1 - Introducción a pandas + numpy + matplotlib.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# 1 - Introducción a pandas + numpy + matplotlib\n",
    "Revisamos algunos **conceptos iniciales** de estas librerías para empezar a analizar datos. Para un tutorial inicial más completo ver [10 Minutes to pandas](http://pandas.pydata.org/pandas-docs/stable/10min.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.1 - Los objetos de pandas: Series\n",
    "Se puede construir un objeto `Series` con cualquier serie de valores. Las `Series` pueden entenderse como listas con **funcionalides adicionales**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    1.0\n",
      "1    3.0\n",
      "2    5.0\n",
      "3    NaN\n",
      "4    6.0\n",
      "5    8.0\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "s = pd.Series([1,3,5,np.nan,6,8])\n",
    "print s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Suma: 23.0 / Cantidad de valores: 5\n"
     ]
    }
   ],
   "source": [
    "print \"Suma:\", s.sum(), \"/ Cantidad de valores:\", s.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Las operaciones sobre `Series` son distintas que las operaciones sobre listas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     2.0\n",
       "1     6.0\n",
       "2    10.0\n",
       "3     NaN\n",
       "4    12.0\n",
       "5    16.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s * 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1.0, 3.0, 5.0, nan, 6.0, 8.0, 1.0, 3.0, 5.0, nan, 6.0, 8.0]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(s) * 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Las `Series` pueden tener índices, además del mero índice posicional. Por ejemplo, se puede especificar un **índice temporal**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2016-11-23    1.0\n",
       "2016-11-24    3.0\n",
       "2016-11-25    5.0\n",
       "2016-11-26    NaN\n",
       "2016-11-27    6.0\n",
       "2016-11-28    8.0\n",
       "Freq: D, dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# en este caso elegimos crear una serie de tiempo\n",
    "s2 = pd.Series([1,3,5,np.nan,6,8], pd.date_range(\"20161123\", \"20161128\"))\n",
    "s2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Todos los valores de las `Series` son del mismo **tipo de dato**. Pandas interpreta los valores de la `Series` para determinar el tipo de dato que mejor se ajuste a su contenido. Se puede **forzar un tipo de datos** distinto del interpretado por `pandas`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2016-11-23      1\n",
       "2016-11-24      3\n",
       "2016-11-25      5\n",
       "2016-11-26    NaN\n",
       "2016-11-27      6\n",
       "2016-11-28      8\n",
       "Freq: D, dtype: object"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s3 = pd.Series([1,3,5,np.nan,6,8], pd.date_range(\"20161123\", \"20161128\"), dtype=str)\n",
    "s3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.2 - Los objetos de pandas: DataFrame\n",
    "Un `DataFrame` es un **conjunto de series** que tienen el mismo índice y la misma cantidad de valores. Si cada `Series` es el análogo a una lista, cada `DataFrame` es el análogo a una tabla."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4  serie_5\n",
       "2016-11-23      1.0       1        4      300\n",
       "2016-11-24      3.0       3        3      300\n",
       "2016-11-25      5.0       5        2      300\n",
       "2016-11-26      NaN     NaN       23      300\n",
       "2016-11-27      6.0       6       15      300\n",
       "2016-11-28      8.0       8       30      300"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame({\n",
    "        \"serie_2\": s2, \n",
    "        \"serie_3\": s3,\n",
    "        \"serie_4\": [4, 3, 2, 23, 15, 30],\n",
    "        \"serie_5\": np.array([300] * 6, dtype='int32')\n",
    "    })\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hay muchas formas de crear un `DataFrame`. `Pandas` tiene toda una serie de **funciones para leer distintos tipos de archivos tabulares** y cargarlos en un `DataFrame`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>321</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "0     123      Juan Pérez    M\n",
       "1     321   Lucía Alvarez    F\n",
       "2      15     Carlos Hale    M\n",
       "3      10     Sofía Cilar    F"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas = pd.read_csv(\"data/dataframe_example.csv\")\n",
    "df_personas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Las **columnas y las filas pueden utilizarse individualmente como series**. En el caso de las columnas, si no se utilizan espacios en los encabezados, incluso se pueden llamar como cualquier otro atributo de un `DataFrame`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    M\n",
       "1    F\n",
       "2    M\n",
       "3    F\n",
       "Name: sexo, dtype: object"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas.sexo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['M', 'F'], dtype=object)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas.sexo.unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Una forma útil para obtener algunas **estadísticas sumarias**de los campos numéricos es el metodo `describe`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>117.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>145.493127</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>13.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>69.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>172.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>321.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           numero\n",
       "count    4.000000\n",
       "mean   117.250000\n",
       "std    145.493127\n",
       "min     10.000000\n",
       "25%     13.750000\n",
       "50%     69.000000\n",
       "75%    172.500000\n",
       "max    321.000000"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2.1 - Ordenar un DataFrame\n",
    "Podemos ordenar un `DataFrame` **por los valores de sus columnas**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>321</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "2      15     Carlos Hale    M\n",
       "0     123      Juan Pérez    M\n",
       "1     321   Lucía Alvarez    F\n",
       "3      10     Sofía Cilar    F"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas.sort_values(\"nombre_apellido\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>321</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "3      10     Sofía Cilar    F\n",
       "2      15     Carlos Hale    M\n",
       "0     123      Juan Pérez    M\n",
       "1     321   Lucía Alvarez    F"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas.sort_values(\"numero\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos ordenar un `DataFrame` **por los valores de sus índices**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>321</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "3      10     Sofía Cilar    F\n",
       "2      15     Carlos Hale    M\n",
       "1     321   Lucía Alvarez    F\n",
       "0     123      Juan Pérez    M"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# ordena las filas\n",
    "df_personas.sort_index(axis=0, ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sexo</th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>M</td>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>321</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>M</td>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>10</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  sexo  numero nombre_apellido\n",
       "0    M     123      Juan Pérez\n",
       "1    F     321   Lucía Alvarez\n",
       "2    M      15     Carlos Hale\n",
       "3    F      10     Sofía Cilar"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# ordena las columnas\n",
    "df_personas.sort_index(axis=1, ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2.2 - Seleccionar elementos en un DataFrame\n",
    "Los objetos de tipo `DataFrame` soportan funcionalidades extendidas para los operadores **[ ]**. Estos **permiten seleccionar filas, columnas o combinaciones de éstas** dentro del `DataFrame`.\n",
    "\n",
    "Para selccionar una columna, equivalente a `df_personas.sexo`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    M\n",
       "1    F\n",
       "2    M\n",
       "3    F\n",
       "Name: sexo, dtype: object"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas[\"sexo\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para seleccionar **varias columnas al mismo tiempo**:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>sexo</th>\n",
       "      <th>nombre_apellido</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>M</td>\n",
       "      <td>Juan Pérez</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>F</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>M</td>\n",
       "      <td>Carlos Hale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>F</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  sexo nombre_apellido\n",
       "0    M      Juan Pérez\n",
       "1    F   Lucía Alvarez\n",
       "2    M     Carlos Hale\n",
       "3    F     Sofía Cilar"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas[[\"sexo\", \"nombre_apellido\"]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Selección por número de filas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4  serie_5\n",
       "2016-11-23      1.0       1        4      300\n",
       "2016-11-24      3.0       3        3      300"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[0:2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Selección por etiqueta (índice) de fila"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4  serie_5\n",
       "2016-11-24      3.0       3        3      300\n",
       "2016-11-25      5.0       5        2      300\n",
       "2016-11-26      NaN     NaN       23      300"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"20161124\":\"20161126\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Si se define una combinación de etiquetas (índices) de columna y fila, resulta equivalente a seleccionar una única celda"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[\"20161124\", \"serie_3\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Si se define una combinación de varias etiquetas, resulta en un subconjunto de valores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           serie_3  serie_5\n",
       "2016-11-24       3      300\n",
       "2016-11-25       5      300"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[\"20161124\":\"20161125\", [\"serie_3\", \"serie_5\"]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Es posible **seleccionar una posición especifica** mediante su índice de fila y columna"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "300"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[1, 3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Y al especificar rangos de valores se selecciona un subjconjunto de celdas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>2</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_4  serie_5\n",
       "2016-11-24        3      300\n",
       "2016-11-25        2      300"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[1:3, 2:4]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Si el subconjunto de celdas que se desea seleccionar no es contiguo, se pueden especificar a mano sus índices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>2</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>30</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_4  serie_5\n",
       "2016-11-24        3      300\n",
       "2016-11-25        2      300\n",
       "2016-11-28       30      300"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[[1, 2, 5], [2, 3]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Selección por expresión booleana"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "0     123      Juan Pérez    M\n",
