pandas_intro.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploración de un conjunto de datos con Python y Pandas\n",
    "## Configuración del ambiente\n",
    "### Prerequisitos\n",
    "### Instalación de Pandas  (si no lo han hecho aún) \n",
    "\n",
    "### Para instalarlo en Thonny hágalo desde la herramienta de gestión de paquetes:  \n",
    "#### Tools --> Manage Packages\n",
    "\n",
    "### Si está trabajando directamente con Python u otro ambiente, desde el Shell\n",
    "\n",
    "#### python -m pip install pandas\n",
    "\n",
    "### Si está trabajando con Anaconda o Conda utilice esta instrucción desde el Shell\n",
    "\n",
    "##### conda install pandas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Hola Pandas\n",
    "#### Lo primero es Importar los módulos a usar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np    # Pandas está basado en NumPy (Numerical Python) por lo que hay que importarlo también\n",
    "import pandas as pd\n",
    "import requests       #  Módulo con funciones para obtener archivos desde la Internet\n",
    "\n",
    "# Pandas tiene su módulo para hacer gráficas basado en otro módulo\n",
    "#    llamado MatPlotLib, se usa esta instrucción para que las gráficas salgan en Notebook\n",
    "%matplotlib inline    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### En la siguiente sección se obtiene una base de datos de la NBA desde la Internet y la guardan en su computador. \n",
    "#### Primero, le damos la dirección de los datos en Internet, y la dirección en nuestro computador donde queremos guardar los datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "url_datos_fuente = \"https://raw.githubusercontent.com/fivethirtyeight/data/master/nba-elo/nbaallelo.csv\"\n",
    "path_en_mi_compu = \"nba_all_elo.csv\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Se procede a descargar y guardar en nuestro computador"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "respuesta = requests.get(url_datos_fuente)\n",
    "respuesta.raise_for_status()\n",
    "with open(path_en_mi_compu, \"wb\") as f:\n",
    "    f.write(respuesta.content)\n",
    "print(\"Descarga, lista!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Una vez en nuestro computador, leemos el archivo desde allí"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba = pd.read_csv(\"nba_all_elo.csv\")\n",
    "\n",
    "type(nba)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Veamos el formato general de los datos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Obtener la longitud de la base de datos (número de registros o entradas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "len(nba)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Obtener el formato de la base de datos (longitud y número de columnas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Desplegar los primeros registros (default = 5 registros)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Si el número de columnas excede el especificado, se trunca al número dado.  None = no truncar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option(\"display.max.columns\", None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Especificar el número de cifras decimales dado, si el número es float (default = 6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option(\"display.precision\", 2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Desplegar los últimos registros (default = 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### Puede mostrar las últimas 3 líneas del `DataFrame`?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "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>gameorder</th>\n",
       "      <th>game_id</th>\n",
       "      <th>lg_id</th>\n",
       "      <th>_iscopy</th>\n",
       "      <th>year_id</th>\n",
       "      <th>date_game</th>\n",
       "      <th>seasongame</th>\n",
       "      <th>is_playoffs</th>\n",
       "      <th>team_id</th>\n",
       "      <th>fran_id</th>\n",
       "      <th>pts</th>\n",
       "      <th>elo_i</th>\n",
       "      <th>elo_n</th>\n",
       "      <th>win_equiv</th>\n",
       "      <th>opp_id</th>\n",
       "      <th>opp_fran</th>\n",
       "      <th>opp_pts</th>\n",
       "      <th>opp_elo_i</th>\n",
       "      <th>opp_elo_n</th>\n",
       "      <th>game_location</th>\n",
       "      <th>game_result</th>\n",
       "      <th>forecast</th>\n",
       "      <th>notes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>126311</th>\n",
       "      <td>63156</td>\n",
       "      <td>201506140GSW</td>\n",
       "      <td>NBA</td>\n",
       "      <td>1</td>\n",
       "      <td>2015</td>\n",
       "      <td>6/14/2015</td>\n",
