{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "63dc4cd0-19bd-4dd3-af8a-dc995310c83f",
   "metadata": {},
   "source": [
    "# Predicción precio vivienda con RNA"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25664ead-b296-4ca1-baf6-69601912323d",
   "metadata": {},
   "source": [
    "Usaremos el set de datos SaratogaHouses del paquete mosaicData de R.  Incluye información de más de 1700 viviendas de NY (Estados Unidos) recogiendo información de 15 variables:\n",
    "\n",
    "    price: precio de la vivienda.\n",
    "    lotSize: metros cuadrados de la vivienda.\n",
    "    age: antigüedad de la vivienda.\n",
    "    landValue: valor del terreno.\n",
    "    livingArea: metros cuadrados habitables.\n",
    "    pctCollege: porcentaje del vecindario con título universitario.\n",
    "    bedrooms: número de dormitorios.\n",
    "    firplaces: número de chimeneas.\n",
    "    bathrooms: número de cuartos de baño (el valor 0.5 hace referencia a cuartos de baño sin ducha).\n",
    "    rooms: número de habitaciones.\n",
    "    heating: tipo de calefacción.\n",
    "    fuel: tipo de alimentación de la calefacción (gas, electricidad o diesel).\n",
    "    sewer: tipo de desagüe.\n",
    "    waterfront: si la vivienda tiene vistas al lago.\n",
    "    newConstruction: si la vivienda es de nueva construcción.\n",
    "    centralAir: si la vivienda tiene aire acondicionado.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fbf65cb1-26ff-4c5d-bde6-acc3555b0026",
   "metadata": {},
   "source": [
    "Veamos si podemos modelar el precio del alquiler en base a dichos datos y variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ff51c8f1-1419-425e-9cd2-d87b69816d8c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "plt.style.use('fivethirtyeight')\n",
    "\n",
    "from sklearn.neural_network import MLPRegressor\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.compose import make_column_selector\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn import set_config\n",
    "import multiprocessing\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3b2d9178-da67-421d-bf8a-ecb6e05826f0",
   "metadata": {},
   "outputs": [],
   "source": [
    "os.chdir (\"/home/mydoctor/Documents/03.Trabajos/01.C2B/21.Deep Learning - C2B (15h)/datos\")\n",
    "df = pd.read_csv(\"saratogaHousing.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1503be6f-cfde-43d9-8186-1371e28f668f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>lotSize</th>\n",
       "      <th>age</th>\n",
       "      <th>landValue</th>\n",
       "      <th>livingArea</th>\n",
       "      <th>pctCollege</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>fireplaces</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>rooms</th>\n",
       "      <th>heating</th>\n",
       "      <th>fuel</th>\n",
       "      <th>sewer</th>\n",
       "      <th>waterfront</th>\n",
       "      <th>newConstruction</th>\n",
       "      <th>centralAir</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>132500</td>\n",
       "      <td>0.09</td>\n",
       "      <td>42</td>\n",
       "      <td>50000</td>\n",
       "      <td>906</td>\n",
       "      <td>35</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>5</td>\n",
       "      <td>electric</td>\n",
       "      <td>electric</td>\n",
       "      <td>septic</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>181115</td>\n",
       "      <td>0.92</td>\n",
       "      <td>0</td>\n",
       "      <td>22300</td>\n",
       "      <td>1953</td>\n",
       "      <td>51</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>6</td>\n",
       "      <td>hot water/steam</td>\n",
       "      <td>gas</td>\n",
       "      <td>septic</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>109000</td>\n",
       "      <td>0.19</td>\n",
       "      <td>133</td>\n",
       "      <td>7300</td>\n",
       "      <td>1944</td>\n",
       "      <td>51</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8</td>\n",
       "      <td>hot water/steam</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>155000</td>\n",
       "      <td>0.41</td>\n",
       "      <td>13</td>\n",
       "      <td>18700</td>\n",
       "      <td>1944</td>\n",
       "      <td>51</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>5</td>\n",
       "      <td>hot air</td>\n",
       "      <td>gas</td>\n",
       "      <td>septic</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>86060</td>\n",
       "      <td>0.11</td>\n",
       "      <td>0</td>\n",
       "      <td>15000</td>\n",
       "      <td>840</td>\n",
       "      <td>51</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3</td>\n",
       "      <td>hot air</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1723</th>\n",
       "      <td>289000</td>\n",
       "      <td>0.38</td>\n",
       "      <td>32</td>\n",
       "      <td>24200</td>\n",
       "      <td>2310</td>\n",
       "      <td>61</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2.5</td>\n",
       "      <td>11</td>\n",
       "      <td>hot water/steam</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1724</th>\n",
       "      <td>285000</td>\n",
       "      <td>0.94</td>\n",
       "      <td>37</td>\n",
       "      <td>36200</td>\n",
       "      <td>2564</td>\n",
       "      <td>61</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2.5</td>\n",
       "      <td>11</td>\n",
       "      <td>hot water/steam</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1725</th>\n",
       "      <td>194900</td>\n",
