Titanic.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Titanic\n",
    "Este dataset contendrá los detalles de un subconjunto de los pasajeros a bordo (891 para ser exactos) del Titanic, lo que es más importante, revelará si sobrevivieron o no. Algunos datos están perdidos, por lo que habrá que realizar un proceso de limpieza."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "titanic_df = pd.read_csv('titanic_dataset.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "\n",
       "                                                Name     Sex   Age  SibSp  \\\n",
       "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
       "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
       "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
       "4                           Allen, Mr. William Henry    male  35.0      0   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  \n",
       "0      0         A/5 21171   7.2500   NaN        S  \n",
       "1      0          PC 17599  71.2833   C85        C  \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S  \n",
       "3      0            113803  53.1000  C123        S  \n",
       "4      0            373450   8.0500   NaN        S  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titanic_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Campos del dataset\n",
    "- PassengerId: Id único de cada pasajero.\n",
    "- **Survived**: Determina si un pasajero sobrevivió (1) o no (0). *Este será nuestro conjunto y*\n",
    "- Pclass: Clase del tiquete.\n",
    "- Name: Nombre del pasajero.\n",
    "- Sex: Sexo del pasajero.\n",
    "- Age: Edad del pasajero.\n",
    "- SibSp: # de hermanas / cónyuges a bordo del Titanic.\n",
    "- Parch: # de padres / hijos a bordo del Titanic.\n",
    "- Ticket: Número del tiquete.\n",
    "- Fare: Valor pagado por el tiquete.\n",
    "- Cabin: Indica cubierta donde estaba ubicado el pasajero.\n",
    "- Embarked: Puerto de embarcación al Titanic."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Proceso de limpieza y modificación de datos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cast de datos de algunas columnas\n",
    "- Sex\n",
    "- Embarked"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "titanic_df['Sex'].replace(['female', 'male'], [0, 1], inplace=True)\n",
    "titanic_df['Embarked'].replace(['Q', 'S', 'C'], [0, 1, 2], inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Datos vacíos de edad\n",
    "Se reemplazan por la media de las edades."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "age_mean = titanic_df['Age'].mean()\n",
    "titanic_df['Age'] = titanic_df['Age'].replace(np.nan, age_mean)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Segmentos de edades a un solo valor\n",
    "Se reemplazan los segmentos tipo 8-15 por un solo valor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "bins = [0, 8, 15, 18, 25, 40, 60, 100]\n",
    "values = ['1', '2', '3', '4', '5', '6', '7']\n",
    "titanic_df['Age'] = pd.cut(titanic_df['Age'], bins, labels=values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Eliminación columnas\n",
    "- Se elimina la columna Cabin puesto que tiene muchos datos perdidos.\n",
    "- Se elimina la columna Name porque no influye en el análisis.\n",
    "- Se elimina la columna Ticket porque no influye en el análisis.\n",
    "- Se elimina la columna PassengerId porque no influye en el análisis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Datos perdidos = 687\n"
     ]
    }
   ],
   "source": [
    "print('Datos perdidos = ' + str(titanic_df['Cabin'].isnull().sum()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "titanic_df.drop(['Cabin'], axis=1, inplace=True)\n",
    "titanic_df.drop(['Name'], axis=1, inplace=True)\n",
    "titanic_df.drop(['Ticket'], axis=1, inplace=True)\n",
    "titanic_df.drop(['PassengerId'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Eliminación filas\n",
    "Se eliminan las filas con valores vacíos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "titanic_df.dropna(axis=0, how='any', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Información básica del dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 889 entries, 0 to 890\n",
      "Data columns (total 8 columns):\n",
      " #   Column    Non-Null Count  Dtype   \n",
      "---  ------    --------------  -----   \n",
      " 0   Survived  889 non-null    int64   \n",
      " 1   Pclass    889 non-null    int64   \n",
      " 2   Sex       889 non-null    int64   \n",
      " 3   Age       889 non-null    category\n",
      " 4   SibSp     889 non-null    int64   \n",
      " 5   Parch     889 non-null    int64   \n",