       "      <td>101</td>\n",
       "      <td>1</td>\n",
       "      <td>CLE</td>\n",
       "      <td>Cavaliers</td>\n",
       "      <td>91</td>\n",
       "      <td>1704.39</td>\n",
       "      <td>1700.74</td>\n",
       "      <td>60.01</td>\n",
       "      <td>GSW</td>\n",
       "      <td>Warriors</td>\n",
       "      <td>104</td>\n",
       "      <td>1809.98</td>\n",
       "      <td>1813.63</td>\n",
       "      <td>A</td>\n",
       "      <td>L</td>\n",
       "      <td>0.23</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126312</th>\n",
       "      <td>63157</td>\n",
       "      <td>201506170CLE</td>\n",
       "      <td>NBA</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>6/16/2015</td>\n",
       "      <td>102</td>\n",
       "      <td>1</td>\n",
       "      <td>CLE</td>\n",
       "      <td>Cavaliers</td>\n",
       "      <td>97</td>\n",
       "      <td>1700.74</td>\n",
       "      <td>1692.09</td>\n",
       "      <td>59.29</td>\n",
       "      <td>GSW</td>\n",
       "      <td>Warriors</td>\n",
       "      <td>105</td>\n",
       "      <td>1813.63</td>\n",
       "      <td>1822.29</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>0.48</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>126313</th>\n",
       "      <td>63157</td>\n",
       "      <td>201506170CLE</td>\n",
       "      <td>NBA</td>\n",
       "      <td>1</td>\n",
       "      <td>2015</td>\n",
       "      <td>6/16/2015</td>\n",
       "      <td>103</td>\n",
       "      <td>1</td>\n",
       "      <td>GSW</td>\n",
       "      <td>Warriors</td>\n",
       "      <td>105</td>\n",
       "      <td>1813.63</td>\n",
       "      <td>1822.29</td>\n",
       "      <td>68.52</td>\n",
       "      <td>CLE</td>\n",
       "      <td>Cavaliers</td>\n",
       "      <td>97</td>\n",
       "      <td>1700.74</td>\n",
       "      <td>1692.09</td>\n",
       "      <td>A</td>\n",
       "      <td>W</td>\n",
       "      <td>0.52</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        gameorder       game_id lg_id  _iscopy  year_id  date_game  \\\n",
       "126311      63156  201506140GSW   NBA        1     2015  6/14/2015   \n",
       "126312      63157  201506170CLE   NBA        0     2015  6/16/2015   \n",
       "126313      63157  201506170CLE   NBA        1     2015  6/16/2015   \n",
       "\n",
       "        seasongame  is_playoffs team_id    fran_id  pts    elo_i    elo_n  \\\n",
       "126311         101            1     CLE  Cavaliers   91  1704.39  1700.74   \n",
       "126312         102            1     CLE  Cavaliers   97  1700.74  1692.09   \n",
       "126313         103            1     GSW   Warriors  105  1813.63  1822.29   \n",
       "\n",
       "        win_equiv opp_id   opp_fran  opp_pts  opp_elo_i  opp_elo_n  \\\n",
       "126311      60.01    GSW   Warriors      104    1809.98    1813.63   \n",
       "126312      59.29    GSW   Warriors      105    1813.63    1822.29   \n",
       "126313      68.52    CLE  Cavaliers       97    1700.74    1692.09   \n",
       "\n",
       "       game_location game_result  forecast notes  \n",
       "126311             A           L      0.23   NaN  \n",
       "126312             H           L      0.48   NaN  \n",
       "126313             A           W      0.52   NaN  "
      ]
     },
     "execution_count": 157,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Explorar un poco los datos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Desplegar los tipos de datos con `.info()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "nba.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mostrar estadísticas básicas con `.describe()`\n",
    "#### Desplegar información sobre columnas con valores numéricos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.describe()   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Desplegar información sobre columnas con valores no numéricos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.describe(include=np.object)   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Una pequeña exploración de datos con Pandas\n",
    "#####  Ver cuántos registros hay de cada equipo.  Pandas permite usar el nombre de las columnas como índice"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba[\"team_id\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Ver cuántos registros hay de cada franquicia."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba[\"fran_id\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Hay algo raro acá.  Los datos de la franquicia Lakers no coinciden con los del equipo LAL.  Busquemos la Id de los equipos para la franquicia Lakers.  (Más abajo hablaremos sobre el .loc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.loc[nba[\"fran_id\"] == \"Lakers\", \"team_id\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Ajá!!!   Apareció otro equipo bajo la franquicia Lakers:   MNL (Minnesota Lakers)\n",
