Devoir_Python_caleb.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Devoir de Python\n",
    "Caleb KASHALA ILUNGA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table de matiére\n",
    "    - Partie scrapping\n",
    "    - Partie nettoyage\n",
    "    - Partie descriptive\n",
    "    - partie apprentissage machine learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Problématique: Comment estimer le meilleur prix possible en fonction de différentes variables présentées?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "from selenium import webdriver\n",
    "from selenium.webdriver.common.by import By\n",
    "from selenium.webdriver.common.keys import Keys\n",
    "from time import sleep, time \n",
    "from selenium.common.exceptions import StaleElementReferenceException\n",
    "import pandas as pd\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.stats\n",
    "import seaborn as sns \n",
    "\n",
    "from sklearn.linear_model import LinearRegression, Lasso, Ridge, ElasticNet\n",
    "from sklearn.neighbors import KNeighborsRegressor\n",
    "from sklearn.gaussian_process import GaussianProcessRegressor\n",
    "from sklearn.svm import SVR\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.neural_network import MLPRegressor\n",
    "\n",
    "from sklearn.model_selection import train_test_split, cross_val_score\n",
    "\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.preprocessing import LabelBinarizer\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.impute import MissingIndicator\n",
    "\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "\n",
    "from sklearn.compose import make_column_transformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.compose import ColumnTransformer\n",
    "\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "from sklearn.datasets import fetch_openml\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.inspection import permutation_importance\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "\n",
    "from sklearn import model_selection as ms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PARTIE SCRAPPING"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "driver = webdriver.Firefox()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "\n",
    "page=driver.get(\"https://fr.shopping.rakuten.com/nav/Tel-PDA_Telephones-mobiles#xtatc=PUB-[PMC]-[E]-[smartphones-tablettes-objets-connectes]-[sous-categorie]-[smartphone]-[Smartphones]-[]\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "i=1\n",
    "carac=[]\n",
    "n=1\n",
    "pk=[]\n",
    "page=driver.get(\"https://fr.shopping.rakuten.com/nav/Tel-PDA_Telephones-mobiles#xtatc=PUB-[PMC]-[E]-[smartphones-tablettes-objets-connectes]-[sous-categorie]-[smartphone]-[Smartphones]-[]\")\n",
    "sleep(5)\n",
    "driver.find_element_by_link_text(\"Plus tard\").click()\n",
    "sleep(5)\n",
    "python_button = driver.find_elements_by_xpath(\"/html/body/div[2]/div/div/div/div/div[2]/span\")[0]\n",
    "python_button.click()\n",
    "while n <30:\n",
    "    pk.append([driver.find_elements_by_class_name(\"productList_layoutContent_NC_.pl-8.f14\")])\n",
    "    sleep(4)\n",
    "    driver.find_element_by_link_text(\"Suivant\").click()\n",
    "    sleep(10)\n",
    "    p=driver.find_elements_by_class_name(\"productList_layoutContent_NC_.pl-8.f14\")\n",
    "    sleep(4)\n",
    "    for i in range(len(p)):\n",
    "        pa=p[i].find_element_by_class_name(\"silver.description_caption_3J5\")\n",
    "        \n",
    "        if len(pa.find_elements_by_class_name(\"f12.description_bulletPoints_Vnz li\"))>1:\n",
    "            try:\n",
    "                reseau=([pa.find_elements_by_class_name(\"f12.description_bulletPoints_Vnz li\")[1].text])\n",
    "            except:\n",
    "                reseau=\"NA\"\n",
    "            try:\n",
    "                ecran=([pa.find_elements_by_class_name(\"f12.description_bulletPoints_Vnz li\")[0].text])\n",
    "            except:\n",
    "                ecran=\"NA\"\n",
    "            try:\n",
    "                exploit=([pa.find_elements_by_class_name(\"f12.description_bulletPoints_Vnz li\")[2].text])\n",
    "            except:\n",
    "                exploit=\"NA\"\n",
    "            try:\n",
    "                prix=([p[i].find_element_by_class_name(\"f20.b\").text])\n",
    "            except:\n",
    "                prix=\"NA\"\n",
    "            m=driver.find_elements_by_class_name(\"lh-title\")\n",
    "            marque=m[i].text\n",
    "        else:\n",
    "            pass\n",
    "        \n",
    "        Resultat={\"Marque\":marque,\"reseau\":reseau,\"exploit\":exploit,\"prix\":prix,\n",
    "                  \"écran\":ecran}\n",
    "\n",
    "        carac.append(Resultat)\n",
    "    \n",
    "    n=n+1\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Data=pd.DataFrame(carac)\n",
    "Data.to_csv(\"Data_Scrapping.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Après avoir lancé le scrapping nous obtenons une base de données avec différentes variables et modalités associées aux variables.\n",
    "* Notre base de données est une base contenant des informations sur différentes marques et modèles de téléphone, dans cette base de données nous avons 1305 observations et 5 variables:\n",
    "    - Marque \n",
    "    - Réseau \n",
    "    - Exploit \n",
    "    - prix \n",
    "    - Ècran \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>Unnamed: 0</th>\n",
       "      <th>Marque</th>\n",
       "      <th>reseau</th>\n",
       "      <th>exploit</th>\n",
       "      <th>prix</th>\n",
       "      <th>écran</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>OnePlus Nord N100 Dual SIM 64 Go Midnight Frost</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 64 Go / RAM 4 Go']</td>\n",
       "      <td>['172,58 €']</td>\n",
       "      <td>['Ecran 6.52\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>Huawei P40 lite 128 Go Double SIM Vert crush (...</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 128 Go / RAM 6 Go']</td>\n",
       "      <td>['196,59 €']</td>\n",
       "      <td>['Ecran 6.4\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Xiaomi Poco X3 NFC Dual-SIM 64 Go Bleu</td>\n",
       "      <td>['Android']</td>\n",
       "      <td>['Double SIM']</td>\n",
       "      <td>['203,79 €']</td>\n",
       "      <td>['Mémoire 64 Go / RAM 6 Go']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>Xiaomi Mi 10T Dual-SIM 128 Go Noir</td>\n",
       "      <td>['Android']</td>\n",
       "      <td>['Double SIM']</td>\n",
       "      <td>['409,99 €']</td>\n",
       "      <td>['Mémoire 128 Go / RAM 8 Go']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>Huawei P30 lite 128 Go (RAM 4 Go) Double SIM B...</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 128 Go / RAM 4 Go']</td>\n",
       "      <td>['214,50 €']</td>\n",
       "      <td>['Ecran 6.15\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>Samsung Galaxy Note9 128 Go Double SIM Noir</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 128 Go / RAM 6 Go']</td>\n",
       "      <td>['374 €']</td>\n",
       "      <td>['Ecran 6.4\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>Samsung Galaxy S9 64 Go Double SIM Noir</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 64 Go / RAM 4 Go']</td>\n",
       "      <td>['270 €']</td>\n",
       "      <td>['Ecran 5.8\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>Huawei P20 lite 64 Go Double SIM Noir minuit</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 64 Go / RAM 4 Go']</td>\n",
       "      <td>['189 €']</td>\n",
       "      <td>['Ecran 5.84\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>Samsung Galaxy A51 128 Go Double SIM Blanc</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 128 Go / RAM 4 Go']</td>\n",
       "      <td>['289,99 €']</td>\n",
       "      <td>['Ecran 6.5\"']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                                             Marque  \\\n",
       "1           1    OnePlus Nord N100 Dual SIM 64 Go Midnight Frost   \n",
       "2           2  Huawei P40 lite 128 Go Double SIM Vert crush (...   \n",
       "3           3             Xiaomi Poco X3 NFC Dual-SIM 64 Go Bleu   \n",
       "4           4                 Xiaomi Mi 10T Dual-SIM 128 Go Noir   \n",
       "5           5  Huawei P30 lite 128 Go (RAM 4 Go) Double SIM B...   \n",
       "6           6        Samsung Galaxy Note9 128 Go Double SIM Noir   \n",
       "7           7            Samsung Galaxy S9 64 Go Double SIM Noir   \n",
       "8           8       Huawei P20 lite 64 Go Double SIM Noir minuit   \n",
       "9           9         Samsung Galaxy A51 128 Go Double SIM Blanc   \n",
       "\n",
       "          reseau                        exploit          prix  \\\n",
       "1  ['Réseau 4G']   ['Mémoire 64 Go / RAM 4 Go']  ['172,58 €']   \n",
       "2  ['Réseau 4G']  ['Mémoire 128 Go / RAM 6 Go']  ['196,59 €']   \n",
       "3    ['Android']                 ['Double SIM']  ['203,79 €']   \n",
       "4    ['Android']                 ['Double SIM']  ['409,99 €']   \n",
       "5  ['Réseau 4G']  ['Mémoire 128 Go / RAM 4 Go']  ['214,50 €']   \n",
       "6  ['Réseau 4G']  ['Mémoire 128 Go / RAM 6 Go']     ['374 €']   \n",
       "7  ['Réseau 4G']   ['Mémoire 64 Go / RAM 4 Go']     ['270 €']   \n",
       "8  ['Réseau 4G']   ['Mémoire 64 Go / RAM 4 Go']     ['189 €']   \n",
