06_feature_celular_que_mas_vio.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn.feature_extraction as fte\n",
    "from sklearn.decomposition import PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "processed = pd.read_csv('processed.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 38829 entries, 0 to 38828\n",
      "Data columns (total 15 columns):\n",
      "person                38829 non-null object\n",
      "mayor_evento          38829 non-null object\n",
      "prod_visto            38829 non-null object\n",
      "dispos                38829 non-null object\n",
      "compras               38829 non-null int64\n",
      "visitas               38829 non-null int64\n",
      "cant_prod_vistos      38829 non-null int64\n",
      "primer_visita         38829 non-null object\n",
      "ultima_visita         38829 non-null object\n",
      "cant_eventos          38829 non-null int64\n",
      "permanencia           38829 non-null int64\n",
      "hace_cuanto_visito    38829 non-null int64\n",
      "eventos_por_dia       38829 non-null float64\n",
      "channel               38829 non-null object\n",
      "visitas_por_dia       38829 non-null float64\n",
      "dtypes: float64(2), int64(6), object(7)\n",
      "memory usage: 4.4+ MB\n"
     ]
    }
   ],
   "source": [
    "processed.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                   Ninguno\n",
       "1                 iPhone 6S\n",
       "2                 iPhone 6S\n",
       "3    Samsung Galaxy A3 Duos\n",
       "4                  iPhone 7\n",
       "Name: prod_visto, dtype: object"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processed.prod_visto.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "vec = fte.text.TfidfVectorizer()\n",
    "svd = PCA(n_components=2)\n",
    "tfidf = vec.fit_transform(processed.prod_visto)\n",
    "reduced = svd.fit_transform(tfidf.todense())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(reduced)\n",
    "df['model'] = processed.prod_visto\n",
    "plot = df.sort_values(by='model')\n",
    "plot['marca'] = plot['model'].str.split(' ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot['marca'] = plot['marca'].transform(lambda x: x[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x7ff03bdd56a0>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 802.5x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(data=plot,x_vars=0,y_vars=1,hue='marca',height=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_features = processed.join(df.drop('model',axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>person</th>\n",
       "      <th>mayor_evento</th>\n",
       "      <th>prod_visto</th>\n",
       "      <th>dispos</th>\n",
       "      <th>compras</th>\n",
       "      <th>visitas</th>\n",
       "      <th>cant_prod_vistos</th>\n",
       "      <th>primer_visita</th>\n",
       "      <th>ultima_visita</th>\n",
       "      <th>cant_eventos</th>\n",
       "      <th>permanencia</th>\n",
       "      <th>hace_cuanto_visito</th>\n",
       "      <th>eventos_por_dia</th>\n",
       "      <th>channel</th>\n",
       "      <th>visitas_por_dia</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0008ed71</td>\n",
       "      <td>checkout</td>\n",
       "      <td>Ninguno</td>\n",
       "      <td>Computer</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2018-05-17 12:27:47</td>\n",
       "      <td>2018-05-17 16:28:37</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>14</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Referral</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>-0.183765</td>\n",
       "      <td>-0.108759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>00091926</td>\n",
       "      <td>viewed product</td>\n",
       "      <td>iPhone 6S</td>\n",
       "      <td>Computer</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>372</td>\n",
       "      <td>2018-05-03 22:08:29</td>\n",
       "      <td>2018-05-31 19:52:03</td>\n",
       "      <td>448</td>\n",
       "      <td>28</td>\n",
       "      <td>0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>Direct</td>\n",
       "      <td>1.214286</td>\n",
       "      <td>0.422713</td>\n",
       "      <td>-0.457605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>00091a7a</td>\n",
       "      <td>brand listing</td>\n",
       "      <td>iPhone 6S</td>\n",
       "      <td>Smartphone</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2018-03-26 14:51:11</td>\n",
       "      <td>2018-03-26 14:56:58</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>66</td>\n",
       "      <td>10.0</td>\n",
       "      <td>Paid</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.422713</td>\n",
       "      <td>-0.457605</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>000ba417</td>\n",
       "      <td>viewed product</td>\n",
       "      <td>Samsung Galaxy A3 Duos</td>\n",
       "      <td>Computer</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>153</td>\n",
       "      <td>2018-05-17 11:11:45</td>\n",
       "      <td>2018-05-26 13:09:22</td>\n",
       "      <td>206</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>20.6</td>\n",
       "      <td>Direct</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>-0.304328</td>\n",
       "      <td>0.020564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>000c79fe</td>\n",
       "      <td>searched products</td>\n",
       "      <td>iPhone 7</td>\n",
       "      <td>Smartphone</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2018-05-29 00:27:47</td>\n",
       "      <td>2018-05-29 00:38:07</td>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>17.0</td>\n",
       "      <td>Paid</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.669254</td>\n",
       "      <td>0.075761</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     person       mayor_evento              prod_visto      dispos  compras  \\\n",
       "0  0008ed71           checkout                 Ninguno    Computer        0   \n",
       "1  00091926     viewed product               iPhone 6S    Computer        0   \n",
       "2  00091a7a      brand listing               iPhone 6S  Smartphone        0   \n",
       "3  000ba417     viewed product  Samsung Galaxy A3 Duos    Computer        1   \n",
       "4  000c79fe  searched products                iPhone 7  Smartphone        0   \n",
       "\n",
       "   visitas  cant_prod_vistos        primer_visita        ultima_visita  \\\n",
       "0        2                 0  2018-05-17 12:27:47  2018-05-17 16:28:37   \n",
       "1       34               372  2018-05-03 22:08:29  2018-05-31 19:52:03   \n",
       "2        1                 3  2018-03-26 14:51:11  2018-03-26 14:56:58   \n",
       "3        6               153  2018-05-17 11:11:45  2018-05-26 13:09:22   \n",
       "4        1                 3  2018-05-29 00:27:47  2018-05-29 00:38:07   \n",
       "\n",
       "   cant_eventos  permanencia  hace_cuanto_visito  eventos_por_dia   channel  \\\n",
       "0             6            1                  14              6.0  Referral   \n",
       "1           448           28                   0             16.0    Direct   \n",
       "2            10            1                  66             10.0      Paid   \n",
       "3           206           10                   5             20.6    Direct   \n",
       "4            17            1                   2             17.0      Paid   \n",
       "\n",
       "   visitas_por_dia         0         1  \n",
       "0         2.000000 -0.183765 -0.108759  \n",
       "1         1.214286  0.422713 -0.457605  \n",
       "2         1.000000  0.422713 -0.457605  \n",
       "3         0.600000 -0.304328  0.020564  \n",
       "4         1.000000  0.669254  0.075761  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_features.to_csv('features_02.csv')"
   ]
  },
  {
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}