    "####  Veamos cuál es la primera fecha en que jugó este equipo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.loc[nba[\"team_id\"] == \"MNL\", \"date_game\"].min()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Ahora veamos cuál es la última fecha en que jugó este equipo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.loc[nba[\"team_id\"] == \"MNL\", \"date_game\"].max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Otra forma de encontrar estas fechas, con una sola instrucción"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.loc[nba[\"team_id\"] == \"MNL\", \"date_game\"].agg((\"min\", \"max\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### Encontrar cuantos puntos han anotado los Boston Celtics a lo largo de todos los partidos que están en este conjunto de datos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "626484"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conociendo las estructuras de datos de Pandas\n",
    "#### Para los siguientes ejercicios construiremos unos datos ficticios\n",
    "### `Series`:    El elemento básico de construcción en Pandas\n",
    "####  Creamos una serie con los siguientes datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos = pd.Series([5555, 7000, 1980])\n",
    "\n",
    "ingresos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Nótese que aparecen dos columnas.  La primera columna es el índice implícito de cada valor, va de 0 a n-1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "type(ingresos.values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Nótese que el tipo de dato aparece como un arreglo de NumPy.  Los arreglos o matrices son el plato fuerte de NumPy\n",
    "##### Ahora creamos otra serie pero esta vez el índice se lo damos nosotros"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades = pd.Series(\n",
    "    [4200, 8000, 6500],\n",
    "    index=[\"Amsterdam\", \"Toronto\", \"Tokyo\"]\n",
    ")\n",
    "\n",
    "ingresos_ciudades\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### `Series de Pandas` vs estructuras que vienen incluídas en Python\n",
    "#### También podemos crear una serie usando Diccionarios"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "conteo_empleados_ciudades = pd.Series({\"Amsterdam\": 5, \"Tokyo\": 8})\n",
    "\n",
    "conteo_empleados_ciudades"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Podemos usar el método keys() con la serie"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "conteo_empleados_ciudades.keys()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Hasta podemos usar operadores que hemos visto con Listas, Diccionarios, etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"Tokyo\" in conteo_empleados_ciudades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"Nueva York\" in conteo_empleados_ciudades"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### `DataFrame`:  La estructura de datos más popular de Pandas\n",
    "####  Podemos crear un Dataframe a partir de dos o más series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades = pd.DataFrame({\n",
    "    \"ingresos\": ingresos_ciudades,\n",
    "    \"conteo_empleados\": conteo_empleados_ciudades\n",
    "})\n",
    "\n",
    "datos_ciudades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Se usa el método axes (ejes) para referirse a ambos filas y columnas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.axes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  axes[0] se refiere a las filas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.axes[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  axes[1] se refiere a las colummnas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.axes[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"Amsterdam\" in datos_ciudades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"ingresos\" in datos_ciudades"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### Apliquemos lo último que hemos aprendido, a nuestros datos de la NBA\n",
    "### Despliegue el índice y los ejes del conjunto de datos `nba`.  \n",
    "### Revise si tiene una columna \"points\". O... se llamará \"pts\"?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=126314, step=1)"
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=126314, step=1)"
      ]
     },
     "execution_count": 187,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['gameorder', 'game_id', 'lg_id', '_iscopy', 'year_id', 'date_game',\n",