       "2      15     Carlos Hale    M"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas[[True, False, True, False]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Los operadores nativos de Python usados para hacer comparaciones booleanas funcionan lugar a lugar dentro de una `Series`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     True\n",
       "1    False\n",
       "2     True\n",
       "3    False\n",
       "Name: sexo, dtype: bool"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas.sexo == \"M\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Esto permite hacer queries facilmente.** Por ej, para seleccionar a las personas de sexo masculino dentro del `DataFrame` se usa un comparador booleano como selector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "0     123      Juan Pérez    M\n",
       "2      15     Carlos Hale    M"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas[df_personas.sexo == \"M\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Este tipo de comparaciones booleanas tambien pueden hacerse para un `DataFrame`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           serie_2 serie_3 serie_4 serie_5\n",
       "2016-11-23   False   False    True    True\n",
       "2016-11-24   False   False   False    True\n",
       "2016-11-25    True    True   False    True\n",
       "2016-11-26   False   False    True    True\n",
       "2016-11-27    True    True    True    True\n",
       "2016-11-28    True    True    True    True"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df > 3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Este tipo de comparaciones tambien pueden hacerse para **visualizar queries dentro de un dataframe**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23.0</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15.0</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30.0</td>\n",
       "      <td>300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4  serie_5\n",
       "2016-11-23      NaN     NaN      4.0      300\n",
       "2016-11-24      NaN     NaN      NaN      300\n",
       "2016-11-25      5.0       5      NaN      300\n",
       "2016-11-26      NaN     NaN     23.0      300\n",
       "2016-11-27      6.0       6     15.0      300\n",
       "2016-11-28      8.0       8     30.0      300"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df > 3] # máscara sobre todos los valores del DataFrame"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La utilización de comparadores puede ser tan compleja como se desee. **Para concatenar condiciones se utiliza `&` como `and` y `|` como `or`**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4  serie_5\n",
       "2016-11-23      NaN     NaN      4.0      NaN\n",
       "2016-11-24      NaN     NaN      NaN      NaN\n",
       "2016-11-25      5.0       5      NaN      NaN\n",
       "2016-11-26      NaN     NaN     23.0      NaN\n",
       "2016-11-27      6.0       6     15.0      NaN\n",
       "2016-11-28      8.0       8     30.0      NaN"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[(df > 3) & (df < 300)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2.3 - Setear elementos en un DataFrame\n",
    "Setear una nueva serie/columna automáticamente ubica los elementos en el lugar que le corresponden según el índice de la serie y del DataFrame. Los elementos que exceden el índice del DataFrame no se incorporan."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2016-11-26    4\n",
       "2016-11-27    3\n",
       "2016-11-28    2\n",
       "2016-11-29    6\n",
       "2016-11-30    7\n",
       "2016-12-01    5\n",
       "Freq: D, dtype: int64"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s4 = pd.Series([4, 3, 2, 6, 7, 5], index=pd.date_range('20161126', periods=6))\n",
    "s4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "      <th>serie_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>300</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>300</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>300</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>300</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>300</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>300</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4  serie_5  serie_6\n",
       "2016-11-23      1.0       1        4      300      NaN\n",
       "2016-11-24      3.0       3        3      300      NaN\n",
       "2016-11-25      5.0       5        2      300      NaN\n",
       "2016-11-26      NaN     NaN       23      300      4.0\n",
       "2016-11-27      6.0       6       15      300      3.0\n",
       "2016-11-28      8.0       8       30      300      2.0"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"serie_6\"] = s4\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Los mismos atributos que permiten seleccionar valores, permiten setear valores**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "      <th>serie_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>500</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>500</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>500</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>500</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>500</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>500</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4  serie_5  serie_6\n",
       "2016-11-23      1.0       1        4      500      NaN\n",
       "2016-11-24      3.0       3        3      500      NaN\n",
       "2016-11-25      5.0       5        2      500      NaN\n",
       "2016-11-26      NaN     NaN       23      500      4.0\n",
       "2016-11-27      6.0       6       15      500      3.0\n",
       "2016-11-28      8.0       8       30      500      2.0"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[:, \"serie_5\"] = 500\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "      <th>serie_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>500</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>500</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>500</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>900</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>500</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>500</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4  serie_5  serie_6\n",
       "2016-11-23      1.0       1        4      500      NaN\n",
       "2016-11-24      3.0       3        3      500      NaN\n",
       "2016-11-25      5.0       5        2      500      NaN\n",
       "2016-11-26      NaN     NaN       23      900      4.0\n",
       "2016-11-27      6.0       6       15      500      3.0\n",
       "2016-11-28      8.0       8       30      500      2.0"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[\"20161126\", \"serie_5\"] = 900\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "      <th>serie_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>arbol</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>arbol</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>arbol</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4 serie_5  serie_6\n",
       "2016-11-23      1.0       1        4   arbol      NaN\n",
       "2016-11-24      3.0       3        3   arbol      NaN\n",
       "2016-11-25      5.0       5        2   arbol      NaN\n",
       "2016-11-26      NaN     NaN       23   arbol      4.0\n",
       "2016-11-27      6.0       6       15   arbol      3.0\n",
       "2016-11-28      8.0       8       30   arbol      2.0"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.iloc[:, 3] = \"arbol\"\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2.4 - Agrupar por categorías"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>321</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "0     123      Juan Pérez    M\n",
       "1     321   Lucía Alvarez    F\n",
       "2      15     Carlos Hale    M\n",
       "3      10     Sofía Cilar    F"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sexo</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>F</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>M</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      numero  nombre_apellido\n",
       "sexo                         \n",
       "F          2                2\n",
       "M          2                2"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas.groupby(\"sexo\").count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>321.0</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15.0</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10.0</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "0     NaN      Juan Pérez    M\n",
       "1   321.0   Lucía Alvarez    F\n",
       "2    15.0     Carlos Hale    M\n",
       "3    10.0     Sofía Cilar    F"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas_nan = df_personas.copy()\n",
    "df_personas_nan.loc[0, \"numero\"] = np.nan\n",
    "df_personas_nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sexo</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>F</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>M</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      numero  nombre_apellido\n",
       "sexo                         \n",
       "F          2                2\n",
       "M          1                2"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas_nan.groupby(\"sexo\").count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.3 - Graficar\n",
    "Los objetos de `pandas` tienen un atributo `plot` que es un **wrapper sobre matplotlib, que simplifica la creación de gráficos**. `matplotlib` por default grafica en una ventana independiente a la consola que ejecuta los scripts. Para que jupyter grafique dentro de la celda de ouput existe el comando magic `%matplotlig.inline` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "plt.style.use(\"ggplot\") # para que los gráficos sean más lindos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3846f427d0>"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgoAAAGKCAYAAABtmWkQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3846f88590>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.serie_4.plot.line()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3846f0f990>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAAFqCAYAAADBfvryAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAEb5f5v/MBbBO1Z/AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3846e97610>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.serie_4.plot.hist(bins=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3844c624d0>"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhoAAAGDCAYAAABtOLkwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzt3Xl4VPW9x/HPyWQ3hCQkmZgEg5G9gEFQpJnLIioq3j5U\nVEDwEhcUo1ifa6nFFRfKLZbS1qp9KlxDUWQRcbluIEVtojbWpcgmSApehARCEgIMCZPk3D/mMjGG\nJZnMyZlM3q/n4Xmc35n5zXe+HpjPnPObM4ZpmqYAAAAsEGZ3AQAAIHQRNAAAgGUIGgAAwDIEDQAA\nYBmCBgAAsAxBAwAAWIagAQAALEPQAAAAliFoAAAAyxA0AACAZYIyaNTU1KigoEB33nmnpk6dqoce\nekg7d+5s9TyFhYUWVNcx0Qsv+uBFHxrRCy/60IheeAWqD0EZNJ599ll99dVXmjlzphYsWKBBgwbp\n8ccfV2VlZavmKSoqsqjCjodeeNEHL/rQiF540YdG9MIrUH0IuqBx/PhxFRcX68Ybb1Tfvn3ldDp1\n3XXXKS0tTWvXrrW7PAAA0ApBFzQaGhrU0NCg8PDwJuORkZHatm2bTVUBAAB/BF3QiI6OVu/evbV6\n9WpVVlaqoaFBH374obZv366qqiq7ywMAAK0QdEFDkmbOnClJmjFjhqZMmaJ3331XLpdLYWGtK7df\nv35WlNchOZ1Ou0sICvTBiz40ohde9KERvfAK1HuoYZqmGZCZLHD8+HG53W4lJCTod7/7nWpqavTL\nX/6y2f0KCwubLVrp16+ffvKTn7RXqQAAhJzXX39dW7dubTKWm5srl8vV4jmCOmiccOTIEc2cOVM3\n3nijLrnkklY9trKyUnV1dRZV1nHEx8erurra7jJsRx+86EMjeuFFHxrRCyk8PFyJiYmBmSsgswTY\nP//5T5mmqfT0dJWWluqFF15QRkaGRo0a1eq56urq5PF4Al9kB2OaJn0QfTiBPjSiF170oRG9CKyg\nDBput1vLli1TRUWF4uLidPHFF2vSpEmtXqMBAADsFZRBY/jw4Ro+fLjdZQAAgDbiEAEAALAMQQMA\nAFiGoAEAACxD0AAAAJYhaAAAAMsQNAAAgGUIGgAAwDIEDQAAYBmCBgAAsAxBAwAAWIagAQAALEPQ\nAAAAliFoAAAAyxA0AACAZQgaAADAMgQNAABgGYIGAACwDEEDAABYhqABAAAsQ9AAAACWIWgAAADL\nEDQAAIBlCBoAAMAy4XYX8EMNDQ1auXKlCgsLVVVVpcTERI0aNUoTJkywuzQAANBKQRc0Xn31Vb33\n3nu66667lJmZqZ07d+qZZ57RWWedpSuuuMLu8gAAQCsEXdDYvn27hg4dqpycHElScnKyCgsL9c03\n39hcGQAAaK2gW6PRp08fbdq0Sfv27ZMk7dq1S19//bUGDx5sc2UAgFBWbxo65DG0ZW+lDnkM1ZuG\n3SWFhKA7ojF+/HgdO3ZM99xzj8LCwmSapiZNmqTc3Fy7SwMAhKh609DG/cc0b8Nu1dabinIYmj06\nS4NSY+QwTLvL69CCLmh89NFHKiws1D333KPMzEzt2rVLBQUFSkpK0ogRI+wuDwAQgo7UyRcyJKm2\n3tS8Dbv13IS+6hphc3EdXNAFjRdeeEE//