      " 6   Fare      889 non-null    float64 \n",
      " 7   Embarked  889 non-null    float64 \n",
      "dtypes: category(1), float64(2), int64(5)\n",
      "memory usage: 56.8 KB\n"
     ]
    }
   ],
   "source": [
    "titanic_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Columnas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Survived', 'Pclass', 'Sex', 'Age', 'SibSp', 'Parch', 'Fare',\n",
       "       'Embarked'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titanic_df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape: (889, 8)\n"
     ]
    }
   ],
   "source": [
    "print('Shape: ' + str(titanic_df.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Algunos datos estadísticos\n",
    "Solo de las columnas numéricas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "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>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>889.000000</td>\n",
       "      <td>889.000000</td>\n",
       "      <td>889.000000</td>\n",
       "      <td>889.000000</td>\n",
       "      <td>889.000000</td>\n",
       "      <td>889.000000</td>\n",
       "      <td>889.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.382452</td>\n",
       "      <td>2.311586</td>\n",
       "      <td>0.649044</td>\n",
       "      <td>0.524184</td>\n",
       "      <td>0.382452</td>\n",
       "      <td>32.096681</td>\n",
       "      <td>1.102362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.486260</td>\n",
       "      <td>0.834700</td>\n",
       "      <td>0.477538</td>\n",
       "      <td>1.103705</td>\n",
       "      <td>0.806761</td>\n",
       "      <td>49.697504</td>\n",
       "      <td>0.515181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.895800</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.454200</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>512.329200</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Survived      Pclass         Sex       SibSp       Parch        Fare  \\\n",
       "count  889.000000  889.000000  889.000000  889.000000  889.000000  889.000000   \n",
       "mean     0.382452    2.311586    0.649044    0.524184    0.382452   32.096681   \n",
       "std      0.486260    0.834700    0.477538    1.103705    0.806761   49.697504   \n",
       "min      0.000000    1.000000    0.000000    0.000000    0.000000    0.000000   \n",
       "25%      0.000000    2.000000    0.000000    0.000000    0.000000    7.895800   \n",
       "50%      0.000000    3.000000    1.000000    0.000000    0.000000   14.454200   \n",
       "75%      1.000000    3.000000    1.000000    1.000000    0.000000   31.000000   \n",
       "max      1.000000    3.000000    1.000000    8.000000    6.000000  512.329200   \n",
       "\n",
       "         Embarked  \n",
       "count  889.000000  \n",
       "mean     1.102362  \n",
       "std      0.515181  \n",
       "min      0.000000  \n",
       "25%      1.000000  \n",
       "50%      1.000000  \n",
       "75%      1.000000  \n",
       "max      2.000000  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titanic_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Machine Learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "#Métodos\n",
    "from sklearn.linear_model import LogisticRegression, SGDClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "#Reportes\n",
    "from sklearn.metrics import classification_report, accuracy_score, precision_score\n",
    "from sklearn.metrics import confusion_matrix, f1_score, mean_squared_error, roc_auc_score\n",
    "#Tiempo\n",
    "import time\n",
    "#Graficas\n",
    "import matplotlib.pyplot as plt\n",
    "import csv\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Separación X y y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.array(titanic_df.drop(['Survived'], 1))\n",
    "y = np.array(titanic_df['Survived'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(889, 7)\n",
      "(889,)\n"
     ]
    }
   ],
   "source": [
    "print(X.shape)\n",
    "print(y.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Se separan los datos en train y test."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(622, 7)\n",
      "(267, 7)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print(X_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Métricas\n",
    "Esta función se define para tener los valores de tiempo de ejecución, matriz de confusión, score en el entrenamiento, accuracy, precisiom, sensivity, specificity,  así como f1 score, y MSE. Todos estos valores se entregan se imprimen y se almacenan en un .csv (Diferenciando cada método)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "def metricas (X_train, y_train, y_test, y_pred, time_ml, method):\n",