       "       'seasongame', 'is_playoffs', 'team_id', 'fran_id', 'pts', 'elo_i',\n",
       "       'elo_n', 'win_equiv', 'opp_id', 'opp_fran', 'opp_pts', 'opp_elo_i',\n",
       "       'opp_elo_n', 'game_location', 'game_result', 'forecast', 'notes'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 188,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## El acceso a los elementos en `listas` de Python y en `Series` de Pandas: Simillitudes y Diferencias\n",
    "### Los beneficios y limitaciones del operador de indización de Python `[]`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Pandas permite el uso de los nombres de las columnas como índices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades[\"Toronto\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  También acepta que se utilicen los indices númericos como se hace con listas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades[-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades[1:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Y todas las reglas que vimos para rodajas \"slices\" de listas son válidas con los nombres de las columnas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades[\"Toronto\":]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Acceso a elementos de `Series` via `.loc` e `.iloc`\n",
    "\n",
    "#### Para los siguientes ejercicios, construimos un nuevo conjunto de datos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "colores = pd.Series(\n",
    "    [\"rojo\", \"morado\", \"azul\", \"verde\", \"amarillo\"],\n",
    "    index=[1, 2, 3, 5, 8]\n",
    ")\n",
    "\n",
    "colores"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  .loc se refiere a que se van a utilizar índices explicitos (dados por nosotros)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "colores.loc[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  .iloc se refiere a que se van a utilizar índices implícitos (dados por Pandas).   Recordemos que Python empieza la numeración con 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "colores.iloc[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### La siguiente imagen muestra el uso de estos indices\n",
    "<img src=\"iloc_vs_loc.png\">\n",
    "\n",
    "#### Desplegar los elementos con los índice implícitos 1 y 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "colores.iloc[1:3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Desplegar los elementos con los índices explícitos entre 3 y 8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "colores.loc[3:8]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "colores.iloc[-2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Accesando los elementos de un `DataFrame`\n",
    "### Accesando las columnas de un `DataFrame`\n",
    "#### Regresemos a nuestros datos sobre ciudades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades[\"ingresos\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "type(datos_ciudades[\"ingresos\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.ingresos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Se puede ver que ambos datos_ciudades[\"ingresos\"] y datos_ciudades.ingresos dan el mismo resultado\n",
    "\n",
    "#### Hay una instancia donde lo anterior no se cumple.  Cuando el nombre de una columna coincide con un atributo o método de los DataFrames, pueden ocurrir cosas inesperadas.  Para ilustrar esto, crearemos un DataFrame nuevo, esta vez con nombres de columnas en inglés:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "juguetes = pd.DataFrame([\n",
    "    {\"name\": \"pelota\", \"shape\": \"esfera\"},\n",
    "    {\"name\": \"Cubo de Rubik\", \"shape\": \"cubo\"}\n",
    "])\n",
    "\n",
    "juguetes[\"shape\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "juguetes.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Accesando a las filas de un `DataFrame` con `.loc` e `.iloc`\n",
    "\n",
    "#### Volvemos con nuestros datos de ciudades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.loc[\"Amsterdam\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.loc[\"Tokyo\": \"Toronto\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.iloc[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### De nuevo, practiquemos lo aprendido, con los datos de la NBA\n",
    "### Despliege la penúltima fila del conjunto de datos `nba`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gameorder               63157\n",
       "game_id          201506170CLE\n",
       "lg_id                     NBA\n",
       "_iscopy                     0\n",
       "year_id                  2015\n",
       "date_game           6/16/2015\n",
       "seasongame                102\n",
       "is_playoffs                 1\n",
       "team_id                   CLE\n",