elPNXz4cElS9+7ddeDAAa1Zs+aUQaOwsFBFRUVNxpxO\np/Ly8hQfHy/TJI1GREQoKSnJ7jJsRx+86EMjeuHV2ftQurfSFzJOqK03VVMvnevsfH0xDO9po4KC\nApWVlTXZlpubK5fL1eK5gi5oHD9+3PcCTzAM47RhweVynfJFV1dXy+PxBLTGjigpKUkVFRV2l2E7\n+uBFHxrRC6/O3ocYh6Eoh9EkbEQ5DEU71Cn7EhERoZSUFOXl5bV5rqBbDDpkyBC98sor+vzzz3Xg\nwAEVFxfrzTff1EUXXWR3aQCAEBUXLs0enaUoh/eDbnR4mGaPzlJc0H0c73gMM8jOK9TU1GjFihUq\nLi5WdXW1EhMT5XK5NGHCBDkcjlbPd+DAAY5oiE8rJ9AHL/rQiF540QfvgtAjdVJNvRTt8IaPzroQ\n9MQRjUAIuqwWHR2tadOmadq0aXaXAgDoRByGqa4R3jUZnT10BVLQnToBAAChg6ABAAAsQ9AAAACW\nIWgAAADLEDQAAIBlCBoAAMAyBA0AAGAZggYAALAMQQMAAFiGoAEAACxD0AAAAJYhaAAAAMsQNAAA\ngGUIGgAAwDIEDQAAYBmCBgAAsAxBAwAAWIagAQAALEPQAAAAliFoAAAAyxA0AACAZQgaAADAMgQN\nAABgGYIGAACwTLjdBfzQnXfeqfLy8mbjY8eO1c0332xDRQAAwF9BFzT+67/+Sw0NDb7b3377rZ54\n4gkNHz7cxqoAAIA/gi5odOnSpcntzz77TGlpaerXr59NFQEAAH8F9RqNuro6/e1vf9Po0aPtLgUA\nAPghqINGcXGx3G63Ro0aZXcpAADAD0EdNDZs2KDBgwcrISHB7lIAAIAfgm6Nxgnl5eX66quvNGvW\nrDPet7CwUEVFRU3GnE6n8vLyFB8fL9M0rSqzw4iIiFBSUpLdZdiOPnjRh0b0wos+NKIXkmEYkqSC\nggKVlZU12ZabmyuXy9XyucwgfRdeuXKl1q9fr2effVZhYf4feDlw4IA8Hk8AK+uYkpKSVFFRYXcZ\ntqMPXvShEb3wog+N6IU3bKWkpARkrqA8dWKapj744AONGjWqTSEDAADYKyjfxb/66iuVl5fzbRMA\nADq4oFyjMWjQIK1YscLuMgAAQBsF5RENAAAQGggaAADAMgQNAABgGYIGAACwDEEDAABYhqABAAAs\nQ9AAAACWIWgAAADLEDQAAIBlCBoAAMAyBA0AAGAZggYAALAMQQMAAFiGoAEAACxD0AAAAJYhaAAA\nAMsQNAAAgGUIGgAAwDIEDQAAYBmCBgAAsAxBAwAAWIagAQAALEPQAAAAlgm3u4CTqaio0Isvvqgv\nv/xStbW1Ovvss3XHHXcoOzvb7tIAAEArBF3QOHr0qB566CENHDhQDzzwgLp06aJ9+/YpLi7O7tIA\nAEArBV3QePXVV5WcnKwZM2b4xlJSUmysCAAA+CvogsZnn32mnJwc/fa3v9XWrVuVlJSkyy+/XGPG\njLG7NKDDqzcNHamTSvdWKsZhKC5cchim3WUBCGFBFzTKysq0du1aXX311brmmmv0zTff6Pnnn1dE\nRIRGjBhhd3lAh1VvGtq4/5jmbdit2npTUQ5Ds0dnaVBqDGEDgGWC7lsnpmkqOztbkyZNUo8ePXTp\npZdqzJgxWrdund2lAR3akTr5QoYk1dabmrdht47U2VwYgJAWdEc0EhMTlZGR0WQsIyNDxcXFp3xM\nYWGhioqKmow5nU7l5eUpPj5epsmntYiICCUlJdldhu06cx9K91b6QsYJtfWmauqlc52dsydS594n\nvo8+NKIXkmEYkqSCggKVlZU12ZabmyuXy9XiuYIuaPTp00d79+5tMrZ3714lJyef8jEul+uUL7q6\nuloejyegNXZESUlJqqiosLsM23XmPsQ4DEU5jCZhI8phKNqhTtsTqXPvE99HHxrRC2/YSklJUV5e\nXpvnCrpTJ+PGjdOOHTu0Zs0alZaWqrCwUH/96191xRVX2F0a0KHFhUuzR2cpyuH9pBIdHqbZo7MU\nF3QfNwCEEsMMwvMKn3/+uZYtW6bS0lKlpqbq6quv1iWXXOLXXAcOHOCIhkjoJ3T2Ppz41klNvRTt\nEN86EfvECfShEb1oPKIRCEH5WeaCCy7QBRdcYHcZQMhxGKa6RnjXZHT2f0gBtI+gO3UCAABCB0ED\nAABYhqABAAAsQ9AAAACWIWgAAADLEDQAAIBlCBoAAMAyBA0AAGAZggYAALAMQQMAAFiGoAEAACxD\n0AAAAJYhaAAAAMsQNAAAgGUIGgAAwDIEDQAAYBmCBgAAsAxBAwAAWIagAQAALEPQAAAAlgn350Gm\naaqurk4RERG+MY/Ho61bt8rtduvcc8+V0+kMWJEAAKBjanXQWL58ud5++23V1taqT58++tnPfqba\n2lo99thjqqiokCQZhqErrrhCeXl5ga4XAAB0IK0KGh9++KHWrFmjCy64QCkpKXr//ff1zDPPSJIG\nDx6siy++WMePH9f69ev19ttvKzs7WyNGjLCkcAAAEPxaFTTeeecdDR06VLNmzZIkZWdn69lnn5XL\n5dJtt93mu9/QoUN1//33a/369QQNAAA6sVYFjX379mnkyJG+2z/60Y8kSUOGDGl232HDhmnNmjWt\nLmjVqlV6+eWXm4ylp6dr4cKFrZ4LAADYq1VBo6GhQQ6Hw3c7OjpakhQfH9/svnFxcfJ4PH4V1b17\ndz388MMyTVOSmjwnAADoOFoVNBITE1VeXu67HRUVpcmTJystLa3ZfcvLy08aQFrC4XD4/dj2Vm8a\nOlInHfU06KyIMMWFSw7DtLssAACCQquCRnZ2tr7++mvf7cjISI0fP/6k9/3yyy917rnn+lXUvn37\ndPvttysyMlK9evXSDTfcoOTkZL/mslK9aWjj/mOat2G3autNRTkMzR6dpUGpMYQNAADUyqBx4403\n6vDhw2e8X3V1tfr06XPStRtn0qtXL+Xn5ys9PV1VVVVatWqVHnnkES1YsMB3qiZYHKmTL2RIUm29\nqXkbduu5CX3VNeIMDwYAoBMwzBMLISxWV1en7du3q0ePHoqNjW3x49xut/Lz8zVt2jSNHj36pPcp\nLCxUUVFRkzGn06m8vDzV1tbKqpe4ZW+lbl+9tdn4nyf0U7/0REue018RERF+r5kJJfTBiz40ohde\n9KERvfBeDysqKkoFBQUqKytrsi03N1cul6vFc/l1ZVB/HDlyRI8++qgeeughDRgwoMWPi42N1dln\nn63S0tJT3sflcp3yRVdXV1u2w8Q4DEU5DN8RDUmKchiKdsh38bJgkZSUFHQ12YE+eNGHRvTCiz40\nohfesJWSkhKQC28G/W+d1NTUqKysTImJwXWEQJLiwqXZo7MU5TAkSdHhYZo9Oktx7RbfAAAIbkH3\nlrh06VINGTJEKSkpqqio0MqVK+VwOJSbm2t3ac04DFODUmP03IS+cnsaFMu3TgAAaCLogsbBgwf1\nhz/8QYcPH1Z8fLz69u2ruXPnqkuXLnaXdlIOw1TXCKlrhCGJgAEAwPcFXdC455577C4BAAAESNCv\n0QAAAB0XQQMAAFiGoAEAACzTbkEjNjZWd9xxhzIzM9vrKQEAgM3atBjU7XZr7dq12rx5sw4dOqTb\nbrtNPXv21JEjR/T+++9r6NChvh9ci4yM1KhRowJRMwAA6CD8DhoHDx7UnDlzVF5errPPPlvfffed\nampqJHl/In7dunU6cOCAbrrppoAVCwAAOha/g8bSpUt17NgxPfnkk4qPj9f06dObbL/wwgv1+eef\nt7lAAADQcfm9RmPjxo268sorlZmZKcMwmm13Op06ePBgm4oDAAAdm99B4/jx44qPjz/l9mPHjvk7\nNQAACBF+B43MzExt3dr8J9JP+PTTT9WjRw9/pwcAACHA76Bx1VVXqaioSK+++qrcbrckqaGhQaWl\npXrqqae0fft2jRs3LmCFAgCAjsfvxaAjRoxQeXm5VqxYoeXLl0uSfvWrX8k0TYWFhWny5Mm66KKL\nAlYoAADoeNp0HY1rrrlGI0aM0CeffKLS0lKZpimn06lhw4bJ6XQGqkYAANBBtfnXW5OTk3X11VcH\nohYAABBi+K0TAABgmRYf0Zg4caIMw9ALL7yg8PBwTZw48YyPMQzDt34DAAB0Pi0OGhMmTJBhGAoL\nC2tyGwAA4FRaHDSuv/76094GAAD4Ib/WaNTW1uq+++7T2rVrA10PAAAIIX4FjaioKO3fv59TJwAA\n4LT8/tZJTk6O/vnPfwayFgAAEGL8DhoTJkzQvn379Ic//EHbtm1TRUWFjhw50uwPAADovPy+YNe9\n994rSdqzZ4+KiopOeb8VK1b4+xQAAKCD8ztotNfXW9esWaPly5frqquu0rRp0yx/PgAAEDh+B432\n+HrrN998o/Xr1ysrK8vy5wIAAIEXsEuQu91uNTQ0BGo61dTU6KmnntKMGTN01llnBWxeAADQftoU\nNHbu3Km5c+dq6tSpuvnmm7VlyxZJUnV1tebPn6/Nmzf7PfeiRYs0ZMgQDRgwoC0lAgAAG/kdNL7+\n+ms9/PDDKi0t1b/927/JNE3ftvj4eLndbq1bt86vuYuKirR7927dcMMN/pYHAACCgN9B46WXXlJG\nRoZ++9vfavLkyc22/+hHP9I333zT6nkPHjyogoICzZw5U+Hhbf4VewAAYCO/38l37typyZMnKyIi\nQjU1Nc22JyUlqbKystXzlpSUqLq6Wvfdd59vrKGhQVu2bNE777yjZcuWNfu2S2FhYbOv2DqdTuXl\n5Sk+Pr7J0ZbOKiIiQklJSXaXYTv64EUfGtELL/rQiF7I9z5bUFCgsrKyJttyc3PlcrlaPJffQcPh\ncJz2DbyiokIxMTGtnnfgwIFasGBBk7Gnn35aGRkZGj9+/Em/UutyuU75oqurq+XxeFpdR6hJSkpS\nRUWF3WXYjj540YdG9MKLPjSiF96wlZKSory8vDbP5XfQ6NWrlz755BONGzeu2baamhq9//776t+/\nf6vnjY6OVmZmZrOxLl26NBsHAADBze81Gtdff71KSko0b948ffHFF5KkXbt2af369frlL3+p6upq\nTZgwIWCFAgCAjscw27CAYdOmTXruuedUWlraZNzpdGrGjBl+HdEItAMHDnDqRBwKPIE+eNGHRvTC\niz40oheNp04CoU1f6xgwYIB+//vfa9euXdq3b59M05TT6dR5550XkOIAAEDH5nfQ2LVrl/bs2SOX\ny6UePXqoR48e+vLLL/WXv/xFHo9HLpdLV111VSBrBQAAHYzfazReeOEFffzxx77b+/fv129+8xvt\n379fkrRkyRK99957ba8QAAB0WH4Hjd27d6tPnz6+2x988IHCwsL061//Wr/61a908cUX+31lUAAA\nEBr8Dhput1tdunTx3f7iiy80aNAgxcfHS5IGDRrUbJEoAADoXPwOGgkJCfruu+8kSZWVlSopKdGg\nQYN822tqak56cS0AANB5+L0Y9MILL9Tbb7+t48ePa+fOnYqIiNBFF13k27579245nc6AFAkAADom\nv4PGpEmTVF1drb/97W+KjY1Vfn6+EhISJHlPq3zyyScaO3ZswAoFAAAdj99BIzo6Wnffffcpt/3p\nT39SZGSk34UBAICOz5LfYQ8LC1NsbKwVUwMAgA7E78WgAAAAZ0LQAAAAliFoAAAAyxA0AACAZQga\nAADAMgQNAABgGYIGAACwDEEDAABYhqABAAAsQ9AAAACWIWgAAADLEDQAAIBlCBoAAMAyBA0AAGAZ\nS34mvi3Wrl2rdevWaf/+/ZKk7t2769prr1VOTo7NlQEAgNYKuqCRnJysKVOmKC0tTZL0/vvva/78\n+Zo/f74yMzNtrg4AALRG0AWNCy64oMntSZMmae3atdqxY0fIBI1609CROumop0FnRYQpLlxyGKbd\nZQEAEHBBFzS+r6GhQR9//LFqa2vVu3dvu8sJiHrT0Mb9xzRvw27V1puKchiaPTpLg1JjCBsAgJAT\nlEHj22+/1YMPPiiPx6Po6GjNmjVLGRkZdpcVEEfq5AsZklRbb2reht16bkJfdY2wuTgAAALMME0z\n6D5G19fXq7y8XEePHtXf//53rV+/Xo8++ugpw0ZhYaGKioqajDmdTuXl5am2tlbB9BK37K3U7au3\nNhv/84R+6peeaNnzRkREyOPxWDZ/R0EfvOhDI3rhRR8a0QvJMAxFRUWpoKBAZWVlTbbl5ubK5XK1\nfK5gDBo/9PjjjystLU3Tp09v9WMPHDgQVDvMIY+h6au3+Y5oSFKUw/j/IxrW/a9ISkpSRUWFZfN3\nFPTBiz40ohde9KERvfCGrZSUlIDM1SGuo2GaZlCFhbaIC5dmj85SlMOQJEWHh2n26CzFBeVJLAAA\n2ibo3t5eeuklDR48WN26ddOxY8dUWFioLVu26MEHH7S7tIBwGKYGpcbouQl95fY0KJZvnQAAQljQ\nBY1Dhw7pj3/8oyorKxUbG6usrCw9+OCDGjBggN2lBYzDMNU1QuoaYUgiYAAAQlfQBY0ZM2bYXQIA\nAAiQDrFGAwAAdEwEDQAAYBmCBgAAsAxBAwAAWIagAQAALEPQAAAAliFoAAAAyxA0AACAZQgaAADA\nMgQNAABgGYIGAACwDEEDAABYhqABAAAsQ9AAAACWIWgAAADLEDQAAIBlCBoAAMAyBA0AAGAZggYA\nALAMQQMAAFiGoAEAACxD0AAAAJYhaAAAAMuE213AD61Zs0bFxcXau3evIiMj1bt3b02ZMkXp6el2\nlwYAAFop6ILGtm3bdOWVVyo7O1sNDQ1atmyZ5s6dq4ULFyoyMtLu8gAAQCsE3amT2bNna8SIEcrM\nzNQ555yj/Px8lZeXq6SkxO7SAABAKwVd0Pght9stSYqLi7O5EgAA0FpBHTRM01RBQYH69u2rzMxM\nu8sBAACtFNRBY9GiRdqzZ4/uueceu0sBAAB+MEzTNO0u4mQWL16szz77TI899piSk5NPe9/CwkIV\nFRU1GXM6ncrLy1Ntba2C9CW2q4iICHk8HrvLsB198KIPjeiFF31oRC8kwzAUFRWlgoIClZWVNdmW\nm5srl8vV8rmCMWgsXrxY//jHPzRnzhw5nc42zXXgwIFOv8NIUlJSkioqKuwuw3b0wYs+NKIXXvSh\nEb3whq2UlJSAzBV0p04WLVqkwsJC3X333YqKilJVVZWqqqp0/Phxu0sDAACtFHTX0Vi3bp0kac6c\nOU3G8/PzNXLkSBsqAgAA/gq6oLFixQq7SwAAAAESdKdOAABA6CBoAAAAyxA0AACAZQgaAADAMgQN\nAABgGYIGAACwDEEDAABYhqABAAAsQ9AAAACWIWgAAADLEDQAAIBlCBoAAMAyBA0AAGAZggYAALAM\nQQMAAFiGoAEAACxD0AAAAJYhaAAAAMsQNAAAgGUIGgAAwDIEDQAAYBmCBgAAsAxBAwAAWIagAQAA\nLBNudwEns3XrVr3++usqKSlRVVWVZs2apaFDh9pdFgAAaKWgPKJRW1urHj166JZbbrG7FAAA0AZB\neUQjJydHOTk5dpcRVOpNQ0fqpKOeBp0VEaa4cMlhmHaXBQDAaQVl0EBT9aahjfuPad6G3aqtNxXl\nMDR7dJYGpcYQNgAAQS0oT52gqSN18oUMSaqtNzVvw24dqbO5MAAAziAkjmgUFhaqqKioyZjT6VRe\nXp7i4+Nlmh37U3/p3kpfyDihtt5UTb10rjOpRXNEREQoKall9w1l9MGLPjSiF170oRG9kAzDkCQV\nFBSorKysybbc3Fy5XK4WzxUSQcPlcp3yRVdXV8vj8bRzRYEV4zAU5TCahI0oh6Foh1RRUdGiOZKS\nklp831BGH7zoQyN64UUfGtELb9hKSUlRXl5em+fi1EkHEBcuzR6dpSiHN2FGh4dp9ugsxYVETAQA\nhLKgfKuqqalRaWmp73ZZWZl27dqluLg4JScn21iZPRyGqUGpMXpuQl+5PQ2K5VsnAIAOIiiDRklJ\niR599FHf7b/85S+SpJEjRyo/P9+usmzlMEx1jZC6RhiSCBgAgI4hKING//79tWLFCrvLAAAAbcQa\nDQAAYBmCBgAAsAxBAwAAWIagAQAALEPQAAAAliFoAAAAyxA0AACAZQgaAADAMgQNAABgGYIGAACw\nDEEDAABYhqABAAAsQ9AAAACWIWgAAADLEDQAAIBlCBoAAMAyBA0AAGAZggYAALAMQQMAAFiGoAEA\nACxD0AAAAJYhaAAAAMsQNAAAgGXC7S7gVN555x298cYbqqqqUo8ePXTTTTepZ8+edpcFAABaISiP\naHz00UdaunSprr/+es2fP19ZWVmaO3euqqur7S4NAAC0QlAGjTfffFOXXnqpRo4cqYyMDE2fPl1R\nUVHasGGD3aUBAIBWCLqgUVdXp5KSEg0cONA3ZhiGBg4cqO3bt9tYGQAAaK2gCxqHDx9WQ0ODunbt\n2mS8a9euqqqqsqkqAADgj6BdDHoyhmG0+jHh4R3qJVrGMAxFRETYXYbt6IMXfWhEL7zoQyN6Edj3\nzqB7F+7SpYvCwsJ06NChJuOHDh1qdpTjhMLCQhUVFTUZ69evn37yk58oMTHRslo7mpSUFLtLCAr0\nwYs+NKIXXvShEb3wev3117V169YmY7m5uXK5XC2fxAxC999/v/nf//3fvtsNDQ3mjBkzzNdee61V\n87T2/qHs+eeft7uEoEAfvOhDI3rhRR8a0QuvQL2HBt0aDUkaN26c3nvvPX3wwQf67rvv9Nxzz6m2\ntlajRo1q1Tw/TGGdWVlZmd0lBAX64EUfGtELL/rQiF54Beo9NOhOnUjSj3/8Yx0+fFgrV670XbDr\ngQceUHx8vN2lAQCAVgjKoCFJY8eO1dixY+0uAwAAtEFQnjoBAAChwTFnzpw5dhdhpXPOOcfuEoIG\nvfCiD170oRG98KIPjeiFVyD6YJimaQagFgAAgGY4dQIAACxD0AAAAJYhaAAAAMsQNAAAgGWC9joa\nbfXOO+/ojTfe8F3w66abblLPnj3tLqvdrFq1Si+//HKTsfT0dC1cuNCmitrP1q1b9frrr6ukpERV\nVVWaNWuWhg4d2uQ+K1as0F//+lcdPXpUffr00fTp05WWlmZTxdY4Ux+eeeYZffDBB00ek5OTo9mz\nZ7d3qZZas2aNiouLtXfvXkVGRqp3796aMmWK0tPTfffxeDxasmSJPv74Y3k8Hp1//vm69dZbT/n7\nSh1VS3oxZ86cZleEvOyyy3Trrbe2d7mWWbt2rdatW6f9+/dLkrp3765rr71WOTk5kjrP/iCduReB\n2B9CMmh89NFHWrp0qW677Tb17NlTb775pubOnavf//73nerqot27d9fDDz+sE18scjgcNlfUPmpr\na9WjRw+NHj1aCxYsaLb91Vdf1TvvvKM777xTqampWr58uebOnauFCxeG1K/9nqkPkjdY3Hnnnb59\nJBR/sXIOK9pwAAAMI0lEQVTbtm268sorlZ2drYaGBi1btsz3/zsyMlKSVFBQoC+//FL33nuvYmJi\ntHjxYi1YsECPPfaYzdUHVkt6YRiGxowZo0mTJvn2i6ioKDvLDrjk5GRNmTLF9+Hi/fff1/z58zV/\n/nxlZmZ2mv1BOnMvArE/hOSpkzfffFOXXnqpRo4cqYyMDE2fPl1RUVHasGGD3aW1K4fDofj4eHXt\n2lVdu3ZVXFyc3SW1i5ycHE2cOFEXXXTRSbe//fbbmjBhgoYOHapzzjlHd911lyoqKlRcXNzOlVrr\nTH2QvMHi+/tIbGxsO1bYPmbPnq0RI0YoMzNT55xzjvLz81VeXq6SkhJJktvt1oYNGzRt2jT1799f\n5557rvLz8/X111/rm2++sbn6wDpTL06Iiopqsl9ER0fbVLE1LrjgAuXk5CgtLU1paWmaNGmSoqOj\ntWPHjk61P0in78UJbd0fQufj2/+rq6tTSUmJfvrTn/rGDMPQwIEDtX37dhsra3/79u3T7bffrsjI\nSPXq1Us33HCDkpOT7S7LVvv371dVVZUGDhzoG4uNjVWvXr20fft2/fjHP7axuva3efNmTZ8+XWed\ndZYGDBigSZMmhXwgdbvdkuR7nSUlJaqvr9eAAQN890lPT1dycrK2b98e0qdcf9iLEwoLC/Xhhx8q\nISFBQ4YM0bXXXus74hFqGhoa9PHHH6u2tla9e/fu1PvD93vRp08f33hb94eQCxqHDx9WQ0NDs3Np\nXbt21d69e22qqv316tVL+fn5Sk9PV1VVlVatWqVHHnlECxYsCLlPJ61RVVUlSSfdP05s6yxycnI0\nbNgwpaamqqysTMuWLdO8efP0xBNPyDAMu8uzhGmaKigoUN++fZWZmSnJu0+Eh4c3O5oT6vvEyXoh\nSS6XSykpKUpMTNTu3bv14osvat++fbr33nttrDbwvv32Wz344IPyeDyKjo7WrFmzlJGRoX/961+d\nbn84WS9OrNsJxP4QckHjdEL1H8+TObGQR/JeQrZnz57Kz8/Xxx9/rNGjR9tYWXAyTVNhYSF5JvGU\nvn/0pnv37jrnnHM0c+ZMbd68ucmnuVCyaNEi7dmzp0Xn2kP9osknevH44483GR8zZozvv7t3766E\nhAQ9/vjj2r9/v1JTU9u7TMtkZGToySef1NGjR/X3v/9df/zjH/Xoo4+e8v6hvD+cqhcZGRkB2R9C\n7l/WLl26KCwsTIcOHWoyfujQoZBcMdxSsbGxOvvss1VaWmp3KbZKSEiQpGb7R3V1dafePyQpNTVV\nXbp0Cdl9ZPHixfriiy80Z84cJSUl+cYTEhJUV1fnO41wQnV1tW9/CTXf70ViYuJp79urVy9JCrn9\nwuFwyOl0Kjs7W5MnT1ZWVpbeeuutTrk/nKoXJ+PP/hByQSM8PFzZ2dn66quvfGOmaWrTpk1Nzjl1\nNjU1NSorKzvjPyqhLjU1VQkJCU32D7fbrR07dnTq/UOSDh48qMOHD4fkPrJ48WL94x//0COPPNJs\nnVJ2drYcDoc2bdrkG9u7d6/Ky8vVu3fv9i7Vcqfrxcn861//kqSQfZM9wTRNeTyeTrc/nMyJXpyM\nP/tDSJ46GTdunJ5++mllZ2f7vt5aW1urUaNG2V1au1m6dKmGDBmilJQUVVRUaOXKlXI4HMrNzbW7\nNMvV1NQ0SdtlZWXatWuX4uLilJycrKuuukqvvPKK0tLSfF9v7datmy688EIbqw680/UhLi5OL7/8\nsoYNG6aEhASVlpbqxRdfVHp6us4//3wbqw68RYsWqaioSL/4xS8UFRXlO88eGxuryMhIxcbG6pJL\nLtGSJUt01llnKSYmRs8//7z69OkTcgv/ztSLsrIyFRYWavDgwerSpYt2796tJUuWqH///iH1a6Yv\nvfSSBg8erG7duunYsWMqLCzUli1b9OCDD3aq/UE6fS8CtT+E7K+3vvvuu3r99dd9F+y6+eabdd55\n59ldVrv53e9+p23btunw4cOKj49X3759NXny5JA6x3oqW7ZsOem51pEjRyo/P1+StHLlSq1fv15H\njx5Vv379dMstt4TcBbtO14dbb71VTz75pHbt2iW3263ExESdf/75mjhxYshda2bixIknHc/Pz9fI\nkSMleS/QtHTpUhUVFcnj8SgnJ0e33HJLyJ1OO1MvDh48qKeeekr/+7//q9raWnXr1k3Dhg3TNddc\nE1KLyP/0pz9p06ZNqqysVGxsrLKysjR+/Hjf2qTOsj9Ip+9FoPaHkA0aAADAfiG3RgMAAAQPggYA\nALAMQQMAAFiGoAEAACxD0AAAAJYhaAAAAMsQNAAAgGUIGgAAwDIEDQAAYBmCBoB2tXLlylNeChtA\n6CFoAGhXhmHIMIx2ea4PP/xQEydO1LRp09rl+QA0x2+dAGhXDQ0NamhoUHi4tT8eXVNTo3vuuUfH\njh2TJC1ZssTS5wNwchzRANAuamtrJUlhYWGWhwxJWr16tWJiYjRkyBDLnwvAqVn/tx1Ah1FTU6Pl\ny5fr008/VVVVle9no6dOnaoePXpIknbs2KGVK1dqx44dqq+v13nnnafJkyerT58+vnlWrlyp1