       "9  ['Réseau 4G']  ['Mémoire 128 Go / RAM 4 Go']  ['289,99 €']   \n",
       "\n",
       "                           écran  \n",
       "1                ['Ecran 6.52\"']  \n",
       "2                 ['Ecran 6.4\"']  \n",
       "3   ['Mémoire 64 Go / RAM 6 Go']  \n",
       "4  ['Mémoire 128 Go / RAM 8 Go']  \n",
       "5                ['Ecran 6.15\"']  \n",
       "6                 ['Ecran 6.4\"']  \n",
       "7                 ['Ecran 5.8\"']  \n",
       "8                ['Ecran 5.84\"']  \n",
       "9                 ['Ecran 6.5\"']  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#import re\n",
    "df1=pd.read_csv(\"Data_Scrapping.csv\")#df['Marque'] = df['Header'].str.extract('([A-S]\\w{0,})')\n",
    "df1[1:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Partie Nettoyage\n",
    "\n",
    "#### Après avoir créé notre base de données provenant du scrapping nous allons commencer par nettoyer notre base de données et apporter quelques modifications."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dans un premier temps nous commençons par transformer toutes nos variables en type \"str\" pour pouvoir facilement utiliser les fonctions extract etreplace.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#str\n",
    "df1['écran']=df1['écran'].astype(str)\n",
    "df1['reseau']=df1['reseau'].astype(str)\n",
    "df1['Marque']=df1['Marque'].astype(str)\n",
    "df1['exploit']=df1['exploit'].astype(str)\n",
    "df1['prix']=df1['prix'].astype(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Unnamed: 0</th>\n",
       "      <th>Marque</th>\n",
       "      <th>reseau</th>\n",
       "      <th>exploit</th>\n",
       "      <th>prix</th>\n",
       "      <th>écran</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Samsung Galaxy Note10 256 Go Double SIM Noir</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 256 Go / RAM 8 Go']</td>\n",
       "      <td>['546,88 €']</td>\n",
       "      <td>['Ecran 6.3\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>OnePlus Nord N100 Dual SIM 64 Go Midnight Frost</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 64 Go / RAM 4 Go']</td>\n",
       "      <td>['172,58 €']</td>\n",
       "      <td>['Ecran 6.52\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>Huawei P40 lite 128 Go Double SIM Vert crush (...</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 128 Go / RAM 6 Go']</td>\n",
       "      <td>['196,59 €']</td>\n",
       "      <td>['Ecran 6.4\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Xiaomi Poco X3 NFC Dual-SIM 64 Go Bleu</td>\n",
       "      <td>['Android']</td>\n",
       "      <td>['Double SIM']</td>\n",
       "      <td>['203,79 €']</td>\n",
       "      <td>['Mémoire 64 Go / RAM 6 Go']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>Xiaomi Mi 10T Dual-SIM 128 Go Noir</td>\n",
       "      <td>['Android']</td>\n",
       "      <td>['Double SIM']</td>\n",
       "      <td>['409,99 €']</td>\n",
       "      <td>['Mémoire 128 Go / RAM 8 Go']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1300</th>\n",
       "      <td>1300</td>\n",
       "      <td>Apple iPhone 6 Plus 16 Go Or</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['iOS 8']</td>\n",
       "      <td>['164,90 €']</td>\n",
       "      <td>['Ecran 5.5\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1301</th>\n",
       "      <td>1301</td>\n",
       "      <td>Samsung Galaxy A5 (2017) Dual SIM 32 Go Or sable</td>\n",
       "      <td>['Android 6.0 (Marshmallow)']</td>\n",
       "      <td>['Double SIM']</td>\n",
       "      <td>['159,98 €']</td>\n",
       "      <td>['Ecran 5.2\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1302</th>\n",
       "      <td>1302</td>\n",
       "      <td>Samsung Galaxy Xcover 4 16 Go Noir</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 16 Go / RAM 2 Go']</td>\n",
       "      <td>['159,99 €']</td>\n",
       "      <td>['Ecran 5\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1303</th>\n",
       "      <td>1303</td>\n",
       "      <td>Samsung Galaxy A5 (2017) 32 Go Or</td>\n",
       "      <td>['Réseau 4G']</td>\n",
       "      <td>['Mémoire 32 Go / RAM 3 Go']</td>\n",
       "      <td>['149,99 €']</td>\n",
       "      <td>['Ecran 5.2\"']</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1304</th>\n",
       "      <td>1304</td>\n",
       "      <td>Google Pixel 3A XL 64 Go Blanc</td>\n",
       "      <td>['Single-SIM (SIM unique)']</td>\n",
       "      <td>nan</td>\n",
       "      <td>['280,60 €']</td>\n",
       "      <td>['Android 9.0 Pie']</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1305 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Unnamed: 0                                             Marque  \\\n",
       "0              0       Samsung Galaxy Note10 256 Go Double SIM Noir   \n",
       "1              1    OnePlus Nord N100 Dual SIM 64 Go Midnight Frost   \n",
       "2              2  Huawei P40 lite 128 Go Double SIM Vert crush (...   \n",
       "3              3             Xiaomi Poco X3 NFC Dual-SIM 64 Go Bleu   \n",
       "4              4                 Xiaomi Mi 10T Dual-SIM 128 Go Noir   \n",
       "...          ...                                                ...   \n",
       "1300        1300                       Apple iPhone 6 Plus 16 Go Or   \n",
       "1301        1301   Samsung Galaxy A5 (2017) Dual SIM 32 Go Or sable   \n",
       "1302        1302                 Samsung Galaxy Xcover 4 16 Go Noir   \n",
       "1303        1303                  Samsung Galaxy A5 (2017) 32 Go Or   \n",
       "1304        1304                     Google Pixel 3A XL 64 Go Blanc   \n",
       "\n",
       "                             reseau                        exploit  \\\n",
       "0                     ['Réseau 4G']  ['Mémoire 256 Go / RAM 8 Go']   \n",
       "1                     ['Réseau 4G']   ['Mémoire 64 Go / RAM 4 Go']   \n",
       "2                     ['Réseau 4G']  ['Mémoire 128 Go / RAM 6 Go']   \n",
       "3                       ['Android']                 ['Double SIM']   \n",
       "4                       ['Android']                 ['Double SIM']   \n",
       "...                             ...                            ...   \n",
       "1300                  ['Réseau 4G']                      ['iOS 8']   \n",
       "1301  ['Android 6.0 (Marshmallow)']                 ['Double SIM']   \n",
       "1302                  ['Réseau 4G']   ['Mémoire 16 Go / RAM 2 Go']   \n",
       "1303                  ['Réseau 4G']   ['Mémoire 32 Go / RAM 3 Go']   \n",
       "1304    ['Single-SIM (SIM unique)']                            nan   \n",
       "\n",
       "              prix                          écran  \n",
       "0     ['546,88 €']                 ['Ecran 6.3\"']  \n",
       "1     ['172,58 €']                ['Ecran 6.52\"']  \n",
       "2     ['196,59 €']                 ['Ecran 6.4\"']  \n",
       "3     ['203,79 €']   ['Mémoire 64 Go / RAM 6 Go']  \n",
       "4     ['409,99 €']  ['Mémoire 128 Go / RAM 8 Go']  \n",
       "...            ...                            ...  \n",
       "1300  ['164,90 €']                 ['Ecran 5.5\"']  \n",
       "1301  ['159,98 €']                 ['Ecran 5.2\"']  \n",
       "1302  ['159,99 €']                   ['Ecran 5\"']  \n",
       "1303  ['149,99 €']                 ['Ecran 5.2\"']  \n",
       "1304  ['280,60 €']            ['Android 9.0 Pie']  \n",
       "\n",
       "[1305 rows x 6 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#str\n",
    "#df['sexe'].astype('int')\n",
    "df1.dropna ( axis = 0 ,  how = 'all' )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "import Nettoyage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dans un premier temps nous commençons par nettoyer la variable marque, pour si faire nous avons fait en sorte de ne garder que les marquent des téléphones et non pas les modèles associés aux marques.\\\n",
    "Ensuite nous avons mis toutes les variables en minuscule, pour facilement pouvoir séparer les données par marques."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\caleb\\Desktop\\Devoir python\\Nettoyage.py:5: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  fichier['Marque'][i]=\"Inconnu\"\n"
     ]
    }
   ],
   "source": [
    "Nettoyage.nettoyage_marque(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "samsung       389\n",
       "apple         275\n",
       "xiaomi        138\n",
       "inconnu       135\n",
       "huawei        111\n",
       "oneplus        43\n",
       "nokia          41\n",
       "sony           38\n",
       "google         26\n",
       "honor          21\n",
       "asus           17\n",
       "oppo           16\n",
       "realme         16\n",
       "lg             13\n",
       "blackberry     10\n",
       "nubia           9\n",
       "alcatel         3\n",
       "oukitel         2\n",
       "cubot           2\n",
       "Name: Marque, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.Marque.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dans la cellule suivante nous avons nettoyé les données provenant de la variable écran. Nous avons enlevé la chaine de caractère \"['Écran \" au début et nous avons enlevé les 3 dernières chaine de caractère"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Nettoyage.nettoyage_ecran(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "del df1['Unnamed: 0']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#str\n",