       "      <td>0.39</td>\n",
       "      <td>9</td>\n",
       "      <td>20400</td>\n",
       "      <td>1099</td>\n",
       "      <td>51</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3</td>\n",
       "      <td>hot air</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1726</th>\n",
       "      <td>125000</td>\n",
       "      <td>0.24</td>\n",
       "      <td>48</td>\n",
       "      <td>16800</td>\n",
       "      <td>1225</td>\n",
       "      <td>51</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7</td>\n",
       "      <td>hot air</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1727</th>\n",
       "      <td>111300</td>\n",
       "      <td>0.59</td>\n",
       "      <td>86</td>\n",
       "      <td>26000</td>\n",
       "      <td>1959</td>\n",
       "      <td>51</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6</td>\n",
       "      <td>hot air</td>\n",
       "      <td>gas</td>\n",
       "      <td>septic</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1728 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       price  lotSize  age  landValue  livingArea  pctCollege  bedrooms  \\\n",
       "0     132500     0.09   42      50000         906          35         2   \n",
       "1     181115     0.92    0      22300        1953          51         3   \n",
       "2     109000     0.19  133       7300        1944          51         4   \n",
       "3     155000     0.41   13      18700        1944          51         3   \n",
       "4      86060     0.11    0      15000         840          51         2   \n",
       "...      ...      ...  ...        ...         ...         ...       ...   \n",
       "1723  289000     0.38   32      24200        2310          61         5   \n",
       "1724  285000     0.94   37      36200        2564          61         4   \n",
       "1725  194900     0.39    9      20400        1099          51         2   \n",
       "1726  125000     0.24   48      16800        1225          51         3   \n",
       "1727  111300     0.59   86      26000        1959          51         3   \n",
       "\n",
       "      fireplaces  bathrooms  rooms          heating      fuel  \\\n",
       "0              1        1.0      5         electric  electric   \n",
       "1              0        2.5      6  hot water/steam       gas   \n",
       "2              1        1.0      8  hot water/steam       gas   \n",
       "3              1        1.5      5          hot air       gas   \n",
       "4              0        1.0      3          hot air       gas   \n",
       "...          ...        ...    ...              ...       ...   \n",
       "1723           1        2.5     11  hot water/steam       gas   \n",
       "1724           1        2.5     11  hot water/steam       gas   \n",
       "1725           0        1.0      3          hot air       gas   \n",
       "1726           1        1.0      7          hot air       gas   \n",
       "1727           0        1.0      6          hot air       gas   \n",
       "\n",
       "                  sewer waterfront newConstruction centralAir  \n",
       "0                septic         No              No         No  \n",
       "1                septic         No              No         No  \n",
       "2     public/commercial         No              No         No  \n",
       "3                septic         No              No         No  \n",
       "4     public/commercial         No             Yes        Yes  \n",
       "...                 ...        ...             ...        ...  \n",
       "1723  public/commercial         No              No         No  \n",
       "1724  public/commercial         No              No         No  \n",
       "1725  public/commercial         No              No         No  \n",
       "1726  public/commercial         No              No         No  \n",
       "1727             septic         No              No         No  \n",
       "\n",
       "[1728 rows x 16 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f155927d-fed5-439a-aab5-9d4c23bdfbb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Renombramos las columnas\n",
    "df.columns = [\"precio\", \"metros_totales\", \"antiguedad\", \"precio_terreno\",\n",
    "                 \"metros_habitables\", \"universitarios\", \"dormitorios\", \n",
    "                 \"chimenea\", \"banyos\", \"habitaciones\", \"calefaccion\",\n",
    "                 \"consumo_calefacion\", \"desague\", \"vistas_lago\",\n",
    "                 \"nueva_construccion\", \"aire_acondicionado\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5dee7346-d018-4e59-8077-ecc6e26dd254",
   "metadata": {},
   "source": [
    "### EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "61777bd7-84ab-42fe-93da-6184cba2e73e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1728 entries, 0 to 1727\n",
      "Data columns (total 16 columns):\n",
      " #   Column              Non-Null Count  Dtype  \n",
      "---  ------              --------------  -----  \n",
      " 0   precio              1728 non-null   int64  \n",
      " 1   metros_totales      1728 non-null   float64\n",
      " 2   antiguedad          1728 non-null   int64  \n",
      " 3   precio_terreno      1728 non-null   int64  \n",
      " 4   metros_habitables   1728 non-null   int64  \n",
      " 5   universitarios      1728 non-null   int64  \n",
      " 6   dormitorios         1728 non-null   int64  \n",
      " 7   chimenea            1728 non-null   int64  \n",
      " 8   banyos              1728 non-null   float64\n",
