120\n",
       "2007    120\n",
       "2008    120\n",
       "2009    120\n",
       "2010    120\n",
       "2011    120\n",
       "2012    120\n",
       "2013    120\n",
       "2014    120\n",
       "2015    120\n",
       "2016    120\n",
       "Name: Zona, dtype: int64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.groupby('Anno')['Zona'].size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Anno\n",
       "2001    108\n",
       "2002    108\n",
       "2003    108\n",
       "2004    108\n",
       "2005    108\n",
       "2006    108\n",
       "2007    108\n",
       "2008    108\n",
       "2009    108\n",
       "2010    108\n",
       "2011    108\n",
       "2012    108\n",
       "2013    108\n",
       "2014    108\n",
       "2015    108\n",
       "2016    108\n",
       "Name: Zona, dtype: int64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.groupby('Anno')['Zona'].count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Raggruppare su più colonne\n",
    "\n",
    "Possiamo utilizzare più di una colonna per raggruppare, fornendo una lista come argomento della `groupby`.\n",
    "\n",
    "Ad esempio, andiamo a calcolare il numero totale di incidenti, diviso per zona e mese."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Zona  Mese\n",
       "1.0   1       1909\n",
       "      2       1755\n",
       "      3       2164\n",
       "      4       2036\n",
       "      5       2466\n",
       "      6       2218\n",
       "      7       2182\n",
       "      8        842\n",
       "      9       2220\n",
       "      10      2566\n",
       "      11      2466\n",
       "      12      2022\n",
       "2.0   1       1006\n",
       "      2        996\n",
       "      3       1149\n",
       "      4       1232\n",
       "      5       1343\n",
       "      6       1310\n",
       "      7       1323\n",
       "      8        694\n",
       "      9       1194\n",
       "      10      1402\n",
       "      11      1217\n",
       "      12      1092\n",
       "3.0   1       1873\n",
       "      2       1802\n",
       "      3       2168\n",
       "      4       2142\n",
       "      5       2408\n",
       "      6       2418\n",
       "              ... \n",
       "7.0   7       1923\n",
       "      8       1038\n",
       "      9       1906\n",
       "      10      2084\n",
       "      11      1871\n",
       "      12      1700\n",
       "8.0   1       1882\n",
       "      2       1743\n",
       "      3       2248\n",
       "      4       2135\n",
       "      5       2535\n",
       "      6       2377\n",
       "      7       2325\n",
       "      8       1179\n",
       "      9       2251\n",
       "      10      2517\n",
       "      11      2430\n",
       "      12      2074\n",
       "9.0   1       2085\n",
       "      2       1936\n",
       "      3       2382\n",
       "      4       2316\n",
       "      5       2633\n",
       "      6       2502\n",
       "      7       2445\n",
       "      8       1402\n",
       "      9       2365\n",
       "      10      2658\n",
       "      11      2457\n",
       "      12      2211\n",
       "Name: Incidenti, Length: 108, dtype: int64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.groupby(['Zona', 'Mese']).sum()['Incidenti']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Raggruppamenti e indice\n",
    "\n",
    "Il comportamento di default è che le variabili indicate nella `groupby` diventino l'indice della Serie o del DataFrame risultante."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "FrozenList(['Zona_int', 'Mese'])"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gruppi = incidenti.groupby(['Zona_int', 'Mese']).sum()\n",
    "gruppi.index.names"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se invece si vuole che siano nuove colonne del DataFrame risultante, bisogna usare l'opzione `as_index`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "FrozenList([None])"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gruppi2 = incidenti.groupby(['Zona_int', 'Mese'], as_index = False).sum()\n",
    "gruppi2.index.names"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Controlliamo esplicitamente che `Zona_int` e `Mese` siano colonne del DataFrame `gruppi2`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Zona_int', 'Mese', 'Anno', 'Feriti', 'Incidenti', 'Morti', 'Zona',\n",
       "       'rapporto', 'prova'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gruppi2.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Apply e Tabelle pivot"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Apply\n",
    "\n",
    "Il metodo `apply` permette di applicare una qualunque funzione a tutte le righe (o, più raramente, le colonne) di un  DataFrame o di una Serie. La funzione riceve direttamente una riga del DataFrame.\n",
    "\n",
    "Una forma più sofisticata permette di applicare una funzione che riceve in input un intero DataFrame.\n",
    "\n",
    "Come primo esempio, costruiamo una nuova variabile `pericolo` che corrisponde al numero di incidenti, più il numero di feriti moltiplicato per 100. Il primo passo è costruire la funzione che calcola tale indice."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calcola_p(riga):\n",
    "    return riga['Incidenti'] + riga['Feriti'] * 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Poi si può usare la `apply`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "      <th>prova</th>\n",
       "      <th>Zona_int</th>\n",
       "      <th>tipo</th>\n",
       "      <th>pericolo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>69</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.380000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>b</td>\n",
       "      <td>6950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.222222</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>a</td>\n",
       "      <td>21071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.263736</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>a</td>\n",
       "      <td>11591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001</td>\n",
       "      <td>187</td>\n",
       "      <td>141</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.326241</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "      <td>18841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2001</td>\n",
       "      <td>154</td>\n",