    "    time_exec = str(time_ml)\n",
    "    print('Tiempo de ejecución = ' + time_exec)\n",
    "    print('\\n')\n",
    "    print('Matriz de confusión')\n",
    "    tn, fp, fn, tp = confusion_matrix(y_test, y_pred).ravel()\n",
    "    print(confusion_matrix(y_test, y_pred))\n",
    "    print('\\n')\n",
    "    if method == 'reg_log':\n",
    "        score_train = str(log_reg.score(X_train, y_train))\n",
    "    elif method == 'svm':\n",
    "        score_train = str(svc.score(X_train, y_train))\n",
    "    elif method == 'knn':\n",
    "        score_train = str(knn.score(X_train, y_train))\n",
    "    elif method == 'rand_for':\n",
    "        score_train = str(rfm.score(X_train, y_train))\n",
    "    elif method == 'sgd':\n",
    "        score_train = str(sgd.score(X_train, y_train))\n",
    "    print('Score train = ' + score_train)\n",
    "    print('\\n')\n",
    "    acc_test = str(accuracy_score(y_test, y_pred))\n",
    "    print('Accuracy test = ' + acc_test)\n",
    "    prec_test = str(precision_score(y_test, y_pred))\n",
    "    print('Precision test = ' + prec_test)\n",
    "    sens_test = str(tp/(tp+fn))\n",
    "    print('Sensivity test = ' + sens_test)\n",
    "    spec_test = str(tn/(tn+fp))\n",
    "    print('Specificity test = ' + spec_test)\n",
    "    \n",
    "    f1_scor = str(f1_score(y_test, y_pred))\n",
    "    print('F1 Score = ' + f1_scor)\n",
    "    mserror = str(mean_squared_error(y_test, y_pred))\n",
    "    print('Mean Squared Error = ' + mserror)\n",
    "    \n",
    "    metricas = [score_train, acc_test, prec_test, sens_test, spec_test, f1_scor, mserror, time_exec, method]\n",
    "    #print(metricas)\n",
    "    headers = ['score_train', 'acc_test','prec_test', 'sens_test', 'spec_test', 'f1_scor', 'mserror', 'time', 'method']\n",
    "    metricas_df = pd.DataFrame([metricas], columns = headers)\n",
    "    metricas_df.head()\n",
    "    \n",
    "    with open('results.csv', mode='a') as write_file:\n",
    "        writer_file = csv.writer(write_file, delimiter=',')\n",
    "        writer_file.writerow(metricas)\n",
    "        #df_results = pd.DataFrame.to_csv(metricas_df, 'results.csv')\n",
    "        \n",
    "def labels_csv():\n",
    "    headers = ['score_train', 'acc_test','prec_test', 'sens_test', 'spec_test', 'f1_scor', 'mserror', 'time', 'method']\n",
    "    with open('results.csv', mode='w') as write_file:\n",
    "        writer_file = csv.writer(write_file, delimiter=',')\n",
    "        writer_file.writerow(headers)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regresión logística"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "start = time.time()\n",
    "\n",
    "log_reg = LogisticRegression(max_iter=1000)\n",
    "log_reg.fit(X_train, y_train)\n",
    "y_pred = log_reg.predict(X_test)\n",
    "\n",
    "end = time.time()\n",
    "time_ml = end - start"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tiempo de ejecución = 0.03164935111999512\n",
      "\n",
      "\n",
      "Matriz de confusión\n",
      "[[136  21]\n",
      " [ 34  76]]\n",
      "\n",
      "\n",
      "Score train = 0.8263665594855305\n",
      "\n",
      "\n",
      "Accuracy test = 0.7940074906367042\n",
      "Precision test = 0.7835051546391752\n",
      "Sensivity test = 0.6909090909090909\n",
      "Specificity test = 0.8662420382165605\n",
      "F1 Score = 0.7342995169082126\n",
      "Mean Squared Error = 0.20599250936329588\n"
     ]
    }
   ],
   "source": [
    "labels_csv()\n",
    "metricas(X_train, y_train, y_test, y_pred, time_ml, 'reg_log')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Área bajo la curva"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7785755645628257"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Support Vector Machines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "start = time.time()\n",
    "\n",
    "svc = SVC(kernel='linear')\n",
    "svc.fit(X_train, y_train)\n",
    "y_pred = svc.predict(X_test)\n",
    "\n",
    "end = time.time()\n",
    "time_ml = end - start"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tiempo de ejecución = 0.9122257232666016\n",
      "\n",
      "\n",
      "Matriz de confusión\n",
      "[[132  25]\n",
      " [ 36  74]]\n",
      "\n",
      "\n",
      "Score train = 0.8022508038585209\n",
      "\n",
      "\n",
      "Accuracy test = 0.7715355805243446\n",
      "Precision test = 0.7474747474747475\n",
      "Sensivity test = 0.6727272727272727\n",
      "Specificity test = 0.8407643312101911\n",
      "F1 Score = 0.708133971291866\n",
