Noviazgo clasificacion out of the box.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "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>fecha</th>\n",
       "      <th>hora</th>\n",
       "      <th>edad</th>\n",
       "      <th>genero</th>\n",
       "      <th>barrio</th>\n",
       "      <th>pregunta_1</th>\n",
       "      <th>pregunta_2</th>\n",
       "      <th>pregunta_3</th>\n",
       "      <th>pregunta_4</th>\n",
       "      <th>pregunta_5</th>\n",
       "      <th>pregunta_6</th>\n",
       "      <th>pregunta_7</th>\n",
       "      <th>pregunta_8</th>\n",
       "      <th>pregunta_9</th>\n",
       "      <th>pregunta_10</th>\n",
       "      <th>pregunta_11</th>\n",
       "      <th>pregunta_12</th>\n",
       "      <th>pregunta_13</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3508</th>\n",
       "      <td>2018-03-04</td>\n",
       "      <td>16:13:00</td>\n",
       "      <td>22</td>\n",
       "      <td>Mujer</td>\n",
       "      <td>Boedo</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5242</th>\n",
       "      <td>2018-02-16</td>\n",
       "      <td>13:07:00</td>\n",
       "      <td>23</td>\n",
       "      <td>Mujer</td>\n",
       "      <td>Recoleta</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>A</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4519</th>\n",
       "      <td>2018-02-18</td>\n",
       "      <td>16:00:00</td>\n",
       "      <td>24</td>\n",
       "      <td>Mujer</td>\n",
       "      <td>Otro</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>700</th>\n",
       "      <td>2019-03-23</td>\n",
       "      <td>10:38:00</td>\n",
       "      <td>22</td>\n",
       "      <td>Hombre</td>\n",
       "      <td>Otro</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4750</th>\n",
       "      <td>2018-02-17</td>\n",
       "      <td>02:57:00</td>\n",
       "      <td>31</td>\n",
       "      <td>Hombre</td>\n",
       "      <td>Almagro</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>A</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "      <td>A</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>A</td>\n",
       "      <td>A</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           fecha      hora  edad  genero    barrio pregunta_1 pregunta_2  \\\n",
       "3508  2018-03-04  16:13:00    22   Mujer     Boedo          B          B   \n",
       "5242  2018-02-16  13:07:00    23   Mujer  Recoleta          A          B   \n",
       "4519  2018-02-18  16:00:00    24   Mujer      Otro          B          B   \n",
       "700   2019-03-23  10:38:00    22  Hombre      Otro          B          B   \n",
       "4750  2018-02-17  02:57:00    31  Hombre   Almagro          A          B   \n",
       "\n",
       "     pregunta_3 pregunta_4 pregunta_5 pregunta_6 pregunta_7 pregunta_8  \\\n",
       "3508          C          B          C          B          C          B   \n",
       "5242          A          A          B          A          B          C   \n",
       "4519          C          B          B          A          B          C   \n",
       "700           B          C          B          B          C          C   \n",
       "4750          A          A          B          A          C          C   \n",
       "\n",
       "     pregunta_9 pregunta_10 pregunta_11 pregunta_12 pregunta_13  \n",
       "3508          C           C           C           C           C  \n",
       "5242          C           C           A           C           A  \n",
       "4519          B           C           B           B           B  \n",
       "700           C           C           C           C           C  \n",
       "4750          C           C           A           A           B  "
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"resultados-del-test.csv\", delimiter = \",\")\n",
    "data.sample(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creo variable target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['target'] = data.pregunta_13 != 'C'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    3963\n",
       "True     1713\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.target.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Defino feature columns, X e y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [],
   "source": [
    "categorical_columns = ['genero', 'barrio'] + [col for col in data.columns if 'pregunta_13' not in col and 'pregunta' in col]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "categorical_features = pd.get_dummies(data.loc[:, categorical_columns])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5676, 19)"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5676, 87)"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "categorical_features.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Concateno las columnas dummy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.concat([data, categorical_features], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5676, 106)"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Creo un array solo con las columnas que utilizare"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_index, test_index = train_test_split(data.index, stratify = data.target, random_state = 42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Normalizo variables continuas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [],
   "source": [
    "numerical_columns = ['edad']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "scaler.fit(data.loc[train_index, numerical_columns])\n",
    "train_numerical_scaled = scaler.transform(data.loc[train_index, numerical_columns])\n",
    "test_numerical_scaled = scaler.transform(data.loc[test_index, numerical_columns])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Creo un array con los nombres de columnas para valores escalados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [],
   "source": [
    "numerical_features = [f\"{col}_scaled\" for col in numerical_columns]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_numerical_df = pd.DataFrame(train_numerical_scaled, index = train_index, columns = numerical_features)\n",
    "test_numerical_df = pd.DataFrame(test_numerical_scaled, index = test_index, columns = numerical_features)\n",
    "numerical_features_dataframe = train_numerical_df.append(test_numerical_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agrego los datos escalados al dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.concat([data, numerical_features_dataframe], axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Genero X, y para train y test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_cols = categorical_features.columns.tolist() + numerical_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = data.loc[:, feature_cols]\n",
    "y = data.target\n",
    "\n",
    "X_train, y_train = X.loc[train_index], y.loc[train_index]\n",
    "X_test, y_test = X.loc[test_index], y.loc[test_index]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Implemento regression logistica"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Entreno la regression logistica"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GaussianNB(priors=None, var_smoothing=1e-09)"
      ]
     },
     "execution_count": 208,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr = GaussianNB()\n",
    "lr.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Veo las metricas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import roc_curve\n",
    "from sklearn.metrics import log_loss\n",
    "from sklearn.metrics import auc\n",
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = lr.predict(X_test)\n",
    "y_pred_proba = lr.predict_proba(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {},
   "outputs": [],
   "source": [
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'ROC Curve')"
      ]
     },
     "execution_count": 212,
     "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": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "sns.lineplot(fpr, tpr)\n",
    "plt.xlabel('1 - Specificty')\n",
    "plt.ylabel('TPR / Sensitivity')\n",
    "plt.title('ROC Curve')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.989269901795787"
      ]
     },
     "execution_count": 213,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log_loss(y_test, y_pred_proba)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6698994218999029"
      ]
     },
     "execution_count": 214,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "auc(fpr, tpr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6807610993657506"
      ]
     },
     "execution_count": 215,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    991\n",
       "True     428\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 216,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    844\n",
       "True     575\n",
       "dtype: int64"
      ]
     },
     "execution_count": 217,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(y_pred).value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}