       "fran_id             Cavaliers\n",
       "pts                        97\n",
       "elo_i                 1.7e+03\n",
       "elo_n                 1.7e+03\n",
       "win_equiv                  59\n",
       "opp_id                    GSW\n",
       "opp_fran             Warriors\n",
       "opp_pts                   105\n",
       "opp_elo_i             1.8e+03\n",
       "opp_elo_n             1.8e+03\n",
       "game_location               H\n",
       "game_result                 L\n",
       "forecast                 0.48\n",
       "notes                     NaN\n",
       "Name: 126312, dtype: object"
      ]
     },
     "execution_count": 209,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Acceder a un subconjunto de filas y columnas en un `DataFrame` con `.loc` e `.iloc`\n",
    "#### Para un DataFrame, los métodos de acceso .loc e .iloc también acceptan un segundo parámetro. El primer parámetro selecciona filas basado en los índices, el segundo parámetro selecciona columnas. Se pueden utilizar estos parámetros juntos para seleccionar un subconjunto de filas y columnas del DataFrame:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_ciudades.loc[\"Amsterdam\": \"Tokyo\", \"ingresos\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### Probemos esto con nuestros datos de la NBA\n",
    "### Déle un vistazo a los juegos con índices 5555 and 5559. Nos interesan unicamente los nombres de los equipos y los resultados."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "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>fran_id</th>\n",
       "      <th>opp_fran</th>\n",
       "      <th>pts</th>\n",
       "      <th>opp_pts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5555</th>\n",
       "      <td>Pistons</td>\n",
       "      <td>Warriors</td>\n",
       "      <td>83</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5556</th>\n",
       "      <td>Celtics</td>\n",
       "      <td>Knicks</td>\n",
       "      <td>95</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5557</th>\n",
       "      <td>Knicks</td>\n",
       "      <td>Celtics</td>\n",
       "      <td>74</td>\n",
       "      <td>95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5558</th>\n",
       "      <td>Kings</td>\n",
       "      <td>Sixers</td>\n",
       "      <td>81</td>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5559</th>\n",
       "      <td>Sixers</td>\n",
       "      <td>Kings</td>\n",
       "      <td>86</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      fran_id  opp_fran  pts  opp_pts\n",
       "5555  Pistons  Warriors   83       56\n",
       "5556  Celtics    Knicks   95       74\n",
       "5557   Knicks   Celtics   74       95\n",
       "5558    Kings    Sixers   81       86\n",
       "5559   Sixers     Kings   86       81"
      ]
     },
     "execution_count": 211,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Consultas (Queries) con Pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "decada_actual = nba[nba[\"year_id\"] > 2010]\n",
    "\n",
    "decada_actual.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "juegos_con_anotaciones = nba[nba[\"notes\"].notnull()]\n",
    "juegos_con_anotaciones.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "terminado_con_ers = nba[nba[\"fran_id\"].str.endswith(\"ers\")]\n",
    "terminado_con_ers.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Uso del query()\n",
    "\n",
    "#### A partir de este punto estaremos trabajando con un método muy fácil de utilizar, el __query()__\n",
    "\n",
    "#### Para los que estén familiarizados con SQL, tiene mucho en común con la instrucción __\"where\"__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.query('(_iscopy == 0) & (pts > 100) & (opp_pts > 100) & (team_id == \"BLB\")')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Sin usar el query(), la instrucción sería así:\n",
    "\n",
    "nba[\n",
    "    (nba[\"_iscopy\"] == 0) &\n",
    "    (nba[\"pts\"] > 100) &\n",
    "    (nba[\"opp_pts\"] > 100) &\n",
    "    (nba[\"team_id\"] == \"BLB\")\n",
    " ]\n",
    " #### Cuál parece más fácil?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### En la primavera de 1992, ambos equipos de Los Angeles tuvieron que jugar un juego de \"casa\" en una cancha ajena. Encuentre esos 2 juegos. \n",
    "#### Ayuda:\n",
    "#### Ambos equipos tienen un ID que empieza con \"LA\".\n",
    "#### Será útil utilizar la columna \"_iscopy\"\n",
    "#### Lo más probable es que la columna \"notes\" no este vacía"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "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>gameorder</th>\n",
       "      <th>game_id</th>\n",