       "      <td>105</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.466667</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>c</td>\n",
       "      <td>15505</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto  prova  Zona_int tipo  \\\n",
       "0  2001      69         50     1      1   NaN  1.380000      0         0    b   \n",
       "1  2001     209        171     1      0   1.0  1.222222      0         1    a   \n",
       "2  2001     115         91     1      1   2.0  1.263736      0         2    a   \n",
       "3  2001     187        141     1      1   3.0  1.326241      0         3    b   \n",
       "4  2001     154        105     1      0   4.0  1.466667      0         4    c   \n",
       "\n",
       "   pericolo  \n",
       "0      6950  \n",
       "1     21071  \n",
       "2     11591  \n",
       "3     18841  \n",
       "4     15505  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti['pericolo'] = incidenti.apply(calcola_p, axis = 1)\n",
    "incidenti.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creare una categoria con apply\n",
    "\n",
    "La soluzione che abbiamo visto in precedenza per creare la categoria `tipo` ha due difetti: è fragile (perchè il valore soglia *1.3* compare in due istruzioni diverse, quindi è più probabile scrivere un valore sbagliato) e richiede un'istruzione distinta per ogni valore possibile di `tipo`.\n",
    "\n",
    "Quindi prima scriviamo una funzione che riceve in input un numero e calcola il valore della categoria."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def categoria(n):\n",
    "    soglie = [('a', 1.3), ('b', 1.4), ('c', None)]\n",
    "    for (categoria, soglia) in soglie:\n",
    "        if soglia is None or n < soglia:\n",
    "            return categoria"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Poi è possibile *applicare* la funzione a `rapporto`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>Anno</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Incidenti</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Morti</th>\n",
       "      <th>Zona</th>\n",
       "      <th>rapporto</th>\n",
       "      <th>prova</th>\n",
       "      <th>Zona_int</th>\n",
       "      <th>tipo</th>\n",
       "      <th>pericolo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>69</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.380000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>b</td>\n",
       "      <td>6950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>209</td>\n",
       "      <td>171</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.222222</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>a</td>\n",
       "      <td>21071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>115</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.263736</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>a</td>\n",
       "      <td>11591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001</td>\n",
       "      <td>187</td>\n",
       "      <td>141</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.326241</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>b</td>\n",
       "      <td>18841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2001</td>\n",
       "      <td>154</td>\n",
       "      <td>105</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.466667</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>c</td>\n",
       "      <td>15505</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Anno  Feriti  Incidenti  Mese  Morti  Zona  rapporto  prova  Zona_int tipo  \\\n",
       "0  2001      69         50     1      1   NaN  1.380000      0         0    b   \n",
       "1  2001     209        171     1      0   1.0  1.222222      0         1    a   \n",
       "2  2001     115         91     1      1   2.0  1.263736      0         2    a   \n",
       "3  2001     187        141     1      1   3.0  1.326241      0         3    b   \n",
       "4  2001     154        105     1      0   4.0  1.466667      0         4    c   \n",
       "\n",
       "   pericolo  \n",
       "0      6950  \n",
       "1     21071  \n",
       "2     11591  \n",
       "3     18841  \n",
       "4     15505  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti['tipo'] = incidenti['rapporto'].apply(categoria)\n",
    "incidenti.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Differenze\n",
    "\n",
    "Nelle slide precedenti abbiamo visto due diverse tipologie di utilizzo di apply:\n",
    "\n",
    "1.  `incidenti.apply(calcola_p, axis = 1)`\n",
    "2.  `incidenti['rapporto'].apply(categoria)`\n",
    "\n",
    "Nel caso 1 diventa essenziale l'opzione `axis = 1` che indica a pandas che la funzione deve essere invocata per ogni riga. In questo caso la funzione `calcola_p` riceve un argomento che è la riga di un DataFrame. Tecnicamente la riga è un dizionario con chiavi corrispondenti alle colonne del DataFrame.\n",
    "\n",
    "Nel caso 2 invece la `apply` viene chiamata su una Serie. In questo caso la funzione `categoria` riceve un singolo valore in input e la parte `axis = 1` non è necessaria.\n",
    "\n",
    "Un terzo caso prevede invece che la funzione riceva un intero DataFrame"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calcolo percentuale\n",
    "\n",
    "Adesso vogliamo aggiungere una colonna che contiene la percentuale del numero di incidenti in un certo mese rispetto a tutti quelli dell'anno nella zona di riferimento. Questo calcolo richiede più passi.\n",
    "\n",
    "Tramite un raggruppamento, possiamo calcolare il numero totale di incidenti per ogni anno e zona."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Anno  Zona_int\n",
       "2001  0            605\n",
       "      1           2207\n",
       "      2           1171\n",
       "      3           2176\n",
       "      4           1782\n",
       "      5           1693\n",
       "      6           1409\n",
       "      7           1828\n",
       "      8           2200\n",
       "      9           2387\n",
       "2002  0            685\n",
       "      1           2177\n",
       "      2           1272\n",
       "      3           2185\n",
       "      4           1713\n",
       "      5           1609\n",
       "      6           1475\n",
       "      7           1774\n",
       "      8           2147\n",
       "      9           2330\n",
       "2003  0            671\n",