      " 9   habitaciones        1728 non-null   int64  \n",
      " 10  calefaccion         1728 non-null   object \n",
      " 11  consumo_calefacion  1728 non-null   object \n",
      " 12  desague             1728 non-null   object \n",
      " 13  vistas_lago         1728 non-null   object \n",
      " 14  nueva_construccion  1728 non-null   object \n",
      " 15  aire_acondicionado  1728 non-null   object \n",
      "dtypes: float64(2), int64(8), object(6)\n",
      "memory usage: 216.1+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70a81377-9c64-4ec6-8488-b43c1b1a49da",
   "metadata": {},
   "source": [
    "Todos las columnas tienen los tipos correctos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "02e7d800-149b-4d6a-a83a-c131e6a66788",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "precio                0\n",
       "metros_totales        0\n",
       "antiguedad            0\n",
       "precio_terreno        0\n",
       "metros_habitables     0\n",
       "universitarios        0\n",
       "dormitorios           0\n",
       "chimenea              0\n",
       "banyos                0\n",
       "habitaciones          0\n",
       "calefaccion           0\n",
       "consumo_calefacion    0\n",
       "desague               0\n",
       "vistas_lago           0\n",
       "nueva_construccion    0\n",
       "aire_acondicionado    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum().sort_values()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ce36ffb-20b9-4ed3-b79e-d3844844a58a",
   "metadata": {},
   "source": [
    "No tenemos problemas de completitud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "d8b285f4-6e7b-4bc3-824b-d2a853ef086f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Comprobamos cómo se distruyen los datos de precio de venta\n",
    "\n",
    "fig, ax = plt.subplots(nrows=1, ncols=1, figsize=(6, 3))\n",
    "sns.histplot(df, x='precio', kde=True,ax=ax)\n",
    "sns.set_style(\"white\")\n",
    "ax.set_title(\"Distribución Precio\")\n",
    "ax.set_xlabel('precio');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "15d69203-a4e5-478f-8a00-ed5a14249f82",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Pintamos las variables numéricas\n",
    "\n",
    "fig, axes = plt.subplots(nrows=3, ncols=3, figsize=(20, 15))\n",
    "axes = axes.flat\n",
    "columnas_numeric = df.select_dtypes(include=['float64', 'int']).columns\n",
    "columnas_numeric = columnas_numeric.drop('precio')\n",
    "\n",
    "for i, colum in enumerate(columnas_numeric):\n",
    "    sns.histplot(\n",
    "        data    = df,\n",
    "        x       = colum,\n",
    "        stat    = \"count\",\n",
    "        kde     = True,\n",
    "        color   = (list(plt.rcParams['axes.prop_cycle'])*2)[i][\"color\"],\n",
    "        line_kws= {'linewidth': 2},\n",
    "        alpha   = 0.3,\n",
    "        ax      = axes[i]\n",
    "    )\n",
    "    axes[i].set_title(colum, fontsize = 15, fontweight = \"bold\")\n",
    "    axes[i].tick_params(labelsize = 15)\n",
    "    axes[i].set_xlabel(\"\")\n",
    "    axes[i].set_ylabel(\"\")\n",
    "    \n",
    "    \n",
    "fig.tight_layout()\n",
    "plt.subplots_adjust(top = 0.9)\n",
    "fig.suptitle('Distribución variables numéricas', fontsize = 10, fontweight = \"bold\");"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f70bb3ab-69fa-43f7-a4bd-72fc5edca2db",
   "metadata": {},
   "source": [
    "La variable chimenea adopta pocos estados, en casos como éstos es conveniente tratarla como categórica."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "2e5503b0-1841-4636-aff4-7e77370898d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    942\n",
       "0    740\n",
       "2     42\n",
       "4      2\n",
       "3      2\n",
       "Name: chimenea, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.chimenea = df.chimenea.astype(\"str\")\n",
    "df.chimenea.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d7086af4-1551-498f-943b-b81152ac2eed",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chimenea</th>\n",
       "      <th>calefaccion</th>\n",
       "      <th>consumo_calefacion</th>\n",
       "      <th>desague</th>\n",
       "      <th>vistas_lago</th>\n",
       "      <th>nueva_construccion</th>\n",
       "      <th>aire_acondicionado</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1728</td>\n",
       "      <td>1728</td>\n",
       "      <td>1728</td>\n",
       "      <td>1728</td>\n",
       "      <td>1728</td>\n",
       "      <td>1728</td>\n",
       "      <td>1728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>1</td>\n",
       "      <td>hot air</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>942</td>\n",
       "      <td>1121</td>\n",
       "      <td>1197</td>\n",
       "      <td>1213</td>\n",
       "      <td>1713</td>\n",
       "      <td>1647</td>\n",
       "      <td>1093</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       chimenea calefaccion consumo_calefacion            desague vistas_lago  \\\n",
       "count      1728        1728               1728               1728        1728   \n",
       "unique        5           3                  3                  3           2   \n",
       "top           1     hot air                gas  public/commercial          No   \n",
       "freq        942        1121               1197               1213        1713   \n",
       "\n",
       "       nueva_construccion aire_acondicionado  \n",
       "count                1728               1728  \n",
       "unique                  2                  2  \n",