      "Mean Squared Error = 0.22846441947565543\n"
     ]
    }
   ],
   "source": [
    "metricas(X_train, y_train, y_test, y_pred, time_ml, 'svm')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Área bajo la curva"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.756745801968732"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## K Neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "start = time.time()\n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors=3, algorithm='kd_tree', weights='uniform')\n",
    "knn.fit(X_train, y_train)\n",
    "y_pred = knn.predict(X_test)\n",
    "\n",
    "end = time.time()\n",
    "time_ml = end - start"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tiempo de ejecución = 0.011999845504760742\n",
      "\n",
      "\n",
      "Matriz de confusión\n",
      "[[122  35]\n",
      " [ 35  75]]\n",
      "\n",
      "\n",
      "Score train = 0.8745980707395499\n",
      "\n",
      "\n",
      "Accuracy test = 0.7378277153558053\n",
      "Precision test = 0.6818181818181818\n",
      "Sensivity test = 0.6818181818181818\n",
      "Specificity test = 0.7770700636942676\n",
      "F1 Score = 0.6818181818181818\n",
      "Mean Squared Error = 0.26217228464419473\n"
     ]
    }
   ],
   "source": [
    "metricas(X_train, y_train, y_test, y_pred, time_ml, 'knn')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Área bajo la curva"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7294441227562246"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "start = time.time()\n",
    "\n",
    "rfm = RandomForestClassifier(n_estimators=70, oob_score=True, n_jobs=-1, random_state=101, max_features=None, min_samples_leaf=1)\n",
    "rfm.fit(X_train, y_train)\n",
    "y_pred=rfm.predict(X_test)\n",
    "\n",
    "end = time.time()\n",
    "time_ml = end - start"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tiempo de ejecución = 0.12302231788635254\n",
      "\n",
      "\n",
      "Matriz de confusión\n",
      "[[132  25]\n",
      " [ 30  80]]\n",
      "\n",
      "\n",
      "Score train = 0.9501607717041801\n",
      "\n",
      "\n",
      "Accuracy test = 0.7940074906367042\n",
      "Precision test = 0.7619047619047619\n",
      "Sensivity test = 0.7272727272727273\n",
      "Specificity test = 0.8407643312101911\n",
      "F1 Score = 0.7441860465116279\n",
      "Mean Squared Error = 0.20599250936329588\n"
     ]
    }
   ],
   "source": [
    "metricas(X_train, y_train, y_test, y_pred, time_ml, 'rand_for')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Área bajo la curva"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7840185292414592"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SGD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "start = time.time()\n",
    "\n",
    "sgd =  SGDClassifier(loss='hinge', shuffle=True,random_state=101)\n",
    "sgd.fit(X_train, y_train)\n",
    "y_pred=sgd.predict(X_test)\n",
    "\n",
    "end = time.time()\n",
    "time_ml = end - start"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tiempo de ejecución = 0.006484031677246094\n",
      "\n",
      "\n",
      "Matriz de confusión\n",
      "[[125  32]\n",
      " [ 33  77]]\n",
      "\n",
      "\n",
      "Score train = 0.8118971061093248\n",
      "\n",
      "\n",
      "Accuracy test = 0.7565543071161048\n",
      "Precision test = 0.7064220183486238\n",
      "Sensivity test = 0.7\n",
      "Specificity test = 0.7961783439490446\n",
      "F1 Score = 0.7031963470319634\n",
      "Mean Squared Error = 0.24344569288389514\n"
     ]
    }
   ],
   "source": [
    "metricas(X_train, y_train, y_test, y_pred, time_ml, 'sgd')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Área bajo la curva"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7480891719745223"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gráficas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Antes de tunear hiper-parámetros"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "results_initial_df = pd.read_csv('results_initial.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "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>score_train</th>\n",
       "      <th>acc_test</th>\n",
       "      <th>prec_test</th>\n",
       "      <th>sens_test</th>\n",
       "      <th>spec_test</th>\n",
       "      <th>f1_scor</th>\n",
       "      <th>mserror</th>\n",
       "      <th>time</th>\n",
       "      <th>method</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.826367</td>\n",
       "      <td>0.794007</td>\n",
       "      <td>0.783505</td>\n",
       "      <td>0.690909</td>\n",
       "      <td>0.866242</td>\n",
