{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "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>Unnamed: 0</th>\n",
       "      <th>COUNTRY_KEY</th>\n",
       "      <th>ITEM_DESC</th>\n",
       "      <th>BARCODE</th>\n",
       "      <th>BEM_CLASS_KEY</th>\n",
       "      <th>BEM_CLASS_DESC_FR</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>FRA</td>\n",
       "      <td>150ML SECHE VERNIS VITRY</td>\n",
       "      <td>3538892526034</td>\n",
       "      <td>1964</td>\n",
       "      <td>DIVERS MAQUILLAGE</td>\n",
       "      <td>maquillage des ongles</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>FRA</td>\n",
       "      <td>15COM.SOMMEIL TRIPLE ACTI.PURE</td>\n",
       "      <td>3701056803443</td>\n",
       "      <td>1964</td>\n",
       "      <td>DIVERS MAQUILLAGE</td>\n",
       "      <td>maquillage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>FRA</td>\n",
       "      <td>2 EPONG.DEMAQUILLER N°42 SANOD</td>\n",
       "      <td>3701509580426</td>\n",
       "      <td>1964</td>\n",
       "      <td>DIVERS MAQUILLAGE</td>\n",
       "      <td>autres produits maquillage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>FRA</td>\n",
       "      <td>200ML DIFF PFUM FIGUE CDP</td>\n",
       "      <td>3551780164033</td>\n",
       "      <td>1964</td>\n",
       "      <td>DIVERS MAQUILLAGE</td>\n",
       "      <td>maquillage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>FRA</td>\n",
       "      <td>2PCS EPONGE DEMAQUILLANTE</td>\n",
       "      <td>3276558220253</td>\n",
       "      <td>1964</td>\n",
       "      <td>DIVERS MAQUILLAGE</td>\n",
       "      <td>autres produits maquillage</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0 COUNTRY_KEY                       ITEM_DESC        BARCODE  \\\n",
       "0           0         FRA        150ML SECHE VERNIS VITRY  3538892526034   \n",
       "1           1         FRA  15COM.SOMMEIL TRIPLE ACTI.PURE  3701056803443   \n",
       "2           2         FRA  2 EPONG.DEMAQUILLER N°42 SANOD  3701509580426   \n",
       "3           3         FRA       200ML DIFF PFUM FIGUE CDP  3551780164033   \n",
       "4           4         FRA       2PCS EPONGE DEMAQUILLANTE  3276558220253   \n",
       "\n",
       "   BEM_CLASS_KEY  BEM_CLASS_DESC_FR                  prediction  \n",
       "0           1964  DIVERS MAQUILLAGE       maquillage des ongles  \n",
       "1           1964  DIVERS MAQUILLAGE                  maquillage  \n",
       "2           1964  DIVERS MAQUILLAGE  autres produits maquillage  \n",
       "3           1964  DIVERS MAQUILLAGE                  maquillage  \n",
       "4           1964  DIVERS MAQUILLAGE  autres produits maquillage  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.read_csv(\"dataset/data_maquillage.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from traning_zone.classe_prediction.classe_prediction import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = [\"KID KITTY FEMME BETY NAILMATI\"]\n",
    "\n",
    "pred = Prediction(X)\n",
    "\n",
    "data = pred.prediction_grand_modele_()\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>item_desc</th>\n",
       "      <th>classe</th>\n",
       "      <th>famille</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>150ML SECHE VERNIS VITRY</td>\n",
       "      <td>{'MAQUILLAGE': 0.017379147931933403, 'AUTRES P...</td>\n",
       "      <td>{'Confiserie': 6.264624244067818e-05, 'Soins C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>15COM.SOMMEIL TRIPLE ACTI.PURE</td>\n",
       "      <td>{'PRODUITS DIÉTÉTIQUES': 0.22845330834388733, ...</td>\n",
       "      <td>{'Confiserie': 0.011916466057300568, 'Soins Ch...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2 EPONG.DEMAQUILLER N°42 SANOD</td>\n",
       "      <td>{'MAQUILLAGE': 0.04227576404809952, 'AUTRES PR...</td>\n",
       "      <td>{'Confiserie': 0.011007937602698803, 'Soins Ch...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>200ML DIFF PFUM FIGUE CDP</td>\n",
       "      <td>{'SOINS DU CORPS': 0.9595848321914673, 'SPÉCIF...</td>\n",
       "      <td>{'Confiserie': 7.310342334676534e-05, 'Soins C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2PCS EPONGE DEMAQUILLANTE</td>\n",