       "      1           2056\n",
       "      2           1160\n",
       "      3           2061\n",
       "      4           1623\n",
       "      5           1628\n",
       "      6           1371\n",
       "      7           1868\n",
       "      8           2034\n",
       "      9           2247\n",
       "                  ... \n",
       "2014  0            673\n",
       "      1           1039\n",
       "      2            567\n",
       "      3           1008\n",
       "      4            833\n",
       "      5            839\n",
       "      6            778\n",
       "      7            944\n",
       "      8           1090\n",
       "      9           1069\n",
       "2015  0            938\n",
       "      1            969\n",
       "      2            522\n",
       "      3           1000\n",
       "      4            786\n",
       "      5            754\n",
       "      6            764\n",
       "      7            845\n",
       "      8           1016\n",
       "      9           1033\n",
       "2016  0            480\n",
       "      1           1009\n",
       "      2            566\n",
       "      3           1029\n",
       "      4            888\n",
       "      5            877\n",
       "      6            721\n",
       "      7            978\n",
       "      8           1060\n",
       "      9           1162\n",
       "Name: Incidenti, Length: 160, dtype: int64"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti.groupby(['Anno', 'Zona_int']).sum()['Incidenti']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tabella pivot\n",
    "\n",
    "Una tabella pivot viene ottenuta a partire da un DataFrame per riassumere i dati di quest'ultimo.\n",
    "In particolare, in una tabella pivot sono presenti tre elementi:\n",
    "*  un insieme di variabili per guidano il raggruppamento di righe\n",
    "*  una variabile (o un insieme di variabili) che etichetta nuove colonne\n",
    "*  una funzione per aggregare i valori.\n",
    "\n",
    "Ad esempio, andiamo a costruire una tabella pivot che riassume per ogni *anno* il numero totale (quindi la *somma) di incidenti (feriti, ecc.), ed in cui si ha una colonna per ogni *zona*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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>Zona_int</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Anno</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>2001</th>\n",
       "      <td>605</td>\n",
       "      <td>2207</td>\n",
       "      <td>1171</td>\n",
       "      <td>2176</td>\n",
       "      <td>1782</td>\n",
       "      <td>1693</td>\n",
       "      <td>1409</td>\n",
       "      <td>1828</td>\n",
       "      <td>2200</td>\n",
       "      <td>2387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2002</th>\n",
       "      <td>685</td>\n",
       "      <td>2177</td>\n",
       "      <td>1272</td>\n",
       "      <td>2185</td>\n",
       "      <td>1713</td>\n",
       "      <td>1609</td>\n",
       "      <td>1475</td>\n",
       "      <td>1774</td>\n",
       "      <td>2147</td>\n",
       "      <td>2330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>671</td>\n",
       "      <td>2056</td>\n",
       "      <td>1160</td>\n",
       "      <td>2061</td>\n",
       "      <td>1623</td>\n",
       "      <td>1628</td>\n",
       "      <td>1371</td>\n",
       "      <td>1868</td>\n",
       "      <td>2034</td>\n",
       "      <td>2247</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004</th>\n",
       "      <td>682</td>\n",
       "      <td>1958</td>\n",
       "      <td>1101</td>\n",
       "      <td>2155</td>\n",
       "      <td>1579</td>\n",
       "      <td>1623</td>\n",
       "      <td>1287</td>\n",
       "      <td>1650</td>\n",
       "      <td>2051</td>\n",
       "      <td>2230</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>716</td>\n",
       "      <td>1867</td>\n",
       "      <td>1023</td>\n",
       "      <td>1880</td>\n",
       "      <td>1479</td>\n",
       "      <td>1512</td>\n",
       "      <td>1232</td>\n",
       "      <td>1558</td>\n",
       "      <td>1898</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>675</td>\n",
       "      <td>1809</td>\n",
       "      <td>1011</td>\n",
       "      <td>1821</td>\n",
       "      <td>1527</td>\n",
       "      <td>1397</td>\n",
       "      <td>1204</td>\n",
       "      <td>1496</td>\n",
       "      <td>1996</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>574</td>\n",
       "      <td>1723</td>\n",
       "      <td>974</td>\n",
       "      <td>1735</td>\n",
       "      <td>1383</td>\n",
       "      <td>1432</td>\n",
       "      <td>1302</td>\n",
       "      <td>1495</td>\n",
       "      <td>1802</td>\n",
       "      <td>1956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>498</td>\n",
       "      <td>1502</td>\n",
       "      <td>916</td>\n",
       "      <td>1715</td>\n",
       "      <td>1322</td>\n",
       "      <td>1309</td>\n",
       "      <td>1132</td>\n",
       "      <td>1425</td>\n",
       "      <td>1747</td>\n",
       "      <td>1763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009</th>\n",
       "      <td>494</td>\n",
       "      <td>1552</td>\n",
       "      <td>844</td>\n",
       "      <td>1507</td>\n",
       "      <td>1266</td>\n",
       "      <td>1232</td>\n",
       "      <td>1120</td>\n",
       "      <td>1308</td>\n",
       "      <td>1560</td>\n",
       "      <td>1647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010</th>\n",
       "      <td>535</td>\n",
       "      <td>1449</td>\n",
       "      <td>767</td>\n",
       "      <td>1468</td>\n",
       "      <td>1217</td>\n",
       "      <td>1196</td>\n",
       "      <td>1002</td>\n",
       "      <td>1170</td>\n",
       "      <td>1493</td>\n",
       "      <td>1615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>2261</td>\n",
       "      <td>1128</td>\n",
       "      <td>702</td>\n",
       "      <td>1273</td>\n",
       "      <td>906</td>\n",
       "      <td>903</td>\n",
       "      <td>879</td>\n",
       "      <td>1016</td>\n",
       "      <td>1103</td>\n",
       "      <td>1328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012</th>\n",
       "      <td>452</td>\n",
       "      <td>1264</td>\n",
       "      <td>778</td>\n",
       "      <td>1311</td>\n",