       "top                    No                 No  \n",
       "freq                 1647               1093  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Comprobamos y pintamos las variables categóricas\n",
    "df.select_dtypes(include=['object']).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3b8b9b3e-1dd0-4850-baeb-5d1f8b41e71d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 7 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=3, ncols=3, figsize=(12, 5))\n",
    "axes = axes.flat\n",
    "columnas_object = df.select_dtypes(include=['object']).columns\n",
    "\n",
    "for i, colum in enumerate(columnas_object):\n",
    "    df[colum].value_counts().plot.barh(ax = axes[i])\n",
    "    axes[i].set_title(colum, fontsize = 14)\n",
    "    axes[i].set_xlabel(\"\")\n",
    "\n",
    "# Se eliminan los axes vacíos\n",
    "for i in [7, 8]:\n",
    "    fig.delaxes(axes[i])\n",
    "    \n",
    "fig.tight_layout()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a3c46ab5-3600-4614-a908-bc40338b9836",
   "metadata": {},
   "source": [
    "Si alguno de los niveles de una variable cualitativa tiene muy pocas observaciones en comparación a los otros niveles, puede ocurrir que, durante la validación cruzada o bootstrapping, algunas particiones no contengan ninguna observación de dicha clase (varianza cero), lo que puede dar lugar a errores. Para este caso, hay que tener precaución con la variable chimenea. Se unifican los niveles de 2, 3 y 4 en un nuevo nivel llamado \"2_mas\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "1fde20d6-0f67-423e-8568-bc918fef0a72",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        740\n",
       "1        942\n",
       "2_mas     46\n",
       "Name: chimenea, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mapeo = {'2': \"2_mas\",\n",
    "               '3': \"2_mas\",\n",
    "               '4': \"2_mas\"}\n",
    "\n",
    "df['chimenea'] = df['chimenea'] .map(mapeo).fillna(df['chimenea'])\n",
    "\n",
    "df.chimenea.value_counts().sort_index()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae3288ce-0f09-4655-9259-9ddf8b25f549",
   "metadata": {},
   "source": [
    "### División de los datos en Train y Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d82f6b0a-64dd-4c41-a830-cdefdd3937da",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "                                        df.drop('precio', axis = 'columns'),\n",
    "                                        df['precio'],\n",
    "                                        train_size   = 0.7,\n",
    "                                        random_state = 123,\n",
    "                                        shuffle      = True\n",
    "                                    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a182e5bb-28c3-4fa1-8f65-90db95b3b8e0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Partición de entrenamento\n",
      "-----------------------\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "count      1209.000000\n",
       "mean     212474.237386\n",
       "std       97233.331785\n",
       "min        5000.000000\n",
       "25%      145000.000000\n",
       "50%      191500.000000\n",
       "75%      259900.000000\n",
       "max      775000.000000\n",
       "Name: precio, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>metros_totales</th>\n",
       "      <th>antiguedad</th>\n",
       "      <th>precio_terreno</th>\n",
       "      <th>metros_habitables</th>\n",
       "      <th>universitarios</th>\n",
       "      <th>dormitorios</th>\n",
       "      <th>banyos</th>\n",
       "      <th>habitaciones</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1209.000000</td>\n",
       "      <td>1209.000000</td>\n",
       "      <td>1209.000000</td>\n",
       "      <td>1209.000000</td>\n",
       "      <td>1209.000000</td>\n",
       "      <td>1209.000000</td>\n",
       "      <td>1209.000000</td>\n",
       "      <td>1209.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.510993</td>\n",
       "      <td>26.751034</td>\n",
       "      <td>34671.025641</td>\n",
       "      <td>1778.014888</td>\n",
       "      <td>55.506203</td>\n",
       "      <td>3.167907</td>\n",
       "      <td>1.923904</td>\n",
       "      <td>7.097601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.731041</td>\n",
       "      <td>28.285677</td>\n",
       "      <td>34460.420428</td>\n",
       "      <td>636.602289</td>\n",
       "      <td>10.214999</td>\n",
       "      <td>0.817298</td>\n",
       "      <td>0.658301</td>\n",
       "      <td>2.365677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.010000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>200.000000</td>\n",
       "      <td>616.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.170000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>15100.000000</td>\n",
       "      <td>1304.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.370000</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>25000.000000</td>\n",
       "      <td>1656.000000</td>\n",
       "      <td>57.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.530000</td>\n",
       "      <td>33.000000</td>\n",
       "      <td>42000.000000</td>\n",
       "      <td>2185.000000</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>12.200000</td>\n",
       "      <td>225.000000</td>\n",
       "      <td>250000.000000</td>\n",
       "      <td>5228.000000</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>12.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       metros_totales   antiguedad  precio_terreno  metros_habitables  \\\n",