       "      <td>{'MAQUILLAGE': 0.0028436225838959217, 'AUTRES ...</td>\n",
       "      <td>{'Confiserie': 0.08245589584112167, 'Soins Che...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91</th>\n",
       "      <td>TAILLE CRAYON JUMBO EYE CARE</td>\n",
       "      <td>{'MAQUILLAGE': 0.026179207488894463, 'AUTRES P...</td>\n",
       "      <td>{'Confiserie': 9.6481955552008e-05, 'Soins Che...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>92</th>\n",
       "      <td>TROUSSE CM + GEL + SAVON C419</td>\n",
       "      <td>{'EAUX DE COLOGNE/DE TOILETTE': 0.191847190260...</td>\n",
       "      <td>{'Confiserie': 0.0001183701169793494, 'Soins C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>TROUSSE CM BOI ROSE OSM C419</td>\n",
       "      <td>{'EAUX DE COLOGNE/DE TOILETTE': 0.013748150318...</td>\n",
       "      <td>{'Confiserie': 0.0005058256792835891, 'Soins C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>94</th>\n",
       "      <td>TRSE CM+GELHYDRO+SAVON GR C419</td>\n",
       "      <td>{'EAUX DE COLOGNE/DE TOILETTE': 0.038684066385...</td>\n",
       "      <td>{'Confiserie': 0.0004881306958850473, 'Soins C...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>ZAP FAP NACRE 106 BIO 3G</td>\n",
       "      <td>{'MAQUILLAGE': 0.8830502033233643, 'AUTRES PRO...</td>\n",
       "      <td>{'Confiserie': 0.0036271295975893736, 'Soins C...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>96 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         item_desc  \\\n",
       "0         150ML SECHE VERNIS VITRY   \n",
       "1   15COM.SOMMEIL TRIPLE ACTI.PURE   \n",
       "2   2 EPONG.DEMAQUILLER N°42 SANOD   \n",
       "3        200ML DIFF PFUM FIGUE CDP   \n",
       "4        2PCS EPONGE DEMAQUILLANTE   \n",
       "..                             ...   \n",
       "91    TAILLE CRAYON JUMBO EYE CARE   \n",
       "92   TROUSSE CM + GEL + SAVON C419   \n",
       "93    TROUSSE CM BOI ROSE OSM C419   \n",
       "94  TRSE CM+GELHYDRO+SAVON GR C419   \n",
       "95        ZAP FAP NACRE 106 BIO 3G   \n",
       "\n",
       "                                               classe  \\\n",
       "0   {'MAQUILLAGE': 0.017379147931933403, 'AUTRES P...   \n",
       "1   {'PRODUITS DIÉTÉTIQUES': 0.22845330834388733, ...   \n",
       "2   {'MAQUILLAGE': 0.04227576404809952, 'AUTRES PR...   \n",
       "3   {'SOINS DU CORPS': 0.9595848321914673, 'SPÉCIF...   \n",
       "4   {'MAQUILLAGE': 0.0028436225838959217, 'AUTRES ...   \n",
       "..                                                ...   \n",
       "91  {'MAQUILLAGE': 0.026179207488894463, 'AUTRES P...   \n",
       "92  {'EAUX DE COLOGNE/DE TOILETTE': 0.191847190260...   \n",
       "93  {'EAUX DE COLOGNE/DE TOILETTE': 0.013748150318...   \n",
       "94  {'EAUX DE COLOGNE/DE TOILETTE': 0.038684066385...   \n",
       "95  {'MAQUILLAGE': 0.8830502033233643, 'AUTRES PRO...   \n",
       "\n",
       "                                              famille  \n",
       "0   {'Confiserie': 6.264624244067818e-05, 'Soins C...  \n",
       "1   {'Confiserie': 0.011916466057300568, 'Soins Ch...  \n",
       "2   {'Confiserie': 0.011007937602698803, 'Soins Ch...  \n",
       "3   {'Confiserie': 7.310342334676534e-05, 'Soins C...  \n",
       "4   {'Confiserie': 0.08245589584112167, 'Soins Che...  \n",
       "..                                                ...  \n",
       "91  {'Confiserie': 9.6481955552008e-05, 'Soins Che...  \n",
       "92  {'Confiserie': 0.0001183701169793494, 'Soins C...  \n",
       "93  {'Confiserie': 0.0005058256792835891, 'Soins C...  \n",
       "94  {'Confiserie': 0.0004881306958850473, 'Soins C...  \n",
       "95  {'Confiserie': 0.0036271295975893736, 'Soins C...  \n",
       "\n",
       "[96 rows x 3 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = df.ITEM_DESC\n",
    "\n",
    "pred = Prediction(X)\n",
    "data = pred.prediction_modele_spacy()\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.to_csv(\"sapcy_spacy.csv\")"
   ]
  }
 ],
 "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": "3.11.3"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}