       "      <td>0.734300</td>\n",
       "      <td>0.205993</td>\n",
       "      <td>0.030045</td>\n",
       "      <td>reg_log</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.802251</td>\n",
       "      <td>0.629213</td>\n",
       "      <td>0.617021</td>\n",
       "      <td>0.263636</td>\n",
       "      <td>0.885350</td>\n",
       "      <td>0.369427</td>\n",
       "      <td>0.370787</td>\n",
       "      <td>0.013098</td>\n",
       "      <td>svm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.874598</td>\n",
       "      <td>0.737828</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.777070</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.262172</td>\n",
       "      <td>0.010212</td>\n",
       "      <td>knn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.950161</td>\n",
       "      <td>0.764045</td>\n",
       "      <td>0.790123</td>\n",
       "      <td>0.581818</td>\n",
       "      <td>0.891720</td>\n",
       "      <td>0.670157</td>\n",
       "      <td>0.235955</td>\n",
       "      <td>0.116731</td>\n",
       "      <td>rand_for</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.811897</td>\n",
       "      <td>0.689139</td>\n",
       "      <td>0.728814</td>\n",
       "      <td>0.390909</td>\n",
       "      <td>0.898089</td>\n",
       "      <td>0.508876</td>\n",
       "      <td>0.310861</td>\n",
       "      <td>0.006156</td>\n",
       "      <td>sgd</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   score_train  acc_test  prec_test  sens_test  spec_test   f1_scor   mserror  \\\n",
       "0     0.826367  0.794007   0.783505   0.690909   0.866242  0.734300  0.205993   \n",
       "1     0.802251  0.629213   0.617021   0.263636   0.885350  0.369427  0.370787   \n",
       "2     0.874598  0.737828   0.681818   0.681818   0.777070  0.681818  0.262172   \n",
       "3     0.950161  0.764045   0.790123   0.581818   0.891720  0.670157  0.235955   \n",
       "4     0.811897  0.689139   0.728814   0.390909   0.898089  0.508876  0.310861   \n",
       "\n",
       "       time    method  \n",
       "0  0.030045   reg_log  \n",
       "1  0.013098       svm  \n",
       "2  0.010212       knn  \n",
       "3  0.116731  rand_for  \n",
       "4  0.006156       sgd  "
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_initial_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Time (s)')"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(results_initial_df.method, results_initial_df.time, 'b-', marker='o')\n",
    "plt.xlabel('Algorithms')\n",
    "plt.ylabel(\"Time (s)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f634a572d30>"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 8))\n",
    "plt.plot(results_initial_df.method, results_initial_df.acc_test, 'b-', marker='o')\n",
    "plt.plot(results_initial_df.method, results_initial_df.prec_test, 'r--', marker='o')\n",
    "plt.plot(results_initial_df.method, results_initial_df.sens_test, 'g-', marker='o')\n",
    "plt.plot(results_initial_df.method, results_initial_df.spec_test, 'y--', marker='o')\n",
    "\n",
    "plt.xlabel('Algorithm')\n",
    "plt.legend(['Accuracy', 'Precission', 'Sensitivity', 'Specificity'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Después de tunear hiper-parámetros"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "results_df = pd.read_csv('results.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>score_train</th>\n",
       "      <th>acc_test</th>\n",
       "      <th>prec_test</th>\n",
       "      <th>sens_test</th>\n",
       "      <th>spec_test</th>\n",
       "      <th>f1_scor</th>\n",
       "      <th>mserror</th>\n",
       "      <th>time</th>\n",
       "      <th>method</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.826367</td>\n",
       "      <td>0.794007</td>\n",
       "      <td>0.783505</td>\n",
       "      <td>0.690909</td>\n",
       "      <td>0.866242</td>\n",
       "      <td>0.734300</td>\n",
       "      <td>0.205993</td>\n",
       "      <td>0.031649</td>\n",
       "      <td>reg_log</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.802251</td>\n",
       "      <td>0.771536</td>\n",
       "      <td>0.747475</td>\n",
       "      <td>0.672727</td>\n",
       "      <td>0.840764</td>\n",
       "      <td>0.708134</td>\n",
       "      <td>0.228464</td>\n",
       "      <td>0.912226</td>\n",
       "      <td>svm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.874598</td>\n",
       "      <td>0.737828</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.777070</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.262172</td>\n",
       "      <td>0.012000</td>\n",