       "count     1209.000000  1209.000000     1209.000000        1209.000000   \n",
       "mean         0.510993    26.751034    34671.025641        1778.014888   \n",
       "std          0.731041    28.285677    34460.420428         636.602289   \n",
       "min          0.010000     0.000000      200.000000         616.000000   \n",
       "25%          0.170000    12.000000    15100.000000        1304.000000   \n",
       "50%          0.370000    19.000000    25000.000000        1656.000000   \n",
       "75%          0.530000    33.000000    42000.000000        2185.000000   \n",
       "max         12.200000   225.000000   250000.000000        5228.000000   \n",
       "\n",
       "       universitarios  dormitorios       banyos  habitaciones  \n",
       "count     1209.000000  1209.000000  1209.000000   1209.000000  \n",
       "mean        55.506203     3.167907     1.923904      7.097601  \n",
       "std         10.214999     0.817298     0.658301      2.365677  \n",
       "min         20.000000     1.000000     1.000000      2.000000  \n",
       "25%         52.000000     3.000000     1.500000      5.000000  \n",
       "50%         57.000000     3.000000     2.000000      7.000000  \n",
       "75%         63.000000     4.000000     2.500000      9.000000  \n",
       "max         82.000000     7.000000     4.500000     12.000000  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chimenea</th>\n",
       "      <th>calefaccion</th>\n",
       "      <th>consumo_calefacion</th>\n",
       "      <th>desague</th>\n",
       "      <th>vistas_lago</th>\n",
       "      <th>nueva_construccion</th>\n",
       "      <th>aire_acondicionado</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1209</td>\n",
       "      <td>1209</td>\n",
       "      <td>1209</td>\n",
       "      <td>1209</td>\n",
       "      <td>1209</td>\n",
       "      <td>1209</td>\n",
       "      <td>1209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>1</td>\n",
       "      <td>hot air</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>667</td>\n",
       "      <td>779</td>\n",
       "      <td>842</td>\n",
       "      <td>853</td>\n",
       "      <td>1202</td>\n",
       "      <td>1149</td>\n",
       "      <td>761</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       chimenea calefaccion consumo_calefacion            desague vistas_lago  \\\n",
       "count      1209        1209               1209               1209        1209   \n",
       "unique        3           3                  3                  3           2   \n",
       "top           1     hot air                gas  public/commercial          No   \n",
       "freq        667         779                842                853        1202   \n",
       "\n",
       "       nueva_construccion aire_acondicionado  \n",
       "count                1209               1209  \n",
       "unique                  2                  2  \n",
       "top                    No                 No  \n",
       "freq                 1149                761  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " \n",
      "Partición de test\n",
      "-----------------------\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "count       519.000000\n",
       "mean     210784.420039\n",
       "std      101285.133770\n",
       "min       20000.000000\n",
       "25%      145000.000000\n",
       "50%      185000.000000\n",
       "75%      255000.000000\n",
       "max      775000.000000\n",
       "Name: precio, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>metros_totales</th>\n",
       "      <th>antiguedad</th>\n",
       "      <th>precio_terreno</th>\n",
       "      <th>metros_habitables</th>\n",
       "      <th>universitarios</th>\n",
       "      <th>dormitorios</th>\n",
       "      <th>banyos</th>\n",
       "      <th>habitaciones</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>519.000000</td>\n",
       "      <td>519.000000</td>\n",
       "      <td>519.000000</td>\n",
       "      <td>519.000000</td>\n",
       "      <td>519.000000</td>\n",
       "      <td>519.000000</td>\n",
       "      <td>519.000000</td>\n",
       "      <td>519.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.475106</td>\n",
       "      <td>30.630058</td>\n",
       "      <td>34292.003854</td>\n",
       "      <td>1701.306358</td>\n",
       "      <td>55.710983</td>\n",
       "      <td>3.123314</td>\n",
       "      <td>1.844894</td>\n",
       "      <td>6.911368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.616682</td>\n",
       "      <td>31.116823</td>\n",
       "      <td>36326.430790</td>\n",
       "      <td>576.279936</td>\n",
       "      <td>10.613302</td>\n",
       "      <td>0.817412</td>\n",
       "      <td>0.655779</td>\n",
       "      <td>2.194202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>600.000000</td>\n",
       "      <td>672.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.170000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>15150.000000</td>\n",
       "      <td>1291.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.380000</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>25000.000000</td>\n",
       "      <td>1604.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.565000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>38050.000000</td>\n",
       "      <td>2073.000000</td>\n",
       "      <td>64.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>8.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>8.970000</td>\n",