       "      <td>1012</td>\n",
       "      <td>1055</td>\n",
       "      <td>926</td>\n",
       "      <td>1077</td>\n",
       "      <td>1294</td>\n",
       "      <td>1356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>531</td>\n",
       "      <td>1137</td>\n",
       "      <td>584</td>\n",
       "      <td>1146</td>\n",
       "      <td>932</td>\n",
       "      <td>941</td>\n",
       "      <td>922</td>\n",
       "      <td>1000</td>\n",
       "      <td>1205</td>\n",
       "      <td>1237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014</th>\n",
       "      <td>673</td>\n",
       "      <td>1039</td>\n",
       "      <td>567</td>\n",
       "      <td>1008</td>\n",
       "      <td>833</td>\n",
       "      <td>839</td>\n",
       "      <td>778</td>\n",
       "      <td>944</td>\n",
       "      <td>1090</td>\n",
       "      <td>1069</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015</th>\n",
       "      <td>938</td>\n",
       "      <td>969</td>\n",
       "      <td>522</td>\n",
       "      <td>1000</td>\n",
       "      <td>786</td>\n",
       "      <td>754</td>\n",
       "      <td>764</td>\n",
       "      <td>845</td>\n",
       "      <td>1016</td>\n",
       "      <td>1033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016</th>\n",
       "      <td>480</td>\n",
       "      <td>1009</td>\n",
       "      <td>566</td>\n",
       "      <td>1029</td>\n",
       "      <td>888</td>\n",
       "      <td>877</td>\n",
       "      <td>721</td>\n",
       "      <td>978</td>\n",
       "      <td>1060</td>\n",
       "      <td>1162</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Zona_int     0     1     2     3     4     5     6     7     8     9\n",
       "Anno                                                                \n",
       "2001       605  2207  1171  2176  1782  1693  1409  1828  2200  2387\n",
       "2002       685  2177  1272  2185  1713  1609  1475  1774  2147  2330\n",
       "2003       671  2056  1160  2061  1623  1628  1371  1868  2034  2247\n",
       "2004       682  1958  1101  2155  1579  1623  1287  1650  2051  2230\n",
       "2005       716  1867  1023  1880  1479  1512  1232  1558  1898  2016\n",
       "2006       675  1809  1011  1821  1527  1397  1204  1496  1996  2016\n",
       "2007       574  1723   974  1735  1383  1432  1302  1495  1802  1956\n",
       "2008       498  1502   916  1715  1322  1309  1132  1425  1747  1763\n",
       "2009       494  1552   844  1507  1266  1232  1120  1308  1560  1647\n",
       "2010       535  1449   767  1468  1217  1196  1002  1170  1493  1615\n",
       "2011      2261  1128   702  1273   906   903   879  1016  1103  1328\n",
       "2012       452  1264   778  1311  1012  1055   926  1077  1294  1356\n",
       "2013       531  1137   584  1146   932   941   922  1000  1205  1237\n",
       "2014       673  1039   567  1008   833   839   778   944  1090  1069\n",
       "2015       938   969   522  1000   786   754   764   845  1016  1033\n",
       "2016       480  1009   566  1029   888   877   721   978  1060  1162"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.pivot_table(incidenti, index = 'Anno', columns = ['Zona_int'], values = 'Incidenti', aggfunc = sum)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tabella pivot con medie\n",
    "\n",
    "Talvolta la funzione per aggregare i valori non è la somma, ma la media. In questo caso si può omettere l'opzione `aggfunc)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>Zona_int</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Anno</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>2001</th>\n",
       "      <td>50.416667</td>\n",
       "      <td>183.916667</td>\n",
       "      <td>97.583333</td>\n",
       "      <td>181.333333</td>\n",
       "      <td>148.500000</td>\n",
       "      <td>141.083333</td>\n",
       "      <td>117.416667</td>\n",
       "      <td>152.333333</td>\n",
       "      <td>183.333333</td>\n",
       "      <td>198.916667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2002</th>\n",
       "      <td>57.083333</td>\n",
       "      <td>181.416667</td>\n",
       "      <td>106.000000</td>\n",
       "      <td>182.083333</td>\n",
       "      <td>142.750000</td>\n",
       "      <td>134.083333</td>\n",
       "      <td>122.916667</td>\n",
       "      <td>147.833333</td>\n",
       "      <td>178.916667</td>\n",
       "      <td>194.166667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>55.916667</td>\n",
       "      <td>171.333333</td>\n",
       "      <td>96.666667</td>\n",
       "      <td>171.750000</td>\n",
       "      <td>135.250000</td>\n",
       "      <td>135.666667</td>\n",
       "      <td>114.250000</td>\n",
       "      <td>155.666667</td>\n",
       "      <td>169.500000</td>\n",
       "      <td>187.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004</th>\n",
       "      <td>56.833333</td>\n",
       "      <td>163.166667</td>\n",
       "      <td>91.750000</td>\n",
       "      <td>179.583333</td>\n",
       "      <td>131.583333</td>\n",
       "      <td>135.250000</td>\n",
       "      <td>107.250000</td>\n",
       "      <td>137.500000</td>\n",
       "      <td>170.916667</td>\n",
       "      <td>185.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>59.666667</td>\n",
       "      <td>155.583333</td>\n",
       "      <td>85.250000</td>\n",
       "      <td>156.666667</td>\n",
       "      <td>123.250000</td>\n",
       "      <td>126.000000</td>\n",
       "      <td>102.666667</td>\n",
       "      <td>129.833333</td>\n",
       "      <td>158.166667</td>\n",
       "      <td>168.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>56.250000</td>\n",
       "      <td>150.750000</td>\n",
       "      <td>84.250000</td>\n",
       "      <td>151.750000</td>\n",
       "      <td>127.250000</td>\n",
       "      <td>116.416667</td>\n",
       "      <td>100.333333</td>\n",
       "      <td>124.666667</td>\n",
       "      <td>166.333333</td>\n",
       "      <td>168.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>47.833333</td>\n",
       "      <td>143.583333</td>\n",
       "      <td>81.166667</td>\n",
       "      <td>144.583333</td>\n",
       "      <td>115.250000</td>\n",
       "      <td>119.333333</td>\n",
       "      <td>108.500000</td>\n",
       "      <td>124.583333</td>\n",