    "df1['écran']=df1['écran'].astype(str)\n",
    "df1['reseau']=df1['reseau'].astype(str)\n",
    "df1['Marque']=df1['Marque'].astype(str)\n",
    "df1['exploit']=df1['exploit'].astype(str)\n",
    "df1['prix']=df1['prix'].astype(str)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pour le nettoyage de du variable réseau nous avons transformé les modalités:\n",
    "    - 5G si le téléphone est doté d'un réseau 5G\n",
    "    - 4G si le téléphone est doté d'un réseau 4G\n",
    "    - 3G si le téléphone est doté d'un réseau 3G\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Nettoyage.nettoyage_reseau(df1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En ce qui concerne le nettoyage des données provenant de la variable prix nous avons remplacé les virgules par des points pour pouvoir ensuite les transformer en type numérique, ensuite nous avons enlevé le signe euro et le \"nan\" présent dans cette colonne."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "Nettoyage.nettoyage_prix(df1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "En ce qui concerne le nettoyage des données provenant de la variable exploit nous avons effacé les chaines de caractère \"['\" et \"']\" présent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Nettoyage.nettoyage_exploit(df1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "À partir de le variable exploit nous avons créé deux variables la variable RAM et la variable mémoire. Dans le cas où l'information sur le RAM et la mémoire serait inexistante dans le variable exploit nous avons défini la modalité \"inconnu\". Et nous avons pour la fin supprimée la colonne associée à la variable exploit.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Nettoyage.col_Ram(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Nettoyage.cor_exploit(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Nettoyage.col_memoire(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>Marque</th>\n",
       "      <th>reseau</th>\n",
       "      <th>prix</th>\n",
       "      <th>écran</th>\n",
       "      <th>Ram</th>\n",
       "      <th>Mémoire</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>oneplus</td>\n",
       "      <td>4G</td>\n",
       "      <td>172.58</td>\n",
       "      <td>6.52</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>huawei</td>\n",
       "      <td>4G</td>\n",
       "      <td>196.59</td>\n",
       "      <td>6.4</td>\n",
       "      <td>RAM 6 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>huawei</td>\n",
       "      <td>4G</td>\n",
       "      <td>214.50</td>\n",
       "      <td>6.15</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>samsung</td>\n",
       "      <td>4G</td>\n",
       "      <td>374</td>\n",
       "      <td>6.4</td>\n",
       "      <td>RAM 6 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>samsung</td>\n",
       "      <td>4G</td>\n",
       "      <td>270</td>\n",
       "      <td>5.8</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>huawei</td>\n",
       "      <td>4G</td>\n",
       "      <td>189</td>\n",
       "      <td>5.84</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>samsung</td>\n",
       "      <td>4G</td>\n",
       "      <td>289.99</td>\n",
       "      <td>6.5</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>inconnu</td>\n",
       "      <td>5G</td>\n",
       "      <td>320.88</td>\n",
       "      <td>6.6</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>google</td>\n",
       "      <td>5G</td>\n",
       "      <td>825</td>\n",
       "      <td>6</td>\n",
       "      <td>RAM 8 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Marque reseau    prix écran        Ram Mémoire\n",
       "1   oneplus     4G  172.58  6.52   RAM 4 Go    64  \n",
       "2    huawei     4G  196.59   6.4   RAM 6 Go   128  \n",
       "5    huawei     4G  214.50  6.15   RAM 4 Go   128  \n",
       "6   samsung     4G     374   6.4   RAM 6 Go   128  \n",
       "7   samsung     4G     270   5.8   RAM 4 Go    64  \n",
       "8    huawei     4G     189  5.84   RAM 4 Go    64  \n",
       "9   samsung     4G  289.99   6.5   RAM 4 Go   128  \n",
       "10  inconnu     5G  320.88   6.6   RAM 4 Go   128  \n",
       "11   google     5G     825     6   RAM 8 Go   128  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.pop('exploit')\n",
    "df1[1:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Inconnu        330\n",
      " RAM 4 Go      181\n",
      " RAM 6 Go      157\n",
      " RAM 8 Go      124\n",
      " RAM 3 Go       84\n",
      " RAM 2 Go       61\n",
      " RAM 1 Go       18\n",
      " RAM 12 Go       9\n",
      " RAM 512 Mo      7\n",
      " Inconnu         3\n",
      " RAM 16 Go       3\n",
      "Name: Ram, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(df1['Ram'].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "Nettoyage.nettoyage_prix(df1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "samsung       267\n",
      "apple         258\n",
      "xiaomi        114\n",
      "huawei         93\n",
      "inconnu        64\n",
      "sony           29\n",
      "oneplus        29\n",
      "nokia          26\n",
      "google         21\n",
      "honor          21\n",
      "asus           15\n",
      "oppo           14\n",
      "lg              9\n",
      "blackberry      8\n",
      "realme          8\n",
      "alcatel         1\n",
      "Name: Marque, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print(df1.Marque.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "df1=df1.reset_index()\n",
    "del df1['index']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Àprès avoir nettoyé toute notre base de données et ajouté deux variables  nous nous retrouvons avec une base de données de 977 observations et 6 variables.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>Marque</th>\n",
       "      <th>reseau</th>\n",
       "      <th>prix</th>\n",
       "      <th>écran</th>\n",
       "      <th>Ram</th>\n",
       "      <th>Mémoire</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>oneplus</td>\n",
       "      <td>4G</td>\n",
       "      <td>172.58</td>\n",
       "      <td>6.52</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>huawei</td>\n",
       "      <td>4G</td>\n",
       "      <td>196.59</td>\n",
       "      <td>6.4</td>\n",
       "      <td>RAM 6 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>huawei</td>\n",
       "      <td>4G</td>\n",
       "      <td>214.50</td>\n",
       "      <td>6.15</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>samsung</td>\n",
       "      <td>4G</td>\n",
       "      <td>374</td>\n",
       "      <td>6.4</td>\n",
       "      <td>RAM 6 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>samsung</td>\n",
       "      <td>4G</td>\n",
       "      <td>270</td>\n",
       "      <td>5.8</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>huawei</td>\n",
       "      <td>4G</td>\n",
       "      <td>189</td>\n",
       "      <td>5.84</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>samsung</td>\n",
       "      <td>4G</td>\n",
       "      <td>289.99</td>\n",
       "      <td>6.5</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>inconnu</td>\n",
       "      <td>5G</td>\n",
       "      <td>320.88</td>\n",
       "      <td>6.6</td>\n",
       "      <td>RAM 4 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>google</td>\n",
       "      <td>5G</td>\n",
       "      <td>825</td>\n",
       "      <td>6</td>\n",
       "      <td>RAM 8 Go</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Marque reseau    prix écran        Ram Mémoire\n",
       "1  oneplus     4G  172.58  6.52   RAM 4 Go    64  \n",
       "2   huawei     4G  196.59   6.4   RAM 6 Go   128  \n",
       "3   huawei     4G  214.50  6.15   RAM 4 Go   128  \n",
       "4  samsung     4G     374   6.4   RAM 6 Go   128  \n",
       "5  samsung     4G     270   5.8   RAM 4 Go    64  \n",
       "6   huawei     4G     189  5.84   RAM 4 Go    64  \n",
       "7  samsung     4G  289.99   6.5   RAM 4 Go   128  \n",
       "8  inconnu     5G  320.88   6.6   RAM 4 Go   128  \n",
       "9   google     5G     825     6   RAM 8 Go   128  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1[1:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PARTIE DESCRIPTIVE"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pour commencer cette partie nous transformons toutes nos variables en type \"category\" sauf la variable correspondant au prix."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#str\n",
    "df1.Ram=df1.Ram.astype(\"category\")\n",
    "df1.reseau=df1.reseau.astype(\"category\")\n",
    "df1.Marque=df1.Marque.astype(\"category\")\n",
    "df1['écran']=df1['écran'].astype(\"category\")\n",
    "df1['Mémoire']=df1['Mémoire'].astype(\"category\")\n",
    "df1.prix=pd.to_numeric(df1[\"prix\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "complet=df1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Avec cette commande nous montrons le nombre de modalités par variable, modalité la plus représentée pour chaque variable et leur nombre d'apparition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "         Marque reseau écran      Ram  Mémoire\n",
      "count       977    977   977      977      977\n",
      "unique       16      3    72       11        9\n",