       "      <th>lg_id</th>\n",
       "      <th>_iscopy</th>\n",
       "      <th>year_id</th>\n",
       "      <th>date_game</th>\n",
       "      <th>seasongame</th>\n",
       "      <th>is_playoffs</th>\n",
       "      <th>team_id</th>\n",
       "      <th>fran_id</th>\n",
       "      <th>pts</th>\n",
       "      <th>elo_i</th>\n",
       "      <th>elo_n</th>\n",
       "      <th>win_equiv</th>\n",
       "      <th>opp_id</th>\n",
       "      <th>opp_fran</th>\n",
       "      <th>opp_pts</th>\n",
       "      <th>opp_elo_i</th>\n",
       "      <th>opp_elo_n</th>\n",
       "      <th>game_location</th>\n",
       "      <th>game_result</th>\n",
       "      <th>forecast</th>\n",
       "      <th>notes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>68901</th>\n",
       "      <td>34451</td>\n",
       "      <td>199205030LAC</td>\n",
       "      <td>NBA</td>\n",
       "      <td>0</td>\n",
       "      <td>1992</td>\n",
       "      <td>5/3/1992</td>\n",
       "      <td>86</td>\n",
       "      <td>1</td>\n",
       "      <td>LAC</td>\n",
       "      <td>Clippers</td>\n",
       "      <td>115</td>\n",
       "      <td>1546.90</td>\n",
       "      <td>1556.0</td>\n",
       "      <td>47.89</td>\n",
       "      <td>UTA</td>\n",
       "      <td>Jazz</td>\n",
       "      <td>107</td>\n",
       "      <td>1647.51</td>\n",
       "      <td>1638.41</td>\n",
       "      <td>H</td>\n",
       "      <td>W</td>\n",
       "      <td>0.50</td>\n",
       "      <td>at Anaheim, CA (Anaheim Convention Center)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>68903</th>\n",
       "      <td>34452</td>\n",
       "      <td>199205030LAL</td>\n",
       "      <td>NBA</td>\n",
       "      <td>0</td>\n",
       "      <td>1992</td>\n",
       "      <td>5/3/1992</td>\n",
       "      <td>86</td>\n",
       "      <td>1</td>\n",
       "      <td>LAL</td>\n",
       "      <td>Lakers</td>\n",
       "      <td>76</td>\n",
       "      <td>1489.65</td>\n",
       "      <td>1474.4</td>\n",
       "      <td>39.40</td>\n",
       "      <td>POR</td>\n",
       "      <td>Trailblazers</td>\n",
       "      <td>102</td>\n",
       "      <td>1655.12</td>\n",
       "      <td>1670.37</td>\n",
       "      <td>H</td>\n",
       "      <td>L</td>\n",
       "      <td>0.41</td>\n",
       "      <td>at Las Vegas, NV</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       gameorder       game_id lg_id  _iscopy  year_id date_game  seasongame  \\\n",
       "68901      34451  199205030LAC   NBA        0     1992  5/3/1992          86   \n",
       "68903      34452  199205030LAL   NBA        0     1992  5/3/1992          86   \n",
       "\n",
       "       is_playoffs team_id   fran_id  pts    elo_i   elo_n  win_equiv opp_id  \\\n",
       "68901            1     LAC  Clippers  115  1546.90  1556.0      47.89    UTA   \n",
       "68903            1     LAL    Lakers   76  1489.65  1474.4      39.40    POR   \n",
       "\n",
       "           opp_fran  opp_pts  opp_elo_i  opp_elo_n game_location game_result  \\\n",
       "68901          Jazz      107    1647.51    1638.41             H           W   \n",
       "68903  Trailblazers      102    1655.12    1670.37             H           L   \n",
       "\n",
       "       forecast                                       notes  \n",
       "68901      0.50  at Anaheim, CA (Anaheim Convention Center)  \n",
       "68903      0.41                            at Las Vegas, NV  "
      ]
     },
     "execution_count": 216,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Agregaciones (Aggregations)\n",
    "### Agregaciones con `Series`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ingresos_ciudades.max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Agregaciones con un `DataFrame`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "puntos = nba[\"pts\"]\n",
    "\n",
    "type(puntos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "puntos.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Agrupamientos (Grouping)\n",
    "####  Agrupemos los datos por la franquicia y obtengamos el total de puntos de cada una"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.groupby(\"fran_id\", sort=False)[\"pts\"].sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  A continuación, note el uso de los paréntesis extremos, esto permite considerar todo el bloque como una sola instrucción.  Si no se usan, daría un error de indentación"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "(nba.query('(fran_id == \"Spurs\") & (year_id > 2010)')\n",
    "    .groupby([\"year_id\", \"game_result\"])[\"game_id\"]\n",