       "      <td>150.166667</td>\n",
       "      <td>163.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>41.500000</td>\n",
       "      <td>125.166667</td>\n",
       "      <td>76.333333</td>\n",
       "      <td>142.916667</td>\n",
       "      <td>110.166667</td>\n",
       "      <td>109.083333</td>\n",
       "      <td>94.333333</td>\n",
       "      <td>118.750000</td>\n",
       "      <td>145.583333</td>\n",
       "      <td>146.916667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009</th>\n",
       "      <td>41.166667</td>\n",
       "      <td>129.333333</td>\n",
       "      <td>70.333333</td>\n",
       "      <td>125.583333</td>\n",
       "      <td>105.500000</td>\n",
       "      <td>102.666667</td>\n",
       "      <td>93.333333</td>\n",
       "      <td>109.000000</td>\n",
       "      <td>130.000000</td>\n",
       "      <td>137.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010</th>\n",
       "      <td>44.583333</td>\n",
       "      <td>120.750000</td>\n",
       "      <td>63.916667</td>\n",
       "      <td>122.333333</td>\n",
       "      <td>101.416667</td>\n",
       "      <td>99.666667</td>\n",
       "      <td>83.500000</td>\n",
       "      <td>97.500000</td>\n",
       "      <td>124.416667</td>\n",
       "      <td>134.583333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>188.416667</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>58.500000</td>\n",
       "      <td>106.083333</td>\n",
       "      <td>75.500000</td>\n",
       "      <td>75.250000</td>\n",
       "      <td>73.250000</td>\n",
       "      <td>84.666667</td>\n",
       "      <td>91.916667</td>\n",
       "      <td>110.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012</th>\n",
       "      <td>37.666667</td>\n",
       "      <td>105.333333</td>\n",
       "      <td>64.833333</td>\n",
       "      <td>109.250000</td>\n",
       "      <td>84.333333</td>\n",
       "      <td>87.916667</td>\n",
       "      <td>77.166667</td>\n",
       "      <td>89.750000</td>\n",
       "      <td>107.833333</td>\n",
       "      <td>113.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>44.250000</td>\n",
       "      <td>94.750000</td>\n",
       "      <td>48.666667</td>\n",
       "      <td>95.500000</td>\n",
       "      <td>77.666667</td>\n",
       "      <td>78.416667</td>\n",
       "      <td>76.833333</td>\n",
       "      <td>83.333333</td>\n",
       "      <td>100.416667</td>\n",
       "      <td>103.083333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014</th>\n",
       "      <td>56.083333</td>\n",
       "      <td>86.583333</td>\n",
       "      <td>47.250000</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>69.416667</td>\n",
       "      <td>69.916667</td>\n",
       "      <td>64.833333</td>\n",
       "      <td>78.666667</td>\n",
       "      <td>90.833333</td>\n",
       "      <td>89.083333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015</th>\n",
       "      <td>78.166667</td>\n",
       "      <td>80.750000</td>\n",
       "      <td>43.500000</td>\n",
       "      <td>83.333333</td>\n",
       "      <td>65.500000</td>\n",
       "      <td>62.833333</td>\n",
       "      <td>63.666667</td>\n",
       "      <td>70.416667</td>\n",
       "      <td>84.666667</td>\n",
       "      <td>86.083333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016</th>\n",
       "      <td>40.000000</td>\n",
       "      <td>84.083333</td>\n",
       "      <td>47.166667</td>\n",
       "      <td>85.750000</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>73.083333</td>\n",
       "      <td>60.083333</td>\n",
       "      <td>81.500000</td>\n",
       "      <td>88.333333</td>\n",
       "      <td>96.833333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Zona_int           0           1           2           3           4  \\\n",
       "Anno                                                                   \n",
       "2001       50.416667  183.916667   97.583333  181.333333  148.500000   \n",
       "2002       57.083333  181.416667  106.000000  182.083333  142.750000   \n",
       "2003       55.916667  171.333333   96.666667  171.750000  135.250000   \n",
       "2004       56.833333  163.166667   91.750000  179.583333  131.583333   \n",
       "2005       59.666667  155.583333   85.250000  156.666667  123.250000   \n",
       "2006       56.250000  150.750000   84.250000  151.750000  127.250000   \n",
       "2007       47.833333  143.583333   81.166667  144.583333  115.250000   \n",
       "2008       41.500000  125.166667   76.333333  142.916667  110.166667   \n",
       "2009       41.166667  129.333333   70.333333  125.583333  105.500000   \n",
       "2010       44.583333  120.750000   63.916667  122.333333  101.416667   \n",
       "2011      188.416667   94.000000   58.500000  106.083333   75.500000   \n",
       "2012       37.666667  105.333333   64.833333  109.250000   84.333333   \n",
       "2013       44.250000   94.750000   48.666667   95.500000   77.666667   \n",
       "2014       56.083333   86.583333   47.250000   84.000000   69.416667   \n",
       "2015       78.166667   80.750000   43.500000   83.333333   65.500000   \n",
       "2016       40.000000   84.083333   47.166667   85.750000   74.000000   \n",
       "\n",
       "Zona_int           5           6           7           8           9  \n",
       "Anno                                                                  \n",
       "2001      141.083333  117.416667  152.333333  183.333333  198.916667  \n",
       "2002      134.083333  122.916667  147.833333  178.916667  194.166667  \n",
       "2003      135.666667  114.250000  155.666667  169.500000  187.250000  \n",
       "2004      135.250000  107.250000  137.500000  170.916667  185.833333  \n",
       "2005      126.000000  102.666667  129.833333  158.166667  168.000000  \n",
       "2006      116.416667  100.333333  124.666667  166.333333  168.000000  \n",
       "2007      119.333333  108.500000  124.583333  150.166667  163.000000  \n",
       "2008      109.083333   94.333333  118.750000  145.583333  146.916667  \n",
       "2009      102.666667   93.333333  109.000000  130.000000  137.250000  \n",
       "2010       99.666667   83.500000   97.500000  124.416667  134.583333  \n",
       "2011       75.250000   73.250000   84.666667   91.916667  110.666667  \n",