       "      <td>knn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.950161</td>\n",
       "      <td>0.794007</td>\n",
       "      <td>0.761905</td>\n",
       "      <td>0.727273</td>\n",
       "      <td>0.840764</td>\n",
       "      <td>0.744186</td>\n",
       "      <td>0.205993</td>\n",
       "      <td>0.123022</td>\n",
       "      <td>rand_for</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.811897</td>\n",
       "      <td>0.756554</td>\n",
       "      <td>0.706422</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>0.796178</td>\n",
       "      <td>0.703196</td>\n",
       "      <td>0.243446</td>\n",
       "      <td>0.006484</td>\n",
       "      <td>sgd</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   score_train  acc_test  prec_test  sens_test  spec_test   f1_scor   mserror  \\\n",
       "0     0.826367  0.794007   0.783505   0.690909   0.866242  0.734300  0.205993   \n",
       "1     0.802251  0.771536   0.747475   0.672727   0.840764  0.708134  0.228464   \n",
       "2     0.874598  0.737828   0.681818   0.681818   0.777070  0.681818  0.262172   \n",
       "3     0.950161  0.794007   0.761905   0.727273   0.840764  0.744186  0.205993   \n",
       "4     0.811897  0.756554   0.706422   0.700000   0.796178  0.703196  0.243446   \n",
       "\n",
       "       time    method  \n",
       "0  0.031649   reg_log  \n",
       "1  0.912226       svm  \n",
       "2  0.012000       knn  \n",
       "3  0.123022  rand_for  \n",
       "4  0.006484       sgd  "
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "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>score_train</th>\n",
       "      <th>acc_test</th>\n",
       "      <th>prec_test</th>\n",
       "      <th>sens_test</th>\n",
       "      <th>spec_test</th>\n",
       "      <th>f1_scor</th>\n",
       "      <th>mserror</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.853055</td>\n",
       "      <td>0.770787</td>\n",
       "      <td>0.736225</td>\n",
       "      <td>0.694545</td>\n",
       "      <td>0.824204</td>\n",
       "      <td>0.714327</td>\n",
       "      <td>0.229213</td>\n",
       "      <td>0.217076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.060994</td>\n",
       "      <td>0.024330</td>\n",
       "      <td>0.041442</td>\n",
       "      <td>0.020929</td>\n",
       "      <td>0.036479</td>\n",
       "      <td>0.025048</td>\n",
       "      <td>0.024330</td>\n",
       "      <td>0.391429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.802251</td>\n",
       "      <td>0.737828</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.672727</td>\n",
       "      <td>0.777070</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.205993</td>\n",
       "      <td>0.006484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.811897</td>\n",
       "      <td>0.756554</td>\n",
       "      <td>0.706422</td>\n",
       "      <td>0.681818</td>\n",
       "      <td>0.796178</td>\n",
       "      <td>0.703196</td>\n",
       "      <td>0.205993</td>\n",
       "      <td>0.012000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.826367</td>\n",
       "      <td>0.771536</td>\n",
       "      <td>0.747475</td>\n",
       "      <td>0.690909</td>\n",
       "      <td>0.840764</td>\n",
       "      <td>0.708134</td>\n",
       "      <td>0.228464</td>\n",
       "      <td>0.031649</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.874598</td>\n",
       "      <td>0.794007</td>\n",
       "      <td>0.761905</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>0.840764</td>\n",
       "      <td>0.734300</td>\n",
       "      <td>0.243446</td>\n",
       "      <td>0.123022</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.950161</td>\n",
       "      <td>0.794007</td>\n",
       "      <td>0.783505</td>\n",
       "      <td>0.727273</td>\n",
       "      <td>0.866242</td>\n",
       "      <td>0.744186</td>\n",
       "      <td>0.262172</td>\n",
       "      <td>0.912226</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       score_train  acc_test  prec_test  sens_test  spec_test   f1_scor  \\\n",
       "count     5.000000  5.000000   5.000000   5.000000   5.000000  5.000000   \n",
       "mean      0.853055  0.770787   0.736225   0.694545   0.824204  0.714327   \n",
       "std       0.060994  0.024330   0.041442   0.020929   0.036479  0.025048   \n",
       "min       0.802251  0.737828   0.681818   0.672727   0.777070  0.681818   \n",
       "25%       0.811897  0.756554   0.706422   0.681818   0.796178  0.703196   \n",
       "50%       0.826367  0.771536   0.747475   0.690909   0.840764  0.708134   \n",