    "    .count()\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### Echémos un vistazo a la temporada 2014-15 (year_id: 2015) de los Golden State Warriors. Cuántos juegos ganados y perdidos tuvieron durante la temporada regular y los playoffs?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "is_playoffs  game_result\n",
       "0            L              15\n",
       "             W              67\n",
       "1            L               5\n",
       "             W              16\n",
       "Name: game_id, dtype: int64"
      ]
     },
     "execution_count": 223,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Manipulación de  Columnas\n",
    "\n",
    "#### Haremos una copia de nuestros datos del NBA para asegurar que los datos originales no se pierdan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = nba.copy()\n",
    "\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Agregar nuevas Columnas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"diferencia\"] = df.pts - df.opp_pts\n",
    "\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"diferencia\"].max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Renombrar Columnas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_renombrado = df.rename(\n",
    "    columns={\"game_result\": \"resultado\", \"game_location\": \"lugar_de_juego\"}\n",
    ")\n",
    "\n",
    "df_renombrado.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Eliminación de Columnas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "columnas_elo = [\"elo_i\", \"elo_n\", \"opp_elo_i\", \"opp_elo_n\"]\n",
    "\n",
    "df.drop(columnas_elo, inplace=True, axis=1)\n",
    "\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cambiar el tipo de datos de Columnas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Pandas tiene un tipo de dato especial para fechas, la base de datos tiene la fecha como cadena, la cambiaremos al nuevo tipo de dato"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"date_game\"] = pd.to_datetime(df[\"date_game\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Cuántos valores diferentes hay para el lugar donde se lleva a cabo el juego?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"game_location\"].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Cuáles son los diferentes valores?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"game_location\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Estos valores son strings.  Para poder manejarlos mejor, cambiémoslos a valores categóricos (discretos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"game_location\"] = pd.Categorical(df[\"game_location\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"game_location\"].dtype"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### Encuentre otra columna que tenga un tipo de datos genérico y cámbie el tipo a categórico"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 236,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ\n",
    "#  La variable que buscamos solo puede tener dos valores, cómo podemos saber cuántos valores\n",
    "#  únicos tiene esa variable?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "L    63157\n",
       "W    63157\n",
       "Name: game_result, dtype: int64"
      ]
     },
     "execution_count": 237,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 239,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ\n",
    "# Cambiar el tipo genérico a uno categórico"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CategoricalDtype(categories=['L', 'W'], ordered=False)"
      ]
     },
     "execution_count": 241,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ\n",
    "# Cambiar el tipo genérico a uno categórico"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Limpiando los Datos\n",
    "####  Quizás la tarea más difícil en el análisis de datos es la limpieza o depuración de los mismos\n",
    "### Valores faltantes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Eliminemos o botemos \"drop\" las filas que tengan datos faltantes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "filas_sin_datos_faltantes = nba.dropna()\n",
    "\n",
    "filas_sin_datos_faltantes.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Note que no especificamos axes para indicar que queremos eliminar filas, esto es porque el default es axes = 0\n",
    "####  Hagamos lo mismo con las columnas que tengan datos faltantes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_sin_columnas_faltantes = nba.dropna(axis=1)\n",
    "\n",
    "datos_sin_columnas_faltantes.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Podemos llenar, o completar, datos faltantes con algún valor deseado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datos_con_anotaciones_default = nba.copy()   #  Hacemos una copia para no perder el original\n",