       "2012       87.916667   77.166667   89.750000  107.833333  113.000000  \n",
       "2013       78.416667   76.833333   83.333333  100.416667  103.083333  \n",
       "2014       69.916667   64.833333   78.666667   90.833333   89.083333  \n",
       "2015       62.833333   63.666667   70.416667   84.666667   86.083333  \n",
       "2016       73.083333   60.083333   81.500000   88.333333   96.833333  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.pivot_table(incidenti, index = 'Anno', columns = ['Zona_int'], values = 'Incidenti')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I dati presenti in una tabella pivot possono essere calcolati con una `groupby`, ma la struttura del DataFrame risultante è diversa. Ogni volta bisogna capire se preferiamo la forma ottenuta con la `groupby` (poche colonne, tante righe) o la tabella pivot (meno righe, più colonne).\n",
    "\n",
    "Di solito la tabella pivot è più semplice da leggere per una persona, ma più complicata da rielaborare."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Elementi estremi\n",
    "\n",
    "Di particolare interesse è capire chi ha realizzato un valore estremo (per valore estremo si intende un minimo o un massimo). Ad esempio vogliamo sapere in quale anno si sono verificati più incidenti.\n",
    "\n",
    "Il primo passo è calcolare il numero di incidenti per ogni anno."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>Incidenti</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Morti</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Anno</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>17458</td>\n",
       "      <td>23998</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2002</th>\n",
       "      <td>17367</td>\n",
       "      <td>23843</td>\n",
       "      <td>76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>16719</td>\n",
       "      <td>22422</td>\n",
       "      <td>72</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Incidenti  Feriti  Morti\n",
       "Anno                          \n",
       "2001      17458   23998     86\n",
       "2002      17367   23843     76\n",
       "2003      16719   22422     72"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_anno = incidenti.groupby('Anno').sum()[['Incidenti', 'Feriti', 'Morti']]\n",
    "incidenti_anno.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Adesso possiamo calcolare il valore massimo della colonna `Incidenti` e selezionare le righe che hanno tale valore."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17458"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "massimo = incidenti_anno['Incidenti'].max()\n",
    "massimo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "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>Incidenti</th>\n",
       "      <th>Feriti</th>\n",
       "      <th>Morti</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Anno</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>17458</td>\n",
       "      <td>23998</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Incidenti  Feriti  Morti\n",
       "Anno                          \n",
       "2001      17458   23998     86"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_anno[incidenti_anno['Incidenti'] == massimo]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Elementi estremi 2\n",
    "\n",
    "Un'alternativa più rapida è fornita dai metodi `idxmax` e `idxmin` che restituiscono il valore dell'*indice* che realizza il valore estremo. Questo valore può essere utilizzato per estrarre la riga che ci interessa."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2001"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_anno['Incidenti'].idxmax()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Incidenti    17458\n",
       "Feriti       23998\n",
       "Morti           86\n",
       "Name: 2001, dtype: int64"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_anno.loc[incidenti_anno['Incidenti'].idxmax()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Confronto**\n",
    "\n",
    "*  La soluzione con `idxmax` è più veloce da scrivere e permette di individuare facilmente l'anno che ci interessa. Ma bisogna ricordare che `idxmax` permette di calcolare un solo valore (in questo caso, un solo anno).\n",
    "*  La soluzione con la selezione è più complicata. Inoltre per calcolare l'anno che ci interessa, dovremmo fare una nuova estrazione. Il vantaggio è che è corretta anche nel caso in cui due anni diversi realizzano entrambi il massimo valore."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Idxmax e raggruppamenti\n",
    "\n",
    "Talvolta ci interessa capire chi ha realizzato un massimo all'interno di ogni gruppo. In altre parole, vogliamo isolare non una sola riga, ma una riga per ogni gruppo.\n",
    "\n",
    "Ad esempio, per ogni anno vogliamo calcolare in quale mese si sono verificati più incidenti. Il primo passo è costruire un DataFrame che dica per ogni coppia (`Anno`, `Mese`) quanti incidenti si sono verificati."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "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>Anno</th>\n",
       "      <th>Mese</th>\n",
       "      <th>Incidenti</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2001</td>\n",
       "      <td>1</td>\n",
       "      <td>1233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2001</td>\n",
       "      <td>2</td>\n",
       "      <td>1188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2001</td>\n",
       "      <td>3</td>\n",
       "      <td>1419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2001</td>\n",
       "      <td>4</td>\n",
       "      <td>1369</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2001</td>\n",
       "      <td>5</td>\n",
       "      <td>1697</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2001</td>\n",
       "      <td>6</td>\n",
       "      <td>1842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2001</td>\n",
       "      <td>7</td>\n",
       "      <td>1657</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2001</td>\n",
       "      <td>8</td>\n",
       "      <td>848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2001</td>\n",
       "      <td>9</td>\n",
       "      <td>1491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2001</td>\n",
       "      <td>10</td>\n",
       "      <td>1675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2001</td>\n",
       "      <td>11</td>\n",