      "top     samsung     4G   6.1  Inconnu  Inconnu\n",
      "freq        267    784    83      330      333\n"
     ]
    }
   ],
   "source": [
    "price = complet.set_index('prix') # transformer la variable en\n",
    "print(price.describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ici nous montrons quelques informations sur le variable prix, notamment la moyenne, le minimum et le maximum."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>prix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>977.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>370.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>321.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>15.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>159.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>255.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>469.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2099.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          prix\n",
       "count   977.00\n",
       "mean    370.50\n",
       "std     321.79\n",
       "min      15.00\n",
       "25%     159.40\n",
       "50%     255.00\n",
       "75%     469.99\n",
       "max    2099.00"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.describe().round(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import numpy\n",
    "import pandas\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DeprecationWarning"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nous remarquons grace à ces deux graphiques que la majorité des téléphones ont un prix compris entre 15 et 500 euros."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20f5a8f91c0>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAD4CAYAAAAD6PrjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAASjklEQVR4nO3df5BdZX3H8feXEAlgUEkCTbPBDU4KCYoCkdohTatARVDBViRMaUNhCLVx1KmdGtRBYcxM7IzgD0oR0CGiTQz4g7TWsZiKDjNoDC1CQgzEsoY1GRIDFRUJIXz7xz37eEk22bvJnr27e9+vmZ17znPPufd7Hy/5eM557nMiM5EkCeCQdhcgSRo5DAVJUmEoSJIKQ0GSVBgKkqTi0HYXcDAmT56c3d3d7S5DkkaV+++//xeZOaW/50Z1KHR3d7N27dp2lyFJo0pE/Gxfz3n6SJJUGAqSpMJQkCQVo/qagiQNhV27dtHb28uzzz7b7lKG1IQJE+jq6mL8+PEt72MoSOp4vb29TJw4ke7ubiKi3eUMicxkx44d9Pb2MmPGjJb38/SRpI737LPPMmnSpDETCAARwaRJkwZ99GMoSBKMqUDocyCfyVCQJBVeU5CkPXQv/uaQvl7P0vOG7LWuvvpq5s2bx1lnnTVkr9mso0NhqP+Hb9VQfkEkdY7du3dz7bXX1voenj6SpBGgp6eHE088kQULFnDyySfzzne+k2eeeYbu7m6uvfZa5s6dyx133MGll17KnXfeyS9/+UtOOOEENm7cCMDFF1/MLbfcctB1GAqSNEJs3LiRhQsX8uCDD3LUUUdx4403Ao3fG9x7773Mnz+/bPuyl72MG264gUsvvZQVK1bw1FNPccUVVxx0DYaCJI0Q06dP54wzzgDgkksu4d577wXgoosu6nf7s88+m9e85jUsWrSIW2+9dUhqMBQkaYTYcwhp3/qRRx7Z7/YvvPACGzZs4PDDD+fJJ58ckhoMBUkaITZv3sx9990HwPLly5k7d+5+t7/++uuZNWsWy5cv57LLLmPXrl0HXUNHjz6SpP60a4TgrFmzWLZsGVdeeSUzZ87k3e9+N5/97Gf73faRRx7h1ltvZc2aNUycOJF58+bx8Y9/nGuuueagajAUJGmEOOSQQ7jpppte1NbT0/Oi9dtuu60sb9iwoSxfd911Q1PDkLyKJGlMMBQkaQTo7u5m3bp17S7DUJAkaEw1PdYcyGcyFCR1vAkTJrBjx44xFQx991OYMGHCoPar/UJzRIwD1gI/z8y3RsTRwFeAbqAHeFdmPlVtexVwObAbeG9mfrvu+iSpq6uL3t5etm/f3u5ShlTfndcGYzhGH70P2AAcVa0vBlZn5tKIWFytfzAiZgPzgZOA3we+ExF/kJm7h6FGSR1s/Pjxg7o72VhW6+mjiOgCzgOaf399PrCsWl4GXNDUviIzd2bmY8Am4PQ665MkvVjd1xQ+Bfwj8EJT27GZuRWgejymap8GPN60XW/V9iIRsTAi1kbE2rF2qCdJ7VZbKETEW4FtmXl/q7v007bXVZ/MvDkz52TmnClTphxUjZKkF6vzmsIZwNsj4lxgAnBURHwJeCIipmbm1oiYCmyrtu8Fpjft3wVsqbE+SdIeajtSyMyrMrMrM7tpXED+r8y8BFgFLKg2WwDcVS2vAuZHxGERMQOYCaypqz5J0t7aMffRUmBlRFwObAYuBMjM9RGxEngYeB5Y5MgjSRpewxIKmXkPcE+1vAM4cx/bLQGWDEdNkqS9+YtmSVJhKEiSCkNBklQYCpKkwlCQJBWGgiSpMBQkSYWhIEkqDAVJUmEoSJIKQ0GSVBgKkqTCUJAkFYaCJKkwFCRJhaEgSSoMBUlSYShIkgpDQZJUGAqSpMJQkCQVhoIkqTAUJEmFoSBJKgwFSVJhKEiSCkNBklQYCpKkwlCQJBWGgiSpMBQkSYWhIEkqDAVJUmEoSJIKQ0GSVBgKkqTCUJAkFYaCJKkwFCRJRW2hEBETImJNRPw4ItZHxDVV+9ERcXdEPFo9vqJpn6siYlNEbIyIN9dVmySpf3UeKewE3pSZrwVeB5wTEW8AFgOrM3MmsLpaJyJmA/OBk4BzgBsjYlyN9UmS9lBbKGTDr6vV8dVfAucDy6r2ZcAF1fL5wIrM3JmZjwGbgNPrqk+StLdarylExLiIeADYBtydmT8Ejs3MrQDV4zHV5tOAx5t2763a9nzNhRGxNiLWbt++vc7yJanj1BoKmbk7M18HdAGnR8Sr97N59PcS/bzmzZk5JzPnTJkyZahKlSQxTKOPMvP/gHtoXCt4IiKmAlSP26rNeoHpTbt1AVuGoz5JUkOdo4+mRMTLq+XDgbOAnwCrgAXVZguAu6rlVcD8iDgsImYAM4E1ddUnSdrboTW+9lRgWTWC6BBgZWb+e0TcB6yMiMuBzcCFAJm5PiJWAg8DzwOLMnN3jfVJkvZQWyhk5oPAKf207wDO3Mc+S4AlddUkSdo/f9EsSSoMBUlSYShIkgpDQZJUGAqSpMJQkCQVLQ1JjYhXZ+a6uovpFN2Lv9mW9+1Zel5b3lfS6NHqkcJN1b0R/q7vV8qSpLGnpVDIzLnAX9KYm2htRPxrRJxda2WSpGHX8jWFzHwU+AjwQeBPgM9ExE8i4s/rKk6SNLxaCoWIODkirgc2AG8C3paZs6rl62usT5I0jFqd++gG4BbgQ5n5277GzNwSER+ppTJJ0rBrNRTOBX7bN2tpRBwCTMjMZzLz9tqqkyQNq1avKXwHOLxp/YiqTZI0hrQaChMy89d9K9XyEfWUJElql1ZD4TcRcWrfSkScBvx2P9tLkkahVq8pvB+4IyL67pk8FbionpIkSe3SUihk5o8i4kTgBCCAn2TmrlorkyQNu8HcjvP1QHe1zykRQWZ+sZaqJElt0eqEeLcDrwIeAHZXzQkYCpI0hrR6pDAHmJ2ZWWcxkqT2anX00Trg9+osRJLUfq0eKUwGHo6INcDOvsbMfHstVUmS2qLVUPhYnUVIkkaGVoekfi8iXgnMzMzvRMQRwLh6S5MkDbdWp86+ArgT+FzVNA34Rl1FSZLao9ULzYuAM4Cnodxw55i6ipIktUerobAzM5/rW4mIQ2n8TkGSNIa0Ggrfi4gPAYdX92a+A/i3+sqSJLVDq6GwGNgOPARcCfwHjfs1S5LGkFZHH71A43act9RbjiSpnVqd++gx+rmGkJnHD3lFkqS2GczcR30mABcCRw99OZKkdmrpmkJm7mj6+3lmfgp4U821SZKGWaunj05tWj2ExpHDxFoqkiS1Taunjz7ZtPw80AO8a8irkSS1Vaujj95YdyGSpPZr9fTR3+/v+cy8bmjKkSS102BGH70eWFWtvw34PvB4HUVJktqj1V80TwZOzcwPZOYHgNOArsy8JjOv6W+HiJgeEd+NiA0RsT4i3le1Hx0Rd0fEo9XjK5r2uSoiNkXExoh488F+OEnS4LQaCscBzzWtPwd0D7DP88AHMnMW8AZgUUTMpjFlxurMnAmsrtapnpsPnAScA9wYEd6zQZKGUaunj24H1kTE12n8svkdwBf3t0NmbgW2Vsu/iogNNO7DcD7wp9Vmy4B7gA9W7SsycyfwWERsAk4H7hvE55EkHYRWRx8tiYhvAX9cNf1NZv5Pq28SEd3AKcAPgWOrwCAzt0ZE330ZpgE/aNqtt2rb87UWAgsBjjvuuFZLkCS1oNXTRwBHAE9n5qeB3oiY0cpOEfFS4KvA+zPz6f1t2k9bf/Mt3ZyZczJzzpQpU1opQZLUolZvx/lRGqd4rqqaxgNfamG/8TQC4cuZ+bWq+YmImFo9PxXYVrX3AtObdu8CtrRSnyRpaLR6pPAO4O3AbwAycwsDTHMREQF8Htiwx+8YVgELquUFwF1N7fMj4rDqKGQmsKbF+iRJQ6DVC83PZWZGRAJExJEt7HMG8FfAQxHxQNX2IWApsDIiLgc205hxlcxcHxErgYdpjFxalJm7W/8okqSD1WoorIyIzwEvj4grgMsY4IY7mXkv/V8nADhzH/ssAZa0WJMkaYgNGArVaaCvACcCTwMnAFdn5t011yZJGmYDhkJ12ugbmXkaYBBI0hjW6oXmH0TE62utRJLUdq1eU3gj8LcR0UNjBFLQOIg4ua7CJEnDb7+hEBHHZeZm4C3DVI8kqY0GOlL4Bo3ZUX8WEV/NzL8YjqIkSe0x0DWF5iGlx9dZiCSp/QYKhdzHsiRpDBro9NFrI+JpGkcMh1fL8LsLzUfVWp0kaVjtNxQy05vcSFIHGczU2ZKkMc5QkCQVhoIkqTAUJEmFoSBJKgwFSVJhKEiSCkNBklQYCpKkwlCQJBWGgiSpMBQkSYWhIEkqDAVJUmEoSJIKQ0GSVBgKkqTCUJAkFYaCJKkwFCRJhaEgSSoMBUlSYShIkgpDQZJUGAqSpMJQkCQVhoIkqTAUJEmFoSBJKg6t64Uj4gvAW4Ftmfnqqu1o4CtAN9ADvCszn6qeuwq4HNgNvDczv11XbZ2qe/E32/bePUvPa9t7S2pdnUcKtwHn7NG2GFidmTOB1dU6ETEbmA+cVO1zY0SMq7E2SVI/aguFzPw+8OQezecDy6rlZcAFTe0rMnNnZj4GbAJOr6s2SVL/hvuawrGZuRWgejymap8GPN60XW/VtpeIWBgRayNi7fbt22stVpI6zUi50Bz9tGV/G2bmzZk5JzPnTJkypeayJKmzDHcoPBERUwGqx21Vey8wvWm7LmDLMNcmSR1vuENhFbCgWl4A3NXUPj8iDouIGcBMYM0w1yZJHa/OIanLgT8FJkdEL/BRYCmwMiIuBzYDFwJk5vqIWAk8DDwPLMrM3XXVJknqX22hkJkX7+OpM/ex/RJgSV31SJIGNlIuNEuSRgBDQZJU1Hb6SGrWrik2nF5DGhyPFCRJhaEgSSoMBUlSYShIkgpDQZJUGAqSpMJQkCQVhoIkqTAUJEmFoSBJKgwFSVJhKEiSCkNBklQYCpKkwlCQJBWGgiSpMBQkSYWhIEkqvB2nxrR23QYUvBWoRiePFCRJhaEgSSoMBUlSYShIkgpDQZJUGAqSpMJQkCQVhoIkqTAUJEmFoSBJKpzmQhpjnNpDB8MjBUlSYShIkgpDQZJUGAqSpMJQkCQVjj6SatLOUUDSgRpxoRAR5wCfBsYBt2bm0jaXJKlF7QpCh8IOnREVChExDvhn4GygF/hRRKzKzIfbW5kk7W0s/iZkRIUCcDqwKTP/FyAiVgDnA4aCpH3yVN3QGWmhMA14vGm9F/jD5g0iYiGwsFr9dURsHOR7TAZ+ccAVdgb7aGD20cDso4EdcB/FJw7qfV+5rydGWihEP235opXMm4GbD/gNItZm5pwD3b8T2EcDs48GZh8NbCT20UgbktoLTG9a7wK2tKkWSeo4Iy0UfgTMjIgZEfESYD6wqs01SVLHGFGnjzLz+Yh4D/BtGkNSv5CZ64f4bQ741FMHsY8GZh8NzD4a2Ijro8jMgbeSJHWEkXb6SJLURoaCJKnoqFCIiHMiYmNEbIqIxe2up50ioiciHoqIByJibdV2dETcHRGPVo+vaNr+qqrfNkbEm9tXeX0i4gsRsS0i1jW1DbpPIuK0qm83RcRnIqK/odaj0j766GMR8fPqu/RARJzb9FxH9VFETI+I70bEhohYHxHvq9pHz/coMzvij8aF658CxwMvAX4MzG53XW3sjx5g8h5t/wQsrpYXA5+olmdX/XUYMKPqx3Ht/gw19Mk84FRg3cH0CbAG+CMav7v5FvCWdn+2mvvoY8A/9LNtx/URMBU4tVqeCDxS9cOo+R510pFCmUIjM58D+qbQ0O+cDyyrlpcBFzS1r8jMnZn5GLCJRn+OKZn5feDJPZoH1ScRMRU4KjPvy8Z/2V9s2mfU20cf7UvH9VFmbs3M/66WfwVsoDFTw6j5HnVSKPQ3hca0NtUyEiTwnxFxfzV1CMCxmbkVGl9u4JiqvZP7brB9Mq1a3rN9rHtPRDxYnV7qOzXS0X0UEd3AKcAPGUXfo04KhQGn0OgwZ2TmqcBbgEURMW8/29p3e9tXn3RiX/0L8CrgdcBW4JNVe8f2UUS8FPgq8P7MfHp/m/bT1tY+6qRQcAqNJpm5pXrcBnydxumgJ6rDVqrHbdXmndx3g+2T3mp5z/YxKzOfyMzdmfkCcAu/O7XYkX0UEeNpBMKXM/NrVfOo+R51Uig4hUYlIo6MiIl9y8CfAeto9MeCarMFwF3V8ipgfkQcFhEzgJk0LoJ1gkH1SXVq4FcR8YZqtMhfN+0zJvX9Y1d5B43vEnRgH1Wf5/PAhsy8rump0fM9avfV+uH8A86lMRrgp8CH211PG/vheBojHn4MrO/rC2ASsBp4tHo8ummfD1f9tpExMlKkn35ZTuP0xy4a/0/t8gPpE2AOjX8YfwrcQDVzwFj420cf3Q48BDxI4x+5qZ3aR8BcGqd5HgQeqP7OHU3fI6e5kCQVnXT6SJI0AENBklQYCpKkwlCQJBWGgiSpMBQkSYWhIEkq/h8hrtJ9CdK42gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df1.plot.