       "      <td>169.000000</td>\n",
       "      <td>412600.000000</td>\n",
       "      <td>4128.000000</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>12.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       metros_totales  antiguedad  precio_terreno  metros_habitables  \\\n",
       "count      519.000000  519.000000      519.000000         519.000000   \n",
       "mean         0.475106   30.630058    34292.003854        1701.306358   \n",
       "std          0.616682   31.116823    36326.430790         576.279936   \n",
       "min          0.000000    0.000000      600.000000         672.000000   \n",
       "25%          0.170000   14.000000    15150.000000        1291.000000   \n",
       "50%          0.380000   19.000000    25000.000000        1604.000000   \n",
       "75%          0.565000   35.000000    38050.000000        2073.000000   \n",
       "max          8.970000  169.000000   412600.000000        4128.000000   \n",
       "\n",
       "       universitarios  dormitorios      banyos  habitaciones  \n",
       "count      519.000000   519.000000  519.000000    519.000000  \n",
       "mean        55.710983     3.123314    1.844894      6.911368  \n",
       "std         10.613302     0.817412    0.655779      2.194202  \n",
       "min         20.000000     1.000000    0.000000      3.000000  \n",
       "25%         52.000000     3.000000    1.500000      5.000000  \n",
       "50%         60.000000     3.000000    1.500000      7.000000  \n",
       "75%         64.000000     4.000000    2.500000      8.000000  \n",
       "max         82.000000     7.000000    4.000000     12.000000  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>chimenea</th>\n",
       "      <th>calefaccion</th>\n",
       "      <th>consumo_calefacion</th>\n",
       "      <th>desague</th>\n",
       "      <th>vistas_lago</th>\n",
       "      <th>nueva_construccion</th>\n",
       "      <th>aire_acondicionado</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>519</td>\n",
       "      <td>519</td>\n",
       "      <td>519</td>\n",
       "      <td>519</td>\n",
       "      <td>519</td>\n",
       "      <td>519</td>\n",
       "      <td>519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>1</td>\n",
       "      <td>hot air</td>\n",
       "      <td>gas</td>\n",
       "      <td>public/commercial</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>275</td>\n",
       "      <td>342</td>\n",
       "      <td>355</td>\n",
       "      <td>360</td>\n",
       "      <td>511</td>\n",
       "      <td>498</td>\n",
       "      <td>332</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       chimenea calefaccion consumo_calefacion            desague vistas_lago  \\\n",
       "count       519         519                519                519         519   \n",
       "unique        3           3                  3                  3           2   \n",
       "top           1     hot air                gas  public/commercial          No   \n",
       "freq        275         342                355                360         511   \n",
       "\n",
       "       nueva_construccion aire_acondicionado  \n",
       "count                 519                519  \n",
       "unique                  2                  2  \n",
       "top                    No                 No  \n",
       "freq                  498                332  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Partición de entrenamento\")\n",
    "print(\"-----------------------\")\n",
    "display(y_train.describe())\n",
    "display(X_train.describe())\n",
    "display(X_train.describe(include = 'object'))\n",
    "print(\" \")\n",
    "\n",
    "print(\"Partición de test\")\n",
    "print(\"-----------------------\")\n",
    "display(y_test.describe())\n",
    "display(X_test.describe())\n",
    "display(X_test.describe(include = 'object'))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fbbf9c5d-3dd8-4702-98a3-d9c74a1c4f01",
   "metadata": {},
   "source": [
    "### Preprocesado y modelado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "ac8f7ff1-8e1c-4a29-9ac5-7982544e8a40",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Identificación de columnas numéricas y categóricas\n",
    "numeric_cols = X_train.select_dtypes(include=['float64', 'int']).columns.to_list()\n",
    "cat_cols = X_train.select_dtypes(include=['object', 'category']).columns.to_list()\n",
    "\n",
    "\n",
    "# Transformaciones para las variables numéricas\n",
    "numeric_transformer = Pipeline(\n",
    "                        steps=[('scaler', StandardScaler())])\n",
    "\n",
    "# Transformaciones para las variables categóricas\n",
    "categorical_transformer = Pipeline(\n",
    "                            steps=[('onehot', OneHotEncoder(handle_unknown='ignore'))])\n",
    "\n",
    "# Para tener en cuenta aquellas columnas que no requieren transformación, necesitamos usar el argumento reminder=passthrough\n",
    "preprocessor = ColumnTransformer(\n",
    "                    transformers=[\n",
    "                        ('numeric', numeric_transformer, numeric_cols),\n",
    "                        ('cat', categorical_transformer, cat_cols)],\n",
    "                    remainder='passthrough')\n",
    "\n",
    "# Combinamos los pasos de preprocesado y el modelo en un mismo pipeline\n",
    "pipe = Pipeline([('preprocessing', preprocessor),\n",