       "75%       0.874598  0.794007   0.761905   0.700000   0.840764  0.734300   \n",
       "max       0.950161  0.794007   0.783505   0.727273   0.866242  0.744186   \n",
       "\n",
       "        mserror      time  \n",
       "count  5.000000  5.000000  \n",
       "mean   0.229213  0.217076  \n",
       "std    0.024330  0.391429  \n",
       "min    0.205993  0.006484  \n",
       "25%    0.205993  0.012000  \n",
       "50%    0.228464  0.031649  \n",
       "75%    0.243446  0.123022  \n",
       "max    0.262172  0.912226  "
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Time (s)')"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(results_df.method, results_df.time, 'b-', marker='o')\n",
    "plt.xlabel('Algorithms')\n",
    "plt.ylabel(\"Time (s)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f6347b0feb0>"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 8))\n",
    "plt.plot(results_df.method, results_df.acc_test, 'b-', marker='o')\n",
    "plt.plot(results_df.method, results_df.prec_test, 'r--', marker='o')\n",
    "plt.plot(results_df.method, results_df.sens_test, 'g-', marker='o')\n",
    "plt.plot(results_df.method, results_df.spec_test, 'y--', marker='o')\n",
    "\n",
    "plt.xlabel('Algorithm')\n",
    "plt.legend(['Accuracy', 'Precission', 'Sensitivity', 'Specificity'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# /-----------------------------------------------------------------------------------------------------------/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predicciones\n",
    "A continuación se muestra un poco del proceso de predicción de cada método, en este dataset no se conoce quién sobrevivió o no pero fue un ejercicio interesante para probar los algoritmos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 336,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_df = pd.read_csv('test.csv')\n",
    "\n",
    "test_df['Sex'].replace(['female', 'male'], [0, 1], inplace=True)\n",
    "test_df['Embarked'].replace(['Q', 'S', 'C'], [0, 1, 2], inplace=True)\n",
    "age_mean = test_df['Age'].mean()\n",
    "test_df['Age'] = test_df['Age'].replace(np.nan, age_mean)\n",
    "\n",
    "bins = [0, 8, 15, 18, 25, 40, 60, 100]\n",
    "values = ['1', '2', '3', '4', '5', '6', '7']\n",
    "test_df['Age'] = pd.cut(test_df['Age'], bins, labels=values)\n",
    "\n",
    "test_df.drop(['Cabin'], axis=1, inplace=True)\n",
    "test_df.drop(['Name'], axis=1, inplace=True)\n",
    "test_df.drop(['Ticket'], axis=1, inplace=True)\n",
    "\n",
    "test_df.dropna(axis=0, how='any', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 337,
   "metadata": {},
   "outputs": [],
   "source": [
    "ids = test_df['PassengerId']\n",
    "pred_x = test_df.drop('PassengerId', axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predicción Regresión Logística"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 338,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>893</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>894</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>895</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>896</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>897</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>898</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>899</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>900</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>901</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived\n",
       "0          892         0\n",
       "1          893         0\n",
       "2          894         0\n",
       "3          895         0\n",
       "4          896         1\n",
       "5          897         0\n",
       "6          898         1\n",
       "7          899         0\n",
       "8          900         1\n",
       "9          901         0"
      ]
     },
     "execution_count": 338,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_log_reg = log_reg.predict(pred_x)\n",
    "out_log_reg = pd.DataFrame({'PassengerId': ids, 'Survived': pred_log_reg})\n",
    "out_log_reg.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predicción SVC "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 340,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>893</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>894</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>895</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>896</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>897</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>898</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>899</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>900</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>901</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived\n",