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20f5b0af760>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZUAAAD4CAYAAAAkRnsLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXhcd33v8fdXo13WbtmW5d2xkzi742wsgRYCSeDiQguYQhNSaJqH0EJp722AtrfthULpA5TQQEhp2oQCITQEXDCkSWjS0MaJnYUQr/ES24plWbb2fZnv/WOOnLGsZSTNmTOSPq/nmUejc85v5jcnE3/0W87vmLsjIiKSDjlRV0BERGYPhYqIiKSNQkVERNJGoSIiImmjUBERkbTJjboCUZo/f76vWLEi6mqIiMwozzzzzAl3rxlt35wOlRUrVrB9+/aoqyEiMqOY2aGx9qn7S0RE0kahIiIiaaNQERGRtFGoiIhI2ihUREQkbRQqIiKSNgoVERFJG4WKZJ2huHP/9iM8vrcp6qqIyCTN6YsfJTt984kDfO6nuwH4/i1XcdmKqohrJCKpUktFskr/YJx/fOIAG5ZXMn9ePv/w831RV0lEJkGhIlll+6FmTnT2c/PVq9h02TKeeKmJ1u7+qKslIilSqEhW+a+9J8jNMV5z1nzevG4hcYfH9mhsRWSmUKhIVnnmUDMXLilnXkEuF9aVU16Ux1MHT0ZdLRFJkUJFsoa7s7uhg3WLywDIyTEuXlrBs4daI66ZiKRKoSJZo76lh46+Qc6tLTu1bf2ySvYe76CjdyDCmolIqhQqkjV2NbQDnBYq5y0uwx32NnZGVS0RmQSFimSNXQ0dmME5i0pPbTs7eL7nWEdU1RKRSVCoSNbY19TJ0spiivNfvSa3rqKIkvwYexsVKiIzgUJFssaR5m6WVRWfti0nx1i7qFQtFZEZQqEiWeNIczdLR4QKwNkLS9mjlorIjKBQkazQ2TfIya7+M1oqAGctmEdzV7+urBeZARQqkhWONHcDsLSq6Ix9y6tLAHj5ZHdG6yQik6dQkawwHCqjtVRWVCe2HTrZldE6icjkKVQkKxweJ1SGx1kOqaUikvVCDRUzu9bM9pjZPjO7bZT9Zma3B/tfMLP1E5U1s78zs93B8Q+aWUXSvk8Gx+8xs7eG+dkkvY629lKcH6O8KO+MfYV5MWrLC3lZLRWRrBdaqJhZDLgDuA5YB7zPzNaNOOw6YE3wuBn4egplHwbOd/cLgb3AJ4My64BNwHnAtcDXgteRGaCxo5dFZYWY2aj7l1cXc1gtFZGsF2ZL5XJgn7sfcPd+4D5g44hjNgL3esJWoMLMascr6+7/4e6DQfmtwJKk17rP3fvc/SCwL3gdmQGOt/eyoKxgzP3Lq0o0UC8yA4QZKnXAkaTf64NtqRyTSlmA3wV+Oon3w8xuNrPtZra9qUn36cgWje19LCwrHHP/8vnFnOjso7NvcMxjRCR6YYbKaP0YnuIxE5Y1s08Dg8C3J/F+uPtd7r7B3TfU1NSMUkQyzd1pbO8dP1SqEtOKNQNMJLuFGSr1wNKk35cAR1M8ZtyyZnYj8Hbg/e4+HBypvJ9kofaeQfoG4ywoHbv7a/j6lVdaejJVLRGZgjBDZRuwxsxWmlk+iUH0zSOO2QzcEMwCuxJoc/eG8cqa2bXAnwLvcPfuEa+1ycwKzGwlicH/p0P8fJImjR29ACwYp6WyuCIIlVaFikg2y534kKlx90Ez+yjwEBAD7nb3HWZ2S7D/TmALcD2JQfVu4KbxygYv/Q9AAfBwMFNoq7vfErz2/cBOEt1it7r7UFifT9KnsT0RKgvHaalUl+RTkJvDUYWKSFYLLVQA3H0LieBI3nZn0nMHbk21bLD9rHHe77PAZ6daX4lGY3sfwLhjKmZGXUWRWioiWU5X1Evkhlsq400pBqirLOKV1t5MVElEpkihIpE73t5LaWHuaTfnGs3i8iIN1ItkOYWKRO54x/jXqAyrqyziRGcfvQMaKhPJVgoVidzxjj5q5o3f9QWJWwsDNLSpC0wkWylUJHItXf1Uzcuf8LjhacWaASaSvRQqErmTXf1Ul0wcKksqdQGkSLZTqEikBobitPUMUJVCqCwsK8RMF0CKZDOFikSqJbjvfCqhkp+bw8LSQoWKSBZTqEikmrtSDxWAxRWF6v4SyWIKFYlUc+dkQ6WIhjaFiki2UqhIpE4GLZXqkomnFMNwqPTy6uLUIpJNFCoSqcmMqQAsLi+kbzB+qttMRLKLQkUidTLo/qoszkvp+NpT16roAkiRbKRQkUg1d/VTUZxHbiy1r+LwVfVHNa4ikpUUKhKp5q7+lLu+AGrLE2uE6ap6keykUJFInezqS+lq+mFVwc26tP6XSHZSqEikmrv6qSxOPVTMjMW6WZdI1lKoSKSau/qpTmExyWS15YU0KFREspJCRSITjzst3amt+5VscUWRZn+JZCmFikSmvXeAobhTleKFj8MWlxdyvKOXwaF4SDUTkalSqEhkXr2afvItlbhDY0dfGNUSkWlQqEhkJruY5LBa3axLJGspVCQyrd0DAJOa/QWJ7i9QqIhkI4WKRKY1WPerIsUlWoZpqRaR7KVQkci09SRaKuWTDJV5BbmUFeZqCXyRLKRQkci0dg8QyzFKC3InXTYxrVihIpJtFCoSmdaefsqL8jCzSZfVtSoi2UmhIpFp7R6gomhyXV/DFlcUqvtLJAspVCQybT0Dkx5PGVZbXkRL9wA9/UNprpWITIdCRSLT2j1A+TRaKqD7qohkG4WKRKatZxrdX+W6AFIkGylUJDKt3f1UTPLCx2GLg2tVGjRYL5JVFCoSiaG40947OOXur4VlhZih+6qIZBmFikSiPbjwcbJX0w/Lz82hZl6BZoCJZBmFikSidZqhAokuMN1WWCS7KFQkEqfW/Sqa2pgKJGaAqftLJLuEGipmdq2Z7TGzfWZ22yj7zcxuD/a/YGbrJyprZu82sx1mFjezDUnbV5hZj5k9HzzuDPOzyfS0TnHdr2S15UU0tPbi7umqlohM0+QXXUqRmcWAO4BrgHpgm5ltdvedSYddB6wJHlcAXweumKDsi8C7gG+M8rb73f3isD6TpE9bsOz9VKcUQ6L7q2dgiNbuASoneU8WEQlHmC2Vy4F97n7A3fuB+4CNI47ZCNzrCVuBCjOrHa+su+9y9z0h1lsyYLj7a6qzvyDpvioarBfJGmGGSh1wJOn3+mBbKsekUnY0K83sOTN73MxeP9oBZnazmW03s+1NTU0pvKSEoa1nEJhmqOi+KiJZJ8xQGW3p2ZGd32Mdk0rZkRqAZe5+CfAJ4DtmVnbGi7jf5e4b3H1DTU3NBC8pYWnt6ae0IJfc2NS/grXBUi2aViySPcIMlXpgadLvS4CjKR6TStnTuHufu58Mnj8D7AfWTqnmErq27qkvJjlsfkkBeTFTS0Uki4QZKtuANWa20szygU3A5hHHbAZuCGaBXQm0uXtDimVPY2Y1wQA/ZraKxOD/gfR+JEmX1p6BaV2jApCTY9SW62ZdItkktNlf7j5oZh8FHgJiwN3uvsPMbgn23wlsAa4H9gHdwE3jlQUws3cCXwVqgJ+Y2fPu/lbgauCvzWwQGAJucffmsD6fTE9rd/+0rlEZVluu+6qIZJPQQgXA3beQCI7kbXcmPXfg1lTLBtsfBB4cZfsDwAPTrLJkSGvPALXBQPt01FUU8dRB/e0gki10Rb1Eom0ad31MVltRyLH2XobiugBSJBsoVCTj3D0tYyqQmFY8FHeOd2iwXiQbKFQk4zr7BhmK+7SuURn26s26FCoi2UChIhnXNrxCcRoG6usqE6FS39I97dcSkelTqEjGtXZPfzHJYUsriwE40qxQEckGChXJuFdbKtMPlaL8GDWlBRw6qVARyQYKFcm44ZbKVO9PP9LyqmIOq6UikhUUKpJxrT3BDbrS0P0FsKxaoSKSLRQqknGnxlTS0P0FsLyqhGPtvfQODKXl9URk6hQqknFtPQMU5uVQmBdLy+stqy7CXTPARLKBQkUyLl3rfg1bVlUCoC4wkSygUJGMa+0eSFvXF8Dy6sS0Ys0AE4meQkUyrq1n+vdSSVZdkk9JfkyhIpIFUgoVM3vAzN5mZgohmba2nvQsJjnMzFiqacUiWSHVkPg68NvAS2b2eTM7J8Q6ySyX7u4vSHSBKVREopdSqLj7I+7+fmA98DLwsJn9j5ndZGbp/ddBZr22NK1QnGx5dQmHm7uJawl8kUil3J1lZtXAB4EPA88BXyERMg+HUjOZlXoHhugZGErb1fTDllUV0z8Y51i7VisWiVJKd340sx8A5wDfAv5XcB95gO+Z2fawKiezT3uw7ldZmru/Vs1PTCs+eKKLxWm4o6SITE2qLZVvuvs6d//ccKCYWQGAu28IrXYy66RzMclkqxfMA2B/U2daX1dEJifVUPnMKNueTGdFZG5oHQ6VNI+pLCgtYF5BLvuPK1REojRu95eZLQLqgCIzuwSwYFcZUBxy3WQWSve6X8PMjNU1Jexv6krr64rI5Ew0pvJWEoPzS4AvJW3vAD4VUp1kFkvnXR9HWl0zjycPnEz764pI6sYNFXe/B7jHzH7T3R/IUJ1kFmvtTix7n84r6oetXjCPHzz3Cp19g8wrSGkOioik2UTdXx9w938FVpjZJ0bud/cvjVJMZExtPQOYQWkI/+ivrglmgDV1ccGS8rS/vohMbKKB+pLg5zygdJSHyKS09SSups/JsYkPnqTVNZoBJhK1ibq/vhH8/KvMVEdmu9bu9K77lWxZdTGxHFOoiEQo1QUlv2BmZWaWZ2aPmtkJM/tA2JWT2ae1Z4DyNF9NP6wgN8ayqmJealSoiEQl1etU3uLu7cDbgXpgLfC/Q6uVzFrD3V9hOXthKXsaO0J7fREZX6qhMvyvwPXAd929OaT6yCzX1t0fWvcXwDm1pbx8soueft2vXiQKqYbKv5vZbmAD8KiZ1QBauU8mrTWEFYqTnbOoDHfYq9aKSCRSXfr+NuAqYIO7DwBdwMYwKyazTzzutIfc/XVubWJS4u5j7aG9h4iMbTIXC5xL4nqV5DL3prk+Mot19A0S9/Qv0ZJsaWUxxfkxdjWopSIShVSXvv8WsBp4HhjurHYUKjIJbd3Di0mGM/sLICfHWLuwVC0VkYik2lLZAKxzd91WT6ZseN2vMFsqkOgC+9mLx3B3zNJ/kaWIjC3VgfoXgUVhVkRmv9aexLpfYQ7UQ2KwvqV7gOMdfaG+j4icKdVQmQ/sNLOHzGzz8GOiQmZ2rZntMbN9ZnbbKPvNzG4P9r9gZusnKmtm7zazHWYWN7MNI17vk8Hxe8zsrSl+NsmQ4WXvw5xSDHDOosRg/a4GdYGJZFqq3V9/OdkXNrMYcAdwDYkLJreZ2WZ335l02HXAmuBxBfB14IoJyr4IvAv4xoj3WwdsAs4DFgOPmNlad9cFC1kiU91f5ywqA2BXQwdvPHtBqO8lIqdLdUrx48DLQF7wfBvw7ATFLgf2ufsBd+8H7uPMacgbgXs9YStQYWa145V1913uvmeU99sI3Ofufe5+ENgXvI5kibaQ7k8/UnlxHkurinjxlbZQ30dEzpTq2l+/B/wbr7YO6oAfTlCsDjiS9Ht9sC2VY1IpO5X3w8xuNrPtZra9qalpgpeUdGrt7qcoL0ZhXiz097qwroIXXmkN/X1E5HSpjqncCrwWaAdw95eAifoVRpt2M3L22FjHpFJ2Ku+Hu9/l7hvcfUNNTc0ELynpFPa6X8kuWFLOkeYemrv6M/J+IpKQaqj0Bd1QAAQXQE70j3w9sDTp9yXA0RSPSaXsVN5PItTaHe4SLckuDG7S9St1gYlkVKqh8riZfQooMrNrgO8D/z5BmW3AGjNbaWb5JAbRR84Y2wzcEMwCuxJoc/eGFMuOtBnYZGYFZraSxOD/0yl+PsmA1gy2VM6vC0KlXl1gIpmUaqjcBjQBvwJ+H9gC/Nl4Bdx9EPgo8BCwC7jf3XeY2S1mdktw2BbgAIlB9X8EPjJeWQAze6eZ1ZNYi+wnZvZQUGYHcD+wE/gZcKtmfmWXsNf9SlZWmMeq+SX8sl4tFZFMSmlKsbvHzeyHwA/dPeXRbXffQiI4krfdmfTcSYzXpFQ22P4g8OAYZT4LfDbV+klmtXYPcOGSzIQKJLrAth7QXRpEMmnclkrQLfWXZnYC2A3sMbMmM/uLzFRPZpPWnv5Q1/0a6YIlFRxr7+V4u+7SIJIpE3V/fZzErK/L3L3a3atIXKT4WjP7o9BrJ7NG78AQvQPxjHV/AVwUDNa/oC4wkYyZKFRuAN4XXEwIgLsfAD4Q7BNJSXuGrqZPtm5xGTkGL2iwXiRjJgqVPHc/MXJjMK6SuX8dZMZrObXsfea+NsX5uaxdWMpzRxQqIpkyUaiMd+WYriqTlA1fhFiVwTEVgEuXV/L84VaG4rprg0gmTBQqF5lZ+yiPDuCCTFRQZofW7kSoVJZkNlQ2rKiko29Q96wXyZBxpxS7e/iLNMmc0ByESlWGQ+XSZVUAPHOohXNryzL63iJzUaoXP4pMS0tXZm7QNdLSqiJqSgt45lBLRt9XZK5SqEhGNHcNUJIfoyA3s41fM+PSZZUKFZEMUahIRrR292d8PGXYhhWVHG7u5niHLoIUCZtCRTKiubufygzP/Bq2fnklAM+qtSISOoWKZERLV3QtlfMXl5Ofm8P2lxUqImFTqEhGtHQPUJXhQfph+bk5XLSknGcOK1REwqZQkYyIsqUCcOnyKl58pY3eAd0NQSRMChUJXf9gnI6+wcjGVACuWFnFwJBrXEUkZAoVCV1UV9Mn27CikliOsfXAycjqIDIXKFQkdMOLSWZ63a9kpYV5nF9XzpMKFZFQKVQkdMOLSVZGNFA/7MpVVTx/pJWefo2riIRFoSKha8mC7i+Aq1ZVMzDkurpeJEQKFQndqWXvIw6Vy1ZUEcsxnjxwxi2CRCRNFCoSuuGB+kwvJjlSSUEuFy4p58n9GlcRCYtCRUIX1WKSo7lqVTUv1LfR1TcYdVVEZiWFioSupbufighnfiV7zer5DMadpw6qtSISBoWKhO5EZx/z52VHqGxYUUlRXozH9jRFXRWRWUmhIqE72dnP/HkFUVcDgMK8GFetrubxvQoVkTAoVCR0iZZKdoQKwBvPruHQyW4OnuiKuiois45CRUIVjzsnu/qpzpLuL4A3rl0AwGN7jkdcE5HZR6EioWrrGWAo7lnVUllWXcyq+SUaVxEJgUJFQnWisw+A+aXZEyoAbzi7hq0HTtLdr6nFIumkUJFQnehMXPg4P+Kr6Ue6Zt1C+gbjaq2IpJlCRUKVrS2VK1ZWU12Sz5ZfNURdFZFZRaEioToZhEp1lrVUYjnGW89fxM93H9fdIEXSSKEioTrR2U8sxyK96+NYrj+/lu7+IXWBiaSRQkVCdaKzj6qSfHJyLOqqnOHKVVVUleTzw+deiboqIrOGQkVCdaKzP+u6voblxnJ41yV1PLKr8dTYj4hMj0JFQnWis4+aLBukT/bey5YyGHcefFatFZF0CDVUzOxaM9tjZvvM7LZR9puZ3R7sf8HM1k9U1syqzOxhM3sp+FkZbF9hZj1m9nzwuDPMzyapOdHZl7UtFYA1C0u5dHkl//rUIYbiHnV1RGa80ELFzGLAHcB1wDrgfWa2bsRh1wFrgsfNwNdTKHsb8Ki7rwEeDX4ftt/dLw4et4TzySRV7s7xjj4WlhdGXZVxffh1Kzl0spufvXgs6qqIzHhhtlQuB/a5+wF37wfuAzaOOGYjcK8nbAUqzKx2grIbgXuC5/cAvxHiZ5BpaOkeoH8wzsLS7A6Vt5y3iFXzS/jaY/twV2tFZDrCDJU64EjS7/XBtlSOGa/sQndvAAh+Lkg6bqWZPWdmj5vZ60erlJndbGbbzWx7U5Omkoapsb0XgEVZ3lKJ5Rgf+bWz2HG0nR8+r7EVkekIM1RGm0M68s/AsY5JpexIDcAyd78E+ATwHTMrO+NF3O9y9w3uvqGmpmaCl5TpOBaEysKy7B2oH/auS+q4aEk5n9uym47egairIzJjhRkq9cDSpN+XAEdTPGa8so1BFxnBz+MA7t7n7ieD588A+4G1afkkMiWNbcOhkt0tFYCcHOOvNp7Pya5+PvmDX6kbTGSKwgyVbcAaM1tpZvnAJmDziGM2AzcEs8CuBNqCLq3xym4Gbgye3wj8CMDMaoIBfsxsFYnB/wPhfTyZSGN74tqPBVk+pjLs4qUV/PFb1vLjFxr4+uP7o66OyIyUG9YLu/ugmX0UeAiIAXe7+w4zuyXYfyewBbge2Ad0AzeNVzZ46c8D95vZh4DDwLuD7VcDf21mg8AQcIu7N4f1+WRix9p7qS7JJz935lwOdcvVq9nV0MEXfraHgUHnD379rKxcDUAkW4UWKgDuvoVEcCRvuzPpuQO3plo22H4SeNMo2x8AHphmlSWNGtt7Z0TXV7KcHOPL77mIvBzjy4/s5bkjLXz2nRdQV1EUddVEZoSZ8yekzDiJUMn+QfqRcmM5fPE9F/H/Np7Hk/tP8qYvPsZXH31JqxmLpEChIqFpbO/N+unEYzEzfueqFTz6x2/gTecs5IsP7+WaLz/Of+w4pkF8kXEoVCQU/YNxTnT2z5hB+rEsqSzmjvev5zsfvoKivBg3f+sZbvqXbRwPpkuLyOkUKhKK4Qsfa2doS2Wk15w1n5/84ev5s7edy9YDJ7n2K0/wn7uPR10tkayjUJFQ1Lf0AIm/9GeLvFgOH379Kn78B69jQWkBN/3LNi3tIjKCQkVCcaSlG4ClVbNv1tRZC0r54a2v5R0XLeYLP9vDpx58kcGheNTVEskKoU4plrmrvqUHM6gtn32hAlCYF+Pv33sxSyqL+Npj++noHeDv33sxuTH9nSZzm0JFQlHf3E1tWeGMuvBxsnJyjP9z7TlUFOfxN1t248BXFCwyxylUJBT1LT2zajxlPDdfvRrD+OyWXeDwlU0KFpm7FCoSiiMt3Vy1ujrqamTM7129CjP4zE92nboqX8Eic5FCRdKufzDOsfbeOdNSGfbh168i7s7fbNlNzOCL77mYmNYNkzlGoSJp19DWgzssrZydg/Tjufnq1QzGnS/8bA85Ocbf/dZFChaZUxQqknZHmmffNSqT8ZE3nsXgkPOlh/eSm2N8/l0XaqVjmTMUKpJ2B090ArByfknENYnOH75pDYNx5/ZHXyKWk8Nnf+N8BYvMCQoVSbv9TV2U5Mdm5ArF6fRHb17DUDzOHf+5n/7BOJ971wWzeoq1CChUJAT7mzpZvWAeZnP7L3Mz40/ecjb5sRhffmQvR1t7uPMDl1JenBd11URCoz+bJO0ONHWxumZe1NXICmbGx968hi+/9yKeOdTCO7/+3xxo6oy6WiKhUahIWnX3D/JKaw+r5vB4ymjeeckS/vXDV9DaPcDGO/6bx/c2RV0lkVAoVCStDjR1AbBKLZUzXL6yih/d+lrqKoq46Z+f5ptPHNAKxzLrKFQkrXY1tANwTm1pxDXJTkurivnBR17DW89bxGd+sos/+f4Luk2xzCoKFUmrHUfbKc6PsaJa3V9jKc7P5Y7fXs8fvXktDzxbz6a7ttLS1R91tUTSQqEiabXzaDvnLCrVVeQTyMlJDODf+YH17Gxo53fufoq27oGoqyUybQoVSZt43NnZ0M55i8ujrsqMce35tXzjdy5l77FObvjnp+nuH4y6SiLTolCRtDnU3E1n3yDrFpdFXZUZ5dfOXsAd71/Pr+pb+dh9zzMU1+C9zFwKFUmb7S83A3Dp8sqIazLzXLNuIX/x9nU8vLORz23ZFXV1RKZMV9RL2mx7uZnyojzO0nTiKfnga1dy8EQX3/zFQVbML+EDVy6Pukoik6ZQkbTZ/nILG5ZXauHEafjzt6/jcHM3/3fzDpZXF/P6NTVRV0lkUtT9JWlxrK2XAye6uGxlVdRVmdFyYzl89bfXs2bBPD7y7WfZd7wj6iqJTIpCRdLi57uPA4lBZ5meeQW5fPPGDRTkxrjpX7ZxsrMv6iqJpEyhImnx6K5GllQWsXahxlPSYUllMf94w6Ucb+/j97/1zJSnGg8MxXn2cAv3bzvCPf/zMlt+1UB9S3eaayvyKo2pyLS19w7wi30neN/ly+b8cvfpdMmySr783ov56Hee5YN3b+Pumy5jXkFq/8u2dvfz7acOc++TL9PYfmZL54K6cj74mhW84+LF5MX0t6Wkj0JFpm3z80fpG4zzrvV1UVdl1rn+glq+sukSPv695/nNr/0Pd7x/PWctGLs1uL+pk3/+74P82zP19A7Eed1Z8/nzt6/jgrpyivNzaWzv5cn9J7l/+xH++Pu/5I7H9vHp68/l189ZoD8IJC1sLq+SumHDBt++fXvU1ZjR3J23f/UXDMWdn37s9fqHKSRPvNTEx+57np7+IT70upW874pl1FUUAdDVN8gTLzVx37YjPLanifxYDhsvXszvvm4l59aOfiGqu/PwzkY+/9PdHDjRxWtWV/PJ687lgiVaDUEmZmbPuPuGUfcpVBQq0/HIzkY+fO92/vY3L+C9ly2Lujqz2rG2Xj7zk538+IUGAGpKC8jLMRo7+hiKOwvLCth02TI+cOVyakpTu5XzwFCcb289xO0/30dzVz8bL17MJ65Zy3ItCCrjUKiMQaEyPb0DQ7zt9icYjDuPfOIN6pvPkIMnunh45zH2H+9iIB6nrqKIK1ZWc+WqKnKn+N+gvXeAbzy+n3/6xUH6BuNcvaaGd1y0mNeeNZ9F5YVp/gQy0ylUxqBQmTp351MPvsh3nz7MPb97