    "                 ('modelo', MLPRegressor(solver = 'lbfgs', max_iter= 6000))])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "054235ad-aeab-4006-b096-41108cf5c538",
   "metadata": {},
   "source": [
    "### Validación cruzada y búsqueda de hiperparámetros"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "6e21507b-aef2-4356-8fc7-add90ba76200",
   "metadata": {},
   "outputs": [],
   "source": [
    "hiperparametros = {\n",
    "    'modelo__hidden_layer_sizes': [(10), (20), (10, 10)],\n",
    "    'modelo__alpha': np.logspace(-3, 3, 10),\n",
    "    'modelo__learning_rate_init': [0.001, 0.01]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "d7f94870-5349-4fc4-a23a-573756d721cf",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5,\n",
       "                   estimator=Pipeline(steps=[('preprocessing',\n",
       "                                              ColumnTransformer(remainder='passthrough',\n",
       "                                                                transformers=[('numeric',\n",
       "                                                                               Pipeline(steps=[('scaler',\n",
       "                                                                                                StandardScaler())]),\n",
       "                                                                               ['metros_totales',\n",
       "                                                                                'antiguedad',\n",
       "                                                                                'precio_terreno',\n",
       "                                                                                'metros_habitables',\n",
       "                                                                                'universitarios',\n",
       "                                                                                'dormitorios',\n",
       "                                                                                'banyos',\n",
       "                                                                                'habitaciones']),\n",
       "                                                                              ('cat',\n",
       "                                                                               Pipeline(steps=[('onehot',\n",
       "                                                                                                OneHotEncode...\n",
       "                   param_distributions={'modelo__alpha': array([1.00000000e-03, 4.64158883e-03, 2.15443469e-02, 1.00000000e-01,\n",
       "       4.64158883e-01, 2.15443469e+00, 1.00000000e+01, 4.64158883e+01,\n",
       "       2.15443469e+02, 1.00000000e+03]),\n",
       "                                        'modelo__hidden_layer_sizes': [10, 20,\n",
       "                                                                       (10,\n",
       "                                                                        10)],\n",
       "                                        'modelo__learning_rate_init': [0.001,\n",
       "                                                                       0.01]},\n",
       "                   random_state=123, return_train_score=True,\n",
       "                   scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid = RandomizedSearchCV(\n",
    "        estimator  = pipe,\n",
    "        param_distributions = hiperparametros,\n",
    "        n_iter     = 50,\n",
    "        scoring    = 'neg_mean_squared_error',\n",
    "        n_jobs     = multiprocessing.cpu_count() - 1,\n",
    "        cv         = 5, \n",
    "        verbose    = 0,\n",
    "        random_state = 123,\n",
    "        return_train_score = True)\n",
    "\n",
    "grid.fit(X = X_train, y = y_train)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eccc5c51-43ee-4a09-bbd1-e070d54a58a0",
   "metadata": {},
   "source": [
    "### Resultados y evaluación final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "9d14a1ba-6b8b-4349-a9ce-bf83c56e95a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>param_modelo__learning_rate_init</th>\n",
       "      <th>param_modelo__hidden_layer_sizes</th>\n",
       "      <th>param_modelo__alpha</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "      <th>std_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0.01</td>\n",
       "      <td>10</td>\n",
       "      <td>0.001</td>\n",
       "      <td>-3.230825e+09</td>\n",
       "      <td>3.616289e+08</td>\n",
       "      <td>-2.450421e+09</td>\n",
       "      <td>1.949371e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.001</td>\n",
       "      <td>10</td>\n",
       "      <td>0.1</td>\n",
       "      <td>-3.288593e+09</td>\n",
       "      <td>2.684194e+08</td>\n",
       "      <td>-2.476251e+09</td>\n",
       "      <td>5.500066e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.001</td>\n",
       "      <td>10</td>\n",
       "      <td>0.021544</td>\n",
       "      <td>-3.301317e+09</td>\n",
       "      <td>4.095670e+08</td>\n",
       "      <td>-2.520127e+09</td>\n",
       "      <td>1.827820e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.01</td>\n",
       "      <td>10</td>\n",
       "      <td>2.154435</td>\n",
       "      <td>-3.302082e+09</td>\n",
       "      <td>2.894838e+08</td>\n",
       "      <td>-2.297498e+09</td>\n",
       "      <td>1.180598e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.01</td>\n",
       "      <td>10</td>\n",
       "      <td>46.415888</td>\n",
       "      <td>-3.304760e+09</td>\n",
       "      <td>3.596774e+08</td>\n",