       "      <td>1583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2001</td>\n",
       "      <td>12</td>\n",
       "      <td>1456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2002</td>\n",
       "      <td>1</td>\n",
       "      <td>1236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2002</td>\n",
       "      <td>2</td>\n",
       "      <td>1192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2002</td>\n",
       "      <td>3</td>\n",
       "      <td>1435</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>2002</td>\n",
       "      <td>4</td>\n",
       "      <td>1439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2002</td>\n",
       "      <td>5</td>\n",
       "      <td>1626</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2002</td>\n",
       "      <td>6</td>\n",
       "      <td>1519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>2002</td>\n",
       "      <td>7</td>\n",
       "      <td>1691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2002</td>\n",
       "      <td>8</td>\n",
       "      <td>792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2002</td>\n",
       "      <td>9</td>\n",
       "      <td>1536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2002</td>\n",
       "      <td>10</td>\n",
       "      <td>1778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2002</td>\n",
       "      <td>11</td>\n",
       "      <td>1700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2002</td>\n",
       "      <td>12</td>\n",
       "      <td>1423</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>2003</td>\n",
       "      <td>1</td>\n",
       "      <td>1236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>2003</td>\n",
       "      <td>2</td>\n",
       "      <td>1146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2003</td>\n",
       "      <td>3</td>\n",
       "      <td>1479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>2003</td>\n",
       "      <td>4</td>\n",
       "      <td>1407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>2003</td>\n",
       "      <td>5</td>\n",
       "      <td>1749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>2003</td>\n",
       "      <td>6</td>\n",
       "      <td>1627</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>162</th>\n",
       "      <td>2014</td>\n",
       "      <td>7</td>\n",
       "      <td>769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>163</th>\n",
       "      <td>2014</td>\n",
       "      <td>8</td>\n",
       "      <td>376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164</th>\n",
       "      <td>2014</td>\n",
       "      <td>9</td>\n",
       "      <td>785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>165</th>\n",
       "      <td>2014</td>\n",
       "      <td>10</td>\n",
       "      <td>875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>166</th>\n",
       "      <td>2014</td>\n",
       "      <td>11</td>\n",
       "      <td>730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>167</th>\n",
       "      <td>2014</td>\n",
       "      <td>12</td>\n",
       "      <td>679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>2015</td>\n",
       "      <td>1</td>\n",
       "      <td>652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>2015</td>\n",
       "      <td>2</td>\n",
       "      <td>631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>170</th>\n",
       "      <td>2015</td>\n",
       "      <td>3</td>\n",
       "      <td>724</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>171</th>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>763</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>172</th>\n",
       "      <td>2015</td>\n",
       "      <td>5</td>\n",
       "      <td>826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>173</th>\n",
       "      <td>2015</td>\n",
       "      <td>6</td>\n",
       "      <td>792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>174</th>\n",
       "      <td>2015</td>\n",
       "      <td>7</td>\n",
       "      <td>808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>2015</td>\n",
       "      <td>8</td>\n",
       "      <td>385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>176</th>\n",
       "      <td>2015</td>\n",
       "      <td>9</td>\n",
       "      <td>764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>177</th>\n",
       "      <td>2015</td>\n",
       "      <td>10</td>\n",
       "      <td>881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>2015</td>\n",
       "      <td>11</td>\n",
       "      <td>776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>2015</td>\n",
       "      <td>12</td>\n",
       "      <td>625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>2016</td>\n",
       "      <td>1</td>\n",
       "      <td>637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>2016</td>\n",
       "      <td>2</td>\n",
       "      <td>706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>2016</td>\n",
       "      <td>3</td>\n",
       "      <td>728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>2016</td>\n",
       "      <td>4</td>\n",
       "      <td>821</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>184</th>\n",
       "      <td>2016</td>\n",
       "      <td>5</td>\n",
       "      <td>866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185</th>\n",
       "      <td>2016</td>\n",
       "      <td>6</td>\n",
       "      <td>767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>186</th>\n",
       "      <td>2016</td>\n",
       "      <td>7</td>\n",
       "      <td>770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187</th>\n",
       "      <td>2016</td>\n",
       "      <td>8</td>\n",
       "      <td>429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>188</th>\n",
       "      <td>2016</td>\n",
       "      <td>9</td>\n",
       "      <td>774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>189</th>\n",
       "      <td>2016</td>\n",
       "      <td>10</td>\n",
       "      <td>873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190</th>\n",
       "      <td>2016</td>\n",
       "      <td>11</td>\n",
       "      <td>819</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>191</th>\n",
       "      <td>2016</td>\n",
       "      <td>12</td>\n",