OW9Yq4v0ZoPG9l6+/dRh7t92hGPtvQBUFOextLKY8qI8ivNjzCvIpawoj8rifCqK81hcUcRZC+axrKpYq1PPEeOFSqgD9WZ2LfAVIAZ8090/P2K/BfuvB7qBD7r7s+OVNbMq4HvACuBl4D3u3hLs+yTwIWAI+EN3fyjMzzdX9Q4M8dc/3sl3nz7MLW9YrUCZRRaWFfKJa9by8TetYdexdrYeaGZ/Uyf1LT109g7Q1NFHZ98g7T0DdPSdPs05PzeHsxeWsq62jPPqyjhvcRnn1pZRnJ++f2Z6B4Zo6uijqbOPpo4+uvoGcW7suLUAAAe3SURBVIe4O6WFuVTPK6C6JJ/a8iKK8mNpe19JXWgtFTOLAXuBa4B6YBvwPnffmXTM9cAfkAiVK4CvuPsV45U1sy8Aze7+eTO7Dah09z81s3XAd4HLgcXAI8Badx/ztnpqqaTG3TnZ1c/LJ7r4xb4TfPfpwzS29/H7b1jFbdeeo8H5OWpgKE5bzwBHmrt56XgnLzV2sKuhgx1H22gJ7g1jBivnl3De4nLOW1zGOYtKmT+vgPKiPOYV5GIG8SAUuvuGaOsZoLWnn5buAY619XC0tZdXWnt4paWHo209tE7injOVxXnUlhexuKKIxRWFLK4oora8kLqKImorilhYWjDl7sK5LqqWyuXAPnc/EFTiPmAjsDPpmI3AvZ5Itq1mVmFmtSRaIWOV3Qi8MSh/D/AY8KfB9vvcvQ84aGb7gjo8me4PtvtYOx/9znOn7i9+KpaT8nn46chjhjPcgy2nfh+R7SmXG7GfMfePqOtp7z365xj+vX8wTv9QHEj8I3Hlymr+/r2XcNXqamTuyovlMH9eAfPnFXDJsldXpnZ3Gtp62XG0nR1H29h5tJ1nD7Xw7788Oun3KC3Mpa4iEQzrl1ewqKyQBaWF1JQWUFNaQElBLrHgj5r23gGau/o50dlHQ1svDUEo1bd08/TBk7T3nt6yyjEoyc8lN2bkxXKCh5Ez2h9JNu6viW0jymX7n1pvPLuGT79tXdpfN8xQqQOOJP1eT6I1MtExdROUXejuDQDu3mBmw+uC1AFbR3mt05jZzcDNAMuWTW22UmFujLMXBvdgt9N+nPbFenXb6MfYiAMteHLm8WPsH/ECqZYbq75jlcmNGYvKCllSWcyG5ZVUluQjMhYzC1oHRVyzbuGp7a3d/ext7KS1u5+2ngE6gn/kYzmGGRTlxagIxmkqivJYWF5IWWFe2urV2TdIQ2sPR9t6OdraQ0NrD519QwzG4wwMxekfdAaG4ozsuxnZmzNq387IPwpHPyqrLCwLZwJGmKEyWlCPPNNjHZNK2am8H+5+F3AXJLq/JnjNUa2YX8Id718/laIic1ZFcT6XR7jg6LyCXNYsLGXN8B+EEoowOxTrgaVJvy8BRrZ/xzpmvLKNQRcZwc/jk3g/EREJUZihsg1YY2YrzSwf2ARsHnHMZuAGS7gSaAu6tsYruxm4MXh+I/CjpO2bzKzAzFYCa4Cnw/pwIiJyptC6v9x90Mw+CjxEYlrw3e6+w8xuCfbfCWwhMfNrH4kpxTeNVzZ46c8D95vZh4DDwLuDMjvM7H4Sg/mDwK3jzfwSEZH008WPmlIsIjIp400p1iRtERFJG4WKiIikjUJFRETSRqEiIiJpM6cH6s2sCTiUhpeaD5xIw+vMZjpHE9M5mpjO0cQycY6Wu/uoK8nO6VBJFzPbPtZMCEnQOZqYztHEdI4mFvU5UveXiIikjUJFRETSRqGSHndFXYEZQOdoYjpHE9M5mlik50hjKiIikjZqqYiISNooVEREJG0UKhMws3eb2Q4zi5vZhhH7Pmlm+8xsj5m9NWn7pWb2q2Df7RbcXjFYlv97wfanzGxFZj9N5pnZtcH52Wdmt0Vdn0wys7vN7LiZvZi0rcrMHjazl4KflUn7JvV9mg3MbKmZ/aeZ7Qr+P/tYsF3nKWBmhWb2tJn9MjhHfxVsz85z5O56jPMAzgXOBh4DNiRtXwf8EigAVgL7gViw72ngKhJ3o/wpcF2w/SPAncHzTcD3ov58IZ+7WHBeVgH5wflaF3W9Mvj5rwbWAy8mbfsCcFvw/Dbgb6f6fZoND6AWWB88LwX2BudC5+nVc2TAvOB5HvAUcGW2niO1VCbg7rvcfc8ouzYC97l7n7sfJHFPmMuDu1GWufuTnviveC/wG0ll7gme/xvwptny19QYLgf2ufsBd+8H7iNxDuYEd/8voHnE5uTvwD2c/t2Y7PdpxnP3Bnd/NnjeAewC6tB5OsUTOoNf84KHk6XnSKEydXXAkaTf64NtdcHzkdtPK+Pug0AbUB16TaMz1jmayxZ64u6mBD8XBNun8n2aVYLu4EtI/CWu85TEzGJm9jyJ26c/7O5Ze45Cu/PjTGJmjwCLRtn1aXf/0SjbIdF8HMnH2T5emdlqrn3e6ZjK92nWMLN5wAPAx929fZwG/Jw8T564i+3FZlYBPGhm549zeKTnSKECuPubp1CsHlia9PsS4Giwfcko25PL1JtZLlDOmd0js8lY52guazSzWndvCLojjgfbp/J9mhXMLI9EoHzb3X8QbNZ5GoW7t5rZY8C1ZOk5UvfX1G0GNgUzulYCa4Cng2Zoh5ldGYyX3AD8KKnMjcHz3wJ+HvRtzlbbgDVmttLM8klMTtgccZ2ilvwduJHTvxuT/T7NeMFn+idgl7t/KWmXzlPAzGqCFgpmVgS8GdhNtp6jqGc2ZPsDeCeJhO8DGoGHkvZ9msTMij0kzaIANgAvBvv+gVdXLigEvk9i4OxpYFXUny8D5+96EjN69pPoToy8Thn87N8FGoCB4Dv0IRJjaI8CLwU/q6b6fZoND+B1JLpgXgCeDx7X6zyddo4uBJ4LztGLwF8E27PyHGmZFhERSRt1f4mISNooVEREJG0UKiIikjYKFRERSRuFioiIpI1CRURE0kahIiIiafP/AYRrP0UNKAnbAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df1.prix.plot(kind = \"kde\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Grâce à ce boxplot nous constatons que le prix du téléphone dépend assez grandement de sa technologie réseau.\\\n",
    "Plus la technologie réseau est élevée plus le prix en moyenne est élevé."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20f5b10fd90>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14, 5))\n",
    "sns.boxplot(x=\"reseau\", y=\"prix\", data=df1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Grâce à ce diagramme en bâton nous observons que les téléphones de marque Samsung et apple sont les plus représentés ensuite viennent les marques telles que xiaomi, huawei et la variable \"inconnu\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20f5b18c7c0>"
      ]
     },
     "execution_count": 31,
     "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": [
    "t = pandas.crosstab(df1.Marque, \"freq\")\n",
    "t.plot.bar()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dans ce graphique nous représentons des boxplot qui nous montrent les prix par marques. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20f5b17cee0>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14, 5))\n",
    "sns.boxplot(x=\"Marque\", y=\"prix\", data=df1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ici nous représentons un camembert qui repreneuse l'effectif des téléphones pour chaque RAM disponiple."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([<matplotlib.axes._subplots.AxesSubplot object at 0x0000020F5B1B2580>],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "t = pandas.crosstab(df1.Ram, \"freq\")\n",
    "t.plot.pie(subplots=True, figsize = (8,8))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ce graphique nous montre la quantité de téléphone en fonction de sa technologie réseau et de sa mémoire. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20f5b565b50>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "t = pd.crosstab(df1.reseau, df1.Mémoire)\n",
    "t.plot.bar( figsize = (8,8))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Partie Apprentissage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pour commencer la partie apprentissage nous commençons par définir les variables explicatives et la variable à expliquer.\\\n",
    "En ce qui nous concerne la variable à expliquer est le prix en fonction des variables marques, réseau, RAM, écran, mémoire."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "X, y = df1[[\"Marque\",\"reseau\",\"Ram\",\"écran\",\"Mémoire\"]], df1[\"prix\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nous vérifions que nous avons bien le même nombre de lignes dans les matrices associé aux variables explicatives et à la variable à expliquer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((977, 5), (977,))"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape, y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Séparation de la base de données en données d'entrainement et données d'apprentissage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "quali = [\"Marque\",\"reseau\",\"Ram\",\"écran\",\"Mémoire\"] \n",
    "\n",
    "qualitrans=Pipeline(steps=[('imputer',OneHotEncoder(handle_unknown=\"ignore\"))])\n",
    "\n",
    "transformer = ColumnTransformer(transformers=[('cat',qualitrans,quali)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*  Grâce à cette fonction nous allons pouvoir modéliser  plusieurs modèles avec différents paramètres.\n",
    "* Les modèles qui sont modélisés sont les suivants:*LinearRegression, Lasso, Ridge, ElasticNet, KNeighborsRegressor, SVR, RandomForestRegressor, MLPRegressor*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 1858602.4993796684, tolerance: 6767.307471662021\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 1948954.9703600053, tolerance: 6772.872292048161\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 1509948.71864085, tolerance: 7108.9050141169355\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 2043054.927793432, tolerance: 7116.687170477754\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 1585327.2739005527, tolerance: 7392.553895035389\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 2194330.353223826, tolerance: 7276.594141967224\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 1748795.3194734687, tolerance: 7184.292997664735\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 1513494.8524612961, tolerance: 7205.1361814040965\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 1967315.8385087969, tolerance: 7080.577867678696\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 2272763.4586276365, tolerance: 7256.084707702792\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 69926.2857366763, tolerance: 6767.307471662021\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 82175.59296747111, tolerance: 6772.872292048161\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 63324.455928333104, tolerance: 7108.9050141169355\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 93582.7753014192, tolerance: 7116.687170477754\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 68789.96738585085, tolerance: 7392.553895035389\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 80866.27344766445, tolerance: 7184.292997664735\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 76499.62041322328, tolerance: 7205.1361814040965\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 74568.51680404693, tolerance: 7080.577867678696\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 178930.90097540803, tolerance: 7256.084707702792\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 212496.7250605654, tolerance: 6767.307471662021\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 105983.81384647265, tolerance: 6772.872292048161\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 133829.99757066742, tolerance: 7108.9050141169355\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 201974.51321439445, tolerance: 7116.687170477754\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 190036.76603031158, tolerance: 7392.553895035389\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 123395.61714218184, tolerance: 7276.594141967224\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 164370.48509822972, tolerance: 7184.292997664735\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 252540.88571946882, tolerance: 7205.1361814040965\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 234424.1334314607, tolerance: 7080.577867678696\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 499822.73349139653, tolerance: 7256.084707702792\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 3996819.9923812756, tolerance: 6767.307471662021\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 3716349.04811756, tolerance: 6772.872292048161\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 3900057.971116151, tolerance: 7108.9050141169355\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 4395180.008292127, tolerance: 7116.687170477754\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 4004830.7958339285, tolerance: 7392.553895035389\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 3423916.5385573497, tolerance: 7276.594141967224\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 3956778.2662176713, tolerance: 7184.292997664735\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 4262094.540383992, tolerance: 7205.1361814040965\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 4289581.2405093135, tolerance: 7080.577867678696\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 4508074.235980541, tolerance: 7256.084707702792\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "\n",