       "0          892         0\n",
       "1          893         0\n",
       "2          894         0\n",
       "3          895         0\n",
       "4          896         0\n",
       "5          897         0\n",
       "6          898         0\n",
       "7          899         0\n",
       "8          900         0\n",
       "9          901         0"
      ]
     },
     "execution_count": 340,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_svc = svc.predict(pred_x)\n",
    "out_svc = pd.DataFrame({'PassengerId': ids, 'Survived': pred_svc})\n",
    "out_svc.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predicción KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 341,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>893</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>894</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>895</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>896</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>897</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>898</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>899</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>900</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>901</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived\n",
       "0          892         0\n",
       "1          893         0\n",
       "2          894         0\n",
       "3          895         0\n",
       "4          896         1\n",
       "5          897         0\n",
       "6          898         1\n",
       "7          899         1\n",
       "8          900         1\n",
       "9          901         0"
      ]
     },
     "execution_count": 341,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_knn = knn.predict(pred_x)\n",
    "out_knn = pd.DataFrame({'PassengerId': ids, 'Survived': pred_knn})\n",
    "out_knn.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predicción Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 345,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>893</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>894</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>895</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>896</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>897</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>898</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>899</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>900</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>901</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived\n",
       "0          892         0\n",
       "1          893         1\n",
       "2          894         0\n",
       "3          895         0\n",
       "4          896         0\n",
       "5          897         0\n",
       "6          898         1\n",
       "7          899         0\n",
       "8          900         1\n",
       "9          901         0"
      ]
     },
     "execution_count": 345,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_rfm = rfm.predict(pred_x)\n",
    "out_rfm = pd.DataFrame({'PassengerId': ids, 'Survived': pred_rfm})\n",
    "out_rfm.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predicción SGD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 346,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>893</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>894</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>895</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>896</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>897</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>898</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>899</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>900</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>901</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived\n",
       "0          892         0\n",
       "1          893         0\n",
       "2          894         0\n",
       "3          895         0\n",
       "4          896         0\n",
       "5          897         0\n",
       "6          898         0\n",
       "7          899         0\n",
       "8          900         1\n",
       "9          901         0"
      ]
     },
     "execution_count": 346,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_sgd = sgd.predict(pred_x)\n",
    "out_sgd = pd.DataFrame({'PassengerId': ids, 'Survived': pred_sgd})\n",
    "out_sgd.head(10)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}