    "\n",
    "datos_con_anotaciones_default[\"notes\"].fillna(\n",
    "    value=\"no hay anotaciones\",\n",
    "    inplace=True\n",
    ")\n",
    "\n",
    "datos_con_anotaciones_default[\"notes\"].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Valores inválidos\n",
    "#### Es buena práctica es revisar si hay valores que no son válidos, o no tienen sentido\n",
    "####  No parecería lógico que hubiera un partido de la NBA con 0 puntos, pero chequeémos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.query('pts == 0')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Wow!!!  Hay un partido donde el resultado final fue 2 a 0, debe ser un error, habrá que verificarlo!  No lo haremos pero sería interesante ver qué fue lo que pasó\n",
    "\n",
    "### Inconsistencias entre valores de columnas diferentes\n",
    "#### Otra buena practica es buscar datos que no tienen sentido entre sí\n",
    "#### No esperaríamos que el equipo con más puntos en un partido no aparezca como el ganador \"w\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.query('(pts > opp_pts) & (game_result != \"W\")').empty"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### o al revés, que el equipo con menos puntos no fuera el perdedor \"L\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.query('(pts < opp_pts) & (game_result != \"L\")').empty"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Trabajando con múltiples conjuntos de datos en Pandas\n",
    "### Combinar Datos basados en el índice con `concat()`\n",
    "#### Para ver esto, regresamos a nuestros datos de ciudades y agreguemos unas ciudades más"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mas_datos_ciudades = pd.DataFrame(\n",
    "    {\"ingresos\": [7000, 3400], \"conteo_empleados\":[2, 2]},\n",
    "    index=[\"Nueva York\", \"Barcelona\"]\n",
    ")\n",
    "\n",
    "todos_los_datos_ciudades = pd.concat([datos_ciudades, mas_datos_ciudades], sort=False)\n",
    "\n",
    "todos_los_datos_ciudades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "paises_ciudades = pd.DataFrame({\n",
    "    \"pais\": [\"Holanda\", \"Japon\", \"Holanda\", \"Canada\", \"España\"],\n",
    "    \"capital\": [1, 1, 0, 0, 0]},\n",
    "    index=[\"Amsterdam\", \"Tokyo\", \"Rotterdam\", \"Toronto\", \"Barcelona\"]\n",
    ")\n",
    "\n",
    "ciudades = pd.concat([todos_los_datos_ciudades, paises_ciudades], axis=1, sort=False)\n",
    "\n",
    "ciudades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.concat([todos_los_datos_ciudades, paises_ciudades], axis=1, join=\"inner\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Combinación de datos basados en columnas arbitrarias con `merge()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "paises = pd.DataFrame({\n",
    "    \"poblacion_millones\": [17, 127, 37],\n",
    "    \"continente\": [\"Europa\", \"Asia\", \"Norte America\"]\n",
    "}, index= [\"Holanda\", \"Japon\", \"Canada\"])\n",
    "\n",
    "pd.merge(ciudades, paises, left_on=\"pais\", right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.merge(\n",
    "    ciudades,\n",
    "    paises,\n",
    "    left_on=\"pais\",\n",
    "    right_index=True,\n",
    "    how=\"left\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualización de datos de un DataFrame de Pandas\n",
    "#### Dicen que una imagen vale mil palabras.  Definitivamente eso es cierto.  Veremos como sacar algunas gráficas\n",
    "#### El \";\" al final de la instrucción no es necesario, pero bloquea un texto de ejecución que no nos es útil\n",
    "#### Veamos el historial de puntos anotados por los Knicks, cada año"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba.query('fran_id == \"Knicks\"').groupby(\"year_id\")[\"pts\"].sum().plot();  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### En la instrucción anterior se puede ver una de las cosas muy poderosas de Pandas:  la capacidad de encadenar métodos, uno tras otro\n",
    "#### Generémos una gráfica de barras, esta vez para ver el número de juegos que ha jugado cada franquicia en su historia (los 10 más altos) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nba[\"fran_id\"].value_counts().head(10).plot(kind=\"bar\");  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ejercicio\n",
    "\n",
    "### En el 2013, el Miami Heat ganó el campeonato. Cree una gráfica de pastel (pie plot) que muestre el número de juegos ganados y perdidos durante esa temporada."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Utilice el espacio de arriba para generar esta solución - NO ESCRIBA ACÁ"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language": "python",
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}