    "modeles = []\n",
    "modeles.append(LinearRegression())\n",
    "\n",
    "for val_alpha in (1e-3, 1e-2, 1e-1, 1):\n",
    "    modeles.append(Lasso(alpha=val_alpha))\n",
    "    \n",
    "for val_alpha in (1e-3, 1e-2, 1e-1, 1):\n",
    "    modeles.append(Ridge(alpha=val_alpha))\n",
    "\n",
    "for val_alpha in (1e-3, 1e-2, 1e-1, 1):\n",
    "    for val_l1 in (0.25, 0.5, 0.75):\n",
    "        modeles.append(ElasticNet(alpha=val_alpha, l1_ratio=val_l1))\n",
    "    \n",
    "for nb_voisins in range(3, 10):\n",
    "    modeles.append(KNeighborsRegressor(n_neighbors=nb_voisins))\n",
    "\n",
    "\n",
    "\n",
    "for val_epsilon in (10 ** n for n in range(-3, 1)):\n",
    "    for val_C in (10 ** n for n in range(-3, 4)):\n",
    "        modeles.append(SVR(epsilon=val_epsilon, C=val_C))\n",
    "\n",
    "\n",
    "for nb_estimateurs in (50, 100, 150, 200):\n",
    "    modeles.append(RandomForestRegressor(n_estimators=nb_estimateurs))\n",
    "    \n",
    "for nb_neurones in  ((100,), (50, 50), (25, 50, 25)):\n",
    "    modeles.append(MLPRegressor(hidden_layer_sizes=nb_neurones))\n",
    "    \n",
    "i=0\n",
    "Classificateur=[]\n",
    "for i in range(len(modeles)):\n",
    "    liste= Pipeline(steps=[(\"processor\",transformer),\n",
    "                        (\"classi\",modeles[i])])\n",
    "    Classificateur.append(liste) \n",
    "    i=i+1\n",
    "    \n",
    "i=0\n",
    "resultats=[]\n",
    "for i in range(len(Classificateur)):\n",
    "    cr=cross_val_score(Classificateur[i],X_train,y_train,cv=10)\n",
    "    resultats.append(cr)\n",
    "    i=i+1\n",
    "DeprecationWarning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Performance des modèles sur les données d'entrainement suite à une cross-validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Dico={}\n",
    "i=0\n",
    "for i in range(len(resultats)):\n",
    "    Dico[modeles[i]]=[resultats[i].mean(),resultats[i].std()]\n",
    "    i=i+1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestRegressor(n_estimators=200)            : [0.8500670896710559, 0.0452415438631804]\n",
      "RandomForestRegressor(n_estimators=50)             : [0.8459837252851836, 0.0425471041385668]\n",
      "RandomForestRegressor(n_estimators=150)            : [0.8447658919372051, 0.04981255906222098]\n",
      "RandomForestRegressor()                            : [0.8445608626103738, 0.04700879066574336]\n",
      "SVR(C=1000, epsilon=1)                             : [0.8098230198250691, 0.07335370999561991]\n",
      "SVR(C=1000)                                        : [0.8095997135884161, 0.07344033867810665]\n",
      "SVR(C=1000, epsilon=0.01)                          : [0.8095796211923375, 0.07345039282427256]\n",
      "SVR(C=1000, epsilon=0.001)                         : [0.8095776309920767, 0.07345137517969176]\n",
      "MLPRegressor(hidden_layer_sizes=(25, 50, 25))      : [0.7880671546626665, 0.06643525081167745]\n",
      "KNeighborsRegressor(n_neighbors=4)                 : [0.7842880696893514, 0.05980865008094562]\n",
      "ElasticNet(alpha=0.001, l1_ratio=0.75)             : [0.7702063318411752, 0.0636469239994682]\n",
      "ElasticNet(alpha=0.001)                            : [0.7698429991044152, 0.06275959493907071]\n",
      "Lasso(alpha=0.1)                                   : [0.7698412062902722, 0.06869858426347178]\n",
      "Ridge(alpha=0.1)                                   : [0.7691111185490316, 0.06462507252305312]\n",
      "KNeighborsRegressor(n_neighbors=7)                 : [0.7683026191659096, 0.07402266643600348]\n",
      "ElasticNet(alpha=0.001, l1_ratio=0.25)             : [0.7677688774982762, 0.062388596575485004]\n",
      "Lasso(alpha=0.01)                                  : [0.767522895219372, 0.06756790678095531]\n",
      "KNeighborsRegressor(n_neighbors=6)                 : [0.7671818537071479, 0.0690334039167989]\n",
      "Ridge(alpha=0.01)                                  : [0.7661004514614943, 0.06642412961329465]\n",
      "LinearRegression()                                 : [0.7657426570448935, 0.06694350285629931]\n",
      "Ridge(alpha=0.001)                                 : [0.7655698753785873, 0.06676112138715604]\n",
      "Lasso(alpha=0.001)                                 : [0.7651284465200551, 0.06603228815959183]\n",
      "KNeighborsRegressor()                              : [0.7636269476248141, 0.06695161885488107]\n",
      "KNeighborsRegressor(n_neighbors=8)                 : [0.763155832677435, 0.07793604646104177]\n",
      "Ridge(alpha=1)                                     : [0.7588592883490233, 0.06169738933320693]\n",
      "KNeighborsRegressor(n_neighbors=9)                 : [0.758750668803495, 0.08357634932684442]\n",
      "ElasticNet(alpha=0.01, l1_ratio=0.75)              : [0.7478867150412051, 0.060729492495686335]\n",
      "KNeighborsRegressor(n_neighbors=3)                 : [0.7474435584855071, 0.06864788697767678]\n",
      "MLPRegressor(hidden_layer_sizes=(50, 50))          : [0.73706482468824, 0.05882102735694653]\n",
      "Lasso(alpha=1)                                     : [0.7340164644619082, 0.060741042801707626]\n",
      "ElasticNet(alpha=0.01)                             : [0.7287063858786114, 0.059124176651952895]\n",
      "ElasticNet(alpha=0.01, l1_ratio=0.25)              : [0.7159576895084248, 0.05826421897785568]\n",
      "SVR(C=100, epsilon=1)                              : [0.695473566875257, 0.10161151403774481]\n",
      "SVR(C=100)                                         : [0.6953442974371866, 0.1017880481651834]\n",
      "SVR(C=100, epsilon=0.01)                           : [0.6953336521047896, 0.10181214582099625]\n",
      "SVR(C=100, epsilon=0.001)                          : [0.6953324961986436, 0.10181461403612185]\n",
      "ElasticNet(alpha=0.1, l1_ratio=0.75)               : [0.6630105553011121, 0.05942113080609634]\n",
      "ElasticNet(alpha=0.1)                              : [0.6161218792365035, 0.061516868709634125]\n",
      "ElasticNet(alpha=0.1, l1_ratio=0.25)               : [0.580505321422616, 0.06161192520668646]\n",
      "ElasticNet(alpha=1, l1_ratio=0.75)                 : [0.42373936097910914, 0.05597026354933814]\n",
      "ElasticNet(alpha=1)                                : [0.32128102448027007, 0.04685775993289569]\n",
      "SVR(C=10, epsilon=1)                               : [0.2901406494417049, 0.0822939287545958]\n",
      "SVR(C=10)                                          : [0.290014799651673, 0.08233369251084098]\n",
      "SVR(C=10, epsilon=0.01)                            : [0.29001463506434344, 0.08233397622092271]\n",
      "SVR(C=10, epsilon=0.001)                           : [0.2900145370556694, 0.08233410130008821]\n",
      "ElasticNet(alpha=1, l1_ratio=0.25)                 : [0.26037999933823075, 0.04012567108874257]\n",
      "MLPRegressor()                                     : [0.1908685637289355, 0.059739153956505385]\n",
      "SVR(C=1, epsilon=0.001)                            : [-0.06895972375212187, 0.06804987049732593]\n",
      "SVR(C=1, epsilon=0.01)                             : [-0.0689672466627875, 0.06804994946110285]\n",
      "SVR(C=1)                                           : [-0.06905284784031904, 0.0680620869930629]\n",
      "SVR(C=1, epsilon=1)                                : [-0.06941555101937173, 0.06819889112851797]\n",
      "SVR(C=0.1)                                         : [-0.13321677654780975, 0.07529650064894326]\n",
      "SVR(C=0.1, epsilon=0.01)                           : [-0.13323051784243803, 0.07535330871458967]\n",
      "SVR(C=0.1, epsilon=0.001)                          : [-0.13323146768701105, 0.07535751627066044]\n",
      "SVR(C=0.1, epsilon=1)                              : [-0.13348751915409646, 0.0751763786681212]\n",
      "SVR(C=0.01, epsilon=1)                             : [-0.13965112664052926, 0.07587679947450317]\n",
      "SVR(C=0.01, epsilon=0.001)                         : [-0.1396785483240938, 0.07659882899870603]\n",
      "SVR(C=0.01, epsilon=0.01)                          : [-0.13968015837997966, 0.07659730792898009]\n",
      "SVR(C=0.01)                                        : [-0.1396848956008851, 0.07655911278723776]\n",