atX\na8GCBVq9erW+/PJLpaam6te//rVv24oVK5o894cffqi33npLe/bsUWRkpAYNGqQbb7xR3bp1a/Xr\nKC0t1VtvvaVZs2bpo48+alNPALQNRzQA+Pz5z3/WunXrNHz4cN16663693//d0VFRWnPnj2SpE2b\nNmnOnDmqqanRddddp8mTJ8vtduuxxx7Tzp07ffOcWIOxcOFCeTwe3XDDDRozZkyTbd/3yiuv6Omn\nn1Z6erqmTZumcePG+Z7L7Xa3+nU8//zzGjBggHJycvxpA4AA4ogGAJ8vvvhCY8aM0dSpU0+6/bnn\nntOAAQM0e/Zs39hll12m//zP/9Ty5cv1wAMPNLl/VlaW7r777tM+Z3l5uVatWqXJkydr/PjxvvFh\nw4Zp1qxZWrt2bZPxM/n888/11Vdf6Te/+U2LHwPAOhzRAOATGxurnTt3qrKystm2Xbt2qbS0VLm5\nuTp8+LDvz7FjxzRgwABt3bq12WMuv/zyMz7nJ598ItM0NXz48CbzxsfH6+yzz9bmzZtbXH9dXZ2W\nLFmiyy+/XOnp6S1+HADrcEQDgM/UqVP1zDPP6I477lB2drYGDx6skSNHKjU1Vfv27ZMkPf300yd9\nrGEYcrvdio2N9Y2lpqae8TlLS0tlmuYpj3y0ZuHo//zP/+jIkSO67rrrWvwYANYiaADwGT58uPr1\n66fi4mJt3LhRb7zxhl577TX9/Oc/14lvwt94442+haE/FB0d3eR2ZGTkGZ/TNE0ZhqH7779fYWHN\nD7L+cM5TcbvdWrNmjcaOHSu32y232y3TNFVTUyPTNHXgwAFFRUUpPj6+RfMBCAyCBoAmEhISdPnl\nl+vyyy9XdXW17rvvPq1Zs0b/8R//IUmKiYnRgAEDAvZ8TqdTpmkqNTVVaWlpfs9z9OhR1dTU6LXX\nXtNrr73WbPtdd92lCy+8UD//+c/bUi6AViJoAJDkvZBWTU1Nk1Mf8fHxSkxMlMfj0XnnnSen06k3\n3nhDubm5zY40VFdX+3W0YNiwYXrppZe0atUqzZw5s9n2I0eOKC4u7ozzdO3aVbNmzWo2/tZbb2nH\njh362c9+poSEhFbXB6BtCBoAJHmvoTFjxgxdfPHFysrKUnR0tDZu3KidO3f6LuE9Y8YMzZs3T/fe\ne69GjRqlpKQkVVRUaPPmzYqNjdUvfvGLVj+v0+nUxIkT9dJLL2n//v268MILFRMTo7KyMn366ae6\n7LLLdPXVV59xnsjISA0dOrTZeHFxsXbu3HnSbQCsR9AAIMn7Rj127Fht3LhRxcXFMk1TaWlpmj59\nui699FJJUv/+/fXEE09o9erVevfdd3Xs2DElJiaqZ8+euuyyy1r8XD+8lsb48eOVnp6uN998U6tX\nr5YkdevWTTk5OQQEoIPjt04AAIBluI4GAACwDKdOAAS948ePn/FS5HFxce3yY20AWoe/lQCC3kcf\nfaRnn332tPd55JFH1L9//3aqCEBLsUYDQNCrqqry/bDbqWRnZzf5ai6A4EDQAAAAlmExKAAAsAxB\nAwAAWIagAQAALEPQAAAAliFoAAAAyxA0AACAZQgaAADAMgQNAABgmf8DO6+J5/LlKXsAAAAASUVO\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3846dcd210>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.plot.scatter(\"serie_4\", \"serie_3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3844be28d0>"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAs4AAAFKCAYAAAANJdl/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzs3XlcU3e+P/7XyUYWSMIeQKEBK4j7UhAXFEWxMl+nTuly\n7bSdjs44KlK8t+21ndbaO97ys3qdaac+OvZ24+p0tU5lXKoFEbAFpIgLgizKIkKISNgCJCQ5vz8o\nZ0zZggJheT8fDx7Vk0/e530SLG8+eZ/Ph2FZlgUhhBBCCCGkTzx7J0AIIYQQQshoQIUzIYQQQggh\nNqDCmRBCCCGEEBtQ4UwIIYQQQogNqHAmhBBCCCHEBlQ4E0IIIYQQYgMqnAkhhBBCCLEBFc6EEEII\nIYTYgApnQgghhBBCbECFMyGEEEIIITagwpkQQgghhBAbUOFMCCGEEEKIDahwJoQQQgghxAZUOBNC\nCCGEEGIDKpwJIYQQQgixARXOhBBCCCGE2IAKZ0IIIYQQQmxAhTMhhBBCCCE2oMKZEEIIIYQQG1Dh\nTAghhBBCiA2ocCaEEEIIIcQGVDgTQgghhBBiAyqcCQGQkJCAkJAQyOVyeHp6Yu3atSguLrYaYzAY\nsGXLFri5ucHJyQkxMTHQarVWY+Lj4zFv3jyIxWLMmTOn1/Pt3bsXgYGBEIvFmDhxIhISEvrMr6Cg\nADExMVCr1eDxeHjnnXe6jcnIyMCaNWvg4+MDHo+HpKSkfq9bp9MhLi4OQUFBkMlk8PPzw/PPP4+m\npqZuYz/55BPMnDkTEokEKpUKW7du7TO2La/XzZs3ER0dDZlMBpVKhZdeegkWi6XfvHfs2AFvb29I\npVKsWLECpaWl3a7rqaeegkKhgLOzMzZs2AC9Xt9v3P3790OtVkMikWD+/PnIyckZ8DURQggZu6hw\nJgSdRefWrVuRnZ2N5ORkdHR0YOXKlWhra+PGxMfH4/jx4/j666+Rnp6O6upqPProo91irV+/Hk8+\n+WSv54qLi8NHH32Effv2oaioCElJSQgJCekzv9bWVgQEBGD37t3w8vLqcYxer8esWbOwf/9+MAxj\n03VXV1ejpqYG+/btQ35+PhITE/Htt99iw4YNVuP27duH1157Da+88goKCgqQnJyMqKioPmP393pZ\nLBasXr0aJpMJWVlZSExMxCeffIIdO3b0GXf37t149913ceDAAZw/fx4ymQxRUVEwGo3cmHXr1qGw\nsBApKSk4fvw40tPTsXHjxj7jfvHFF/iP//gPvPHGG8jLy8PMmTMRFRWFuro6m6+JEELIGMcSQrq5\nffs2yzAMm5GRwbIsyzY2NrIikYg9cuQIN+batWsswzBsdnZ2t+fv3LmTnT17drfjBQUFrFAoZEtK\nSu45twceeIB9++23+xzDMAx79OjRe4r/1VdfsWKxmDWbzSzLsqxOp2OlUimbmppqcwxbXq8TJ06w\nAoGAvX37Njfmb3/7G6tUKtmOjo5eY3t5ebH79u2zOpdYLGa/+OILlmU7X2OGYdgLFy5wY7799luW\nz+ezNTU1vcYNDQ1l4+LiuL9bLBbWx8eH3b17t83XRAghZGyjGWdCetDQ0ACGYeDi4gIAyM3Nhclk\nwvLly7kxgYGB8PX1RWZmps1xjx07hoCAACQlJcHf3x9qtRq/+93voNPpBv0a7lVDQwPkcjl4vM7/\nPXz33XdgWRY3b95EcHAwJk6ciCeeeAJVVVW9xrDl9crKysL06dPh5ubGjYmKikJjYyOuXr0KAKio\nqACPx0N6ejoAoKysDBqNxiquXC5HaGioVVxnZ2fMnj2bGxMZGQmGYZCdnc0d4/F4+L//+z8AQEdH\nB3Jzc63iMgyDyMhILu6PP/44KN8DhBBCRi8qnAn5GZZlER8fj0WLFiE4OBgAoNFoIBKJIJfLrcZ6\nenpCo9HYHPvGjRsoLy/H4cOHcejQISQmJiI3NxePPfbYoF7Dvaqrq8OuXbus2hpu3LgBs9mMhIQE\nvPPOO/j6669RX1+PFStWwGQy9RjHltdLo9HA09Oz2+NdjwGAUChEUFAQpFIpd5xhmB6fd3dcDw8P\nq8f5fD5cXFys3qugoCAoFAruus1mc59xa2trB+V7gBBCyOglsHcChIw0mzdvRkFBAc6dO9fvWJZl\nbe4nBjr7eo1GIw4ePIiAgAAAwIcffoi5c+eipKQEYrGYK9YZhsErr7yC7du339uF/ExCQgLefPNN\nLnZBQQEmTJjAPd7c3Izo6GhMmzYNr7/+ulXOJpMJf/3rX7nZ1s8++wwqlQqpqalYsWKFzTnY+np1\njfH29kZBQcGgxP35mKGKSwghZOyiwpmQu8TGxuLEiRPIyMiAt7c3d1ylUsFoNKKpqclqxlGr1Xab\npeyLl5cXBAIBVzQDwJQpUwAAlZWViIiIwKVLl7jHulpFBsOmTZvwxBNPcH+/+/paWloQFRUFpVKJ\nI0eOgM/nW+V8d54A4ObmBjc3N1RWVvZ4LlteL5VK1W3VitraWgDo9TVVqVRgWRa1tbVWY7RaLdea\noVKpuq10YTabodPpeo3r5uYGPp/Pnb+3fAfje4AQQsjoRa0ahPwkNjYWR48eRWpqKnx9fa0emzt3\nLgQCAVJSUrhjxcXFqKysRFhYmM3nWLhwIUwmE8rKyrhjRUVFYBgGfn5+4PF48Pf3576USuX9X9hP\nlEqlVeyuHubm5masXLkSEokESUlJEIlE3XLuyrNLfX096urq4Ofn1+O5+nq9FixYAAAICwvDlStX\nrFatOH36NBQKBTfr/nNqtRoqlcoqblNTE7Kzs63iNjQ0IC8vjxuTkpIClmURGhraY1yhUIi5c+da\nxWVZFikpKVzcwfoeIIQQMorZ6aZEQkaUTZs2sUqlkk1PT2c1Gg331dbWZjXmgQceYFNTU9kff/yR\nXbBgAbto0SKrOKWlpWxeXh67ceNGNigoiL148SJ78eJF1mg0sizbuVLDvHnz2KVLl7J5eXnsjz/+\nyM6fP59dtWpVn/kZjUb24sWLbF5eHuvt7c2+9NJL7MWLF9nS0lJuTEtLCzeGYRj2z3/+M3vx4kW2\nsrKy17jNzc1saGgoO3PmTPbGjRtW1961qgbLsuwjjzzCTp8+nf3hhx/YK1eusL/4xS/Y6dOnsyaT\nqc/XtK/Xy2w2szNmzGBXrVrFXrp0if32229ZDw8P9tVXX+XG3Lp1iw0KCmJzcnK4Y7t372ZdXFzY\npKQk9vLly+wvf/lLdtKkSazBYODGPPzww+zcuXPZ8+fPs+fOnWMnT57M/vrXv7bKLzAwkP3mm2+4\nv3/xxResWCxmExMT2cLCQvb3v/896+Liwmq1WpuviRBCyNhGhTMhbOfybTwer9tXYmIiN6a9vZ2N\njY1lXV1dWUdHRzYmJoatra21irN06dIe41RUVHBjampq2JiYGFYul7NeXl7s+vXrWZ1O12d+5eXl\nPeYYERHBjTl79myPY5577rle4549e7bb+K4Yd+fc3NzMbtiwgXVxcWHd3NzYmJgYtqqqqs+cbXm9\nKisr2ejoaFYmk7EeHh7sSy+9ZFWwl5eXszwej01LS7N63uuvv856eXmxEomEXblyZbfl/XQ6HfvU\nU0+xcrmcVSqV7IYNG1i9Xm815ufvL8uy7P79+1k/Pz9WLBaz8+fPtyrYbb0mQgghYxfDsixr71lv\nQgghhBBCRjrqcSaEEEIIIcQGVDgTQgghhBBiAyqcCSGEEEIIsQEVzoQQQgghhNiACmdCCCGEEEJs\nQIUzIYQQQgghNqAttwkhZIixnWvmc3+++3hP/+0NwzA9Hrv7q7dxhBBC7h8VzoQQMgAsy8JisXBf\nRqMRRqMRJpMJZrMZLMvCbDbDYrHAbDZzXx0dHejo6IDJZILJZOL+3NHRAaPRaPVfAODxeD1+8fl8\nMAzD/dnBwQFisRgODg4QiUQQiUTg8/m9Pl8gEEAkEsHBwYEb9/PCmxBCSM+ocCaEkJ/cXRQbjUa0\ntbWho6MDZrOZK3jb29vR1NSE+vp66HQ6tLa2oq2tjfsyGo32voweMQwDiUQCqVQKmUwGmUwGpVIJ\npVIJJycnSKVSCAQC8Pl8CAQCCAQCiMViiMViCAQCqwKbEELGK9o5kBAyrnQVx2azmSt2u2aDW1tb\nUV9fjzt37uDOnTtoaWlBc3Mz9Hq9vdMeVgzDwNHREQqFAi4uLvD09ISrqyscHR0hFAq5wloikUAi\nkXCz4FRUE0LGOiqcCSFjUleBbDQa0dzczLVCGAwG1NfXo7q6GlqtFvX19WhpabF3uqMKj8eDXC6H\nUqmESqWCj48PnJ2d4eDgAIFAAJlMRgU1IWRMosKZEDLqdfUVt7W1oaWlBR0dHWhra0NtbS3Ky8tR\nW1sLnU4Hi8Vi71THPEdHR7i5ucHHxwc+Pj5QKBRc77WTkxMcHBy4tg9CCBltqHAmhIwqXUWyXq+H\nXq+H0WiEXq/HrVu3UF5eDq1WSzPII5BCoYC3tzcmTZoET09Prn/ayckJQqEQPB6tjkoIGfmocCaE\njGgsy8JkMqGxsRHt7e1obW3FzZs3UVJSgpqaGhgMBnunSO4Bj8eDm5sbfH194e/vD2dnZ4hEIshk\nMjg6OnJtHoQQMpJQ4UwIGVG6+pIbGxthMBjQ1NSE69evo6ysDFqtltotxjCRSASVSoWAgACo1WpI\npVJIpVLI5XIqpAkhIwIVzoQQu2JZFh0dHWhoaEB7ezvq6upw7do1VFZWoqGhwd7pETvi8Xjw8vLC\n5MmToVarudlomUxGhTQhxC6ocCaEDKuu1S5aWlq4r9LSUhQWFqKurs7e6ZERTCAQwMfHB1OnToWP\njw+kUimUSiV3wyEhhAw1KpwJIUOOZVkYDAY0NDSgra0NNTU1yM/PR2VlJUwmk73TI6OUVCrF5MmT\nMX36dCgUCjg5OXH90YQQMhSocCaEDAmWZdHW1oaGhgY0Nzfj6tWruHbtGpqamuydGhmD+Hw+/Pz8\nMGvWLHh6enI7I1JLByFkMFHhTAgZNBaLhSuWm5qacOXKFRQWFqKtrc3eqZFxxsPDAzNmzOB6o11c\nXCAUCqmIJoTcFyqcCSH3xWKxoLW1FY2NjWhoaMClS5dQVFREy8SREUOhUGD27NkIDAzkimiaiSaE\n3AsqnAkhA9bVs1xfXw+dTofc3FyUlJSgo6PD3qkR0id3d3fMmzcPDzzwAJycnKBQKKgnmhBiMyqc\nCSE2M5lM0Ol0aG5uRn5+PvLy8qgNg4xaPj4+CAkJgbe3N+RyOZycnGh1DkJIn6hwJoT0qWvpuMbG\nRlRWViI7OxtardbeaREyaHg8HtRqNcLCwuDq6gpXV1fqhyaE9IgKZ0JINyzLwmg04s6dO9DpdMjO\nzkZpaSnt2kfGPEdHR4SEhCAwMBAKhQJyuZxmoQkhHCqcCSEci8WC5uZmNDY2orCwENnZ2WhtbbV3\nWoQMO4Zh4O/vjwULFsDFxYVmoQkhAKhwJoQAMJvN3OxyZmYmioqKQP9rIKSTk5MTQkNDMXnyZCiV\nSjg5OVEBTcg4RYUzIeNUVztGXV0dqqurkZaWRlteE9IHhmEwZcoULFy4EM7OzlAqldTGQcg4Q4Uz\nIeMMy7LQ6/Wor69HQUEBsrOzaWUMQgbI19cXS5cuhaurK9zc3GhJO0LGCSqcCRknWJZFS0sL6uvr\ncf78eeTl5dHNfoTcJzc3N0RERMDLywvu7u4QCoX2TokQMoSocCZkjGNZFk1NTVz/8uXLl6l/mZBB\n5ujoiPDwcPj7+8PNzQ0ODg7UB03IGESFMyFjFMuyaGxsRH19Pb7//nsUFBRQwUzIEBOJRFi0aBGC\ng4Ph4eEBkUhk75QIIYOICmdCxhiWZdHQ0ID6+nqkp6ejqKjI3ikRMu6IxWIsXboUDz74IDw8PKiF\ng5AxggpnQsYIlmXR3NyMuro6pKamoqSkxN4pETLuyWQyLFu2DGq1Gh4eHhAIBPZOiRByH6hwJmSU\nY1kW7e3tuH37NrKyspCXl0ctGYSMME5OTlixYgUmTpwIDw8PWoWDkFGKCmdCRjGj0QitVov8/Hyc\nO3cOJpPJ3ikRQvqgVCoRFRUFHx8fuLm50TrQhIwyVDgTMgqZzWZotVqUlZXhu+++o3WYCRllfHx8\nsHr1ari7u9NOhISMIlQ4EzKKsCyLO3fuoLq6GidPnoROp7N3SoSQ+zB37lzMnz8fHh4ecHBwsHc6\nhJB+UOFMyCjAsiza2tqg1Wpx6tQplJWV2TslQsggcXBwQGRkJCZNmkT9z4SMcFQ4EzLCdbVl5Ofn\n4+zZszCbzfZOiRAyBFxdXREdHQ2VSgVnZ2dq3yBkBKLCmZARqmsDk5qaGiQlJVFbBiHjRGBgIJYv\nXw6VSkUbqBAywlDhTMgIZDQaUVtbi/T0dFy5csXe6RBChplIJEJ0dDS3hTfNPhMyMtA6OISMICzL\nQqvVIjc3Fx988AEVzYSMU0ajEf/4xz9w5MgRVFZWwmAw2DslJCQkICQkBHK5HJ6enli7di2Ki4ut\nxhgMBmzZsgVubm5wcnJCTEwMtFqt1Zj4+HjMmzcPYrEYc+bM6fV8e/fuRWBgIMRiMSZOnIiEhIQ+\n8ysoKEBMTAzUajV4PB7eeeedbmMyMjKwZs0a+Pj4gMfjISkpqd/r1ul0iIuLQ1BQEGQyGfz8/PD8\n88+jqanJahyPx7P64vP5+PLLL/uN/dRTT0GhUMDZ2RkbNmyAXq+3GnP58mWEh4dDIpHAz88Pe/bs\n6TdnW96HmzdvIjo6GjKZDCqVCi+99BIsFku/sXfs2AFvb29IpVKsWLECpaWlA76m0YwKZ0JGCKPR\niJs3b+LIkSM4evToiPhBSQixr7KyMnzwwQfIzc2FVqu16+ZGGRkZ2Lp1K7Kzs5GcnIyOjg6sXLnS\najnM+Ph4HD9+HF9//TXS09NRXV2NRx99tFus9evX48knn+z1XHFxcfjoo4+wb98+FBUVISkpCSEh\nIX3m19raioCAAOzevRteXl49jtHr9Zg1axb2799v8yx+dXU1ampqsG/fPuTn5yMxMRHffvstNmzY\n0G1sYmIiamtrodFoUFNTg0ceeaTP2OvWrUNhYSFSUlJw/PhxpKenY+PGjdzjzc3NiIqKglqtxoUL\nF7Bnzx7s3LkTH3zwQZ9x+3sfLBYLVq9eDZPJhKysLCQmJuKTTz7Bjh07+oy7e/duvPvuuzhw4ADO\nnz8PmUyGqKgoGI1Gm69ptKNWDULsjGVZ6HQ6lJWVISkpiQpmQkiPJk6cyN08KBaL7Z0O6urq4OHh\ngfT0dCxatAhNTU1wd3fH559/jrVr1wIAioqKMGXKFGRlZXUrfN944w0cPXoUFy5csDpeWFiImTNn\noqCgAJMmTbqn3NRqNbZt24a4uLhex/B4PHzzzTdYs2bNgOMfPnwYTz/9NPR6PbeJzUDjXbt2DcHB\nwcjNzcXs2bMBAKdOnUJ0dDSqqqqgUqnw3nvv4bXXXoNGo+G2a3/55Zdx9OhRFBQU9BjXlvfh5MmT\nWLNmDWpqauDm5gYAOHDgALZv347bt2/3ujW8t7c3XnzxRWzbto07l6enJxITE/H444+jsLAQU6dO\n7fOaRjuacSbEjsxmM27duoUTJ07gq6++oqKZENKrmzdv4oMPPkBOTg5u375t19lnAGhoaADDMHBx\ncQEA5ObmwmQyYfny5dyYwMBA+Pr6IjMz0+a4x44dQ0BAAJKSkuDv7w+1Wo3f/e53I+oG6YaGBsjl\n8m47P27ZsgXu7u4IDQ3Fxx9/3GeMzMxMODs7cwUmAERGRoJhGGRnZwMAsrKyEB4eblXIRkVFoaio\nCI2NjQCAtLQ08Hg8VFZWArDtfcjKysL06dO5orkrbmNjI65evQoAqKioAI/HQ3p6OoDOTz80Go1V\nXLlcjtDQUKu4/V3TaEeFMyF20LViRlFRET755BPk5+fbOyVCyChgMplw/PhxfPPNN6iqqrLb8pQs\nyyI+Ph6LFi1CcHAwAECj0UAkEkEul1uN9fT0hEajsTn2jRs3UF5ejsOHD+PQoUNITExEbm4uHnvs\nsUG9hntVV1eHXbt2dWs/+NOf/oQvv/wSycnJiImJwebNm/Huu+/2Gkej0cDDw8PqGJ/Ph4uLC/d6\naTQaeHp6Wo3p+nvXGKlUiqCgIAiFQu54f++DLXGFQiGCgoIglUq54wzD9Pi8u+P2d02jXc9z8YSQ\nIWOxWKDRaJCdnT1mfgMnhAyv69ev4+OPP8Zjjz0GlUoFpVI5rCtvbN68GQUFBTh37ly/Y1mWHVBu\nFosFRqMRBw8eREBAAADgww8/xNy5c1FSUgKxWMwV6wzD4JVXXsH27dvv7UJ+JiEhAW+++SYXu6Cg\nABMmTOAeb25uRnR0NKZNm4bXX3/d6rl//OMfuT/PnDkTLS0t2LNnD2JjYweUQ3+vV9cnDV1jHnro\noV7bNgYSt0vXGG9v70GLO9DvgZGMCmdChpHRaER1dTUOHz6Muro6e6dDCBnFpkyZArFYisobbTD7\n8uDqJh+W4iQ2NhYnTpxARkYGvL29ueMqlQpGoxFNTU1Ws51arbbbLGVfvLy8IBAIuKIZ6LxWAKis\nrERERAQuXbrEPdbVKjIYNm3ahCeeeIL7+93X19LSgqioKCiVShw5cqTfHR5DQ0Oxa9cuGI3GHtfj\nVqlU3Va6MJvN0Ol0XC+wSqVCbW2t1Ziu5/T2mtryPqhUKuTk5Fg9r+s8fcVlWRa1tbVWY7RaLdea\n0dc1DeR7YCSjVg1ChkFXa0ZhYSE++ugjKpoJIfdMpVLhued+iwf9Q/Fdkh5Z6Q1IO61FTbUOJpNp\nSM8dGxuLo0ePIjU1Fb6+vlaPzZ07FwKBACkpKdyx4uJiVFZWIiwszOZzLFy4ECaTCWVlZdyxoqIi\nMAwDPz8/8Hg8+Pv7c19KpfL+L+wnSqXSKnZXD3NzczNWrlwJiUSCpKQkmzamycvLg7Ozc69jw8LC\n0NDQgLy8PO5YSkoKWJblbqQMCwtDenq6VUvO6dOnERgYCIVC0WPcvt6HBQsWcHGvXLli9bPo9OnT\nUCgU3Gz+z6nVaqhUKqu4TU1NyM7Otorb2zWFhob2/EKNMvydO3futHcShIxlXWszZ2dn4/jx47Rl\nNiHknojFYqxevRpzZoci51wHbhQbgJ/uD2zVW1BR2gqFCyBy4EEoFAz67PPmzZvx6aef4vDhw/Dy\n8oJer4der4dAIIBAIICDgwNqamqwf/9+zJo1C3fu3MEf/vAH+Pn54bXXXuPiXL9+HZWVlTh79ixK\nS0uxePFi1NbWwtXVFXw+H/7+/jh27BhOnjyJWbNm4datW9i8eTOmTp3a5yoZHR0dyM/Ph0ajwcGD\nB6FWq+Hh4QG9Xs/NSuv1ehQWFkKj0eDAgQMICQmBRCJBR0dHr4VoS0sLVqxYgba2Nnz66acwm83c\ntctkMjAMg2PHjuHcuXNwcHBAU1MTvvjiC+zYsQP//u//jiVLlvQY183NDdnZ2fj8888xe/ZslJeX\n4w9/+ANWrVqFZ555BgAwefJkvPfee8jPz0dgYCDOnDmDP/7xj/iv//ovbg3snJwcREZG4tFHH4WT\nk1Of78Orr74KAPD398eRI0eQnJyMGTNm4OLFi4iLi8OmTZsQGRkJoHMZvpCQEMyfP5+beTebzUhI\nSEBwcDCMRiPi4uJgMBjwzjvvgM/n23RNox0tR0fIEDKbzaipqcGxY8dQXl5u73QIIaPUnDlz8NBD\nISgtsOBGcXuv4xgGmPmQHJMC5XBxdRrU4pnH4/UY7+OPP+aKIoPBgBdeeAGfffYZDAYDVq1ahf37\n91vdMBYREcGt1HC3srIybhZbo9Fg69atOH36NGQyGVavXo29e/f2ObtcUVEBtVrdLcclS5bgzJkz\nADpXoIiIiOg25tlnn8VHH33UY9y0tDQsW7bM6lhXz25XzqdOncLLL7+M69evg2VZTJo0CZs3b+5x\nree7NTQ0IDY2Fv/85z/B4/EQExODt99+m7shDwCuXLmC2NhY5OTkwM3NDXFxcXjhhRe65Xf362fL\n+3Dz5k1s2rQJZ8+ehUwmw29+8xskJCRws+wVFRXw9/dHamoqwsPDueft3LkT77//PhoaGrB48WLs\n37/fatlAW65pNKPCmZAh0trailu3buHLL79Ec3OzvdMhhIxCnp6eePjh1TC0inE+Qw8bNnYDAEzw\nE2PeAme4eyi6LZlGCLl3VDgTMsi6+pmvXbuGf/7zn9SaQQgZMAcHB6xcuRIqz4nIOdeORt3Ae5cV\nzgKEr3CFp0rZ741shBDbUOFMyCBiWRZ37txBTk4O0tLS7J0OIWQUmj17NkJCQnH9Govr19r6f0If\nxBIelka5QeUth4ND/ze0EUL6RoUzIYOka5me1NRUq6WSCCHEFu7u7oiO/gWMbRJkp7fY3JbRHz4f\nWBzpCp+Jcjg6SQYnKCHjFBXOhAyCrk1NkpKS6CZAQsiAODg4YMWKFfBS+d5zW4YtQhYr4T9JDrlC\nNmY2oyBkuFHhTMh9MpvNuHXrFr766ivcvn3b3ukQQkaRmTNnYv78MJQVsSgpvL+2DFtMm+2EKdMV\ncHYZ3BU3CBkvqHAm5D4YjUZUVVXhs88+o5UzCCE2c3NzQ3T0L9BhkCI7bfDaMmzhP1mKWQ85w819\neHYaJGQsocKZkHtkMBhQUVGBQ4cOwWg02jsdQsgoIBKJsGLFCnh7+yEnY+jaMvrj5eOA+eEucPdU\nUvFMyABQ4UzIPTAYDCgvL8ff//53KpoJITaZMWMGwuYvQHkpi+KrQ9+W0R9XdyEWR7rBg4pnQmxG\nhTMhA0RFMyFkIDrbMqJhMsiQndECs30mmXvk6iHE4uVUPBNiKyqcCRmArqL50KFD6OjosHc6hJAR\nTCQSYfnySEyY8AByv2+H7s4Iqpjv4u4pxKJlbtS2QYgNqHAmxEZUNBNCbDVt2jQsXLAIFddZFOXb\nvy2jP+4qERZGuNIW3YT0Q2DvBAgZDQwGA8rKyvD3v/+dimZCSK9cXV0RHR0Ni8kRKcdHVltGT4Qi\nBpODZZigFqO4rgXtDAtfD2eaeSakFzTjTEg/Ojo6UFZWhoMHD1LRTAjpkVAoxPLlkfCdqMaPI7gt\no4tcIcCC7iqIAAAgAElEQVTU2Y5wdBbgdPkdJObWwGQBFvgp8MJSf0xwp7YNQnpCn8cQ0gez2Yyq\nqip8+umnVDQTQno0depUPPfcejAmX3yX1DJii2aGAXx8HbD8F66YHu6ED0tq8MRX+fgwp7NoBoAf\nKhrx3g8VqK5rxHDNqyUkJCAkJARyuRyenp5Yu3YtiouLrcYYDAZs2bIFbm5ucHJyQkxMDLRardWY\n+Ph4zJs3D2KxGHPmzOn1fHv37kVgYCDEYjEmTpyIhISEPvMrKChATEwM1Go1eDwe3nnnnW5jMjIy\nsGbNGvj4+IDH4yEpKanf69bpdIiLi0NQUBBkMhn8/Pzw/PPPo6mpqcfx9fX1mDBhAvh8fq9j7o79\n1FNPQaFQwNnZGRs2bIBer7cac/nyZYSHh0MikcDPzw979uzpN2db3oebN28iOjoaMpkMKpUKL730\nEiw2LFS+Y8cOeHt7QyqVYsWKFSgtLR3wNQ0HKpwJ6UXXjoCffvop2tvb7Z0OIWSEcXFxwdNPP4M5\ns5Yg9Xgrrl0Zmb3MQhGDqbMcsfIRN4gnCRCfXILn/lGIHyoaexz/XUk9DuZWQdswPJs6ZWRkYOvW\nrcjOzkZycjI6OjqwcuVKtLX96/WMj4/H8ePH8fXXXyM9PR3V1dV49NFHu8Vav349nnzyyV7PFRcX\nh48++gj79u1DUVERkpKSEBIS0md+ra2tCAgIwO7du+Hl5dXjGL1ej1mzZmH//v02z9RXV1ejpqYG\n+/btQ35+PhITE/Htt99iw4YNPY5fv349Zs2aZVPsdevWobCwECkpKTh+/DjS09OxceNG7vHm5mZE\nRUVBrVbjwoUL2LNnD3bu3IkPPvigz7j9vQ8WiwWrV6+GyWRCVlYWEhMT8cknn2DHjh19xt29ezfe\nffddHDhwAOfPn4dMJkNUVJTVylX9XdNwoVYNQnrAsiya7tyEuaMdVbUNOPVdGu7cuWPvtAghI4BQ\nKMSyZcvg5xuAC5kG3Lk9Mj+NkisFmDqrsx3jVNkd/N+Ff80s2+KFJX5YFewFhaN06JLsQV1dHTw8\nPJCeno5FixahqakJ7u7u+Pzzz7F27VoAQFFREaZMmYKsrKxuhe8bb7yBo0eP4sKFC1bHCwsLMXPm\nTBQUFGDSpEn3lJtarca2bdsQFxfX6xgej4dvvvkGa9asGXD8w4cP4+mnn4Zer7e6SfO9997DV199\nhddeew2RkZHQ6XSQy+U9xrh27RqCg4ORm5uL2bNnAwBOnTqF6OhoVFVVQaVS4b333sNrr70GjUYD\ngaDzdreXX34ZR48eRUFBQY9xbXkfTp48iTVr1qCmpgZubm4AgAMHDmD79u24ffs2d66f8/b2xosv\nvoht27Zx5/L09ERiYiIef/xxFBYWYurUqX1e03ChGWdCfoZlWegbNHC4+TmcSt7Eg+xZ/P7JCGzb\n8hwWLlxAd5wTMo4FBwfjued+C75Fje+SWkZc0cwwgI+fA5ZFu2L6Yid8UNzZjvHRjwMrmgFgX3oF\nrlTVo32Y16tvaGgAwzBwcXEBAOTm5sJkMmH58uXcmMDAQPj6+iIzM9PmuMeOHUNAQACSkpLg7+8P\ntVqN3/3ud9DpdIN+DfeqoaEBcrnc6udMQUEBdu3ahYMHD9r08yczMxPOzs5cgQkAkZGRYBgG2dnZ\nAICsrCyEh4dbFbJRUVEoKipCY2PnJxFpaWng8XiorKwEYNv7kJWVhenTp3NFc1fcxsZGXL16FQBQ\nUVEBHo+H9PR0AEBZWRk0Go1VXLlcjtDQUKu4/V3TcKFVNQj5mbbmO+BpTkHQ3PmPXNh8FcLmq5AI\nnBDx4GIsmvcstPWtOJWSgerqajtnSwgZDs7OzoiO/gUY1gmpJ1pgGln1stXqGKWNbdiWUoLalvsr\neC0s8Mq3pdi/VojgCW7g8/mDlG3vWJZFfHw8Fi1ahODgYACARqOBSCTqNsPq6ekJjUZjc+wbN26g\nvLwchw8fxqFDh2AymRAfH4/HHnsMycnJg3od96Kurg67du2yaj8wGo1Yt24d9u7dCx8fn259vz3R\naDTw8PCwOsbn8+Hi4sK9XhqNBv7+/lZjPD09uccUCgWkUimCgoIgFAq54/29DxqNhovTU9yZM2dC\nKBQiKCgIUqmUO84wTI/Puztuf9c0XKhwJuQuhrYmWOpyIb6d2u0xnqkZYu0JOGhPQCINwLO/XA6j\nYBUuFZThbFoGTKaReUMQIeTeCYVCRERE4IEHJiEv04A6bYu9U7Ly83aMF78uHfDMcl/aOizYfqIY\nb/+SjwAv1yFfaWPz5s0oKCjAuXPn+h3LsuyA8rFYLDAajTh48CACAgIAAB9++CHmzp2LkpISiMVi\nrlhnGAavvPIKtm/ffm8X8jMJCQl48803udgFBQWYMGEC93hzczOio6Mxbdo0vP7669zx7du3Izg4\nGP/2b/8GANwNm/fSZdvf69UVs2vMQw891GvbxkDiduka4+3tPWhxB/o9MBiocCbkJx0dRhjrSyCu\n+qzPcQwAQet1CCquQ8qTYIHPfMzdvA4NejOSz2bh+vXrw5MwIWRIBQVNQXh4OG6VM/ju6MgpmBkG\n8PZ1QNB0R7QzFryfewuZlX2vsnA/tC0d+O/k63gjioeJQ7jGc2xsLE6cOIGMjAx4e3tzx1UqFYxG\nI5qamqxmO7VabbdZyr54eXlBIBBwRTMATJkyBQBQWVmJiIgIXLp0iXusq1VkMGzatAlPPPEE9/e7\nr6+lpQVRUVFQKpU4cuSI1cx+amoq8vPz8dVXXwHoLBRZloW7uzv++Mc/WhXZXVQqVbeVLsxmM3Q6\nHdcLrFKpUFtbazWm6zm9vaa2vA8qlQo5OTlWz+s6T19xWZZFbW2t1RitVsu1ZvR1TQP5HhgMVDgT\ngs7/GbU13IKk7H/BwPbf5BlLG0R1qRDVpUIqnoAnIiNhjF6Ca9erkXImzequcELI6ODs7Izo1dHg\nMfIR1Zbxr3YMCUobWxGffP/tGLbKr9Xjw/NV2LSQD5WLYtDjx8bG4ujRo0hLS4Ovr6/VY3PnzoVA\nIEBKSgp3U1pxcTEqKysRFhZm8zkWLlwIk8mEsrIyqNVqAJ03tzEMAz8/P/B4vG7tC4NFqVRCqVR2\nO961uoVEIkFSUhJEIpHV40eOHLH6OXL+/HmsX78e586d6zXXsLAwNDQ0IC8vjys8U1JSwLIsdyNl\nWFgYXn31VZjNZq5QP336NAIDA6FQ9Pz+9vU+LFiwgIv75ptvoq6ujutzPn36NBQKBTeb/3NqtRoq\nlQopKSmYMWMGgM6bA7Ozs7Fly5Z+ryk0NLTHuEOFVtUg417XzYCC6+9D0Hrj/uMxInQo58LgshB6\nIx9nMn606WMpQoh9CQQCLF26FP7+k5GXZUBd7ciomOXKnzYrUQrw7Y07OJg38Bv9BstLSx/A6mne\nkEnEgxZz8+bN+Oyzz5CUlITJkydzxxUKBcRiMTfm5MmT+Pjjj+Hk5IS4uDjweDxkZGRw469fv47m\n5mb87W9/Q1paGj7//HMAnTd0CoVCrnB0dHTEn//8Z5jNZsTGxkKpVOLkyZO95tfR0YGCggKwLIvo\n6Gj8+te/xrp16+Do6MjNXuv1epSWloJlWcyZMwf79u1DREQEXFxcMHHixB7jtrS0IDIyEu3t7fjH\nP/7B9fwCgLu7e483AqalpSEiIoK7ibA3q1evhlarxXvvvQej0Yjf/va3CAkJwcGDBwF0FqZBQUFY\nsWIF/vM//xNXrlzB+vXr8fbbb2P9+vUAgJycHDzzzDM4c+YMtwxff++DxWLB7Nmz4e3tjd27d6Om\npgbPPPMMfv/73+NPf/oTgM5l+JYvX46DBw9i3rx5AIC33noLu3fvxieffIIHHngAr732Gq5evYqr\nV69yv0z0d03DhQpnMu61tzbBfCsZDjVHBz222cEDRrflMEr8UV5Vh1PJZ/tduJ4QMvwCAwOxZMlS\n1FQyyM9rtXc6w96OYSshj8F7j07BdF+PQVthiMfj9dj+8fHHH+OZZ54B0LnxxgsvvIDPPvsMBoMB\nq1atwv79+61uGIuIiOBWarhbWVkZN4ut0WiwdetWnD59GjKZDKtXr8bevXt7nA3uUlFRAbVa3S3H\nJUuW4MyZMwD+VdD+fMyzzz6Ljz76qMe4aWlpWLZsmdWxrp7du3Pu6Tl9LUcHdK7OERsbi3/+85/g\n8XiIiYnB22+/bVWcX7lyBbGxscjJyYGbmxvi4uLwwgsvdDvX3bnY8j7cvHkTmzZtwtmzZyGTyfCb\n3/wGCQkJ3PdLRUUF/P39kZqaivDwcO55O3fuxPvvv4+GhgYsXrwY+/fvt1o20JZrGg5UOJNxzWw2\no1VbAHHJ/wyoRWOgWIYPk3wGDK5L0GaR4Fz2FeTm5g7Z+QghtlEqlYiOjoaAp0Dm2RZ0DO/Ka93c\n3Y5R0tCKtzNvDls7hq28nER4+5dT8IDKhbblJuMOFc5k3GJZFi13KuFQ/D/gmXreQWsoWIRKGF2X\nwug0FdXaRpxKTsft27eH7fyEkM62jCVLlmBSQCAuZhuh1di3OB1J7Ri2WPmgC+KW+MPTufdZT0LG\nIiqcybilb6wFv+wQBE2X7XJ+FgxMjlNgdIuAgadAzsVifP9DJiyWEfzTkpAx4MEHJyMiIgI1Nxnk\nX7BfW8bd7RhtjAX/O0LaMWz136sCsCTQGw4iob1TIWTYUOFMxiVjux7Gmh8gvnnI3qkAACx8R3S4\nLoZRPht1jW04lXIOVVVV9k6LkDFFoVAgOjoaIoESmaktGOYN8Th3t2MUN7TinRHYjmELqZCH92OC\nMdnHnVo2yLhBhTMZdywWC1puF0NS/BYY1mzvdKywAMxSfxjdl8Mg8MDlwnKkpZ+D0V4/4QkZA/h8\nPpaEL8GDDwbh4nkjtDX2+fc02toxbDHTyxE7oyZjgnvvN9cRMpZQ4UzGnRbdLQiL3wHfMLzbdA4U\nyxOjw2U+DMpQNOrNSE7Ltmm7VULIv0yaNAnLIpZBc4uHK7nD35bBMICPrwMCpzuiDRa8f+EWskZR\nO4YtXl2uxqppE6hlg4wLVDiTccVobIfxVjrEN/9u71QGxCz2gcE9EkaHCSi+UYPkM2lobbX/klmE\njFRyuRzR0dFwEDoj82wrjIbhndodK+0YtnAU8fF+TDAmebtRywYZ86hwJuMGy7JovlMBybU3wVgM\n9k7nnrCM8KfNVRah1chH6vcXkJ+fb++0CBkx+Hw+Fi8OR+DkKbicY4SmeniL1bHYjmGL1UGuiAsP\ngKvC0d6pEDKkqHAm40abXgeUfQFhfaa9UxkUFpEHDG4R6JBOQnn1HZz67iwaG4dvWT1CRpqAgAAs\nW7Yc2moeLv84fJ/I/Lwd40DuLWTfHFvtGLZ471dTMEftOWgboxAyElHhTMYFi8UCvfYaxMW7MdY+\nSGTB69xcxW0J2i1SfJ+Tj5ycH+2dFiHDxsnJCdHRv4DEobMtw9A+PFO8P2/HeDvzJrRjtB3DFr5K\nMfatCYKfp4u9UyFkyFDhTMYFva4GgpK/gt9+y96pDKnOzVWWwOg0DTW3m3AqOR1ardbeaREyJPh8\nPhYtWoSgoGBc/tEETdXwtGDd3Y5x8sYdHBon7Ri2iF/ki7WzJ0IqdrB3KoQMCSqcyZjX0WGE4db3\nEFd+Yu9Uhk3n5iqBMLotg4GnxI+XSnDu+x9ocxUyZvj7+2P5skjcruXj0nn9kJ+P2jFs48Bn8OHj\nU2ltZzJmUeFMxrSubbXFRf8fGPP4XIXCwpehw3URjPI5uNPYjtOpP6CystLeaRFyT5ycnLB6dTRk\nEhdkpraifYjbMkQiBg8GO2KCWozihlb85YdK3NZ3DOk5R7uHA12xLWISnJ1k9k6FkEFHhTMZ09pb\nG2EpPwJR3Vl7p2J3LACL9AEY3FfAIPBAflEFUs9m0OYqZFTg8XhYuHARgoOnIj/XhOqbQ9uWoVAK\nEEztGPeExwAfPT4VwRM9aNaZjDlUOJMxi2VZNNeVQVL4JzCgn3h3Y3kO6HAOhUE5H01tFiSnnUdJ\nSYm90yKkR2q1GsuXR+JOrQAXh7Atg9oxBs9itRIvRz4Id6WTvVMhZFBR4UzGLENbC8xlhyGqS7V3\nKiOaWewNg1skOsQTUVxWg+Qz6dDrh75nlJD+ODo6YvXqaDjKXDvbMtqG5hdgq3YMXSv+kkntGAMl\n5DEIcJVg3gQ55njJ4CLmwZHPwtdvIs06kzGFCmcyJnGbnRTuAsPSD0BbdG6uMgdGl0XQdwiQ9sNF\nXL582d5pkXGIx+NhwYIFmDp1+pC2ZVA7xr1zlggwzVOG+b5yTHIRQ8ZjIbp9C5IL6ej44QyguwP+\nvEVQxL0KmaubvdMlZNBQ4UzGJKOhHR0Vx+GgSbJ3KqOSReQGg9sydEgfRGVNPU4lp0Gn09k7LTIO\n+Pn5YcWKlai/LUBe1uB/8kHtGAPHY4AHnCWY4+OEh3wc4SEVwKGjDZLr+eBlnYH5YjbQy4o9jns/\ngTJoKs06kzGDCmcyJjXXVUJ87U0wljZ7pzKqdW6uMg0G16VoZ2X44ccCnD9/3t5pkTHI0dERqx9e\nDUdHN2SdbUVb6+BO/VI7hu0cRXxM8eicTQ52l8CRD4h0Wkgufw/TuRSg1vb18AULl0GxaTukzrQp\nChkbqHAmY05HRwcMVWcgvvmpvVMZUywCBYxunZuraOpacCo5HbW1tfZOi4xyPB4PYWFhmDZtJgry\nOlBVMbhtGQqlAFNnO0Gq5ONbasfohgEwQeGAmV5OCJ3oBG9HAcSWDkgqi8DPToX5x3PA/ay8w/Dg\n+D+fQDl5Cs06kzGBCmcy5rTUV0F07S3wTI32TmVMYsHAJJsMo/tyGHhK5F4pxfffZ8JkMtk7NTLK\n+Pr6YuXKlWioEyI3c/DaMqgdo3diAQ+B7lKETJBjhkoKhZCBsKUB0vxssD8kw1JeOujnFK5aC+f1\n8XCQ0rrOZPSjwpmMKWazGW23siAuf9/eqYwLFr4MHS4LYVTMRX2TAafOfE+bq5B+yWQyPPzww5A7\neSA7rRWt+sGZAhaJGEye6gifB6gdo4unowjTVTKETZTDT+kACWuCQ005RDlnYco8A7QNw8ZQEimc\n9n4M5QMBQ38uQoYYFc5kTNHraiAo/jP4BmohGE4sAIvEDwb3SBiFKlwtvokzZ9NhMAztJhVkdGEY\nBvPnz8eMGbNQkGcatLaMrtUxZEoBvr1+B4cujs92DCGPwSQ3Ceb5yDHHWwYXBx6EbS2QFV3onE2+\ndsVuuUlf2QPnsCXg8Xh2y4GQwUCFMxkzWJZFS20hJMW77Z3KuNa5uUoIDMowNLezSE47j+LiYnun\nRexs4sSJWLkyCo31QuT+cP9tGT9vx/hbbhXO32wehExHDxeJAFM9ZQjzVWCSiwOkPBYi7U9Lwn2f\nAjTW2ztFDn/GPChe2EVL05FRjwpnMmYYDG0wX/+CNjwZQcxir582V/FDaYUG36WkoaWlxd5pkWEk\nlUrx8MMPQ6nwRNbZ+2/LGK/tGDwGULtIMMfbCQ9NcIS7pHNJOGnpFSAzBZbLOb0uCTciMDw4/uUg\nlAGT6SZBMqpR4UzGjJb6KogKE8AzU2E20rCMAB2K2TC6LEarSYS0zIu4dOmSvdMiQ4hhGISGhmLm\nzNkouGhCVfn9tWWMt3YMJwc+prh3Lgk3xV0CRz4Lke42JJfOwfR9MlBbY+8UB0y8Ph7Oa56EQCCw\ndyqE3DMqnMmYwLIsWjT5kJTstXcqpB8WketPm6tMRqVGh1PfnaXNVcYYHx8frFr1MJp0Qvz4/b23\nZYyXdgwGwASlA2Z5OSF0ghO8nYQQm42QVBSBl30G5pwMYAysWsN4TYDijb/CyWeivVMh5J5Rlz4Z\nE4yGNvDrs+ydBrEBz3gHkuqv4FSagEBeBjaui0T85t8gNDTE3qmR+ySVSvGrX/0Kq6J+ieyzpnsu\nmkUiBtNmO2HlI+4Q+PMR910xfvtN4ZgpmsUCHmZ5OeL3Id549/9NQuKvHsSeh5ywoSEL095/Gcq4\nRyDe9jjYv7wGc2bqmCiaAYCtqUKH7g6GYr4uISEBISEhkMvl8PT0xNq1a7vdW2EwGLBlyxa4ubnB\nyckJMTEx0Gq1VmPi4+Mxb948iMVizJkzp9fz7d27F4GBgRCLxZg4cSISEhL6zK+goAAxMTFQq9Xg\n8Xh45513uo3JyMjAmjVr4OPjAx6Ph6Sk/ne+1el0iIuLQ1BQEGQyGfz8/PD888+jqelfyy/W19fj\n4Ycfho+PD8RiMXx9fbF161Y0N/f970mn0+Gpp56CQqGAs7MzNmzYAL3e+t/05cuXER4eDolEAj8/\nP+zZs6ffnG15H27evIno6GjIZDKoVCq89NJLsNjQirRjxw54e3tDKpVixYoVKC21Xl7RlmvqD31e\nMso1NTXh/fffxyOPPIJJkybZOx27MervQNxwwd5pkAFgYIGw6QqETVcgFcixIjgcS+Y/B21dC75N\nyYBGo7F3isRGDMMgJCQEs2bNwbVLZuR9f2/tUgpnAabO+mmzkut38MLXJWOiHUPlKMIML0eETnSC\nn6JrSbgbEJ0/BlPWWW5JOLN90xwWxtTjMPo/OOhrOmdkZGDr1q2YN28eTCYTXn75ZaxcuRKFhYWQ\nSCQAOovikydP4uuvv4ZcLseWLVvw6KOPIiMjwyrW+vXrkZ2djcuXL/d4rri4OCQnJ2Pfvn2YNm0a\n6uvrUV/f942Yra2tCAgIwOOPP45t27b1OEav12PWrFn47W9/i0cffdSm666urkZNTQ327duHKVOm\noKKiAhs3bkRNTQ2+/PJLAJ2bDD3yyCP47//+b7i7u6O0tBSbN2+GTqfDoUOHeo29bt061NbWIiUl\nBUajEb/5zW+wceNG7jnNzc2IiorCypUrceDAAVy5cgXPPfccV5D2pr/3wWKxYPXq1fD29kZWVhaq\nq6vx9NNPQyQSYdeuXb3G3b17N959910kJiZCrVbj1VdfRVRUFAoLCyESiWy6JltQq8Yod/v2bcTG\nxuLFF1/EvHnz7J2OXVgsFug1VyAp3WfvVMh96txcZRKM7pEw8lxwIf86Ms59T5urjGDe3t5Yteph\ntDSKkHNu4DPMDAP4+IkROE02JtoxhHwGD7pKMW+CE+Z4OcLZgQdRWzOk13LB/pACS5H9loQbEaQy\nyPd8BMUQr+lcV1cHDw8PpKenY9GiRWhqaoK7uzs+//xzrF27FgBQVFSEKVOmICsrCyEh1p94vfHG\nGzh69CguXLCekCksLMTMmTNRUFBwz5NVarUa27ZtQ1xcXK9jeDwevvnmG6xZs2bA8Q8fPoynn34a\ner2+1+X//vrXv2Lv3r2oqKjo8fFr164hODgYubm5mD17NgDg1KlTiI6ORlVVFVQqFd577z289tpr\n0Gg0XN/6yy+/jKNHj6KgoKDHuLa8DydPnsSaNWtQU1MDN7fOVVgOHDiA7du34/bt2732yHt7e+PF\nF1/kfjFpamqCp6cnEhMT8fjjj6OwsBBTp07t85psQTPOI9yzzz7b5+Ndv/e8/fbb3D+QxMTEIc9r\nJDEaWiG487290yCDgAELob4EQn0JpHwpFvktQMiMX0PXYsTpMz+gvLzc3imSn0gkEqxatQquLl7I\nSmuFvnlgRbPIgcHk4M7VMYoa9Ij7rnhUro7hIhVgmqcM8yd2LQlngYP2FsS5R9HxQQrQ1ABgfMwm\n26RVD3N9HVg//yFdXaOhoQEMw8DFxQUAkJubC5PJhOXLl3NjAgMD4evri8zMzG6Fc2+OHTuGgIAA\nJCUl4d133wXLsoiMjMRbb70FZ2fnIbmWgWpoaIBcLu+1aK6ursaRI0ewdOnSXmNkZmbC2dmZKzAB\nIDIyEgzDIDs7G7/85S+RlZWF8PBwq0I2KioKb731FhobG6FQKJCWloaIiAiUl5fD19fXpvchKysL\n06dP54rmrribNm3C1atXMXPmTFRUVECtVuPs2bMIDw9HWVkZNBqNVVy5XI7Q0FBkZmbi8ccfR1ZW\nVr/XZAsqnEe49vZ2iMVizJ07t8ffstrb25GdnY0HH3zQ6ptsPOlorYe48aK90yCDjDG3wuF2MkS3\nkyGV+GLdqhUwCpehoKQKZ86mo7293d4pjksMw2DevHmYM2ceivLNuPjDwNoyutoxZEo+To6ydgwe\nA/i7SDDHxwkP+fy0JJyxFZLSK0DSGasl4UbfrwDDpyP3B3RMm819fD7YWJZFfHw8Fi1ahODgYACA\nRqOBSCSCXC63Guvp6TmgtrAbN26gvLwchw8fxqFDh2AymRAfH4/HHnsMycnJg3od96Kurg67du3C\nxo0buz22bt06HD16FG1tbVizZg3+93//t9c4Go0GHh4eVsf4fD5cXFy410uj0cDf399qjKenJ/eY\nQqGAVCpFUFAQhEIhd7y/90Gj0XBxeoo7c+ZMCIVCBAUFQSqVcscZhunxeXfH7e+abEGF8wi3ZcsW\nfPrppygrK8Ozzz6LWbNmWT2u1WqRnZ2N1atXj8tWDZZlAWMTGAvtUDdWMQD4bZWQVn4ICU+EeW4h\nmP67GLQYGKSk5+DatWv2TnHc8PLywsMPr4a+SYTvkvRgbSx4R2s7hpMDH8EenUvCBblJ4MRnIdJp\nIb6YDNNnKcDtziXhRkndP2KYfvwehujHIPKeMCTxN2/ejIKCApw7d67fsSzLDmjm22KxwGg04uDB\ngwgI6Gw3+fDDDzF37lyUlJRALBZzxTrDMHjllVewffv2e7uQn0lISMCbb77JxS4oKMCECf96DZub\nmxEdHY1p06bh9ddf7/b8v/zlL9i5cyeKiorwyiuvYNu2bdi/f/+Acujv9er6FLxrzEMPPdRr28ZA\n4nbpGuPt7T1ocQf6PUCF8wgXHh6OkJAQfP3119izZw+mTZuGZ599Ft7e3gAw7heSN5vN4Omv2zsN\nMkwYixGi+nMQ1Z+D1EGFX4VHomPlQlyvrMXp5LO0ucoQEYvFWLVqFdxcvXE+vQ3NTba1ZYgc/rVZ\nSfgVtwkAACAASURBVJFOj62ni1DXOjL71RkAE5VizPJ2ROgEJ3g5CiE2GyApLwIv5XOYfzzHrW4x\nMq9g9LBUlcPUev+7R/YkNjYWJ06cQEZGBvdzEgBUKhWMRiOampqsZju1Wm23Wcq+eHl5QSAQcEUz\nAEyZMgUAUFlZiYiICKs16rtaRQbDpk2b8MQTT3B/v/v6WlpaEBUVBaVSiSNHjoDP53d7voeHBzw8\nPDB58mS4uLhg8eLF2LFjR4/Xr1Kpuq10YTabodPpuF5glUqF2tpaqzFdz+ntNbXlfVCpVMjJybF6\nXtd5+orLsixqa2utxmi1Wq41o69rGsj3ABXOo4BYLMZTTz2F5cuX4+OPP8YLL7yAlStX4rHHHrN3\nanbXYWgBvynf3mkQO+AbNJBWHQLL8DFNPhuTf/P/0GoWIT3zEi5epNadwTJv3jzMnfsQivPNuJRp\n2y8mCmcBps52gkzBx/HSOvz7lyUjblZWIuQh0F2K0IlyTPeUQS4ARM31kORngT2WDEvlDQAAC+pP\nHgoWzS1Y/B/stQ/3XsTGxuLo0aNIS0uDr6+v1WNd7Y4pKSncTWnFxcWorKxEWFiYzedYuHAhTCYT\nysrKoFarAXTe3MYwDPz8/MDj8bq1LwwWpVIJpVLZ7XjX6hYSiQRJSUk2tcCYzWYwDAODoedPa8PC\nwtDQ0IC8vDyu8ExJSQHLslw/eFhYGF599dX/v707j4+yvNcGfj3P7DOZyWTfQ0JYwr7KJqCsQWhd\nCooCahVXRIrV0p5Tt/Z45NB68NVX3lrrUoqtWi0WXFBkCwiETfaEJUAWQib7ZDJJZn2e94+YkQiB\nCQSeTHJ9P5/5tJm5557fMxPh4p57gd/vDwT19evXo3fv3ggPD79ov5f6HMaMGRPo9+WXX0ZlZWVg\nCur69esRHh4eGM3/sfT0dMTHx2Pjxo0YOHAggKbFgbt27cITTzxx2WsaOXLkZd+zZtxVIwR99913\nWLlyJZxOJyZNmoQ1a9Z02V01nFXF0OX9F6dqEABA0kTBHT0B3rDeOGuz4+sN2aiqqlK6rJCUkJCA\nadNuQaNTj93fOi87LSMwHWOACQ2yhDf3nsWesx1nOka8WYuB8WEYlWJBargWBtkH/bnT0Oze3LQl\nnKtR6RK7FM2UWxHx2BLo9Pp26W/BggX44IMPsHbtWvTq1Stwf3h4OPTfv8aCBQuwbt06vPfeezCb\nzVi0aBFEUWyxHd2pU6dQV1eHN998E9nZ2fjwww8BAH379oVGowkEx7CwMLz66qvw+/1YuHAhrFYr\n1q1b12p9Xq8Xubm5kGUZM2bMwLx58zBnzhyEhYUFRq/r6+uRn58PWZYxdOhQLF++HBMmTEBkZCRS\nUi5+aIzT6cTkyZPhcrnw6aefBub8Ak0jzIIgYN26dSgrK8MNN9yAsLAwHDlyBEuWLEF0dDSys7Nb\nrXn69OkoLy/Hn/70J3g8Hjz44IMYMWIEVq1aBaApmGZmZmLKlCn49a9/jcOHD2P+/Pl47bXXMH/+\nfADAnj17cN9992HTpk1ISEgI6nOQJAlDhgxBYmIili1bhtLSUtx333145JFH8F//9V8AmhY4Tpo0\nCatWrQpknz/84Q9YtmwZ/vrXvyItLQ3PPfccjh49iqNHjwb+MXG5awoGg3OI8vl8+Pzzz/Hpp5/C\n5XJ1yeAsyzLqKvJhPNb6vo7UNckQ4TP3gSd6IlwwY9d3ecjZtTuoDfS7Or1ej6lTsxAXm4RdWxtR\nV3vp8dbAdIxuTbtj/J/tRYpPx9CoBPSMNuKGZDOGxDdvCeeAMW8f5B0bIJ04qmh9BAiJqQj/3esw\nt9M8Z1EULzp18b333sN9990HoOngjWeeeQYffPAB3G43pk2bhhUrVrRYMDZhwgRs3br1gn7OnDkT\nGMW22Wx48sknsX79ephMJkyfPh2vvPLKRUeDmzXvAvHjGm+66SZs2rQJAAI7UPy4zf3334933333\nov1mZ2dj4sSJLe5rnrPbXPOWLVvw29/+Fnl5eXC73UhJScHMmTPx61//+oJFeuez2+1YuHAhPvvs\nM4iiiFmzZuG1115rEc4PHz6MhQsXYs+ePYiOjsaiRYvwzDPPXFDf+e9fMJ9DcXExHn/8cWzZsgUm\nkwk///nPsXTp0sA3FIWFhejevTs2b96M8ePHB5734osv4q233oLdbse4ceOwYsWKFtsGBnNNl8Pg\nHOIaGhpQX1+P8PDwC76e8fl8OHHiBNLS0tr0SxEqJElCw9kd0Be0vjKYSFJb4IkaB495ECpq6vHV\nhm0oLS1VuqwOadiwYRg+fATyc/04feLSu5acvzvGF/mVeH+/TbHpGFFGDfrHmzAqxYIeEXoYBD90\n5cXQ79sK745NgS3hqAMRBJhf+zusGb0u35aoA2Fw7sTsdjseffRRPPfcc+jfv7/S5bQ7t6sRUv5K\naKp3Kl0KhQAZaDpcJXoyPKooHDh6GtnbvuXhKmhaNHPLLbfAVW/A7m1OtDYw3xGmY6h+tCVctEEN\nracBxpOHgJ0bIR3Zh1YvgDoU00v/DxGDb+jyi9wptHBxIIUsn6sWGid31KDgCAA09fnQ1OfDKBow\nJmUMhi+YC7vTh/Wbd+DMmTNKl3jd6XQ6ZGVlIS42Gbu3uuCovfjivwt2x/j6+u2OYfnRlnBhKhna\n6rKmLeH+8Q1Q2bTanlE59PjyDsE/YGirJ8ERdUT8baWQJfkaIXjKL9+Q6EcEqRG6yo3QVm6E0ZCC\ne7Imw6OdgGOnSrBhY3aXOFxlyJAhGDFiJE7lSdiw6+KBuXl3DGO4Cl9eh90xBACp520JFx+mgd7n\nhr7gGFQbPoB/33ZuCdeJ+E4cgc/tZnCmkMLfVgpJsiwDvgbwCz66Gk2HqxTDWPQeDIIWQyNuQL/v\nD1fZtG0v8vLylC6x3cXGxmL69BnwNOqxYW39BbMaBAFITtOjV/+m6Rh/uobTMYwaEb1jTBiZYkb/\nOCMsagFaRxWMR3ZA+mwDpOKCQFtuCdf5yBU2+JwOwGRSuhSioDE4U0iSZRmClwt+qP0Isgfa6u3Q\nVm+HUReHn42dBM+UMThdXI5vNmbD4XAoXeJV0el0mDJlKhLiU7DnWxdqa1oeQHH+dIxj1ddmOkai\nRYsBzVvCWbTQy17oS05Bk7MWvl1bAlvCcTS5a5AqyyC1so8wUUfF4EwhSZIkCG7uz0vXhspdBkPJ\nP6AXVOhvGYye901Ho1+PbbsO4bvvvlO6vDYbPHgwRo4chdPHZGzY03JaxrWajqENbAlnwZAEEyK0\nIjQNtU1bwq3cACn/h+NyGZS7KGcdJJ9X6SqI2oTBmUKSJEkQXDaly6BOTpD90NTug6Z2H4yaSEwf\ncjMmj3sQJWV2fPVNxz9cJSYmBtOnz4DXbcSGtT/slhGYjtHPhAa0z3SMaKMG/eJNGJ1iQUakHkb4\noS0vhn7vanjf2gTU1QLglAtqSeaIM4UYBmcKSZK3EaKnY4cW6lxEbzX0pauhg4AeYX3wyOyJcAkW\n7DlwHDt25nSow1W0Wi2mTJmKxMRU7NnmQm1N0yizViegd78wJDZPx1h/ZdMxVAKQEWXE0KQwDE80\nI9qggs5dD8PJg8CnmyAd2htoy/FEuhTZ1Rg4sIMoFHAf507M4/Fgx44dGDx48CVPNApFDbU2qI69\nCpWbo86kHElthidyHLzhg1Fe3YD1m75FSUmJojUNHDgQo0ePQcEJGSdym+YM/3g6RlsPKwnXq9E3\n1ohRKRb0DmwJZ4Nh/7fwbt8Y2BKOqK0MT72IyInTAyfCEXV0DM4hpqGhAevXr8fRo0dRW1uLRx55\nBD169IDT6cSWLVswfPhwxMfHK13mNeesPgtd7ksQpEalSyGCDMBvzIA7ZjI86mgczC3A1m3fwuPx\nXLcaoqOjMWPGDPjcJuRkOyHLLXfHCPawElH4fku4hDCMTLEgzqSG3ueG4UwuxF2b4d+7A5A4K5na\nh+7OBxEx92FoNBqlSyEKCqdqhJCqqiq8+OKLqKysREJCAkpKSgL7zYaFheGbb75BRUUFHnjgAYUr\nvfZkyQ8wNFMHIQBQN5yCuvBU0+EqSaMw7LF7YHf6sCE7B6dOXbuDerRaLSZPnoLkpG7Yu92Femc9\n+g02/zAd4zK7Yxg1IjJjTRiZYkH/WAMsGgGa2ioYD2+Hf80GyCWFAL7/x8E1uwrqqvxnz0Dy+QAG\nZwoRDM4hZNWqVWhsbMQf//hHWCwWPPzwwy0ev+GGG0Jyxf8VkTzcw5k6JEFqhLZyM7SVm2HUJ2P2\n5MnwzLgJx06dw8ZN2WhsbL9/8A0YMABjxtyIgpMy9nzrQr/BZhjCVfgivxJ/b2V3jESLDgPjTRiV\nYkFKuBYGyQv92Xyod/wbvpxswNP0j3GOKdP1IJWXwlfngM5gULoUoqAwOIeQQ4cOYcaMGUhOTkZd\n3YVfucbFxXX4Vf7thl8VUwhQuc7CWPxXGAQNhlqHo99DP0O9R4XN3+7D0aNHr7jf6OhoTJ8+A7Lf\nhBNH/MjoY0RUuoQVe85ib8kPfzboVAJ6xRgxPNmCwfEmROhEaOrtMOXtg7TyG0j5xwJt+V8UKUGu\n55Z0FFoYnEOIx+OBxWJp9fH2HMnq+PilMYUOQfZCW7MT2pqdMOpicfuYSZg+aT4KSirw9Tdbgj5c\nRavVYuLESeiWmg5HrRrhkRrIyT9Mx4gxaTAhIwKjUszoHqGHAX7oy4qg3fsv+P68ObAlHEMydRh+\nPyB3nB1piC6HwTmEJCcnIy8vD1OmTLno43v27EFaWtr1LYqI2kTlLoeh5APoBRX6WgYi475b0CgZ\nsH33Yezdu6/V5/Xr1w9jx46DTmdEowf4tqYKuw7WYnCiGf9xczdE6VXQuuthOnkA8icbIR35YdoW\ngzJ1WD4vIHGPAgodDM4hZPr06VixYgVSU1MxevRoAE0HgdhsNnz88cc4ceIEnn76aYWrvF44w5lC\nW9PhKvuhqd0Po8aKaQNvxsQbH0RpuQNfbchGRUUFACAqKgozZsyA1RoBr1+NAocLehUwuZsJ08Mc\nMOz/Ct73NwJV5QD4XQyFGL8fMhicKXRwO7oQs3r1anz88ceQZTmwabwsyxBFEbNnz8btt9+udInX\nRZ3tKAwn/qB0GUTtSoYAX1gmPNET4RYjUFnrgTUiEqJKA48kQddYB7G0CNK5IqADHbhCdMXUGuhG\njoM1uZvSlRAFhSPOIeZnP/sZxo8fj5ycHNhsNsiyjLi4OIwcORJxcXFKl3cdccSZOh8BMlTOk/Bb\nh6NBb8AZzxfwlja0bKQH0F2R8ojanVo0YLiB43cUOhicQ1B0dDR+8pOfKF0GEbUjCYA79hbUhg/C\n4bIvcPzU15DbdL4fUejRqy24odtcpcsgChqDMxGRwjyWYaiJm4KC2r34LvcZ+CSX0iURXReioIYg\n8BtECh0Mzh3Y7NmzIQgC3n//fajVasyePfuyzxEEAR9++OF1qI6IrpbP0A01SXei3FWCncefRaPX\nrnRJRNeVKKg48Y5CCoNzBzZz5kwIggBRFFv8TADnOFMok9ThsKfeC7vsx/bTy2FvLFK6JCJFCIIK\ngKh0GURBa9Nv69KlSzFixAhYLBbExcXhjjvuwIkTJ1q0cbvdeOKJJxAdHQ2z2YxZs2ahvLy8RZvF\nixdj+PDh0Ov1GDp0aKuv98orr6B3797Q6/VISUnB0qVLL1lfbm4uZs2ahfT0dIiiiNdff/2CNtu2\nbcOtt96KpKQkiKKItWvXXva6a2pqsGjRImRmZsJkMqFbt274xS9+0eLQgpUrV0IURahUKoiiGLip\nVCpUVlZesu+5c+ciPDwcEREReOihh1BfXw8AuOuuu3DnnXfiyJEjGD9+PO6//34888wzKCgowJ13\n3tnqDQjucyguLsaMGTNgMpkQHx+PJUuWQApipf7zzz+PxMREGI1GTJkyBfn5+UFfU7sR+ActhR4J\natSm3IdzaQ9jS8k/8cWx3zI0U5dm0FihUWuVLoMoaG1KH9u2bcOTTz6JXbt2YcOGDfB6vZg6dWqL\nE+sWL16ML774Av/617+wdetWnDt3DjNnzrygr/nz5+Puu+9u9bUWLVqEd999F8uXL8fx48exdu1a\njBgx4pL1NTQ0ICMjA8uWLUNCQsJF29TX12Pw4MFYsWJF0KO3586dQ2lpKZYvX44jR45g5cqV+Oqr\nr/DQQw8F2tx9992w2WwoLS2FzWaDzWZDVlYWbr75ZkRHR7fa95w5c5CXl4eNGzfiiy++wNatW/Ho\no48GHq+rq0NWVhZSU1Mxd+5czJs3Dy+++CLefvvtS9Z8uc9BkiRMnz4dPp8POTk5WLlyJf7617/i\n+eefv2S/y5YtwxtvvIE///nP2L17N0wmE7KysuDxeIK+pnYh8ssSCh0SgPq4GSjr+Qx21+7Dv4/+\nEiWO7y77PKLOLkwbA73WrHQZREG7qn2cKysrERsbi61bt2Ls2LFwOByIiYnBhx9+iDvuuAMAcPz4\ncfTp0wc5OTkXBN/f/e53WLNmDb77ruVfIHl5eRg0aBByc3PRo0ePK6otPT0dTz31FBYtWtRqG1EU\n8e9//xu33nprm/v/5JNPcO+996K+vj4wleJ8lZWVSEpKwnvvvYc5c+ZctI9jx46hb9++2LdvH4YM\nGQIA+PrrrzFjxgycPXsW8fHx+NOf/oTnnnsONpsNDz/8MObMmYNNmzZhzZo1yM3NvWi/wXwO69at\nw6233orS0tJAsP/zn/+M3/zmN6ioqIBaffFgmpiYiF/96ld46qmnAq8VFxeHlStX4q677kJeXh76\n9et3yWtqD3VVRdAffR4CN86nDs4TPhw1sZNx2r4LB879Ez7JrXRJRB3GsKS5GNPrvlb/ziHqaK7q\n+2673Q5BEBAZGQkA2LdvH3w+HyZNmhRo07t3b6SmpmLnzp1B9/v5558jIyMDa9euRffu3ZGeno6H\nH34YNTU1V1Nuu7Lb7bBYLBcNzUDT1A2TyXTR0fZmO3fuRERERCBgAsDkyZMhCAJ27doFAMjJycH4\n8eOhVqsxePBgHDx4EFlZWTh+/Dhqa2sBANnZ2RBFEUVFTV/5BvM55OTkYMCAAS1Gw7OyslBbW4uj\nR48CAAoLCyGKIrZu3QoAOHPmDGw2W4t+LRYLRo4c2aLfy11TexBEFWR1WLv1R9TefMZ0VPT8FY6H\nZeCzY89i79lVDM1EPxJuSGr171GijuiKf1tlWcbixYsxduxY9O3bFwBgs9mg1WphsVhatI2Li4PN\nZgu679OnT6OgoACffPIJ3n//faxcuRL79u0LzN9VWmVlJV566aVLTj947733MHfuXOh0ulbb2Gw2\nxMbGtrhPpVIhMjIy8H7ZbLbAwSYzZ85EaWkp1q9fD6BpxNrpdAIAevXqBY/HA6fTGdTncH6/5z/e\n/BgAaDQaZGZmwmg0Bu4XBOGizzu/38tdU3sQVVrI6vB264+ovUjqCFR3fxKFcVOxLv+P2HL6f9Ho\n424ZRBcTpo3moncKKVf83ciCBQuQm5uLb7/99rJtm4+GDpYkSfB4PFi1ahUyMjIAAO+88w6GDRuG\nkydPQq/XB8K6IAj4z//8T/zmN7+5sgv5kaVLl+Lll18O9J2bm4vk5OTA43V1dZgxYwb69++PF154\n4aJ97Ny5E3l5eXj//fevqIbW3q+nn34aAFBbWwtZlvG73/0OZnPT3LBBgwbht7/9LQC0eux2sJ9D\nc5vExMRWp4O0td+2/g5cjqgxQtJGQuU62259El0NSdSiLnkO6rRR2FX8V5xzHFS6JKIOT63SMzhT\nSLmi4Lxw4UJ8+eWX2LZtGxITEwP3x8fHw+PxwOFwtBjtLC8vb9Nx0AkJCVCr1YHQDAB9+vQBABQV\nFWHChAk4ePCHv5Sap4q0h8cff7zFfsnnX5/T6URWVhasVitWr14NlUp10T7efvttDB48GIMHD77k\na8XHx1+w04Xf70dNTU1gLnB8fDzKysoA/LAd3dGjR7F+/XrMnj07MBp8vujo6Mt+DvHx8dizZ0+L\n5zW/TmufVXx8PGRZRllZWYs25eXlgakZl7qm9jwSXKXWwmNMARyH2q1PoishAWiMvxUOc18cLluD\n4xUbAM69JwqKWtX6t7JEHVGbg/PChQuxZs0aZGdnIzU1tcVjw4YNg1qtxsaNGwOL0k6cOIGioiKM\nHj066Ne48cYb4fP5cObMGaSnpwNoWtwmCAK6desGURTRvXv3tpYeFKvVCqvVesH9zbtbGAwGrF27\nFlrtxbfPqa+vx8cff4xly5Zd9rVGjx4Nu92O/fv3B4Lnxo0bIctyYCHl6NGj8eyzz8Lv9+Ouu+4C\nAOzfvx+ZmZm4//77L9qvw+Fo9XMYM2ZMoN+XX34ZlZWVgXnO69evR3h4eGA0/8fS09MRHx+PjRs3\nYuDAgYHX2rVrF5544onLXtPIkSMv+54ES6VSQTYktVt/RFfCYx2JmtgJOFW9EweO/hJ+2XP5JxER\nAEAt6qHiDkkUYto0x3nBggX4+9//jn/84x8wmUwoKytDWVkZXK6m42EtFgvmz5+PX/7yl9iyZQv2\n7duHBx54ADfeeGOLHTVOnTqFAwcOoLS0FI2NjTh48CAOHjwIr9cLoGkx2dChQ/Hggw/iwIED2Ldv\nHx577DFMnTr1krtseL1eHDx4EAcOHIDH40FJSQkOHjyIU6dOBdrU19cH2gBN86kPHjyI4uLiVvt1\nOp2YMmUKGhoa8Pbbb8Nutweu/cf7Hn/44Yfw+/2t7qRxvszMTGRlZeHhhx/Gnj17sH37djz55JO4\n5557AiPOc+bMgVarxYMPPojc3Fx89NFHeP311/H000+joaEBkiRhz5496NOnD0pLSy/7Odxwww0A\ngKlTp6Jv37649957cejQIXz99dd47rnnsHDhQmg0GgBN2/D16dMHe/fuDdS8ePFivPTSS/jss89w\n+PBh3HfffUhOTsZtt90W9DW1B0EQIGsi2q0/orbwGbujoucSHDOlYm3ef2Jfyd8ZmonayKSNglql\nUboMojZp03Z0oihedC7Se++9h/vuuw9A08EbzzzzDD744AO43W5MmzYNK1asaLFgbMKECYGdGs53\n5syZwCi2zWbDk08+ifXr18NkMmH69Ol45ZVXLjoa3KywsBDp6ekX1HjTTTdh06ZNAJp2oJgwYcIF\nbe6//368++67F+03OzsbEydObHFf85zd82sGmkbLMzIy8Le//a3VOs9nt9uxcOFCfPbZZxBFEbNm\nzcJrr73WYgrG4cOHsXDhQuzZswdWqxWDBg1CVFQUfD4fnn32WVRVVWHixIlYsmQJ5s2bh379+gX1\nORQXF+Pxxx/Hli1bYDKZ8POf/xxLly4NrHAuLCxE9+7dsXnzZowfPz7wvBdffBFvvfUW7HY7xo0b\nhxUrVrT4B00w19Qe6ipOwZD3+3btk+hSJE0kalLnocbvwo7CN1HrOqd0SUQhK9U6EtP6/QdMRu7j\nTKHjqvZxpuvr+PHj+P3vf4/IyEj0798fmzZtwnPPPYf+/fsDaAq0VqsVixcvVrjS68NZUwJt3v9A\n9Dku35joKjQt/JsHh9aKnKJ3Yas7onRJRCFvXPqTGJL+01bXCxF1RJxcFEI++OADJCUl4b//+7/R\n2NgYGEVv1q9fP2RnZytU3fWn1lvhN/WAWMsT2OjaaFr4dxsc5kwctK3Gycot4MI/ovYRHZbBPZwp\n5DA4h5BTp07hnnvugUajCcwrP19kZGSHOiTmWtNo9Wi0DoaGwZmuAXfEaNijb0J+zXYcPPpL+GWv\n0iURdRoqQQudOoxb0VHIYXAOISqVCpeaWVNdXQ2DwXAdK1KWKIqQDImXb0jUBj5TT1Qn3oHS+lPY\ndew/4eZUIKJ2F2XKgEHH018p9DA4h5CePXsiJycHM2bMuOAxl8uFLVu2tLqVXGckCAJEtR6yaIAg\nNSpdDoU4SRuFmuR7Ue13YsfJpXC4S5UuiajTSrEOg0nf+mJ/oo6KwTmE3HXXXXjxxRexdOlS3Hjj\njQCAgoIClJWV4bPPPoPD4cDMmTMVrvL6UuvC4Td1h7ruqNKlUIiSRD0cKfPgUJuRU/QOypyXPy2T\niK5OomUA1GpGEAo93FUjxBw5cgR/+ctfYLPZWtwfFxeHxx57rEuNOAOAz+eDu3gjdMX/ULoUCjES\ngIaEn8ER1hMHSv+FU1XZ4MI/outBwOzBbyIxpvVzGYg6KgbnEFVQUIDS0lLIsoy4uLgWx5N3JbIs\nw1l+HIbjS5UuhUKIK2Is7NHjcLI6G4dKV0OSfUqXRNRlWPXJ+OmAlxFt5emvFHr4PUkIKSgowNmz\nZzF27FikpaUhLS0NBw4cwN/+9jd4vV6MHTsW06dPV7rM60oQBEBthCxoIHDXA7oMn6k3qhNvw7n6\nk9id92u4/U6lSyLqcpLCh8BsiFK6DKIrwuAcQt5//33odDqMHTsWAFBeXo5XXnkFZrMZERERWLly\nJbRaLSZPnqxwpdeXWm+B35wJteOw0qVQB+XXxsCeMg9VPgd2nHwZdW7b5Z9ERNdEj5ibodXqlC6D\n6IowOIeQwsJC/PSnPw38nJ2dDVEUsWzZMlgsFrz66qv45ptvulxw1hksqI+5icGZLtC08O9eONQm\n7Cz8C8rrjyldElGXZtREwqSL5P7NFLIYnENIQ0MDzGZz4Of9+/dj4MCBsFgsAICBAwfiwIEDSpWn\nGFEUAX0cZFEHQXIrXQ51ABLE7xf+ZWD/uX/idPU2pUsiIgC9YibDGhardBlEV4zBOYRYrVaUlJQA\nAGpqanD69GncfPPNgcddLleX/Ve81hgFn3U4NNXblS6FFOaKvAn26DE4UbkFhwvf4MI/og6ke9RY\naDWcpkGhi8E5hNxwww1Yt24dPB4PTp06BY1GgxEjRgQeLywsRFxcnIIVKkerN8IZPYbBuQvzhvVB\nTcKtKHHmYXfuEnj89UqXRETnsegSYdSFd9kBHuocGJxDyN133w2Hw4Ft27bBaDRiwYIFsFqb7W+t\noAAAH2lJREFUTl5qaGhATk4OsrKyFK5SGYIgQNBaIanDIfpqlS6HriO/Ng41KXNQ5bNjx4n/gtNT\nrnRJRHQRfeKmwRrWNQd3qPPgPs6dhCRJcLlc0Gq1XfY0Jo/HDd+ZT6EtW6d0KXQdSKIRtSnz4FAb\nsLPwL6ioP6F0SUR0CbMG/j8kx/biiDOFtK6ZsDohURRhNBqVLkNRGo0WroihDM6dnAQRDYmz4DCl\n4btzH+EMp+cQdXhRxgyY9JymQaGPwZk6DUEQoNJa4NclQOUuVbocugZcURNgjxqF45UbcaTgdS78\nIwoR/eJ/AospWukyiK4agzN1KrqwaLjiJkNVtErpUqgdec39UBP/U5ytO4I9ub+Cx9+gdElEFCSt\nyojE8P5QqVRKl0J01RicqVNRq9WQzb0gqUwQuatCyPPpEmBPuQeVnirsPPE7OD0VSpdERG00MGEm\nIsyJnKZBnQKDM3U6OnM8vPHToSv5WOlS6ApJogm1qffCIWqx/cz/Q1VDvtIlEdEVUIlaZESPg45H\nbFMnweBMnY5Go4E7fADk0s8hSI1Kl0NtIEFEfdJdcBhTsa/kAxTW7FS6JCK6Cv3iZsBqjuNoM3Ua\nDM7U6QiCAJ05Dp64qdCVrlG6HApSY/Qk1EbegLyKDTh65lXIsl/pkojoKoiCGr1js2DQmZQuhajd\nMDhTp6TR6uCOGAq5bB0EyaN0OXQJXvMA1MTPQFHdIezNXQIvF/4RdQo9oyfCGhbL0WbqVBicqVMS\nBAFacyy8MROhLftK6XLoInz6JNiT70aFuxw7T7yIek+l0iURUbsRMCDhNpgMFqULIWpXDM7UaWm1\nBtRFjoKmfAME7vfbYUjqMNQm34taUYUdZ95AVcMppUsionaWFjkaFlMMR5up02Fwpk5LEATowmLg\njRoPbeUmpcvp8iSIqE++Gw59EvaW/B1F9t1Kl0RE14SAoUmzYTFFKl0IUbtjcKZOTas3oS5mPDRV\n30KQOddZCRIAd8xU1EYMRW7518g99QpkSEqXRUTXSGbsVESEcd9m6pwYnKlTEwQBeksCPEk/g+7s\nh0qX0+V4LYNQHXcLihz7sffoM/BJLqVLIqJrSKsyYVDiTJhNEUqXQnRNMDhTp6fV6eGJGAh/xRao\n3Daly+kSfPpk1CTPRoXbhp3Hn0eDt1rpkojoOhiT9iiirckcbaZOi8GZugRjeALqU++F/uQfwT/O\nrx1JbUZtyr2wQ8CO06+hurFA6ZKI6DqJMmYgOWIQtBqeEkidF4MzdQmiKEJtSYIv8kZoqrcrXU6n\nI0ENZ/LdqNMnYPfZVThbu1fpkojouhIwrvtCRIUnKl0I0TXF4Exdht5oRV18FtS1+yHwkI120bTw\nbxpqrYNxtHwd8k79gQv/iLqgPrG3INqSAlEUlS6F6JpicKYuQxAEGMIT4E6+G/rCd5UuJ+R5w4eg\nOi4LBfZ9+C6XC/+IuiqdKgwDE+9AmNGqdClE1xyDM3UpGo0WnvDe8BvSoOL82yviM6SiJmk2yl0l\n2HnsOTR6a5QuiYgUNCb9McRwQSB1EQzO1OUYLLFwps6F4fhSCJxWEDRJbYE99T7YZT92nF6OmsZC\npUsiIoXFhvVBsnUgNBqt0qUQXRcMztTliKIInSUBnsQ7oDv3L6XL6fAkqOFMuQcOXRz2nP0bztZ+\np3RJRNQBaFVhuDljMaKsXBBIXQeDM3VJOoMZvugR8DmOQu08pnQ5HZIEwB07HfbwgThS9jmOV6zn\nwj8iCpjU89eIjUjlFA3qUgRZlmWliyBSgizLqKsqgv7EHyH66pQup0PxhA9HTexknLHvwf5zH8In\nuZUuiYg6kEEJMzE87W5YwiKVLoXoumJwpi7N7/ejoTwX+pP/CwH8T8FnSENN0p0ocxUjp+htNHrt\nSpdERB1MlDED0/o8h5gILgikrofBmbo8d6MDvpLN0J1brXQpipHUVthT70WN7MWOgjdhd51VuiQi\n6oA0ogG3D/hfJEb35J7N1CVxjjN1eTqDBb7okfA5crvcfGdJ0KAuZQ7qtNHYXbwSJY4DSpdERB3Y\nhB7PIDYijaGZuiwGZyIAxvA41KXOgdhF5jtLABrjfgqHpR8Ol63FiYoNXPhHRJfUN24GkiP7Q6vR\nKV0KkWI4VYPoe11lvrMnfARq4ibidE0O9p/7J/ySR+mSiKiDs+pTML3v7xEbmcJ5zdSlMTgTncfd\n6IDv3FboSj5WupR25zN2R03STNgaCpFT9A5cvlqlSyKiEGBQW/GTfv+DxOgMTtGgLo9TNYjOozNY\n4I8ZCY+nCtqKTUqX0y4kdQRqut0Lu9+N7fl/RC0X/hFRkNSiHrf0+R0SorozNBOBwZnoAgZzNBoS\nsuD12KEJ4VPyJFGLuuS5cGgjsKvoPZTWHVa6JCIKIYKgwrTMF5AQ2RMqlUrpcog6BAZnoh8RBAHG\n8Dg4U+6E4HNAXZ+vdEltIgFojL8NDnMfHLR9ipOVm4BOPGebiK6NST2WIDmqL7RaLgYkasY5zkSt\nkGUZzuqz0OavgMpdqnQ5QXFHjII9ZgLyq7fj4LlP4Je58I+I2m50t4fRLykLZlOE0qUQdSgMzkSX\nIEkSnFWF0Of/H4gd+BQ9n6kHahJ/htKG08gpehdun0PpkogoRPWPvx3Du81GhCVW6VKIOhwGZ6LL\n8Pv9qK86DcPJ5RD8DUqX04KkjUJN8jxUS/XYUfAWHO5zSpdERCEsLWIMxmU8jihrAredI7oIBmei\nIHi9XriqTkB/8lUIslfpcpoW/qXcC4fagpyid2FzHlW6JCIKcbFhvTG512+4VzPRJTA4EwXJ43HB\nXXEU+tMrIMh+RWqQADQk3AFHWC8cLF2N/Kot4MI/Irpa0caemNL7N4iL6sbQTHQJDM5EbeBxNcBd\nfQL6U29c95Fnd8SNqIkej/zqrThUuhr+DjDyTUShLzasNyb1XMLQTBQEBmeiNvJ4XHBXn4I+/3UI\nkuuav57P1BPViXegtD4fu4rfg9tXd81fk4i6hviwvpjQ6xnERaYyNBMFgcGZ6Ap4vR40Vp+GPv91\niP76a/Iafm0MalLmocbnwPbCP6PObbsmr0NEXVOieRBu7rmYc5qJ2oDBmegK+Xw+NFSfgT7//0L0\n1bZbv5KohyNlHhzqMOwsegflzrx265uICAC6RYzCjd0fQ2xEMkMzURswOBNdBb/fj/rqQuhPrYDo\nqbyqviSI3y/864H95z7G6eqt7VQlEdEPesdMxQ3d5iHamsjQTNRGDM5EV0mSJDiri6E79SZUV7iP\nsityPOzRN+Jk5RYcsn0KSfa1c5VERMCghFkYlHw7IsPjGZqJrgCDM1E7kCQJ9TVnoT39DlSNBUE/\nz2fKRHXibShxHsPu4pXw+J3Xrkgi6tJGps5Hn4TJsJpjGJqJrhCDM1E7kWUZzuoSaIo/gtpx6JJt\n/do42FPmoMpXix2Ff0adu+w6VUlEXY1K1GJSj18jJWogLGGRSpdDFNIYnInakSzLaKgtA8q3Qmv7\nAj8e05FEA2pT7kWd2oCdhW+jvP64InUSUddg0kYjq/dziI/MgE5rULocopDH4EzUzmRZhqveDl91\nLvSF70KQfU0L/xJnwmFKx3fnPsKZ6u1Kl0lEnVyCeQBuyliE2MhuUKlUSpdD1CkwOBNdIx53A1w1\nBRBq9qM2YhiOV27CEdsaLvwjomtuYMIdGJh0O6KtSZzPTNSOGJyJriG/34/K6hJsPfMGztbuU7oc\nIurkREGNmzN+ibToYbCERTE0E7UzBmeia0yWZVTUlODwuTU4VLpa6XKIqJMyaKyY2vs5JEb0gl5v\nVLocok6JwZnoOpBlGQ5nFYqrD2Fz/ivwSW6lSyKiTiTG1BMTe/4KcZFpnM9MdA0xOBNdR26PC2XV\np7Hh5FLUuq7ssBQioh8IGJQ4E/0TfoJoaxJEUVS6IKJOjcGZ6DqTJAkVNWdxtPRzHCxdDYD/CRJR\n2xk1UZjUcwnirD0QZgznfGai64DBmUgBsizD2WBHmf00NuX/AfWeSqVLIqIQ0itmMoYm343YiFRO\nzSC6jhiciRTk9/tRXlOMA2c/wrGK9UqXQ0QdnFYVhgk9fonEiL4ID4vmKDPRdcbgTKSw5oWDpfbj\n2Jz/Clw+h9IlEVEHlBw+DGPSH0GsNRUajVbpcoi6JAZnog7C6/Wiwl6MXYXvoaBmh9LlEFEHoRK1\nGJv+BLpFDkNkeDxHmYkUxOBM1IHIsowaRzlKag4j+/Rr8PoblC6JiBQUY+qJmzKeQqw1FTqdQely\niLo8BmeiDsjjdaOipgg5he+iyL5b6XKI6DrTqowY3e0RpEQOQVR4IreZI+ogGJyJOihZlmGvq0CF\n4wy2nn4ddW6b0iUR0TUnoE9sFgYmzkS0NQlajU7pgojoPAzORB2c3+9HZW0JCqpysLtoJXySS+mS\niOgaiDJmYFz3hYiypMBstHIuM1EHxOBMFCJc7kZU1Zbgu5IPkV+5WelyiKid6FRhGJP+GJKsAzgt\ng6iDY3AmCiFNW9dVo6quCFtP/1/UNBYqXRIRXSEBIvrGzUD/hFsRY03mFnNEIYDBmSgESZKEqtpz\nOFtzEDsK3oLH71S6JCJqgxhTb4zrvgCR5mQel00UQhiciUKYx+tGpb0ER2xrkVv2JWTZr3RJRHQJ\nZl08Rnd7CHGW3ogMj+e0DKIQw+BMFOJkWUZ9Yx3sThsOl67B8fL1kCEpXRYRnSdMG4NR3eYjztIb\nUeFJUKlUSpdERFeAwZmok2gK0A7YnTYcOvdvnKjYwABNpDCjJgqjuj2IeEtfRIUnQq1WK10SEV0F\nBmeiTkaWZTgbamF3luHQudU4WbmJAZroOjNorBiR8sD3O2UkMTATdRIMzkSd1PkB+uC5T5BfuYUB\nmuga06stuCHlfiRHDEaUJQkajUbpkoioHTE4E3VyTQHaDruzDAdKPsGpqmwGaKJ2plObMTx5HlIj\nhiMqPJFbyxF1UgzORF2ELMuoa7Cj1mnDkdLPcLJyE/yyV+myiEJauD4RQ5PnIN6ciUhLIrRaHpFN\n1JkxOBN1MbIso8HlRK2zHEX2fThY8gkavFVKl0UUQgQkhw/FkKS7EG6MR4Q5nnOYiboIBmeiLkqW\nZfh8PlQ7zqGm4Sz2nf0A5c48pcsi6rDUoh6ZsdOQGTsVFmMUzKZI7sNM1MUwOBMRJElCbV0FHI1V\nyLV9jpOVm+GXPUqXRdQhhGljMSR5NpIsA2E1x0GvM/KkP6IuisGZiAI4jYPoB/Hm/hiafA8ijEmI\ntCRwOgYRMTgT0YV+PI3j0LlPcc5xkLtxUKdn0EQgMyYL3aPHwmyIQnhYNKdjEFEAgzMRXZIkSair\nr4GzsQZldXk4YluLqobTSpdF1G7Uoh7pkTeiT9w0hOmiYDXHQavRcToGEV2AwZmIgubz+VBbX4kG\nlx0FNbtwrOwrOD3lSpdF1GYCRCRYBqB/wm2INKbAYoqBUR/GsExEl8TgTERtJssyvF4v7M4y1Ltr\ncKJiA/Irs+HxO5UujeiSrIZU9I//KRItAxBmiODOGETUJgzORHRVZFmGy92I2voyOBrLcNT2BYrt\ne3i4CnUYJm00ekZPREbUOBj14bCGxUGlUnF0mYjajMGZiNqNJElocDlQ11CNOlc58iuzUVizCy5f\nrdKlUZciIMbUAz1jJiLBMgB6bRjCTTGct0xEV43BmYiuCUmS4Pa44KivgMtbh7P2/civzEZ14xml\nS6NOSC3qkRQ+GL1jJsNqTIFRZ4HZGMmRZSJqVwzORHTNybIMv9+P2voKNLqdqG4owImKjTjnOAi/\nxINW6MqYtNFIj7wR3aPGwqiNgNkYCaPezDnLRHTNMDgT0XXn9/vR6KpDXWM16t3VOF21HYU1Odyh\ngy5JFNSIMfVE96hxSAwfAL3G3DQFQ8spGER0fTA4E5GiZFmGx+OGo6ESbm8DnO4KFNv3oaT2AGoa\niwDwj6iuqjkod4schURLU1A26MwIM0RwCgYRKYLBmYg6FEmS4PN5UddQBbe3AQ0eO87VHkRx7T5U\n1p+CJPuULpGuEY1oQExYL3SLGIFYcyYMGguDMhF1KAzORNShNR//Xe+qQaPbCZe3DuXO4yiozkGF\n8wS8UqPSJdIVCtPGIi6sD7pFjkCEMRVatRFGfTiMOjODMhF1SAzORBRSmhcaNrjr0OCqhdfXiHpP\nNcrq8lBadxTVDWfg9tUpXSadRyVoYTWkICasJxIs/RGuT4JObYRGrUeYPgJarR6iKDIoE1GHx+BM\nRCFPkqSmBYceJxrddfD6XHB5HahuKIDNkYvqxgLUuko4zeM6MGgiEGlIQ7ylH2LDesOojYBWrYdO\nbYRRb4VarWZIJqKQxeBMRJ1S88i0z+dFvdsOj9cFr98Fl7cWVfWnUVl/Gg63DU53GRq9dsiQlC45\nZKhFPcy6OITpYhFlTEOEMRUWfSJ0KhM0ah2MOgt0WhPUajUDMhF1KgzORNSlNAdqv98Hl8cJt68R\nfr8XPr8bXsmNek8VahvPoqr+DOrcZXC6y9Hosytd9nUlCCqYNFEw6+IQbkhClDEd4YYk6NRmaFQ6\nqEQNNGodDNowaNRN0yw4ikxEXQF3iSeiLkUQBKjVauh0eoSboxEbkYKE6O5IieuD9PhB6JcyAaN6\nzEVWv9/gJwN+j58NehV3D/4LZg58A7f1ewW3ZP4ON2f8EiNSH0C/uFuRFnEjEswDEGlIg1ETBZWo\nVfoSL0qrMsKsi0eMqSeSw4ehZ/QkDEm6G2PTn8CUXs/ip32X4Y4Br2LWoBWYPfjPuH3gHzF9wAuY\n1OcXGNb9DvRKGoW0+P5IiumJ+Kg0RIUnwGgwQ6PRKLKQb+nSpRgxYgQsFgvi4uJwxx134MSJEy3a\nuN1uPPHEE4iOjobZbMasWbNQXt5yr/DFixdj+PDh0Ov1GDp0aKuv98orr6B3797Q6/VISUnB0qVL\nL1lfbm4uZs2ahfT0dIiiiNdff/2CNtu2bcOtt96KpKQkiKKItWvXXva6a2pqsGjRImRmZsJkMqFb\nt274xS9+AYfD0aLdnj17MHnyZERERCAyMhLTpk3DoUOHLtl3MO9XcXExZsyYAZPJhPj4eCxZsgSS\ndPlva55//nkkJibCaDRiypQpyM/Pv+C65s6di/DwcEREROChhx5CfX39ZftdsWIF0tPTYTAYMGrU\nKOzZs6fN10TUFmqlCyAi6igEQYAgCBBFEWq1Gnq9vsXjsixfcPNLPnh8Lvh8bvgkLyTJ33STffBL\nPkiyFz7JDZ/fA7/c9LNf8sIve+GXPPBLXvgkz/f/3wOf5GlqL3kgyT6IEKESddCo9FCrdFCLOqhF\nLVSB/9U2/a+ggUrUQiWqIQoaiIIKKlEDUVB/f58KoqiCRq2FRmWARq0LXGvzdYfSiPG2bdvw5JNP\nYvjw4fD5fPiP//gPTJ06FXl5eTAYDACaQvG6devwr3/9CxaLBU888QRmzpyJbdu2tehr/vz52LVr\nV6vBctGiRdiwYQOWL1+O/v37o7q6GtXV1Zesr6GhARkZGbjrrrvw1FNPXbRNfX09Bg8ejAcffBAz\nZ84M6rrPnTuH0tJSLF++HH369EFhYSEeffRRlJaW4p///Geg31tuuQW33347/vSnP8Hn8+H555/H\ntGnTUFxcDJVKddG+L/d+SZKE6dOnIzExETk5OTh37hzuvfdeaLVavPTSS63WvGzZMrzxxhtYuXIl\n0tPT8eyzzyIrKwt5eXnQapv+oTlnzhyUlZVh48aN8Hg8+PnPf45HH30U77//fqv9fvTRR3j66afx\n1ltvYcSIEXj11VeRlZWFEydOIDo6OqhrImozmYiIrilJkmS/3x+4+Xw+2efzyV6vN3DzeDyyx+OR\n3W534NbY2CC7XI2Bn5vbND+nuZ/mfiVJCty6moqKClkQBHnbtm2yLMtybW2trNVq5dWrVwfaHDt2\nTBYEQd61a9cFz3/xxRflIUOGXHB/bm6urNFo5JMnT15xbWlpafJrr712yTaCIMhr1qy5ov4//vhj\nWa/Xy36/X5ZlWd67d68siqJ89uzZQJvDhw/LoijKp06dumgfwbxfX375paxWq+WKiopAmzfffFO2\nWq2y1+tttb6EhAR5+fLlLV5Lr9fLH330kSzLTe+xIAjyd999F2jz1VdfySqVSi4tLW2135EjR8qL\nFi0K/CxJkpyUlCQvW7Ys6GsiaitO1SAiusaaR3abbyqVCiqVCmq1OnDTaDTQaDTQarWBm15vgE6n\nD/zc3Kb5Oc39nD/HONRGjtuL3W6HIAiIjIwEAOzbtw8+nw+TJk0KtOnduzdSU1Oxc+fOoPv9/PPP\nkZGRgbVr16J79+5IT0/Hww8/jJqamna/hitlt9thsVggik1/pffu3RtRUVF455134PV60djYiLff\nfht9+/ZFWlraRfsI5v3KycnBgAEDAqO5AJCVlYXa2locPXoUAFBYWAhRFLF161YAwJkzZ2Cz2Vr0\na7FYMHLkyBb9RkREYMiQIYE2kydPhiAI2LVrV+A+URTxt7/9DQDg9Xqxb9++Fv0KgoDJkycH+t27\nd2+7/A4QnY/BmYiIQposy1i8eDHGjh2Lvn37AgBsNhu0Wi0sFkuLtnFxcbDZbEH3ffr0aRQUFOCT\nTz7B+++/j5UrV2Lfvn2488472/UarlRlZSVeeuklPProo4H7wsLCsHnzZqxatQoGgwFmsxnr16/H\nl19+GQjXPxbM+2Wz2RAXF3fB482PAYBGo0FmZiaMRmPgfkEQLvq88/uNjY1t8bhKpUJkZGSLzyoz\nMxPh4eGB6/b7/Zfst6ysrF1+B4jOx+BMREQhbcGCBcjNzcUHH3xw2bayLLdpRF6SJHg8HqxatQpj\nxozB+PHj8c4772DTpk04efIkiouLYTabYTabYbFY8D//8z9XcyktLF26tEXfZ8+ebfF4XV0dZsyY\ngf79++OFF14I3O9yuTB//nyMGzcOu3fvxo4dO9C/f39Mnz4dbre7TTUE+341t0lMTERubi6GDx9+\n1f3+uE1ubi5uu+22du+XqC24OJCIiELWwoUL8eWXX2Lbtm1ITEwM3B8fHw+PxwOHw9FixLG8vPyC\nUcpLSUhIgFqtRkZGRuC+Pn36AACKioowYcIEHDx4MPBY81SR9vD4449j9uzZgZ/Pvz6n04msrCxY\nrVasXr26xYK/v//97ygsLEROTk6L+yIiIrBmzRrcddddF7xWMO9XfHz8BbtWlJWVAUCr72l8fDxk\nWUZZWVmLNuXl5YGpGfHx8RfsdOH3+1FTU9Nqv9HR0VCpVIHXb63e9vgdIDofR5yJiCgkLVy4EGvW\nrMHmzZuRmpra4rFhw4ZBrVZj48aNgftOnDiBoqIijB49OujXuPHGG+Hz+XDmzJnAfcePH4cgCOjW\nrRtEUUT37t0DN6vVevUX9j2r1dqi7+ZpFnV1dZg6dSoMBgPWrl0b2JmiWWNj4wVTMprnvre2ddyl\n3q8xY8YAAEaPHo3Dhw+jsrIy0Gb9+vUIDw8PTJH5sfT0dMTHx7fo1+FwYNeuXS36tdvt2L9/f6DN\nxo0bIcsyRo4cedF+NRoNhg0b1qJfWZaxcePGQL/t9TtA1IJSqxKJiIiu1OOPPy5brVZ569atss1m\nC9waGxtbtElLS5M3b94s7927Vx4zZow8duzYFv3k5+fL+/fvlx999FE5MzNTPnDggHzgwAHZ4/HI\nsty0U8Pw4cPlm2++Wd6/f7+8d+9eedSoUfK0adMuWZ/H45EPHDgg79+/X05MTJSXLFkiHzhwQM7P\nzw+0cTqdgTaCIMivvvqqfODAAbmoqKjVfuvq6uSRI0fKgwYNkk+fPt3i2pt31Th27JhsMBjkBQsW\nyHl5efKRI0fkefPmyREREbLNZrvke3qp98vv98sDBw6Up02bJh88eFD+6quv5NjYWPnZZ58NtCkp\nKZEzMzPlPXv2BO5btmyZHBkZKa9du1Y+dOiQfNttt8k9evSQ3W53oM0tt9wiDxs2TN69e7f87bff\nyr169ZLnzZvXor7evXvL//73vwM/f/TRR7Jer5dXrlwp5+XlyY888ogcGRkpl5eXB31NRG3F4ExE\nRCFHEARZFMULbitXrgy0cblc8sKFC+WoqCg5LCxMnjVrllxWVtain5tvvvmi/RQWFgbalJaWyrNm\nzZItFouckJAgz58/X66pqblkfQUFBRetccKECYE2W7ZsuWibBx54oNV+t2zZckH75j7Or3nDhg3y\nuHHj5IiICDkqKkqePHmyvHv37kvWHMz7VVRUJM+YMUM2mUxybGysvGTJkkBgb75uURTl7OzsFs97\n4YUX5ISEBNlgMMhTp069YHu/mpoaee7cubLFYpGtVqv80EMPyfX19S3a/PjzlWVZXrFihdytWzdZ\nr9fLo0aNahHYg70morbgkdtEREREREHgHGciIiIioiAwOBMRERERBYHBmYiIiIgoCAzORERERERB\nYHAmIiIiIgoCgzMRERERURAYnImIiIiIgsDgTEREREQUBAZnIiIiIqIgMDgTEREREQWBwZmIiIiI\nKAgMzkREREREQWBwJiIiIiIKAoMzEREREVEQGJyJiIiIiILA4ExEREREFAQGZyIiIiKiIDA4ExER\nEREFgcGZiIiIiCgIDM5EREREREFgcCYiIiIiCgKDMxERERFREBiciYiIiIiCwOBMRERERBQEBmci\nIiIioiAwOBMRERERBYHBmYiIiIgoCAzORERERERBYHAmIiIiIgoCgzMRERERURAYnImIiIiIgsDg\nTEREREQUhP8P1OtP0PlCWXsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3844bf6f50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.plot.pie(\"serie_4\", legend=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.4 - Mergear `DataFrame`s\n",
    "### 1.4.1 - Mergear mediante concatenación\n",
    "Vamos a extender la información de un dataframe concatenándolo con otro que tiene **nuevas filas**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>numero</th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>123</td>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>321</td>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15</td>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10</td>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   numero nombre_apellido sexo\n",
       "0     123      Juan Pérez    M\n",
       "1     321   Lucía Alvarez    F\n",
       "2      15     Carlos Hale    M\n",
       "3      10     Sofía Cilar    F"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>numero</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ignacio Heredia</td>\n",
       "      <td>400</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   nombre_apellido  numero sexo\n",
       "0  Ignacio Heredia     400    M"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas2 = pd.DataFrame({\n",
    "        \"numero\": [400],\n",
    "        \"nombre_apellido\": [\"Ignacio Heredia\"],\n",
    "        \"sexo\": [\"M\"]\n",
    "    })\n",
    "df_personas2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>numero</th>\n",
       "      <th>sexo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>123</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>321</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>15</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>10</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Ignacio Heredia</td>\n",
       "      <td>400</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   nombre_apellido  numero sexo\n",
       "0       Juan Pérez     123    M\n",
       "1    Lucía Alvarez     321    F\n",
       "2      Carlos Hale      15    M\n",
       "3      Sofía Cilar      10    F\n",
       "4  Ignacio Heredia     400    M"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# concatena dataframes \"uno arriba del otro\"\n",
    "df_personas_concat = pd.concat([df_personas, df_personas2], ignore_index=True)\n",