       "      <td>-2.435577e+09</td>\n",
       "      <td>4.262295e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.001</td>\n",
       "      <td>10</td>\n",
       "      <td>0.001</td>\n",
       "      <td>-3.332898e+09</td>\n",
       "      <td>4.559129e+08</td>\n",
       "      <td>-2.445700e+09</td>\n",
       "      <td>2.004258e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.01</td>\n",
       "      <td>10</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>-3.362608e+09</td>\n",
       "      <td>3.908840e+08</td>\n",
       "      <td>-2.323996e+09</td>\n",
       "      <td>1.783919e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>0.01</td>\n",
       "      <td>10</td>\n",
       "      <td>0.004642</td>\n",
       "      <td>-3.421465e+09</td>\n",
       "      <td>5.413912e+08</td>\n",
       "      <td>-2.525470e+09</td>\n",
       "      <td>1.977407e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>0.001</td>\n",
       "      <td>10</td>\n",
       "      <td>0.004642</td>\n",
       "      <td>-3.426928e+09</td>\n",
       "      <td>3.088534e+08</td>\n",
       "      <td>-2.337111e+09</td>\n",
       "      <td>1.130859e+08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>0.01</td>\n",
       "      <td>10</td>\n",
       "      <td>0.464159</td>\n",
       "      <td>-3.437133e+09</td>\n",
       "      <td>4.856893e+08</td>\n",
       "      <td>-2.276094e+09</td>\n",
       "      <td>2.748345e+08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   param_modelo__learning_rate_init param_modelo__hidden_layer_sizes  \\\n",
       "25                             0.01                               10   \n",
       "6                             0.001                               10   \n",
       "20                            0.001                               10   \n",
       "3                              0.01                               10   \n",
       "4                              0.01                               10   \n",
       "15                            0.001                               10   \n",
       "8                              0.01                               10   \n",
       "32                             0.01                               10   \n",
       "26                            0.001                               10   \n",
       "44                             0.01                               10   \n",
       "\n",
       "   param_modelo__alpha  mean_test_score  std_test_score  mean_train_score  \\\n",
       "25               0.001    -3.230825e+09    3.616289e+08     -2.450421e+09   \n",
       "6                  0.1    -3.288593e+09    2.684194e+08     -2.476251e+09   \n",
       "20            0.021544    -3.301317e+09    4.095670e+08     -2.520127e+09   \n",
       "3             2.154435    -3.302082e+09    2.894838e+08     -2.297498e+09   \n",
       "4            46.415888    -3.304760e+09    3.596774e+08     -2.435577e+09   \n",
       "15               0.001    -3.332898e+09    4.559129e+08     -2.445700e+09   \n",
       "8               1000.0    -3.362608e+09    3.908840e+08     -2.323996e+09   \n",
       "32            0.004642    -3.421465e+09    5.413912e+08     -2.525470e+09   \n",
       "26            0.004642    -3.426928e+09    3.088534e+08     -2.337111e+09   \n",
       "44            0.464159    -3.437133e+09    4.856893e+08     -2.276094e+09   \n",
       "\n",
       "    std_train_score  \n",
       "25     1.949371e+08  \n",
       "6      5.500066e+07  \n",
       "20     1.827820e+08  \n",
       "3      1.180598e+08  \n",
       "4      4.262295e+08  \n",
       "15     2.004258e+08  \n",
       "8      1.783919e+08  \n",
       "32     1.977407e+08  \n",
       "26     1.130859e+08  \n",
       "44     2.748345e+08  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "resultados = pd.DataFrame(grid.cv_results_)\n",
    "resultados.filter(regex = '(param.*|mean_t|std_t)').drop(columns = 'params').sort_values('mean_test_score', ascending = False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "40f0f187-5af6-4b61-a0df-e23c7414baa1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Error de test (rmse):  67771.2859907507\n"
     ]
    }
   ],
   "source": [
    "modelo_final = grid.best_estimator_\n",
    "predicciones = modelo_final.predict(X = X_test)\n",
    "rmse = mean_squared_error(\n",
    "        y_true = y_test,\n",
    "        y_pred = predicciones,\n",
    "        squared = False\n",
    "       )\n",
    "print('Error de test (rmse): ', rmse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "79006a66-86db-4389-ad3d-48ff1f9caa1a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'activation': 'relu',\n",
       " 'alpha': 0.001,\n",
       " 'batch_size': 'auto',\n",
       " 'beta_1': 0.9,\n",
       " 'beta_2': 0.999,\n",
       " 'early_stopping': False,\n",
       " 'epsilon': 1e-08,\n",
       " 'hidden_layer_sizes': 10,\n",
       " 'learning_rate': 'constant',\n",
       " 'learning_rate_init': 0.01,\n",
       " 'max_fun': 15000,\n",
       " 'max_iter': 6000,\n",
       " 'momentum': 0.9,\n",
       " 'n_iter_no_change': 10,\n",
       " 'nesterovs_momentum': True,\n",
       " 'power_t': 0.5,\n",
       " 'random_state': None,\n",
       " 'shuffle': True,\n",
       " 'solver': 'lbfgs',\n",
       " 'tol': 0.0001,\n",
       " 'validation_fraction': 0.1,\n",
       " 'verbose': False,\n",
       " 'warm_start': False}"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "modelo_final['modelo'].get_params()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}