      "SVR(C=0.001, epsilon=1)                            : [-0.14024969239578106, 0.07598076725192401]\n",
      "SVR(C=0.001, epsilon=0.001)                        : [-0.14027052355811656, 0.07668965903353325]\n",
      "SVR(C=0.001, epsilon=0.01)                         : [-0.14027213390387971, 0.07668814061742786]\n",
      "SVR(C=0.001)                                       : [-0.1402817053415087, 0.07663298789779514]\n"
     ]
    }
   ],
   "source": [
    "for key, value in (sorted(Dico.items(), key=lambda x: x[1],reverse=True)) :\n",
    "    print(f'{repr(key):50} : {value}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 2163020.204727728, tolerance: 7907.571891470534\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 84049.36738547683, tolerance: 7907.571891470534\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 210658.7577155307, tolerance: 7907.571891470534\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_coordinate_descent.py:512: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Duality gap: 4539450.678364381, tolerance: 7907.571891470534\n",
      "  model = cd_fast.sparse_enet_coordinate_descent(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "C:\\Users\\caleb\\Anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:582: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    }
   ],
   "source": [
    "modfit=[]\n",
    "for i in range(len(Classificateur)):\n",
    "    L = Classificateur[i].fit(X_train, y_train)\n",
    "    modfit.append(L)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Performance des modèles sur les données de test suite à une cross-validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "Dicoscoretest={}\n",
    "i=0\n",
    "for i in range(len(modeles)):\n",
    "    Dicoscoretest[modeles[i]]=[modfit[i].score(X_test, y_test)]\n",
    "    i=i+1\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestRegressor(n_estimators=200)            : [0.8241934501947241]\n",
      "RandomForestRegressor(n_estimators=50)             : [0.8230916624617003]\n",
      "RandomForestRegressor(n_estimators=150)            : [0.8226891585261457]\n",
      "RandomForestRegressor()                            : [0.8225567096535018]\n",
      "MLPRegressor(hidden_layer_sizes=(25, 50, 25))      : [0.8199349473220139]\n",
      "SVR(C=1000, epsilon=1)                             : [0.8061313813667815]\n",
      "SVR(C=1000)                                        : [0.8060535127113734]\n",
      "SVR(C=1000, epsilon=0.01)                          : [0.8060404785478497]\n",
      "SVR(C=1000, epsilon=0.001)                         : [0.8060391439345816]\n",
      "KNeighborsRegressor(n_neighbors=9)                 : [0.7997880650492766]\n",
      "KNeighborsRegressor(n_neighbors=8)                 : [0.7977441366148311]\n",
      "KNeighborsRegressor(n_neighbors=6)                 : [0.7890954586828117]\n",
      "KNeighborsRegressor(n_neighbors=7)                 : [0.7835892341103244]\n",
      "KNeighborsRegressor()                              : [0.7824742624418706]\n",
      "SVR(C=100, epsilon=1)                              : [0.7798069749108093]\n",
      "SVR(C=100)                                         : [0.7795511266850292]\n",
      "SVR(C=100, epsilon=0.01)                           : [0.7795199543309009]\n",
      "SVR(C=100, epsilon=0.001)                          : [0.7795167615145757]\n",
      "MLPRegressor(hidden_layer_sizes=(50, 50))          : [0.7758662735787456]\n",
      "KNeighborsRegressor(n_neighbors=4)                 : [0.7553108015022969]\n",
      "Lasso(alpha=0.01)                                  : [0.7468772179063484]\n",
      "LinearRegression()                                 : [0.7468680817479408]\n",
      "ElasticNet(alpha=0.001, l1_ratio=0.75)             : [0.7464079915366889]\n",
      "Ridge(alpha=0.01)                                  : [0.7463450932926271]\n",
      "Ridge(alpha=0.1)                                   : [0.7462453911340905]\n",
      "Ridge(alpha=0.001)                                 : [0.7460690213617304]\n",
      "Lasso(alpha=0.001)                                 : [0.7459852488514643]\n",
      "ElasticNet(alpha=0.001)                            : [0.745786676012393]\n",
      "ElasticNet(alpha=0.001, l1_ratio=0.25)             : [0.7450137827610144]\n",
      "Ridge(alpha=1)                                     : [0.7424028894442701]\n",
      "Lasso(alpha=0.1)                                   : [0.7417129320910988]\n",
      "ElasticNet(alpha=0.01, l1_ratio=0.75)              : [0.7374514304010316]\n",
      "KNeighborsRegressor(n_neighbors=3)                 : [0.7343349939837853]\n",
      "ElasticNet(alpha=0.01)                             : [0.7280178121118541]\n",
      "ElasticNet(alpha=0.01, l1_ratio=0.25)              : [0.7201684283338825]\n",
      "Lasso(alpha=1)                                     : [0.7081107205870469]\n",
      "ElasticNet(alpha=0.1, l1_ratio=0.75)               : [0.6797373086550054]\n",
      "ElasticNet(alpha=0.1)                              : [0.640329807886225]\n",
      "ElasticNet(alpha=0.1, l1_ratio=0.25)               : [0.6084014280888481]\n",
      "ElasticNet(alpha=1, l1_ratio=0.75)                 : [0.45390505225860756]\n",
      "SVR(C=10, epsilon=0.001)                           : [0.40044649646477726]\n",
      "SVR(C=10, epsilon=0.01)                            : [0.4004394419302848]\n",
      "SVR(C=10)                                          : [0.40036797658152656]\n",
      "SVR(C=10, epsilon=1)                               : [0.400053671580084]\n",
      "ElasticNet(alpha=1)                                : [0.347710388577246]\n",
      "ElasticNet(alpha=1, l1_ratio=0.25)                 : [0.28345319563502747]\n",
      "MLPRegressor()                                     : [0.26164906361977336]\n",
      "SVR(C=1)                                           : [-0.009079485324649195]\n",
      "SVR(C=1, epsilon=0.01)                             : [-0.009087247140311838]\n",
      "SVR(C=1, epsilon=0.001)                            : [-0.009088023496207409]\n",
      "SVR(C=1, epsilon=1)                                : [-0.00965578734455641]\n",
      "SVR(C=0.1, epsilon=1)                              : [-0.08578905232147926]\n",
      "SVR(C=0.1, epsilon=0.001)                          : [-0.08583309877427703]\n",
      "SVR(C=0.1, epsilon=0.01)                           : [-0.08583309877427703]\n",
      "SVR(C=0.1)                                         : [-0.08583309877427703]\n",
      "SVR(C=0.01, epsilon=1)                             : [-0.0927329407541233]\n",
      "SVR(C=0.01, epsilon=0.001)                         : [-0.0930006957785301]\n",
      "SVR(C=0.01, epsilon=0.01)                          : [-0.0930006957785301]\n",
      "SVR(C=0.01)                                        : [-0.0930006957785301]\n",
      "SVR(C=0.001, epsilon=1)                            : [-0.09323254078863497]\n",
      "SVR(C=0.001, epsilon=0.001)                        : [-0.09372002390543743]\n",
      "SVR(C=0.001, epsilon=0.01)                         : [-0.09372002390543743]\n",
      "SVR(C=0.001)                                       : [-0.09372002390543743]\n"
     ]
    }
   ],
   "source": [
    "for key, value in (sorted(Dicoscoretest.items(), key=lambda x: x[1],reverse=True)) :\n",
    "    print(f'{repr(key):50} : {value}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Àprès avoir testé et ordonné tous les modèles sur les données d'entrainement et les données de test du meilleur au plus mauvais, nous constatons que les modèles sont les Random Forest. Et plus précisement celle avec un n_estimators de défaut c'est à 200(dans mon cas). \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "import machine"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Nous utilisons le modfit[57] car elle correspond au modèle random forest avec un n_estimators par défaut."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "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>predicted</th>\n",
       "      <th>true</th>\n",
       "      <th>difference</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>188.416029</td>\n",
       "      <td>172.58</td>\n",
       "      <td>15.836029</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    predicted    true  difference\n",
       "1  188.416029  172.58   15.836029"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "machine.prediction(df1,modfit[57],1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "La fonction prediction provenant de machine.py fait une estimation du prix par rapports aux différentes variables explicatives.Elle estime le meilleur prix possible en fonction des données.\n",
    "\n",
    "Ici les données proviennent de la base de données déjà existante que nous avons mais nous aurions bien évidemment pu modifier la fonction pour que nous-mêmes entrons les valeurs des variables explicatives et non pas qu'elles proviennent de la base de données déjà existante ou fictive.Et faire bien plus encore"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nous avons laissé certaine fonction comme celle du scrraping et celle sur les modèles(cross-validation) pour montrer la complexité."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# fin"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Aucun(e)",
  "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
}