    "df_personas_concat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos a extender la información de un dataframe concatenándolo con otro que tiene **nuevas columnas**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>nombre_apellido</th>\n",
       "      <th>numero</th>\n",
       "      <th>sexo</th>\n",
       "      <th>profesion</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Juan Pérez</td>\n",
       "      <td>123</td>\n",
       "      <td>M</td>\n",
       "      <td>programador</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Lucía Alvarez</td>\n",
       "      <td>321</td>\n",
       "      <td>F</td>\n",
       "      <td>programador</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Carlos Hale</td>\n",
       "      <td>15</td>\n",
       "      <td>M</td>\n",
       "      <td>programador</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Sofía Cilar</td>\n",
       "      <td>10</td>\n",
       "      <td>F</td>\n",
       "      <td>artista</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Ignacio Heredia</td>\n",
       "      <td>400</td>\n",
       "      <td>M</td>\n",
       "      <td>artista</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   nombre_apellido  numero sexo    profesion\n",
       "0       Juan Pérez     123    M  programador\n",
       "1    Lucía Alvarez     321    F  programador\n",
       "2      Carlos Hale      15    M  programador\n",
       "3      Sofía Cilar      10    F      artista\n",
       "4  Ignacio Heredia     400    M      artista"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_personas3 = pd.DataFrame({\"profesion\": [\"programador\"]*3 + [\"artista\"]*2})\n",
    "\n",
    "# axis = 1 para concatenar dataframes \"uno al lado del otro\"\n",
    "pd.concat([df_personas_concat, df_personas3], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.4.2 - Merge\n",
    "Vamos a mergear un dataframe con otro que contiene el nombre de los días de la semana, utilizando el índice de tiempo que tienen los dos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "      <th>serie_6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>arbol</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>arbol</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>arbol</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4 serie_5  serie_6\n",
       "2016-11-23      1.0       1        4   arbol      NaN\n",
       "2016-11-24      3.0       3        3   arbol      NaN\n",
       "2016-11-25      5.0       5        2   arbol      NaN\n",
       "2016-11-26      NaN     NaN       23   arbol      4.0\n",
       "2016-11-27      6.0       6       15   arbol      3.0\n",
       "2016-11-28      8.0       8       30   arbol      2.0"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "date_range = pd.date_range(\"20161120\", periods=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dia_semana_ingles</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-20</th>\n",
       "      <td>Sunday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-21</th>\n",
       "      <td>Monday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-22</th>\n",
       "      <td>Tuesday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>Wednesday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>Thursday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>Sunday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>Monday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-29</th>\n",
       "      <td>Tuesday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           dia_semana_ingles\n",
       "2016-11-20            Sunday\n",
       "2016-11-21            Monday\n",
       "2016-11-22           Tuesday\n",
       "2016-11-23         Wednesday\n",
       "2016-11-24          Thursday\n",
       "2016-11-25            Friday\n",
       "2016-11-26          Saturday\n",
       "2016-11-27            Sunday\n",
       "2016-11-28            Monday\n",
       "2016-11-29           Tuesday"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dia_semana_ingles = pd.DataFrame({\"dia_semana_ingles\": date_range.weekday_name}, index=date_range)\n",
    "df_dia_semana_ingles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "      <th>serie_6</th>\n",
       "      <th>dia_semana_ingles</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Wednesday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Thursday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>arbol</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>arbol</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Sunday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>arbol</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Monday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4 serie_5  serie_6 dia_semana_ingles\n",
       "2016-11-23      1.0       1        4   arbol      NaN         Wednesday\n",
       "2016-11-24      3.0       3        3   arbol      NaN          Thursday\n",
       "2016-11-25      5.0       5        2   arbol      NaN            Friday\n",
       "2016-11-26      NaN     NaN       23   arbol      4.0          Saturday\n",
       "2016-11-27      6.0       6       15   arbol      3.0            Sunday\n",
       "2016-11-28      8.0       8       30   arbol      2.0            Monday"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"left\": el dataframe principal es el de la izquierda\n",
    "merged_dia_semana = pd.merge(\n",
    "    df, df_dia_semana_ingles, left_index=True, right_index=True, how=\"left\")\n",
    "merged_dia_semana"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "      <th>serie_6</th>\n",
       "      <th>dia_semana_ingles</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-11-20</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Sunday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-21</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Monday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-22</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Tuesday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-23</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4.0</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Wednesday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-24</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Thursday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-25</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2.0</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-26</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23.0</td>\n",
       "      <td>arbol</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-27</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15.0</td>\n",
       "      <td>arbol</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Sunday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-28</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30.0</td>\n",
       "      <td>arbol</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Monday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-11-29</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Tuesday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            serie_2 serie_3  serie_4 serie_5  serie_6 dia_semana_ingles\n",
       "2016-11-20      NaN     NaN      NaN     NaN      NaN            Sunday\n",
       "2016-11-21      NaN     NaN      NaN     NaN      NaN            Monday\n",
       "2016-11-22      NaN     NaN      NaN     NaN      NaN           Tuesday\n",
       "2016-11-23      1.0       1      4.0   arbol      NaN         Wednesday\n",
       "2016-11-24      3.0       3      3.0   arbol      NaN          Thursday\n",
       "2016-11-25      5.0       5      2.0   arbol      NaN            Friday\n",
       "2016-11-26      NaN     NaN     23.0   arbol      4.0          Saturday\n",
       "2016-11-27      6.0       6     15.0   arbol      3.0            Sunday\n",
       "2016-11-28      8.0       8     30.0   arbol      2.0            Monday\n",
       "2016-11-29      NaN     NaN      NaN     NaN      NaN           Tuesday"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"right\": el dataframe principal es el de la derecha\n",
    "pd.merge(df, df_dia_semana_ingles, left_index=True, right_index=True, how=\"right\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos a **mergear dos dataframes utilizando una columna que tienen en común**, para traducir los días de la semana al español."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dia_semana_español</th>\n",
       "      <th>dia_semana_ingles</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Lunes</td>\n",
       "      <td>Monday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Martes</td>\n",
       "      <td>Tuesday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Miércoles</td>\n",
       "      <td>Wednesday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Jueves</td>\n",
       "      <td>Thursday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Viernes</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Sábado</td>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Domingo</td>\n",
       "      <td>Sunday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  dia_semana_español dia_semana_ingles\n",
       "0              Lunes            Monday\n",
       "1             Martes           Tuesday\n",
       "2          Miércoles         Wednesday\n",
       "3             Jueves          Thursday\n",
       "4            Viernes            Friday\n",
       "5             Sábado          Saturday\n",
       "6            Domingo            Sunday"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traduccion_dia_semana = pd.DataFrame({\n",
    "    \"dia_semana_ingles\": [\"Monday\", \"Tuesday\", \"Wednesday\", \"Thursday\", \"Friday\", \"Saturday\", \"Sunday\"],\n",
    "    \"dia_semana_español\": [\"Lunes\", \"Martes\", \"Miércoles\", \"Jueves\", \"Viernes\", \"Sábado\", \"Domingo\"]\n",
    "})\n",
    "traduccion_dia_semana"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>serie_2</th>\n",
       "      <th>serie_3</th>\n",
       "      <th>serie_4</th>\n",
       "      <th>serie_5</th>\n",
       "      <th>serie_6</th>\n",
       "      <th>dia_semana_ingles</th>\n",
       "      <th>dia_semana_español</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Wednesday</td>\n",
       "      <td>Miércoles</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>Jueves</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.0</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>arbol</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "      <td>Viernes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>arbol</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Saturday</td>\n",
       "      <td>Sábado</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6.0</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>arbol</td>\n",
       "      <td>3.0</td>\n",
       "      <td>Sunday</td>\n",
       "      <td>Domingo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>8.0</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>arbol</td>\n",
       "      <td>2.0</td>\n",
       "      <td>Monday</td>\n",
       "      <td>Lunes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   serie_2 serie_3  serie_4 serie_5  serie_6 dia_semana_ingles  \\\n",
       "0      1.0       1        4   arbol      NaN         Wednesday   \n",
       "1      3.0       3        3   arbol      NaN          Thursday   \n",
       "2      5.0       5        2   arbol      NaN            Friday   \n",
       "3      NaN     NaN       23   arbol      4.0          Saturday   \n",
       "4      6.0       6       15   arbol      3.0            Sunday   \n",
       "5      8.0       8       30   arbol      2.0            Monday   \n",
       "\n",
       "  dia_semana_español  \n",
       "0          Miércoles  \n",
       "1             Jueves  \n",
       "2            Viernes  \n",
       "3             Sábado  \n",
       "4            Domingo  \n",
       "5              Lunes  "
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(merged_dia_semana, traduccion_dia_semana, on=\"dia_semana_ingles\", how=\"left\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}