       "      <td>580</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>192 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Anno  Mese  Incidenti\n",
       "0    2001     1       1233\n",
       "1    2001     2       1188\n",
       "2    2001     3       1419\n",
       "3    2001     4       1369\n",
       "4    2001     5       1697\n",
       "5    2001     6       1842\n",
       "6    2001     7       1657\n",
       "7    2001     8        848\n",
       "8    2001     9       1491\n",
       "9    2001    10       1675\n",
       "10   2001    11       1583\n",
       "11   2001    12       1456\n",
       "12   2002     1       1236\n",
       "13   2002     2       1192\n",
       "14   2002     3       1435\n",
       "15   2002     4       1439\n",
       "16   2002     5       1626\n",
       "17   2002     6       1519\n",
       "18   2002     7       1691\n",
       "19   2002     8        792\n",
       "20   2002     9       1536\n",
       "21   2002    10       1778\n",
       "22   2002    11       1700\n",
       "23   2002    12       1423\n",
       "24   2003     1       1236\n",
       "25   2003     2       1146\n",
       "26   2003     3       1479\n",
       "27   2003     4       1407\n",
       "28   2003     5       1749\n",
       "29   2003     6       1627\n",
       "..    ...   ...        ...\n",
       "162  2014     7        769\n",
       "163  2014     8        376\n",
       "164  2014     9        785\n",
       "165  2014    10        875\n",
       "166  2014    11        730\n",
       "167  2014    12        679\n",
       "168  2015     1        652\n",
       "169  2015     2        631\n",
       "170  2015     3        724\n",
       "171  2015     4        763\n",
       "172  2015     5        826\n",
       "173  2015     6        792\n",
       "174  2015     7        808\n",
       "175  2015     8        385\n",
       "176  2015     9        764\n",
       "177  2015    10        881\n",
       "178  2015    11        776\n",
       "179  2015    12        625\n",
       "180  2016     1        637\n",
       "181  2016     2        706\n",
       "182  2016     3        728\n",
       "183  2016     4        821\n",
       "184  2016     5        866\n",
       "185  2016     6        767\n",
       "186  2016     7        770\n",
       "187  2016     8        429\n",
       "188  2016     9        774\n",
       "189  2016    10        873\n",
       "190  2016    11        819\n",
       "191  2016    12        580\n",
       "\n",
       "[192 rows x 3 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_mese = incidenti.groupby(['Anno', 'Mese'], as_index = False).sum()[['Anno', 'Mese', 'Incidenti']]\n",
    "incidenti_mese"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Applicando `idxmax` otteniamo il singolo mese in cui si sono verificati più incidenti, ma noi vogliamo un mese per ogni anno (quindi un mese per il 2001, un mese per il 2002, ecc)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Anno         2001\n",
       "Mese            6\n",
       "Incidenti    1842\n",
       "Name: 5, dtype: int64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_mese.loc[incidenti_mese['Incidenti'].idxmax()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Quindi sappiamo che il picco del numero di incidenti si è verificato nel mese di Giugno del 2001. Ma ciò non è quello che desideriamo."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Idxmax e raggruppamento 2\n",
    "\n",
    "Raggruppare per anno le righe di `incidenti_mese` è il primo passo: bisogna applicare la funzione `idxmax`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Anno\n",
       "2001      5\n",
       "2002     21\n",
       "2003     28\n",
       "2004     45\n",
       "2005     52\n",
       "2006     69\n",
       "2007     76\n",
       "2008     93\n",
       "2009    105\n",
       "2010    112\n",
       "2011    130\n",
       "2012    141\n",
       "2013    153\n",
       "2014    160\n",
       "2015    177\n",
       "2016    189\n",
       "Name: Incidenti, dtype: int64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_mese.groupby('Anno')['Incidenti'].idxmax()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Siccome `idxmax` restituisce i valori dell'indice corrispondenti alle righe che realizzano un massimo, bisogna passare per una `loc` per ottenere i mesi corrispondenti."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "scrolled": false
   },
   "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>Anno</th>\n",
       "      <th>Mese</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2001</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2002</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>2003</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>2004</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>2005</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69</th>\n",
       "      <td>2006</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>2007</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>2008</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105</th>\n",
       "      <td>2009</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>112</th>\n",
       "      <td>2010</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>130</th>\n",
       "      <td>2011</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141</th>\n",
       "      <td>2012</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>2013</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160</th>\n",
       "      <td>2014</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>177</th>\n",
       "      <td>2015</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>189</th>\n",
       "      <td>2016</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Anno  Mese\n",
       "5    2001     6\n",
       "21   2002    10\n",
       "28   2003     5\n",
       "45   2004    10\n",
       "52   2005     5\n",
       "69   2006    10\n",
       "76   2007     5\n",
       "93   2008    10\n",
       "105  2009    10\n",
       "112  2010     5\n",
       "130  2011    11\n",
       "141  2012    10\n",
       "153  2013    10\n",
       "160  2014     5\n",
       "177  2015    10\n",
       "189  2016    10"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "incidenti_mese.loc[incidenti_mese.groupby('Anno')['Incidenti'].idxmax(), ['Anno', 'Mese']]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}